The Lvov-Warsaw School. Past and Present. Ángel Garrido - Urszula Wybraniec-Skardowska (Eds.)

Studies in Universal Logic
Ángel Garrido
Urszula Wybraniec-Skardowska
Editors
The Lvov-Warsaw
School. Past and
Present
Series Editor
Jean-Yves Béziau (Federal University of Rio de Janeiro and Brazilian Research Council,
Rio de Janeiro, Brazil)
Editorial Board Members
Hajnal Andréka (Hungarian Academy of Sciences, Budapest, Hungary)

Mark Burgin (University of California, Los Angeles, USA)
Răzvan Diaconescu (Romanian Academy, Bucharest, Romania)
Andreas Herzig (Centre National de la Recherche Scientifique, Toulouse, France)
Arnold Koslow (City University of New York, USA)
Jui-Lin Lee (National Formosa University, Huwei Township, Taiwan)
Larissa Maksimova (Russian Academy of Sciences, Novosibirsk, Russia)
Grzegorz Malinowski (University of Łódź, Poland)
Francesco Paoli (University of Cagliari, Italy)
Darko Sarenac (Colorado State University, Fort Collins, USA)
Peter Schröder-Heister (University Tübingen, Germany)
Vladimir Vasyukov (Russian Academy of Sciences, Moscow, Russia)
This series is devoted to the universal approach to logic and the development of a general
theory of logics. It covers topics such as global set-ups for fundamental theorems of logic
and frameworks for the study of logics, in particular logical matrices, Kripke structures,
combination of logics, categorical logic, abstract proof theory, consequence operators, and
algebraic logic. It includes also books with historical and philosophical discussions about
the nature and scope of logic. Three types of books will appear in the series: graduate
textbooks, research monographs, and volumes with contributed papers.
More information about this series at http://www.springer.com/series/7391

Ángel Garrido • Urszula Wybraniec-Skardowska
Editors
The Lvov-Warsaw School.

Past and Present
Editors
Ángel Garrido Urszula Wybraniec-Skardowska
UNED Cardinal Stefan Wyszyński University
Department of Fundamental Mathematics in Warsaw
Madrid, Spain Warsaw, Poland
ISSN 2297-0282 ISSN 2297-0290 (electronic)

ISBN 978-3-319-65429-4 ISBN 978-3-319-65430-0 (eBook)
https://doi.org/10.1007/978-3-319-65430-0
Library of Congress Control Number: 2018943846
Mathematics Subject Classification (2010): 01A72, 03-03
© Springer International Publishing AG, part of Springer Nature 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of
the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation,
broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage
and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or
hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does
not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective
laws and regulations and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are
believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors
give a warranty, express or implied, with respect to the material contained herein or for any errors or omissions
that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps
and institutional affiliations.
Printed on acid-free paper
This book is published under the trade name Birkhäuser, www.birkhauser-science.com by the registered
company Springer International Publishing AG part of Springer Nature.
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Preface
This book is about the Lvov-Warsaw School. I did not “order” this book, but I am very
glad to have received this proposal and the book to be published in a book series I have
created inspired by this school.
The Lvov-Warsaw School is one of the most important schools in the history of logic
but is still not very well known outside of Poland and a circle of aficionados. This book
with contributions about the main members of this school and their works will certainly
help to fill the gap, reinforcing the already existing attraction and promoting new interests
for this school.
I would like to thank the editors of this book, Urszula and Angel, for their considerable
efforts to gather all the papers included in this big volume, as well as all the contributors
of this book and Birkhäuser staff in Basel.
Editor-in-Chief Jean-Yves Beziau

Rio de Janeiro, Brazil
December 19, 2017
v
Contents
Part I Twardowski’s School: The Period of Crystallization of LWS
Introduction. The School: Its Genesis, Development and Significance . . . . . 3

1 A Short History and Influence of the School . . . . . . . . . . . . . . . . . . . 4
2 On the Structure and Contents of the Anthology . . . . . . . . . . . . . . . . . 13
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Kazimierz Twardowski: A Great Teacher of Great Philosophers . . . . . . . . 15

Anna Brożek
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
2 Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
3 Writings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
5 Philosophical School . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
Kazimierz Twardowski’s Works Mentioned in the Paper . . . . . . . . . . . . . . . 32
Jan Łukasiewicz: A Creator of New Ideas in Logic and a Reinterpreter

of Its History . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
Jacek Jadacki
1 Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3 Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
4 Metaphysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5 Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
6 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
7 Teacher . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
Jan Łukasiewicz’s Works Mentioned in the Paper . . . . . . . . . . . . . . . . . . . 45
vii
viii Contents
Kazimierz Ajdukiewicz: The Cognitive Role of Language . . . . . . . . . . . . 47

Anna Jedynak
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
2 Main Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
3 Scientific Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48
4 Disciples and Continuators . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
On Ajdukiewicz’s Project of the Semantic Theory of Knowledge . . . . . . . . 65

Adam Olech
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
2 Presentation of the Meta-Epistemological Project of the Semantic
Theory of Knowledge and Analysis Thereof . . . . . . . . . . . . . . . . . . . 70
3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
Categorial Grammars and Their Logics . . . . . . . . . . . . . . . . . . . . . . . 91

Wojciech Buszkowski
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
2 Basic Categorial Grammars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
3 Lambek Calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
Material Implication and Conversational Implicature in Lvov-Warsaw

School . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
Rafal Urbaniak and Michał Tomasz Godziszewski
1 Ajdukiewicz’s Equivalence Argument . . . . . . . . . . . . . . . . . . . . . . . 117
2 Jackson’s Argument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118
3 Ajdukiewicz on Disjunction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
4 Ajdukiewicz vs. Quine on Assertibility . . . . . . . . . . . . . . . . . . . . . . 121
5 Expressing vs. Stating . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123
6 Ajdukiewicz on the Diagnostics of Improper Use . . . . . . . . . . . . . . . . 124
7 Generalization, Weakening, Moore’s Paradox . . . . . . . . . . . . . . . . . . 125
8 Apparent Connection Between Disjuncts . . . . . . . . . . . . . . . . . . . . . 125
9 Ajdukiewicz’s Diagnostics and Grice’s Cancellability . . . . . . . . . . . . . . 127
10 Goła̧b on Indicative Conditionals . . . . . . . . . . . . . . . . . . . . . . . . . . 130
11 Słupecki’s Reply to Goła̧b . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
12 Ajdukiewicz’s Solution and Goła̧b’s Problem . . . . . . . . . . . . . . . . . . . 131
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
Tadeusz Czeżowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Dariusz Łukasiewicz
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2 Main Papers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
4 Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
5 Pupils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
Contents ix
Tadeusz Czeżowski’s Theory of Knowledge and Beliefs . . . . . . . . . . . . . . 137

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
2 Modern Epistemological Individualism: Faith and Testimony
in Epistemological Framework . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
3 Methodism, Particularism and Belief-Formation Process . . . . . . . . . . . . 141
4 What Can Justify Beliefs: Basic Beliefs, Internalism and Externalism . . . . . 142
5 Czeżowski’s Theory of Knowledge and Epistemological Individualism . . . . 143
6 Czeżowski’s Internalism and Fallibilism . . . . . . . . . . . . . . . . . . . . . 147
7 Czeżowski and Some Views on Belief-Formation Processes . . . . . . . . . . 149
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
What Is Reasoning? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Robert Kublikowski
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
2 The Definition of Reasoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . 155
3 The Conditions of Correctness of Reasoning . . . . . . . . . . . . . . . . . . . 157
4 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
Tadeusz Kotarbiński: Socrates of Warsaw . . . . . . . . . . . . . . . . . . . . . . 165

Jacek Jadacki
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
2 Writings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 166
4 Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5 Pupils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 173
Tadeusz Kotarbiński’s Works Mentioned in the Paper . . . . . . . . . . . . . . . . . 173
Agency in a Praxiological Approach . . . . . . . . . . . . . . . . . . . . . . . . . 175

Wojciech W. Gasparski
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 175
2 Agency in Tadeusz Kotarbiński’s Approach . . . . . . . . . . . . . . . . . . . . 176
3 Other Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 183
4 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 186
Zygmunt Zawirski: A Creator of New Ideas in Methodology of Science

and Scientific Metaphysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
Krzysztof Śleziński
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
2 Writings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192
4 Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 196
5 Pupils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
Zygmunt Zawirski’s Works Mentioned in the Paper . . . . . . . . . . . . . . . . . . 198
x Contents
Zygmunt Zawirski’s Concept of Scientific Metaphysics . . . . . . . . . . . . . . 201

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201
2 Relation Between Science and Metaphysics . . . . . . . . . . . . . . . . . . . 202
3 Synthesis of Human Knowledge and Metaphysics . . . . . . . . . . . . . . . . 204
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 207
Stanisław Leśniewski: Original and Uncompromising Logical Genius . . . . . 209

Peter Simons
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209
2 Phases of Activity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
3 Early Writings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211
4 Antinomies, Classes, Parts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
5 Formalization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213
6 Consolidation and Publication . . . . . . . . . . . . . . . . . . . . . . . . . . . 215
7 Final Years . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
8 Metalogic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 216
9 Teaching and Students . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
10 Colleagues and Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
11 Legacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218
Stanisław Leśniewski’s Works Mentioned in the Paper . . . . . . . . . . . . . . . . 219
Izydora Dambska:
˛ The First Lady of the Twentieth-Century Polish
Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
Anna Brożek and Jacek Jadacki
1 Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
2 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
3 Work . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224
4 The Lvov-Warsaw School Versus Neopositivism and Linguistic
Philosophy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225
5 Metaphysics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 226
6 Semiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 227
7 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229
8 Axiology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231
9 Significance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233
Izydora Dambska’s
˛ Works Mentioned in the Paper . . . . . . . . . . . . . . . . . . 233
Maria Kokoszyńska-Lutmanowa: The Great Polemist . . . . . . . . . . . . . . . 235

Filip Kawczyński
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235
2 Main Papers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 236
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 237
4 Pupils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239
5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 240
Contents xi
Seweryna Łuszczewska-Romahnowa . . . . . . . . . . . . . . . . . . . . . . . . . 241

Roman Murawski and Jerzy Pogonowski
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 241
2 Main Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 242
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
4 Resonance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
5 Pupils . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 245
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 246
On Existential Dependence and Independence in the World of Thoughts

and States of Affairs (with Reference to Eugenia Ginsberg-Blaustein’s
and Roman Ingarden’s Analyses) . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
Urszula M. Żegleń
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
2 Ingarden’s Definitions and Theorems . . . . . . . . . . . . . . . . . . . . . . . 251
3 Dependence in the Mental Sphere of Thoughts . . . . . . . . . . . . . . . . . . 254
4 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
Józef I.M. Bocheński . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

Korneliusz Policki
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263
2 Main Research Interests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
3 Bocheński’s Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 265
4 Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
5 Criticism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
6 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271
J.M. Bocheński’s Theory of Signs . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

Korneliusz Policki
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
2 Ontological Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 273
3 Psychological Assumptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 274
4 The Concept of the Sign . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 275
5 The Theory of Signs and Authority . . . . . . . . . . . . . . . . . . . . . . . . 276
6 The Theory of Signs and ‘Philosophical Superstitions’ . . . . . . . . . . . . . 276
7 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
Jan Salamucha (1903–1944) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

Kordula Świ˛etorzecka
1 Beginnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
2 Education . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
3 Didactics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 280
4 Priesthood . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
5 Death . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281
xii Contents
6 Intellectual Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Selected Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283
Struve and Biegański: Towards Modern Approach to Logic . . . . . . . . . . . 285

Roman Murawski
1 Henryk Struve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
2 Władysław Biegański . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
3 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
Part II Warsaw School of Logic, Its Main Figures and Ideas: The
Period of Prosperity
Łukasiewicz and His Followers in Many-Valued Logic . . . . . . . . . . . . . . 301

Grzegorz Malinowski
1 Łukasiewicz’s Way to Many-Valuedness . . . . . . . . . . . . . . . . . . . . . 301
2 Matrix Semantics for n-Valued Logics . . . . . . . . . . . . . . . . . . . . . . . 306
3 Post Logics and Słupecki’s Functional Completeness Programme . . . . . . . 309
4 Algebraic Interpretations of Łukasiewicz Logics . . . . . . . . . . . . . . . . . 312
5 In Search for Interpretation of Łukasiewicz Logics . . . . . . . . . . . . . . . 315
6 On Applications and Influence of Łukasiewicz Work . . . . . . . . . . . . . . 320
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 326
Tomorrow’s Sea-Battle and the Beginning of Temporal Logic . . . . . . . . . . 329

Kazimierz Trz˛esicki
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 329
2 Language and Semantics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 331
3 Branching Time Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 333
4 Temporal Logic of Possible Worlds . . . . . . . . . . . . . . . . . . . . . . . . 334
5 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
Leśniewski and Mereology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
Peter Simons
1 How Leśniewski Came to Mereology . . . . . . . . . . . . . . . . . . . . . . . 337
2 Leśniewski’s Understanding of Sets or Classes . . . . . . . . . . . . . . . . . . 339
3 Terminology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
4 The Early System . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 341
5 Remarks About This System . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344
6 Symbolic Reconstruction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 345
7 Later Improvements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
8 Whitehead’s Alternative Mereology . . . . . . . . . . . . . . . . . . . . . . . . 349
9 The Polish Continuation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 351
10 The Harvard Variant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 352
11 Subsequent Developments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354
12 Prospects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 358
Contents xiii
Alfred Tarski (1901–1983) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Jan Woleński
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 370
Some Philosophical Aspects of Semantic Theory of Truth . . . . . . . . . . . . . 373

Jan Woleński
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 387
Tarski’s Influence on Computer Science . . . . . . . . . . . . . . . . . . . . . . . 391

Solomon Feferman
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 402
The Absence of Multiple Universes of Discourse in the 1936 Tarski

Consequence-Definition Paper . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
John Corcoran and José Miguel Sagüillo
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 406
2 Corcoran’s Awareness of the Issue in the 1960s and 1970s . . . . . . . . . . . 410
3 The Place of the Monistic-Pluralistic Distinction . . . . . . . . . . . . . . . . . 413
4 The Origin of the Modern form of the Monistic Framework . . . . . . . . . . 417
5 Concluding Remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 419
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
Alfred Tarski: Auxiliary Notes on His Legacy . . . . . . . . . . . . . . . . . . . . 425
Jan Zygmunt
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 425
2 Set Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 432
3 Geometry and Measure Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . 436
4 Decidable and Undecidable Theories . . . . . . . . . . . . . . . . . . . . . . . 439
5 Selected Works of Tarski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 448
Stanisław Jaśkowski: Life and Work . . . . . . . . . . . . . . . . . . . . . . . . . 457

Andrzej Indrzejczak
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 457
2 Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 459
3 Influence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 462
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 463
Stanisław Jaśkowski and Natural Deduction Systems . . . . . . . . . . . . . . . 465

Andrzej Indrzejczak
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 465
2 Natural Deduction in General . . . . . . . . . . . . . . . . . . . . . . . . . . . . 466
3 Jaśkowski’s Research on ND . . . . . . . . . . . . . . . . . . . . . . . . . . . . 468
4 Other Approaches to ND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 479
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 483
xiv Contents
Variations on Jaśkowski’s Discursive Logic . . . . . . . . . . . . . . . . . . . . . 485

Barbara Dunin-K˛eplicz, Alina Powała, and Andrzej Szałas
1 Prelude . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485
2 Theme: Jaśkowski’s Discursive Logic D2 . . . . . . . . . . . . . . . . . . . . 487
3 Movement: Belief Bases and Belief Structures . . . . . . . . . . . . . . . . . . 488
4 Variations Part I: D4 —A New Framework for Discursive Logics . . . . . . . . 491
5 Variations Part II: Relation to Dialogue . . . . . . . . . . . . . . . . . . . . . . 492
6 Variations Part III: Relation to Argumentation . . . . . . . . . . . . . . . . . . 493
7 Coda . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 495
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 496
Czesław Lejewski: Propagator of Lvov-Warsaw Ideas Abroad . . . . . . . . . . 499

Peter Simons
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 499
2 Main Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 500
4 Influence, Teaching, Personality . . . . . . . . . . . . . . . . . . . . . . . . . . 503
Works by Czesław Lejewski Mentioned in the Paper . . . . . . . . . . . . . . . . . 503
Adolf Lindenbaum, Metric Spaces and Decompositions . . . . . . . . . . . . . . 505
Robert Purdy and Jan Zygmunt
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505
2 A Short Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 506
3 Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 520
4 Decomposition of Point Sets, and Their Equivalence by Decomposition . . . 529
5 Decompositions and Equivalence of Polygons in Elementary Geometry . . . . 534
6 Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 537
7 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 547
Andrzej Mostowski: A Biographical Note . . . . . . . . . . . . . . . . . . . . . . 551

Marcin Mostowski
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
2 Main Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 551
3 Main Scientific Achievements . . . . . . . . . . . . . . . . . . . . . . . . . . . 552
4 Influence of His Works . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
5 Students . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
6 Last Words . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 553
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 554
Foundations and Philosophy of Mathematics in Warsaw, the School

of Andrzej Mostowski and Philosophy . . . . . . . . . . . . . . . . . . . . . . . . 555
Marcin Mostowski
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 555
2 Some Mathematical Works in the Foundations . . . . . . . . . . . . . . . . . . 556
3 Old School and Old Ideas . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 562
4 The Old School and New Ideas: Computations . . . . . . . . . . . . . . . . . . 563
5 What for Philosophy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 563
6 Who Is Your Master? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 564
Contents xv
Jerzy Słupecki (1904–1987) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567

Jan Woleński
References and Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 572
Rejection in Łukasiewicz’s and Słupecki’s Sense . . . . . . . . . . . . . . . . . . 575

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 575
2 Reconstruction of Concepts of Rejection and Decidability: The Notions
Introduced and Used by Łukasiewicz . . . . . . . . . . . . . . . . . . . . . . . 578
3 The Problem of Decidability of Aristotle’s Syllogistics, Set by
Łukasiewicz, and Its Solution Given by Słupecki . . . . . . . . . . . . . . . . . 579
4 Notions of Rejection in a Deductive System and Notions
of Ł-Decidability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 585
5 More Important Findings Concerning Ł-Decidability of Deductive
Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 587
6 Rejection Operation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 591
7 Rejection Operation as a Primitive Notion . . . . . . . . . . . . . . . . . . . . 593
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 594
Bolesław Sobociński: The Ace of the Second Generation of the LWS . . . . . . 599
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 600
2 National Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 602
3 Academic Career . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 604
4 Interests and Achievements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 607
5 Sobociński Personality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 609
6 Selected Publications by Bolesław Sobociński . . . . . . . . . . . . . . . . . . 610
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 612
Bolesław Sobociński on Universals . . . . . . . . . . . . . . . . . . . . . . . . . . 615

Kordula Świ˛etorzecka and Marek Porwolik
1 Leśniewski’s Nominalist Argumentation in Sobociński’s Version . . . . . . . 617
2 Sobociński’s Metaconceptualism . . . . . . . . . . . . . . . . . . . . . . . . . . 624
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 631
Many-Valued Logics in the Iberian Peninsula . . . . . . . . . . . . . . . . . . . . 633

Angel Garrido
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 634
2 Many-Valued Logics and the Lvov-Warsaw School . . . . . . . . . . . . . . . 636
3 Reception of Many-valued Logics and Fuzzy Logic in the Iberian
Peninsula . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 641
4 Final Note . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 642
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 643
Ontology of Logic and Mathematics in Lvov-Warsaw School . . . . . . . . . . 645

Roman Murawski
1 Jan Łukasiewicz: (Neo)Platonism . . . . . . . . . . . . . . . . . . . . . . . . . 646
2 Stanisław Leśniewski: Nominalism . . . . . . . . . . . . . . . . . . . . . . . . 648
3 Alfred Tarski: Nominalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 650
xvi Contents
4 Tadeusz Kotarbiński: Reism . . . . . . . . . . . . . . . . . . . . . . . . . . . . 652

5 Leon Chwistek: Nominalism . . . . . . . . . . . . . . . . . . . . . . . . . . . . 655
6 Kazimierz Ajdukiewicz and Ontology of Mathematics . . . . . . . . . . . . . 656
7 Andrzej Mostowski: Nominalism, Reism, Constructivism . . . . . . . . . . . 657
8 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 659
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 660
Part III The War and Post-War Period
A View of Revival of Mathematical Logic in Warsaw, 1945–1975 . . . . . . . . 665

Victor W. Marek
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 665
2 The Beginnings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666
3 Andrzej Mostowski and His Students . . . . . . . . . . . . . . . . . . . . . . . 667
4 Helena Rasiowa and Her Group . . . . . . . . . . . . . . . . . . . . . . . . . . 668
5 Andrzej Grzegorczyk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 669
6 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 671
Andrzej Mostowski and the Notion of a Model . . . . . . . . . . . . . . . . . . . 673

Wilfrid Hodges
1 The Emergence of a New Discipline . . . . . . . . . . . . . . . . . . . . . . . . 673
2 Mostowski’s Writings in Model Theory . . . . . . . . . . . . . . . . . . . . . . 675
3 Background and Notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 676
4 Some of Tarski’s Proposals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 677
5 ‘Theories’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 679
6 ‘Models’ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 681
7 Thirty Years of Foundational Studies . . . . . . . . . . . . . . . . . . . . . . . 684
8 Robinson’s Complaint . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 685
9 Mostowski’s Attributions in General . . . . . . . . . . . . . . . . . . . . . . . . 687
10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 688
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 689
All Quantifiers Versus the Quantifier All . . . . . . . . . . . . . . . . . . . . . . 693

Stanisław Krajewski
1 The Universal Quantifier and Its Dual . . . . . . . . . . . . . . . . . . . . . . . 694
2 A Digression on the Range of the Familiar Quantifiers . . . . . . . . . . . . . 695
3 The Power of ∀ and ∃ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696
4 Generalized Quantifiers in Logic . . . . . . . . . . . . . . . . . . . . . . . . . . 697
5 Characterizing Context-Independent Quantifiers . . . . . . . . . . . . . . . . . 699
6 Formalism-Free Definition of ∀-Definable Quantifiers? . . . . . . . . . . . . . 701
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 701
Helena Rasiowa (1917–1994) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703

Andrzej Jankowski and Andrzej Skowron
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 709
Contents xvii
Post Algebras in the Work of Helena Rasiowa . . . . . . . . . . . . . . . . . . . . 711

Ewa Orłowska
1 Plain Semi-Post Algebras . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 712
2 Post Algebras of Order m . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 715
3 Post Algebras of Order ω+ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716
4 Post Algebras of Order ω + ω∗ . . . . . . . . . . . . . . . . . . . . . . . . . . . 717
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 719
Andrzej Grzegorczyk, a Logician Par Excellence . . . . . . . . . . . . . . . . . . 723

1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 723
2 Logical Accomplishments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 724
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 726
4 Influence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 727
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 728
A Mystery of Grzegorczyk’s Logic of Descriptions . . . . . . . . . . . . . . . . . 731

Joanna Golińska-Pilarek and Taneli Huuskonen
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 732
2 The Logic of Descriptions LD . . . . . . . . . . . . . . . . . . . . . . . . . . . 733
3 Properties of LD . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 738
4 Extensions and Modifications of LD . . . . . . . . . . . . . . . . . . . . . . . . 741
5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 744
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 744
Roman Suszko: Logician and Philosopher . . . . . . . . . . . . . . . . . . . . . . 747

Mieczysław Omyła
1 Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 747
2 Main Papers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 748
3 Views . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 750
4 Influences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 751
5 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 752
From Formal Theory of Knowledge to Non-Fregean Logic . . . . . . . . . . . . 753

Mieczysław Omyła
1 Diachronic Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 754
2 Non-Fregean Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 757
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 762
Categories of First-Order Quantifiers . . . . . . . . . . . . . . . . . . . . . . . . 763

1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
2 Problem of Quantifiers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 764
3 Some Intuitive Foundations of the Theory of Categorial Languages . . . . . . 765
4 The Solution of the Problem of Quantifiers of 1st-Order . . . . . . . . . . . . 769
5 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 775
xviii Contents
The Lvov-Warsaw School: A True Mythology . . . . . . . . . . . . . . . . . . . . 779

Jean-Yves Beziau
1 From Bahía Blanca to Wrocław . . . . . . . . . . . . . . . . . . . . . . . . . . 780
2 The Atopicity of the Lvov-Warsaw School . . . . . . . . . . . . . . . . . . . . 787
3 The Future of the Lvov-Warsaw School . . . . . . . . . . . . . . . . . . . . . . 798
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 812
Part I
Twardowski’s School: The Period
of Crystallization of LWS
Introduction. The School: Its Genesis,
Development and Significance
Abstract The Introduction outlines, in a concise way, the history of the Lvov-Warsaw
School—a most unique Polish school of worldwide renown, which pioneered trends
combining philosophy, logic, mathematics and language. The author accepts that the
beginnings of the School fall on the year 1895, when its founder Kazimierz Twardowski,
a disciple of Franz Brentano, came to Lvov on his mission to organize a scientific
circle. Soon, among the characteristic features of the School was its serious approach
towards philosophical studies and teaching of philosophy, dealing with philosophy and
propagation of it as an intellectual and moral mission, passion for clarity and precision, as
well as exchange of thoughts, and cooperation with representatives of other disciplines.
The genesis is followed by a chronological presentation of the development of the School
in the successive years. The author mentions all the key representatives of the School
(among others, Ajdukiewicz, Leśniewski, Łukasiewicz, Tarski), accompanying the names
with short descriptions of their achievements. The development of the School after
Poland’s regaining independence in 1918 meant part of the members moving from Lvov
to Warsaw, thus providing the other segment to the name—Warsaw School of Logic.
The author dwells longer on the activity of the School during the Interwar period—the
time of its greatest prosperity, which ended along with the outbreak of World War 2.
Attempts made after the War to recreate the spirit of the School are also outlined and the
names of continuators are listed accordingly. The presentation ends with some concluding
remarks on the contribution of the School to contemporary developments in the fields of
philosophy, mathematical logic or computer science in Poland.
Keywords Lvov-Warsaw School · Twardowski · Warsaw School of Logic ·

Łukasiewicz · Leśniewski · Tarski · Lvov-Warsaw School Phenomenon · Polish logic
after the War · Mostowski · Rasiowa · Grzegorczyk · Suszko
Mathematics Subject Classification (2000) Primary 01-02, 03-03
© Springer International Publishing AG, part of Springer Nature 2018 3

Á. Garrido, U. Wybraniec-Skardowska (eds.), The Lvov-Warsaw
School. Past and Present, Studies in Universal Logic,
https://doi.org/10.1007/978-3-319-65430-0_1
4 U. Wybraniec-Skardowska
1 A Short History and Influence of the School1
1.1 Twardowski’s School: The Period of Crystallization of the L-WS
The beginnings of the Lvov-Warsaw School (L-WS by acronym) date back to the end of
the nineteenth century in Lvov, precisely to 1895, when Kazimierz Twardowski (1866–
1938), who, at the age of 29 (in 1895), was appointed professor of philosophy in Lvov,
then an Austrian town (Fig. 1).
Kazimierz Twardowski took the chair of Philosophy at Lvov University.
He came from Vienna, where he belonged to the last group of students of Franz
Brentano. His main aim was to build and organize a strong philosophical circle in Poland,
developing the scientific philosophy and methodology in the Brentano’s spirit. Poland
was at that time partitioned between Austro-Hungary, Germany and Russia; and Lvov
belonged to the Austro-Hungarian Empire. Organizing regular and modern philosophical
studies was not easy. In Lvov at that time there was no relevant library, or respectable
philosophical studies; neither were there any philosophical seminars. Twardowski himself
organized such a seminar and a library, donating his own book collection. His incredibly
serious treatment of philosophical investigations and philosophy teachings consisting in
promoting a clear and critical way of thinking and cultivating scientific, objective truth, his
moral attitude towards philosophy in connection with his amazing charisma as a teacher
and tutor, and friendly relations with students, were the main reasons of his becoming
highly recognized by both students and reputable representatives of other fields of science,
not only Lvov philosophy. Also his scientific-organizational achievements and ventures
which he accomplished played a significant role. All those features and abilities drew
many young people towards philosophy, and their numbers began to grow quickly. Many
Fig. 1 Kazimierz
Twardowski
1 Basedon the entry from Stanford Encyclopedia of Philosophy ‘Lvov-Warsaw School’ elaborated by Jan
Woleński [21], his famous book [20] and his article [22] and also the book edited by Jacek Jadacki and
Jacek Paśniczek [7], the book by Jacek Jadacki [6] and the essay by Urszula Wybraniec-Skardowska [23].
Introduction. The School: Its Genesis, Development and Significance 5
of them became later Twardowski’s devoted disciples. Twardowski promoted many PhD
students, who later earned their professorships not only in Poland, but also in various
countries around the world, cultivating the tradition and the program of the School.
Among the characteristic features of the School was its serious approach towards philo-
sophical studies and teaching of philosophy, dealing with philosophy and propagation of it
as an intellectual and moral mission, passion for clarity and precision, as well as exchange
of thoughts, cooperation with representatives of other disciplines at home and abroad, and
also fruitful collaboration with mathematicians. The L-WS found its own scientific style
of philosophizing and met international standards of training, rigor, professionalism and
specialization. The Lvov-Warsaw School was the first of its kind in Poland. At the same
time, its community managed to establish contacts between its philosophers and world
philosophy.
In this philosophy program, logic—in its broader sense (formal logic, semantics and
methodology of sciences)—was a meaningful element and although Twardowski was not
a logician and was never a devotee of mathematical logic, he did lecture logic as well (see
Fig. 2).
He credited Stanisław Piatkiewicz
˛ with forming the foundations of formal logic (then
called logistics) in Poland, and recognized Lvov as the place of its birth.
The first students of Twardowski were Władysław Witwicki (1878–1948) and Jan
Łukasiewicz (1878–1956). Both earned their Doctorate degrees under his supervision.
Witwicki specialized in psychology. Łukasiewicz was Twardowski’s first student to
Fig. 2 Student’s certificates of Jan Trzciniecki from Lvov University Faculty of Philosophy, lectures in
Psychology of Thinking and Logic, signed by Twardowski (photos from the family archives of Irena
Trzciniecka-Schneider)
be interested in logic. After obtaining his PhD (1902) and the habilitation (1906),
he systematically lectured logic (since 1906). The list of other, trained mainly by
Twardowski, eminent individuals strongly interested in logic, who also attended courses
conducted by Łukasiewicz include (in alphabetical order): Kazimierz Ajdukiewicz (1882–
1963), Tadeusz Czeżowski (1889–1981), Tadeusz Kotarbiński (1886–1981), Zygmunt
Zawirski (1882–1948) and Stanisław Leśniewski (1886–1939), who joined the group
later, in 1910.
Twardowski promoted 30 full professors of different domains of science; he graduated
many more PhD students.
Ajdukiewicz played an important role in logic at Lvov University. He also lectured
for mathematicians (before Leon Chwistek got professorship in mathematical logic).
Ajdukiewicz gave philosophy a clearly logicizing character.
Twardowski and Ajdukiewicz trained a group which included three women: Izydora
Dambska
˛ (1904–1983), Seweryna Rohman (1904–1978), Maria Kokoszyńska (1905–
1981); Henryk Mehlberg (1904–1978) and Zygmunt Schierer (1900?–1943) also were a
part of the group. One of the prominent listeners of Twardowski’s and Ajdukiewicz’s lec-
tures was Eugenia Ginsberg-Blaustein (1905–1944), together with her husband Leopold
Blaustein (1905–1944). They all belonged to the second generation of Twardowski’s
School. So did Janina Hosassion-Lindenbaum (later Mrs. Lindenbaum; 1899–1942), who
was a philosopher, and a member of Kotarbiński’s strong philosophical group.
The L-WS was also joined by a group of catholic philosophers, including Father
Innocenty (Józef Maria) Bocheński (1902–1995) and Father Jan Salamucha (1904–1944).
All the representatives of the L-WS mentioned here are, in this book, classified to
belong to the period of Crystallization (1900–1918). This period was terminated by the
outbreak of the First World War, which interrupted regular scientific activities.
1.2 Warsaw School of Logic, Its Main Figures and Ideas:

The Period of Prosperity
The third generation of the students who later reached a great significance in developing
the teachings and ideas of the L-WS consists essentially of well-known representatives
of the so-called Warsaw School of Logic (by acronym—WSL), acting in the interwar
period (1918–1939). In this book, the period is called the Prosperity period of the L-WS.
It began when in 1916 the University of Warsaw was reactivated and Poland regained its
independence in 1918. Kotarbiński and Wytwicki were offered positions as professors at
the University of Warsaw at the time.
Many scholars from Lvov moved to Warsaw to organize and build national academic
life. Twardowski’s School ceased to be a school of Lvov and became a nationwide school.
During the interwar period, his former students held chairs in philosophy departments at
all of the Polish universities, with the exception of the Catholic University of Lublin.
The main centers of the L-WS since 1918 till 1939 were Lvov and Warsaw. It should
be added that Lvov and Warsaw were also the two centers of the Polish mathematical
school, which acted in parallel, but not independently of the L-WS. The cooperation of
the Lvov-Warsaw philosophers with mathematicians evolved into establishing a logical

school based in Warsaw (WSL).
Lvov was still a strong philosophical center, where under the influence of Twardowski,
Ajdukiewcz and their disciples there was an exchange of ideas by representatives of
various generations of the L-WS. Among the philosophical specialties there prevailed
methodology of science and philosophy of language. Kotarbiński, who moved from Lvov
to Warsaw in 1919, for the professorship, created a strong group of scholars working
mainly in philosophy of science. Kotarbiński educated a whole pleiad of disciples in
Warsaw, among others, Dina Sztejnbarg (later Mrs. Janina Kotarbińska; 1901–1997),
Edward Poznański (1901–1976), and Aleksander Wundheiler (1902–1957) (the last two
of them were the authors of the famous article on the concept of truth in physics).
Warsaw was mainly the center of logic. The development of logic in Warsaw was to a
large extent the merit of the mathematician Zygmunt Janiszewski (1888–1920), belonging
to the first generation of Twardowski’s students, also vitally interested in logic and
foundations of mathematics. He was one of the main initiators of the mutual cooperation
between mathematicians and philosophers of Twardowski’s School.
According to Janiszewski’s program for development of mathematics, mathematical
logic and foundations of mathematics had a great importance and played a special role.
Logic in Lvov did not play such a role as it did in Warsaw. This happened mainly due to
the fact that two philosophers, namely Łukasiewicz and Leśniewski, became professors of
the University of Warsaw at the Faculty of Mathematical and Natural Sciences. They were
put in charge of chairs in the mathematical environment and community at the Faculty
(Łukasiewicz since 1915, Leśniewski since 1919).
The Warsaw School of Logic (WSL) grew out of Twardowski’s School. The WSL had
double roots: philosophical and mathematical. The combination of logic and philosophy
in the history of the WSL did not limit itself to the fact that it was genetically related to
philosophy. There was a close contact between its logicians and philosophers, in particular
with Tadeusz Kotarbiński.
Stanisław Leśniewski and Jan Łukasiewicz, the WSL’s two founders, and their student,
Alfred Tarski, who obtained his doctor’s degree in 1924 with an impressive scientific
output, are the three outstanding representatives of the Warsaw School (Fig. 3):
Fig. 3 Jan Łukasiewicz, Stanisław Leśniewski and Alfred Tarski

Łukasiewicz and Leśniewski were logicians and philosophers graduated at Lvov.

They continued the philosophical thoughts, teaching methods and scientific research
organization of Kazimierz Twardowski, but were protagonists of formal techniques in sci-
ence, similar to those applied in mathematics. And the connection between philosophers
and mathematics, initiated by Janiszewski and his program for mathematical research,
manifested itself, among others, in the fact that Warsaw leading mathematicians (Wacław
Sierpiński, Stefan Mazurkiewicz, Kazimierz Kuratowski) assigned a considerable role
of logic to Jan Łukasiewicz and Stanisław Leśniewski, philosophers by education. Both
began teaching mathematical logic not only in the mathematical environment, but also in
the philosophical one. Logic became an attractive subject to study in Warsaw. And even
though the WSL was a joint creation of philosophers and mathematicians, logic—as a
subject of its research—was not regarded as a part of mathematics or philosophy, but as
an autonomous science.
Well-known representatives of WSL (in alphabetical order) were: Stanisław Jaśkowski
(1906–1965), Adolf Lindenbaum (1904–1941?), Czesław Lejewski (1913–2001), An-
drzej Mostowski (1913–1975), Moses Presburger (1904?–1943), Jerzy Słupecki (1904–
1987), Bolesław Sobociński (1904–1980), and Mordechaj Wajsberg (1902–1942?). Most
were mathematicians, with the exception of Sobociński, who graduated in philosophy,
and Lejewski, who studied classical philology.
In the WSL, unprecedented results were achieved. Heinrich Scholz from Münster said:
Warsaw became the main centre of logical studies.
Thus, in the lifetime of one generation, Polish logic grew from ground level to the
acme of international acclaim. In the well-known book of A. Fraenkel, Y. Bar-Hillel, and
A. Levy [5, p. 200], it is stated that:
Probably no other country, taking into account the size of its population, has contributed so greatly
to the development of mathematical logic and foundations of mathematics as Poland.
and that:
this curious fact should be explained sociologically.
The masters of WSL—Leśniewski, Łukasiewicz and Tarski—were individualists with

different personalities, who shaped their students in many ways. Leśniewski gave the
WSL its theme of synthesis, Łukasiewicz its dynamics, and Tarski the contact with
mathematics (see Woleński [20]). As Jan Woleński writes [20], there were no divisions
in the School into ‘old’ and ‘young’, ‘beginner’ and ‘advanced’. The School had a
strong emphasis on cooperation, irrespective of social position, views, or character of
members. Thus, it combined scientists active in public and academic life. Members also
differed in socio-political and religious views, or their personalities: Łukasiewicz and
Sobociński were conservatives; Lindenbaum and Presburger leaned towards communism.
Some were devout Catholics, others were followers of Judaism, some were atheists.
Except Leśniewski, who was quite reserved, most were persons of a rather friendly nature.
Regardless of their differences, they united around a shared scientific idea, the charisma
of their teachers, awareness of their exceptionality, and their position in the development
of logic in the world.
The problems which the WSL dealt with belonged mostly to mathematical logic.
Still, it all started in philosophy. Both Łukasiewicz and Leśniewski earned their Doctor’s
degrees in philosophy at the University of Lvov: the first one—in 1902, and the other
one—in 1912; both under the supervision of Kazimierz Twardowski. Łukasiewicz started
with methodology of empirical sciences. But his monograph About the Principle of
Contradiction in Aristotle (published in Polish in 1910, see [11]) adds a short lecture
on ‘algebraic logic’. The works by Łukasiewicz, besides those of Jan Śleszyński, are
the first in Poland on mathematical logic. Łukasiewicz never resumed his research into
the methodology of science. Leśniewski’s studies in the time before the First World War
concerned primarily problems of semantics of colloquial language and antinomies.
The philosophical education of the WSL founders strongly influenced their disciples
and their output in mathematical logic, inducing great care for intuitive value. The
representatives of the WSL connected philosophical questions to those of formal logic,
solving classical problems of philosophy by its means. For instance, Łukasiewicz was
convinced that the three-valued logic he had created in 1920 (see [12]) indeed cast new
light on the problem of determinism. And of course, in the famous work [18] (see [19])
on the concept of truth (published in Polish in 1933 and translated into many languages),
Tarski solved one of the fundamental questions of the theory of knowledge in such
an undisputable manner, that probably no other account could claim. As to the self-
image, Leśniewski called himself a ‘philosopher-apostate’, while Łukasiewicz considered
himself a philosopher. Still, Łukasiewicz also did purely formal work, while to Leśniewski
logic was always a tool for philosophical questions.
Another feature of the WSL was a drive for full, precise and simplest solutions
to problems. This ‘perfectionism’ (Łukasiewicz) caused the Warsaw logicians to often
release results with a delay, even at the risk of losing priority. They delighted in formally
perfecting systems, simplifying axioms several times. A peak achievement was reduction
of axioms to only one, as short as possible in terms of symbols. The most surprising
results in this area were achieved by Łukasiewicz and Sobociński, and it is worth recalling
that Łukasiewicz and Leśniewski created two types of original and inventive logical
symbolism.
We will not describe concrete achievements in detail, since there are many sources.
A core subject in the WSL was methodology of the propositional calculus, initiated by
Łukasiewicz. Tarski’s results on the notion of truth are the most outstanding. Famous other
contributions were by Lindenbaum, Wajsberg on intuitionism, or Jaśkowski’s natural
deduction system. Tarski’s subsequent accomplishments extended to the methodology
of all deductive systems. Starting from 1930, he initiated the abstract study of axiomatic
systems, but also a standard semantic viewpoint in his paper on the notion of truth [18, 19].
Tarski led a very busy didactic activity, and his students who obtained outstanding results
already before the World War 2 included Andrzej Mostowski and Wanda Szmielew.
A highly original strand in the Warsaw School was Leśniewski’s creation of the
systems of ‘prototetics’, ‘ontology’ and ‘mereology’ in 1929–1930 (see [8, 9]), trying
to improve on the mathematical foundations of Russell and Whitehead. Especially,
Leśniewski’s ontology has continued to attract interest, for instance from Jerzy Słupecki,
Andrzej Grzegorczyk and Bar-Hillel, and in the early 1970s Boguslaw Iwanuś. But
perhaps his most famous system is the mereology, studied by Tarski and many others,
that still finds applications in geometry, biology, and linguistics today.
The notion of a scientific school is a complex one. Members should address a common
problem with shared methods of investigation. The WSL satisfied this to a high degree.
Also, a true school should have results that create a valuable whole.
What was the lasting contribution of the WSL? The 400-page volume Polish Logic
1920–1939 (edited by S. McCall [16] in 1967), which has translations of 17 articles by
Polish logicians. All, except two, are from the WSL. Also, during that period, the Selected
Works of Łukasiewicz [14] and [15] appeared; the first edited by J. Słupecki, the second
by L. Borkowski. Half of them were papers from the interwar period. Also there was a
monograph by E.C. Luschei on The Logical Systems of Leśniewski [10]. A wide selection
of Tarski’s pre-war articles was Logic, Semantics, Mathematics [19].
A scientific school should also be characterized by personal contacts of its members,
an atmosphere of constant discussion and exchange of thoughts. The following excerpt
from Łukasiewicz’s preface to his Elements of Mathematical Logic [13] illustrates what
the cooperation at Warsaw University looked like:
I owe the most to the scientific atmosphere created at Warsaw University in the field of
mathematical logic. It is in discussions with my colleagues, mainly Professor Leśniewski and
Assistant Professor Tarski, and often also with students of theirs and mine that I had a chance
to comprehend many a notion, absorb new ways of expressing myself, and learning many a new
result whatever their authors were like.
The second period of the L-WS was a phase of splendid creative prosperity: talents
that crystalized during the first period brought results in the form of original scientific
work—ideas, conceptions and scientific systems. It needs to be emphasized that the L-
WS, although known first of all for its achievements in the field of logic, presented a most
pluralistic character with regards to interests and views.
1.3 The War and Post-War Period

1.3.1 Annihilation
1939 is regarded as the last year of the WSL. Twardowski and Leśniewski died before
1939. The invasion of Poland by Germany on 1 September 1939 and by the Soviet Union
on 17 September 1939 began the Second World War. Poland was again divided, under the
German-Soviet pact of 23 August 1939. Warsaw was completely destroyed. Many key
members of the L-WS were forced to leave Warsaw (Fig. 4).
Fig. 4 Warsaw after bombardment
A period of unprecedented persecution and extermination of Polish citizens (especially

Polish Jews) began, as well as pillage on a massive scale. Most people mentioned above
lost their lives: the Lindenbaums, Presburger, Salamucha, Schmierer and Wajsberg. The
Polish Government-in-Exile was constituted.
Numerous representatives of the L-WS mentioned above emigrated from Poland
during the Second World War or shortly after it: Łukasiewicz (Dublin), Tarski (Berkeley),
Lejewski (Manchester), Mehlberg (Toronto, Chicago), Sobociński (Notre Dame), Wun-
deheiler (New York), Bocheński (Fribourge) and Poznański (Jerusalem, before 1939).
Zawirski died in 1947.
On the home front in Poland itself, the resistance movement gradually evolved into an
Underground State. Although teaching was conducted in secret, still scientific and cultural
activity was interrupted.
1.3.2 WW2 Impact
Polish Logic after the Second World War, since 1945, never regained the renown of the
WSL, not so much by losses in human resources as in rhythms of scientific activity,
and losses to libraries and manuscripts. The new academic life, and lack of freedom in
expressing thoughts, did not favor logic either. Lvov found itself outside Poland (it has
been in the Ukraine since the War). Polish country was exposed to the darkest many years
of communist terror. In the new political situation many scholars of the L-WS continued
teaching and working.
Philosophers and logicians from Warsaw, who did not leave Poland, went elsewhere.
For instance, Stanisław Jaśkowski found himself in Toruń (the city of Nicholas Coperni-
cus) in 1945, continuing his scientific and didactic activity in logic, but also in general
mathematics.
Jerzy Słupecki was a typical illustration of how the Warsaw School spread in new
ways. After the War, he found himself in Lublin and then in Wrocław (in 1948). Wrocław
was then a home of displaced people from all over Poland, and expatriates from Lvov.2
Słupecki as his main scientific goal chose to continue, popularize and extend the
studies of the WSL (in particular the outputs of Łukasiewicz, Tarski and Leśniewski).
Słupecki also started working for the Opole Teacher’s Training College, 80 km to the east
of Wrocław, where he founded his own center of logic.
But after the War, a new Warsaw center of mathematical logic and foundations was
started by Andrzej Mostowski, who stayed in Warsaw at the time when the situation of
logic was difficult.
Mostowski kept close contacts with mathematicians like Kazimierz Kuratowski and
Wacław Sierpiński. Continuing Tarski’s work, his center studied set theory, model theory,
decidability, algebraic and topological methods in logic. One persistent topic in his center
has been generalized quantifiers, introduced by him in 1957 [17]. Famous names from
this center are Helena Rasiowa (1917–1994) and Andrzej Grzegorczyk (1922–2014),
who continued—he is deceased the pre-war traditions as a logician, mathematician,
philosopher and ethicist.
Mostowski and Rasiowa educated a numerous group of logicians and mathematicians,
who are classified to a new generation of the L-WS, and who kept up its renown.
Polish logic after the Second World War owed much to Tadeusz Kotarbiński and
Kazimierz Ajdukiewicz, who still taught actively. Kotarbiński’s work initiated the new
discipline—praxeology. Ajdukiewicz’s early ideas opened a way to study questions and
answers in modern logic [1], the categorial grammar [4], and the notion of meaning [2, 3].
It should be added that in 1955 Ajdukiewicz founded the journal Studia Logica; Studia
Logica is now a significant international journal (incidentally, one of many founded by
Polish logicians before the War).
The figures of the Warsaw post-war logic should also include Roman Suszko (1919–
1978)—a logician and philosopher (a disciple of Zawirski and Ajdukiewicz) who,
together with Ryszard Wójcicki, created in the late 1960s an active logical center at Polish
Academy of Science.
2A strong group of logicians and philosophers gathered in Wrocław: mathematicians, philosophers and
logicians. After the War, Wrocław became perhaps one of the leading mathematical and logical centers
in Poland; two famous names are Czesław Ryll-Nardzewski and Jerzy Łoś—prominent logicians and
mathematicians.
1.3.3 Polish Logic Today
Polish logic and philosophy still function with vigor today. We have already mentioned
the strong ongoing tradition of mathematical logic, emanating eventually from the WSL
described in the above. As for formal and philosophical logic, a continued influence of
the L-WS is described by Jacek Jadacki in the book [7] edited by himself and by Jacek
Paśniczek.
It should also be mentioned here that many Polish logicians of the new generation
found a new scope for their talents, namely, in the field of computer science (that did not
exist at all in the pre-War period) and many of them are internationally famous.
2 On the Structure and Contents of the Anthology
The book is divided into three parts related to each other. After Introduction the next three
parts are in accordance with Sects. 1.1–1.3. Thus, the main contents of the anthology are
included into the following parts:
Part I Twardowski’s School: The Period of Crystallization of L-WS,
Part II Warsaw School of Logic, Its Main Figures and Ideas: The Period of Prosperity,
Part III The War and Post-War Period.
Every part consists of biographies of almost all the leaders of the L-WS, mentioned
in particular sections, respectively. They will be followed most often by articles related
to their outputs. All the biographies and articles were elaborated by Polish and foreign
experts on the achievements of the L-WS representatives.
In Part II we omit biographies of Łukasiewicz and Leśniewski—the founders of
Warsaw School of Logic—which are presented in Part I.
Acknowledgements First of all, the editors would like to thank Professors Jacek J. Jadacki, Jan Woleński
and Andrzej Skowron for their kind support and advice, which helped to give the anthology its present
form. We would like to express our utmost gratitude to all the authors of this volume, who belong to a
new generation of L-WS, for their input and unrelenting efforts to accomplish sometimes very demanding
tasks. Their contributions will certainly help to preserve the history of Lvov-Warsaw School for future
generations.
Individual thanks are due to Iwona Bajek for the English-language verification of a number of texts in
this book, and Dr. Zbigniew Bonikowski for the preparation of many works for publication in LaTeX and
the style of Springer.
References
1. Ajdukiewicz, K.: Analiza semantyczna zdania pytajnego [A semantic analysis of a question sentence].
Ruch Filozoficzny X, 194–195 (1926)
2. Ajdukiewicz, K.: O znaczeniu wyrażeń [On meaning of expressions]. In: Ksi˛ega Pamiatkowa ˛
Polskiego Towarzystwa Filozoficznego we Lwowie, Lvov (1931)
3. Ajdukiewicz, K.: Sprache und Sinn. Erkenntnis IV, 100–138 (1934)
4. Ajdukiewicz, K.: Die syntaktische Konnexität. Studia Philosophica 1, 1–27 (1935) (English transla-
tion: Syntactic connection. In: McCall, S. (ed.) Polish Logic in 1920–1939, pp. 202–231. Clarendon
Press, Oxford (1967))
5. Fraenkel, A., Bar-Hillel, Y., Levy, A.: Foundations of Set Theory. North-Holland Publishing,
Amsterdam (1958)
6. Jadacki, J.: Polish Analytical Philosophy. Semper, Warsaw (2009)
7. Jadacki, J., Paśniczek, J. (eds.): The Lvov-Warsaw School — The New Generation. Poznań Studies
in the Philosophy of Sciences and Humanities, vol. 89. Rodopi, Amsterdam/New York (2006)
8. Leśniewski, S.: Grundzüge eines neuen Systems der Grundlagen der Mathematik. Fundam. Math. 14,
1–81 (1929)
9. Leśniewski, S.: Über die Grundlagen der Ontologie. Comptesrendus des séances de la Société des
Sciences et des Lettres de Varsovie, Classe II 23, 111–132 (1930)
10. Luschei E.C.: The Logical Systems of Leśniewski. North-Holland Publishing, Amsterdam (1962)
11. Łukasiewicz, J.: O zasadzie sprzeczności u Arystotelesa. Studium krytyczne [About the Principle of
Contradiction in Aristotle. Critical study]. PAU, Kraków (1910)
12. Łukasiewicz, J.: O logice trójwartościowej [About the three-valued logic]. Ruch Filozoficzny 5, 169–
171 (1920)
13. Łukasiewicz, J.: Elementy logiki matematycznej [Elements of Mathematical Logic]. Nakładem
Komisji Wydawniczej Koła Matematyczno-Fizycznego Słuchaczów Uniwersytetu Warszawskiego,
Warszawa (1929). Presburger, M. (ed.) (2nd edn. PWN, Warszawa (1958))
14. Łukasiewicz, J.: Z zagadnień logiki i filozofii. Pisma wybrane [From Issues of Logic and Philosophy.
Selected Works]. PWN, Warszawa (1961). Selection, introduction and notes by J. Słupecki
15. Łukasiewicz, J.: In: Borkowski, L. (ed.) Selected Works. North-Holland Publishing, Amsterdam
(1970)
16. McCall, S. (ed.): Polish Logic in 1920–1939. Clarendon Press, Oxford (1967)
17. Mostowski, A.: On a generalization of quantifiers. Fundam. Math. XLIV, 12–36 (1957)
18. Tarski, A.: Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych [The Concept of Truth in Languages of
Deductive Sciences]. Towarzystwo Naukowe Warszawskie, Warszawa (1933) (German version: Der
Warhheitsbegriff in den formalisierten Sprachen. Studia Philosophica I, 261–405 (1936); English
translation: The Concept of Truth in Formalized Languages. In: Tarski, A. Logic, Semantics,
Metamathematics: Papers from 1926 to 1938, pp. 151–278. Clarendon Press, Oxford (1956))
19. Tarski, A.: Logic, Semantics, Metamathematics: Papers from 1926 to 1938. Clarendon Press, Oxford
(1956). Trans. J.H. Woodger (2nd edn edited and introduced by John Corcoran, Hackett Publishing
Company, Indianapolis (1983))
20. Woleński, J.: Logic and Philosophy in the Lvov-Warsaw School. Kluwer Academic Publishers,
Dordrecht (1989)
21. Woleński, J.: Lvov-Warsaw School. In: Zalta, E.N. (ed.) The Stanford Encyclopedia of Philosophy
(Winter 2015 edn.). https://plato.stanford.edu/archives/win2015/entries/lvov-warsaw/
22. Woleński, J.: Logic and the foundations of mathematics in Lvov. Banach Cent. Publ. 87(1), 27–44
(2009)
23. Wybraniec-Skardowska, U.: Polish Logic, Some Lines from a Personal Perspective. Publications of
Institute for Logic, Language and Computation, Amsterdam (2009)
U. Wybraniec-Skardowska ()
Cardinal Stefan Wyszyński University in Warsaw, Warsaw, Poland
e-mail: [email protected]
Kazimierz Twardowski: A Great Teacher
of Great Philosophers
Anna Brożek
Who really is a patriot, is not hungry for fame.

His inner conviction is enough for him,
that he makes what he should make,
and none of his effort will be wasted;
his sweat and — when it is needed — his blood
produces the most beautiful and most wonderful seeds:
the greatness and happiness of the Fatherland.
On patriotism (Twardowski [7, p. 454])
Abstract Kazimierz Twardowski was the founder of the Lvov-Warsaw School, one of
the greatest phenomena in the European culture. The School had its representatives in all
scientific disciplines, logic and mathematics including. Among his pupils are such great
figures of European philosophy like: Jan Łukasiewicz, Stanisław Leśniewski, Tadeusz
Kotarbiński, Władysław Tatarkiewicz and Kazimierz Ajdukiewicz. The paper presents
his life and various fields of his rich activity, as well as the list of his main works and
the greatest achievements in philosophy. He is best known in the world as the author of
the distinction presentation-object-content, act-product, as well as the supporter of the
postulate of clarity of thought and speech and the critic of psychologism, relativity of
truth, symbolomania and pragmatophobia in science.
Keywords Act-product · Clarity · Kazimierz Twardowski · Lvov-Warsaw School ·

Pragmatophobia · Presentation · Psychologism · Relativity · Symbolomania
Mathematics Subject Classification (2000) Primary 01A70, 03A05; Secondary 03B65
1 Life
He was born on October 20, 1886, in Vienna; he died on February 11, 1938 in Lvov.
After several years of education at home and in public school, Twardowski, through
the efforts of his father, obtained a place in a prestigious high school in Vienna—
Theresianum. This school provided him with a comprehensive classical education;

16 A. Brożek
in Theresianum, he was also familiar with philosophical problems. Thanks the iron
discipline prevailing in the school, Twardowski got accustomed to systematic, persistent
work—and thanks to the example instilled in him at home, from an early age he wanted
to serve Poland and the Poles.
After graduating from Theresianum Twardowski began his studies at the University
of Vienna, initially at the Faculty of Law. For some time he lived at the property of Earl
Wojciech Dzieduszycki in Jezupol where he was teaching the children of the earl. Perhaps
partly due to Dzieduszycki—a philosopher, a politician and a writer—with whom he had
cordial relations, Twardowski moved to the Philosophical Faculty. Studying philosophy,
he was influenced to the greatest degree by Franz Brentano—who became for Twardowski
the model of a teacher and a researcher.
After graduating Twardowski performed annual military service. The doctoral degree
he earned in 1891 was based on a dissertation on the views of Descartes. The official
promoter of this work was Robert Zimmermann (at this time, Brentano could not officially
be a promoter of any doctorate).
In 1892, Twardowski married Kazimiera Kołodziejska, with whom he had three
daughters: Helena, Aniela and Maria. All were well-educated, and the youngest of them
married Kazimierz Ajdukiewicz, one of Twardowski’s closest disciples.
Thanks to a scholarship from the Austrian Government, in 1892 Twardowski traveled
to Leipzig and Munich. After returning to Vienna, he took a job as an official, so that he
could earn money as a tutor while preparing his habilitation. He received it in 1894 for the
thesis Zur Lehre und Gegenstand der Inhalt der Vorstellung. This dissertation was, until
recently, the most famous of Twardowski’s works, since it was published in German. After
obtaining habilitation, in the academic year 1894/1895, Twardowski taught in Vienna as
a Privatdozent.
As early as the next year, thanks to a happy coincidence, as the only 29-year-
old scholar, Twardowski took over the chair of philosophy at the University of Lvov.
Twardowski quickly proved to be an equally charismatic teacher, like his Viennese master.
Twardowski remained faithful to the University of Lvov, working there until retirement,
i.e. until 1930, and after that he continued to take part in its life.
2 Personality
He was one of the most important and most versatile characters of Polish culture of the
first half of the twentieth century.
As a scholar, he was the author of excellent work from all the areas of philosophy—
from logic, through ontology and epistemology, to aesthetics and ethics—and also works
on its history, on psychology and on didactics; he also composed songs full of expression,
most of which were unfortunately lost during the turmoil of the wars.
As a teacher, he was a master of several generations of Polish philosophers and, more
broadly, humanists.
As a citizen, he set the tone of the teaching environment, presiding, in the first decade
of the twentieth century, over the influential Society of Teachers of Higher Education and
leading an extensive lifelong correspondence with many opinion-forming representatives
Kazimierz Twardowski: A Great Teacher . . . 17
of the country. Thanks to cautious, but firm and persistent action, he led, among other
things, to fully repolonizing education in the Austrian partition of Poland.
For the elementary component of citizenship he felt patriotism. The latter—understood
as an actually active attitude—Twardowski distinguished from ceremonial patriotism. A
ceremonial patriot has a sentimental attitude to his country and nation, gladly manifesting
this attitude on certain occasions. However, he can not afford to sacrifice his own good
for the good of his countrymen.
3 Writings
The most important of Twardowski’s book publications published in his lifetime—are:

Zur Lehre vom Inhalt und Gegenstand der Vorstellungen [1], Imageries and concepts [3],
The essential concepts of didactics and logic [4], Six lectures on medieval philosophy [5],
Addresses and papers [6] and Philosophical dissertations and papers [9].
After his death the following appeared: Selected philosophical writings [10], Collected
psychological and pedagogical writings [11], Ethics [12], Diaries. Vol. I–II [13],
Philosophy and music [15], Youthful diaries [16] and Thought, speech and action. Vol.
I–II [17].
In English, the following of Twardowski’s books were published, i.a.: On the content
and object of presentations [1], On actions, products and other topics in philosophy [14]
and On prejudices, judgments and other topics in philosophy [18].
4 Views
It is often said that the greatness of Twardowski lies exclusively in his organizational and
pedagogical activities: in particular, that he was the founder of the Lvov-Warsaw School.
Such an assessment is given even by many representatives of the School itself.
However, it should be kept in mind that only some of the philosophical views of
Twardowski were expressed in print during his lifetime. In this case, Twardowski was
like his master—Franz Brentano.
This involved, among others, the fact that Twardowski not only preached, but also
realized, to the highest degree, the postulate of accuracy and precision. Anyone who
obeys this postulate—knows how difficult it is to give final form to our own thoughts.
Twardowski severely judged those who lacked awareness of this situation:
How many authors publish philosophical papers, essays, and even extensive volumes, which
directly swarming with equivocations, incomprehensibilities, thoughts, hopes, inconsistencies,
vague expressions and paralogisms, and who, together with their publishers, loses nothing,
provoking sound and often admiration in wide reading circles! I could envy them the ease of
their literary production; but, in fact, I underestimate the products of their pens and I resent
their behavior which is highly detrimental to the logical development of philosophical thinking
[8, p. 32].
Meanwhile, an analysis of even only the printed legacy of Twardowski shows that he
has won important theoretical results in all disciplines of philosophy. Here we have an
outline of these results.
18 A. Brożek
4.1 Metaphilosophy
4.1.1 Philosophy and Worldview
Twardowski regarded philosophy as a conglomerate of disciplines, which consists of

logic, psychology, metaphysics, ethics, aesthetics and the history of philosophy. He
pointed also to a particular trait of all philosophical disciplines: the objects of philosoph-
ical research are given in internal experience (introspection).
The word “philosophy” is sometimes used as a synonym for the word “worldview”.
Twardowski’s postulate was separating philosophical problems, cultivated with scientific
methods, from the problems of a worldview, whose field is one of personal beliefs; giving
the latter the form of a theory which he considered pure speculation.
In Twardowski’s relation to faith—we need to distinguish its relationship to the
Catholic dogma, to Catholic ethics and Catholic ceremony. As for the first and second
case, Twardowski wanted, in Christianity, a dominance of the ethical dimension over—let
us call it so—the dogmatic.
4.1.2 Brightness of Language
One of the main causes of errors in philosophy, he recognized, is the lack of clarity of
thought, coupled closely with the lack of clarity of speech; the lack of clarity of speech can
at the same time be regarded as a symptom of the lack of clarity of thought, as discursive
thinking can not be extralinguistic.
Accordingly, philosophical terminology required reconstruction. In this reconstruction,
he recommended using the analytical method (which was also used by him). Such a
reconstruction is carried out in two stages. The first stage is to examine the essential
properties of several typical designates of a given term and to formulate its analytical
definition. In the second stage this definition is adjusted by checking whether the
analytical theses implicated by this definition are true for other designata of the defined
term.
Twardowski advocated the careful use of formal logic in the analysis of philosophical
problems: he was the enemy of (as he expressed it) symbolomania, i.e. abusing logical
symbolism, and of pragmatophobia, related to the first, i.e. focusing our attention on
the syntactic aspects of this symbolism. For logical structures—including those that are
formally flawless—are not always based on well defined intuitions, and these structure do
not often have any model in the field of philosophical studies.
4.1.3 Independence of Thought
Twardowski was a firm defender of the independence of thought.

In his opinion, our beliefs are often dependent on: those deemed by us to be authorities;
the environment in which we live; the language we use; our mental organization and
finally, our emotions.
To the question of whether we can think independently, Twardowski responds in a

complex way. There are things or domains, in which we all are doomed to be—more or
less—dependent in thinking; however, some of us, in some cases, can think independently.
Independence of thought is therefore an ideal which—according to Twardowski—we
ought to and, which is important, we are able to pursue successfully.
4.1.4 Anti-psychologism
Twardowski regarded logic and psychology to be propaedeutic disciplines of philosophy.

Earlier, however, he had rejected the psychologism prevailing at the time of his youth, i.e.
the view, according to which, logic is a part of psychology or should rely upon it.
Psychologism was, according to Twardowski, untenable for the following reasons: (1)
logic was created and developed independently of psychology; (2) theses in psychology
(which are generalizations of experiential data) are only probable, whereas theses in logic
are reliable (and, consequently, unquestionable), and being reliable, they can not follow
psychological theses; (3) psychology is the theory of the actual activity of thinking—
whereas logic deals with evaluating the typical forms of the products of thinking (i.e.
thoughts) in terms of their accuracy.
4.2 Metaphysics
Twardowski never disavowed metaphysics as such, but he had a very rigorous view of the
conditions that must be met by philosophical research, if their results could be included
in scientific knowledge. However, he did not share the (positivist) belief that since the
object of study of metaphysics—and wider, in philosophy—is neither the world of sensual
phenomena, nor the world of mental phenomena, it is not science. The idea is that outside
these phenomena there are still objects of another type (e.g. relationships)—and they are
examined in, among other disciplines, metaphysics in a manner adopted in other sciences,
in particular, by describing these objects and using inductive and deductive reasoning to
define and justify statements about these objects, just as in the natural sciences.
At the same time, Twardowski was a supporter of the cumulative conception of
philosophy. The long-term goal of philosophy is to construct a synthesis of the scientific
theories which would be the theory of all objects (and not only a theory of sensual or
mental phenomena); such a synthesis has still not been achieved, but its elements already
exist and we should patiently seek others. For, according to Twardowski, it is better to
accept no synthesis at all—than to accept a fallacious one.
We owe to Twardowski a developed theory of objects, formulated, by the way, before

Alexius Meinong.
According to Twardowski, each object—regardless of its existential and metaphysical
category—is a homogeneous whole, composed of various properties. The components
of objects are whatever can be distinguished in these objects: a particular component is
something that can be distinguished in fact; a component of the abstract is something
20 A. Brożek
that can be distinguished only intentionally. All properties of objects and all relationships
between the parts of these objects are abstract components. For example, the stem and
thorns of a certain rose are its concrete components, whereas its redness and its being-a-
plant are its abstract components.
4.2.2 Categorization of Objects
As one of their tasks, metaphysicians always sought to make a categorization of objects.

However, few of them realized this goal in a logically satisfactory way, such that these
categorizations took the form of logically correct classifications; fortunately Twardowski
belonged to this smaller group.
He carried out logically correct classifications of objects based on different existential
aspects and different ontic aspects. In the first case, he found three dichotomous classifi-
cations: into possible and impossible objects, into existing and non-existing objects, and
into real and unreal objects. In the second case: into individual and general objects, into
simple and complex objects, as well as into corporal and spiritual objects. This made it
possible to recognize the traditional categorizations as the result of not fully intentionally
crossing the classifications made by Twardowski.
4.2.3 Actions and Products
An important step on the way to Twardowski’s rejecting psychologism, was his analysis
of actions and the products of actions.
Actions are states of a special type. They are associated with certain specific objects,
which he called “products”: and so a picture is the product of painting (or drawing), an
inscription is the product of writing, a thought is the product of thinking, etc. Among
the artifacts, there are relatively unstable products, which can be distinguished from the
corresponding actions only mentally (scil. by abstraction), and relatively stable products.
A jump is an unstable product of jumping and a dance is an unstable product of dancing.
A picture as a product of painting, a hole as a product of digging—are relatively stable
products of their relevant actions.
Products of physical actions—i.e. physical products—are either unstable (e.g. a cry as
a product of crying, a jump as a product of jumping, a rotation as a product of rotating)
or stable (e.g. a print as a product of printing, a braid as a product of braiding). Stable
products of physical actions exist more than the actions by which they arise. However,
all the products of mental actions—i.e. mental products—are unstable (e.g. a thought as a
product of thinking, an experience as a product of experiencing, a decision as the product
of deciding).
Some actions are directed at certain objects. Objects—especially things—at which
physical actions are directed, create the material (scil. basis) of these actions (e.g. sand,
which reflects the rate of person walking along this sand).
The product of a mental action directed at a certain material is not a material itself,
but is a new structure of this material (created by this action). The object, at which a
certain mental action is directed, is the object of this action (e.g. a landscape imagined by
someone). Actions that are directed at something—are intentional actions.
Certain properties of products are not properties of the actions that create these
products. For example—it happens that our dream is false, but action of dreaming itself
is not false; a question may be confusing, but it cannot be the confusing action of asking
a question.
Twardowski made a distinction of actions and the contents of presentations in terms
of actions and products. He considered stable products to be psychophysical objects; he
identified cultural artifacts just with such objects.
4.2.4 The Soul
Twardowski proposed an original approach to the problem of the ontological status of the
soul.1
According to Twardowski, the thesis about the existence of the soul—as a substratum
of experiences—is an obvious thesis and as such it does not need any proof. An opponent
to this thesis—an asubstantialist—could not use the pronoun “I” in the ordinary sense
of this pronoun. The so-called group of mental phenomena, which is identified by
asubstantialists with “I”, would not be able to determine that a given experience belongs
just to it (a similar difficulty would arise with any internal perception); for a similar reason,
this group would not say of itself that it knows something.
Twardowski justified the thesis of the simplicity of the soul in such a way. If a
substratum of experiences consisted of parts, and some two experiences (e.g. a visual and
an auditory impression) were located in two different parts of the substratum, it would not
be possible to compare them; meanwhile, such a comparison is a fact. Since the soul is
simple (scil. it is a mental atom), it is eternal, because its possible creation or annihilation
cannot be naturally explained. It remains to be assumed that the natural eternity of souls-
atoms goes hand in hand with the fact that they are created by God (creating any objects
by an eternal God does not require that that they occurred later in time than God—and
thus were non-eternal) and at one moment (e.g. at the moment of the birth of a human
being) are revealed in the temporal-spatial world.
The thesis of the simplicity of the soul is one of the premises of reasoning justifying the
thesis of the immortality of the soul. The second premise is the principle of conservation
of energy—taken from the natural sciences. While Twardowski was aware of certain
imperfections of his reasoning, he considered it to be conclusive.
4.3 Psychology and Epistemology
Standing on the position of anti-psychologism—Twardowski devoted much of his

attention to psychological and meta-psychological considerations, with respect to con-
siderations bordering on psychology, epistemology and methodology. Until the end of his
life, he believed that psychology is the basic philosophical discipline due to the fact that
it is the theory of (real) thinking.
1 Twardowski’s views concerning this subject are little known. Only recently was his extensive German
manuscript from 1895 published. Cf. [1].
22 A. Brożek
Twardowski practiced empirical psychology, i.e. psychology that justifies its state-
ments on the basis of experience—unlike, for example, political history, whose tested
facts (as past) need to be reconstructed on the basis of indirect data, i.e. testimonies, and
«natural» history, when investigating the history of certain fragments of nature. At the
same time, empirical psychology appeals to external experience (scil. extraspection) and
internal experience (scil. introspection), introspection being its ultimate foundation. So
psychology may be considered a quasi-historical science: lying on the border of empirical
and historical disciplines—wherein some fields of psychology are closer to the first (as
e.g. examining the psyche of healthy people), and some to the others (as e.g. psychiatry).
4.3.1 Components of Consciousness
Experiential components of consciousness—mental facts—are spiritual actions (which

can be only mentally separated from each other) and their products. The source of
knowledge concerning psychic facts cannot be located in the senses, but rather self-
consciousness. In such a direct way, only states of our own consciousness are knowable.
The basic kind of spiritual action is presenting something for the substratum of
these actions. It is a necessary condition for all other—secondary—types of spiritual
actions, in particular: judging, reasoning (or “pondering, hesitating, comparing, devising,
synthesizing, distinguishing etc.”), feeling and deciding. On the other hand, judging is a
necessary condition of feeling—in any case convictional feeling—and deciding.
Twardowski thought, at the same time, that presenting and judging are kinds of
thinking; this is suggested, among other things, by the fact, that there are two contexts of
the word “to think”, namely: “to think of X”, and “to think that p”. To think of X—is the
same as—to present X (scil. to imagine or comprehend), and to think that p—is the same
as—to judge that p. One of the kinds of thinking is also reasoning, which—according to
Twardowski—is judging about judgments.
Both primary and secondary spiritual actions are intentional actions: they require an
object.
The spiritual conditions of spiritual actions are dispositions. Dispositions are only
hypothetical components of consciousness. According to the law of practice, it is assumed
that the occurrence of any mental fact leaves a disposition to reproduce it.
4.3.2 Acts, Contents and Objects of Presentations
The legitimacy, which Twardowski permanently secured in the history of European

philosophy, was in carrying a precise delimitation of the three aspects of presentation:
act, content (i.e.—as he put it—the product of action) and object. Twardowski accepted
this distinction throughout all his scientific activity, but he made some modifications to
the original conception of this distinction.
Act and content of presentation are the metaphysical (scil. abstract) parts of presen-
tation. In languages, presentations correspond to names; the content of presentations
corresponds to the meaning of names, and the object of presentations—to what names
designate.
The object of a given presentation is different from the content of this presentation,
because: (a) when we perform a negative judgment of the type “A-being-B does not
exist”, then we deny the existence of the object A-being-B given in presentation, which
forms the basis of the performed judgment, but, at the same time, we realize that the
content of this presentation contains something (namely B), which means that the object
of this presentation does not exist in reality; (b) there are various presentations of the same
object (i.e. we can imagine the same A, in one time, as P, and in another time, as Q). The
expression “presented object” is ambiguous; it may refer (in the determining meaning)
to a real object, which is presented by someone—or (in the modifying meaning) to the
mental presentation of this real object.
Twardowski divided presentations into images (scil. concrete, pictorial presentations)
and concepts (scil. abstract, non-pictorial presentations). At the same time, he distin-
guished perceptive images—from reconstructive (memory-based) and productive (based
on imagination) images. The difference between them is qualitative rather than only
quantitative. Perceptive presentations—unlike reconstructive and productive ones—are
more vivid and independent of our will.
Twardowski defined concepts as presentations consisting of a vague underlying image
and presented judgments which assign to the object of this image properties constituting
the connotation of the expression with which this concept is linked.
By virtue of their object—concepts can be divided into synthetic and analytic. Objects
of synthetic concepts are things or persons (e.g. the concept of a tree, of a kingdom, of
God). The underlying image of a synthetic concept is the (reconstructive or productive)
image of an object similar to the one that has to be conceptualized. One of the components
creating the content of a given concept is an image of the statement stating that the object
of the underlying image has properties, which de facto are not possessed by this object.
Both types of concepts—i.e. synthetic and analytic—have two forms: proper and
shortened, scil. hemisymbolic or symbolic. In the hemisymbolic form, the act of
conceptualizing contains the image of the name designating the object of this concept and
the accurate underlying image. In the symbolic form, there is only the first component
(scil. the image of a name).
4.3.3 Judgments
The starting point in the theory of judgments was to Twardowski the idiogenic conception
of judgment. According to this conception, judging is recognizing or rejecting the
existence of a certain object.2 Just as within the domain of presentations, Twardowski
distinguished, in the case of judgments, content and the object of them. The object of a
judgment—is what we recognize as existing or reject as non-existing. The content of a
2 InTwardowski, we find a characteristic hesitation, which will be also visible in many of his disciples:
either a logical value should be considered a defining property of “judgment”, or “judgment” should
be defined without using the notion of truth with respect to falsehood. Here we are dealing with the
second solution. But sometimes Twardowski called “judgment” a mental action, “in which truth or
falsehood is contained”. This involves “truth” and “falsehood” in the proper sense. The external criterion
of truthfulness understood in such a manner is conformity with the truth, and of falsehood—fallacy.
The internal criterion is whether “true” and “false” are adjectives determining nouns (and thus they are
determinators), where they stand—or adjectives which retain the changed sense of those nouns (and
thus—in the latter case—they are modifiers).
24 A. Brożek
positive judgment is the existence of the object, the content of an appropriate negative
judgment is the non-existence of the object. Each judgment is a superstructure on an
underlying presentation. The underlying presentations of judgments have an existential
paraphrase as in the judgments “There is a city where Kazimierz Twardowski was born”
and “There is a city where the peace conference was held after the Napoleonic wars”,
which have the same object: the city of Vienna. These judgments are different because
they are different as to the content of their underlying presentation: in the first case,
Vienna is presented as a city which is the birthplace of Kazimierz Twardowski; in the
second case—as a city where the peace conference was held after the Napoleonic wars.
Twardowski noticed, however, that the traditional idiogenic theory encounters some
difficulties—especially, e.g., in the case of judgments on past objects as well as in the
case of judgments about relationships.
The problem of judgments about a future object was reconstructed by Twardowski
as follows. Consider the judgment “There was a king”. Brentano would interpret this
judgment as acceptance of the existence of a past king. However, if “to be” means the
same as “to be now”, and “past king” designates such a king who was formerly and now
is gone—then we are in contradiction: because we receive the judgment “There exists
something that once existed, but no longer exists”.
Twardowski sketched therefore a modification of the theory judgment, avoiding the
identified difficulties. He contrasts in it two types of judgments: the existential and
the relational, the term “existential judgment” coming from the content, and the term
“relational judgment” coming from the object of presentation. The content of existential
judgment is namely existence, and the object of relational judgment—wherein the term
“relationship” was used by Twardowski in the meaning similar to that, in which the term
“state of affairs” is used today. In relational judgment, a certain relationship is accepted
or rejected, and the content of such a judgment is subsistence (Germ. vorhandensein
Bestehen). In this case, the overall form of judgment is: ±bA, where ‘±’ means the
acceptance or the rejection, ‘b’—subsistence, and ‘A’—an appropriate relationship. The
characteristic fact here is the presence of the difference between the judgments: “God
exists” and “There was a king”. While the content of the first judgment is the existence
of God, the content of the second is the subsistence of a relationship of possession taking
place between a certain king and a preceding period (or otherwise: the subsistence of
a relationship of possession occurring between a certain king and the property of being
contemporary with some past period).
4.3.4 Truth and Error
Twardowski gave convincing arguments in favor of alethic absolutism, showing that

relativists, supposing that the logical value of judgments may vary in time and space, have
in mind de facto not judgments, but rather their imperfect language expressions which are
rarely reliable sentences. For example, maintaining that the judgment that it is raining
is relatively true, since it is true only in certain circumstances—is a misunderstanding.
When we say “It is raining”—we express the judgment that it is raining at such and not
another place and at such and not another time, and the judgment that it is raining in these
and not other circumstances is absolutely true or absolutely false.
Twardowski devoted a lot of attention to the analysis of prejudices.
Prejudice, according to Twardowski, is a false belief, asserting the occurrence of a

certain regularity—a belief which is supported by someone without sufficient justification.
Twardowski distinguished two particular kinds of prejudices: superstitions and survivals.
A superstition is a prejudice concerning the accuracy occurring between natural and
supranatural phenomena. In contrast, survival is a prejudice which was once common
but its detailing is not a prejudice.
4.4 Semiotics
4.4.1 Functions of Speech
According to Twardowski—language is a double instrument by the fact that particular

language signs perform a dual representative function (scil. functions of expressing) in the
face of spiritual actions (primarily, in the face of emotional-volitional actions; secondarily,
in the face of intellectual actions): objective and subjective functions.
Objective functions—significative and denotative ones—rely in turn on the fact that a
language sign signifies a product of represented spiritual actions and designates the object
of these actions. Subjective functions—indicative and evocative ones—rely in turn on the
fact that a language sign indicates an action on the part of the sender of this sign and
inspires action in its recipient; the matter here is about actions, the products of which are
meant by this sign. It is thanks to the indicative function that the language signs (of natural
speech) may indicate, among other things, emotional shades (scil. emotional moments),
marked not only in the choice of vocabulary (cf. e.g. a pair of words: “nag”—“horse”, “to
give up the ghost”—“to die”,3 the first positions of which are occupied by the words of
a specific color), but also in the manner of pronouncing words. On the other hand, it is
thanks to the evocative function, that language signs can be understand at all.
Twardowski notes that language signs are usually ambiguous. The same kind of sign
can have several logical meanings (e.g.: “my portrait” may mean the same as “a portrait
done by myself”, “a portrait, which belongs to me” or “portrait, which presents me”).
Speech features also multinominality. One type of thinking corresponds sometimes with
many kinds of words of speech.
Ambiguity and multinominality speak against a parallelism between speaking and
thinking; and the fact that we can speak without thinking and we can think without
speaking (also continuously—as it is, for example, in the case of dumb men) speaks
against the identity of speaking and thinking. But there is a strong correlation between
speaking and thinking. Firstly, signs of speech—as symbols of corresponding thoughts—
remain in a causal relationship with them. Secondly, in thinking—especially more abstract
thinking—there are numerous «shortened» concepts (scil. hemisymbolic and symbolic
concepts, as mentioned above). Thirdly, the relationship between thinking and speaking
means that people thinking vaguely cannot clearly express their thoughts.
The set of language signs has two subsets: sentences and parts of sentences. The
latter can in turn be presenting or cathegorematic signs, or connecting or syncathe-
3 In this paper, all examples come from Twardowski himself.

26 A. Brożek
gorematic signs. Cathegorematic signs perform representative functions in relation to

presentations—they mean something; syncathegorematic signs are only co-meaning.
Typical cathegoremata are names; typical syncathegoremata are connectors.
4.4.2 Names
Of the grammatical parts of speech, names tend to generally be proper nouns (e.g.:
“Lvov”, “Stanislaus Augustus”) and common nouns (e.g.: “father”, “soul”), pronouns
(e.g.: “he”, “we”, “something”), adjectives (e.g.: “white”, “gloomy”), numerals (e.g.:
“two,” “tenth”, “fifty seven and a half”) and verbs (e.g.: “to run”, “to learn”, “to expire”)—
namely those which are grammatical subjects (except sentences deprived of a predicate),
predicates (except sentences deprived of a subject), predicatives or complements of simple
sentences. Names may be not only single nouns, but also whole noun phrases, and
therefore fusions of nouns with other expressions (e.g.: “human eye”, “this man”, “a
man”, “the second son”, “the highest mountain in Europe”, “the son who insulted his
father”).
According to Twardowski—as there are no objectless presentations, so too are there
no empty names: each name designates a certain entity—and exactly one entity (even a
complex name). On the other hand, the fact that a given object is designated does not
imply the existence of this object.
4.4.3 Declaratives
Among sentences, Twardowski distinguished not only expressions representing judg-

ments, but also those representing orders, wishes, requests, questions or curses.
He called sentences representing judgments—“declaratives”. A declarative means the
frame of a given judgment (scil. existence); it designates the object of this judgment; it
indicates a judging performed by the speaker and raises performing—or only presenting
(and thus understanding)—the judgment by the listener. Moreover, not every declarative
can be described as a “sentence” in the grammatical sense (cf. e.g. the word “Fire!”,
representing sometimes the judgment that it is burning).
A special type of declaratives are definitional formulas of the structure “A is (identical
with) B” (e.g. “A state is a public community which includes a sedentary population within
a given territory as a community of rulers and the ruled”). These declaratives do not de
facto represent the judgments that A is (identical with) B (especially in the example, given
above, we claim nothing about a state), but rather judgments, that ‘A’ designates B (in
particular we claims in this definition that by the word “state” we designate a certain
object, whose representation has been constructed in this way).
Truthfulness, as well as evidence and probability—and their inverses—could be
assigned to declaratives only indirectly: due to judgments represented by them. Absolute
truthfulness is namely possessed by a declarative representing only true judgments. Such
declaratives should be complete ones. About a declarative which sometimes represents
true judgment and sometimes wrong, we can say that it is relatively true. Among such
declaratives, there are elliptical declaratives (e.g.: “It is raining”, “A cold bath is a healthy
thing”, “Apartments in Lvov are expensive”). Similarly, if talking about synonymy in
relation to judgments is ridiculous, so much is it so in relation to declaratives. Namely,

two declaratives are synonymous when they represent the same judgment.
4.4.4 Connectors
According to Twardowski, among lato sensu connectors, a special place is occupied by

determinators, i.e. nouns, pronouns, adjectives, adverbs and numerals—and the phrases
equivalent to them—which occur in attributive contexts (e.g.: “speed OF MOTION”, the
“father OF SOCRATES”, “the gable wall OF A HOUSE”; “THIS man”, “A man”, “NO man”).
In predicative contexts, they are cathegoremata.
A determinator—together with the name for which it stands—co-represents the
presentation of a judgment relating to the object designated by the determined expression.
Depending on the subject of this judgment, an ascriptive function is performed by a
determinator in relation to a given determined expression, consisting in determination,
abolution, confirmation or modification (i.e. both abolution and determination). The
determining and abolutioning function consists in changing the meaning and, in particular,
either in its enrichment (cf. “good man”), or its depletion (cf. “alleged shape”). The
confirming function consists in strengthening or restoring some of its components (cf.
“actual fact”). The determining, abolutioning and confirmating functions are simple
ones. On the other hand, the modifying function is complex, consisting in simultaneous
enrichment and depletion of meaning (cf. “the former minister”).
Stricto sensu connectors contain conjunctions which may occur either in a nominal
context (e.g. “non-”) or in a propositional context (e.g. “no”, “or”, “if”).
The conjunction “non-” in the nominal context (e.g. “non-Greek”) performs a specific
modifying function in relation to the meaning of its nominal element: the infinitive
function namely. It changes the meaning of its argument so that the meaning of the
whole context becomes identical with the meaning of a generic name (genus proximum)
superior in relation to this element (here: the meaning of the name “man”), enriched with
a presentation of the judgment refusing to the object of this name—the specific features
of the object of the negated name (here: the name “Greek”). With all this, the rule of
infinitation is in force, allowing the addition of “non” only to such names which are
subordinate a certain name. This rule is violated by the expression “non-entity”, because
for the word “entity” we can not establish a genus. Thus “non-entity” is a nonsense term.
Therefore, if we considered the noun “nothing” as synonymous with the expression “a
non-entity”, we would also consider “nothing” to be a nonsense term. In fact, “nothing” is
a syncathegorematic element of negative sentences: it is either a propositional connector
or it may be the negation of an existential quantifier. The phrase “Nothing is eternal”
therefore means the same as “There is no entity which would be eternal”.
As for conjunctions in propositional contexts, e.g. the conjunction “or”, they fulfil
the modifying function in relation to sentences which are combined with the aid of
them: these sentences do not represent in this context performed judgments but presented
judgments, and a whole alternative sentence expresses the probability of these presented
judgments. The degree of this probability remains in inverse proportion to the number
of main elements of this context. A similar modifying function—at least in certain
contexts—is fulfilled by the connective “if” in relation to sentences-elements (scil.
predecessor and successor). The whole of such a context expresses the judgment that
28 A. Brożek
the logical relation of consequence occurs between the presented judgments. We are
talking here about so-called formal (scil. logical) truths—i.e. the judgments considered
consequences of the relevant reasons—that is the “truth” is opposed sometimes to material
truths. However, if these “truths” are true, it is precisely in the same sense in which any
materially true judgment is true. Thus, formal truths are a kind of material truths, i.e.
judgments which state what is or deny what is not.
4.5 Ethics
Twardowski was a forerunner (anyway, in Poland) of the program of so-called inde-

pendent ethics, i.e. an ethics free from ideological assumptions. Among the traditional
ethical problems, he separated problems which, according to him, could be considered
scientifically. He claimed that the task of scientific ethics consisted in a description of
the conditions in which the aspirations and actions of individual people and communities
could be reconciled to the greatest extent.
4.5.1 Ethical Absolutism
Just as in the theory of truth, he also presented in ethics a profound critique of axiological
relativism and skepticism.
In support of ethical agnosticism, the following arguments were formulated, among
others: (a) there are no absolutely obligatory ethical norms (resp. criteria); (b) there
are no commonly obligatory norms (at all times and in all societies); (c) the terms
“good” and “bad” have different meanings in different times and places. According to
Twardowski—these must be rejected because: (a) when we precisely formulate norms
which are considered general judgments, they appear to be norms restricted to a certain
domain; (b) the fact that, at certain times or in certain communities, certain norms are not
obligatory, should be understood in such a way that either they are not (at that time or in
these communities) recognized as obligatory, and they can be mistakenly recognized as
such, or that they (despite the general formulation) do not apply de facto to all cases—i.e.
they are not applicable in all of them; (c) the variability of notions does not imply the
variability of norms.
Among the arguments in favor of ethical relativism, sometimes the consequences that
the theory of evolution entails to ethics are mentioned. Twardowski—without ruling on
the merits of the latter—carried out a detailed analysis of these alleged consequences.
He showed that the spread of the theory of evolution caused the appearance of the
ethical conception, according to which ethical convictions were the result of natural
selection. Moreover, some people began to regard as moral every, and only such, an action,
which contributes to maintaining our own life and the survival of the species. This leads to
changing ethical beliefs along with changing living conditions and, consequently, to—as
it was called by Twardowski—“ethical anarchy”.
It is hard—Twardowski emphasized—not to acknowledge the fact that people differ in
their assessments as to the validity of various things: in the assessments of ethical actions
(performed “on the basis of conscience”), in aesthetic assessments (performed “on the
basis of taste”), as well as in logical assessments, i.e. assessments of beliefs in terms of

truth and falsehood (performed “on the basis of reason”). However, this does not imply
that there are no generally obligatory norms in matters of conscience, taste and reason.
Differences of opinions in these matters may have a source in the fact that we do not have
a theory to order these matters satisfactorily.
These are not “moral truths”—as Twardowski termed them—developed in the course
of evolution, but the development of humanity itself, improving in terms of “reason,
conscience and aesthetic taste.”
4.5.2 Freedom of Will
The question of free will is linked with determinism, when the freedom of will of X is
recognized as a lack of causes for the decision-making acts (scil. resolutions) of X. The
motives and personality of X (scil. the totality of X’s dispositions) could be such causes.
Twardowski was interested not in resolving the controversy of determinism-
indeterminism—although he was convinced that determinism is the most probable of
all possible positions in this issue. He wanted to determine whether determinism entails
such ominous effects for ethics as anti-determinists sometimes thought. Twardowski’s
answer was negative: if resolutions had in fact causes-motives and the personality of the
acting person influenced them, then nothing stood in the way of assessing ethical actions
just because of these motives and this personality. Because, generally speaking, “never
and nowhere is the necessity of a certain phenomenon an obstacle to evaluating it”.
But the question arises as to whether X is responsible for the actions of X taken
under the resolution of X, determined by the motives and personality of X. According
to Twardowski, if a certain action of X is taken in such circumstances—that can be (as
Twardowski used to say) “accounted” as X’s action, and therefore X is a perpetrator of
this action—then X is responsible for this action to Y (if, in particular, Y has the right to
obtain compensation): this action is his merit (if good) or his guilt (if bad), and therefore
it is worthy of reward or punishment.
Twardowski thought that although the existence of free will was not necessary for
moral improvement, for some people, believing (alternatively: instinctive believing) in
its existence was needed. If the thesis of determinism was true, then the reason for the
existence of practical ethics (eo ipso for formulating moral norms) could be that some
people carried about “the positive value of resolutions”.
4.6 Aesthetics
In the field of aesthetics—Twardowski conducted, among other things, an analysis of two

questions examined in the context of experimental, and therefore scientific aesthetics: the
question of the criterion of beauty in art, and the question of how music is able to perform
its evocative function.
Experimental aesthetics is practiced by a person, who verifies, with the aid of care-
fully selected experiments, hypotheses posed by aestheticians and concerning aesthetic
assessments and preferences. One such hypothesis is that “only sizes of figures based
30 A. Brożek
on the line of the so-called golden ratio could be considered beautiful”. As has been
shown by experimental research, this hypothesis—according to Twardowski—is verified
by “aesthetically educated minds”, and is falsified by “aesthetically uneducated ones”.
Twardowski posed the question of how music evokes musical experiences in the
audience. He answered that question as follows: pleasure arises as a result of listening to
music (irrespective of any text to which it is associated), because: (a) this listening does
not require any effort; (b) sounds of music themselves are “pleasant”; (c) listening to
music, we have a “sense of both diversity and homogeneity of a certain number of
impressions”. But a musical composition which is listened to also evokes experience with
an analogical structure—with its own (melodic, rhythmic, chordal) structure.
5 Philosophical School
Twardowski conducted a de facto reorganization of the philosophical life in Lvov.

He led to the origin of modern philosophical studies, to the establishment of various
philosophical institutions—and he instilled in his students and all the people affected by
him—the love of truth and diligent work. He instilled in his pupils not so much individual
ideas, but rather certain methodological assumptions.
In terms of philosophical and ideological beliefs, Twardowski’s students differed very
much: among them were ontological dualists and monists; axiological absolutists and
relativists; opponents and proponents of multi-valued logic; conservatives and socialists;
priests and atheists. What they shared was the above-mentioned precision in expressing
thoughts and attention to properly justifying accepted theses—and, associated with those
qualities, carefulness in solving problems and a critical attitude in relation to results:
primarily one’s own—but subsequently, also others.
The didactic activity of Twardowski had a lot of components.
First, he delivered cycles of so-called introductory lectures concerning in fact—
successively—all the systematic disciplines of philosophy and the history of philosophy.
Secondly, he had lessons for a narrower group of students, forming in them the ability
of independent philosophical thinking. Thirdly, he delivered lectures to a wider audience:
within the university, designed for all faculties—and outside the university, including non-
academic groups. Fifthly, he participated in the discussions of the Philosophical Circle,
existing at Lvov University before his arrival in Lvov, and then in the Polish Philosophical
Society, created by him in 1904. Sixthly, the program to create a serious study of
philosophy in Poland also involved creating the magazine Philosophical Movement
(1911) and involving in the organization of the existing Philosophical Review.
Twardowski’s program has brought great results. After World War I, Twardowski’s
School spread to other Polish philosophical centers: the Lvovian wing was, first of all,
supplemented by Warsawian one. This happened because chairs of philosophy and logic
in Warsaw were taken by some of the first generation of Twardowski’s pupils. Thus
it came to the formation of the Lvov-Warsaw School, the largest, most influential and
most famous in the world—so far—Polish school of philosophy. It included philosophers
of the caliber of Jan Łukasiewicz, Władyslaw Witwicki, Zygmunt Zawirski, Stefan
Baley, Stanisław Leśniewski, Tadeusz Kotarbiński, Władysław Tatarkiewicz, Tadeusz

Czeżowski, Kazimierz Ajdukiewicz and Izydora Dambska.
˛
The value of Twardowski’s work is shown in the fact that at the time of his death—his
immediate disciples held chairs in philosophical domains (and many of Twardowski’s
pupils also held chairs in other fields) at almost all of the Polish universities (i.e. in
Warsaw, Poznań, Cracow, Lvov and Vilna).
6 Conclusions
The basis of Twardowski’s philosophical views is a pluralistic conception of being,

a bipolar conception of actions, an intentional conception of consciousness, a triadic
conception of presentations, a presentative conception of images, an allogenic conception
of concepts, a constructivist conception of universals, a gradational conception of the
accuracy of concepts, and an idiogenic conception of judgments.
These views form the metaphysical basis of his descriptive semiotics. This is
an abstract theory of language phenomena, i.e. a system of definitions and their
consequences—built with the method of logical analyses, preceded by an inventory
and classification system supplementing these phenomena. The core of this theory is a
psycho-physical conception of signs and a functional conception of expressions.
In semiotics, Twardowski is to be respected fourfold: as a critic of other people’s
conceptions, a creator of his own analyses, refinements and theses, a precursor of new
ideas and methods, and an impeller of later polemics.
It was under Twardowski’s influence that Jan Łukasiewicz became an anti-
psychologist. As a result of his criticism of relativism, the classical theory of truth was
adopted widely in Poland and became the background of the semantics of Alfred Tarski.
Distinguishing content and the object of presentation in the form proposed by Twardowski
has been assimilated and further established by Meinong. The criticism of idealism,
based on these distinctions, was continued later, fruitfully, by Ajdukiewicz. The view that
probability is possessed by presentations of judgments and not by judgments themselves,
was taken over from Twardowski, i.a. by Łukasiewicz. Coming from Twardowski’s idea
of the relationship of action-product—as a relationship of causation different from causal
connection—resulted in the future praxeology of Kotarbiński. The idea of examining
objects as correlates of mental actions, i.e. as existentially neutral objects—was also
present in the ontology of Meinong and in the phenomenology of Edmund Husserl—and
was revived later in Saul Kripke’s semantics of possible worlds. The idea of images—as
a possible underlying presentations of concepts—has allowed the extension of normal
defining beyond the classic formula (per genus). The idea of concepts as presentations
containing, i.a. presented judgments in its content—can be regarded as the announcement
of a Russellian reduction of notions to propositional functions. The idea of presented
judgment as a presentation of performed judgment resulted in the Ingardenian conception
of quasi-judgments. The idea of a work of art as a product of an artist’s actions, different
from the material of these actions—was developed in detail by Roman Ingarden in his
intentional aesthetics. In Twardowski, we should look for the prototypes of Czeżowski’s
32 A. Brożek
method of analytical description and the Ajdukiewiczian method of semantic translation

(of interpreted texts).
All this is enough to consider Twardowski as, not only an extraordinary teacher, but
also an excellent scholar.
Kazimierz Twardowski’s Works Mentioned in the Paper
1. Zur Lehre vom Inhalt und Gegenstand der Vorstellungen. Eine Psychologische Untersuchung. Wien,
Hölder (1894) (English translation: On the content and object of presentation. A psychological
investigation. Martinus Nijhof, The Hague (1977))
2. Die Unsterblichkeitsfrage. Wydawnictwo WFiS UW, Warszawa (2009) (Manuscript 1895)
3. Wyobrażenia i poj˛ecia [Images and concepts]. In: Wybrane pisma filozoficzne [Selected philosophical
writings]. PWN, Warszawa (1965), pp. 114–197 (Original work published 1898. English translation:
Imageries and concepts. Axiomathes, vol. I, 79–104 (1995))
4. Zasadnicze poj˛ecia dydaktyki i logiki [The essential concepts of didactics and logic]. Nakładem
Towarzystwa Pedagogicznego, Lwów (1901)
5. O filozofii średniowiecznej wykładów sześć [Six lectures on medieval philosophy]. H. Altenberg,
Warszawa (1910)
6. Mowy i rozprawy z okresu działalności w Towarzystwie Nauczycieli Szkół Wyższych [Addresses
and papers from the period of activity in the Association of Teachers in Higher Education]. Nakładem
TNSW, Lwów (1912)
7. O patriotyzmie [On patriotism]. In: Myśl, mowa i czyn [Thought, Speech and Actions]. Vols. I–II.
Copernicus Center Press & Wydawnictwo Naukowe Semper, Kraków & Warszawa (2013–2014),
vol. I, pp. 436–455 (Original work published 1919)
8. Autobiografia filozoficzna [Philosophical autobiography]. In: Przeglad˛ Filozoficzny vol. I(1), pp. 19–
33 (1992)(Original work published 1926)
9. Rozprawy i artykuły filozoficzne [Philosophical dissertations and papers]. Ksi˛egarnia SA
Ksiażnica-Atlas
˛ TNSW, Lwów (1927)
10. Wybrane pisma filozoficzne [Selected philosophical writings]. PWN, Warszawa (1965)
11. Wybór pism psychologicznych i pedagogicznych [Collected psychological and pedagogical writings].
WSiP, Warszawa (1992)
12. Etyka [Ethics]. Wydawnictwo Adam Marszałek, Toruń (1994)
13. Dzienniki [Diaries]. Vols. I–II. Wydawnictwo Adam Marszałek, Toruń (1997)
14. Twardowski, K.: On Actions, Products and Other Topics in Philosophy. Rodopi, Amsterdam (1999)
15. Filozofia i muzyka [Philosophy and music]. Wydawnictwo Naukowe Semper, Warszawa (1995)
16. Dzienniki młodzieńcze [Youthful diaries]. Wydawnictwo Naukowe Semper, Warszawa (2013)
17. Myśl, mowa i czyn [Thought, Speech and Actions]. Vols. I–II. Copernicus Center Press &
Wydawnictwo Naukowe Semper, Kraków & Warszawa (2013–2014)
18. Brozek, A., Jadacki, J.: On Prejudices, Judgments and Other Topics in Philosophy. Rodopi,
Amsterdam (2014)
A. Brożek ()
Institute of Philosophy, University of Warsaw, Warsaw, Poland
Jan Łukasiewicz: A Creator of New Ideas in
Logic and a Reinterpreter of Its History
Jacek Jadacki
Poetic works do not differ from scientific in more imaginative fantasy.

[. . . ] However, scientists differ from poets in this respect, that the
former REASON always and everywhere.
About creativity in science (Łukasiewicz [3, p. 32])
Abstract Jan Łukasiewicz was a leading figure of the Warsaw School of Logic—one
of the branches of the Lvov-Warsaw School. The paper presents his personality, life and
didactic activity, as well as the list of his main works and the greatest achievements in
logic, its history and philosophy. In propositional logic, he invented the bracketless sym-
bolism and constructed many systems of this logic. One of his greatest achievements
was discovering three-valued logic. He also reconstruct the history of logic with the
apparatus of modern logic. In ontology, Łukasiewicz made a logical analysis of the notion
of causality and determinism. In epistemology, he established the precise formulation of
the paradox of the liar.
Keywords Determinism · History of logic · Jan Łukasiewicz · Lvov-Warsaw School ·

Many-valued logic · Propositional logic
Mathematics Subject Classification (2000) Primary 01A70, 03B05; Secondary 03B50
1 Personality
People dabbling in science are eminent, if they see fundamental questions in their
discipline and find original responses to these questions. If the theoretical construction
proposed by an eminent scholar—in particular, an eminent logician—also has the value
of simplicity and beauty, one can say about him that he is a genius. Łukasiewicz was
certainly a genius in this sense.
It was said of him that he was shy, sensitive and irritable. He was sensitive to how
others judged him—and whether he was appreciated by them. He could not hide the fact
that he cared about his recognition.

34 J. Jadacki
Sometimes, talents are born—as in the Polish phrase—on the stone; but a talent may
not generally develop so as to become a genius: to do this, one needs the appropriate
soil. This soil—in the case of Łukasiewicz—was the mental environment brought to
life by Kazimierz Twardowski, that is, in short, the philosophical Lvov-Warsaw School.
Łukasiewicz grew in this environment—and then he co-created this environment, not
without mutual theoretical interaction also with his own colleagues and students.
What was the relationship of Łukasiewicz to that environment?
He identified himself with the Brentanian roots of the School—but not with all of
them. Twardowski was highly estimated by him—but not for everything. He appreciated
Stanisław Leśniewski—as a logician—at least initially, higher than himself.
Three women played a great role in the life of Łukasiewicz: firstly—his mother,
Leopoldyna née Holtzer; then—princess Maria Józefina Sapieżanka, who was the object
of his great love, but without reciprocity; in the end—his wife, Regina née Barwińska,
who was his bedrock especially in the last years of his life (even though—let us add—a
sometimes troublesome bedrock due to her antagonistic character).
He was a great Polish patriot, but his ancestors were Ruthenian (paternal) and Tyrolean
(maternal). To strangers, with whom he came into contact, he referred differently. He had
friends among the Germans—but he did not feel good in Germany. He treated Ukrainians
with sympathy—but he was, i.a., an opponent of ukrainizing the Lvov University. He felt
aversion to some Poles of Jewish origin—but he fiercely opposed anti-Jewish movements
in the academic circles. In the last decade of the life, Ireland became his second
homeland—but his attitude to the Irish people was rather (unfairly!) dry.
On the one hand he was a man of deep faith: in particular, a Catholic and a practicing
Catholic. On the other hand, he avoided the public «confession of faith». He also avoided
political declarations—which does not mean that he did not have an explicit (conservative)
view on these issues.
2 Life
He was born on December 21, 1878, in Lvov; he died on February 13, 1956 in Dublin.
He studied law and later philosophy with Twardowski at the Lvov University—and at
the universities of Berlin and Louvain. After his doctorate (1902) and habilitation (1909),
he was a participant in the seminar of Alexius Meinong in Graz (1910). From the years
1911–1915, he was a professor of philosophy and logic at Lvov University; from the years
1915–1939 at the Warsaw University (with breaks from 1918–1920 and 1924–1929),
where he served twice as the rector (1922/1923 and 1931/1932). In 1919 he was the
Minister of Religious Denominations and Public Education. During one of the German
raids at Warsaw in September 1939, his library collections and rich manuscript legacy
burned totally. In 1944, as an avowed anti-communist—in the face of the approaching
front of the Soviet troops, he decided to go to Switzerland, but because of the tense
political situation after the assassination attempt on Adolf Hitler, he had to stop in Münster
(under the supervision of Heinrich Scholz). After the war, he settled first in Brussels and
then in Dublin, where he was professor of logic at the Royal Irish Academy.
Jan Łukasiewicz: A Creator of New Ideas in Logic . . . 35
He was a member of the Polish Academy of Sciences, the Polish Scientific Society
in Exile, and received honorary doctorates from universities in Münster and Dublin. He
was a leading figure in the Warsaw School of Logic, being an essential component of the
philosophical Lvov-Warsaw School. The spirit of the School found in his philosophical
and logical works its most perfect incarnation.
3 Works
Among the most important publications of Łukasiewicz, there are the following books: On
the principle of contradiction in Aristotle [2], Die Grundlagen der logischen Wahrschein-
lichkeitsrechnung [5] and Aristotle’s syllogistic from the standpoint of modern formal
logic [11].1
Most of his papers were collected in two Polish volumes: Selected problems of logic
and philosophy [13] and Logic and metaphysics. Miscellanea [19].
Many of his works appeared also in translation into other languages, including English
[13], French [16, 20, 22], Spanish [17], Japanese [18] and Russian [12, 21].
Łukasiewiczian Elements of mathematical logic [8] is the model of a manual. This
work includes a lecture on the axiomatic system of classical propositional calculus, propo-
sitional calculus with quantifiers and a fragment of the calculus of names (Aristotelian
syllogistic included). An unusual source for the history of Polish culture of the twentieth
century is Łukasiewicz’s Diary [23].
The most important scientific achievements were Łukasiewiczian logic, the philosophy
and history—as well as ontology and methodology of philosophy.
4 Metaphysics
4.1 Ontology
Of ontological issues, Łukasiewicz was especially interested in two problems (though

interrelated in many ways): what is a causal relationship and what are the reasons for
determinism.
4.1.1 Causal Relationship
An attempt to define “causal relationship” was made by Łukasiewicz in his classic

dissertation “Analysis and construction of the concept of cause” [1]. The analytical part of
this dissertation has little equal in philosophical literature—and not only in Polish writing.
1 Thiswork is a continuation of the Polish monograph, prepared in 1939, which unfortunately burned
during World War 2.
36 J. Jadacki
In the constructive part, Łukasiewicz proposed reducing the notion of a causal

relationship to the notion of necessity. This reduction could be simplified in such a way:
The fact that object P1 has feature c1 is the reason for the fact that object P2 has feature
c2 —when if object P1 has feature c1 , then object P2 must have feature c2 , where P2 must
have feature c2 , when if object P2 did not have feature c2 , then object P2 would be an
internally contradictory object.
Łukasiewicz initially thought that he gave, in this way, an equivalent definition of
“causal relationship”. But it finally turned out, that the necessity of the relationship
between the two states of affairs does not settle the fact that this is a causal relationship,
since the impossibility of the occurring state of affairs S2 without the occurring state of
affairs S1 can take place, i.a. in the case where the occurring state of affairs S1 is later
than the occurring state of affairs S2 , or when the two states of affairs are timeless (like,
e.g., the fact that a certain figure is a square necessarily involves the fact that this figure is
a rectangle)—whereas it is assumed that every state of affairs is later than its cause.
Consequently, the definition proposed by Łukasiewicz can be treated at most as an
inclusive definition, indicating only a necessary condition for the occurrence of a causal
connection.
4.1.2 Determinism
The ontological thesis of determinism (in one of its versions) holds that each state of
affairs occurring in the real world is uniquely determined by causes preceding it.
Łukasiewicz pointed out that among the premises forming the basis of the thesis
of determinism there are two principles: the principle of the excluded middle and the
principle of causality [6]. The first—let us remember—states (in a certain version) that
for any state of affairs, this state of affairs occurs or does not occur. The second one
states (in a certain version) that every state of affairs has a cause in some previous state
of affairs. Following the principle of the excluded middle applied to the future, we must
recognize that the occurrence of any future state of affairs has been already determined;
but—as Łukasiewicz says—there is no compulsion to accept the principle of the excluded
middle (on the grounds of, e.g., trivalent logic, this rule does not apply). Supposing the
principle of causality, we must accept the fact that causal-effectual chains are infinite «in
the past» (i.e., they are eternal); but—as Łukasiewicz says—one can accept the hypothesis
of infinite causal-effectual chains and at the same time recognize that in a certain—
distant enough—moment, there is «already» no reason for a given state of affairs, if only
one assumes that time intervals between successive links in the causal-effectual chain
decrease «back» unlimitedly; under such a condition, adopting the principle of causality
is compatible with indeterminism.
4.2 Epistemology
Łukasiewicz’s attitude to epistemology was more than critical: he considered most

epistemological problems to be apparent ones.
The problem of truth was one of the real problems to which he devoted a lot of
attention; it was understandable: truth is one of the logical values, and Łukasiewicz
worked on constructing logical systems in which more than two logical values could be
admissible, and not only truth and falsehood.
By “truth”—or more precisely: a “true sentence”—Łukasiewicz understood a sentence
that “admits this property to an object, which is really possessed by this object, or that
refuses this property, which is not really possessed” [5, p. 55]. We use different criteria
for the truth of sentences, but we are not able to justify that these or other criteria are valid;
an attempt to give such a justification always ends in either a vicious circle or regressus
in infinitum.
Łukasiewicz was probably the first to establish the precise formulation of the so-called
paradox of the liar. As the source of the paradox, he indicated a sentence of the type
“Sentence S is false”: such that, if sentence Z, that is, the sentence “Sentence S if false”,
is true, it is just as it says, so it is false; if sentence S is false, then it is just not the case,
so sentence S is not false—or it is true. This formulation became an inspiration for the
semantic conception of truth, proposed by the student of Łukasiewicz, Alfred Tarski.
5 Logic
5.1 Propositional Logic
A particular subject of Łukasiewicz’s interest was a classic propositional calculus,

i.e. a logical theory which reports the meaning of conjunctions (functors) connecting
sentences—such as “and”, “or”, “if . . . , then”, “always and only if”—and the negation
“it is not the case that”.
Let us start with the fact that Łukasiewicz invented for this calculus a special
symbolism, called “bracketless symbolism” and later “Polish symbolism”. It consists
in recording complex sentences in such a way that at first we give a functor, followed
by sentences «bonded» by this functor. For example, the formula “If p, then (q and r)”
is recorded in such a symbolism as: CpKqr (where ‘C’ signifies functor “if . . . , then”,
and ‘K’—functor “and”). In the parenthetical symbolism, this formula has the form:
p → (q ∧ r). Note that the bracketless formula, mentioned above, has two symbols
less than its equivalent parenthetical formula.
Now: propositional calculus in the form of axiomatic theory is a set of statements about
these conjunctions, each of which is either an axiom (and therefore the claim accepted
without proof), or can be derived from axioms using determined rules called “rules
of inference”. Łukasiewicz constructed many versions of such propositional calculus—
differing, among other, as to adopted axioms and their number, rules of inference and
their number, as well as which functors are considered primary and which secondary, i.e.
definable with the use of the former. Among those versions, one was considered a classic
one: it is a system which operates one functor (i.e. the functor “always if . . . ”—or the
functor of implication), universal quantifier (“every . . . ”) binding propositional variables,
four rules of inference, symbols of truth and falsehood, and symbols of accepting and
rejecting sentences.
38 J. Jadacki
Łukasiewicz’s desire was to create a system which would contain as little as possible
axioms, rules of inference etc. He succeeded in this respect, firstly, in inventing a
system of implicational propositional calculus (or a calculus operating with only one
functor: the functor of implication) based on one axiom numbering—in the bracketless
symbolism—13 symbols (CCCpqrCCrpCsp), and then to prove that it is the shortest
axiom of that implicational calculus. Secondly, Łukasiewicz honed the rules of inference.
He invented, among others, a useful version of the rule of substitution, i.e. the rule
according to which in a sentence containing variables we are free to insert in their
place other variables or constants, as long as we are doing it consequently (i.e. to insert
these variables or constants into the places of each appearance of a substituted variable).
He developed a more precise reasoning called “generalizing deduction”, i.e. the method
of proving general statements on the basis of their particular cases. He codified the
matrix characteristics of the functor of propositional calculus and the method examining
the tautologicality of formulas of that calculus, involving the compilation of possible
combinations of substituting propositional variables in these formulas by symbols of truth
and falsehood.
5.2 Propositional Logic with Quantifiers
In turn, the system of propositionalcalculus with quantifiers was based by Łukasiewicz

on: (a) the universal quantifier ( ) and the functor of implication (C) as primary
concepts; (b) the three axioms (Tarski and Paul Bernays’s ones:CqCpq, CCCpqpp
and CCpqQQqrCpr); (c) the definition of negation (Np = Cp pp) and (d) five
rules of inference: the rule of substitution (modified in comparison to the analogical
rule in propositional calculus without quantifiers), the rule modus ponens, the rule
of replacement, the rule of combining, and the rule of skipping quantifiers. In this
system, Łukasiewicz presented proofs of 19 theorems (he left five theorems without
proof), including the proofs of three axioms of the system of propositional logic without
quantifiers, described above.
5.3 Reconstruction of Syllogisms
Łukasiewicz has reconstructed the most significant part of the logic of Aristotle, i.e. “as-
sertoric” syllogistic (scil. non-modal) [11]. His intention was to make this reconstruction:
on one hand—according to the intentions of the great philosopher; on the other hand—
developed in the spirit of modern logic.
According Łukasiewicz—Aristotle’s syllogistic is a part of the logic of names, namely
a formal theory of three constants: “all . . . are . . . ”, “none . . . are . . . ”, “some . . . are . . . ”
and “some . . . are not . . . ”, where the values of the variables representing arguments of
those functors of two arguments are adopted only in general terms (in particular, with the
exclusion of empty and negative names). The so constructed syllogistic is superstructured
over propositional logic; in particular, it contains the following constants of this logic: “if
. . . , then . . . ”, “ . . . and . . . ” and (in some proofs), “it is not the case that . . . .”
Aristotle tried to axiomatize syllogistic, taking as its basis four modes of the first
syllogistic figure, eventually reducing finally to two of them (Barbara and Celarent). It
turned out, however, that we need to add to them two laws of conversion, and (in some
cases) two laws of identity. The simplest axiomatic base contains as primary terms—the
constants “all . . . are . . . ” and “ . . . some . . . are . . . ” (the rest can be defined with their
help and the negation of sentences), and as axioms—two laws of identity and the moods
Barbara and Datani (or: Barbara and Dimaris).
Reduction of imperfect syllogisms to perfect ones, postulated by Aristotle, was
interpreted by Łuksiewicz as the proof of theorems of the system (scil. deriving them
from axioms). According to Łukasiewicz, Aristotle gave not only proofs of the true
syllogistic formulas, but he also tried to show that all the others formulas are wrong, and
as such should be rejected. He rejected inconclusive formulas, usually using the method
of exemplification by means of appropriate concrete terms (which satisfy “premises”, but
do not satisfy—“conclusions”).
5.4 Many-Valued Logic
One of Łukasiewicz’s most important achievements was discovering three-valued

logic2 —and more generally: logics more than two-valued; the philosophical importance
of the three-valued logic was compared, by Łukasiewicz himself, to the importance of
non-Euclidean geometry in mathematics.
Existing logical calculi based on the principle of bivalence—i.e. on the assumption that
every sentence has exactly one logical value: it is either true or false: tertium non datur.
Łukasiewicz generalized the concept of logical value in such a manner that he allowed
the existence of “intermediate” values between truth and falsehood: one—in the case of
the three-valued logic, two—in the case of the four-valued logic, or more—up to the
infinitely-many-valued (in short: ℵ0 -valued) logic.
Behind the idea of the three-valued logic there were the following insights. There are
sentences which at the given moment can not be principally (and not, e.g., because of
someone’s ignorance) determined to be true or false. These are sentences about future
events, which at present are not determined (i.e. those that there are currently neither
the cause of their occurrence, nor the cause of their non-occurrence—or sentences about
past events, the effects of which have completely “expired” (i.e. those for which there are
currently no effects of their occurrence). These are just sentences having a third logical
value. So sentences can be not only true or false, but also undetermined.3
An example of intuitive interpretation of the four-valued logic is its interpretation in
terms of modal logic, that is, one in which there are functors of the type “it is necessary
2 The first outline of non-Chrisipian logic dates back to 1917, and the first system of it—to 1920.
3 Itmay be worth noting that one of the consequences of adopting the so characterized indefiniteness as a
third value in the three-valued logic is that we should to accept the view that sentences change their logical
value in time: e.g. sentence undetermined at a certain moment may come to be true (or false) at some later
moment, namely, one in which adequate causes have already occurred. Someone who would not want to
agree with such a consequence, could not also agree with the sketched interpretation of the third value.
Questioning the interpretation does not involve, of course, questioning the interpreted calculus.
40 J. Jadacki
that . . . ” and “it is possible that . . . .” Łukasiewicz made this interpretation when it turned
out that the characteristics of propositional functors in the conceptual apparatus of the
three-valued logic is unsatisfactory.
As regards the infinitely-many-valued logic, Łukasiewicz claims (at least initially) that
logical values present in this logic can be identified with degrees of probability.
5.5 Metalogic
Among Łukasiewicz’s great achievements, there were the results of his metalogical
analysis, and in particular: his results of research on consistency, independence and the
completeness of axioms of the propositional calculus. The issue at stake was to determine
whether or not the constructed sets of axioms contained axioms negative to each other
such that some of them resulted in others, and finally, whether these axioms are sufficient
to prove all the theses of the system, which appears to be true.
It was also demonstrated here that classical propositional calculus is a fragment
of intuitionistic logic. If one considers that at the core of the latter belief there is
the conviction that only proved theorems can be considered as theorems of a certain
mathematical theory—and therefore among its theorems there is not, e.g., the law of
excluded middle (“p or it is not the case that p”)—then Łukasiewicz’s result sounds
surprising (removing this impression requires suitably enriching intuitionistic logic itself).
5.6 Philosophy of Logic
In the second half of the nineteenth century—not without relation to the rapid devel-
opment of psychological research—tendencies appeared to reformulate all philosophical
issues, including problems of logic, in such a way that every philosophical (and logical)
problem was replaced by its psychological paraphrase.
Łukasiewicz was one of those people who revolted earliest and most strongly against
such a psychologization—especially the psychologization of logic. His anti-psychologism
was reflected, among other things, in the fact that he precisely contrasted laws of logic
with their alleged equivalents in the form of psychological laws, and—let us add—
ontological ones. He stressed, e.g., the difference between logical, psychological and
ontological principles of contradiction: the first principle (in one of its versions) is the
law of propositional logic, according to which it is not the case that both p and not-p; the
second principle (in one of its versions) claims that in reality is never so that a certain
state of affairs occurs and does not occur at the same time; the third principle (in one of
its versions) is the view that no one alive can hold a certain conviction and its negation at
the same time.
Łukasiewicz—analyzing these versions of the principle of contradiction—noted that
an experience is not able to confirm the ontological version of these principle. Here are
excerpts of this argument—a beautiful example of Łukasiewicz’s philosophical prose:
Any movement [. . . ] takes place in such a way that the changing object loses some features that it
possessed, and acquires new ones that it did not possess. In both cases, contradiction would arise,
if there were not different temporal determinations.
If the change is CONTINUOUS, e.g. the movement of an arrow released from a bow, [. . . ] then
in every smallest interval of time, the changing object loses in turn some features, and purchases
second ones. The moving arrow is in any two moments of time in DIFFERENT places. [. . . ] What
[. . . ] happens when this distance decreases to zero when we shall consider only ONE moment as
unextended on timeline?
Once, we heard the fable that when a princess pricked her little finger on a spindle, she fell
immediately into a hundred-year’s deep dream, and all life around her slept as well. In such a way,
the legendary Popiel’s court froze in the blink of an eye, enchanted by Rzepicha in the songs of
King of Spirit. Suppose that what is only a poetic fantasy has become reality. [. . . ] The arrow would
[then] rest motionless in a certain place. But how do we know that it would be only in ONE place?
Why, in an unextended moment, in a temporal point of the section, could it not be in at least two
different places and thus be in a certain place and not be there at the same time? [. . . ] Experience
is silent on this issue. [. . . ] All the perceived phenomena LAST shorter or longer and SHOULD
last for a minimum period of time to be noticed by us at all. We do not know what happens in an
unextended moment. However the principle of contradiction applies to such a moment; because if
we say that the arrow cannot AT THE SAME TIME be and not be at the same place, the phrase “at
the same time” refers to THE SAME, so the only ONE, unextended moment. [2, pp. 136–138]
So much for the ontological version of the principle of contradiction. As for the
empirical justification of the psychological version, it would require painstaking research,
which has not yet been undertaken and whose desirability at all is in question in
light of the statements of some people (including philosophers and mystics) that they
entertain contradictory beliefs. Finally—the only justification for accepting the logical
version of the principle of contradiction is that it makes it possible to prove with a
certainly unattainable in other cases that someone is mistaken or lying; because we
recognize (assuming the principle of contradiction) a conviction entailing a contradiction
as mistaken—and we have (assuming this principle) the basis to believe that someone
is lying when that person has once said, that p, and some other time, that not-p; the
inability to prove that someone is wrong or lying, would have far-reaching—negative—
consequences for social life, because, among other things, it would make it difficult, and
in some cases even prevent issuing of righteous judgments by the courts.
5.6.2 Metaphysical Neutrality
Łukasiewicz argued not only for depsychologizing logic, but also for—so to speak—the
metaphysical neutralizing of it. The idea was that—already in the twentieth century—
there were tendencies to treat calculi of logical systems as systems implying one or
another ontological or epistemological positions. Additionally, to show the groundless-
ness of such tendencies—it is enough to exactly distinguish the logical system from its
permissible interpretation—and from metaphysical positions occupied by a logician who
is the creator of this system or its interpreter. Logicians tend to be of defined metaphysical
views—or of a defined worldview—but they feed these views as metaphysicians or
members of a particular philosophical orientation, not as logicians: logical systems, as
such, neither assume nor imply any metaphysical theses or any theses of faith or unbelief.
Moreover, practicing formal logic does not require, e.g., accepting nominalism—or
the view that logical systems are systems of unretrievable, «senseless» (i.e. having no
42 J. Jadacki
reference to reality) symbols-strings. If it were so, logical systems would be always sets
of a finite number of theses—because we can not «produce» (or build) an infinite number
of symbols-strings. Meanwhile, these systems are infinite sets (it is enough to note that if
the statement “If p, then p” is a thesis of the propositional calculus, then the statements
“If p, then (p or q)”, “If p, then [p or (q or r)]” etc. ad infinitum are such theses also.
Similarly, e.g., the existence of many logical systems does not support relativism and
conventionalism—or the view that our images of the world are determined by the freely
accepted conceptual apparatus, so that none of them can reasonably be considered as
a «true» image; because in this case, the issue is decided by experience, providing us
knowledge about what the world «truly» is—not the mere fact of the existence (scil.
constructability) of various its images.
5.7 History of Logic
Łukasiewicz’s basic postulate in relation to historical research on logic (and more

generally—philosophy) was the postulate to reconstruct history with the apparatus of
contemporary logic; the first work realizing this postulate was the dissertation On the
principle of contradiction in Aristotle [2], although the postulate itself was explicitly
formulated much later [9].
According to this postulate—Łukasiewicz analyzed two great logical systems of
antiquity: Aristotle’s syllogistic and Stoic logic. Results of this analysis were surprising—
and they questioned existing views on both systems.
It turned out, firstly, that Aristotle’s syllogistic is not a theory of inference (as was
previously thought), but a calculus of names: in particular, Aristotle’s syllogisms are not
rules of inference (of the type: If we accept the premise “Each A is B” and the premise
“Each B is C”, we must accept the conclusion “Each A is C”); they are theses of the
calculus of names (of the type: If each A and each B is C, then each A is C). By the way,
it turned out that Aristotle is the inventor of nominal variables and his syllogistic was the
first (admittedly imperfect) axiomatized system in the history.
Secondly, Łukasiewicz showed that the Stoic logic is not (as previously thought) a
calculus of names, but historically the first the system of propositional logic having its
extension in the Middle Ages under the name of the theory consequences: in opposition
to Aristotelian syllogisms—Stoic syllogisms are the rules of inference.
6 Methodology
In methodology—Łukasiewicz proposed an original classification of reasoning and

reinterpreted the notion of inductive reasoning, the notion of probability and the notion of
magnitude.
6.1 Reasoning
According Łukasiewicz—to reason (let us add: validly) is the same as to select for
a certain sentence S a reason (or a sentence which is followed by sentence S) or a
consequence (i.e., a sentence which follows sentence S); in the first case we are talking
about reduction; in the second case—about deduction. Sentences—given and selected—
may be accepted or not accepted (as true sentences) by the reasoner before beginning
to reason. Depending on which of these two cases takes place—reduction is either an
explanation (when we select an unaccepted as yet reason for an accepted sentence) or
proof (when we select an accepted reason for an unaccepted as yet sentence); on the
other hand deduction is inference (when we select an unaccepted as yet consequence to
an accepted sentence) or confirmation (when we select an accepted consequence to an
unaccepted as yet sentence).
The reasoning consisting in the fact that a certain law is accepted—therefore a general
statement stating occurrence of certain general regularities—on the basis of earlier
accepted individual sentences stating occurrence of a certain number of cases of this
regularity is called “inductive reasoning” (more accurately: “incomplete enumerating
induction”). Before Łukasiewicz induction was believed to be a kind of deduction. In light
of Łukasiewiczian conception of reasoning, this is a misconception—inductive reasoning
is a kind of explanation (in which a reason is selected to individual sentences), and thus it
is a kind of reductive, and not deductive reasoning.
6.2 Probability
In the traditional interpretation of probability, the probability of the occurrence of a certain

events is discussed, e.g. about the probability that picking up a random a ball from a face-
down box in which there are four balls: white, black, red and blue—I shall take up the
white ball. Łukasiewicz proposed a «propositional» interpretation of probability. Certain
probability is entitled not to the fact that (e.g.) I shall take up the white ball from the
box, but to the sentence “I shall take up the white ball from the box”. This sentence is a
substitution of the formula “I shall take up ball x from the box”. In this situation we can
insert in place ‘x’ a name of one of four colors: white, black, red or blue—but only in the
case of one substitution, will the formula be transformed into a true sentence. Therefore,
we can say that the probability of the sentence “I shall take up the white ball from the
box” is equal to 1/4 in this case.
6.3 Magnitude
Łukasiewicz proposed—in the place of a very complicated «classic» (but logically flawed)
definition of “magnitude” given by the mathematician Stanisław Zaremba—a very simple
definition, according to which magnitude is the same as an element of a certain well-
ordered set, or such a set, whose elements are «arranged» sequentially one after another,
44 J. Jadacki
because it is determined that a suitable relationship for them is both asymmetrical,

transitive and consistent (i.a. the relation of being-greater-than is such an ordering relation
in the class of natural numbers).
6.4 Methodology of Philosophy
The methodology of science can be regarded as a description of how—and, in particular,

by means of which research methods—science is actually done. We can also treat it as a
set of rules defining how science should be done.
Łukasiewicz’s contribution to the methodology of philosophy belongs mainly to
methodology in this second sense. Łukasiewicz’s expectations in this regard were very
far-reaching, although he expressed them in a simple postulate:
Metaphysical problems [have not been] resolved, but I do not think that [they are] unsolvable. We
only need to approach them with the scientific method: with the same proven method which is
used by a mathematician or a physicist. And above all, we need to learn to think: clearly, logically
and strictly. [7, p. 368]
In practice, Łukasiewicz advocated practicing «axiomatic» philosophy, i.e. a philoso-

phy with a structure similar to axiomatic logical systems.
The basis of such a philosophy should be its axioms. Łukasiewicz wrote:
Every sentence, especially those that will be the basis of philosophy, [. . . ] [should be] formulated
as precisely as possible because only then can we duly justify or know the direct evidence of these
sentences. [10, p. 372]
Evaluation of the current—especially so-called modern—philosophy was negative:

The state of pre-Kantian philosophy: on the one hand, fantastic dreams, not withstanding scientific
criticism; on the other hand, radical, dogmatic, unfounded skepticism. [But then also:] as
we approach Kantian philosophy with the requirements of scientific criticism, its construction
collapses like a house of cards. At every step, we have vague notions, incomprehensible sentences,
unfounded statements, contradictions and logical errors. [7, p. 368]
The conclusion was crushing:

Caused by a negligence of logic and by the ensuing mental dressage, the whole of modern
philosophy was paralyzed by an impotence of strict and clear thinking. [10, p. 373]
7 Teacher
Even an approximate list of Łukasiewicz’s students is not known; neither the audience
of his lectures, nor the participants of his seminars. Even the list of masters and doctors
promoted by him is incomplete and it raises doubts at various points.
Certainly, his postgraduate students were: Maria Ossowska, in the years 1923–1927,
a senior assistant at the Seminary of Philosophy of Łukasiewicz; Mordchaj Wajsberg
and Zygmunt Kobrzyński, both of whom died during World War 2—so they did not
survive their promoter; finally, Stanisław Jaśkowski and Jerzy Słupecki. It seems that
under the guidance of Łukasiewicz a doctorate on the Stoic logic was prepared by Czesław
Lejewski but there was no promotion because of the outbreak of the war. However, it is
not clear, e.g., whether Bolesław Sobociński was a postgraduate student of Łukasiewicz
or Leśniewski.
Łukasiewicz’s assistants—though not postgraduate students—were: Tarski (from
1929) and Henryk Hiż (from 1940 to 1944).
Those who witnessed Łukasiewicz’s lectures—colleagues and students—stressed that
he was an excellent didactician. This was manifested in the fact that he could make
contact with the participants of his lessons, and his lectures were strict and at the same
time interesting and affordable, with a clear structure, delivered fluidly and in beautiful
language.
Jan Łukasiewicz’s Works Mentioned in the Paper
1. Analiza i konstrukcja poj˛ecia przyczyny [Analysis and construction of the notion of cause]. In: Z
zagadnień logiki i filozofii. Pisma wybrane [Selected Problems of Logic and Philosophy]. PWN,
Warszawa (1961), pp. 9–62 (Original work published 1906)
2. O zasadzie sprzeczności u Arystotelesa. Studium krytyczne [On the principle of contradiction in
Aristotle. A critical study]. Akademia Umiej˛etności, Kraków (1910)
3. O twórczości w nauce [On creativity in science]. In: Logika i metafizyka [Logic and Metaphysics].
Miscellanea. Wydawnictwo WFiS UW, Warszawa (1998), pp. 9–33 (Original work published 1912)
4. Die logischen Grundlagen der Wahrscheinlichkeitsrechnung. Spółka Wydawnicza Polska, Kraków
(1913)
5. Zagadnienia prawdy [Problems of truth]. In: Logika i metafizyka [Logic and metaphysics]. Miscel-
lanea. Wydawnictwo WFiS UW, Warszawa (1998), pp. 55–56 (Original work published 1913)
6. O determinizmie [On determinism]. In: Z zagadnień logiki i filozofii. Pisma wybrane [Selected
Problems of Logic and Philosophy]. PWN, Warszawa (1961), pp. 114–126 (Original work published
1922–1923)
7. Kant i filozofia nowożytna [Kant and modern philosophy]. In: Logika i metafizyka [Logic and
metaphysics]. Miscellanea. Wydawnictwo WFiS UW, Warszawa (1998), pp. 365–368 (Original work
published 1924)
8. Elementy logiki matematycznej [Elements of mathematical logic]. Wydawnictwo Koła
Matematyczno-Fizycznego Słuchaczów Uniwersytetu Warszawskiego, Warszawa (1929)
9. Z historii logiki zdań [From the history of propositional logic]. In: Z zagadnień logiki i filozofii.
Pisma wybrane [Selected problems of logic and philosophy]. PWN, Warszawa (1961), pp. 178–194
(Original work published 1934)
10. Kartezjusz [Descartes]. In: Logika i metafizyka [Logic and Metaphysics]. Miscellanea. Wydawnictwo
WFiS UW, Warszawa (1998), pp. 370–374 (Original work published 1938)
11. Aristotle’s Syllogistic from the Standpoint of Modern Formal Logic. Clarendon Press, Oxford (1951)
12. Аристотелевская силлогистика с точки зрения современной формальной логики. Изда-
тельство Иностранной Литературы, Мoсква (1959)
13. Z zagadnień logiki i filozofii. Pisma wybrane [Selected Problems of Logic and Philosophy]. PWN,
Warszawa (1961)
14. Elements of Mathematical Logic. Pergamon Press, Oxford (1966)
15. Selected Works. North-Holland Publishing Company & PWN. Amsterdam and Warszawa (1970)
16. La syllogistique d’Aristote. Armand Colin, Paris (1972)
17. Estudios de lógica y filozofía. Biblioteca de la Reista de Occidente, Madrid (1975)
18. Sūri rondigaku genron. Bunka Shobō Hakubunsha, Tokio (1992)
19. Logika i metafizyka [Logic and metaphysics]. Miscellanea. Wydawnictwo WFiS UW, Warszawa
(1998)
20. Du principe de contradiction chez Aristote. L’Éclat, Paris (2000)
46 J. Jadacki
21. О принципе противоречия у Аристотеля. Критическое исследование. Центр Гуманитар-

ных Инициатив, Мoсква & Санкт-Петербург (2012)
22. Écrits logiques et philosophique. Vrin, Paris (2013)
23. Pami˛etnik [Diary]. Wydawnictwo Naukowe Semper, Warszawa (2013)
J. Jadacki ()
Kazimierz Ajdukiewicz: The Cognitive Role
of Language
Anna Jedynak
Abstract Kazimierz Ajdukiewicz was an eminent representative of the Lvov-Warsaw

School. His main interest was the cognitive role of language. His radical conventionalism
intended to explain rapid and fundamental changes in science. He used the method of
philosophical paraphrase to make traditional metaphysical questions decidable. Then he
drew metaphysical conclusions from the so called “semantic epistemology” based—
according to his programme—on semantics (which played an important role in his
research) and formal logic. His categorial grammar aiming to formulate the general
criteria of syntactic coherence was the first grammar based exclusively on the structural
properties of expressions. He also undertook a number of methodological issues, both
general and detailed. He was interested in the theory of definition, the theory of
questions, the problem of rationality of fallible inferential methods, the foundation of
sentences, classification of reasonings, of sciences and of axiomatic systems, and in the
reconstruction and evaluation of scientific procedures.
Keywords Ajdukiewicz · Meaning-rules · Methodology of science · Radical

conventionalism · Semantic epistemology · Scientific procedures.
Mathematics Subject Classification (2000) Primary 03A10, 03A99, 03B42, 03B65;

Secondary 03B48
1 Life
Kazimierz Ajdukiewicz (1890 Tarnopol–1963 Warsaw) was an eminent representative

of the Lvov-Warsaw School. In Lvov, he studied philosophy, mathematics, and physics.
He was a disciple (and later on, also a son-in-law) of the founder of the School,
Twardowski, but also studied under other teachers, such as Łukasiewicz, Sierpiński, and
Smoluchowski. After having obtained his PhD degree in 1912 (On the Relation between
the Apriorism of Space in Kant and the Question of the Genesis of the Representation
of Space), he deepened his studies in Göttingen, where he had the opportunity to attend
lectures by Husserl and Hilbert. During World War I, he served in the Austrian army, to

48 A. Jedynak
which he was conscripted, and later, in 1920, during the Bolshevik War, as a volunteer
in the Polish army. In 1921, he received his habilitation (From the Methodology of
Deductive Sciences). In the interwar period, he was first a professor at the University
of Warsaw, and then, since 1928, at the University of Lvov. He survived World War II in
Lvov, involved in administrative work, clandestine teaching, and—in periods free of the
German occupation—lectures in scientific institutes. After the war, he was a professor
at the University of Adam Mickiewicz in Poznań (where in the years 1948–1952 he
served as rector) and the University of Warsaw (since 1954). He served many functions
related to the animation and organization of scientific life. He participated in national
and international congresses, organizing some of them; he edited prestigious scientific
journals (“Studia Philosophica” and “Studia Logica”), travelled abroad in connection
with his scientific activities (Great Britain, Austria, USA), was an active member of
the Polish Academy of Sciences and other scientific associations, and established and
led the Department of Logic at the University of Warsaw. In 1962, the University of
Clérmont-Ferrand awarded him an honoris causa doctorate. The University of Adam
Mickiewicz in Poznań had a similar intention, which was not carried out because of
the death of Ajdukiewicz. In the postwar period, Ajdukiewicz and Kotarbiński were
the main representatives of the Lvov-Warsaw School in Poland. They continued the
School tradition of practicing broadly conceived logic, which included semiotics and
methodology, and of using it to resolve philosophical issues. Ajdukiewicz described
himself as a philosopher and a logician (in that order). He initiated and coordinated
many research themes and educated students who later continued this work. A strong
personality, he was considered a great authority among his colleagues and students.
2 Main Publications
The most important publication of Ajdukiewicz’s papers is Language and Cognition,

comprising a selection of his texts from the years 1920–1939 [2] and 1945–1963 [7].
The essential texts from that publication have also been translated into English and
published as The Scientific World-Perspective and Other Essays, 1931–1963 [31]. Results
of the author’s research in methodology were published posthumously as Pragmatic Logic
[12], conceived as a university textbook. His Problems and Theories of Philosophy [11]
remained the main philosophical textbook for several generations of Poles. Both have also
been translated and published in English.
3 Scientific Activity
3.1 Meaning-Rule Conception and Radical Conventionalism
The driving force of young Ajdukiewicz’s philosophical inquiry was his interest in
the cognitive role of language [38]. Inspired by the ideas of French conventionalists,
especially Le Roy, he wanted to present the latter’s free philosophical insights in a strictly
Kazimierz Ajdukiewicz: The Cognitive Role of Language 49
scientific manner. As a result, he created his radical conventionalism. First, however, he

needed a precise conception of meaning, because the question he investigated was how
the choice of language, and the meanings of its expressions in particular, affected the
process and outcome of cognition. For this reason, he took up studies in semiotics [23, 40].
He developed the meaning-rule conception of meaning, with language understood as
a kind of deductive system governed by what he called meaning-rules [20]. These
meaning-rules prohibited the language user to refuse accepting some of its sentences—
either unconditionally (axiomatic meaning-rules), or by virtue of prior acceptance of
other sentences (deductive meaning-rules), or in the face of certain experiential data
(empirical meaning-rules)—otherwise the meaning of expressions of the language would
be violated. Sentences accepted under those meaning-rules constituted, respectively,
axiomatic, deductive, and empirical theses of the language. Ajdukiewicz considered as
synonymous both expressions of the same language, which were interchangeable within
its meaning-rules, providing the latter remained intact, as well as expressions from
different languages, which occupied an analogous structural position, each within the
meaning-rules of their respective languages. In view of this, meaning was the property—
defined by abstraction—common to all synonymous expressions.
Ajdukiewicz proceeded to investigate only the languages he called closed and con-
nected, that is, those in which all expressions were interconnected in terms of meaning
by way of meaning-rules and to which no new meanings could be added since they
already contained all the meanings from the given relationship network. He called the
set of meanings of a closed and connected language a conceptual apparatus, and the set
of theses of such a language a world-perspective. Then he formulated the thesis of radical
conventionalism: a world-perspective depends on the choice of conceptual apparatus [35].
When applied to languages other than closed and connected ones, this conception did
not claim anything unusual: only that sentences accepted by virtue of the meaning of
their expressions depended on those very meanings. Its originality lay in its application
to closed and connected languages and in the assumption that various such languages
existed, based on different conceptual apparatuses. Two such languages (since they could
not be enriched in a connected way) either used identical conceptual apparatuses and
differed only by the sound of their expressions, or used completely different apparatuses
that had nothing whatsoever in common. In the second case, those languages were
completely mutually untranslatable, that is, no expression of either language had a
synonym in the other one. Moreover, a change of meaning of even one expression of
a closed and connected language led to a change of the meanings of all its expressions,
and so a shift to another apparatus. This in turn meant a transition to a fundamentally
different cognitive perspective.
Radical conventionalism was supposed to give an account of the profound trans-
formations in science, later called scientific revolutions. According to Ajdukiewicz,
what changed were not the theories as such but something more fundamental, namely,
languages and associated conceptual apparatuses by means of which theories were
expressed. A shift to another apparatus made the old theory and the new one mutually
untranslatable. Science, therefore, did not develop cumulatively. However, Ajdukiewicz
pointed out some evolutionary trends of conceptual apparatuses, which allowed progress
in science to take place. He abandoned this conception partly due to the shortcomings that
Alfred Tarski had found in the underlying meaning-rule conception of meaning, and partly
because he had concluded that the notion of a closed and connected language was empty,
50 A. Jedynak
and as a result, that radical conventionalism itself did not have a real exemplification. He
continued promoting the main thesis of radical conventionalism in a weakened version
and a different formulation, applying it to ordinary, actually used languages, though as a
result, it lost its clarity and originality.
3.2 Radical Empiricism
While radical conventionalism emphasized the cognitive role of language, in later years
Ajdukiewicz seemed to shift to the position of radical empiricism, minimizing that role.
He pondered whether it was possible to construct a language governed exclusively by
empirical rules of meaning. First, he pointed out the possibility of abandoning axiomatic
rules [21], and then also deductive ones [26]. He concluded that although it had not been
practised before, it was possible to acquire and express knowledge in a language free of all
a priori components, in which even logic would be based on experience. He also noticed
that constructing such a language required a reconstruction of the notion of meaning,
which he managed to sketch only roughly [26]. This evolution of Ajdukiewicz’s views
can, however, be seen differently and interpreted as a shift from conventionalism not so
much to empiricism as to metaconventionalism [37]. This is because we can choose the
kind of language we need: is it supposed to be governed by all types of rules, or only
by empirical ones? This choice determines not the meanings of expressions but the idea
about what meaning actually is, the acceptance not of the sentences in question but of
epistemological theses. Ajdukiewicz also drew close to empiricism by demonstrating that
sentences traditionally considered to be analytical required resorting to experience, and
specifically to existential premises [28]. He thereby initiated a lively debate in Polish
philosophical literature on the notion of an analytic sentence.
3.3 Rationalism, Realism, Classical Logic
Of special importance for Ajdukiewicz were rationalist standards of pursuing philosophy.

He shared the view, common in the Lvov-Warsaw School, that the necessary condition
of cognitive rationality of problems one investigated was their communicability and
intersubjective testability. He coined the term anti-irrationalism, which meant rationalism
in a broad sense, including empirical methods, as opposed to narrow rationalism that
excluded empiricism. He consistently espoused the classical notion of truth, the reality
of the outside world, and also classical logic, upholding the principle of bivalence and
ungradability of truth. He successfully defended the law of non-contradiction against
ideologues who, in the postwar period, resorting to marxist principles, argued that the
universality of movement and changes in the world led to inevitable contradictions [15].
3.4 Semantical Epistemology and Metaphysics
Although the criterion of intersubjective testability resembles the neopositivist criterion

of sense, the Lvov-Warsaw School—in contradiction to the Vienna Circle—did not avoid
traditional metaphysical problems. Ajdukiewicz reconciled metaphysical aspirations with
the criterion of testability by using his method of philosophical paraphrase: his way of
explaining philosophical concepts was meant to make notoriously vague metaphysical
problems, expressed through those concepts, graspable and decidable. He tried to combine
maximalistic aspirations with the “toolbox” of a minimalist. In his view, metaphysics
could be based on epistemology (which he did not see as part of metaphysics). He
formulated and pursued the following programme of semantic epistemology: all cognition
manifested in language, therefore the study of cognition could be brought down to the
study of its linguistic results, i.e. the sentences of the language in which they were
expressed. Those sentences, their mutual connections and relationships with reality, were
the domain of semantics, based on the achievements of logic. And semantics and logic
provided well-founded conclusions. According to his method, only after having arrived at
such beliefs about the nature of cognition could we draw metaphysical conclusions about
the nature of existence.
In the early period of Ajdukiewicz’s work semantics was riddled with antinomies.
Therefore, the meaning-rule conception was not based on semantic concepts. Instead of
the notion of truth, the notion of a thesis of language played an important role in it.
However, Ajdukiewicz did not equate these two notions. On the contrary—as soon as
semantics had been cleared of antinomies, he carefully differentiated them. Therefore, it is
difficult to understand why later he encountered the ideologically motivated charge that he
equated these concepts, and that he based metaphysics on views relativized to language,
thereby allegedly turning out to be an idealist [39]. This criticism (which Ajdukiewicz
did not neglect to respond to [36]) was especially poignant for him because the views
imputed to him were not only fundamentally different from his own but also ones he
himself argued against.
3.4.1 Polemic with Neo-Kantian Objective Idealism
It was precisely for the purpose of this polemic that Ajdukiewicz made a distinction
between the concepts of truth and thesis [32]. The polemic itself was a model example
of his application of the paraphrase method and the use of semantical epistemology
in metaphysics. In his view, the main thesis of idealism—that the world did not exist
independently but was a correlate of an objective spirit or the transcendental subject—
was unclear, and its central idea required an explanation. At the time, idealists conceived
of this superindividual subject not as some higher self but a system of ideas and judgments
dictated by transcendental norms. These norms were supposed to contain the criteria
of truth, while the world as a correlate of transcendence was supposed to depend on
truth. This established an ontic order completely contrary to Ajdukiewicz’s philosophical
beliefs. In his opinion, criteria did not define truth and truth did not define the world.
The truth was (on the basis of a given language) secondary to events in the world, and
the criteria of truth were secondary to truth itself. To make the problem of idealism
52 A. Jedynak
graspable, Ajdukiewicz interpreted transcendental norms as meaning-rules of language,

and judgments constituting the objective spirit as language theses dictated by meaning-
rules. If the notion of a thesis did not differ from the notion of truth, idealism paraphrased
in such a manner would claim that the set of true sentences depended on linguistic rules.
And as a consequence, that the world had to conform to the rules of language. In order to
reject this view, Ajdukiewicz had to demonstrate that the notion of a thesis differed from
the notion of truth. Then language rules would determine the set of theses but not the set
of truths, and thus, not the external world. To do this, Ajdukiewicz resorted to Gödel’s
theorem, according to which in sufficiently rich languages (and such languages are used
in science) one could formulate undecidable problems, that is, such questions that any
answer, as well as its negation, were not theses. At the same time, by virtue of the law
of excluded middle, one of the two contradictory sentences had to be true. Ajdukiewicz
gave an example of an undecidable problem, pointing out that one of the possible answers
(though which one, was not known) was true, but it was not a thesis. He concluded that
not all truths were theses, and that the directives did not exhaust the richness of the world,
therefore idealism was false. If this result may raise doubts, it is only in connection with
the accuracy of the paraphrase of objective idealism.
3.4.2 Polemic with Subjective Idealism
By taking it up, Ajdukiewicz explained in more detail the programme of semantical

epistemology [17]. Now, in his opinion, metaphysical conclusions could be drawn from
reflection on cognition only when the investigated cognitive results were expressed in
the language of semantics. Such reflection then addressed the relations of language to
the outside world, somehow taking this world into account and offering a cognitive
transition to its affairs. This, however, was not possible when such reflection referred
only to intralinguistic relations, or possibly also mental phenomena, but did not touch the
sphere independent of consciousness. There was then no transition to that sphere and one
could not predicate anything about it. According to Ajdukiewicz, this was actually the
kind of language, free of object-related expressions, that subjective idealists used: they
spoke only about ideas and their perception, and this did not justify any beliefs about
the real external world. Moreover, in contradiction to natural language, they redefined
certain expressions (e.g. they conceived of the body as a system of ideas). The resulting
equivocation created the impression that they used objective language, and so their
conclusions about existence—although counterintuitive—seemed justified. In actual fact,
all the reflections of subjective idealists took place in the sphere of experiences and
sensations, without reaching external reality. Therefore, they could not even deny the
existence of that reality. If, however, they uttered such sentences, it meant they were
trying to restore object language. This would not do them much good, however, because it
immediately violated the meaning-rules of the common language. After all, in the face of
certain data, those rules led to the acceptance of such sentences as these are cats, and as a
consequence, deductive rules required the acceptance of such sentences as cats exist. And
so the idealists either did not get involved in the object language at all, and had no means to
unequivocally deny the real existence of things, or did get involved in it but broke its rules
by stating their views. Proponents of the existence of intentional entities were in a better
position than the idealists [24]. This was because intentional existence was ascertained
by empirical criteria (e.g. one had to check what Homer had written about), expressed in
object language, the rules of which allowed distinguishing real existence from intentional
existence.
3.4.3 Polemic with Reism
Ajdukiewicz’s ontological interests also included the issue of universals [25]. Referring
to T. Kotarbiński’s reism, he noted that the ontological version of that view, according
to which things existed but not universals, was incompatible with the semantic version,
understood as a programme of not using apparent names, that is, names other than those of
specific things [5]. This was because if reists wanted to express the negative part of their
ontological argument, they had to use the apparent name universals. Granted, they did
allow apparent names in sentences, from which those names could be eliminated without a
change in meaning, but was this the case here? If the negative argument could be translated
into a sentence only about things, it would be at odds with the reists’ intentions. And if it
could not, the reists would violate their own programme by uttering it. Ajdukiewicz had
an impact on the reception of reism in the Lvov-Warsaw School. His critique was resisted
only by a weakened version of semantic reism, which came down to the search for reistic
substitutes of abstract names.
Moreover, Ajdukiewicz pointed out that natural language did not determine whether
all names constituted one semantic category (and accordingly, whether everything existed
in the same way) or general names differed categorically from individual ones (and
accordingly, whether different entities might exist in different ways). He concluded
that the process of language clarification could develop in various directions, thereby
leading to different concepts of existence and different answers to the question about
the existence of universals. He held against reism not that it had taken avail of one of
those possibilities but that it had ascribed an absolute value to it. He himself preferred to
use a language that made ontological room for universals, even though reistic language
tempted him with the simplicity of some of its solutions. He stressed the necessity of
relativizing ontological considerations to the language in which they were conducted.
After all, depending on the language rules, the word existence could acquire different
shades of meaning. Ajdukiewicz had thereby anticipated the ontology later proposed by
Quine, which came down to the ontological commitments of language. In this sense,
Ajdukiewicz’s ontological interests fit into his broadly conceived research on the cognitive
role of language.
3.4.4 Logic and Natural Language
Nevertheless, Ajdukiewicz did not see in logic the nostrum for all philosophical problems.
He delimited the applicability of pure logic to philosophical issues [22]. For him, this
limit was the necessity of validating each time the paraphrase of a philosophical problem
in the language of logic. On the other hand, he criticized the view that logic and ordinary
language were incompatible, proposing a pragmatic solution to the paradox of material
implication [16]. What this paradox comes down to is that certain implicational sentences,
true by virtue of the logical interpretation of the implication functor, are unacceptable,
54 A. Jedynak
which seems to undermine this interpretation. Ajdukiewicz made a distinction between

what the sentence asserted and what it expressed. He showed through examples that
sometimes a sentence asserted a true state of affairs but expressed such conviction states
of the person uttering them, about which we knew from context that he or she could not
hold them. In Ajdukiewicz’s opinion, that was the reason of the unacceptability of the
aforementioned sentences.
3.5 Investigation of the Syntax and Structure of Expressions
Ajdukiewicz was the creator of categorial grammar and of the notation specifying the
syntactic position of expressions, which he successfully used when addressing semiotic
issues.
3.5.1 Categorial Grammar
The resolution of semantic antinomies resulted in Ajdukiewicz’s work not only in seman-
tic epistemology but also in taking up by him of the issue of syntactic meaningfulness
of expressions. His categorial grammar—historically, the first grammar based exclusively
on the structural properties of expressions—formulates the general criteria of syntactic
coherence [33]. In other words, it indicates what predetermined steps should be taken
with respect to any expression, so that the obtained results automatically show whether
the words in this expression are assembled to form a meaningful whole. Ajdukiewicz
took the notion of semantic category from Husserl and Leśniewski. He considered a
category to consist of expressions that were exchangeable in any meaningful context.
He distinguished categories of sentences, names, and functors. He divided functors
into subcategories, depending on the category that the expressions those functors built
belonged to and on the number and category of arguments those functors operated on.
He developed a notation that assigned to each expression an index of its category. To
sentences and names, he assigned simple indices (in the form of letters s and n), and
to functors, indices resembling fractions, with the numerator containing the index of the
expression that the given functor built, and the denominator containing the indices of all
the arguments. To investigate the syntactic meaningfulness of any expression, one had
to assemble the indices of all its component expressions, and then to “reduce” indices
similarly to reducing fractions, i.e. if any index had its counterpart in the denominator of
another index, both of those homomorphous symbols should be cancelled. An expression
was syntactically coherent if and only if after making all possible cancellations we
obtained a single index. It was not coherent if we were left with more than one index
after all possible cancellations had been made. Thus, all sentences, names, and functors
were meaningful, but not their accidental combinations.
3.5.2 Syntactical Position of Expressions
Apart from categorial indices, Ajdukiewicz eventually proposed another notation, which
unequivocally determined the syntactic positions of all component expressions in a
compound expression [30] (which was not the case with categorial notation). Syntactic
symbols of individual words in a compound expression allowed a recreation of its
structure even if those words were given in a random order. They gave the notion of a
syntactic position a precise meaning, independent of common intuitions. Ajdukiewicz
used this notation, when towards the end of his life he sketched a new conception of
meaning. He assumed that the meaning of a compound expression was a relationship
that ascribed the denotation of each of its component expressions to its syntactic position
within the whole. This was based on a simple idea that to understand an expression,
one needed to know which fragments of reality the individual words of this expression
related to and in what way they had been connected with one another within it. Another
late semiotic idea of his was that synonymous expressions evoked in language speakers
thoughts that were identical in some essential respects, like the object reference of an
expression or its emotional colouration [12, pp. 7–15]. He did not manage to develop this
idea further—exactly which aspects of thought were significant in terms of the meaning
of the experiences eliciting them, and which were not, and whether those insignificant
ones were incommunicable, and thus non-rational. He also used the notation of syntactic
symbols to analyze the relationship between the subject and the predicate of a sentence
[34] and for sketching the method of eliminating intensional expressions [6, 18]. His aim
was not to remove those expressions from the language but to reinterpret them in a way
that would make them extensional. He concluded that the imprecise natural language
allowed this reinterpretation, though it did not require it. His concept was to have the
functor dissemble the expression not into a few compound arguments, but only into
the simplest component expressions. The denotation of a sentence could change when a
compound argument of the main functor changed to another one with the same denotation
but not when arguments were taken to be only the simplest expressions (including their
syntactic symbols within subordinate sentences). An analogous exchange on the lowest
structural level retained the meaning of the whole and its denotation.
3.6 Methodology
Ajdukiewicz valued scientific knowledge highly and devoted a great deal of attention to
it. His scientism was evidenced by reconstruction of scientific procedures, ideas aimed at
improving them, and the postulate to base philosophical methods on scientific ones (which
he himself pursued). He distinguished two disciplines dealing with science: metascience
and methodology [13]. The first one was an exact theory of deductive systems. The second
one belonged to the humanities and dealt with the activities of scientists and the purposes
they seemed to pursue. It also attempted to understand the mechanisms of science and its
developmental trends.
Science requires precise, specialized terminology and the ability to formulate problems
and to find their valid solutions. Hence Ajdukiewicz’s interest in the theory of definition,
theory of questions, and the issue of justification. His texts on methodology, innovative at
56 A. Jedynak
the time of publication, do not differ much from the contents of contemporary textbooks of
broadly understood logic. It had been due to his contribution, however, that these contents
have been accepted, popularized, and developed further.
3.6.1 Theory of Definition
Ajdukiewicz returned to the issue of definition a number of times. He compiled matters

relating to definitions, systematized conceptually the foreground of definition theory, and
created its foundations. Taking the procedures used in formalized theories as a starting
point, he came to the conclusion that not all aspects of the definition theme were reflected
there. He distinguished real definitions (that unambiguously characterize an object),
nominal definitions (that give translation rules), and arbitrary definitions (or meaning
postulates). He pointed out that these three intersecting notions of definitions could not
be contained in a uniform general theory, because they were so different in terms of their
intension that a superordinate genus did not exist for them [1].
It is not possible to determine the general form of even just nominal definitions, as they
have to be relativized to their languages, and the latter differ in terms of meaning rules.
Only some principles of constructing definitions apply to all languages. Ajdukiewicz
distinguished analytic and synthetic definitions. He characterized types of definitions
with different structures. He recapitulated and collected the conditions of correctness and
usefulness of definitions: formal (e.g. conditions of consistency, non-creativity, existence,
and uniqueness), semantic (e.g. adequacy condition), and pragmatic (e.g. avoiding the
fallacy of ignotum per ignotum). He noted that not all kinds of definitions gave the
translation of the defined term directly, but under certain formal conditions they allowed
translatability. He showed how to formulate definitions to ensure translatability.
He did not share the view of nominalists who, denying the existence of species,
considered all definitions nominal. He believed that such an approach left no room for
inquiry into what certain things were, limiting attention to what their names meant.
He noticed that at an early stage of development of a scientific discipline, primarily
criterial definitions were formulated. Then, as research moved deeper into phenomena,
criteriality gave way to grasping the essence of things, understood as those properties
of things, from which, based on natural laws, the other properties could be deduced. In
this way, the conceptual framework was gradually penetrated by knowledge of empirical
origin. This corresponded to Ajdukiewicz’s observations that conceptual apparatuses
tended towards rationalization (i.e. the process of making originally empirical problems
decidable in a purely conceptual sense, by way of language modification) and that
empirical components did exist in sentences traditionally considered analytic, in particular
in arbitrary definitions.
3.6.2 Theory of Questions
Ajdukiewicz proposed a useful terminology that referred to Twardowski’s proposals and

proved convenient for the logical analysis of questions and answers [19]. He defined the
range of the unknown of a question, positive and negative assumptions of questions, as
well as didactic, captious, and suggestive questions. He distinguished decision questions
and complementation questions; questions posed properly and improperly, and proper,
improper, and correcting answers; partial, complete, and exhaustive answers; and direct
and indirect answers.
3.6.3 Rationality of Inferences and Foundation of Sentences
In the time when induction logic was taking its first steps, Ajdukiewicz tackled the
problem of the rationality of fallible methods of inference [29]. He considered a fallible
inference rational when in the long run, actions based on it brought more profits than
losses, that is, the degree of subjective certitude with which the conclusion was accepted
did not exceed the degree of reliability of the scheme of the applied inference. He assumed
the degree of acceptance of the conclusion could be calculated based on behavioral
criteria, and that the degree of infallibility of the scheme was the relative frequency
of achieving true conclusions based on it, providing the premises were true (he was
aware this explanation was incomplete because in practice the data for calculations was
unavailable). He pointed out that whereas previously the rationality of inference had
been based on probability theory and had relied only on a sense of obviousness, he
demonstrated the correctness of such an approach in view of a general pragmatic theory
of rationality, associated with the balance of profits and losses.
However, Ajdukiewicz did not see in this solution a way leading to the formulation of
criteria of justifying sentences. On the contrary, in the field of justification he perceived
still unsolved problems that others did not. He came to the conclusion that the very
notion of justification, central in methodology, was unclear. At the same time he admitted
that in science, justification or rather the sense of justification was a fact. He outlined a
methodological programme aimed at explicating that notion [27]. In his view, even though
methodology had reconstructed various types of induction, we still did not know what
distinguished inferences approved by scientists, since after all, they did not consider each
generalization justified. This would have to be investigated, and based on this, the notion
of justification recreated. As far as deductive sciences were concerned, it would involve
determining what conditions imposed upon assumptions and rules of transformation
would guarantee justification of secondary theorems. From a broader perspective, such
a programme would proceed to a reconstruction of the scientific method and to the
determination of its capabilities and limitations. Ajdukiewicz returned to the issue of
justification many times, taking up more detailed problems.
3.6.4 Classification of Reasonings
He performed a methodologically useful classification of reasonings, combining their

various divisions: into simple and complex ones; into deductive and nondeductive ones,
and among the latter, into rational and logically valueless ones; further, into those based
on inference and those based only on hypothetical derivation of given sentences from
others; into spontaneous and task guided ones, and among the latter, into those directed
by decision questions, by complementation questions, and by the task of performing a
proof. He took a critical view of the earlier classification elaborated by Łukasiewicz and
modified by Czeżowski, which in his opinion missed part of the logical reality [8].
58 A. Jedynak
3.6.5 Classification of Sciences and Axiomatic Systems
Regarding types of inferences, Ajdukiewicz distinguished deductive and nondeductive

sciences, and regarding ultimate premises that were not subject to justification, he
distinguished deductive sciences based on axioms, empirical sciences based additionally
on observational sentences, and the humanities, based additionally on interpretations
of other people’s broadly understood utterances. He specified stages of development
of deductive sciences (preaxiomatic, intuitive axiomatic, and formalized axiomatic).
Actually, the very beginnings of his activity involved reflection on axiomatic systems,
thus paving the way in Poland for the development of deductive disciplines [10]. Only
later did he turn from metascience to methodology (and even though he had considerable
achievements in the latter, in the last years of his life he expressed regret about not having
been involved in it enough earlier on). In the period of his metascientific investigations he
formulated structural definitions of the notions of logical proof and logical consequence,
and got involved in proofs of consistency and the concept of existence in mathematics; he
pointed out how to strengthen the foundations of contemporary predicate calculus in order
to derive traditional syllogistic from it; and he investigated the conditions of reversibility
of premise and consequence.
Later on, he classified axiomatic systems, combining two of their divisions: into
reductive and deductive ones, and into hypothetical (i.e. neutral) and assertive ones [14].
He performed a critical analysis of the ways of justifying axioms in assertive-deductive
systems, and presented the possibility of limiting deductive sciences to uninterpreted,
formalized neutral systems. He saw hypothetical-reductive systems in empirical sciences
that passed from consequences to premises and at the same time aspired to the formulation
of precise theories. Such a methodologically uniform classification partially eliminates the
gap between deductive and empirical sciences.
3.6.6 Reconstruction and Evaluation of Scientific Procedures
Ajdukiewicz reconstructed such procedures as generalization of observations, testing

of hypotheses, formulation of theories, or revision of principles [12]. He distinguished
various types of natural laws, according to the dependencies they described. He devoted
particular attention to statistical laws. For various kinds of fallible reasonings, he
presented a method for calculating the degree to which assumptions increased the
probability of the conclusion on the basis of accepted knowledge. He came up with a
number of interesting theorems about the impact of evidence on the degree of validation
of hypotheses and generalizations. They allowed a comparative evaluation of the degree
of validation of laws, but not an absolute evaluation (and here Ajdukiewicz saw the
possibility of further research). He proved the rationality of strategies usually used in
science for increasing the degree of validation of laws.
He critically examined ways of justification adopted in the empiricist tradition.
He considered justification that appealed to observations unscientific because direct
observation was neither intersubjective nor reproducible (another observation, although
similar, was never the same as the previous one). The scientific method, however,
required intersubjectivity and repeatability. Scientificity began with generalizations that—
paradoxically—were based on an unscientific foundation. However, Ajdukiewicz did
not join the critics of direct justification, who referred to the theory ladenness of
observations. On one hand, he minimized the role of terminological conventions, and on
the other, he emphasized it. He minimized it by demonstrating that for justifying analytic
sentences, conventions required supplementing with existential premises. He emphasized
it by demonstrating that problems in science, which were seemingly solved arbitrarily
(e.g. what time measurement is really accurate?), were based on implicitly accepted
conventions that specified the meaning of certain vague expressions (e.g. a reliable time
measurement device), thereby assuring a precise meaning, graspability, and decidability
of certain problems (e.g. are two given time periods equal?).
Following Twardowski, Ajdukiewicz understood the humanities antinaturalistically.
He categorized them as idiographical ones (e.g. history), nomothetical ones (e.g. psychol-
ogy), and evaluative ones (e.g. cultural studies). The evaluative ones were supposed to
reveal the values being the goal of human actions, and to evaluate the effectiveness of
those actions. In psychology, contrary to the neo-positivists, he rejected physicalism and
psychophysical reductionism but accepted introspection.
3.7 Axiology
Ajdukiewicz did not focus on axiological problems only in the context of methodology of
the humanities. His metaethical views were close to intuitionism, although he considered
the concept of intuition vague and preferred to talk about feeling or conscience [4]. He
considered values objective and knowable through feelings. He objected to defining values
in terms of feelings. This, he wrote, would resemble defining the properties of things in
terms of sense impressions. In his view, empirical properties and values were primary,
and the reception of both secondary. He saw also other analogies between axiological
and empirical cognition: in both cases one began with direct experience and in both cases
the cognitive faculties—whether feeling or the physical senses—were sometimes fallible.
Then experience was generalized and principles (moral or empirical) formulated. After
that, in both areas individual sentences (resp. judgments or descriptions) could be justified
directly or indirectly, by inferring them from principles.
Ajdukiewicz was interested in psychology and human behaviour, and he characterized
a number of important concepts in that area. Prudence and the ability to reflect were the
traits in people he valued more than spontaneous abandon. In terms of seeking satisfaction
in life, he encouraged keeping a healthy balance between the pursuit of specific goals and
the enjoyment of one’s current activity [9]. He performed a thorough examination of the
notion of justice [3].
One of his concerns was education in logic and philosophy. He called for the
promulgation of logical culture and presented concrete ideas how to go about it. He
authored a few logic and philosophy textbooks designed for students of various levels.
60 A. Jedynak
4 Disciples and Continuators
Ajdukiewicz was a great animator of scientific life. He initiated many kinds of research,
sketching programmes of their further development. Some of the problems he had
addressed were later continued by his disciples, many of them prominent in their own
right. He supervised Master’s or PhD dissertations of the following persons: Zygmunt
Schmierer (who did not survive the war and could not develop his interests), Stefan
Swieżawski (metaphysics, history of philosophy), Jerzy Giedymin (methodology of his-
tory and social sciences, conventionalism), Roman Suszko (logic, epistemology) Henryk
Skolimowski (analytic philosophy, axiology), Adam Nowaczyk (logical foundations of
language and cognition). He also had an impact on the scientific development of such
philosophers as Henryk Mehlberg (philosophy of mathematics and of empirical sciences),
Maria Kokoszyńska-Lutmanowa (philosophy of science, methodology), Izydora Damb- ˛
ska (semiotics, history of philosophy), Seweryna Łuszczewska-Rohmanowa (theory of
knowledge, language of science), Janina Hosiasson-Lindenbaum (probabilistic validation
of fallible inferences), Klemens Szaniawski (logic of fallible inferences in the context
of decision theory and the theory of rationality), Halina Mortimer (induction logic and
its history), Jerzy Pelc (logical semiotics), Witold Marciszewski (pragmatic aspects of
cognition, artificial intelligence), Marian Przeł˛ecki (semantic reconstruction of empirical
theories, metaethical intuitionism), and Ryszard Wójcicki (pragmatic reconstruction of
empirical theories).
References
1. Ajdukiewicz, K.: Three concepts of definition. Logique et Anal. 1, 115–126 (1958)

2. Ajdukiewicz, K.: J˛ezyk i poznanie. Tom I. Wybór pism z lat 1920–1939 [Language and Cognition.
Vol. 1. Selected Papers 1920–1939]. PWN, Warsaw (1960)
3. Ajdukiewicz, K.: O sprawiedliwości [On justice]. In: J˛ezyk i poznanie. Tom I. Wybór pism z lat
1920–1939 [Language and Cognition. Vol. 1. Selected Papers 1920–1939]. PWN, Warsaw (1960),
pp. 365–376 (Originally published in: Kultura i wychowanie 2, 109–121 (1939))
4. Ajdukiewicz, K.: Post˛epowanie człowieka [Human conduct]. In: J˛ezyk i poznanie. Tom I. Wybór
pism z lat 1920–1939 [Language and Cognition. Vol. 1. Selected Papers 1920–1939]. PWN, Warsaw
(1960), pp. 317–364 (Originally published in: Propedeutyka filozofii dla liceów ogólnokształcacych,
˛
pp. 156–214. Ksiażnica-Atlas,
˛ Lvov-Warsaw (1938))
5. Ajdukiewicz, K.: Reizm [Reism]. In: J˛ezyk i poznanie. Tom I. Wybór pism z lat 1920–1939
[Language and Cognition. Vol. 1. Selected Papers 1920–1939]. PWN, Warsaw (1960), pp. 79–101
(Originally published in: Przeglad
˛ Filozoficzny 33, 140–160 (1930))
6. Ajdukiewicz, K.: A method of eliminating intensional sentences and sentential formulae. In: Atti del
XII Congresso Internazionale di Filosofia, Venice 1958, vol. 5, pp. 17–24. G.C. Sansoni, Florence
(1960)
7. Ajdukiewicz, K.: J˛ezyk i poznanie. Tom II. Wybór pism z lat 1945–1963 [Language and Cognition.
Vol. 2. Selected Papers 1945–1963]. PWN, Warszawa (1965)
8. Ajdukiewicz, K.: Klasyfikacja rozumowań [A classification of reasonings]. In: J˛ezyk i poznanie. Tom
II. Wybór pism z lat 1945–1963 [Language and Cognition. Vol. 2. Selected Papers 1945–1963]. PWN,
Warszawa (1965), pp. 206–225 (Originally published in: Stud. Logica 2, 278–299 (1955))
9. Ajdukiewicz, K.: Pochwała życia pracowitego [Praise of a busy life]. In: J˛ezyk i poznanie. Tom II.
Wybór pism z lat 1945–1963 [Language and Cognition. Vol. 2. Selected Papers 1945–1963]. PWN,
Warszawa (1965), pp. 77–82 (Originally published in: Wiedza i życie 17(4), 307–312 (1948))
10. Ajdukiewicz, K.: From the methodology of deductive sciences. Stud. Logica 19, 9–46 (1966). Trans.
J. Giedymin (Originally published as: Z metodologii nauk dedukcyjnych. Wydawnictwo Polskiego
Towarzystwa Filozoficznego, Lvov (1921))
11. Ajdukiewicz, K.: Problems and Theories of Philosophy. Cambridge University Press, Cambridge
(1973). Trans. H. Skolimowski and A. Quinton (Originally published as: Zagadnienia i kierunki
filozofii. Czytelnik, Warsaw (1949))
12. Ajdukiewicz, K.: Pragmatic Logic. Synthese Library, vol. 62, Reidel. Dordrecht-Boston (1974).
Trans. O. Wojtasiewicz (Originally published as: Logika pragmatyczna. PWN, Warszawa (1965))
13. Ajdukiewicz, K.: Methodology and metascience. In: Przeł˛ecki, M., Wójcicki, R. (eds.) Twenty-
Five Years of Logical Methodology in Poland. Synthese Library, vol. 87, pp. 1–12. Reidel,
Dordrecht-Boston; PWN, Warsaw (1977). Trans. E. Ronowicz (Originally published as: Metodologia
i metanauka. życie Nauki 6, 4–15 (1948))
14. Ajdukiewicz, K.: The axiomatic systems from the methodological point of view. In: Giedymin, J.
(ed.) The Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp.
282–294. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: Systemy
aksjomatyczne z metodologicznego punktu widzenia. Stud. Logica 9, 205–218 (1960))
15. Ajdukiewicz, K.: Change and contradiction. In: Giedymin, J. (ed.) The Scientific World-Perspective
and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 192–208. Springer, Dordrecht/Boston
(1978). Trans. J. Giedymin (Originally published as: Sprzeczność i zmiana. Myśl Współczesna 8/9,
35–52 (1948))
16. Ajdukiewicz, K.: Conditional statement and material implication. In: Giedymin, J. (ed.) The
Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 222–238.
Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: Okres warunkowy i
implikacja materialna. Stud. Logica 4, 117–134 (1956))
17. Ajdukiewicz, K.: Epistemology and semiotics. In: Giedymin, J. (ed.) The Scientific World-Perspective
(1978). Trans. J. Giedymin (Originally published as: Epistemologia i semiotyka. Przeglad ˛ Filo-
zoficzny 44, 336–347 (1948))
18. Ajdukiewicz, K.: Intensional expressions. In: Giedymin, J. (ed.) The Scientific World-Perspective
(1978). Trans. D. Pearce (Originally published in: Stud. Logica 20, 63–86 (1967). Translated by
O. Wojtasiewicz from an unpublished typescript of a lecture in the United States in 1959)
19. Ajdukiewicz, K.: Interrogative sentences. In: Giedymin, J. (ed.) The Scientific World-Perspective
(1978). Trans. J. Giedymin (Originally published as: Zdania pytajne. In: Logiczne podstawy
nauczania. Nasza Ksi˛egarnia, Warszawa (1934))
20. Ajdukiewicz, K.: Language and meaning. In: Giedymin, J. (ed.) The Scientific World-Perspective and
Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 35–66. Springer, Dordrecht/Boston (1978).
Translated from German by J. Wilkinson (Originally published as: Sprache und Sinn. Erkenntnis 4,
100–138 (1934))
21. Ajdukiewicz, K.: Logic and experience. In: Giedymin, J. (ed.) The Scientific World-Perspective
(1978). Trans. J. Giedymin (Originally published as: Logika i doświadczenie. Przeglad˛ Filozoficzny
43, 3–21 (1947))
22. Ajdukiewicz, K.: On the applicability of pure logic to philosophical problems. In: Giedymin, J. (ed.)
The Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 90–
94. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: O stosowalności
czystej logiki do zagadnień filozoficznych. Przeglad
˛ Filozoficzny 37, 323–327 (1934))
23. Ajdukiewicz, K.: On the meaning of expressions. In: Giedymin, J. (ed.) The Scientific World-
Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 1–34. Springer, Dor-
drecht/Boston (1978). Trans. J. Giedymin (Originally published as: O znaczeniu wyrażeń. In: Ksi˛ega
Pamiatkowa
˛ Polskiego Towarzystwa Filozoficznego we Lwowie, pp. 31–77 (1931))
24. Ajdukiewicz, K.: On the notion of existence: some remarks connected with the problem of idealism.
In: Giedymin, J. (ed.) The Scientific World-Perspective and Other Essays, 1931–1963. Synthese
Library, vol. 108, pp. 209–221. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally
published as: W sprawie poj˛ecia istnienia. Kilka uwag w zwiazku˛ z zagadnieniem idealizmu. Stud.
Philos. 4, 7–22 (1949/50))
62 A. Jedynak
25. Ajdukiewicz, K.: On the problem of universals. In: Giedymin, J. (ed.) The Scientific World-
Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 95–110. Springer,
Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: W sprawie ‘uniwersaliów’.
Przeglad
˛ Filozoficzny 37, 219–234 (1934))
26. Ajdukiewicz, K.: The problem of empiricism and the concept of meaning. In: Giedymin, J. (ed.)
The Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 306–
319. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: Zagadnienie
empiryzmu a koncepcja znaczenia. Stud. Filozoficzne 1, 3–14 (1964))
27. Ajdukiewicz, K.: The problem of foundation. In: Giedymin, J. (ed.) The Scientific World-Perspective
(1978). Trans. J. Giedymin (Originally published as: Zagadnienie uzasadniania. Stud. Filozoficzne 2,
3–16 (1963))
28. Ajdukiewicz, K.: The problem of the foundation of analytic sentences. In: Giedymin, J. (ed.) The
Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 254–
268. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: Zagadnienie
uzasadnienia zdań analitycznych. Stud. Logica 8, 259–272 (1958))
29. Ajdukiewicz, K.: The problem of the rationality of non-deductive types of inference. In: Giedymin, J.
239–253. Springer, Dordrecht/Boston (1978). Trans. D. Pearce (Originally published as: Zagadnienie
racjonalności zawodnych sposobów wnioskowania. Stud. Filozoficzne 4, 14–29 (1958))
30. Ajdukiewicz, K.: Proposition as the connotation of a sentence. In: Giedymin, J. (ed.) The Scientific
World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 348–361. Springer,
Dordrecht/Boston (1978). Trans. D. Pearce (Originally published in: Stud. Logica 20, 87–98 (1967).
Translated by O. Wojtasiewicz from an unpublished typescript of a lecture in the United States in
1959)
31. Ajdukiewicz, K.: In: Giedymin, J. (ed.) The Scientific World-Perspective and Other Essays, 1931–
1963. Synthese Library, vol. 108. Springer, Dordrecht/Boston (1978)
32. Ajdukiewicz, K.: A semantical version of the problem of transcendental idealism. In: Giedymin, J.
140–154. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published as: Problemat
transcendentalnego idealizmu w sformułowaniu semantycznym. Przeglad ˛ Filozoficzny 40, 271–87
(1937))
33. Ajdukiewicz, K.: Syntactic connexion. In: Giedymin, J. (ed.) The Scientific World-Perspective and
Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 118–139. Springer, Dordrecht/Boston
(1978). Translated from German by H. Weber (Originally published as: Die syntaktische Konnexität.
Stud. Philos. 1, 1–27 (1935))
34. Ajdukiewicz, K.: Syntactical connections between constituents of declarative sentences. In:
Giedymin, J. (ed.) The Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library,
vol. 108, pp. 269–281. Springer, Dordrecht/Boston (1978). Trans. J. Giedymin (Originally published
as: Zwiazki
˛ składniowe miedzy członami zdań oznajmujacych. ˛ Stud. Filozoficzne 6, 73–86 (1960))
35. Ajdukiewicz, K.: The world-perspective and the conceptual apparatus. In: Giedymin, J. (ed.) The
Scientific World-Perspective and Other Essays, 1931–1963. Synthese Library, vol. 108, pp. 67–89.
Springer, Dordrecht/Boston (1978). Translated from German by J. Wilkinson (Originally published
as: Das Weltbild und die Begriffsapparatur. Erkenntnis 4, 259–287 (1934))
36. Ajdukiewicz, K.: My philosophical ideas. In: Sinisi, V., Woleński, J. (eds.) The Heritage of Kazimierz
Ajdukiewicz. Poznań Studies in the Philosophy of the Sciences and the Humanities 40, pp. 13–33.
Rodopi, Amsterdam-Atlanta, GA (1995). Trans. R. Harandon (Extensive excerpts from the original
publication: W sprawie artykułu prof. A. Schaffa o moich pogladach ˛ filozoficznych [Concerning
professor A. Schaff’s paper on my philosophical views]. Myśl Filozoficzna 2, 292–334 (1953))
37. Giedymin, J.: Radical Conventionalism, its Background and Evolution: Poincaré, Le Roy, Aj-
dukiewicz. Editor’s Introduction to Giedymin, J. (ed.) The Scientific World-Perspective and Other
Essays, 1931–1963. Synthese Library, vol. 108, pp. xix–liii. Springer, Dordrecht/Boston (1978)
38. Jedynak, A.: Ajdukiewicz. Wiedza Powszechna, Warsaw (2003)
39. Schaff, A.: Poglady˛ filozoficzne Kazimierza Ajdukiewicza [Philosophical views of Kazimierz
Ajdukiewicz]. Myśl Filozoficzna 1, 209–256 (1952)
40. Wójcicki, R.: Ajdukiewicz: teoria znaczenia [Ajdukiewicz: Theory of meaning]. Pró-szyński i S-ka,
Warsaw (1999)
A. Jedynak ()
The University of Warsaw, Warszawa, Poland
On Ajdukiewicz’s Project of the Semantic
Theory of Knowledge
Adam Olech
Abstract Ajdukiewicz’s the project of semantic theory of knowledge comes down

to the statement that reflection upon logically understood concepts and judgments—
which constitute the logically understood knowledge—is equivalent to the reflection
upon terms and sentences whose meanings are those concepts and judgments. The
value of this equivalence lies in the fact that it allows to apply the results obtained
in metamathematics, treating of expressions and sentences, to epistemology, which
pursuant to the mentioned equivalence is also treating of expressions and sentences, and
which—similarly to mathematics—is also a metatheoretical discipline. What is more,
this equivalence allows to maintain the connection of semantic theory of knowledge, that
speaks about expressions and sentences, with traditional theory of knowledge, that speaks
about concept and judgments understood in a logical way.
This equivalence was justified by Ajdukiewicz. Yet in his justification he did not refer
to any philosophy of a language, including the philosophy of language and knowledge.
Not every philosophy of language is a philosophical language reason of this justification,
that is not every philosophy of language is a philosophical language justification of this
justification. The one which is appropriate for this project is, in my opinion, Edmund
Husserl’s intentional theory of expressions from his Logical investigations. Indicating the
appropriateness of this philosophy of language for this project is the main task which I
undertake in this paper.
Keywords Ajdukiewicz’s metaepistemological opinions · Logical understood concepts

and judgments · Linguistic meanings of expressions · Concepts and judgments
determined as to their contents · Husserl’s intentional theory of meaning
Mathematics Subject Classification (2000) Primary 99Z99; Secondary 00A00
This paper is based on a part of research results, which have been presented in my monograph [39].

66 A. Olech
1 Introduction
1.1 The semantic theory of knowledge as the meta-epistemological

proposal and as its implementations
The name “semantic theory of knowledge” was introduced by Kazimierz Ajdukiewicz.

He used it for the first time in a paper titled A Semantical Version of the Problem
of Transcendental Idealism [Problemat transcendentalnego idealizmu w sformułowaniu
semantycznym], in which he wrote that the semantic theory of knowledge is such a manner
of practicing the theory of knowledge
[. . .] in which epistemological problems are programmatically studied from the perspective of
language as a system of expressions endowed with meaning. Its theses are formulated in such a
way that they concern expressions, i.e. sentences and terms, but sentences and terms of a definite
language which endows them with meaning.1 [12, p. 142]
That paper is composed of two parts: a lecture explaining what the semantic theory of
knowledge is, which lecture is de facto a meta-epistemological credo of this philosopher,
and of an implementation of this theory consisting in analysis of the principal thesis of
Heinrich Rickert’s transcendental idealism, and in consequence in refuting this thesis as
being contradictory to Gödel’s first incompleteness theorem. I isolate this lecture from
the entire paper, since it enjoys a specific, meta-epistemological autonomy, and I describe
it as a “meta-epistemological project of a semantic theory of knowledge”. This project
(proposal) and its implementations—since apart from the one indicated above, there were
others—are one of the most profound developments of the twentieth century analytical
philosophy.
The proposal itself, in terms of volume, was presented on two pages of the abovemen-
tioned paper, and is de facto an introduction thereto. That I attach such importance to
these two pages and focus my deliberations around them, results from the fact that I do
not find in the entire philosophical-analytical literature an argumentation which would as
convincingly support the thesis that cognitive questions should be approached from the
language perspective.
Those Ajdukiewicz’s papers which semiotically analyse traditional philosophical
problems, in particular those whose subject is the dispute between epistemological and
metaphysical idealism and epistemological and metaphysical realism are the imple-
mentations of the abovementioned meta-epistemological proposal. If an epistemological
analysis of these problems was described as “semantic theory of knowledge”, and exactly
this name is used in this case, then the proposal of the semantic theory of knowledge can
1 This paper is an extended version of his lecture, which he gave at the Third Polish Philosophical
Congress in Cracow in 1936. A historical-terminological remark: the concept of “the semantic theory of
knowledge” was used for the first time by Ajdukiewicz during that lecture and not in the abovementioned
paper. One needs to remember however that using the term “semantics” in the 1930s, Ajdukiewicz
understood this term in the same manner in which the term “semiotics” is understood contemporarily.
If we therefore abstracted from this historical context and applied the contemporary terminology to
Ajdukiewicz’s semantic-epistemological studies, then we should describe these studies as “semiotic
theory of knowledge”, since Ajdukiewicz’s semantic theory of knowledge engages all three components
of logical semiotics, i.e. syntax, semantics and pragmatics.
On Ajdukiewicz’s Project of the Semantic Theory of Knowledge 67
be described as “an answer to the question, whether a semantic theory of knowledge is

possible, and if yes, under what conditions”.
The semantic theory of knowledge analyses traditional philosophical problems ap-
proaching them from the language perspective, and does so in a manner that makes it
possible to apply logical tools, which is particularly important for a semantic episte-
mologist. Thus, the proposal of the semantic theory of knowledge is an answer to the
question whether it is possible to approach knowledge from the language perspective,
and if it is, how to do it, as knowledge understood in a traditional manner is composed
of the cognitive subject’s cognitive acts and their results. In other words, the proposal of
the semantic theory of knowledge is an answer to the question, whether it is possible to
translate a cognitive act and its result into a linguistic act and result, and it if its, how to
do it, for this translation to be a translation as to the scope, and at the same a translation
which perceives the language as a logically analysed medium.2
I therefore treat the question about the philosophy of language as interesting, or even as
indispensable for the proposal of the semantic theory of knowledge, since the philosophy
of language is fundamental for the discussed proposal, as well as for every meta-
philosophical declaration characteristic for an analytical philosopher proclaiming that
instead of concepts and propositions it is possible and advisable to speak of expressions,
i.e. of names and sentences, whose meanings these concepts and propositions are. When
presenting his proposal, Ajdukiewicz did not legitimize it from the point of view of
philosophy of language and to be precise from the point of view of the philosophy of
language and knowledge understood as cognitive acts and results connected with them,
i.e. as cognitive creations. And I find this question, the question of philosophical-linguistic
legitimisation of the meta-epistemological proposal, to be of key importance, which
is reflected in this article. And a as philosophy of language adequate in this respect
I indicate the philosophy of Edmund Husserl presented in the second volume of his
Logical Investigations,3 which—similarly to the solutions contained in both volumes of
Logical Investigations—influenced the philosophical convictions of many representatives
of the Lvov-Warsaw School, and apart from them—which is obvious—also the most
2 Ajdukiewicz contends, speaking of the semantic theory of knowledge, that speaking of sentences is
equivalent to speaking of propositions, which are the meanings of those sentences—c.f. [12, p. 141].
He therefore contends—I understand his pronouncement this way—that those two ways of speaking
are a translation as to the scope. The context in which the equivalence claim appears, is as follows:
“So, for example, sentences about the relation of consequence or about the relation of inconsistency,
etc. between judgments are equivalent with sentences asserting suitable relations between the sentences
whose meanings are those judgments” [12, p. 141]. The question which I am posing at this point, but to
which I am not giving a definite answers is as follows: Perhaps we should be speaking in this case not
only of a translation as to the scope, but also of a translation as to the content. What motivates me to ask
this question is the fact that the philosophy of language underlying Ajdukiewicz’s theory of knowledge
is—of which I am deeply convinced—the Husserlian intentional theory of expressions. And from the
perspective of this theory it is impossible to speak of the meanings of sentences, without at the same time
speaking of sentences, and the other way round. What is more, it is impossible to speak of meanings of
expressions, i.e. of the contents of acts understood in specie, without speaking of these acts, as well as of
expressions, in which these acts are involved as their meaning intentions.
3 In the English language edition of Logische Untersuchungen Untersuchung I (Investigation I) titled
Ausdruck und Bedeutung [Expression and Meaning] is included in volume one, whereas in the original it
is included in volume two. Referring to the original, I will be thinking of [17].
68 A. Olech
distinguished Husserl’s student, i.e. Roman Ingarden. This is the philosophy that makes it
possible to reasonably speak of a close link between thinking and speaking, which I wish
to particularly emphasize, and provides thereby a philosophical-linguistic legitimisation
for the meta-epistemological proposal. And by legitimizing this proposal, it makes it
possible to demonstrate the links between the traditional theory of knowledge with the
semantic theory of knowledge.
The fact that I perceive Husserl’s philosophy of language as underlying Ajdukiewicz’s
meta-epistemological proposal and its implementation, i.e. underlying the semantic the-
ory of knowledge, is not only an assumption of mine, but—which I try to demonstrate—an
assumption that is fully justified. Although, when justifying his proposal, Ajdukiewicz
does not rely on Husserl’s philosophy of language, nonetheless when speaking of
knowledge determined as to its content and of knowledge undetermined as to its content,
and the concepts of knowledge understood in this way are of key importance for this
proposal, he makes a reference to the “spirit” of Husserl’s deliberations contained in
Logical Investigations, and, which is symptomatic, Ajdukiewicz makes use of a similar
example as was used by Husserl in Investigation V when he spoke of the intentional
and semantic essence of an act, which concepts correspond to Ajdukiewicz’s concept of
“knowledge determined as to its content”.4 Even if in the fragment of his paper which
discusses knowledge determined as to its content (and knowledge understood in this
manner is based on the close link between thought and language) Ajdukiewicz makes
use of Husserl’s establishments in an implicit manner, yet 3 years later in a paper titled
J˛ezyk i znaczenie [Language and Meaning], when writing in another contest about the link
between thought and language, he explicitly relies on Husserl, and to be more precise: on
Investigation I: Expression and Meaning from volume one of Logical Investigations.5
1.2 The nominalistic guise of the meta-epistemological project
The project of semantic theory of knowledge originated not only in the Husserlian and
anti-psychological climate, but also in the nominalist and logical climate. The latter
climate in Polish philosophy was created by such philosophers as Jan Łukasiewicz,
Stanisław Leśniewski, Tadeusz Kotarbiński, Alfred Tarski. And this other climate also
needs to be borne in mind, if ones speaks of the semantic theory of knowledge understood
both as a proposal and as implementations of this proposal, and this climate was
postulated, in order to present philosophical problems in a nominalist form. This is what
4 Fragments of Ajdukiewicz’s and Husserl’s work, which are convergent in this respect, are: [12, pp.
142–143] and [19] (Investigation V: On intentional experiences and their ‘contents’, §20 The difference
between the quality and the matter of an act and §21, The intentional and the semantic essence), pp. 121–
125, as well as [18] (Vol. 2, Part 1 of the German Editions) (Investigation I: Expression and Meaning, §28
Variations in meanings as variations in the act of meaning), p. 223.
5 See [8, p. 147] and/or [15, p. 37]. A sentence from this pages which is worth citing is: “The representation
[of a sentence—A.O.] enters fully into the judgment-process and, indeed, forms its essential part. This
has been convincingly demonstrated by Husserl”. This sentence ends in a footnote referring to E. Husserl
[17, Investigation I: Ausdruck und Bedeutung].
J. Łukasiewicz wrote about the nominalist form of contemporary logic, and what mutatis
mutandis applies to analytical philosophy pursued in a logical manner:
Contemporary logistic has a nominalistic guise. It refers not to concepts and judgements, but to
terms and propositions, and treats those terms and propositions not as flatus vocis, but—having a
visual approach—as inscription having certain forms. In accordance with that assumption, logistic
strives to formalize all logical deductions, that is, to present them so that their agreement with the
rules of inference, i.e., the rules of transforming inscriptions, can be checked without any reference
to the meanings of the inscriptions. [27, p. 198] and/or [32, p. 222]
This is what Łukasiewicz wrote, but then he added that the nominalist form is indeed
only a form, since the language of logic is impossible to be treated consistently in a
nominalist manner, without encountering insurmountable difficulties thereby. Consequent
nominalism requires for example that the language be treated finistically, i.e. it allows
one to operate on a finite number of inscriptions only, despite the fact that both in case
of artificial languages as well as in case of natural languages one needs to take into
account an infinite number, in case of the latter group of languages—at least a potentially
infinite number. Nominalism requires also—let us add—that expressions be treated only
as expressions-specimens, and does not allow expressions-types, yet it is impossible to
manage without the latter.
Łukasiewicz, unlike Leśniewski, Kotarbiński and Tarski, was not always a nominalist,
after all in the 1930s he no longer was one, nonetheless each of these philosophers
and logicians, irrespective of the fact whether he was a factual nominalist, presented
philosophical and logical-philosophical questions in a nominalist way.6 And it is this
presentation made in a nominalist way that makes it possible to analyse these issues with
the use of the contemporary logical apparatus, in which one speaks of expressions, and not
of concepts and propositions, which were handled by traditional logic, and which were
and are handled by the traditionally conceived theory of knowledge.
If we were to elaborate on this compactly put postulate, ordering us to present
philosophical questions in a nominalist form, and to adjust it to the needs of epistemology
understood in an anti-psychological manner, this postulate could be as follows:
speaking of products of cognitive acts needs to be replaced by speaking of expressions, including
of sentences of language L, since the logical-linguistic ways of understanding them are not
entangled in disputes, in which products of cognitive acts constituting the meanings of expressions,
i.e. logically understood concepts and propositions, are entangled. What is more, such presentation
of the issue is heuristically promising, since it provides a possibility to use in epistemology results
obtained in meta-mathematics, i.e. in a meta-theoretical discipline analogous to epistemology. And
6 At some point Łukasiewicz was a nominalist, however his nominalism resulted from his philosophical
immaturity. This is what he wrote on nominalism in an article [28] from 1937: “I will quite frankly
admit that if someone had asked me not that long ago whether as a logistician [i.e. a logician—A.O.] I
professed to nominalism, I would have given an affirmative answer without hesitation, since I had not
been giving much thought to the nominalist doctrine itself, and I only focused on logistic practice. And
so, logistics aspires to the greatest accuracy possible, and this accuracy can be attained by construction
of a language that would be as precise as possible.[. . .] [Yet] is this strive for language precision and
formalisation nominalism already? I do not believe so. Logistic would assume a nominalist standpoint, if
it treated names and sentences only as signs of a certain form, without caring whether these signs meant
anything. Logistic would then become a study of some ornaments or figures, which we draw and order
in accordance with certain rules, playing with them as if with a game of chess. It would be impossible
for me to accept this view. [. . .] I could no longer assume today a nominalist standpoint in logistic. But
I am saying this as a philosopher and not as a logician. Logic cannot resolve this problem, since it is not
philosophy” ([29, pp. 212–214] and/or [31, pp. 239–241]).
70 A. Olech
it is justified to use these results, if one demonstrates that the products of cognitive acts, described
by non-psychologically understood epistemology, are meanings of expressions of language L, in
which the cognitive subject verbalizes his cognitive acts. In such case, speaking of expressions of
language L, in which cognitive acts are verbalized, will be tantamount to speaking of products of
these acts.7
I contend that the proposal of the semantic theory of knowledge was an implementation
of the above postulate. What is more, it was the fullest and the most meaningful
implementation. Contending the above, I do not forget that Ajdukiewicz was at the same
time far from factual nominalism, which is demonstrated by Ajdukiewicz’s appreciation
that he always had for the Husserlian theory of language and his repeated defence of the
universals (Cf. the same author on this: [1, 7, 13]).
2 Presentation of the Meta-Epistemological Project

of the Semantic Theory of Knowledge and Analysis Thereof
2.1 On the necessity to approach knowledge from the perspective

of language
The theory of knowledge can be pursued as a theory of cognitive acts or a theory of objec-
tive cognitive contents. Objective contents can be understood in various ways—Kazimierz
Twardowski understood them as products of cognitive acts, Bernard Bolzano—as Begriffe
and Sätze ‘an sich’, Gottlob Frege—as Sinne and Gedanken, Edmund Husserl—as
expression meanings (ausdrückliche Bedeutungen), and also as meanings ‘in themselves’
(Bedeutungen ‘an sich’), and others—as creations of an objective sprit.
If the theory of knowledge is to be a theory of objective and at the same time
determined cognitive contents, a theorist of knowledge must approach knowledge from
the perspective of language, and this means that he needs to treat logical concepts and
propositions, which comprise objective cognitive contents, as linguistic meanings of terms
and sentences. And the theory of knowledge understood in such way is identical to the
semantical theory of knowledge, which does not directly describe such concepts and
propositions, but describes them indirectly, since it directly describes expressions, whose
logical meanings are those concepts and propositions.8
7 To emphasize the significance which I attach to these contentions, I am writing down this postulate using
a different font.
8 The contents described in this paragraph is the matter of the meaning intention act together with the
quality of this act. These two components of the act understood in such way constitute its meaning
essence, and the meaning essence understood in specie is the logical meaning of the expression—the
expression in which this act is involved. The meaning essence of an act ascribing meaning to an expression
is an immanent content of the act, i.e. the subjective, empirically-real content. Yet this essence understood
in specie, i.e. the product of its idealising abstraction, gives us the ideal content, i.e. logical, objective
content. I will return to the issue which I am signalling right now at the end of this chapter. And the fact
that when I speak of cognitive contents I understand them in the Husserlian way is justified by the fact
that more or less clearly this is the way they were understood by Ajdukiewicz, as well as by the fact
that such understanding of cognitive contents makes it possible to justify the strict connection between
That we need to approach knowledge from the perspective of the language if we

want to speak of knowledge determined as to its content, results from the fact that it
is impossible to name a certain cognitive content, i.e. some designated logical concept or
proposition, other than by characterising this concept or proposition as a linguistic (log-
ical) meaning of a certain term or sentence. Relativization of concepts and propositions
understood in such way, results in the fact that the theory of knowledge describing them
is not a theory of knowledge is general, i.e. it is not a theory of concepts and propositions
‘hanging in the air’, as father J.M. Bocheński called them, but is a theory of knowledge
determined as to its content. And if knowledge in logical sense is understood as identical
to language meanings of expressions, then each sentence which asserts something about
concepts and propositions in a logical sense, corresponds to an equivalent sentence
which asserts something about terms and sentences, whose meanings these concepts and
propositions are, i.e. each sentence of the traditional theory of knowledge, describing
knowledge understood production-wise, corresponds to an equivalent sentence of the
semantic theory of knowledge.9
Ajdukiewicz in the following manner justifies the necessity to approach knowledge
from the perspective of language, i.e. he justifies in the following manner what I call the
“proposal of a semantic theory of knowledge”—below I quote extensive fragments of this
justification, which are at the same time the essence of the semantic theory of knowledge,
i.e. the thing which is the subject of analysis in this paper:
[. . .] It is impossible to name a given concept or judgment except by characterizing them as the
meanings of certain terms or sentences. Prima facie, the contrary appears to be the case, namely
that one does not have to study concepts and propositions indirectly through language. For to refer
to certain determinate cognitions (with definite content) one uses such phrases as, for example, ‘the
concept of a triangle’, ‘the proposition that 2×2 = 4’, etc. In such phrases one does not mention—
or so it appears—any expressions whose meanings would be those cognitions. This is an illusion,
however. In a phrase like ‘the concept of a triangle’, provided it is to serve as a singular name of a
determinate concept, the word ‘triangle’ is not being used in the usual way as, for example, in the
sentence ‘a triangle is a plane figure’, where ‘triangle’ denotes a class of geometrical figures; in
the phrase ‘the concept of a triangle’ the word ‘triangle’ occurs in suppositione materiali, i.e. as its
own name. For if we were to regard ‘the concept of a triangle’ as containing ‘triangle’ in the normal
supposition, i.e. as a name of triangle, then that expression would have, from this point of view, the
same syntactic structure as, for example, ‘John’s father’. ‘John’s father’ denotes the only object,
which stands in the fatherhood-relation to John. Similarly, other expressions of the same structure,
provided they are singular names, denote a unique object which stands in a certain relation to
the object denoted by the term in the genitive case. The relation in question is indicated by the
noun in the nominative case. (In Russell’s logical symbolism, so-called descriptive function ‘R’x’
corresponds to expressions of this sort.) One may say, therefore, that John’s father is identical with
the (unique) object which stands in the relation of fatherhood to John. Now if the ‘concept of a
triangle’ were to be regarded as having the same syntactical structure as ‘John’s father’, then one
would have to say that ‘the concept of a triangle’ denotes the only object which stands to triangle
thought and language, i.e. the strict connection between knowledge determined as to its contents with the
logical (language) meaning of expressions, which I have already described in the Introduction.
9 The traditional theory of knowledge is both about cognizing and knowledge, i.e. it is both about cognitive
acts, as well as about products of these acts (speaking the language of actions and products of Kazimierz.
Twardowski). If we—in the name of anti-psychologism—limited the subject of study of the theory of
knowledge to products of cognitive acts, then we would receive the abovementioned equivalence of the
traditional and at the same time anti-psychologically oriented theory of knowledge (speaking of concepts
and propositions in the logical sense) and the semantic theory of knowledge (speaking of expressions of
language L, whose meanings these concepts and propositions are).
72 A. Olech
in the relation of being its concept or—in other words—that the concept of a triangle is identical
with the only object which stands to triangle in the relation of being its concept. By analogy, if
we consider in a similar fashion ‘the concept of a trilateral figure’ one would have to say that the
concept of a trilateral figure is identical with the only object which stands to the trilateral figure
in the relation of being its concept. However, a triangle is the same as a trilateral figure. If, on
the other hand, a is identical with b, then whatever the relation R, the only object which stands
in the relation R to a is identical with the only object which stands in the relation R to b. Therefore,
the only object which stands to a triangle in the relation of being its concept, i.e. the concept of a
triangle, would be identical with the only object which stands to a trilateral figure in the relation of
being its concept, i.e. the concept of a trilateral figure. This is not the case, however; the concept
of a triangle is not identical with the concept of a trilateral figure (nota bene, if the expressions
used here, ‘the concept of a triangle’, and ‘the concept of trilateral figure’ are to serve as singular
names of certain determinate concepts and not as universal terms denoting classes of concepts
→
−
which correspond to the symbolic schema ‘ R ’x’).
In order to talk about a determinate concept of a triangle, we must not use ‘the concept of
a triangle’ as if the term ‘triangle’ in the normal supposition (as the name of triangle) occurred
in it. Rather we ought to understand that term as containing the word ‘triangle’ in the material
supposition, i.e. as the name of itself. Accordingly, expressions like ‘the concept of a triangle’—
if they are to be singular names of determinate concepts—ought to be used as abbreviations of
expressions such as “the concept which constitutes the meaning of the term ‘triangle’ ” i.e. as
expressions each characterizing its designatum as an object which is the meaning of the term
‘triangle’. If we want to name a determinate concept we ought, therefore, to write either “the
concept which constitutes the meaning of the word ‘triangle’ ” or—briefly—“the concept of
‘triangle’ ” keeping in mind that the word ‘triangle’ is used here in suppositione materiali, i.e.
as the name of itself. This does not mean that ‘the concept of a triangle’ with ‘triangle’ in the
normal rather than material supposition, is not a grammatically correct expression. It is; one has
to bear in mind though that then it is not a singular name of a determinate concept, but rather a
universal term to whose extension belong all concepts which constitute the meanings of the terms
co-extensive with the term ‘triangle’.
The semantic theory of knowledge is thus intentionally on the same path which had been
followed by the epistemologists (without their realizing it) whenever they talked about certain
determinate concepts, judgments, etc., or, which should have been followed by them had they
expressed their ideas more precisely. [12, pp. 142–143] and/or [5, pp. 266–267]
2.2 The analysis of the necessity to approach knowledge

from the perspective of language
Below I present an analysis and a commentary to the above justification, which I will
begin with elementary, but—as it will prove in the course of the analysis—indispensable
distinctions:
1. The designatum of the concept of a triangle is a triangle. The designatum of the concept
of a trilateral figure is a trilateral figure, and a triangle is the same as a trilateral figure.
But:
2. The designatum of the name “concept of a triangle” is any concept referring to a
triangle, and there are many such concepts. One of them is the concept of a triangle
constituting the meaning of the name “triangle”.
The designatum of the name “concept of a trilateral figure” is any concept referring
to a trilateral figure, and there are many such concepts. One of them is the concept
constituting the meaning of the name “trilateral figure”.
Both the first and the second name are general names—each of them has more than
one designatum.
3. The designatum of the name “concept of ‘triangle’ ” is the concept of a triangle being
the meaning of the name “triangle”, and there is only one such concept.
The designatum of the name “concept of ‘trilateral figure’ ” is the concept of a trilateral
figure being the meaning of the name “trilateral figure”, and there is only one such
concept.
Both the first and the second name are singular names—each of them has one
designatum.
However:
4. The denotation of the name “concept of a triangle” is a set of concepts of the same
scope as the name “triangle”.
The denotation of the name “concept of a trilateral figure” is a set of concepts of the
same scope as the name “trilateral figure”,
and therefore: these names have the same denotation, which is a multi-component set.
5. Denotation of the name “concept of ‘triangle’ ” is a one-component set, which is the
concept of a triangle being the meaning of the name “triangle”.
Denotation of the name “concept of ‘trilateral figure’ ” is a one-component set, which
is the concept of a trilateral figure being the meaning of the name “trilateral figure”,
and therefore: these names have different denotations.
I wish to emphasize a material difference between point (1) and points (2) and (3): in
point (1) one speaks of the designatum of the concept of a triangle, whereas in points (2)
and (3) ones speaks of the designata of the names: “concept of a triangle” (“concept of
a trilateral figure”) and “concept of ‘triangle’ ” (“concept of ‘trilateral figure’ ”). There
also is a difference between point (2) and point (3), which is demonstrated by different
use of the word “triangle”: in the first case the word is used in the normal supposition—in
point (2), and in the second case it is used in suppositione materiali—in point (3). And
this makes the names in point (2) general names, i.e. names with multiple designata, and
each of the designata is a concept referring to a triangle, but the names in point (3) are
singular names—their designata are elementary concepts,10 each of which is the meaning
of the name put in the internal quotation marks. I will return to these distinction is further
analyses, for now I will express them in the shortest manner possible:
The name “triangle” designates a triangle, the concept of a triangle also designates
a triangle, but the name “concept of a triangle” designates any concept referring to a
triangle, and the name “concept of ‘triangle’ ” designates one concept, the concept of a
triangle, which is at the same time the meaning of the name “triangle” and which refers
to a triangle.
10 In this case I am speaking of elementary concepts, and not of singular concepts. I call them elementary
concepts since they are elements of sets, which in this case are general concepts. And so, for example the
concept of a “triangle” being the designatum of the name “concept of ‘triangle’ ”, is an element of a set
of concepts with the same scope, i.e. of concepts each of which refers to a triangle. We cannot however
say about the concept of a triangle that it is a singular concept. Singularity of a concept means, this is
the language usus in this respect, that the concept has one designatum only. Yet, any triangle, and there
are many of them, is a designatum of the concept of a “triangle”. With respect to the name “concept of
‘triangle’ ” we can indeed say that it is singular, since there exists only one such concept, which is the
designatum of the name “concept of ‘triangle’ ”, and which at the same time is the meaning of the name
“triangle”.
74 A. Olech
Let us therefore use the above distinctions in commentaries and analyses of the quoted
justification: If we are anti-psychologists, then following Ajdukiewicz, we differentiate
concepts in the psychological sense from concepts in the logical sense, as well as—
propositions in the psychological sense from propositions in the logical sense; and bearing
in mind the distinctness of both types of concepts and of both types of propositions, we
include both the former and the latter into the scope of epistemological deliberations.11
The former, i.e. the concepts, are respectively—psychological and logical meanings of
nominal expressions; the latter—are psychological and logical meanings of sentences.
By restrictive understanding of cognitive acts, i.e. by such understanding of cognitive
acts which classifies only verbalized cognitive acts as acts deserving to be called
“cognitive” and by acceptance of the Husserlian philosophy of language,12 concepts in
the psychological sense are identical to acts of meaning intention involved in intuitive
presentations of nominal expressions, whereas propositions in the psychological sense are
identical to acts of meaning intention involved in intuitive presentations of propositions.
And consequently, concepts in the logical sense are identical to understood in specie
intentional essences of acts involved in intuitive presentations of nominal expressions,
11 The direct subject of deliberations of the semantic theory of knowledge are expressions whose logical
meanings are concepts and propositions understood logically, and thus an indirect subject of deliberations
of this theory are those propositions and concepts—deliberations concerning these propositions and
concepts are equivalent to deliberations concerning expressions. The semantic theory of knowledge
programmatically narrows the object of its interest to knowledge understood logically. This does not
mean however that the author of the semantic theory of knowledge eliminates from the scope of
epistemological deliberations knowledge understood psychologically, i.e. concepts and propositions
understood psychologically. He never did so. In any case, it is impossible to speak of logically understood
concepts and propositions in abstraction from their psychological counterparts, since they are, these
logical units, ideally understood meaning essences of the acts of conceiving and judging involved
in nominal expressions and sentences. This is how they were described by the author of Logical
Investigations, and following this author, i.e. Husserl, Ajdukiewicz understood them in the same way. And
the fact that he explicated them in a syntactic-pragmatic manner in his directive concept of knowledge
is a separate matter, which is not mutually exclusive with the former. One should add one additional
remark to the above: despite the fact that the semantic theory of knowledge directly describes expressions
equipped in meaning, which means it is pursued based on language interpreted intentionally (but after the
breakthrough caused by the semantic works of Tarski, the author of the semantic theory of knowledge
included in a significant manner the semantic aspect of language into his conceptual apparatus), it needs
to be born in mind that the author of the semantic theory of knowledge treats linguistic expressions
as three-layer objects: the physical sign—the act of meaning intention involved in that act (being the
psychological meaning of the expression)—the logical meaning (being the ideal understanding of the
intentional essence of a meaning generating act). And since it is impossible to speak of an expression
without the meaning intention act, it needs to be stated that this is a subjective-objectivist view of language
expressions. The objectivist component is the expressed sign and the logically understood meaning; the
subjective component is the act of meaning intention involved in the expression sign.
12 Speaking of the Husserlian philosophy (concept) of language (expression or meaning) in this article, I
always mean the one which this philosopher presented in Logical Investigations. One needs also to bear
in mind that Ajdukiewicz accepted this philosophy throughout the entire period of his academic activity,
and not only at the time when after his doctoral thesis he was going in 1913 for Göttingen, to Husserl and
Hilbert, being under the impression of Logical Investigations, but also at the time when he was writing
his last work [10], an unfinished work, interrupted by the author’s death. ([10] was prepared for print by
H. Mortimer and K. Szaniawski, Foreword was written by K. Szaniawski, the work was published 2 years
after the author’s death.) I have described the fact that Ajdukiewicz accepted the Husserlian concept of
the language also as the author of [10] in my articles, e.g. [35, 40].
whereas propositions in the logical sense are identical to understood in specie intentional
essences of acts involved in intuitive presentations of sentences.
Bearing in mind the above, and limiting oneself to the example of the ambiguous
name “concept of a triangle” (written down without caring for potential internal quotation
marks), one is able to indicate the following meanings, which are connected with this
name:
1. If in the name “concept of a triangle” the word “triangle” is in normal supposition, i.e.
if it is not put in quotation marks, then this name has the following form: “concept of
a triangle”, and as such it is a general name.
2. If in the name “concept of a triangle” the word “triangle” is in suppositione materiali,
i.e. if this word in put in quotation marks, then this name has the following form:
“concept of ‘triangle’ ”, and as such it is a singular name.
Ad (1) Similarly as in the case of other names, also in this case we differentiate the
objective references of names: “concept of a triangle” and “concept of ‘triangle’ ”, from
the meanings of these names. Both the objective references of these names—and when
saying this I mean their designata—as well as their meanings are concepts, however
these are not the same concepts.13 Let us deal with the objective references of these
names, since they, and not their meanings, are possible to be taken into account, when we
analyse the question of knowledge both determined and undetermined as to its content,
i.e. the question whose purpose it is to legitimize the principal thesis of the semantic
theory of knowledge. And this thesis proclaims that instead of concepts and propositions
understood logically, we can speak of expressions, whose meanings these propositions
and concepts are. What is more: we not only can, but we ought to speak of expressions,
if we wish to discuss knowledge determined as to its content. And this is tantamount to
the assumption that we ought to speak of those concepts and propositions, which are the
meanings of these expressions.
The name “concept of a triangle” is a general name, and its designatum is every concept
which refers to a triangle, i.e. every concept whose designatum is a triangle; and there are
many such concepts. The designatum of the name “concept of a triangle” is therefore:
the concept of “triangle”, the concept of “trilateral figure”, the concept of “polygon with
the sum of internal angles equal to 180◦” etc. When, however, we speak of a concept, we
could have in mind a concept in the psychological sense or a concept in the logical sense;
thus the name “concept of a triangle” is not only a general name, but at the same time an
ambiguous name—each of its meanings indicates, as its scope, a set of concepts in the
psychological sense or a set of concepts in the logical sense.
A concept in the psychological sense is an act of consciousness of specified quality and
specified matter, i.e.—in case of the latter—of specified sense of objective understanding.
Unity of the both of them, i.e. unity of the quality and of the matter of the act, is the
13 That concepts are meanings of these names results from the fact that the meaning of each name is a
concept. However, the meanings of these names are concepts of a higher tier as compared to concepts
being the designata of these names. And so, the name “concept of ‘triangle’ ” has a meaning which is a
concept, and it has a designatum which also is a concept. The latter, i.e. the designatum of the “notion
of ‘triangle’ ” is the meaning of the name “triangle”; the former is the meaning of the name “notion of
‘triangle’ ”. And thus the meaning of the name “concept of ‘triangle’ ” is a concept of a concept, to be
precise: a concept of a concept of a “triangle”.
76 A. Olech
intentional essence of the act, whereas a concept in the logical sense is the intentional
essence of the act understood in specie or to put it in other words: its ideal abstraction.
Speaking of concepts in the psychological sense, I obviously mean acts ascribing meaning
to nominal expressions, i.e. acts of meaning intention involved in intuitive presentation of
these expressions; the intentional essences of these acts are meaning essences of these
acts. These essences understood in specie, i.e. the logical meanings of expressions—
irrespective of the fact whether this is a nominal or a propositional expression14—are
described by Husserl with the general name of “expression meanings” (ausdrückliche
Bedeutungen) and juxtaposed with meanings ‘in themselves’ (Bedeutungen ‘an sich’),
which have not ‘happened’ to be meant by any expression.
Let us call the ideal abstractions of meaning essences of acts ascribing meaning to
expressions—“ideal contents”.15 We will then say that if we spoke of the name “concept
of a triangle”, and took into consideration its logical meaning, then the designata of this
name would be ideal concepts, i.e. ideal (logical) concepts, such as: concept of a triangle,
concept of a trilateral figure, concept of a polygon with the sum of internal angles equal
to 180◦ , etc., each of which refers to a triangle. In other words, speaking of the name
“concept of a triangle”, we would have in mind the plurality of its ideal contents. Thus to
the question: what do we have in mind when we speak of the name “concept of a triangle”
we should answer: we have in mind the plurality of these ideal concepts. And if this is so,
then the name “concept of a triangle” is not determined as to its content or, to put it in
other words, it is not specified as to its content.16
All of these ideal contents, which comprise the plurality of ideal contents, can be linked
to a relevant nominal expression, and then the general name “concept of a triangle” will
‘break up’ into a plurality of singular names, each of which is determined as to its content.
This linking will consist in that, that a particular ideal content, by being linked to a relevant
expression, shall become the meaning of this expression. It therefore turns out that the
abovementioned plurality of ideal contents is nothing else than the plurality of potential
meanings of expressions determined as to their content.
What has been said so far with respect to this issue, can be expressed in the following
manner: If in the name “concept of a triangle” the expression “triangle” is used in normal
supposition,17 then the name “concept of a triangle” will not be a singular name of the
14 I understand a propositional expression as an expression whose logical (language) meaning is a
proposition in the logical sense.

15 This is exactly how E. Husserl describes them (and he adds that they can be also described as logical
or intentional contents, as an intending sense or sense, or simply as meaning), when he juxtaposes them
with subjective (immanent, empirically-real) contents, i.e. those on which the idealizing abstraction is
carried out—cf. [18, pp. 196, 200, 202]. This immanent content would be meaning in the psychological
sense (psychologically understood concept or proposition), whereas the ideal (logical) content would be
meaning in the logical sense (a logically understood concept or proposition).
16 An analogous reasoning can be carried out, if speaking of the name “concept of a triangle” we are
taking into consideration its psychological meaning as well. Then instead of concepts in logical sentence,
i.e. of ideal contents of cognitive acts, we will be speaking of concepts in the psychological sense, i.e. of
immanent contents of these acts.
17 In [12] which is analysed in this point of the deliberations, Ajdukiewicz uses two types of suppositions:
the suppositione materiali and the normal supposition, and then—making a reference to the normal
supposition—he uses the name “ordinary” and makes no comment on this. I believe that Ajdukiewicz’s
use of the name “ordinary supposition”, in the context of the previously introduced—in the analysed
concept being the meaning of the name “triangle”, but will be a general name, denoting
an entire class (set)18 of elementary concepts, each of which is a meaning of the name
equivalent to “triangle”, such as for example the name: “trilateral figure”, “polygon with
the sum of internal angles equal to 180◦”, etc.
Und thus, if an elementary concept being the meaning of the name “triangle” is the
designatum of the singular name “concept of ‘triangle’ ”, then the designata of the general
name “concept of a triangle” are concepts being the meanings of a possible complex name
in the form of: a1 ∧ a2 ∧ . . . ∧ an , where:
1. any ak is the symbol of a name designating a partial content, comprising the full
content of the general concept of a triangle;
2. symbol “∧” is used to denote conjunction understood as a name-forming functor with
name arguments;
3. n belongs to the set of natural numbers, whereas k is greater or equal to 1 and smaller
or equal to n.
An example of such a complex name is: “a triangle and a trilateral figure and polygon
with the sum of internal angles equal to 180◦ and . . .”. Account taken of the number of
elementary concepts comprising the scope of the name “concept of a triangle” which is
nearing infinity, i.e. the full content of the general concept of a triangle, also the number
of components of the complex name in the form of: a1 ∧ a2 ∧ . . . ∧ an will be nearing
infinity.
Conclusion: the name “concept of a triangle” does not mean, unlike the name “concept
of ‘triangle’ ”, a concept determined as to its content, that is—in other words—the concept
of a triangle is not, unlike the concept of “triangle”, a concept determined as to its content.
And saying this, we bear in mind that the name “concept of ‘triangle’ ” is an abbreviated
form of the name “concept being the meaning of the name ‘triangle’ ”.
Ad (2) The name “concept of ‘triangle’ ” is a singular name. Its sole designatum is
the concept being one of the ideal concepts comprising the abovementioned pluralities
of ideal concepts—to be precise: it is the ideal content connected with the expression
“triangle” and being the meaning of this expression. And since it is linked with the
expression “triangle”, therefore the fact of this link needs to be marked, by putting this
word into special quotation marks, which—being an integral part of this expression—
cause this expression to occur in suppositione materiali. Therefore, the name “concept of
‘triangle’ ” is de facto an abbreviation of the expression “concept being the meaning of
the expression ‘triangle’ ”, and the name “concept of ‘trilateral figure’ ” is an abbreviation
of the expression “concept being the meaning of the expression ‘trilateral figure’ ”, and
the name “concept of ‘polygon with the sum of internal angles equal to 180◦ ’ ” is an
abbreviation of the expression “concept being the meaning of the expression ‘polygon
with the sum of internal angles equal to 180◦ ’ ”, etc. And thus names: “concept of
paper—distinction into suppositione materiali and the normal supposition, means that when he speaks
of the normal supposition and he means the one which was described by Piotr Hiszpan as suppositio
personalis, and which in a different terminology has been described as suppositio simplex, i.e. as regular
(or ordinary) supposition or as simple supposition.
18 In this paper I am using these terms interchangeably. Ajdukiewicz speaks in this case of a class of
concepts (cf. the extensive fragment of his justification cited above concerning the indispensability of
approaching knowledge from the language perspective).
78 A. Olech
‘triangle’ ”, “concept of ‘trilateral figure’ ” etc. are names whose designata are concepts
determined as to their content, i.e. such concepts as: concept of “triangle”, concept of
“trilateral figure”, etc. are concepts determined as to their content, but such names as:
concept of a triangle, concept of a trilateral figure are not.
2.3 Husserl and the issue of being determined as to content

(definite in content)
There is a passage in Logical Investigations pertaining to the issue of being determined as

to content, which corresponds to what has been discussed above, but which moreover has
a certain more general value—both ontologically and epistemologically. It is worth citing
this passage, here it is:
Everything that is, can be known “in itself”. Its being is a being definite in content [ist inhaltlich
bestimmtes Sein—A.O.], and documented in such and such “truths in themselves”. What is, has
its intrinsically definite properties and relations [. . .]. But what is objectively quite definite, must
permit objective determination, and what permits objective determination, must, ideally speaking,
permit expression through wholly determinate word-meanings. To being-in-itself correspond
truths-in-themselves, and, to these last, fixed, unambiguous assertions. Of course, to be able to
say all this actually, would require, not merely the necessary number of well-distinguished verbal
signs, but a corresponding number of expressions having precise meanings—in the strict sense
of expressions. We must be able to build up all expressions covering all meanings entering into
our theory, and to identify or distinguish such meanings with self-evidence. [18, p. 223] and/or
[17, p. 90]
This passage can be commented on in the following manner: Also with respect to a
general concept, for example with respect to the general concept of a triangle, it is possible
to say that it is determined as to its content, however this is a determinateness of another
kind, namely a determinateness in itself and only in itself. On the other hand, with respect
to an elementary concept, e.g. with respect to the elementary concept of “triangle” or with
respect to the elementary concept of “trilateral figure”, we will say that it is a concept
specified as to its content for cognitive subject S speaking language L, and a cognitive
subject always speaks some language. What is more, one is a cognitive subject, only if at
the same time one is a linguistic subject—this is what an analytical philosopher would say.
And this means that one is a cognitive subject, if his cognitive acts are at the same time acts
involved in intuitive presentations of word creations, that is if his cognitive acts are acts
ascribing meaning to expressions. This issue was understood in this way by Ajdukiewicz,
who at the same time claimed that a convincing, phenomenological description of this
issue had been provided by Husserl in the second volume of Logical Investigations.19
19 See [14, p. 37, footnote 2]. This footnote refers to the abovementioned Investigation titled Expression
and Meaning from Husserl’s Logical Investigations. And this is this fragment from Language and
Meaning, which ends in a reference to this investigation and which is written entirely in accordance
with the spirit and the letter of the Logical Investigations: “Articulate judging takes place mostly (if
not always) in reference to voiced or unvoiced speaking, viz. reading, writing, listening, etc. That is to
say, articulate judging is a composite psychic process in which usually there can be discerned a more or
less fragmentary intuitive representation of a word-image. This intuitive representation is then mixed with
certain others (without analysis of the distinguishable components) into the unity of the articulate judging.
2.4 What are the concepts and propositions “in themselves”?
When an analytical philosopher equalizes concepts and propositions understood in the

logical sense with the logical (linguistic) meanings of expressions, what he means
are concepts understood as singular (elementary) objects, and not general concepts, and
the analytical philosopher does so in the name of the postulate demanding determinateness
of knowledge as to its content. This determinateness is assured by the above mentioned
equalization, which at the same time is a close link between knowledge and language.
Determinateness of knowledge understood in the above way constitutes also the linguistic
objectification (in the sense of inter-subjectification) of knowledge, i.e. it makes knowl-
edge universally important.
An analytical philosopher rejects concepts and propositions ‘in themselves’ since
they are not meanings of actual expressions, but are at most meanings of possible
expressions—analogous to the aforementioned possible complex name in the form of:
a1 ∧a2 ∧. . .∧an . General concepts understood this way are a class of possible meanings of
a thinkable multi-element nominal expression, whose every element—being an argument
of the name-forming conjunction functor—means one of many contents comprising a
general concept and each of these contents is an elementary concept. Lack of acceptance
for concepts ‘in themselves’ in this case means as much as negating a contention saying
that these concepts make up knowledge understood in the logical sense, which at the same
time has the value of objectivity in the sense of: actual inter-subjectivity, i.e. in the sense
of: actual determinateness as to its content.
Strictly speaking, an analytical philosopher does not have to reject concepts ‘in
themselves’ just because they are general concepts. He does not have to reject them
also because in their content constitution they are transcendent with respect to language,
which means that they are transcendent with respect to the subject. Such concepts can
occur in an analytical philosopher’s ontology, as long as they do not aspire to be actual
knowledge; since—in an analytical philosopher’s opinion—only knowledge determined
as to its content deserves to be described that way.20
We consider it fallacious to characterize matters in such a way that in the cases above judging is linked
to the sentence-representation simply on the basis of association. The representation enters fully into
the judgment-process and, indeed, forms its essential part. This has been convincingly demonstrated by
Husserl. An articulate judgment-process whose essential part is the intuitive representation of a sentence
we shall call ‘verbal judging’. (Here we leave open the question whether there are in general articulate
but non-verbal judgings.) Scientific judgment-processes in mature form are always of the verbal sort” [14,
p. 37].
The last sentence of this quotation deserves attention, since this sentence says the same thing that
has already been said above, namely that a cognitive subject is cognitive in the strict meaning of the
work “cognition” (“knowledge”) if, and only if the cognitive subject is a linguistic subject. What is
also noteworthy is the fact that the justification of this thesis has been taken by the author from the
phenomenological linguistic descriptions of expressions and their meanings written by the father of
phenomenology, E. Husserl, and yet the representatives of this philosophical belief do no acknowledge
this thesis which perceives a cognitive subject as a linguistic subject (I have devoted the final remarks
contained in point 7 of this article to this issue).
20 With respect to the issue of the possibility of existence in an ontology of an analytical philosopher
of concepts and propositions not being meanings of expressions, I would like to note the following
thing: And so, in an article [6] from 1948 Ajdukiewicz distinguishes three meanings of the term
80 A. Olech
2.5 The general concepts as classes and as collectives
An analytical philosopher, when speaking of knowledge, can have in mind either possible
or actual knowledge, whereby he is willing to ascribe epistemological value only to the
latter:
(a) Possible knowledge—means knowledge undetermined as to its content, but which is
possible to determine. Such knowledge is composed of logical propositions, which
are possible meanings of sentences, and of general concepts composed of elementary
concepts (elementary contents), each of which is a possible meaning of some nominal
expression.
(b) Actual knowledge—means knowledge determined as to its content. Such knowledge
is composed of logical propositions which are actual meanings of sentences and from
elementary concepts being meanings of actual expressions, which are the components
of general concepts.
Speaking of general and elementary concepts, what I mean are logical concepts, also
known as ideal concepts. With respect to general concepts understood in such way I
contended that they are a class of possible meanings of a thinkable multi-element nominal
expression, each element of which—being an argument of the name-forming conjunction
functor—means one of many contents comprising the general concept, and each of these
contents is an elementary concept. Ajdukiewicz described general concepts as classes of
concepts, and I would like to add that they are classes of elementary concepts.21
R. Ingarden, whose philosophy of language also conceives of such concepts, treated
them, as one can suspect, as wholes, i.e. as collectives. He understood meanings of
expressions as actualizations of a part of ideal senses contained in a general concept.22 If
“science”: the psychological meaning, by which science is composed of concepts and propositions in
the psychological sense; the logical meaning, by which science is composed of concepts and propositions
in the logical sense, being the logical (ideal) meaning of terms and sentences, and the ideal meaning,
by which science is composed of concepts and propositions in the logical sense, to which it could
have happened that they never became meanings of any expressions. Thus, the last understanding of
the term “science” is composed of meanings ‘in themselves’ (the Husserlian Bedeutungen ‘an sich’),
to which it could have happened that they never became meanings of any expressions (the Husserlian
ausdrückliche Bedeutungen). Although when speaking of ideal understanding of “science” Ajdukiewicz
speaks only of ideally understood sentences, and not of ideal meanings ‘in themselves’, nonetheless
speaking of such ideal sentences, i.e. de facto of sentences ‘in themselves’, is equivalent to speaking
of meanings ‘in themselves’; what is sure is that the ontic status of these beings is the same—
they are for sure radically atemporal, unlike concepts and propositions understood psychologically,
but also unlike concepts and propositions understood logically, being at the same time the logical
meanings of expressions (ausdrückliche Bedeutungen), which concepts and propositions, being involved
in expressions of a particular language L, are thus involved in time, i.e. in the temporality of this language
and the history of science pursued in the language.
21 Cf. [12, p. 143]. Ajdukiewicz writes in there that such expressions as “concept of a triangle” or “concept
of a trilateral figure” mean entire classes of concepts. He does not write that they mean the entire classes
of elementary concepts. The adjective “elementary” is my qualification in this respect used to designate a
concept determined as to its content.
22 Cf. [25, p. 139] and/or [24, Chapter V § 16]. Strictly speaking, Ingarden does not use the expression
“collective”, he does use the expression “part”, however. And this is the signalled view of Ingarden,
which needs to be quoted in full, since the matter of ideal concepts and of the relation between them and
he indeed conceived of ideal concepts as collectives, then I find Ajdukiewicz’s, rather than
Ingarden’s, approach to be right, among others due to the fact that Ajdukiewicz’s approach
makes it possible to speak of content-deprived general concepts, and I am inclined to
consider transcendental concepts to be such content-deprived general concepts.23 Yet, it
is impossible to reasonably speak of an empty whole or an empty collective, since from
the perspective of mereology empty sets do not exist.
If one acknowledges that a general concept is a class of elementary concepts, one
needs to acknowledge at the same time, that a general concept is an object of a logical
type superior towards the elementary concept corresponding thereto, and in consequence,
names which mean general concepts (e.g. the name “concept of a triangle”) are of a
higher logical type (are names of a higher syntactical category), than the names marking
elementary concepts (e.g. the name “concept of ‘triangle’ ”). And if this is so, then the
conjunction being a name-forming functor with name arguments, as already discussed,
needs to be understood in an enumerative sense, and not in a synthesizing sense. We
would be dealing with synthesizing sense, if general concepts undetermined as to their
content were collectives of elementary concepts determined as to their content.24
meanings of expressions, is one of the most important, if not the most important matter in the philosophy
of language: “[. . . ] m e a n i n g of the word ‘square’ within its material content contains c u r r e n t l
y only a certain p a r t of what is contained in the c o n c e p t of a square, resp. in the idea of a square.
On the other hand, the meaning of the expression ‘rectangular equilateral parallelogram’ contains c u r r
e n t l y a n o t h e r part of the content of t h e s a m e concept, and namely the part thanks to which
its object gets constituted through the multiplicity of ideal qualities equivalent to squareness” [25, p. 139,
translated by A.O.].
23 I write about transcendental concepts, as content-deprived general concepts in an article [34]. On this
topic see also my article [37]. When I speak of the content emptiness of transcendental concepts, what I
mean, which I have expressed in the abovementioned articles, is the objectivist understanding of content,
i.e. such understanding by which the content of a given concept is composed of features inherent to
all designata of a given concept, i.e. the characteristic features of the designata of that concept. Not
going into details connected with this issue, in this footnote I will only state that account taken of all
designata of the transcendent concept of “being”, we will not demonstrate any such feature C, which
could be common for all designata of the distributively understood concept of “being”. And this means
that the concept of “being” is content-deprived. Yet, the content described in these deliberations is the
content connected with the act of consciousness, and to be more precise: the immanent content of the
act understood in specie, or in other words, this is the content being the result of idealising abstraction
carried out on the immanent content—and thus an ideal, objective content, which is also described as
“the sense of the objective understanding” or in short as “sense”. If however the transcendent concept
of “being” is content-deprived, within the aforementioned meaning, i.e. such meaning where content is
defined in categories of features, then it is also content-deprived within the second meaning of the term
“part”, in case of which speaking of content one means the sense of objective understanding. This is so,
because features are also objects (but objects of a different tier than the objects they are inherent to), and
being objects, they are thus objects of objective understanding, which is inherent to sense. Absence of a
feature common for all designata of the transcendent concept of “being” entails thus the absence of the
sense of objective understanding, ergo: the absence of logically understood content. At a side, the same
feature C can be, as any other object, presented with the use of acts of various content, and thus various
senses can refer to the same feature C—this is the main reason why we should differentiate between
content specified in categories of features from contents understood logically, that is, in other words—
from content understood as intentions of expressions. This issue, only signalled at this point, deserves to
be treated separately.
24 Conjunction understood as a name-forming functor with name arguments, in an expression “A and B
are C” has: (1) enumerative sense, when both the object marked with “A” as well as the object marked
82 A. Olech
2.6 Ajdukiewicz’s expression meanings as Husserl’s ausdrückliche

Bedeutungen
This content, being the focus of our deliberations in this paper, whose central theme
is knowledge determined as to its content, is the matter of the meaning intention act
stemming from Logical Investigations, which, as a dependent moment of the act, coexists
together with another moment, inseparably connected with the latter and as dependant,
i.e. with the quality of the act, described several years later by Husserl in Ideas as the
“theoretical character of the act”. Unity of them both, being an immanent content of an
act, constitutes the intentional essence of the act. And the latter understood in specie is
the objective, i.e. ideal content of the act.
Bearing in mind the influence of Husserl’s philosophy’s of language on the philosophy
of language of the author of the semantic theory of knowledge, one must—speaking
of these matters—use Husserlian terminology, and moreover one must, in relation to
Sect. 2.1. of this paper and the footnote to that subparagraph, describing the subjectively
and objectively understood cognitive content, refer to a relevant passage from Logical
Investigations. This passage constitutes a philosophical-linguistic background and a
philosophical-linguistic legitimization of the semiotic justification of the thesis, quoted
at the beginning of this chapter, that if one wishes to speak of knowledge determined as
to its content, one must necessarily approach it from the perspective of the language.
Our distinction posited two sides in every acts: its quality, which stamped it as, e.g., presentation
or judgement, and its matter, that lent it direction to an object, which made a presentation, e.g.,
present this object and no object. [. . . ] One can readily see, In fact, that even if quality and objective
direction are both fixed at the same time, certain variations remain possible. Two identically
qualified acts, e.g. two presentations, may appear directed, to the same object, without full
agreement in intentional essence. The ideas equilateral triangle and equiangular triangle differ
in content, though both are directed, and evidently directed, to the same object: they present the
same object, although ‘in a different fashion’. [. . . ] The matter, therefore, must be that element in
an act which first gives it reference to an object, and reference so wholly definite that it not merely
fixes the object meant in a general way, but also the precise way in which it is meant. [. . . ] It is
the objective, the interpretative sense (Sinn der gegenständlichen Auffasung, Auffasungsinn) [. . . ].
In so far as quality and matter now count for us(as will be shown later) as the wholly essential,
and so never to be dispensed with, constituents of an act, it would be suitable to call the union of
both, forming one part of the complete act, the act’s intentional essence. To pin down this term,
and the conception of the matter it goes with, we simultaneously introduce a second term. To the
extent that we deal with acts, functioning in expressions in sense-giving fashion, or capable of
so functioning—whether all acts are so capable must be considered later—we shall speak more
specifically of the semantic essence of the act. The ideational abstraction of this essence yields a
‘meaning’ in our ideal sense.25 [19, pp. 121–123]
with “B” belong to the class of objects marked with “C” and when at the same time A and B are different
objects; (2) synthesizing sense, if the objects marked with “A” and with “B” taken together create a new,
different object marked with “C”.
25 During the Lectures on Logic Semiotic, which Ajdukiewicz gave at the John Casimir University of
Lvov in Autumn 1930, he spoke in the same manner of the logical meaning of expressions. In the last
lecture from that cycle, i.e. in Lecture XVIII given on 9 December 1930, he repeated a fragment of
Investigation V from the second volume of Logical Investigations, which I have already quoted above.
Lecture XVIII, being the last of the cycle, was a summary of that cycle, it therefore presented conclusions
which Ajdukiewicz accepted, and one of the main topics, which these lectures were devoted to, was the
Meaning-generating acts, as mentioned above, are described by Husserl in Logical

Investigations also as die Bedeuten or as “meaning intentions” (die Bedeutungsinten-
tionen). The ideal content of these acts, obtained as a result of the abovementioned
abstraction, is the expression meaning (ausdrückliche Bedeutung), meaning connected
with the language, unlike the meaning ‘in itself’ (Bedeutung ‘an sich’), which is extra-
linguistic. Ajdukiewicz explicated these expression meanings in a semiotic manner and in
accordance with the spirit of Hilbert,26 as common properties of synonymous expressions
in a given language L. Yet he did so not with the use of syntactic categories, but with
the help of syntactic-pragmatic categories, since the latter, the pragmatic categories—
describing acts of understanding of expressions or acts of acceptation of sentences—are
capable of coping with explication of quality and matter of a meaning-generating act.
Therefore, expression meanings—as described by Husserl in the above mentioned work,
and as described by Ajdukiewicz after him—are understood in specie essences of meaning
intention acts, i.e. they are—roughly speaking—meaning acts in specie. Individual
meaning intentions fall under an ideal species, and this species is something that—as
in Aristotelian conceptual realism—exists in individuo (cf. [21, p. 438]) within the scope
of meaning-generating acts, i.e. within the scope of meaning intentions.
Expression meanings are those meanings ‘in themselves’ that ‘happened’ to be meant
by a given expression, but apart from them there exist also such meanings ‘in themselves‘
(Bedeutungen ‘an sich’), which even due to our limited cognitive capabilities will never
be expressed in meaning-generating acts, i.e. they will never
[. . . ] become real in human mental life. [18, p. 233]
This is because acts of meaning intention are their realization and not their quasi-
realisation, as it is the case with the already mentioned views of R. Ingarden in this
respect. What is ideal, objective and transcendent through operation of an intending act
issue of meaning of expressions. I edited and published Lecture XVIII as a component of my paper [38].
I quoted fragments of Lecture XVIII in article [40]. As to the answer to the question, what the logically
(linguistically) understood meaning of expressions is, I would like to cite the first and the last sentence of
this Lecture—here they are: “One of the best solutions in this case is what was done by Husserl . . . ” [38,
p. 171]. Then Ajdukiewicz present the Husserlian understanding of the meaning of expressions, repeating
what Husserl said in the above quotation from Investigation V, and this is followed by a conclusion,
constituting the last sentence of this lecture, and at the same time the last sentence of the entire cycle of
lectures on logical semiotic, which is as follows: “We can say now that the meaning of a word in such
and such form is the meaning essence of thoughts, which must be involved in this word for this word to
be used as a word of such and such language” [38, p. 172]. If one wanted to correct Ajdukiewicz, and it
is necessary to do this at this point, then one would need to say that the logical meaning of an expression
(and this is the understanding of meaning that he has in mind) is not the meaning essence of an act of
meaning intention involved in this expression, but it is this essence understood in specie or, in other words,
the ideal abstraction of this essence. This is so, because the essence of an act of meaning intention is the
quality and the matter or this act, and those comprise what you can call the immanent content of the act,
and the objective (ideal) content is the immanent content understood in specie.
26 During the session of the International Philosophical Congress in Prague in 1934, Ajdukiewicz
presented his most important work to that date in the following manner: “As the first person in Poland—as
it seems—he formulated (under the influence of Hilbert) an idea of strictly formalized deductive study
of structural (and thus abstracting from the meaning of words) directives of reasoning. Following this
idea, he tried to define the meaning of words as a logical structure of certain relations existing between
expressions of a given language, which apart from the sound of the expressions is necessary for a
characteristic of the language” [2, p. 405, translated by A.O.].
84 A. Olech
falls within the scope of this act. And since without this act there is no expression, since
the act makes an expression an expression, finding its support in a physical expression
sign, then an expression is a physical-psychological-logical creation. Such understanding
of expressions cannot affect the proposal of semantic theory of knowledge claiming that
speaking of expressions is tantamount to speaking of logically understood concepts and
propositions, which are the meanings of expressions.
The psychological components of these creations—are the contents of the meaning-
generating acts, which are subjective (immanent) cognitive contents. Those, in turn,
understood in specie, are objective cognitive contents and are identical to the logical
meanings of expressions, and logical meanings are identical to linguistic meanings.27 And
so, by the aforementioned determination of cognitive acts in terms of acts of meaning
intentions involved in expressions, the cognitive subject is a linguistic subject, to be
precise: it is a part thereof, it is not identical therewith, since not every linguistic activity
is a cognitive activity, but every cognitive activity is at the same time a linguistic activity.
The last sentence requires a commentary. An analytical philosopher, and I mean in
this a case a philosopher who accepts the Husserlian theory of expressions, does not
find cognitive acts to be identical to meaning intention acts, but for certain agrees with
the thesis that each cognitive act is de facto an act of meaning intention involved in
a sentence, which is a verbalization of an act of judgement. This is so, since only
verbalized cognitive acts—and what is more only literally verbalized acts, which will
be discussed below—deserve, in the opinion of an analyst, to be called cognitive acts,
and a verbalized cognitive act is an act whose component is a more or less exact visual
presentation of a word creation. And therefore, to be precise, the abovementioned thesis
has—in another, equivalent wording—the following form: the scope of the concept
“cognitive act” is subordinate to the scope of the concept of “linguistic act” or, to put
27 I demonstrate the fact that Ajdukiewicz’s directive theory of meaning defining meaning of expressions
as a class of abstraction (common feature) of synonymous expressions is an explication of Husserl’s

intentional theory of meanings of expressions in an article [35] and [40]. The main topic of this article
is to present the explicating sequence from the act of meaning expression in specie (ausdrückliche
Bedeutung) up to the language expression as a common feature of synonymic expressions. The
conclusion of this article—which I quote literally—is as follows: Ajdukiewicz’s explicatum for the
Husserlian ausdrückliche Bedeutung is a shared property of synonymous expressions, hence identical
ones. Husserlian identity of species is Ajdukiewicz’s synonymy. The possible sequence of expressions
demonstrating that Ajdukiewicz’s explicatum is translatable with no change of sense to Husserlian
categories (and conversely, too, of course) would be as follows:
1. linguistic meaning—the same as shared property of expressions (The precise shape of this sentence
would be this: the meaning of expression E in language L is the same as the shared property of
synonymous expressions with this expression in language L. The remaining sentences of the sequence
ought to be made more precise in a similar fashion, which I have given up for stylistic considerations),
(if an expression—the same as to be used as an expression, then)
2. linguistic meaning—the same as shared property of the uses as expressions of language,
(if used as an expression of language—the same as to be understood, then)
3. linguistic meaning—the same as shared property of the acts of understanding,
(if an act of understanding—the same as an act of meaning-intention, then)
4. linguistic meaning—the same shared property of acts of meaning-intentions,
(if a shared property of acts of meaning-intentions—the same as their essence, that is species, then)
5. linguistic meaning—the same as an act of meaning-intention in specie [40, p. 151].
it in other words, the scope of the concept “cognitive activity” is subordinate with respect
to the scope of the concept “linguistic activity”. Therefore, consequently, it needs to be
said with respect to the cognitive subject that it is not identical to the linguistic subject,
but is ‘in the power’ of the linguistic subject, or in other words, that the cognitive subject
is a component of the linguistic subject.
2.7 On relation between concepts: “cognitive act”

and “linguistic act”
When discussing these things, it is impossible not to mention the phenomenological

standpoint in this respect—the thesis of a phenomenologist concerning the issue of
the relation between the scopes of the concepts: cognitive act and linguistic act is that
the scopes of these concepts intersect, which means that only some of the cognitive acts
are linguistic acts, i.e. some cognitive acts are linguistic acts and some are not. Crossing
of the scopes of these concepts, and to be precise their independence,28 results from the
fact that a phenomenologist believes, contrary to an analytical philosopher, that although
some cognitive acts and their results are not verbalized, yet they have every right to be
classified as cognitive. Cognitive acts are the acts and results of direct knowledge, specific
for this philosophical belief.29
In this situation it could seem that the standpoint of an analyst and the standpoint of
a phenomenologist are contradictory, since the following categorical sentence: Every C
is L (where C symbolizes cognitive acts, and L—linguistic acts), expressing the thesis of
an analyst, and the complex sentence: Some C are L and some C are not L, expressing
the thesis of a phenomenologist, seem to be in a relation of contradiction, i.e. seem to be
disjunctive. It turns out, however, that different ways of understanding of the cognitive act
by the analyst and the phenomenologist result in the fact that this dispute is an apparent
one, since each of them substitutes a different value for the name variable C. Distinctness
of these values results from the different answers given to the following questions: What
deserves to be called “a reason legitimizing knowledge”? Should we limit ourselves only
to those reasons which result in knowledge having the feature of objectivity, understood as
inter-subjective verifiability and communicatibility? Or should we perhaps, at the expense
of objectivity understood in such way, valorise in knowledge an objectivity understood
otherwise, namely as objective validity? In other words, what should an epistemologist
value more: intersubjectivity of knowledge at the expense of a smaller number of truths,
or should he valorise a greater number of truths at the expense of the intersubjectivity of
knowledge? As it is easily noticeable, answers to these questions are not axiologically
28 If scopes of two names (concepts) are crossing, and the logical sum of those scopes is properly contained
in the class, within which their mutual relation is considered, then such crossing of scopes is referred to as
“independence of names (concepts)”. If however a logical sum of the crossing scopes of names (concepts)
is equal to the class, within which their mutual relation is considered, then this relation is described as
“sub-opposition of names (concepts)”—on this cf. [26, pp. 98–101].
29 As to the non-verbality of cognitive acts and results in phenomenological research—see e.g. [20]. What
deserves particular attention in this respect are also the lectures of R. Ingarden from 1948/1949 given
at the Jagiellonian University, concerning the role of language in science (in knowledge)—see [23, in
particular p. 95 et seq.].
86 A. Olech
neutral, and therefore the dispute between an analyst and a phenomenologist with respect
to the role of the language in knowledge is not a dispute in the strict sense of the word,
since, first, they understand the term “knowledge” differently, and second, the dispute as
to the manner of understanding the term “knowledge” is not a dispute in the strict meaning
of the word, since it is axiologically involved. A closer examination of this dispute
would lead us to the conclusion, that each of them, the analyst and the phenomenologist,
differently understands not only the cognitive actions and results, but also the issue of
verbalization of cognitive actions and results, i.e. each of them differently understands
the term “linguistic actions”, which is involved in this dispute—for the phenomenologist
these can be a kind of linguistic actions that are unacceptable for the analyst, since the
analyst does not accept for example the language of comparisons and metaphors—this
issue is tackled in the fragment from Ajdukiewicz’s speech at the 1935 Paris Philosophical
Congress, which I present below.
As to the aforementioned dispute in the strict meaning of the word, we would be
dealing with it, if the standpoints of the disagreeing parties were possible to express with
sentences in the logical sense and if these sentences were in the relation of contradiction
or in the relation of exclusive or ordinary alternative or in the relation of disjunction.30
2.8 Ajdukiewicz’s and Husserl’s attitude towards the knowledge

definite in content
One should bear in mind the differences between the standpoint of an analytical
philosopher (Ajdukiewicz) and of a phenomenologist (Husserl or Ingarden) with respect
to the question of the relation: cognitive acts and results vs. linguistic acts and results
(and therefore, the differences between their standpoints with respect to the question of
the relation: cognitive subject vs. linguistic subject), when one speaks of the fact that
the intentional theory of expressing and meaning, presented in the second volume of
Logical Investigations, was used by Ajdukiewicz to present verbalized cognitive acts
and results connected therewith in the Husserlian way, as well as, which is connected
with the above, to understand the linguistic meaning of expressions in the Husserlian
way, and in consequence, to understand in this way knowledge determined as to its
content. Ajdukiewicz did not acknowledge however, and this needs to be borne in mind,
the sources of direct knowledge which are characteristic for phenomenology. And, for
example, in the course of a discussion with R. Ingarden, after his congress paper, he
concluded:
If some concept of proposition (in the logical sense) does not constitute a meaning of any
expression, then it is impossible to say anything about it, which would pertain to its content.
All that pertains to such concepts and propositions would therefore be inexpressionable, and
therefore it could not belong to any science, if science is understood as something that is publically
(inter-individually) available. This remark pertains in particular to the alleged theory of indirect
knowledge. [4, p. 338]
30 I write more about the dispute between the analyst and the phenomenologist on the role of language in
knowledge an article [36].

He expressed similar views at the aforementioned Paris International Philosophy

Congress. Discussing the meta-philosophy accepted in the Lvov-Warsaw School, he said
that:
In order to make science, it is not enough to observe the rules of intellectual honesty, i.e. to allow
oneself to be guided by expressing opinions by nothing else than an honest conviction rooted
in deep reflection. It is also necessary to express oneself in an intersubjectively communicative
language, and claim only that which one is able to determine and justify, being certain of such
justification.
Further, for a linguistic utterance to be intersubjectively communicative, it is not enough for
such utterance to be easily understandable for others—it is also necessary that one is able to make
sure that the utterance has been understood in its correct sense. We consider a justification to
be sure, if it is controllable by others, who could verify and repeat it. Intellectual work, which
were unable to meet the presented requirements, could never become an object of cooperation and
would not deserve the right to be called science.
Observing these two postulates determines the method and the language, limiting at the same
time the field of our philosophical activity. If we want to satisfy both of the requirements we are
discussing, we can make use neither of the Bergsonian intuition nor of Husserl’s “Wesenschau”
(with all due respect for their significance in the fields relevant for them), nor of any other
similar methods, until the results, to which they lead, can be formulated in an intersubjectively
communicative manner and can be verified in a manner which will allow us to be sure. Although
theses arrived at with the use of these methods can be verbally communicative, yet use of words
in this case is only suggestive, which means that words are used only to invoke desired mental
reactions of the listener. One therefore expresses these theses in a metaphorical manner, and uses
thereby comparisons and examples, one is unable however to formulate them in expressions of
direct character, i.e. such expressions which to be received correctly need only to be understood
literally.31 [16, pp. 123–124]
3 Conclusion
Semantic theory of knowledge grew out of the spirit of the anti-psychological break-
through, and therefore—according to this theory—if we want to speak of logically
(linguistically) understood knowledge determined as to its content, then instead of logical
concepts and propositions, comprising knowledge understood in the logical way, we
should speak of expressions, whose linguistic meanings these concepts and propositions
are—and this will be semiotically substantiated. And moreover, what I have been trying
to demonstrate, this will be also substantiated from the perspective of the philosophy
of language, if this philosophy is the philosophy presented in the Logical Investigations.
What is more, by proceeding this way, we can apply, in the theory of knowledge conceived
of in such manner, any results obtained in meta-mathematics, which is a meta-theoretical
discipline analogous to epistemology. Logically understood concepts and propositions,
being the linguistic meanings of expressions, are their logical meanings, and being at
31 During that speech Ajdukiewicz used the name “the Lvov-Warsaw School”—perhaps this was the first
time this name was used for the first time.

88 A. Olech
the same time intensions32 of expressions, are identical to the contents of mental acts
understood in specie—the acts which are expressed by these expressions.
References
1. Ajdukiewicz, K.: W obronie uniwersaliów (On Defense of Universals). Ruch Filozoficzny 13, 40b–
41b (1932)
2. Ajdukiewicz, K.: Logistyczny antyirracjonalizm w Polsce [Logistic Anti-irrationalism in Poland].
Przeglad˛ Filozoficzny 37(4), 399–408 (1934) (Reprint in German as Ajdukiewicz, K. (ed.) Der
logistische Antiirrationalismus in Polen. Erkenntnis 5, Leipzig, 151–164 (1935))
3. Ajdukiewicz, K.: Der logistische Antiirrationalismus in Polen. Erkenntnis 5, Leipzig, 151–164 (1935)
4. Ajdukiewicz, K.: Zagadnienie idealizmu w sformułowaniu semantycznym [The Issue of Idealism
in the Semantical Formulation]. Przeglad ˛ Filozoficzny 39(4), 334–340 (1936). Ksi˛ega Pamiatkowa
˛
III Polskiego Zjazdu Filozoficznego. Kraków 1936 [Commemorative Volume of the Third Polish
Philosophical Congress. Cracow 1936]
5. Ajdukiewicz, K.: Problemat transcendentalnego idealizmu w sformułowaniu semantycznym.
Przeglad˛ Filozoficzny 40(5), 271–287 (1937) (Reprint in Polish in Ajdukiewicz, K. (ed.) J˛ezyk i
poznanie. Wybór pism z lat 1920–1939 [Language and Knowledge. Selection of Papers from Years
1920–1939], vol. 1, pp. 264–277. Państwowe Wydawnictwo Naukowe, Warszawa (1960). Reprint in
English as Ajdukiewicz, K. (ed.) A Semantical Version of the Problem of Transcendental Idealism.
In: Introduction by J. Giedymin (ed.) The Scientific World-Perspectives and Other Essays 1931–1963,
pp. 140–154. D. Reidel Publishing Company, Dordrecht/Holland/Boston (1978))
6. Ajdukiewicz, K.: Metodologia i metanauka [Methodology and Meta-science]. życie Nauki 6(31–32),
4–15 (1948) (Reprint in Polish in Ajdukiewicz, K. (ed.) J˛ezyk i poznanie. Wybór pism z lat 1945–
1963 [Language and Knowledge. Selection of Papers from Years 1945–1963], vol. 2, pp. 117–126.
Państwowe Wydawnictwo Naukowe, Warszawa (1965))
7. Ajdukiewicz, K.: Trzy poj˛ecia definicji (Three Concepts of Definitions). Studia Filozoficzne 5(8),
3–16 (1958) (Reprint in Polish in Ajdukiewicz, K. (ed.) J˛ezyk i poznanie. Wybór pism z lat 1945–
1963 [Language and Knowledge. Selection of Papers from Years 1945–1963], vol. 2. Państwowe
Wydawnictwo Naukowe, Warszawa (1965))
8. Ajdukiewicz, K.: J˛ezyk i poznanie. Wybór pism z lat 1920–1939 [Language and Knowledge.
Selection of Papers from Years 1920–1939], vol. 1. Państwowe Wydawnictwo Naukowe, Warszawa
(1960)
9. Ajdukiewicz, K.: J˛ezyk i znaczenie. In: Ajdukiewicz, K. (ed.) J˛ezyk i poznanie. Wybór pism z lat
1920–1939 [Language and Knowledge. Selection of Papers from Years 1920–1939], vol. 1, pp.145–
174. Państwowe Wydawnictwo Naukowe, Warszawa (1960). Trans. from German into Polish by
F. Zeidler (Reprint in English as Ajdukiewicz, K. (ed.) Language and Meaning. In: Introduction
by J. Giedymin (ed.) The Scientific World-Perspectives and Other Essays 1931–1963, pp. 35–66.
D. Reidel Publishing Company, Dordrecht/Holland/Boston (1978). The orig.: Sprache und Sinn.
Erkenntnis 4, Leipzig, 100–138 (1934))
10. Ajdukiewicz, K.: In: Mortimer, H., Szaniawski, K. (ed.) Logika pragmatyczna [Pragmatic Logic].
Państwowe Wydawnictwo Naukowe, Warszawa (1965). Wst˛ep [Foreword] by K. Szaniawski
32 In the 1930s Ajdukiewicz did not deal with the question of intensionality. The first work in which the
concept of “intensionality” was described more extensively is Ajdukiewicz’s article [9] from 1958 in
which—additionally to the topic indicated in the title—he argues that due to the intensional character of
certain expressions, it is impossible to eliminate real definitions from the general theory of definitions
and to replace them with nominal definitions in objective stylistics, without encountering insurmountable
difficulties. The need for such replacement is motivated by some philosophers with their nominalist views,
since the universale is the thing to which the real definition refers to.
11. Ajdukiewicz, K.: J˛ezyk i poznanie. Wybór pism z lat 1945–1963 [Language and Knowledge.
Selection of Papers from Years 1945–1963], vol. 2. Państwowe Wydawnictwo Naukowe, Warszawa
(1965)
12. Ajdukiewicz, K.: A Semantical Version of the Problem of Transcendental Idealism. In: Introduction
by J. Giedymin (ed.) The Scientific World-Perspectives and Other Essays 1931–1963, pp. 140–154.
D. Reidel Publishing Company, Dordrecht/Holland/Boston (1978)
13. Ajdukiewicz, K.: On the problem of universals. In: Introduction by J. Giedymin (ed.) The Scientific
World-Perspectives and Other Essays 1931–1963, pp. 95–110. D. Reidel Publishing Company,
Dordrecht/Holland/Boston (1978)
14. Ajdukiewicz, K.: Language and Meaning. In: Introduction by J. Giedymin (ed.) The Scientific
World-Perspectives and Other Essays 1931–1963, pp. 35–66. D. Reidel Publishing Company,
Dordrecht/Holland/Boston (1978)
15. Ajdukiewicz, K.: In: Introduction by J. Giedymin (ed.) The Scientific World-Perspectives and Other
Essays 1931–1963. D. Reidel Publishing Company, Dordrecht/Holland/Boston (1978)
16. Ajdukiewicz, K.: Przemówienie powitalne delegacji polskiej na Mi˛edzynarodowym Kongresie
Filozofii Naukowej na Sorbonie w roku 1935 [Welcoming Speech of the Polish Delegation at the
International Scientific Philosophy Congress in Sorbonne]. Principia. Pisma koncepcyjne z filozofii i
socjologii teoretycznej 10–11, 123–125 (1994). Trans. from French into Polish by J. Hartman (The
orig. in: Actes du Congrés international de philosophie scientifique, vol. 1. Philosophie scentifique et
empirisme logique, Sorbone, Paris (1935))
17. Husserl, E.: Logische Untersuchungen. Untersuchungen zur Phänomenologie und Theorie der
Erkenntnis. Bd. 2, Teil 1. Max Niemeyer, Halle (1928)
18. Husserl, E.: Logical Investigations, vol. 1. Routledge, London and New York (2001) (J.N. Findlay,
trans. from the 2nd German edition of Logische Untersuchungen)
19. Husserl, E.: Logical Investigations, vol. 2. Routledge, London and New York (2001) (J.N. Findlay,
trans. from the 2nd German edition of Logische Untersuchungen)
20. Ingarden, R.: Dażenia
˛ fenomenologów [The endeavours of phenomenologists]. Przeglad ˛ Filozoficzny
22(3), 118–156, and 22(4), 315–351 (1919) (Reprint in Ingarden, R. (ed.) Z badań nad filozofia˛
współczesna˛ [From Studies of Contemporary Philosophy], pp. 269–379. Państwowe Wydawnictwo
Naukowe, Warszawa (1963))
21. Ingarden, R.: Główne fazy rozwoju filozofii E. Husserla [The Main Gestation Periods of Husserl’s
Philosophy]. In: Ingarden, R. (ed.) Z badań nad filozofia˛ współczesna˛ [From Studies of Contemporary
Philosophy], pp. 383–452. Państwowe Wydawnictwo Naukowe, Warszawa (1963). (the modified
reprint from: idem, Główne linie rozwoju pogladów ˛ filozoficznych Edmunda Husserla [The Main
Lines of Development of Husserl’s Philosophical Ideas]. Przeglad ˛ Filozoficzny 42(2), 25–76 (1939–
1945))
22. Ingarden, R.: Z badań nad filozofia˛ współczesna˛ [From Studies of Contemporary Philosophy].
Państwowe Wydawnictwo Naukowe, Warszawa (1963)
23. Ingarden, R.: Z teorii j˛ezyka i filozoficznych podstaw logiki [From the Theory of Language and the
Philosophical Grounds of Logic]. Państwowe Wydawnictwo Naukowe, Warszawa (1972)
24. Ingarden, R.: The Literary Work of Art. Northwestern University Press, Evanston, IL (1973). Trans.
by George G. Grabowicz
25. Ingarden, R.: O dziele literackim. Badania z pogranicza ontologii, teorii j˛ezyka i filozofii literatury,
2nd edn. Państwowe Wydawnictwo Naukowe, Warszawa (1988). Trans. from German into Polish
by M. Turowicz (The orig.: Das literarische Kunstwerk. Eine Untersuchung aus dem Grenzgebiet
der Ontologie, Logik und Literaturwissenschaft. Niemeyer, Halle (1931). Reprint in English as The
Literary Work of Art. Northwestern University Press, Evanston, IL (1973). Trans. by George G.
Grabowicz)
26. Jadacki, J.J.: Spór o granice j˛ezyka [Dispute on the Limits of Language]. Wydawnictwo Naukowe
Semper, Warszawa (2002)
27. Łukasiewicz, J.: Logistyka a filozofia [Logistic and Philosophy]. Przeglad ˛ Filozoficzny 39(2), 115–
131 (1936) (Reprint in Polish in Łukasiewicz, J. (ed.) Z zagadnień logiki i filozofii. Pisma wybrane
[Topics in Logic and Philosophy. Selected Writings], pp. 195–209. Państwowe Wydawnictwo
Naukowe, Warszawa (1961). Reprint in English as Łukasiewicz, J. (ed.) Logistics and Philosophy.
In: Borkowski, L. (ed.) Selected Works, pp. 218–235. North-Holland Publishing Company, Amster-
dam/London (1970). PWN — Polish Scientific Publishers, Warszawa (1970))
90 A. Olech
28. Łukasiewicz, J.: W obronie logistyki. Myśl katolicka wobec logiki współczesnej [In Defence of
Logic. Catholic Thought in Comparison with Contemporary Logic]. Studia Gnesnensia 15 (1937)
(Reprint in Polish as Łukasiewicz, J. (ed.) W obronie logistyki [In defense of logistic]. In: Z zagadnień
logiki i filozofii. Pisma wybrane [Topics in Logic and Philosophy. Selected Writings], pp. 210–219.
Państwowe Wydawnictwo Naukowe, Warszawa (1961). Reprint in English as Łukasiewicz, J. (ed.) In
defense of logistic. In: Borkowski, L. (ed.) Selected Works, pp. 236–249. North-Holland Publishing
Company, Amsterdam/London (1970). PWN — Polish Scientific Publishers, Warszawa (1970))
29. Łukasiewicz, J.: W obronie logistyki [In defense of logistic]. In: Łukasiewicz, J. (ed.) Z zagadnień
logiki i filozofii. Pisma wybrane [Topics in Logic and Philosophy. Selected Writings], pp. 210–219.
30. Łukasiewicz, J.: Z zagadnień logiki i filozofii. Pisma wybrane [Topics in Logic and Philosophy.
Selected Writings]. Państwowe Wydawnictwo Naukowe, Warszawa (1961). Selected by and provided
with an introduction and footnotes by J. Słupecki
31. Łukasiewicz, J.: In defense of logistic. In: Łukasiewicz, J.: In: Borkowski, L. (ed.) Selected
Works. North-Holland Publishing Company, Amsterdam/London (1970). PWN — Polish Scientific
Publishers, Warszawa (1970)
32. Łukasiewicz, J.: Logistics and Philosophy. In: Borkowski, L. (ed.) Selected Works, pp. 218–
235. North-Holland Publishing Company, Amsterdam/London (1970). PWN — Polish Scientific
33. Łukasiewicz, J.: In: Borkowski, L. (ed.) Selected Works. North-Holland Publishing Company,
Amsterdam/London (1970). PWN — Polish Scientific Publishers, Warszawa (1970)
34. Olech, A.: W sprawie poj˛eć uniwersalnych i transcendentalnych [On Universal and Transcendental
Concepts]. Filozofia Nauki 6(2/22), 35–49 (1998)
35. Olech, A.: Ajdukiewicz a Husserl wobec kwestii znaczenia wyrażeń [Ajdukiewicz and Husserl on the
Issue of the Meaning of Expressions]. Studia Semiotyczne 24, 141–161 (2001) (Reprint in English
as Olech, A. (ed.) Ajdukiewicz and Husserl on the issue of the meaning of expressions. Stud.
Semiotyczne — English Supplement 24, 130–154 (2015/2001). http://studiaes.pts.edu.pl/volume-
xxiv.html)
36. Olech, A.: W sprawie sporu o rol˛e j˛ezyka w poznaniu [On the Dispute Over the Role of Language in
Knowledge]. Przeglad ˛ Filozoficzny — Nowa Seria 14(2/54), 221–229 (2005) (Reprint in: Centrum
Badań Filozoficznych im. R. Ingardena; http://www.roman-ingarden.phils.uj.edu.pl/artykuly.php)
37. Olech, A.: O poj˛eciach uniwersalnych i transcendentalnych [Concerning Universal and Transcen-
dental Concepts]. In: Żurkowska, G., Blandzi, S. (eds.) Rezonujacy ˛ rozum nauki a rozumność
intuicji. Ksi˛ega dedykowana Profesor Alinie Motyckiej [The Resonating Reason of Science and the
Rationality of Intuition. A Commemorative Volume Dedicated to Professor Alina Motycka], pp. 364–
378. Wydawnictwo IFiS PAN, Warszawa (2009)
38. Olech, A.: O Ajdukiewiczu i Ingardenie – uczniach Husserla. W setna˛ rocznic˛e ukazania si˛e Ksi˛egi
I Idei (On Ajdukiewicz and Husserl – the Husserl’s Pupils. On the Hundredth Anniversary of
the First Edition of Book I of Husserl’s Ideen). In: Woźniczka, M. (ed.) Filozofia polska na tle
filozofii europejskiej (Polish Philosophy on the Background of European Philosophy), pp. 159–176.
Wydawnictwo Akademii im. Jana Długosza w Cz˛estochowie, Cz˛estochowa (2014)
39. Olech, A.: Semantyczna teoria poznania [Semantic Theory of Knowledge]. Wydawnictwo im.
S. Podobińskiego Akademii im. Jana Długosza w Cz˛estochowie, Cz˛estochowa (2014)
40. Olech, A.: Ajdukiewicz and Husserl on the issue of the meaning of expressions. Stud. Semiotyczne
— English Supplement 24, 130–154 (2015/2001). http://studiaes.pts.edu.pl/volume-xxiv.html
A. Olech ()
Institute of Philosophy, Jan Dlugosz University in Cz˛estochowa, Cz˛estochowa, Poland
Categorial Grammars and Their Logics
Wojciech Buszkowski
Abstract This paper surveys the development of categorial grammars (also called type
grammars) with emphasis on type logics (i.e. logical calculi underlying these grammars)
and their relation to the origin in Ajdukiewicz (Studia Philosophica 1:1–27, 1935).
Keywords Categorial grammar · Type · Lambek calculus · Substructural logic ·

Parsing
Mathematics Subject Classification (2000) Primary 03B65; Secondary 68T50
1 Introduction
In the modern literature Kazimierz Ajdukiewicz is commonly accepted as the father of

categorial grammars: formal grammars assigning logical types to expressions. His seminal
paper [4] provided a clear idea of these grammars. Ajdukiewicz acknowledged an impact
of Husserl and Leśniewski. From Husserl [27] he took the idea of semantical categories
which can be defined in terms of mutual substitution of expressions in meaningful
or sentential contexts. From Leśniewski he took a classification of categories in basic
categories and functor categories. Here I cannot cite any single publication of Leśniewski,
since in his works he never wrote a longer passage on this matter. Ajdukiewicz and other
authors cite [41], but this paper on new foundations of mathematics merely contains short
notes on ‘the theory of semantical categories’; different logical symbols are characterized
as functors of a particular category (no special symbols for categories are introduced). It
seems that more elaborated considerations only appeared in Leśniewski’s oral lectures.
Ajdukiewicz introduced a system of indices for categories. This system is employed
in his procedure for verifying the ‘syntactic connexion’ of expressions. In [4] he writes:
“We shall base our work here on the relevant results of Leśniewski, adding on our part
a symbolism, in principle applicable to almost all languages [and enabling us to build a
calculus], which makes it possible to formally define and examine the syntactic connexion

92 W. Buszkowski
of a word pattern.”1 (The passage in brackets has been omitted in the English translation in
[6]; I add it, since the word ‘calculus’ is quite important.) In fact, the indices for categories
were introduced in Ajdukiewicz’s earlier paper [3], where they were used in a semantical
analysis of the problem of universals.
Let me briefly comment on terminology. The indices for categories will be called types,
according to modern standards in logic. Categories can be understood as some sets of
expressions or sets of ontological objects (having a common type). Although Leśniewski
and Ajdukiewicz use the term ‘semantical category’ (after Husserl), the term ‘syntactic(al)
category’ is more appropriate. This was noticed by Ajdukiewicz [5]: “The concept
of semantical categories must be clearly distinguished from the concept of syntactical
categories. The term ‘semantical category’ was introduced for the first time by Husserl;
however, the concept he associated with it would correspond better to the term ‘syntactical
category’. For Husserl pointed out that the expressions of a language may be classified
according to the role they can play within a sentence. He defined, therefore, the categories
from the syntactical viewpoint.” Ajdukiewicz [5] outlined a theory of semantical cate-
gories: the type of an expression is determined by the ontological type of the denotation
of this expression. The first semantical interpretation of [4] is due to Bocheński [9].
The connections of syntactic (semantic) types with categories defined by mutual
substitution are by no way obvious, nor simple. They are quite tight for deterministic
(or: rigid) grammars which assign at most one type to one expression, but become less
regular for categorially ambiguous grammars. A thorough discussion of this topic can be
found in [15].
The present paper focuses on categorial grammars: how they developed from the origin
in [4] to their modern forms. Categorial grammars are also called ‘type grammars’, and
the latter term seems better. The classification of expressions in categories appears in
different grammar formalisms (e.g. phrase structure grammars), whereas logical types
assigned to expressions are characteristic of the grammars considered here. To emphasize
this Morrill [47] and others use an even more explicit term ‘type logical grammar’. In the
present paper both terms are used: the former in traditional names of grammars, the latter
in general considerations.
The impact of [4] can be seen in several areas of formal linguistics and logical
philosophy of language. Type grammars belong to formal linguistics, since their main
intention is to describe natural language. They are closely related to type-theoretic
semantics of natural language, initiated by Monatgue [42], with an explicit reference
to [4], and extensively studied by many authors as Montague Grammar. Some other
works may be counted to the logical turn: they study types and categories in formal
languages of logic and mathematics. This direction was represented in Poland by Suszko
[58, 59], Wybraniec-Skardowska [67] and others. Suszko elaborated a formal framework
for syntax and semantics of higher-order languages. Wybraniec-Skardowska presented a
general theory of ‘categorial languages’ with a distinction between expression-tokens and
abstract expressions (this theory, however, does not directly address natural language).
Tałasiewicz [60] provided a philosophical analysis of Ajdukiewicz’s approach, applied to
natural language, with an interpretation in terms of situation semantics,
1 Allcitations from Ajdukiewicz are based on the English translations of Ajdukiewicz’s original papers,
collected in [6].
Categorial Grammars and Their Logics 93
It is a bit surprising that just the linguistic turn leads to new logical calculi (Lambek
logics) and models (residuated algebras), whereas the logical turn usually focuses on some
standard logical systems (higher-order logics, type theories). To keep this survey in a rea-
sonable size I will mainly write on the new logics elaborated for type grammars and only
briefly note some links with other developments. Montague Grammar and its descendants
cannot be discussed in detail; the reader is referred to [65] and the items cited there.
This survey is addressed to a wide community, not necessarily experts in type
grammars. Therefore I omit mathematical subtleties. I do not even discuss all important
mathematical results in this area; I only briefly note some of them to clarify the main
ideas. Nonetheless an acquaintance with general logic and formal linguistics may help
the reader to follow the text. I provide a couple of linguistic examples, but all are quite
simple. The reader is referred to [40, 45–48] for a more advanced linguistic material.
Section 2 is concerned with basic categorial grammars, a framework directly related
to Ajdukiewicz’s proposal (modified in [8]). Section 3 discusses the Lambek calculus and
several analogous systems with a particular emphasis on their role in type grammars. At
the end, I defend the view that Lambek logics are important, general logics of syntactic
and semantic types (besides other applications), but not as good for efficient parsing.
An optimal strategy seems the following: (1) to apply Lambek logics in metatheory and
on the lexical level, (2) to preserve the Ajdukiewicz system as a parsing procedure for
compound expressions.
2 Basic Categorial Grammars
Ajdukiewicz (following Leśniewski) distinguishes two basic categories: sentence (type

s) and name (type n), but stipulates that in general “nothing could be decided about the
number and kind of basic semantic [categories] and functor categories, since these may
vary in different languages.” The types of functor categories have the form of fractions:
α
;
β1 . . . βn
an expression of this type with arguments of type β1 , . . . , βn forms a compound
expression of type α. For example, an intransitive verb is of type ns , a transitive verb
of type nsn , a sentential connective of type sss , an adverb of type αα for α = ns , and so on.
The procedure of checking the ‘syntactic connexion’ of a compound expression is
designed as follows. First, the expression is rewritten in prefix notation: each functor
directly precedes its arguments. So one writes likes John wine instead of John
likes wine and hardly works John instead of John works hardly (my
examples). Second, one considers the sequence of types corresponding to the words of
the rearranged expression. For these two examples one obtains the sequences:
s
s n s
, n, n and s , , n.
nn n n
Third, one reduces a block of adjacent types:

α
, β1 , . . . , βn
β1 . . . βn
94 W. Buszkowski
to α and repeats this step as many times, as possible. If this reduction ends in a single
type, the expression is qualified to be ‘syntactically connected’ and assigned the resulting
type. The Ajdukiewicz reduction procedure applied to our examples yields:
s
, n, n ⇒ s in one step,
nn
s
s s
n
s , , n ⇒ , n ⇒ s in two steps.
n n n
So both expressions are syntactically connected (of type s). In fact, Ajdukiewicz’s original
procedure was more restrictive: at each step one reduces the left-most occurrence of a
reducible pattern, but this constraint narrows its applications [13].
This approach reveals two characteristic components of modern type grammars: (1)
the type lexicon, i.e. an assignment of types to words, (2) the type processing machinery,
i.e. a procedure of checking the grammatical correctness of arbitrary expressions and at
the same time deriving types of them. In terms of contemporary computational linguistics,
(2) is a parsing procedure. Ajdukiewicz was the first who clearly formulated the problem
of parsing and proposed a parsing algorithm (20 years before mathematical linguistics
was founded by Noam Chomsky).
The Ajdukiewicz procedure requires the rewriting of the parsed expression in prefix
notation. In practice this restricts its applications to some formal languages. In fact
Ajdukiewicz acknowledged that one of his goals was a generalization of the parenthesis-
free notation, elaborated by J. Łukasiewicz for propositional logics, toward richer formal
languages. On the other hand, his examples came from natural languages, and he expected
a wide applicability of his method. Probably he admitted various modifications of the
original procedure, when applied in practice.
Bar-Hillel [7] adjusted this approach to natural language. He introduced directional
types of the form:
α
β1 . . . βm ; γ 1 . . . γ n
and the reduction procedure based on the rule:

α
β1 , . . . , βm , , γ1 , . . . , γn ⇒ α.
β1 . . . βm ; γ 1 . . . γ n
s
Now transitive verbs are assigned type n;n , and John likes Mary is parsed as:
s
n, , n ⇒ s in one step.
n; n
In [8], this approach was modified. After Lambek [36], functor types were restricted
to α\β and α/β. An expression of type α\β (resp. β/α) with an argument of type α on
the left (resp. on the right) forms a compound expression of type β. So α\β corresponds
β α α
to α; in the former notation, α/β to ;β , and the fraction β;γ is represented as β\(α/γ ) or
(β\α)/γ . The representation of many-argument types by (nested) one-argument types is
closely related to ‘currying’, i.e. the representation of many-argument functions by one-

argument functions of higher order, a routine in modern type theories.
The reduction procedure is based on two rules:
(RED.1) α, α\β ⇒ β, (RED.2) α/β, β ⇒ α .
In [8], a categorial grammar is formally defined as a triple G = (, I, s) such that

is a nonempty finite set, I is a finite relation between elements of and types, and s is an
atomic type. The elements of are interpreted as the words of a natural language (then
is referred to as the lexicon) or symbols of a formal language (then is referred to as the
alphabet). Nowadays I is called the type lexicon or the initial type assignment. Often I is
represented as a map which assigns finite sets of types to elements of . In examples we
write v : α for α ∈ I (v). One refers to s as the designated type. One admits an arbitrary
finite set of atomic types.
Finite sequences of elements of are called strings (on ). The empty string is
denoted by . The string (v1 , . . . , vn ) is usually written as v1 . . . vn . One says that
G assigns type α to the string v1 . . . vn , if there exist types α1 , . . . , αn , belonging to
I (v1 ), . . . , I (vn ), respectively, such that the sequence α1 , . . . , αn reduces to α by finitely
many applications of rules (RED.1), (RED.2). The language of G consists of all strings
on which are assigned type s by G.
In the modern literature, categorial grammars in the sense of [8] are called basic
categorial grammars (BCGs) or: classical categorial grammars, AB-grammars (a credit
to Ajdukiewicz and Bar-Hillel).
The main mathematical theorem of [8] establishes the weak equivalence of BCGs and
Chomsky’s (-free) context-free grammars (CFGs). Recall that a CFG is defined as a
quadruple G = (, N, s, P ) such that and N are disjoint finite sets (whose elements
are treated as simple symbols), s ∈ N, and P is a finite set of pairs (a, x), where a ∈ N
and x is a string on ∪ N. The elements of (resp. N) are called terminal symbols
(resp. nonterminal symbols or variables), and s is called the start symbol. The pairs in P
are called production rules. one writes a → x for (a, x) and interprets it as a rewriting
rule: the string yaz can be rewritten as yxz according to this rule (by xy one denotes the
concatenation of x and y). The language of G (or: generated by G) consists of all strings
on which can be derived from s by finitely many applications of the production rules. A
CFG is -free, if it contains no nullary rule of the form a → . The equivalence theorem
states that BCGs and -free CFGs generate the same class of languages. More precisely,
for any BCG G there exists an -free CFG G such that L(G) = L(G ), and conversely.
The first part of this theorem can easily be proved: a BCG G = (, I, s) generates
the same language as the CFG with the terminal alphabet , the nonterminal alphabet
consisting of all types involved in G and their subtypes (i.e. subterms), the start symbol
s, and the production rules reversing (RED.1), (RED.2) (restricted to the types in the
nonterminal alphabet) plus the lexical rules α → v, for α ∈ I (v). The second part is
more difficult; the proof in [8] yields, in fact, the Greibach normal form theorem for
CFGs (independently proved a few years later).
In opposition to CFGs, BCGs are lexical: the whole information on the described
language is contained in the type lexicon, whereas the parsing procedure is independent of
this particular language (it employs the language-independent rules (RED.1), (RED.2)).
This is not the case for CFGs. For instance, a standard non-lexical rule for English is
96 W. Buszkowski
s → np, vp (a sentence consists of a noun phrase and a verb phrase). The lexicality
is a characteristic feature of all type grammars, considered nowadays. Sometimes it is
convenient to admit certain simple non-lexical rules, e.g. pn ⇒ np (a proper noun is a
noun phrase), but one tends to eliminate them, whenever possible.
The nonterminal symbols of a CFG can be interpreted as names of syntactic categories,
like types in a BCG. Types, however, can be compound terms, not just simple symbols.
This is significant for lexicality and makes it possible to study logics of types, expressing
deeper relations between types.
Although CFGs are weakly equivalent to BCGs, the strong equivalence does not hold;
this means that the structured languages differ for the two classes of grammars. For a
CFG, each derivation of a string from a nonterminal symbol determines a unique phrase
structure of this string. For instance, the grammar with production rules:
s → np, vp vp → tv, np
np → John np → tee tv → drinks
admits the derivation:
s ⇒ np, vp ⇒ np, tv, np ⇒ · · · ⇒ John, drinks, tee
which yields the phrase structure (John (drinks tee)) or, more explicitly, (Johnnp (drinkst v
teenp )vp )s . These phrase structures can be depicted as binary trees; see Fig. 1.
Similarly, each reduction in a BCG gives rise to a unique phrase structure of the input
string. With the type lexicon:
John : np drinks : (np\s)/np tee : np
one obtains the reduction:
np, (np\s)/np, np ⇒ np, np\s ⇒ s ,
which yields the same phrase structure (John (drinks tee)). We need an auxiliary notion.
The degree of type α, denoted by d(α), is defined as follows: d(α) = 0 if α is atomic,
d(α\β) = d(β/α) = d(β) + 1. For any phrase structure generated by a BCG G, depicted
as a tree, and for any node of this tree, the length of shortest paths from this node to a leaf
is not greater than the maximal degree of types involved in G. Therefore a BCG cannot
Fig. 1 Derivation trees for a drinks: tv tee: np

CFG
A
A
A
drinks tee A
AA John: np vp
A \
John \
\ \
\
\ \
s
generate languages of phrase structures with arbitrarily long shortest paths from a node to
a leaf. On the contrary, a CFG can generate such languages.
For instance, the CFG with rules s → s, s and s → 0 generates all possible phrase
structures on the alphabet {0}. The BCG with the type lexicon 0 : s/s, 0 : s generates
the same language of strings, which consists of all nonempty strings on {0}, but not the
same language of phrase structures. One only gets the phrase structures: 0, (00), (0(00)),
(0(0(00))), and so on, but not ((00)0).
For BCGs, one also considers functor-argument structures (fa-structures), i.e. phrase
structures augmented with functor markers. For the BCG considered above, the phrase
structure (John (drinks tee)) can be refined to the fa-structure (John (drinks tee)1 )2 , which
means that (drinks tee) is the functor in the whole structure and drinks is the functor in
(drinks tee). Every reduction in a BCG determines a unique fa-structure of the recognized
string. The languages of fa-structures and phrase structures take an essential part in the
theory of BCGs; see [11, 15]. In particular, syntactic categories can be defined as certain
sets of fa-structures rather than strings, which results in a more elegant theory.
The type lexicon of a BCG can assign several types to one word. This reflects
the syntactic ambiguity of words in natural language. For instance, and appears as a
sentential connective, but also as a noun connective, verb connective, adverb connective,
and others. As a rule, in logical and mathematical formalisms one symbol can be assigned
a unique type, which completely characterizes the syntactic role of this symbol. These
languages can be described by rigid (or: deterministic) BCGs (I is a function from to
the set of types).
Worthy of noting, not all languages of formal logic can be described by rigid BCGs.
The standard example is the language of (type-free) lambda calculus. Also in the language
of first-order logic, a unary function symbol f requires two types t/iv, t/t, where iv is
the type of individual variables and t of terms (quantifiers are typed (s/s)/iv, where s is
the type of formulas). Alternatively, one can assign only t/t to f and admit a non-lexical
rule iv ⇒ t.
The type of quantifiers, given above, adequately characterizes their role in the syntax
of first-order logic (in modern setting): the quantifier followed by a variable, next by a
formula, yields a formula. It is also fully compatible with Tarskian semantics for this
logic. It, however, does not express the variable-binding role of quantifiers. The final part
of [4] is devoted to the special status of variable-binding operators, and several authors
continue this issue; see [49, 58, 59, 67]. I do not discuss this matter here, since it goes too
far from the main topics of this paper.
One of the leitmotives of type grammars is a close relationship between syntax and
semantics (the dictum syntax mirrors ontology). I have already noted that Bocheński [9]
proposed the semantical interpretation of the theory of Leśniewski and Ajdukiewicz, and
this turn was adopted by Ajdukiewicz [5]. According to the latter, the basic types are
i (individual) and w (truth value; ‘value’ corresponds to Polish ‘wartość’ and German
‘Wert’). Intransitive verbs are typed wi , as they denote functions from the set of individuals
to the set of truth values, transitive verbs iwi , as they denote two-argument functions of this
kind, (binary) sentential connectives www , as they denote binary truth-value functions,
and so on. Ajdukiewicz [5] brings a radical idea of a purely flectional language: the
types of words only account for semantical categories of these words (i.e. the ontological
status of their denotations), whereas their syntactic roles are described by certain new
98 W. Buszkowski
indices, indicating the position of these words in syntactic trees (some representations of
fa-structures). This idea seems very interesting, but the symbolism, proposed in [5], has
a limited value, since the new indices show the positions of words in one particular tree,
not in any well-formed syntactic tree, containing the given word.
On the other hand, the semantical interpretation is quite fundamental and—modulo
terminology and notation—has been commonly adopted in modern type-theoretic seman-
tics. Syntactic types are translated into semantic types: atomic types and compound types
α → β, where α, β are simpler types. Each atomic type p corresponds to a semantic
domain (or: ontological category) Dp . One defines Dα→β as the set of all functions from
Dα to Dβ . For instance, s is translated into t (the type of truth values) and n into e (the
type of entities). Let α • denote the translation of α. One recursively defines:
(α\β)• = (β/α)• = α • → β • .
So n\s is translated into e → t (the type of sets of entities, identified with their
characteristic functions), s/(n\s) into (e → t) → t (the type of families of sets of
entities), and so on. The latter agrees with the interpretation of complete noun phrases as
generalized quantifiers; see van Benthem [62].
In semantics, the reduction rules (RED.1), (RED.2) can be interpreted as the applica-
tion of a function f ∈ Dα • →β • to an argument a ∈ Dα • , which yields f (a) ∈ Dβ • .
Thus, given some fixed denotations of all words of the parsed expression, whose semantic
types correspond to their syntactic types, as above, one can determine the denotation of
this expression by the (iterated) application of functions to their arguments, following the
syntactic reduction procedure. This fully agrees with the principle of compositionality, a
central idea of logical semantics.
3 Lambek Calculus
3.1 Basic Systems
An essential refinement of BCGs is due to Lambek [36]. His Syntactic Calculus,

nowadays called Lambek Calculus and denoted by L, is regarded as a basic type logic.
Lambek presented his system as an improvement of BCGs: “[. . . ] this paper is concerned
with a development of the technique of Ajdukiewicz and Bar-Hillel in a mathematical
direction. We introduce a calculus of types, which is related to the well-known calculus
of residuals. The decision procedure for this system is solved affirmatively, following a
procedure first proposed by Gentzen for the intuitionistic propositional calculus.”
Types are built from atomic types by \, / and · (product; some authors write ⊗). An
axiomatization of L employs simple sequents of the form α ⇒ β, where α, β are types.
L admits the following axioms and inference rules.
(Id) α ⇒ α
(A.1) (α · β) · γ ⇒ α · (β · γ ) (A.2) α · (β · γ ) ⇒ (α · β) · γ
α·β ⇒γ α·β ⇒γ
(Res.1) (Res.2)
β ⇒ α\γ α ⇒ γ /β
α⇒β β ⇒γ
(Cut.1)
α⇒γ
The double line in (Res.1), (Res.2) means that these rules can be used in both directions:
top-down and bottom-up.
By dropping the associativity axioms (A.1), (A.2), one obtains Nonassociative Lambek
Calculus (NL), due to Lambek [37]. The counterparts of (RED.1), (RED.2):
(Red.1) α · (α\β) ⇒ β , (Red.2) (α/β) · β ⇒ α
are provable in NL, using (Id), (Res.1), (Res.2). We, however, obtain (infinitely) many
other laws. Here are some examples.
• (L1) α ⇒ (β/α)\β and α ⇒ β/(α\β),
• (L2) α ⇒ β\(β · α) and α ⇒ (α · β)/β,
• (L3) (α\β) · (β\γ ) ⇒ α\γ and (α/β) · (β/γ ) ⇒ α/γ ,
• (L4) α\β ⇒ (γ \β)\(γ \α) and α/β ⇒ (α/γ )/(β/γ ),
• (L5) (α\β)/γ ⇔ α\(β/γ ) (⇔ stands for both ⇒ and ⇐).
(L1), (L2) are provable in NL, but (L3), (L4), (L5) in L only. Other laws can be
obtained, by using the monotonicity rules: from α ⇒ β infer γ · α ⇒ γ · β, α · γ ⇒ β · γ ,
γ \α ⇒ γ \β, β\γ ⇒ α\γ , α/γ ⇒ β/γ , γ /β ⇒ γ /α, which are derivable in both
systems.
The most general algebraic models of NL are residuated groupoids, i.e. ordered
algebras (A, ·, \, /, ≤) such that (A, ≤) is a partially ordered set, and ·, \, / are binary
operations on A, satisfying the residuation laws:
(RES) a · b ≤ c iff b ≤ a\c iff a ≤ c/b, for all a, b, c ∈ A .
The operations \, / are called the residual operations for product. Residuated semigroups
are residuated groupoids such that · is associative; they are models for L. Both systems
are strongly complete with respect to the corresponding models: the sequents provable in
the system from a set of nonlogical hypotheses are precisely those sequents which are
true in all models, for all valuations μ, satisfying the hypotheses. α ⇒ β is true for μ, if
μ(α) ≤ μ(β).
According to Lambek [36], the intended models for L are language models, i.e. some
algebras of languages (by a language one means a set of strings). By + we denote the
set of all nonempty strings on . For L1 , L2 ⊆ + , one defines:
L1 · L2 = {xy : x ∈ L1 , y ∈ L2 } ,
L1 \L2 = {y ∈ + : xy ∈ L2 for any x ∈ L1 },
L1 /L2 = {x ∈ + : xy ∈ L1 for any y ∈ L2 },

100 W. Buszkowski
where xy denotes the concatenation of strings x and y. It is easy to show that the powerset
of + with ·, \, / defined as above and inclusion as the order, is a residuated semigroup.
α ⇒ β is true for μ in this model if and only if μ(α) ⊆ μ(β) (equivalently: every string
of type α is of type β). The term ‘language model’ is due to Pentus [52]; this paper shows
the weak completeness of L with respect to language models (the sequents provable in L
are precisely those which are valid in all language models). The strong completeness does
not hold, but it holds for the product-free L [10].
Analogously, the intended models for NL are algebras of languages consisting of
phrase structures. Let P denote the set of all phrase structures on . On the powerset of
P one defines ·, \, / as above except that + is replaced by P and xy by (x, y). The
weak and the strong completeness (with respect to these models) hold for the product-free
fragment of NL only [23, 33].
The intended models exhibit Lambek’s interpretation of categories, which is not
the same as in BCGs. For a BCG, the category of type α consists of all (structured)
expressions which are assigned this type by the grammar. According to Lambek, the
basic categories, i.e. those which are assigned atomic types, generate all other categories
by operations ·, \, /, interpreted in the algebra of languages. In particular, if y is of type
α\β (resp. β/α), then, for any x of type α, xy (resp. yx) is of type β in a BCG. Lambek
replaces ‘if . . . then’ by ‘if and only if’. This is an essential difference; it leads to new
reduction patterns, like (L1)–(L5), not admitted in BCGs.
This novel understanding of types caused, probably, a relatively small impact of
Lambek’s approach on his contemporaries. Only in the 1980s there began more systematic
studies in Lambek calculi and their role in type grammars and type-theoretic semantics,
initiated by W. Zielonka and the present author in Poznań and J. van Benthem and his
students (especially M. Moortgat) in Amsterdam. This research was reported in two
collection volumes [17, 50]; the second one also contains reprints of some earlier papers.
The books [11, 43] elaborate on logical and algebraic properties of Lambek calculi and
grammars.
Lambek grammars are defined like BCGs except that the reduction procedure is
replaced with the provability in L, NL or a related system. One employs sequents of
the form α1 , . . . , αn ⇒ β; in algebras, each comma is interpreted as product. For
nonassociative systems, the antecedents of sequents take the form of bracketed sequences,
e.g. (α, (β, γ )), which is different from ((α, β), γ ). So (Red.1), (Red.2) can be written as
(RED.1), (RED.2), and similarly for other laws. WARNING: (RED.1), (RED.2) have been
called reduction rules in Sect. 2, but now the term ‘rule’ is reserved for inference rules of
type logics, e.g. (Res.1), (Res.2), (Cut.1), whereas the provable sequents are referred to as
laws.
Both L and NL can be presented as sequent systems [36, 37]. For L, the axioms are (Id)
and the inference rules are as follows ( and stand for finite, possibly empty, sequences
of types).
, α, β, ⇒ γ ⇒α ⇒β
(· ⇒)
(⇒ ·)
, α · β, ⇒ γ , ⇒ α · β
, β, ⇒ γ ⇒ α α, ⇒ β
(\ ⇒)
(⇒ \)
, , α\β, ⇒ γ ⇒ α\β
, α, ⇒ γ ⇒ β , β ⇒ α
(/ ⇒) (⇒ /)
, α/β, , ⇒ γ ⇒ α/β
, α, ⇒ β ⇒ α
(Cut)
, , ⇒ β
One assumes that is nonempty in (⇒ \), (⇒ /). In sequents, one omits outer
parentheses of antecedent sequences and writes , for the concatenation of and .
The sequent system for NL is similar. The antecedents of sequents are bracketed
sequences of types, hence all rules look a bit differently; see [11, 45].
Clearly these systems are certain intuitionistic sequent systems, types play the role of
formulas, and atomic types of variables (or nonlogical constants). The rules (· ⇒)-(⇒ /)
are the introduction rules for connectives, and (Cut) is the cut rule.
By dropping (Cut), one obtains the cut-free (sequent system for) L. Lambek [36]
proved the cut elimination theorems for L (and in [37] for NL): every provable sequent is
provable in the cut-free system. As a consequence, both systems possess the subformula
property: every provable sequent possesses a proof such that each formula appearing in
this proof is a subformula of a formula occurring in this sequent. Since, additionally, each
introduction rule increases the size of sequents, then the provability in either system is
decidable. It is easy to extract language-restricted fragments. For instance, the product-
free fragment admits product-free formulas only and drops rules (· ⇒), (⇒ ·). L is a
conservative extension of its language-restricted fragments, and similarly for NL.
The product-free L, restricted to (Id), (\ ⇒) and (/ ⇒) ((Cut) is admissible), yields
precisely the correct reduction patterns of BCGs. This system is sometimes denoted by
AB. L is much stronger than AB, but both systems coincide for sequents of the form
α1 , . . . , αn ⇒ p such that p is an atom and no αi contains a compound type on the
argument place. In other words, the order of each αi is at most 1. The order of α, denoted
by o(α), is recursively defined as follows: o(p) = 0 for atomic p,
o(α\β) = o(β/α) = max(o(β), o(α) + 1), o(α · β) = max(o(α), o(β)).
For example, p\q, p\(q\r), (p\q)/r are of order 1, p/(q\p) is of order 2, and so on
(p, q, r are atoms). The product-free L is stronger than any extension of AB by finitely
many new reduction patterns, provable in L [69].
Bar-Hillel et al. [8] shows that every -free CFG G is equivalent to a BCG G with
all types of order at most 1. By the above, the language of G does not change, if one
replaces AB by L. Consequently, every -free CFG is equivalent to a Lambek grammar.
The converse holds as well [51]. Analogous results for NL were obtained in [11, 34].
Due to new laws, Lambek grammars provide a more flexible description of natural
language. We consider atomic types s, n, as above, and n∗ for plural nouns. In BCGs we
get:
1. John likes Jane. n, (n\s)/n, n ⇒ s.
2. John works here. n, n\s, s\s ⇒ s.
3. John never works. n, (n\s)/(n\s), n\s ⇒ s.
4. John works for Jane. n, n\s, (s\s)/n, n ⇒ s.
5. John works and Jane rests. n, n\s, (s\s)/s, n, n\s ⇒ s.
102 W. Buszkowski
6. men work. n∗ , n∗ \s ⇒ s.
7. poor men work. n∗ /n∗ , n∗ , n∗ \s ⇒ s.
8. men works. n∗ , n\s ⇒ s.
9. John work. n, n∗ \s ⇒ s.
Here ⇒ means that the sequent is not provable in AB. The sequents in 8, 9 are unprovable
in L, either.
Now assign s/(n\s) to he and (s/n)\s to her; We abbreviate these types as nps and
npo , respectively, since they correspond to (singular) noun phrase as subject and noun
phrase as object.
10. he likes Jane. s/(n\s), (n\s)/n, n ⇒ s.
11. John likes her. n, n\(s/n), (s/n)\s ⇒ s.
12. he likes her. s/(n\s), (n\s)/n, (s/n)\s ⇒ s.
13. John works for her. n, n\s, (s\s)/n, (s/n)\s ⇒ s.
The sequent in 12 remains unprovable in AB, if even one replaces (n\s)/n by n\(s/n).
In L, these two types are equivalent, by (L5), and this sequent is provable: use (Red.1)
s/n, (s/n)\s ⇒ s, (L3) (to the first and the second type of 12) and (Cut). Also the sequent
in 13 is provable in L. Notice that the student follows the teacher can be
parsed like 12 and John works for a friend like 13 (assign nc to common nouns
and nps /nc , npo /nc to articles).
These examples, similar to those in [36], well illustrate the power of Lambek
grammars. In a BCG we need at least two types of likes (see 10, 11); they are equivalent
in L, hence only one of them is sufficient. To parse 12 in a BCG we need additional types
of words, e.g. (s/n)/((n\s)/n) of he; s/(n\s) ⇒ (s/n)/((n\s)/n) is an instance of
(L4), hence s/(n\s) is sufficient in a Lambek grammar. Even in NL one proves n ⇒ nps ,
n ⇒ npo as instances of (L1). This shows that L provides some logical transformations
of types and explains certain syntactic ambiguities of expressions. Of course, not all; we
still need n\s and n∗ \s for worked, n/n and n∗ /n∗ for poor.
Only four atomic types appear in these examples. Realistic grammars for a natural
language employ much more atoms. Lambek [40] uses 33 atomic types for a fragment of
English, described by a pregroup grammar (see Sect. 3.2.4). We list some of them.
π = subject
π1 = first person singular subject
π2 = second person singular and any plural personal subject
π3 = third person singular subject
s = statement (declarative sentence)
s1 = statement in present tense
s2 = statement in past tense
q̄ = question
q = yes-or-no question
q1 = yes-or-no question in present tense
q2 = yes-or-no question in past tense
i = infinitive of transitive verb
j = infinitive of complete verb phrase
j¯ = complete infinitive with to
o = direct object
n = name
n0 = mass noun
n1 = count noun
n2 = plural noun
n̄ = complete noun phrase
p1 = present participle
p2 = past participle
For semantic considerations, however, it is more natural to reduce the number of atomic
types. Lambek’s n̄ can be defined as s/(n\s) or (s/n)\s, depending on the role in a
sentence (subject or object). Some authors choose np (noun phrase) as an atom and assign
(np\s)/np to transitive verbs, instead of (n\s)/n (this neglects tense and number).
Every proof of ⇒ α in the sequent system of L determines a unique bracketing of
, and similarly for NL. This induces a unique phrase structure of the parsed expression.
Due to associativity, L is ‘structurally omnipotent’: every possible bracketing of comes
from some proof of ⇒ α (if (Cut) can be used). Consequently, Lambek grammars
based on L are not sensitive to phrase structures; they describe languages of strings.
On the contrary, Lambek grammars based on NL naturally describe languages of
phrase structures. Kandulski [32] shows the strong equivalence of these grammars and
BCGs. Therefore some linguists prefer this weaker logic. It is quite weak, indeed; neither
12, nor 13 can be parsed in NL, if the same types are used. Worthy of notice, with NL
one can interchange the roles of functors and arguments. From x : α and y : α\β we infer
(x, y)2 : β, but, using (L1), we obtain x : β/(α\β), hence also (x, y)1 : β.
3.2 Extensions
3.2.1 Multi-Modal Systems
To make it more flexible Moortgat [45] and other authors extend NL in different ways:
↓
admit several products ⊗i with residuals \i , /i and unary modalities ♦i , i , which form
↓
a residuation pair (♦i α ⇒ β and α ⇒ i β are equivalent in models and derivable from
↓
each other in the formal system). From ♦i α ⇒ ♦i α we obtain α ⇒ i ♦i α, and from
↓ ↓ ↓
i α ⇒ i α we obtain ♦i i α ⇒ α.
↓
Let us consider an example from [46]. np can be lifted up to both n ♦n np and
↓
a ♦a np, where the subscripts abbreviate nominative and accusative. We assign np to
↓ ↓ ↓ ↓
John, Mary, n ♦n np to he, she, a ♦a np to him, her, and (n ♦n np\s)/a ♦a np
to likes. The resulting grammar assigns s to John likes Mary, he likes her,
but not to her likes Mary.
Another example comes from [45]. Let r be the type of relative clause, e.g. that
Kazimierz wrote (in the book that Kazimierz wrote). With L we can
assign r/(s/np) to that, which yields that Kazimierz wrote: r, since
r/(s/np), np, (np\s)/np ⇒ r

104 W. Buszkowski
is provable. This sequent, however, is not provable in NL (with any bracketing). We assign
↓
r/(s/♦a a np) to that and admit the weak associativity axiom:
(α · β) · ♦a γ ⇒ α · (β · ♦a γ ).
In NL we prove (np, ((np\s)/np, np)) ⇒ s, and consequently,
(np, ((np\s)/np, ♦a ↓a np)) ⇒ s.
↓
This yields ((np, (np\s)/np), ♦a a np) ⇒ s, by the new axiom, hence
(np, (np\s)/np) ⇒ s/♦a ↓a np,
↓
by (⇒ /). Thus, Kazimierz wrote: s/♦a a np and that Kazimierz wrote: r.
These examples show the spirit of multi-modal Lambek grammars, extensively studied
by a group of contemporary linguists. The unary modalities are used to construct subtypes
and super-types of some types and to restrict associativity and commutativity to some
special cases. (By a subtype of α we mean a type β such that β ⇒ α is true, not a
subformula of α.) This resembles the usage of exponentials !, ? in linear logic, where
structural rules (weakening, contraction) are limited to formulas !α, ?α. More information
on the multi-modal framework can be found in [44–47].
Morrill [48] elaborated Discontinuous Lambek Calculus, a special multi-modal and
multi-sorted logic, intended to process types of discontinuous expressions. In language
models, one admits strings with some occurrences of | (separator); α ⊗i β denotes the
substitution of β for the i-th separator in α. The interpretation of product as substitution
also appeared in [11].
3.2.2 Substructural Logics
One can add lattice connectives ∧, ∨, satisfying the lattice laws. It suffices to add:
α⇒β α⇒γ
α∧β ⇒ α α∧β ⇒ β
α ⇒β∧γ
α⇒γ β⇒γ
α ⇒ α∨β β ⇒ α∨β
α∨β ⇒ γ
to the first axiomatization of L or NL. The corresponding sequent systems (we skip
details) admit cut elimination, hence both logics are decidable. Here are the distributive
laws, provable in NL with ∧, ∨.
α · (β ∨ γ ) ⇔ (α · β) ∨ (α · γ ) (α ∨ β) · γ ⇔ (α · γ ) ∨ (β · γ )
α\(β ∧ γ ) ⇔ (α\β) ∧ (α\γ ) (α ∧ β)/γ ⇔ (α/γ ) ∧ (β/γ )
(α ∨ β)\γ ⇔ (α\γ ) ∧ (β\γ ) α/(β ∨ γ ) ⇔ (α/β) ∧ (α/γ )

Let us note some simple applications of types with ∧, ∨ in type grammars. Lambek
[37] noticed that a type assignment I (v) = {α1 , . . . , αn } could be replaced with the rigid
type assignment I (v) = α1 ∧ · · · ∧ αn . Another application concerns subtypes. Lambek
[40] needs nonlogical assumptions πi ⇒ π, sj ⇒ s, for i = 1, 2, 3, j = 1, 2. Instead one
can define s = s1 ∨ s2 , π = π1 ∨ π2 ∨ π3 and apply a pure logic with ∨ but no nonlogical
assumptions (according to the paradigm of lexicality). Kanazawa [31] proposed feature-
decomposition types: works is of type (np ∧ sg)\s, work of type (np ∧ pl)\s, worked
of type np\s, and became of type ((np\s)/(np ∨ ad), where np, sg, pl, ad are types of
noun phrase, singular, plural, and adjective, respectively.
By L1 we denote L with constant 1 and the axioms:
1·α ⇔α α·1⇔ α.
L1 is strongly complete with respect to residuated monoids, i.e. residuated semigroups

with an element 1 (the unit for product). The sequent system is obtained from that for L
by admitting sequents ⇒ α (interpreted in algebras as 1 ≤ μ(α)), allowing to be empty
in rules (⇒ \), (⇒ /), and adding:
, ⇒ α
(1 ⇒) (⇒ 1) ⇒ 1 .
, 1, ⇒ α
NL1 can be presented in a similar way. Notice that in L1 one proves new laws, not
containing 1, nor the empty antecedent, e.g. α/(α\α) ⇒ α; ⇐ is provable in L. To prove
the former, from α ⇒ α infer ⇒ α\α, by (⇒ \), then apply (/ ⇒). This proof works in
NL1 as well.
The language models for L1 are algebras of subsets of ∗ = + ∪ {}; the operations
·, \, / for languages are defined as above except that + is replaced with ∗ . The
language {} is the unit for product. The intended models for NL1 employ languages of
phrase structures, now enriched with the empty structure such that (, x) = (x, ) = x
for any phrase structure x.
Some linguists object the suitability of L1 as a logic for type grammars. α/α ⇔
(α/α)/(α/α) is provable in L1, hence nc /nc and (nc /nc )/(nc /nc ) are equivalent, but
the former is a natural type of adjectives and the latter of adverbs.
On the other hand, logicians prefer L1 and its extensions. In these systems, some
formulas are provable (a formula α is said to be provable, if ⇒ α is provable); for
example, α\α in NL1 and (α\β)$(γ \α)\(γ \β)) in L1. Furthermore, every sequent is
deductively equivalent to a formula, e.g. α, β ⇒ γ to β\(α\γ ). Accordingly, these logics
can be presented in the form of Hilbert style systems and more easily compared with other
nonclassical logics. For example, the product-free L1 can be axiomatized as a Hilbert style
system with the following axioms and rules.
(a.1) 1 (a.2) 1\(α\α) (a.3) ((α\β)/γ )\(α\(β/γ )) (a.4) (α\(β/γ ))\((α\β)/γ )
(a.5) (α\β)\((γ \α)\(γ \β)) (a.6) ((α/γ )/(β/γ ))/(α/β)
α α\β α\β
(mp$ (\ − /)
β β/α
106 W. Buszkowski
For the 1-free fragment, (a.1), (a.2) are replaced by (id) α\α. Other axiom systems can be
found in [70] and for richer logics in [24].
L1 with ∧, ∨ is called Full Lambek Calculus (FL) and regarded as a basic substructural
logic [24]. Substructural logics can be defined as axiom and rule extensions of FL. They
correspond to some classes (usually varieties or quasi-varieties) of residuated lattices,
i.e. lattice-ordered residuated monoids. One often adds a new constant 0 and defines
negations: ∼ α = α\0, −α = 0/α (0 is interpreted as an arbitrary element of the
residuated lattice).
The term ‘substructural logics’ refers to the fact that sequent systems for these logics
lack some structural rules, characteristic of the Gentzen system for intuitionistic logic:
exchange (e), contraction (c), left weakening or integrality (i), right weakening (o). The
first three rules have the following forms.
, α, β, ⇒ γ , , , ⇒ α
(e) (c)
, β, α, ⇒ γ , , ⇒ α
, ⇒ β
(i)
, α, ⇒ β
They express some algebraic properties of product: (e) a · b = b · a, (c) a ≤ a · a, (i)

a · b ≤ a, a · b ≤ b (with 1 this amounts to a ≤ 1). (o) involves sequents of the form ⇒
(interpreted as μ() ≤ 0). They can be eliminated, and (o) can be replaced by the axiom
, 0, ⇒ α.
For logics with (e), corresponding to commutative algebras, α\β is equivalent to β/α
(in algebras a\b = b/a), and one writes α → β for both. The sequent systems are simpler
(we omit details). Also ∼ α is equivalent to −α, and one writes ¬α for both.
From the algebraic point of view, substructural logics treat implication(s) as residual(s)
of the product operation; the latter usually differs from the lattice meet. Many well-
known nonclassical logics belong to this family: relevant logics, many-valued logics,
fuzzy logics, and intuitionistic and classical logics as the limit cases. For instance,
intuitionistic logic amounts to FL with (e), (c), (i), (o) (in fact, (e) is derivable with
(c), (i)), and Łukasiewicz infinitely valued logic to FL with (e), (i), (o) and the axiom
α ∨ β ⇔ (α → β) → β [24].
Linear logic of Girard [26] can be presented as FL with 0, (e) and the double negation
axiom ¬¬α ⇒ α (⇐ is provable); we neglect exponentials !, ?. Noncommutative versions
are due to Yetter [68] and Abrusci [1]. The former can be presented as FL with 0 and the
axioms α\0 ⇔ 0/α (hence ∼ α and −α collapse in ¬α) and ¬¬α ⇒ α; the latter as FL
with 0 and the axioms ∼ −α ⇒ α, − ∼ α ⇒ α (again ⇐ are provable). Both logics are
conservative extensions of FL without 0 [2]. In [24] they are called Cyclic Involutive Full
Lambek Calculus (CyInFL) and Involutive Full Lambek Calculus (InFL), respectively.
Cut-free sequent systems of linear logics look differently. Formulas are built from
atoms by ⊗, ⊕ and negation(s), where ⊗ stands for product and ⊕ for the dual product
(‘par’). In algebras, a ⊕ b = ¬(¬b ⊗ ¬a) for Girard’s logic and CyInFL and a ⊕ b =∼
(−b ⊗−a) for InFL. One employs classical sequents ⇒ or one-sided sequents only:
either ⇒ (Schütte style), or ⇒ (dual Schütte style). Each comma in is interpreted
as product and in as dual product. In InFL, presented in this way, one can define \, /
as follows: α\β =∼ α ⊕ β, α/β = α ⊕ −β.
In the literature on linear logics, ⊗, ⊕, \, 1, 0 and negation(s) are referred to as

multiplicatives and ∧, ∨ (also constants , ⊥, interpreted as the greatest and the least
element) as additives. According to a different tradition, they are intensional and
extensional connectives and constants, respectively. L1 is often characterized as the
intuitionistic fragment of multiplicative linear logic.
Type grammars usually employ basic intuitionistic substructural logics, often not
admitting empty antecedents of sequents and being restricted to multiplicative connectives
(also multi-modal). Nonetheless the impact of linear logics (which are ‘classical’) can be
seen in current developments. I have already noted an analogy between modalities in type
grammars and exponentials of Girard [26]. Also proof nets, i.e. a representation of proofs
in multiplicative linear logics by means of some graphs of links between formulas, are
used as representations of syntactic structures in type grammars, either directly, or in a
form suitable for intuitionistic fragments. We cannot discuss this matter here; the reader
is referred to [46].
In language models, ∧, ∨ can be interpreted as the set theoretic intersection and union
of languages. Then, we obtain a distributive lattice. The distributive laws for ∧, ∨ are not
provable in FL, nor other logics, discussed above. One can add them as new axioms; it
suffices to add:
(D) α ∧ (β ∨ γ ) ⇒ (α ∧ β) ∨ (α ∧ γ ) .
Nevertheless, some interesting linguistic interpretations of logics without (D) are

possible. Clark [18] introduced syntactic concept lattices as a special kind of concept
lattices from lattice theory. Let L0 ⊆ ∗ be a fixed language. Pairs (x, y), for x, y ∈ ∗ ,
are called contexts. For a set of contexts X, one defines X as the set of all z ∈ ∗ such
that xzy ∈ L0 , for all (x, y) ∈ X. The sets of the form X are called syntactic concepts
for L0 . They can be interpreted as the syntactic categories determined by L0 , a reasonable
generalization of Husserl’s idea, followed by Ajdukiewicz. Since L provides nontrivial
laws α ⇒ β, syntactic categories in Lambek grammars cannot be equivalence classes
(substitution classes). The family of syntactic concepts for L0 is a complete residuated
lattice with operations: X ∧ Y = X ∩ Y , X ∨ Y (resp. X ⊗ Y , 1) equal to the smallest
concept containing X ∪ Y (resp. X · Y , {}), and \, / defined as for languages.
3.2.3 Semantic Types
The product-free L with (e) was studied by van Benthem [62, 63] as a logic of semantic
types; we call this logic the Lambek-van Benthem calculus (LB). Proofs in a natural
deduction system (ND-system) for LB can be encoded by some terms of typed lambda
calculus, namely linear terms (i.e. every λ binds exactly one occurrence of a variable),
satisfying the additional constraint: no subterm is closed. This is an adaptation of the
‘Curry-Howard isomorphism’ between ND-proofs and lambda terms [55]. Since every
ND-proof in L can be translated into an ND-proof in LB, the former determines a
unique lambda term; this lambda term, interpreted in a standard type-theoretic model
(see Sect. 2), denotes a semantic transformation corresponding to the syntactic parsing
in the grammar. Size limits do not allow us to discuss this framework in detail. Let us
108 W. Buszkowski
consider one example. Recall that the characteristic inference rules of ND-systems are
the introduction rules and the elimination rules for connectives.
From n ⇒ n and n\s ⇒ n\s we get n, n\s ⇒ s, by the \-elimination rule: from
⇒ α and ⇒ α\β infer , ⇒ β (in an ND-system for L). This is translated in LB
as: from e → t ⇒ e → t and e ⇒ e infer e → t, e ⇒ t, by the →-elimination rule (in
an ND-system for LB). In L we obtain n ⇒ s/(n\s), by the /-introduction rule, which is
translated into e ⇒ (e → t) → t in LB. The ND-proof in LB is encoded by the term:
λy e→t .y e→t x e .
This is a linear term, satisfying the additional constraint. In a type-theoretic model, if x e is

valuated as a ∈ De (an individual), then this term denotes the family of all (characteristic
functions of) X ⊆ De such that a ∈ X. Thus, the syntactic law n ⇒ s/(n\s) corresponds
to the semantic transformation which sends an individual into the family of all properties
(interpreted extensionally) of this individual. In particular, any proper noun (denoting an
individual) can be treated as a noun phrase (denoting a generalized quantifier, i.e. a family
of sets of individuals).
We have explained a semantic interpretation of (L1). Similarly, (L3) correspond to the
composition of functions, (L5) to the interchange of arguments, and so on. An interesting
theorem of [63] shows that every provable sequent of LB admits only finitely many
different proofs up to the equality in lambda calculus; so every expression possesses only
finitely many ‘semantic readings’. Further studies on this topic can be found in [43, 45].
Analogous correspondences were elaborated for richer logics [66]. Lambek [38]
studied category-theoretic interpretations of L and its extensions. Abstract Categorial
Grammars, introduced by de Groote [20], employ linear lambda-terms as representations
of both syntactic structures and semantic structures (logical forms) of expressions in
natural language with certain homomorphisms between them.
3.2.4 Pregroup Grammars
Lambek [39] proposed another extension of L1, called compact bilinear logic (CBL). It
corresponds to pregroups, i.e. ordered algebras (A, ·,r ,l , 1, ≤) such that (A, ·, 1, ≤) is a
partially ordered monoid and r ,l are unary operations, satisfying the adjoint laws:
a l · a ≤ 1 ≤ a · a l and a · a r ≤ 1 ≤ a r · a ,
for any a ∈ A. a r (resp. a l ) is called the right (resp. left) adjoint of a. (This
terminology is transferred from category theory.) Pregroups coincide with the algebras
for the multiplicative fragment of noncommutative linear logic of Abrusci [1] such that ⊗
equals ⊕ and 1 = 0. The residuals of product are defined by: a\b = a r · b, a/b = a · bl .
We have: a rl = a lr = a, (a · b)r = br · a r , and similarly for l . Adjoints reverse the
ordering: if a ≤ b then b r ≤ a r and b l ≤ a l .
CBL is a logic of free pregroups. From atoms p, q, r, . . . one builds simple types p(n) ,
where n is an integer. p(0) is interpreted as p, p(n) , n > 0, as pr...r (n times), and p(n) ,
n < 0, as pl...l (|n| times). Pregroup types are finite strings of simple types.
One also assumes that the set of atoms is partially ordered by a relation . The relation
⇒, between pregroup types, is defined by the following rewriting rules:
• (Contraction) X, p(n) , p(n+1) , Y ⇒ X, Y ,
• (Expansion) X, Y ⇒ X, p(n+1) , p(n) , Y ,
• (Induced Step) X, p(n) , Y ⇒ X, q (n) , Y , if either p q and n is even, or q p and n
is odd.
U ⇒ V holds, if U can be transformed into V by finitely many applications of these
rules.
Pregroup grammars are defined as Lambek grammars except that L is replaced by
CBL (and types of L by pregroup types). Lambek [39] shows that (Expansion) can be
eliminated from proofs of X ⇒ p, where p is an atom. Furthermore, (Induced Step) and
(Contraction) can be collapsed with one rule of generalized contraction:
• (GCON) X, p(n) , q (n+1) , Y ⇒ X, Y , with the same condition as in (Induced Step).
If α1 , . . . , αn are assigned to v1 , . . . , vn , respectively, then the grammar assigns p to
v1 . . . vn , if the concatenation α1 · · · αn reduces to p by a finite number of applications of
(GCON) and, possibly, (Induced Step) at the end of the reduction. Such derivations can
be presented by means of links, joining the reduced types of (GCON).
For example, we assume she: π3 , will: π r s1 j l , see: iol and him: o, where i, j
are types of infinitive of intransitive verb and infinitive of any complex verb phrase, and
π, π3 , o are understood as above. We also assume π3 π, i j . Then, she will
see him is assigned s1 , since:
π3 , π r s1 j l , iol , o ⇒ s1 .
The reduction can be depicted as follows:
π3 , π r s1 j l , i ol , o
where each link corresponds to one application of (GCON).

With man: n1 (count noun), whom: nr1 n1 oll s l , saw: π r s2 ol and s2 s, one assigns n1
to man whom she saw, by the reduction:
n1 , nr1 n1 oll s l , π3 , π r s 2 ol .
These examples come from [40] (up to minor changes), where Lambek analyzed many
basic grammatical constructions of English within the pregroup framework. In other
publications he and his collaborators applied this approach to several languages: German,
French, Italian, Polish and some non-European languages; see [40] for references.
Parsing by pregroups is computationally simple; it runs in polynomial time [14],
whereas L is NP-complete [53]. CBL is stronger than L1: (p/((p/p)/p))/p ⇒ p is
provable in CBL (define / as above), but not in L1. The logical meaning of the new laws is
not clear; the latter does not hold even in classical logic (interpret / as implication with the
antecedent on the right). No type-theoretic semantics for pregroup grammars is known.
110 W. Buszkowski
It seems that CBL is an algebraic calculus rather than a genuine logic. This opinion is
confirmed by the fact that bounded pregroups are trivial (one-element) algebras, hence
CBL with is inconsistent [14]. (The latter paper shows that pregroup grammars are
equivalent to CFGs.)
On the other hand, all linguistic examples, analyzed by Lambek and other authors by
means of pregroups, can easily be parsed with L. We return to man whom she saw.
The pregroup types, given above, are translations of L-types; e.g. whom: (n1 \n1 )/(s2 /o),
saw: (π\s2 )/o. The sequent:
n1 , (n1 \n1 )/(s/o), π3 , (π\s2 )/o ⇒ n1
is provable in L augmented with s2 ⇒ s, π3 ⇒ π. Therefore, the semantics for these

examples can be transferred from L.
3.2.5 Modal Logics
At the end, we consider other modal logics, extending L and NL. Buszkowski and
Farulewski [16] studied NL with ∧, ∨, which satisfy the laws of a distributive lattice,
and its extensions with either classical negation (BFNL), or intuitionistic implication
and , ⊥ (HFNL); these logics were presented as sequent systems with cut. Hilbert-
style systems for the latter logics, denoted by NLC and NLI, were studied in [29, 30].
The connectives are ∧, ∨, ⇒, ¬ (now ⇒ stands for the classical or intuitionistic
implication, and ¬ for the classical or intuitionistic negation) and Lambek connectives
·, \, /. Lambek’s sequents α ⇒ β are treated as conditionals. NLC (resp. NLI) can be
axiomatized by all tautologies of classical (resp. intuitionistic) propositional logic in the
extended language and the rules: modus ponens for ⇒, (Res.1), (Res.2). In the associative
versions LC, LI one adds axioms (A.1), (A.2). The following formulas, similar to the
modal axiom (K), are provable in NLI, hence also in NLC, LI, LC (we assume that \, /
bind stronger than ⇒).
γ \(α ⇒ β) ⇒ (γ \α ⇒ γ \β) (α ⇒ β)/γ ⇒ (α/γ ⇒ β/γ )
It should be emphasized that the theorems (i.e. provable formulas) of these systems
are -theorems: they satisfy μ(α) = in algebras. In substructural logics one usually
considers 1-theorems (1 ≤ μ(α) in algebras). Both notions collapse for substructural
logics with (i). LC is a conservative extension of L, and NLC of NL.
NLC, LC and NLI, LI are, in fact, some classical and intuitionistic multi-modal
logics; product and its residuals are binary modalities. This perspective was already
admitted in Arrow Logic of van Benthem [64] and multi-modal versions of Lambek
calculi. Kaminski and Francez [29, 30] study relational frames for NLC, LC, NLI, LI,
proving some completeness and decidability results. Interestingly, the undecidability of
LC follows from some results of [35], whereas LI is decidable [30]. For NLC, NLI even
the consequence relations are decidable [16].
3.3 Lambek Versus Ajdukiewicz
Although Lambek logics are much stronger than AB, the parsing procedure in Lambek
grammars can be carried out in a similar way as in BCGs. The action of L and related
systems can be reduced to the lexical level: the type lexicon is extended by new types,
derivable from the initial types in the system.
For example, if α, β, γ ⇒ δ is provable in L, then α ⇒ (δ/γ )/β is provable, by
(⇒ /), and the sequent:
(δ/γ )/β, β, γ ⇒ δ
is provable in AB. For NL, if α, (β, γ ) ⇒ δ is provable, then α ⇒ δ/(β · γ ) is provable,

and δ/(β ·γ ), (β, γ ) ⇒ δ is provable in AB (with product; see [32]). To eliminate product,
one can use β ⇒ (α\δ)/γ and prove in AB:
α, (α\δ)/γ , γ ⇒ δ .
For vi : αi , i = 1, . . . , n, a successful parsing can be arranged as in Fig. 2 (αi ⇒ βi is

provable in a Lambek logic).
In the same way one can arrange semantic derivations: the semantic transformations,
definable in (a fragment of) lambda calculus, can be performed on the initial denotations
of words in a type-theoretic model, and the denotations of compound expressions are
obtained by the (iterated) application of functions to their arguments.
Such laws as (L1), (L2), (L4) produce infinitely many types β derivable from a single
type α. For instance, starting from n, one derives:
n ⇒ s/(n\s) ⇒ s/((s/(n\s))\s) ⇒ · · · .
Nonetheless only finitely many of them are really needed to parse any expression in a
particular grammar. Buszkowski [12] shows that every type grammar G, based on L, is
equivalent to a BCG G whose type lexicon extends that of G by finitely many new types,
derivable in L from those in the type lexicon of G. The same was earlier shown for NL in
[32].
These results seem to support the opinion that Lambek logics can be regarded as
general logics of syntactic or semantic types rather than type processing systems in type
grammars. The former explain deeper reasons for syntactic ambiguities of expressions and
guide our choice of lexical types. On the other hand, parsing can be based on the classical
type reduction procedure, proposed by Ajdukiewicz, with necessary modifications.
Fig. 2 Reducing L-parsing v1 ... vn

to AB-parsing α1 ... αn ⇒ s
⇓ ... ⇓
AB
β1 ... βn ⇒ s
112 W. Buszkowski
This opinion is non-orthodox. Many authors maintain the priority of Lambek logics,
directly applied in grammars, according to the general paradigm of parsing as deduction.
They, however, usually ignore the problems of efficiency. Parsers for BCGs can be
designed like for CFGs; they run in cubic time in the length of the parsed expression. This
is impossible for type grammars based on L, which is NP-complete [53]. Type grammars
with NL remain polynomial [21], but parsers are not as simple as for BCGs.
At the end of this subsection, let me mention some developments in type grammars,
which are closer to Ajdukiewicz.
Combinatory Categorial Grammars (CCGs), developed by M. Steedman, A. Szabolcsi
and others, enrich AB with finitely many new reduction patterns, semantically corre-
sponding to some combinators, i.e. closed lambda-terms; see [57] for an overview. This
direction continues certain ideas of Curry [19] and Shaumyan [54]. Some of the new
patterns are provable in L, but others require a stronger logic (some instances of exchange
and contraction). The Ajdukiewicz procedure enriched with composition laws (similar to
(L3), (L4)) was earlier proposed by Geach [25].
Categorial Unification Grammars (CUGs), studied by Uszkoreit [61], admit poly-
morphic types, containing variables, which range over a family of types. The simplest
example is (x\x)/x as the type of and. In the course of parsing, one applies the reduction
rules of BCGs and a unification algorithm. For instance, α, β\γ ⇒ σ (γ ), where σ is a
substitution such that σ (α) = σ (β).
L with ∧, ∨ can generate some non-context-free languages, e.g. the intersection of
two context-free languages [31]. This also holds for grammars based on AB with ∧, ∨.
Other frameworks going beyond the context-free world are Tupled Pregroup Grammars
[56] and Categorial Dependency Grammars [22]. Both approaches employ very restricted
types only; the resulting grammars might be presented as BCGs with all types of order at
most 1 and certain constraints imposed on reductions.
References
1. Abrusci, V.M.: Phase semantics and sequent calculus for pure noncommutative classical linear logic.
J. Symb. Log. 56, 1403–1451 (1991)
2. Abrusci, V.M.: Classical conservative extensions of Lambek calculus. Studia Logica 71, 277–314
(2002)
3. Ajdukiewicz, K.: W sprawie ‘uniwersaliów’ [On the problem of ‘universals’]. Przeglad ˛ Filozoficzny
37, 219–234 (1934)
4. Ajdukiewicz, K.: Die syntaktische Konnexität (Syntactic connexion). Studia Philosophica 1, 1–27
(1935)
5. Ajdukiewicz, K.: Zwiazki
˛ składniowe mi˛edzy członami zdań oznajmujacych ˛ (Syntactic connections
between constituents of declarative sentences). Studia Filozoficzne 6(21), 73–86 (1960)
6. Ajdukiewicz, K.: Giedymin, J. (ed.) The Scientific World-Perspective and Other Essays, pp. 1931–
1963. D. Reidel, Dordrecht (1978)
7. Bar-Hillel, Y.: A quasi-arithmetical notation for syntactic description. Language 29, 47–58 (1953)
8. Bar-Hillel, Y., Gaifman, C., Shamir, E.: On categorial and phrase structure grammars. Bull. Res.
Counc. Isr. F9, 155–166 (1960)
9. Bocheński, J.: On the syntactical categories. New Scholasticism 23, 257–280 (1949)
10. Buszkowski, W.: Some decision problems in the theory of syntactic categories. Zeitschrift f.
mathematische Logik und Grundlagen der Math. 28, 539–548 (1982)
11. Buszkowski, W.: Logiczne podstawy gramatyk kategorialnych Ajdukiewicza-Lambeka [Logical

foundations of Ajdukiewicz-Lambek categorial grammars]. Państwowe Wydawnictwo Naukowe
(1989)
12. Buszkowski, W.: Extending Lambek grammars to basic categorial grammars. J. Log. Lang. Inf. 5,
279–295 (1996)
13. Buszkowski, W.: The Ajdukiewicz calculus, Polish notation and Hilbert-style proofs. In: Woleński, J.
(ed.) The Lvov-Warsaw School and Contemporary Philosophy, pp. 241–252. Springer Netherlands,
Dordrecht (1998)
14. Buszkowski, W.: Lambek grammars based on pregroups. In: Logical Aspects of Computational
Linguistics. Lecture Notes in Computer Science, vol. 2099, pp. 95–109. Springer, Berlin (2001)
15. Buszkowski, W.: Syntactic Categories and Types: Ajdukiewicz and Modern Categorial Grammars. In:
Brożek, A., et al. (eds.) Tradition of the Lvov-Warsaw School: Ideas and Continuations, pp. 35–71.
Leiden, Boston, Brill-Rodopi (2016)
16. Buszkowski, W., Farulewski, M.: Nonassociative Lambek calculus with additives and context-free
languages. In: Languages: From Formal to Natural. Lecture Notes in Computer Science, vol. 5533,
pp. 45–58. Springer, Berlin (2008)
17. Buszkowski, W., Marciszewski, W., van Benthem, J. (eds.): Categorial Grammar. John Benjamins,
Amsterdam (1988)
18. Clark, A.: A learnable representation of syntax using residuated lattices. In: Formal Grammar. Lecture
Notes in Computer Science, vol. 5591, pp. 183–198. Springer, Berlin (2011)
19. Curry, H.B.: Some logical aspects of grammatical structure. In: Jakobson, R. (ed.) Structure of
Language and Its Mathematical Aspects. Proceedings of Symposium in Applied Mathematics, pp.
56–68. American Mathematical Society, Providence, RI (1961)
20. de Groote, Ph.: Towards abstract categorial grammars. In: Proceedings of 39th Annual Meeting
and 10th Conference of the European Chapter of ACL, pp. 148–155. Association of Computational
Linguistics (2001)
21. de Groote, Ph., Lamarche, F.: Classical Non-associative Lambek calculus. Studia Logica 71, 355–388
(2002)
22. Dekhtyar, M., Dikovsky, A., Karlov, B.: Categorial dependency grammars. Theor. Comput. Sci. 579,
33–63 (2015)
23. Došen, K.: A brief survey of frames for the Lambek calculus. Zeitschrift f. mathematische Logik und
Grundlagen der Math. 38, 179–187 (1992)
24. Galatos, N., Jipsen, P., Kowalski, T., Ono, H.: Residuated Lattices: An Algebraic Glimpse at
Substructural Logics. Elsevier, Amsterdam (2007)
25. Geach, P.T.: A program for syntax. Synthese 22, 3–17 (1971)
26. Girard, J.-Y.: Linear logic. Theor. Comput. Sci. 50, 1–102 (1987)
27. Husserl, E.: Logische Untersuchungen Halle (1900–1901)
28. Jakobson, R. (ed.): Structure of Language and Its Mathematical Aspects. In: Proceedings of
Symposium in Applied Mathematics. American Mathematical Society, Providence, RI (1961)
29. Kaminski, M., Francez, N.: Relational semantics of the Lambek calculus extended with classical
propositional logic. Studia Logica 102, 479–497 (2014)
30. Kaminski, M., Francez, N.: The Lambek calculus extended with intuitionistic propositional logic.
Studia Logica 104, 1051–1082 (2016)
31. Kanazawa, M.: The Lambek calculus enriched with additional connectives. J. Log. Lang. Inf. 1, 141–
171 (1992)
32. Kandulski, M.: Phrase-structure languages generated by categorial grammars with product. Zeitschrift
f. mathematische Logik und Grundlagen der Math. 34, 373–383 (1988)
33. Kandulski, M.: The non-associative Lambek calculus. In: Buszkowski, W., Marciszewski, W., van
Benthem, J. (eds.) Categorial Grammar, pp. 141–151. John Benjamins, Amsterdam (1988)
34. Kandulski, M.: The equivalence of nonassociative Lambek categorial grammars and context-free
grammars. Zeitschrift f. mathematische Logik und Grundlagen der Math. 34, 41–52 (1988)
35. Kurucz, A., Németi, I., Sain, I., Simon, A.: Decidable and undecidable logics with a binary modality.
J. Log. Lang. Inf. 4, 191–206 (1995)
114 W. Buszkowski
36. Lambek, J.: The mathematics of sentence structure. Am. Math. Mon. 65,154–170 (1958)
37. Lambek, J.: On the calculus of syntactic types. In: Jakobson, R. (ed.) Structure of Language and
Its Mathematical Aspects. In: Proceedings of Symposium in Applied Mathematics, pp. 166–178.
American Mathematical Society, Providence, RI (1961)
38. Lambek, J.: Deductive systems and categories I. J. Math. Syst. Theory 2, 287–318 (1968)
39. Lambek, J.: Type grammars revisited. In: Logical Aspects of Computational Linguistics. Lecture
Notes in Computer Science, vol. 1582, pp. 1–27. Springer, Berlin (1999)
40. Lambek, J.: From Word to Sentence: A Computational Algebraic Approach to Grammar. Polimetrica,
Milano (2008)
41. Leśniewski, S.: Grundzüge eines neuen System der Grundlagen der Mathematik. Fundam. Math. 14,
1–81 (1929)
42. Montague, R.: English as a formal language. In: Thomason, R. (ed.) Formal Philosophy: Selected
Papers of Richard Montague, pp. 188–221. Yale University Press, New Haven (1974)
43. Moortgat, M.: Categorial Investigations. Logical and Linguistic Aspects of the Lambek Calculus.
Foris, Dordrecht (1988)
44. Moortgat, M.: Multimodal linguistic inference. J. Log. Lang. Inf. 5, 349–385 (1996)
45. Moortgat, M.: Categorial type logics. In: van Benthem, J., ter Meulen. A. (eds.) Handbook of Logic
and Language, pp. 93–177. Elsevier/MIT, Amsterdam/Cambridge (1997)
46. Moot, R., Retoré, C.: The Logic of Categorial Grammars: A Deductive Account of Natural Language
Syntax and Semantics. Lecture Notes in Computer Science, vol. 6850. Springer, Berlin (2012)
47. Morrill, G.: Type-Logical Grammar: Categorial Logic of Signs. Kluwer, Dordrecht (1994)
48. Morrill, G.: Categorial Grammar: Logical Syntax, Semantics, and Processing. Oxford University
Press, Oxford (2011)
49. Nowaczyk, A.: Categorial languages and variable-binding operators. Studia Logica 37, 27–39 (1978)
50. Oehrle, R.T., Bach, E., Wheeler, D. (eds.): Categorial Grammars and Natural Language Structures.
D. Reidel, Dordrecht (1988)
51. Pentus, M.: Lambek grammars are context-free. In: Proceedings of 8th IEEE Symposium on Logic in
Computer Science, pp. 429–433 (1993)
52. Pentus, M.: Models for the Lambek calculus. Ann. Pure Appl. Logic 75, 179–213 (1995)
53. Pentus, M.: Lambek calculus is NP-complete. Theor. Comput. Sci. 357, 186–201 (2006)
54. Shaumyan, S.: Applicational Grammar as a Semantic Theory of Natural Language. Edinburgh
University Press, Edinburgh (1977)
55. Sørensen, M.H., Urzyczyn, P.: Lectures on the Curry-Howard Isomorphism. Elsevier, Amsterdam
(2006)
56. Stabler, E.: Tupled pregroup grammars. In: Casadio, C., Lambek, J. (eds.) Computational Algebraic
Approaches to Natural Language, pp. 23–52. Polimetrica, Milano (2008)
57. Steedman, M.: Categorial grammar. Lingua 90, 221–258 (1993)
58. Suszko, R.: Syntactic structure and semantical reference I. Studia Logica 8, 213–244 (1958)
59. Suszko, R.: Syntactic structure and semantical reference II. Studia Logica 9, 63–91 (1960)
60. Tałasiewicz, M.: Philosophy of Syntax. Foundational Topics. Trends in Logic, vol. 29. Springer,
Berlin (2010)
61. Uszkoreit, H.: Categorial unification grammar. In: Proceedings of 11th International Conference on
Computational Linguistics, pp. 187–194 (1986)
62. van Benthem, J.: Essays in Logical Semantics. D. Reidel, Dordrecht (1986)
63. van Benthem, J.: The semantics of variety in categorial grammar. In: Buszkowski, W., Mar-
ciszewski, W., van Benthem, J. (eds.) Categorial Grammar, pp. 37–55. John Benjamins, Amsterdam
(1988)
64. van Benthem, J.: Language in Action. Categories, Lambdas and Dynamic Logic. North-Holland,
Amsterdam (1991)
65. van Benthem, J., ter Meulen. A. (eds.): Handbook of Logic and Language. Elsevier/MIT, Amster-
dam/Cambridge (1997)
66. Wansing, H.: The Logic of Information Structures. Ph.D. Thesis, University of Amsterdam (1992)
67. Wybraniec-Skardowska, U.: Theory of Language Syntax. Kluwer, Dordrecht (1991)
68. Yetter, D.N.: Quantales and (non-commutative) linear logic. J. Symb. Log. 55, 41–64 (1990)
69. Zielonka, W.: Axiomatizability of Ajdukiewicz-Lambek calculus by means of cancellation schemes.

Zeitschrift f. mathematische Logik und Grundlagen der Math. 27, 215–224 (1981)
70. Zielonka, W.: Weak implicational logics related to the Lambek calculus - Gentzen versus Hilbert
formalisms. In: Towards Mathematical Philosophy. Trends in Logic, vol. 28, pp. 201–212. Springer,
Berlin (2009)
W. Buszkowski ()
Adam Mickiewicz University, Poznań, Poland
Material Implication and Conversational
Implicature in Lvov-Warsaw School
Rafal Urbaniak and Michał Tomasz Godziszewski
Abstract The relation between indicative conditionals in natural language and material
implication wasn’t a major topic in the Lvov-Warsaw school. However, a major defense
of the claim that the truth conditions of these two are the same has been developed by
Ajdukiewicz (Studia Logica IV:117–134, 1956). The first major goal of this paper is to
present, assess, and improve his strategy. It turns out that it is quite similar to the approach
developed by Grice (Studies in the Way of Words, Harvard University Press, Cambridge,
MA, 1991), so our second goal is to compare these two and to argue that the accuracy of
Ajdukiewicz’s explanation is less dependent on controversial properties of a systematic
but convoluted general theory of cooperative communicative behavior. In Lvov-Warsaw
school the relation between material implication and indicative conditionals was also
discussed by Gołab ˛ (Matematyka 3(5):27–29, 1949) and Słupecki (Matematyka 4(6):32–
35, 1949), so the third part of our paper is devoted to their discussion and relating it to
Ajdukiewicz’s views.
Keywords Material implication · Counterfactuals · Conversational implicature ·

Natural language · Assertibility · Pragmatics · Cancellability
Mathematics Subject Classification (2000) Primary 03A05, 03B05; Secondary 09B65,

94A99
1 Ajdukiewicz’s Equivalence Argument
Ajdukiewicz meant to establish that any conditional of the form:
If p, then q
has the same truth conditions as the material implication
p→q

118 R. Urbaniak and M. T. Godziszewski
(which is false just in case p is true and q false, and true otherwise). His argument has
two main stages:
STAGE 1 Showing that natural language disjunction—not-p or q—has the same truth
conditions as its corresponding natural language conditional—if p, then q.
STAGE 2 Showing that this natural language disjunction has the same truth conditions
as its corresponding classical disjunction—p ∨ q.
If we mark having the same truth conditions by ‘⇔’, the structure of the argument is
this (we make no distinction between “either . . . or . . . ” and “. . . or . . . ”):
[If p, then q] ⇔ [Either not-p or q] (by Stage 1) (1)

[Either not-p or q] ⇔ [¬p ∨ q] (by Stage 2) (2)
[¬p ∨ q] ⇔ [p → q] (by logic) (3)
[If p, then q] ⇔ [p → q]( by transitivity of ⇔) (4)
Let’s focus on STAGE 1 now (we’ll get to STAGE 2 in Sect. 3). Ajdukiewicz gives two
separate arguments for two directions of the equivalence.
⇐:
w.t.s.: [Either not-p or q] ⇒ [If p, then q]
Step 1 (suppose)
Either not-p or q.
Step 2 (from Step 1 by disjunctive syllogism)
If it is not the case that not-p, then q.
Step 3 (from Step 2 by double negation elimination in the antecedent)
If p, then q.
⇒:
w.t.s.: [If p, then q] ⇒ [Either not-p or q]
Step 1 (suppose for contraposition)
It is not the case that [either not-p or q].
Step 2 (from Step 1 by De Morgan’s law)
Neither (not-p) nor q.
Step 3 (from Step 2, assuming conditionals with true antecedents
and false consequents are false)
It is not the case that [if p, then q].
2 Jackson’s Argument
Before we continue with the discussion of Ajdukiewicz’s approach to conditionals, let’s

pause to observe that Jackson [4, pp. 4-6], without reference to Ajdukiewicz (and, we take
Material Implication and Conversational Implicature in Lvov-Warsaw School 119
it, quite independently) formulated a very similar argument. He starts with the following
principles:
Truth-functionality (TF) It is correct to represent English ‘not’, ‘or’ and ‘and’ with
¬, ∨, ∧.
Uncontested principle (UP) For an indicative conditional to be false it is sufficient that
its antecedent is true and the consequent false.
Passage principle(s) (PP) The following reasoning patterns are valid:
p or q. Therefore if not-p, then q.
Not-(p and q). Therefore if p, then not-q.
To pass from the material conditional to the indicative conditional, Jackson relies on
PP and gives two variants of the argument. The first variant is pretty much the same as in
Ajdukiewicz’s argument:
w.t.s.: [Either not- p or q] ⇒ [If p, then q]

Step 1 (suppose)
Either not- p or q.
Step 2 (from Step 1 by first PP)
If it is not the case that not- p, then q.
Step 3 (from Step 2 by double negation elimination)
If p, then q.
The second variant is a bit different. First we notice that the material implication has
the same truth conditions as not-(p and not-q). Then we argue:
w.t.s.: Not (p and not-q) ⇒ [If p, then q]

Step 1 (suppose)
Not (p and not- q).
Step 2 (from Step 1 by second PP)
If p, then not-(not- q).
Step 3 (from Step 2 by double negation elimination)
If p, then q.
For the other direction, Jackson observes that the uncontested principle allows one to
pass from ¬(p → q) to not-(if p, then q) and applies contraposition. This results in a
move very similar to Ajdukiewicz’s argument for ⇒.
3 Ajdukiewicz on Disjunction
Let’s turn to Ajdukiewicz’s STAGE 2. The correspondence between classical disjunction

and natural language ‘or’ isn’t terribly problematic. Yet, Ajdukiewicz’s way of handling
it is quite interesting, because while dealing with it he is forced to make a distinction
between truth and assertibility. Let’s start with quoting Ajdukiewicz in extenso. First,
Ajdukiewicz makes the audience agree that the truth conditions of the relevant sentences
are the same:
I take two pieces of chalk and hide them in my hands, so that the audience doesn’t see in which
hand I placed each piece: both in my right hand, both in my left hand, or one in my left hand
and one in my right hand. Having done this I ask the audience: “am I holding chalk in my left or
right hand?” and the response is unanimous and positive. I point out to the audience that there are
three possibilities (chalk only in my left hand, chalk only in my right hand, chalk in both hands)
and I ask whether no matter which of these is the case they still sustain their agreement with the
disjunction: “[a piece of] chalk is in my right or left hand” and whether the only case in which they
would consider the disjunction to be false would be if I didn’t have chalk in any of my hands—and
again, the response is positive. In this way I once again obtain agreement that disjunction in natural
language is true if at least one of its arguments is true and false if none of them is. [1, p. 252]
Next, he points out that agreement doesn’t have to go hand in hand with truth:
Now I open my hands and show the audience that both pieces of chalk are, say, in my left hand.
In this situation I ask again the same question: am I holding chalk in my left or right hand? This
time, the audience is far from unanimosity, and the most popular opinion is that once it is known
that chalk is in my left hand, one cannot give a positive response to the disjunctive question. Some
would even say that in such a situation one can give neither a positive nor a negative response.
This semantic experiment teaches us that we accept a disjunction as long as we know that one of
the disjuncts is true, but we don’t know which one. Once we find out which of the disjuncts is true,
we no longer accept the disjunction. [1, p. 252]
Ajdukiewicz argues that our reluctance to accept the disjunction notwithstanding,

it is still true. The disjunction is about pieces of chalks and hands, not about any-
one’s knowledge, and since the states of affairs involving these pieces of chalk and
Ajdukiewicz’s hands haven’t changed, nor did the truth-value of the disjunction.
Indeed, he points out that suggesting otherwise and claiming that “p or q” really means
“At least one of p and q is true and I don’t know which” quickly leads to absurdity. For
instance:
For any natural number x, x is even, or x is odd.
is true, whereas:
For any natural number x, at least one of “x is even” and “x is odd” is true, but I don’t know which.
is plainly false, since it implies, among other things:

at least one of “2 is even” and “2 is odd” is true, but I don’t know which.
So, it seems, we refuse to assert the disjunction even though it is true, and more
is required for a sentence to be assertible than its truth. What would be the principles
governing assertibility resulting in this disparity? Ajdukiewcz starts with a discussion of
Quine’s proposal.
4 Ajdukiewicz vs. Quine on Assertibility
Ajdukiewicz doesn’t think he is original and attributes the distinction between acceptabil-
ity and truth to Quine.1 He follows Quine in bringing up two examples:
Example 1 Once we know France is in Europe, we don’t utter “France is in Europe or
the sea is sweet.”
Example 2 Once we know that every S is P (and that there are S), we don’t say “at
least some S are P .”
Quine explains the examples by pointing out that in such contexts the longer sentences
are less informative and that shorter and more informative sentences are usually more
likely to be uttered for pragmatic reasons.
Yet, Ajdukiewicz is quite unhappy about the use of the notion of informativeness in this
account—because he doesn’t find this notion clear enough. He claims the only sensible
explication of this notion as used by Quine is that p is more informative than q iff q
follows logically from p but p doesn’t follow from q.
While Ajdukiewicz finds this definition clear enough, on this account, he claims, it is
false that if one wants to be as helpful and honest in a conversation as possible, one should
always prefer the shorter and more informative claim. For instance, he insists, when we
present a reasoning, we often utter statements weaker than the premises in the process, and
this would be prohibited by the general principle suggested by Quine. Thus, we might say
that Ajdukiewicz discards Quine’s general principle by arguing that following it would
prohibit us from uttering our arguments.
In the context of our discussion of Ajdukiewicz’a ideas it is worthwhile to clarify
Quine’s views on the relation between material implication and natural language condi-
tional.
In Methods of Logic, [6] discusses briefly whether material implication conforms to
the ordinary indicative conditional ‘if-then’. He first states that restricting our attention to
simple indicative conditionals is justified:
the material conditional ‘p → q’ is put forward not as an analysis of general conditionals such
as (1) [‘If anything is a vertebrate, it has a heart’], nor as an analysis of contrafactual conditionals
such as (4) [‘If Eisenhower had run, Truman would have lost.’], but, at most, as an analysis of the
ordinary singular conditional in the indicative mood. . . [6, p. 15]
arguing that (1) must be viewed as affirming a bundle of individual conditionals which in
this context means that it shall be analyzed in terms of first-order logic with quantifiers
instead of propositional calculus only, and that. . .
. . . any adequate analysis of the contrafactual conditional must go beyond mere truth values and
consider causal connections, or kindred relationships, between matters spoken of in the antecedent
of the conditional and matters spoken of in the consequent. . . [6, p. 14]
Quine observes that in natural language some counterfactuals with false antecedents
and false consequents may be true while other counterfactuals with false antecedents
and false consequents can be false. From this, he concludes that the semantics of
1 He does so without giving any references, but Quine’s discussion of these issues can be found in [7] and
[6].
counterfactuals is definitely not truth functional, whatever the adequate analysis of their
meaning is.
After settling that the material implication may be an analysis only of indicative
conditionals, he claims that analyzing conditionals as material implications may be
unnatural when there is no relevance between the antecedent and the consequent:
Even as an analysis of such conditionals, the version ‘p → q’ is sometimes felt to be unnatural,
for it directs us to construe a conditional as true no matter how irrelevant its antecedent may be to
its consequent, so long as it is not the case that the antecedent is true and the consequent false. [6,
p. 15]
Quine argues that if we have a conditional in which the compounds are irrelevant to
each other, it is equally strange to consider it true and to consider it false, irrespective of
the truth values of the antecedent the and consequent. He claims that the strangeness of
conditionals such as ‘If France is in Australia then the sea is sweet’ stems from the fact
that linguistic practice usually allows for forming conditionals (of the form ‘if p, then q’)
out of compounds (p and q) which have unknown truth values:
. . . for it is not usual in practice to form conditionals out of component statements whose truth or
falsity is already known unconditionally. . . In practice, one who affirms ‘If p then q’ is ordinarily
uncertain as to the truth or falsehood individually of ‘p’ and of ‘q’ but has some reason merely for
disbelieving the combination ‘p and not q’ as a whole. [6, p. 15]
Thus, the question of the nature of the reason to disbelieve ‘p and not q’ rises. Quine
replies, as mentioned above, that the reason is to be provided by some kind of relevance
between the compounds of the conditional. However, he claims that this relevance does
not have an influence on the meaning of the conditional, but rather that useful applications
of a conditional in linguistic practice are dependent on it:
Only those conditionals are worth affirming which follow from some manner of relevance
between antecedent and consequent—some law, perhaps, connecting the matters which these
two component statements describe. But such connection underlies the useful application of the
conditional without needing to participate in its meaning. Such connection underlies the useful
application of the conditional even though the meaning of the conditional be understood precisely
as ‘¬(p ∧ q)’. [6, p. 16]
In Mathematical Logic, [7] presents essentially the same view, adding a remark on
the particular role of the truth table for material implication in assessing conditionals in
ordinary practice:
What the truth table adds, in thus deciding the cases beyond the range of ordinary usage,
is essentially theoretical; no supplementary practical use of ‘if-then’ is thereby prescribed. In
practice, even in the light of the truth table, one would naturally not bother to affirm a conditional
if he were in position to affirm the consequent outright or to deny the antecedent—any more than
one bothers to affirm an alternation when he knows which component is true. [7, p. 17]
Hence, according to Quine, there is a difference between the meaning (semantics) of

an expression and its useful applications (pragmatics)—the meaning of a conditional is
exhausted by the truth tables of material implication, whereas practical useful application
of a conditional demands there to be some kind of (e.g. causal) connection or relevance
between the matters expressed in the compounds of a given conditional. And, most
importantly, the rule is not to assert a conditional if we are already certain of the truth
values of its compounds. Consequently, the general rule of reasoning that Quine suggests
and the one he takes as an explanation of the strangeness of conditional with mutually
irrelevant compounds is that in practice communicating agents choose to assert statements

that are as short and as strong as possible. So, when we know the truth values of ‘p’ and
of ‘q, it is useless to use the conditional ‘if p then q” in our reasoning:
Why affirm a long statement like ‘If France is in Europe then the sea is salt.’ or ‘If France is in
Australia then the sea is salt.’ when we are in position to affirm the shorter and stronger statement
‘The sea is salt’? And why affirm a long statement like ‘If France is in Australia then the sea is
salt.’ or ‘If France is in Australia then the sea is sweet.’ when we are in position to affirm the
shorter and stronger statement ‘France is not in Australia’? [6, p. 15]
Instead of following Quine, Ajdukiewicz, having found Quine’s general principle

too strong, and lacking a more successful general principle, settles tentatively with
acknowledging the following phenomenon:
DISJUNCTION One normally doesn’t utter a disjunction if one knows which of the
disjuncts is true.
and goes on to find a more plausible explanation for DISJUNCTION.
5 Expressing vs. Stating
At this point, to provide a more general account of why such a principle should hold,
Ajdukiewicz makes a distinction between what a sentence states (what has to be the case
for it to be true) and what it expresses:
To say that an utterance W of a person O expresses, given the linguistic habits, his state S, is the
same as saying that W uttered by O is for the audience (who know of those [linguistic] habits) a
sign of O’s being in state S, or that uttering W by O allows the audience familiar with those habits
to figure out that O is in state S. [1, p. 255]
On this approach, by uttering something, speakers not only state the fact required for
the truth of the statement, but also express their states, associated with particular types
of utterances by linguistic habits of a given linguistic community. While the statement
might be true or false in virtue of whether what it states obtains, it is proper or improper
in virtue of whether the utterer is in the state expressed by the statement.
Another important aspect of the distinction is that while to come to believe what is
stated, one has to believe the statement to be true, while to come to believe what is
expressed, one doesn’t have to accept the statement itself. It is enough to understand it
and to know the relevant linguistic habits.
Now we have reached Ajdukiewicz’s general principle:
AJDUKIEWICZ One is unwilling to accept improper claims, even if they are true.
Ajdukiewicz doesn’t say anything about this issue, but notice that Ajdukiewicz’s
criticism of Quine’s solution doesn’t apply to his own approach. After all, there are many
cases of deductions which can be presented properly, because the inference steps do not
express anything that contradicts what the premises express.
6 Ajdukiewicz on the Diagnostics of Improper Use
One way of diagnosing our reasons to not assert a particular utterance p suggested by
Ajdukiewicz is to ask ourselves: are we willing to accept ¬p?2 If we refuse to assert p
because we think it’s false, our answer should be positive:
Whether the refusal to accept a certain sentence is motivated by the unwillingness to accept
something false, or the unwillingness to use an expression improperly, can be recognized, among
other facts, by the fact, that in the first case the refusal is accompanied by the readiness to accept
the negation [of the sentence], while in the second case such a readiness is missing. [1, p. 256]
Now, this seems a bit too hasty. After all, if I have no information about some
p (say “aliens exist”), I will refuse to accept it, and I will refuse to accept ¬p. By
Ajdukiewicz’s criterion, this would mean that my refusal to accept p is not motivated
by the unwillingness to accept something false. But this doesn’t sound right—the main
reason why I don’t accept sentences about whose truth values I have no information
is because as far as I know they might be false, and I wouldn’t want to accept a false
sentence.
In all fairness, however, when I refuse to assert “aliens exist”, it’s not only because I
don’t want to accept a (potentially) false statement, but also because this would suggest
to the audience that I do think that I know that aliens exist, and that would be false. But
this holds for any sentence which we don’t know to be true, including those sentences that
we know to be false. So, for instance, I’m unwilling to assert “2+2=5”, and one reason
for this is that this would express the claim that I know that 2+2=5, which is false. But at
the same time, I am willing to assert “2 + 2 = 5”, and so, by Ajdukiewicz’s criterion, I
am refusing to accept “2+2=5” for truth-related reasons, and not just because it would be
simply improper to assert it.
So, we submit, Ajdukiewicz’s account of this diagnostics should be revised. First of
all, refusing to assert a sentence because it would be improper (because it would express
something false) doesn’t exclude refusing to assert it because it’s false. If we believe that
a sentence is false, we should also believe that asserting it would be improper, because the
assertion would express our knowledge that it is true (and we don’t have that knowledge).
Second, as our example with aliens shows, refusing to assert a sentence because it
would be improper (because it would express something false) doesn’t exclude refusing
to assert it because we wouldn’t want to accept something false, even if we don’t know
the sentence to be false.
Third, the distinction should be rather between sentences that we believe to be false and
refuse to accept, and sentences that we don’t believe to be false and refuse to accept. To
this distinction Ajdukiewicz’s diagnostic criterion correctly applies: if I believe something
to be false, I will assert its negation, and if I don’t believe it to be false, I won’t.
The problem is, however, that now the criterion fails to divide sentences where
Ajdukiewicz would like it to. After all, if I refuse to accept a disjunction while refusing
to accept its negation, all I know is that I don’t believe it to be false. Whether it is further
the case that I only refuse to accept it because I believe it is true, but it would be improper
2 Forthe purposes of this paper we follow Ajdukiewicz in taking acts of acceptance to be public, and so
we ignore the distinction between accepting and asserting.
to assert it, or whether I also think I have no good reasons to think it’s true is a separate
issue that needs to be separately discussed.
There is a charitable and instructive way of reading Ajdukiewicz’s remarks about the
diagnostic method, though. Consider the example with pieces of chalk. The audience
agreed to “p or q”. Upon discovering that p is true and q is false, they cease to assert
this disjunction. Now, if they thought the fact that p is true and q is false made the
disjunction false, they would not only refrain from asserting it, but also were ready to
assert its negation. Hence, it seems, Ajdukiewicz is still right that this shift doesn’t have
much to do with truth conditions of disjunctions.
7 Generalization, Weakening, Moore’s Paradox
One remark is in place here. Ajdukiewicz suggests that by uttering a disjunction we

express the fact that we know it to be true (and, that we don’t know which disjunct is
true). And in general, one might think that when uttering any sentence meant to state a
fact, in standard contexts we express that we know that the sentence is true.
Both the suggestion about disjunction, and its generalization seem a bit hasty, though.
For Ajdukiewicz would like any assertion which expresses a condition which is not
satisfied to be improper. But there seems to be nothing improper to assert a true disjunction
I think I know to be true which I don’t really know to be true (and of which I think I don’t
know which disjunct is true), Gettier-like cases abound. And so, we submit, it would be
sensible, and it would not damage Ajdukiewicz’s approach to conditionals, to weaken
the claim to saying that when one, in a standard context, asserts a disjunction (or any
sentence meant to state a fact), one expresses the fact that one believes (not: knows) that
disjunction (sentence) to be true. In what follows, we’ll keep this remark in mind, but
we’ll still formulate arguments mostly in terms of knowledge, assuming that unless stated
otherwise, considerations still apply after replacing knowledge with belief.
On the other hand, this easy generalization (and, at the same time, weakening) of
Ajdukiewicz’s suggestions bears directly on what, we think, Ajdukiewicz would say
about Moore’s paradox, where one says things like “p, but I don’t believe that p”.
Namely, Ajdukiewicz would point out that by asserting a conjunction one expresses
one’s belief in both conjuncts, and so one expresses one’s belief in p, and one’s
belief in not believing that p. Observe further that assuming introspection, this leads
to straightforward contradiction in expressed judgments. The idea is that the expressed
content is Bp ∧ B¬Bp. The first conjunct, by introspection yields BBp, and so the whole
thing together entails that the belief set is inconsistent.
8 Apparent Connection Between Disjuncts
To use AJDUKIEWCZ to explain DISJUNCTION, Ajdukiewicz observes that a disjunction,

apart from stating that at least one of the disjuncts is true, expresses the fact that the
speaker knows that at least one of the disjuncts is true, and that the speaker is ignorant as
to which of them it is. For this reason, he suggests, once we know which disjunct is true,
we refuse to utter a disjunction.
Ajdukiewcz asks: how can we know that a disjunction of the form “p or q” is true?
Well, one option is that we know p, or we know q. In such a case, asserting “p or q”
would be improper. Another option is that we’re ready to assert q if we find out that p is
false: that is, we’re ready to infer q from ¬p.
So, on this view, in the context of proper assertion, expresses speaker’s readiness to
assert q if they find out that p is false: to infer q from ¬p. Where does this readiness
comes from, though? Isn’t it supported by some connection between the disjuncts, which
is not only expressed, but also stated by a disjunction? Ajdukiewicz disagrees.
What sort of connection would that be? One option would be that it would be
psychological: on this approach, the connection is that a disjunction makes one willing or
ready to infer one disjunct from the negation of the other one. Ajdukiewicz observes that
this notion would relativize truth-conditions of a disjunction to the speaker, and since he
finds the idea of a disjunction being true for one person but false for another unpalatable,
he rejects this account.
Another idea would be that the connection consists in making the inference under
discussion legitimate. But in what sense? If all that is meant here is that if one of the
disjuncts is false, the other one is true, then Ajdukiewicz agrees—but this is exactly
what a disjunction states, and what is captured by the standard truth table for classical
disjunction.
On a stronger interpretation, the claim is that the negation of one disjunct logically
entails the other disjunct. But this, Ajdukiewicz observes, doesn’t seem to hold for natural
language disjunctions. He uses the following example: I will die on a day with an even
date, or on a day with an odd date. This, he holds, is clearly true.3 Yet, without additional
premises that he will die some day and that each day has either an odd date or an even
date, the negation of one of the disjuncts doesn’t logically entail the other disjunct.
Could the second proposal be fixed to avoid Ajdukiewcz’s criticism by saying that
the negation of one of the disjuncts should entail the other disjunct with some additional
premises? Not easily, for reasons similar to those for which the cotenability approach to
conditionals is unsuccessful. For one would have to specify which additional premises can
be used. If any true premises can be used, no interesting connection between disjuncts is
required, it is just enough that one of them really is false. Once p is false, ¬p is true, and
p or q with ¬p (a true sentence after all) logically entails q. One might be tempted to
avoid this by saying that only those extra premises can be used to infer q from p or q and
¬p which are connected with q, but this would make the account circular.
Let’s observe, however, that Ajdukiewicz’s criticism isn’t lethal. He lists three
interpretations of what it would mean for there to be a connection between the relevant
sentences, excludes two of those options, and shows that the third one is exactly the one
he proposes. What is missing is an argument to the effect that this is a complete list of
sensible interpretations. But, absent other interpretations that would avoid criticism, this
should be good enough for now.
3 In fact, he died on April 12, 1963, and so the first disjunct is true.
9 Ajdukiewicz’s Diagnostics and Grice’s Cancellability
Now that we have presented Ajdukiewicz’s defense of the material implication account
of indicative conditionals, let’s compare it to a much better known defense of the same
claim, due to Grice. While analyzing the problem of material implication in his Logic and
Conversation [3, pp. 3–143], P. Grice considers the so-called Indirectness Condition (IC)
associated with the conditional. According to IC, there should be non-truth-functional
grounds for accepting p → q as the meaning of ‘if p then q’. That is, if a subject asserts
‘if p then q’, they are conventionally committed both to the proposition p → q and to
IC, which amounts to claiming some causal (or other, but still non-truth-functional) link
between p and q.
Grice’s analysis of conditionals lies in the scope of his theory of maxims of conver-
sation and of conversational implicature. The main idea is that rational communicative
interaction is governed by certain principles and maxims, which despite their prescriptive
phrasing actually describe how agents behave in order to achieve effective communication
in conversation. The most general rule is the cooperative principle (CP) which says:
Make your contribution as is required, when it is required, by the conversation in which you are
engaged.
According to Grice, the descriptive content of CP consists in the fact that speakers
(generally) observe the cooperative principle, and listeners (generally) assume that
speakers are observing it. Fulfilling CP consists then in obeying the so-called maxims
of conversation. This means that the requirements of CP are explicated by the following
rules:
Maxim of Quality: Contribute only what you know to be true. Do not say false things.
Do not say things for which you lack evidence.
Maxim of Quantity: Make your contribution as informative as is required. Do not say
more than is required.
Maxim of Relevance: Make your contribution relevant.
Maxim of Manner: Avoid obscurity, avoid ambiguity, be brief and be orderly.
Grice claims that certain utterances during a conversation convey meanings that are not
explicitly expressed nor logically implied (entailed) in what is said, but nevertheless can,
in some sense, be inferred for pragmatic reasons. Such meanings (or pragmatic inferences)
are called conversational implicatures:
I am now in a position to characterize the notion of conversational implicature. A man who,
by (in, when) saying (or making as if to say) that p has implicated that q, may be said to have
conversationally implicated that q, provided that (1) he is to be presumed to be observing the
conversational maxims, or at least the cooperative principle; (2) the supposition that he is aware
that, or thinks that, q is required in order to make his saying or making as if to say p (or doing so
in those terms) consistent with this presumption; and (3) the speaker thinks (and would expect the
hearer to think that the speaker thinks) that it is within the competence of the hearer to work out,
or grasp intuitively, that the supposition mentioned in (2) is required. [3, pp. 30–31].
The idea is that a conversational implicature is a pragmatic inference that the listener
has to make if she is about to maintain that the speaker is cooperative. More precisely,
following [5, p. 113] we might say that the speaker S’s saying that p conversationally
implicates q if:
1. S is presumed to be observing the maxims, or at least (in the case of floutings) the
cooperative principle (cooperative presumption).
2. In order to maintain this assumption it must be supposed that S thinks that q
(determinacy).
3. S thinks that both S and the addressee H mutually know that H can work out that to
preserve the assumption in (1), q is in fact required (mutual knowledge).
One of the most important features of conversational implicatures4 is cancellability
and Grice even claims that conversational implicatures differ from semantic implicatures
exactly in being cancellable. This means that they can be consistently dismissed by the
speaker or in light of the context. For instance, the speaker might consistently add to the
conversation some content that entails the negation of an already introduced implicature—
assume that the speaker said ‘Some of the students passed the test.’ This implicates that
not all of the students passed the test. The implicature might be canceled then by uttering:
‘Indeed, all of the students passed the test.’ Another example is when the speaker adds to
the conversation some content that indicates that she is not committed to the implicature
or its negation, as in ‘Some, maybe all, of the students passed the test.’
Grice also makes one distinction that will be relevant to our concerns. He claims that
there is a difference between particularized and generalized conversational implicatures:
I have so far considered only cases of what I might call particularized conversational implicature—
that is to say, cases in which an implicature is carried by saying that p on a particular occasion
in virtue of special features of the context, cases in which there is no room for the idea that an
implicature of this sort is NORMALLY carried by saying that p. But there are cases of generalized
conversational implicature. Sometimes one can say that the use of a certain form of words in
an utterance would normally (in the ABSENCE of special circumstances) carry such-and-such
an implicature or type of implicature. Noncontroversial examples are perhaps hard to find, since
it is all too easy to treat a generalized conversational implicature as if it were a conventional
implicature. [3, p. 37]
A generalized conversational implicature is one which does not depend on particular

features of the context, but is instead typically associated with the proposition expressed.
Not surprisingly, a particularized conversational implicature is one which depends on
particular features of the context.
For instance, imagine a following conversation:
A: Will Sally be at the meeting?
B: Her car broke down.
In this situation, it is implicated by B that Sally will not be at the meeting, but the
implicature is particularized, as there is nothing in the content of the expressed proposition
that would suggest such an inference. Consider, however, another example:
Mary has 3 children.
4 Among others, such as: calculability, non-conventionality, non-detachability, indeterminacy or re-

inforcerability. An interested Reader can consult e.g. Levinson [5].
When such a sentence is uttered in a conversation, it is implicated that Mary has no

more than 3 children and the implicature is generalized—we can associate it with the
proposition expressed, irrespective of the context.
Grice defends the view that the meaning of a conditional on any particular occasion
of utterance is simply equivalent to the meaning of material implication. He claims that
IC is a generalized conversational implicature. Just as all generalized conversational
implicatures, it can be cancelled without contradiction, either by circumstances in the
context or by explicit denials. Grice considers two examples of assertions of conditionals
which carry no implicature of IC. These are the following:
If I have a red king, I also have a black king
uttered during a card game, and

If Mr Jones has black pieces, Mrs Jones has too
uttered during the run of a particular logical puzzle where the participants are about
to guess the identity of the characters in the game and are provided with a piece of
information, such as the one given in the conditional above. Grice claims that the total
contents of the utterances above is just p → q.
One of the maxims of Girce’s pragmatics is the one of Quantity—it dictates to make,
in a given context, the most informative statements of interest as possible. Grice calls the
pragmatic virtue of this maxim ‘conversational helpfulness’. Its point is to make natural
language communication as efficient as possible. This resembles Quine’s suggestion that
we should utter stronger rather than weaker claims However, as Grice puts it:
An infringement of the first maxim of Quantity, given the assumption that the principle of
conversational helpfulness is being observed, is most naturally explained by the supposition of
a clash with the second maxim of Quality (“Have adequate evidence for what you say”), so it is
natural to assume that the speaker regards himself as having evidence only for the less informative
statement . . . [3, p. 33]
The above amounts to the claim that any utterance of ‘if p then q’ will, unless
prevented by context, give rise to the implicature that the speaker does not have definite
information about the truth values of p and q. According to Grice, some utterances of ‘if
p then q’ might implicate a stronger condition than the one provided by the truth-tables
of p → q. However, instead of being a part of the meaning, what is implicated then
is founded on the Cooperative Principle and the maxims of conversational implicature.
Conditionals, then, play, apart from their semantics (given by the truth-table for →),
also roles of pragmatic nature: they enable people to ponder the consequences of certain
choices during a conversation. In the light of Gricean theory one might say that it would
simply be irrational to use a conditional in certain contexts, for instance when there is no
doubt about the truth of the antecedent.
Interestingly, the very same analysis also applies, according to Grice, to disjunctions:
a natural-language disjunction of the form ‘p or q’ shares the logical meaning of ‘p ∨
q’, but in addition carries a generalized implicature that they are not both true. If the
speaker was in a position to offer the more informative form ‘p and q’, then it would be
conversationally more helpful to do so.
Prima facie, both Grice and Ajdukiewicz achieve seemingly equivalent results—the
former’s conversational implicature and the latter’s expression of a mental state are quite
similar to each other. It needs however to be noted that Grice, as Quine, has a general
principle from which the Maxim of Quantity follows, whereas Ajdukiewicz gets to his
result in a different manner. He first acknowledged a simple linguistic phenomenon
(DISJUNCTION—see above) and found a relatively uniform yet simple explanation for
it. Without assuming generalities of logical (as Quine) or of pragmatic (as Grice) sort,
he simply associated effective utterances (i.e. the ones that, during a conversation,
successfully deliver the content intended by the speaker) of a given expression with the
possession of a certain epistemic state. In contrast with Grice’s theory, the accuracy of
his explanation does not depend on strongly controversial properties of a systematic but
convoluted general theory of cooperative behaviour. Secondly, the nature of the criterion
needs to be taken into account—for Grice it seems to be purely pragmatic, conversational
and practical; for Ajdukiewicz—it is, in a sense, doxastic—it pertains to expressing the
agent’s knowledge or belief.
However, it actually might be questioned if Ajdukiewicz’s account allows for express-
ing the content of the mental states only. As he puts it himself, conditionals express certain
type of a lack of knowledge of the agent who makes the assertion (that she does not know
that the antecedent is true and she does not know that the consequent is false). Moreover,
conditionals express that the agent is ready to infer the consequent from the antecedent.
As for the former condition, it is rather clearly doxastic or epistemic. But as for the latter,
one might ask: what does it exactly mean to express a disposition to make a certain type
of inference? It is not completely clear whether one should understand it as expressing a
mental state or rather just a disposition to be in a certain type of mental state a disposition
to perform a particular type of behaviour.
10 Goła̧b on Indicative Conditionals
Material implication as an interpretation of natural language indicative conditionals was

also discussed in Lvov-Warsaw school by Gołab ˛ [2] and Słupecki [8]. The former
proposed a difficulty to the interpretation and the latter replied. Goła̧b complains that
when he teaches introductory logic, the claim that a conditional whose antecedent is
false and consequent true is true usually startles the audience, because they have strong
intuitions there are false conditionals which satisfy these conditions. Goła̧b’s example is:
If today is Monday, tomorrow is Wednesday.
The sentence, uttered on Tuesday, should be true, if the material reading is adequate, but
people usually don’t have this intuition.
Another problem brought up by Goła̧b is with contraposition—he credits prof.
Harassek from Lublin with the following counterexample:
If I’m hungry, I won’t swallow a needle.

If I swallow a needle, I won’t be hungry.
11 Słupecki’s Reply to Goła̧b
As for contraposition, Słupecki suggests that the conditionals formulated in the coun-
terexample don’t express what is really being claimed. The actual and correct premise of
the reasoning should be Even if I’m hungry, I won’t swallow a needle. But if that’s the
case, the problem doesn’t arise, Słupecki claims, because contraposition doesn’t work for
even . . . if conditionals. He claims that for such conditionals there is no rule analogous to
the principle of contraposition.
Słupecki’s response to Goła̧b’s first worry is somewhat dismissive. He insists that the
sentences in question are true, but it’s simply that people wouldn’t normally utter them
due to some unspoken principles of parsimony, which prevent them from stating useless
sentences not worth of interest:
I suppose that the source of these intuitions is a kind of unformulated and not fully recognized
principle of economy which we also follow when we classify the following sentence as nonsense:
2 + 2 = 4 and 2 + 2 = 4 and 2 + 2 = 4
—although in logical terms (logically) it is completely correct (sound)—or when we are not
willing to accept (assert) compound sentences, the components of which are not materially
(factually, objectively) linked or when such a link is bizarre (fanciful, odd, singular). Perhaps
we would not be always inclined to count such sentences as false, but rather as superfluous,
inexpedient or not worth of serious attention.
Słupecki also suggests a certain teaching trick that is meant to facilitate the digestion of
the truth conditions of material implication. The trick consists in asking the students to
describe a state of affairs in which a given conditional would be false. For instance
If it rains, John is not having a walk
The trick is to ask when this sentence would become false. The expected (obvious)
answer should be, and usually is: if and only if it rains and John is having a walk. Then,
it should be an easy way from this point for the students to see that if the above is not the
case, the entire conditional must be true.
The problem with the trick is that it doesn’t seem to overcome the difficulties that
give rise to the questions concerning the semantics of conditionals. It is not too surprising
that such heuristics works for sentences in which the antecedent and the consequent are
causally related. It is, however, doubtful whether this teaching method would work for
problematic examples such as if today is Monday, tomorrow is Wednesday or if it is raining
and not raining, then the Moon is made of green cheese.
12 Ajdukiewicz’s Solution and Goła̧b’s Problem
Obviously, Ajdukiewicz’s approach was not available at the time of Goła̧b’s and Słu-
pecki’s discussion, but we can ask whether the solution he proposed successfully handles
the case Goła̧b brought up (Ajdukiewicz doesn’t seem to be aware of Goła̧b’s and
Słupecki’s papers on the topic).
A natural thing to say about the conditional if today is Monday, tomorrow is Wednesday
from Ajdukiewicz’s perspective is that when we utter it on Tuesday, we know that the
consequent is false, and yet the conditional expresses the claim that we don’t have such
knowledge.
But what if this reason for the utterance being improper is removed? Suppose it is
Tuesday, but neither the speaker nor the audience know what day of the week it is (nor do
they have any beliefs about it). The implication in question expresses that the speaker
doesn’t know which of the disjuncts is true, which is the case. Isn’t this a problem for
Ajdukiewicz’s account? Not really—for there is another reason why an utterance of this
conditional in such situation would be improper, stemming from the discussion in Sect. 8.
How would the speaker come to know the conditional? Well, one way is by knowing
either in the consequent, or in the negation of the antecedent (this is, by the way, quite a
useless way of knowing a conditional, because a conditional known this way cannot be
properly uttered). But the speaker doesn’t know what day of the week it is, so this way of
knowing it is not an option. Another would be, perhaps, to know of a reason why today’s
being Monday would allow the speaker to infer that tomorrow is Wednesday. But there
can’t be such a reason, and so the speaker can’t also come to know the conditional in this
manner. Thus, the speaker is never in position to know the conditional, and consequently,
never in position to utter it properly, for the expressed claim that the speaker knows the
conditional to be true would always be false.
Acknowledgment This work was completed with the support of the Research Foundation Flanders and
Polish National Science Centre (grant DEC-2016/22/E/HS1/00304)
References
1. Ajdukiewicz, K.: Okres warunkowy a implikacja materialna [Conditional sentence and material
implication]. Studia Log. IV, 117–134 (1956)
2. Gołab,
˛ S.: O kilku trudnościach zwiazanych
˛ z prostymi kwestiami logicznymi [On a few difficulties
connected to simple logical issues]. Matematyka 3(5), 27–29 (1949)
3. Grice, P.: Studies in the Way of Words. Harvard University Press, Cambridge (1991)
4. Jackson, F.: Conditionals. Blackwell, Oxford (1987)
5. Levinson, S.C.: Pragmatics. Cambridge University Press, Cambridge (1983)
6. Quine, W.V.: Methods of Logic, 2nd edn. Holt, Rinehart and Winston, Austin (1966)
7. Quine, W.V.: Mathematical Logic, Revised edn. Harvard University Press, Cambridge (1981)
8. Słupecki, J.: O własnościach okresu warunkowego [On the properties of a conditional]. Matematyka
4(6), 32–35 (1949)
R. Urbaniak ()
Institute of Philosophy, Sociology and Journalism, University of Gdańsk, Gdańsk, Poland
M. T. Godziszewski
Department of Logic, Institute of Philosophy, University of Warsaw, Warsaw, Poland
Tadeusz Czeżowski
Abstract The aim of this paper is to present life, main works and ideas of Tadeusz
Czeżowski, who was a Polish logician, philosopher and an eminent representative of
the Lvov-Warsaw School. Czeżowski’s contribution to the contemporary philosophy
and logic is of universal value and includes such achievements as the classification of
reasoning types, the analysis of singular propositions, the analysis of Aristotle’s modal
logic from the point of view of modern logic, the reconstruction of Brentano’s syllogistic
and the theory of transcendental concepts.
Keywords Logic · Philosophy · Classification of reasoning types · Kazimierz

Twardowski
Mathematics Subject Classification (2000) Primary 01A60, 01A5; Secondary 03B42
1 Life
Tadeusz Czeżowski was born in Vienna on July 26, 1889 in a middle class family. His
father was Prefect and later Counsellor of the Governorate of Galicia and was transferred
to Lvov in 1899; his mother, Helena Kusché, belonged to the petite bourgeoisie of Lvov.
He died in Toruń on February 28, 1981.
In 1907 he enrolled at the Faculty of Philosophy of the University of Lvov to study
philosophy, mathematics and physics. In 1912 he qualified as a teacher of mathematics
and physics and taught in a Lvov grammar school from 1912 to 1914. Along with teaching
duties, Czeżowski continued his scientific activity. Under Twardowski’s supervision he
wrote a dissertation on The theory of Classes (Teoria klas), for which he obtained his
doctoral degree in 1914. At the outbreak of the First World War in 1914, he arrived in
Vienna and was appointed by the Rector of the University of Lvov, Kazimierz Twardowski
as responsible for the accommodation of war refugees who wished to continue their
studies there. After resuming his university studies he returned to Lvov, where he held
the position of director of University Chancellor’s Office for 3 years (1915–1918). In
1918, Czeżowski was given a post at the newly established Polish Republic Ministry
for Religious Confessions and Public Education, where he worked until 1923, first as

134 D. Łukasiewicz
a clerk and then as Director of the Department of Science and Secondary Education.
In 1919, he was sent to Vilnius to help in reorganizing the administration of Stefan
Batory University. In 1920, Czeżowski obtained his habilitation degree for the work
Variables and Functions (Zmiennie i funkcje), thanks to which he became Assistant
Professor. During the Polish-Soviet war in 1920 he was a soldier and was awarded for
courage.
In 1923 Czeżowski was offered the Chair of Philosophy at the University of Vilnius
when it became vacant. From 1933 to 1935 he was Vice Rector there and, from 1935
to 1937, Dean of the Faculty of Humanities. He was awarded the title of Full Professor
in 1936. In 1928 he founded Vilnius Philosophical Society. During all those years in
Vilnius, he was a correspondent of Przeglad ˛ Filozoficzny (Philosophical Review), one
of the leading Polish philosophical journals, and he collaborated with Ruch Filozoficzny
(Philosophical Movement), another important Polish philosophical journal, founded and
edited by Kazimierz Twardowski.
When the University was closed during the Second World War, Czeżowski continued to
teach clandestinely, holding as many as 143 lectures, mostly on ethical subjects. During
the war Czeżowski was imprisoned twice. He saved the lives of many Jewish people,
for which, in 1963, he was awarded the title of Righteous among the Nations by Yad
Vashem Institute, together with his wife Antonina and his daughter Teresa, and, in 2012,
he posthumously obtained the honorary citizenship of Israel.
After the war, Czeżowski had to leave the University of Vilnius and, like other
professors of that university, he was moved to a newly established University of Toruń
(Nicolaus Copernicus University), becoming one of its main organizers. He held the Chair
of Philosophy there till 1951 and then the Chair of Logic. In 1946 he founded Toruń
Philosophical Society and, for many years, he was elected the chairman thereof. In 1948
he became the editor-in chief of Ruch Filozoficzny. He was also the initiator and organizer
of a series of conferences on the history of logic, still regularly taking place in Kraków.
Czeżowski retired in 1960. After retirement he continued to teach students and young
scholars of Nicolaus Copernicus University interested in philosophy, holding as many as
173 seminars.
In total, he was the author of some 190 scientific publications, including his
philosophical books (Odczyty Filozoficzne, Filozofia na rozdrożu, Główne zasady
nauk filozoficznych), textbooks on logic, numerous articles and reviews. He also
gave some interviews to magazines popularizing science. It is worth noting that
during the years of communist dictatorship, which started in Poland immediately
after the war, Czeżowski till his death refused any awards offered by the communist
state.
Regarding his private life, in 1929 he married Antonina Packiewicz, who had attended
his lectures as an external student. The couple had two children: the son who died
soon after birth and the daughter Teresa; she died at the age of twenty. After the war,
they adopted another girl, Eleonora. In private life, Czeżowski had a true passion for
mountains; he was an experienced mountaineer and skier. He was also a lover of music
and dance.
Tadeusz Czeżowski 135
2 Main Papers
• O metafizyce, jej kierunkach i zagadnieniach [On metaphysics, its trends and prob-
lems] 1948
• Główne zasady nauk filozoficznych [Main principles of philosophical sciences] 1959
• Filozofia na rozdrożu [Philosophy at a crossroads] 1965
• Odczyty Filozoficzne [Philosophical lectures] 1969
• Transcendtentalia – przyczynek do ontologii [Transcendentals and ontology] 1977
3 Views
In philosophy and methodology of science, he accepted holism, empiricism, epistemic

individualism and fallibilism (in principle). In philosophy of language, Czeżowski argued
for the conception of semantic transparency of the sign. His major achievements in
logic include: establishing the number of relations between the scope of the terms of
the categorical propositions, an extended version of Aristotle’s syllogistic, the analysis
of singular propositions, an original reconstruction of Aristotle’s theory of modal
propositions. Czeżowski’s idea was that Aristotle’s calculus of modal propositions is
dual with calculus of categorical propositions. In metaphysics, Czeżowski elaborated an
original theory of transcendental concepts such as goodness, beauty and truth. He assumed
that they should be interpreted as modi essendi rather than as properties of objects. In
meta-ethics, he defended anti-naturalism, cognitivism and infallibility of basic moral
principles. In the domain of applied ethics, Czeżowski advocated the egalitarian ethics
based on the principle of equal measure, i.e. that human should not value their own good
higher than the good of other people.
4 Resonance
The most influential ideas of Czeżowski are his theory of syllogism, theory of ethics as
empirical and inductive science and his doctrine of transcendentals.
5 Pupils
To the Vilnius group of his pupils belonged: Benedykt Woyczyński, philosphers of law
Józef Zajkowski and Jan Rutski, Saul Sarnaker—a physician interested in philosophy,
Aleksandra Zajkowska-Znamierowska, Józef Reutt, Edward Csató, Stefan Buhardt,
Abraham Fessel, Sawa Frydmann, Maria Renata Mayenowa, Barbara Skarga, Stefan
Wołoszyn, Lidia Wołoszyn. To the Toruń group of Czeżowski’s disciples belonged
Valdemar Voisé, Leon Gumański, Wacław Kubik, Henryk Moese, Janusz Skarbek,
Bogusław Wolniewicz and Zbigniew Zwinogrodzki.
136 D. Łukasiewicz
6 Summary
Czeżowski’s contribution to the contemporary philosophy and logic is of universal value

and includes such achievements as the classification of reasoning types, the analysis of
singular propositions, the analysis of Aristotle’s modal logic from the point of view of
modern logic, the reconstruction of Brentano’s syllogistic and the theory of transcendental
concepts.
Bibliography
1. Czeżowski, T.: O metafizyce, jej kierunkach i zagadnieniach, 1st edn. Wydawnictwo Uniwersytetu
Mikołaja Kopernika, T. Szcz˛esny, Toruń (1948)
2. Czeżowski, T.: Główne zasady nauk filozoficznych, 3rd edn. Zakład Narodowy im. Ossolińskich,
Wrocław (1959)
3. Czeżowski, T.: Filozofia na rozdrożu. PWN, Warszawa (1965)
4. Czeżowski, T.: Odczyty Filozoficzne, 2nd edn. Towarzystwo Naukowe w Toruniu, Toruń (1969)
5. Czeżowski, T.: Transcendtentalia – przyczynek do ontologii. Ruch Filozoficzny 1–2, 54–56 (1977)
D. Łukasiewicz ()
Kazimierz Wielki University, Bydgoszcz, Poland
Tadeusz Czeżowski’s Theory of Knowledge
and Beliefs
Abstract The aim of the paper is to present and critically evaluate Tadeusz Czeżowski’s
views concerning knowledge and belief in the context of some non-Brentanian epistemo-
logical concepts and theories. Czeżowski belonged to the most eminent representatives
of the Lvov-Warsaw School, which was generally under a strong and many-sided
influence of Franz Brentano’s philosophy. Czeżowski himself was a disciple of Kazimierz
Twardowski, the founder of the Lvov-Warsaw School and one of the most prominent
of Brentano’s students. That is why contemporary discussions about the philosophy of
the Lvov-Warsaw School, and about Czeżowski’s philosophy in particular, are often set
against Brentanian ideas. However, I would like to consider Czeżowski’s epistemology
in a historical context which is broader than the Brentanian philosophy only. In fact, it
is a context to which also the latter belongs; I mean the epistemological doctrines of
Descartes, Locke, Hume, Reid and the contemporary views which refer to them.
Keywords Belief · Epistemological individualism · Faith · Fallibilism · Knowledge ·

Truth
Mathematics Subject Classification (2000) Primary 01A60, 03B42; Secondary 01A70
1 Introduction
The aim of the paper is to present and critically evaluate Tadeusz Czeżowski’s views
concerning knowledge and belief in the context of some non-Brentanian epistemological
concepts and theories. Czeżowski belonged to the most eminent representatives of the
Lvov-Warsaw School, which was generally under a strong and many-sided influence
of Franz Brentano’s philosophy. Czeżowski himself was a disciple of Kazimierz Twar-
dowski, the founder of the Lvov-Warsaw School and one of the most prominent of
Brentano’s students. That is why contemporary discussions about the philosophy of the
Lvov-Warsaw School, and about Czeżowski’s philosophy in particular, are often set
against Brentanian ideas. However, I would like to consider Czeżowski’s epistemology
in a historical context which is broader than the Brentanian philosophy only. In fact, it
is a context to which also the latter belongs; I mean the epistemological doctrines of

138 D. Łukasiewicz
Descartes, Locke, Hume, Reid and the contemporary views which refer to them.1 By
doing this I do not want to undermine the great importance of the Brentanian heritage
for Polish analytic tradition, but I just aim to broaden a historical context into which that
philosophy, as I believe, can be rightly included. To achieve this aim, in the first part of the
paper I will present some major ideas of early modern and contemporary epistemology
which are relevant to our further considerations and, next, in that context I will discuss
Czeżowski’s views on knowledge and belief.
In contemporary discussions concerning knowledge, belief and justification in general,
there are some concepts and dichotomies which outline the mainstream of the debate,
such as epistemological individualism vs. collectivism, methodism vs. particularism,
internalism vs. externalism, foundationalism vs. fallibilism or Cartesian vs. Spinozan
belief-formation models. All these topics have been widely discussed in the epistemo-
logical literature, but I decided to recapitulate them briefly below because they are rather
absent from works on the Lvov-Warsaw School. Neither were these concepts used by the
philosophers of the School; some of them could not be used simply because they were not
formulated yet at the time when the Lvov-Warsaw School developed.
2 Modern Epistemological Individualism: Faith and Testimony

in Epistemological Framework
Let us begin with the clarification of the doctrine of epistemological individualism since,
as it will become clear below, it plays a particularly significant role in Czeżowski’s theory
of knowledge. The idea of epistemological individualism can be elucidated by quoting
Antoine Arnauld and Pierre Nicole of the Port Royal School, who themselves were not
proponents of that view, but made useful conceptual distinctions. They write as follows:
For there are two general paths that lead us to believe that something is true. The first is knowledge
we have of it ourselves, from having recognized and examined the truth either by the senses or by
reason. This can generally be called reason, because the senses themselves depend on a judgment
by reason, [. . . ]. The other path is the authority of persons worthy of credence who assure us that a
certain thing exists, although by ourselves we know nothing about it. This is called faith or belief,
following the saying of St. Augustine: Quod scimus, debemus rationi, quod credimus, autoritati
(What we know we owe to reason, what we believe, to authority). [2, p. 260]
Similar ideas can be traced in Thomas Aquinas’ views, according to which:

[. . . ]it is needful that he [man] be able to stand with as much certainty on what another knows but
of which he himself is ignorant, as upon the truths which he himself knows. Hence it is that in
human society faith is necessary in order that one man give credence to the words of another, and
this is the foundation of justice [. . . ]. [1, III, art. i. 3]
1 It
is worth mentioning here Jacek Jadacki’s remark: “during the last 50 years traditional epistemology
has not been cultivated, in principle, in the Lvov-Warsaw School. It was Ajdukiewicz’s principle which
was decisive here: according to him, epistemological problems could be studied only after a suitable
paraphrase and after such a paraphrase they become indistinguishable from respective methodological
problems” [14, p. 97]. In my paper I would like to discuss Czeżowski’s epistemology just in the context
of “traditional epistemology”, whose problems are today probably more appealing and popular among
epistemologists than the formal approach preferred by the followers of Ajudukiewicz’s paradigm.
Tadeusz Czeżowski’s Theory of Knowledge and Beliefs 139
In both quotations we find a sharp distinction between beliefs acquired by reason

and beliefs acquired by faith. The former are beliefs based on perception or inference
and the latter include testimonial beliefs which we owe to someone else’s authority.
The concept of authority used above is broad and embraces individual agents as well
as institutional bodies. A crucial presupposition of the distinction between faith and
reason is that individuals are unable to acquire all beliefs which are worthy of holding
by themselves. Both Thomas Aquinas (and surely other medieval thinkers as well) and
the two authors of the Port Royal Logic claim that testimonial beliefs, or simply “faith”,
make a valuable complement to perceptual and inferential beliefs, or simply, to “reason”.
What is important here is that beliefs belonging to the body of faith can be true; and truth
is a great value in itself. That is why faith is a valuable and legitimate complement to
reason.
Contrary to that tradition, Rene Descartes stressed very firmly and clearly that beliefs
should be acquired by reason alone:
[. . . ] hence I thought it virtually impossible that our judgments should be as unclouded and firm as
they would have been if we had had the full use of our reason from the moment of our birth, and
if we had always been guided by reason alone. [9, p. 117]
Furthermore, beliefs held by others are worthless in our gaining knowledge:

[. . . ] and yet a majority vote is worthless as a proof of truths that are at all difficult to discover; for
a single man is much more likely to hit upon them than a group of people. [9, p. 119]
The same approval of an individual search for knowledge one may find in John Locke’s
seminal Essay:
[. . . ] we should make greater progress in the discovery of rational and contemplative Knowledge, if
we sought it in the Fountain, in the consideration of Things themselves; and made use rather of our
own Thoughts, than other Men’s to find it. For, I think, we may as rationally hope to see with other
Men’s Eyes, as to know by other Men’s Understandings. So much as we our selves consider and
comprehend of Truth and Reason, so much we possess of real and true Knowledge. The floating
of other Men’s Opinions in our brains makes us not one jot the more knowing, though they happen
to be true. What in them was Science, is in us but Opiniatry, whilst we give up our Assent only
to reverend Names, and do not, as they did, employ our own Reason to understand those Truths
which gave them reputation [. . . ] In the Sciences, every one has so much, as he really knows and
comprehends: What he believes only, and takes upon trust, are but shreds; which however well in
the whole piece, make no considerable addition to his stock, who gathers them. [17, p. 58]
Thus both Descartes and Locke claim that only beliefs acquired by reason can be
regarded as knowledge and can be accepted as legitimate. Since beliefs acquired by
reason are beliefs which an individual epistemic agent can gain solely by her own efforts,
that position is called epistemological individualism. It is clear that epistemological
individualism is a view which was not quite shared by the medieval scholars and their
early modern followers (the Port Royal School), who secured some room in the epistemic
framework for authority-based beliefs.
It is worth mentioning that epistemological individualism was strongly defended also
by the ancient Greeks: Socrates, Plato and Aristotle. They all subscribed to the view
that there is a distinction between two kinds of knowledge, doxa and episteme. Doxa
concerns perishable things given in perception and includes testimonial beliefs (opinions).
Episteme concerns eternal forms and ideas; it is necessary and self-evident. Knowledge
about eternal forms of perishable things—defined by Plato as justified and true belief—
140 D. Łukasiewicz
can only be acquired by inference or a specific kind of transsensorial perception called

noesis.
Importantly, epistemological individualism has some ethical implications, perhaps
most clearly formulated by Locke. If knowledge is of great value for humans and
can be gained only by an individual effort, there are some epistemic duties and moral
responsibility resulting from this fact.
However, this strong epistemic individualism is not the only approach to knowledge
that we can find among early modern epistemologists. Apart from the Port Royal School,
which maintained the medieval divide between reason and faith, we find some insights
appreciating the role of faith in human life in David Hume’s philosophy. In his Enquiry
Concerning Human Understanding, Hume writes as follows:
[. . . ] there is no species of reasoning more common, more useful, and even necessary in human
life, than that which is derived from the testimony of men, and the reports of eye-witnesses and
spectators. [12, p. 111]
The idea defended by Hume is that if we strictly followed Descartes’s and Locke’s
epistemological principles, we would have significantly less true beliefs than in a situation
when we do not base all our beliefs on perception and reasoning only. Needless to say, if
one consequently adhered to the Cartesian and Lockean ideals, this would result in one’s
severe cognitive deprivation in everyday life.
However, it was not Hume but Thomas Reid who most vehemently and convincingly
defended the status of beliefs based on the word of others, i.e. the beliefs which Arnauld
and Nicole had called “faith”. Reid claims that it is futile to try to justify the immense body
of true beliefs that we owe to others and their reports by consulting our own intellectual
resources only—in that case we would be left with little knowledge indeed [19]. In Reid’s
philosophy of common sense, there are two principles which are assumed to guide our
cognitive behavior and which are deeply rooted in human nature; the principle of veracity
and the principle of credulity [20, pp. 193–194]. The first principle is “a propensity
to speak truth, and to use the signs of language so as to convey our real sentiments”.
The principle of credulity, in turn, says that humans quite naturally tend to believe what
they are told. Reid’s position regarding the nature of testimonial beliefs (authority-based
beliefs included) is, however, more radical than the views defended by Thomas Aquinas
and the Port Royal School. On his view, testimonial beliefs are part of our knowledge,
not faith, and the act of testimony itself counts as legitimate and sufficient justification
for them. Reid even claims that testimonial beliefs have the same epistemological status
as perceptual and inferential beliefs. In other words, beliefs based on testimony are as
much justified as beliefs based on perception (introspection and memory included) and
reasoning. Reidian bold epistemic ideas are grounded in a more general doctrine of human
nature. According to this doctrine, human beings are sincere, truthful and willing to rely
on the word of others. Sincerity and trust are imposed by the requirements of our everyday
life and cognitive behavior, which are inescapably social, not solitary, in nature.
3 Methodism, Particularism and Belief-Formation Process
Connected with the question of what may count as a proper justification for our
beliefs (and which beliefs may in effect constitute knowledge) is the divide between
epistemological particularism and methodism.2 Richard Fumerton spelled out the core
of the distinction between the two positions as follows:
Does one first decide what one knows and then try to learn from paradigmatic examples of
knowledge the necessary and sufficient conditions for knowledge? Or does one discover first the
necessary and sufficient conditions for knowledge and apply what one learns do discover what one
knows? [10, p. 36].
Epistemological methodism, summarized in the second of the above questions, is an

approach typical of Cartesian-style philosophizing; we define first what knowledge is
and then we apply the adopted criteria to particular beliefs to check if they meet those
criteria and may be considered knowledge. The particularist understanding of knowledge,
articulated in the former question and consistent with Reid’s approach, recommends
a reversed procedure. At first we accept some self-evident, common sense beliefs as
indubitable truths that we know and on that basis we proceed to define what knowledge
is. Needless to say, since particularism takes our common-sense beliefs as the starting
point in defining knowledge, it offers a broader and more inclusive concept thereof—in
contrast to methodism whose criteria-first approach typically results in a more restrictive
understanding of what it means to know something.
A very interesting epistemological question, much related to the divide between
methodism and particularism, is the following: how do we come to believe that a
given proposition is true? In other words, what does a paradigmatic belief-formation
process involve? Again, there are two competing approaches to the problem of how our
beliefs are formed—named after their famous proponents the Cartesian and Spinozan
models, respectively. On Descartes’s view, in order to believe something, we must first
comprehend the propositional content of a (potential) belief which is given to the mind and
only subsequently, if the content is clear enough and some sufficient evidence supports
it, can we accept the content as true belief; or we can reject it as false if the evidence
we possess is unfavorable. On the Spinozan view, our comprehension of belief’s content
is simultaneous with our initial acceptance of it; we accept any incoming belief as true,
which act is inseparable from our understanding thereof, and only later can the proposition
be rejected if we realize that it is incompatible with some other beliefs we hold [24, Part 2,
49].3 Obviously, the Spinozan account of how we must initially accept beliefs in order to
understand them is absolutely congruent with the workings of Reid’s credulity principle;
human beings have a strong, psychologically grounded tendency to believe.
2 The divide concerns the so-called Problem of the Criterion famously formulated anew and brought to
the fore by Roderick Chisholm [3].
3 As pointed to by Gilbert [11, p. 108], Baruch Spinoza was the first thinker to question the Cartesian
divide into the passive (comprehension) and active (assessment) powers of the mind, therefore the second
of the above-mentioned models of belief formation processes can be called Spinozan. For more on the
two models of belief formation, see also Lechniak [16, pp. 73–115].
142 D. Łukasiewicz
4 What Can Justify Beliefs: Basic Beliefs, Internalism

and Externalism
Another pair of vitally important epistemological concepts which we will refer to when
discussing Czeżowski’s views on knowledge and belief are internalism and externalism.
These two concepts recur in any contemporary discussion on epistemic justification,
therefore it will suffice here to mention only the major claims of both. Epistemological
internalism is a fairly traditional view connecting belief’s justification with an individual’s
awareness of it and epistemic responsibility. A particular belief is justified only if the
epistemic agent holding that belief has mental access to and is aware of what confers
justification on that belief. Epistemological externalism, which is a more recent view,
rejects the foregoing claim that in order to have justified belief, the epistemic agent
must be aware of or must have potential access to what justifies that belief. Externalist
reliabilists, for example, maintain that a belief can be sufficiently justified solely by the
fact of being generated by a reliable, truth-conducive cognitive process, of which fact the
epistemic agent need not be aware whatsoever.
Finally, in our further discussion of Czeżowski’s views, we will refer to the concept
of basic beliefs. Let us remember that in order to have the status of being basic (or
foundational), the belief in question is to be justified non-inferentially, by immediate
experience—be it introspection, sensory perception or a kind of intellectual intuition
(noesis). In other words, a basic belief cannot be justified by itself or by any other belief
because then we would encounter the infinite regress problem. Since basic beliefs do
not require justification from other beliefs but they provide justification for other, non-
basic beliefs, they make a foundation for the whole body of our knowledge—that is
the main principle of foundationalism of any strand. In its classical version—defended
by Descartes—foundationalism requires that basic beliefs be infallible; they cannot be
refuted by any future experimental data. Therefore, knowledge built on them is certain
and indubitable. A moderate foundationalism (fallibilism) is a less restrictive view; it
holds that there exist basic beliefs but they need not be infallible, i.e. it is possible that
they will be refuted by some experimental data in the future. According to moderate
foundationalism, for example, perception-based beliefs can be regarded as properly basic
even though they are not infallible. There has been much discussion and critique in the
contemporary epistemology regarding the very notion of basic beliefs, what justifies them,
and the possibility of non-inferential justification of knowledge. As argued by Sellars
[23] and many others, the main problem is whether basic beliefs as such are possible
at all, i.e. whether any belief can be justified by non-propositional and non-judgmental,
purely sensoric experiential data. Another set of problems that any foundationalism faces
is related to how a set of basic beliefs—rather limited in number—is to confer justification
on the whole body of empirical knowledge that we claim to have. It is not necessary and
rather impossible to dwell here on the numerous controversies generated by the concept
of basic beliefs; clearly the problems involved are much deeper than the sheer question of
whether a basic belief may be fallible or infallible.
5 Czeżowski’s Theory of Knowledge and Epistemological

Individualism
Having introduced some main ideas and dichotomies of the early modern and contempo-
rary theory of knowledge, I would like to defend the thesis that Tadeusz Czeżowski’s
epistemological views could be characterized as epistemic individualism, methodism,
internalism and fallibilism (to a certain degree), and that Czeżowski assumed the
Cartesian model of belief-formation, not the Spinozan one. It is also worth noting that
his theory of knowledge may be considered, perhaps with some minor reservations, to
represent the majority view in the Polish analytic camp in the twentieth century. There
are, however, some problems arising when Czeżowski’s theory of knowledge is analyzed
in detail; we will discuss some of them below.
Let us begin with epistemological individualism. Undoubtedly, Czeżowski sharply
distinguished reason (knowledge) and faith [7, p. 56]. In his view, a belief which is
regarded as part of knowledge is self-evident, obvious and firm; its object is compre-
hended in a clear and distinct way. A belief which is part of faith is not self-evident
but vague, shaky and somewhat sketchy. It is typically grounded in emotions, desires or
authority. In Czeżowski’s theory of knowledge, a crucial role is played by the concept
of self-evidence (obviousness). There are four kinds of self-evidence and self-evident
beliefs: apodictic, introspective, perceptual and inferential (“demonstrative”). Apodictic
self-evidence is predicated of beliefs whose content are basic ontological principles such
as the principle of non-contradiction, the axioms of deductive systems or the principle
of causality [4, 8]. According to Czeżowski, it is simply the content of an apodictically
self-evident belief which is responsible for its self-evidence. Introspective self-evidence
is predicated of beliefs whose objects are mental states of a conscious subject. The
introspective self-evidence is not apodictic because, as Czeżowski writes, mental states
are “ontologically contingent” [7, p. 56]. Thus, we may infer—Czeżowski himself does
not say it explicitly—that ontological principles (the principle of non-contradiction
included) are ontologically necessary. Perceptual self-evidence, in turn, is predicated of
perceptual beliefs. It is vividness and persistence of perception which are responsible
for self-evidence of perceptual beliefs. These qualities of a perception act can be saved
in memory and as long as memory-based beliefs are vivid and persistent, they can be
counted as self-evident. Inferential self-evidence is predicated of beliefs which have been
logically deduced from self-evident beliefs (premises); an inferential chain transfers and
preserves self-evidence of its premises. These four kinds of self-evidence correspond
with three kinds of propositions: analytic propositions, empirical propositions (perceptual
and introspective propositions included) and inferential propositions.4 Czeżowski clearly
4 Czeżowski writes: “Cztery epistemologicznie rozróżnione rodzaje zdań odpowiadajace

˛ wymienionym
czterem rodzajom przekonań oczywistych, mianowicie zdania analityczne, zdania empiryczne doświad-
czania wewn˛etrznego i doświadczenia zmysłowego oraz zdania uzyskane droga˛ wynikania logicznego
ze zdań pierwszych trzech rodzajów wyczerpuja˛ zakres zdań wchodzacych
˛ w skład wiedzy. Wiara nie
wymaga oczywistości.” [7, p. 56].
144 D. Łukasiewicz
states that our knowledge is composed only of the foregoing analytic, empirical and
inferential propositions.5
I would like to make three remarks concerning Czeżowski’s views on self-evidence and
knowledge. Firstly, we find in Czeżowski’s philosophy two different concepts of empirical
self-evidence and empirical propositions. The first one is narrow and embraces only
sense perception and introspection. The second one is broad and includes also axiological
perception (moral and aesthetic perception of values) and intellectual intuition (noesis).
In 1945, when Czeżowski was writing his paper in which he provided the foregoing
classification of propositions, he held a narrow concept of perception.6 However, later
on he changed his mind and classified axiological insights as axiological self-evidence. It
was not meant as a new kind of self-evidence but simply as a component of perceptual
knowledge.
Secondly, Czeżowski considered ontological principles to be analytic, but they could
be regarded as synthetic if we assume the existence of an ontological perception
(noesis) in which the basic ontological principles can be given. This interpretation could
be supported by Czeżowski’s metaphysical theory of transcendentals: truth, goodness
and beauty [6]. If we can “perceive” goodness and beauty in axiological insights—as
Czeżowski decisively claims—why not assume that we are capable of “perceiving” the
truth of ontological principles?
The third point is that, in Czeżowski’s framework, beliefs/propositions based on
introspection and perception conceived of in a broad sense are directly justified and
inferential propositions are justified indirectly [7, p. 56]. This means that a belief can be
justified by perception, introspection, memory or inference, or, if it is an analytic belief,
it is justified directly by its content, and only beliefs justified in that way can be regarded
as knowledge. Testimonial beliefs cannot be counted as knowledge and they should be
taken as a matter of faith only. For this reason, Czeżowski’s theory of knowledge is
an exemplification of epistemological individualism, paradigmatically represented in the
early modern philosophy by Descartes.
There is another important and philosophically interesting point in Czeżowski’s
philosophy which should be mentioned while discussing the Cartesian inspirations in his
theory of knowledge. It is a relation between self-evidence and truth. Czeżowski claims
that every true belief is self-evident but not conversely, i.e. not every self-evident belief
is true [7, p. 57].7 But let us note that if every true belief is self-evident and every self-
evident belief is justified, then it follows that every true belief (proposition) is justified.
But this conclusion seems to be simply false and a bit surprising. Surely, not all true
beliefs are epistemically justified; there are true beliefs acquired by wishful thinking,
guess or chance. A correlation between truth and self-evidence suggested by Czeżowski
5I do not want to go here into a detailed discussion on the structure of belief and the relation between
proposition and belief. Suffice it to say that in Czeżowski’s case a proposition is the content of a belief
and a state of affairs is an object of a belief [18].
6 I mean the paper published first in French, “Quelques problems ansciens sous la forme modern” (Studia
Philosophica 1939/1946, 101–113), and next in Polish as “Niektóre dawne zagadnienia w nowoczesnej
postaci”, the Polish version was published first in Odczyty filozoficzne in 1958 and later in the second
edition of Odczyty filozoficzne in 1969 [7].
7 This is contrary to later Brentano’s view which claims that every self-evident belief is true. Artur
Rojszczak [21] discussed Brentano’s conception of truth in more detail.

is not quite Cartesian, but it is very strong, though asymmetrical. That is why Czeżowski’s
theory of knowledge leads to a false conclusion.
There is another striking point in Czeżowski’s theory of knowledge which might
be elucidated by the aforementioned strong but asymmetrical correlation between truth
and self-evidence. The definition of knowledge which Czeżowski used says simply that
knowledge is a set of true beliefs/propositions [5, pp. 8, 57]. Therefore, Czeżowski’s
definition of knowledge is clearly less demanding than the classical definition thereof,
formulated by Plato, which defines knowledge as true and justified belief. However, if
we take into account that every true belief is self-evident and every self-evident belief
is justified (by perception, introspection, memory or inference), then every true belief
belongs to the body of our knowledge. One can say that all true beliefs are justified and
belong to our knowledge. Needless to say, two propositions: every true belief is justified
and every true and justified belief is knowledge are not logically equivalent. The first of
them is simply false. I suggest that we should treat Czeżowski’s definition of knowledge
as a semi-Cartesian component of his epistemology.
As argued above, Czeżowski’s stance in epistemology is epistemic individualism.
However, in his writings there are some remarks and comments which might undermine
this interpretation and therefore they deserve our attention.8 In 1959, Czeżowski states
that we owe only a small portion of all our knowledge to our own experience and
reasoning [5, p. 43].9 Most of our beliefs are “second-hand beliefs”, which we owe to
what other people tell us, what we are taught at different levels of education, what is said
in the media, etc. The problem is that if a belief is based on the authority of teachers or
books or it is based on other people’s reports, then it should be counted as faith rather than
knowledge. It seems that Czeżowski’s position might be defended as coherent and tenable
because, as he claims, we accept belief p on the basis of our perception and understanding
of its content, and not simply because someone else (agent O) informed us about p.
Czeżowski assumes that, having perceived visual or auditory signs and having understood
the content these signs stand for, we ourselves evaluate their cognitive and epistemic value
and finally accept the belief as true or reject it as false. However, there arises a problem
here. Given Czeżowski’s asymmetrical correlation between truth and self-evidence, if p is
true, then there must be some relevant self-evidence corresponding with p. But if p says,
for example, that Napoleon was defeated at Waterloo in 1815, what kind of self-evidence
corresponds with p? Surely, it cannot be the mere perception of visual or auditory signs
and comprehension of the meaning to which they refer. Czeżowski himself points to the
mental states of agent O who said or wrote that p. But what kind of self-evidence O
could have, if O was not an eyewitness of the reported event? Most probably O referred
8A reflection about epistemological individualism in Czeżowski’s philosophy is not only a matter of

historical reconstruction but it can have more general philosophical significance.
9 Czeżowski writes: Jednakże we współczesnych warunkach życia tylko nieznaczny ułamek posiadanej
wiedzy zawdzi˛ecza każdy z nas własnemu doświadczeniu i własnej myśli. Wi˛ekszość przekonań
nabywamy niejako z drugiej r˛eki, przejmujac ˛ gotowe cudze przekonania droga˛ informacji. Dzieje si˛e
to za pośrednictwem słowa mówionego i pisanego, w rozmowie i w listach, przez dzienniki i przez
radio, w szkołach przez ksiażki
˛ i czasopisma naukowe. Motywem przekonania jest we wszystkich tych
przypadkach słuchowe lub wzrokowe spostrzeżenie mowy wraz ze zrozumieniem tego, co si˛e spostrzega;
pośrednio jednak trzeba odnieść powstawanie przekonań ta˛ droga˛ do cudzych przeżyć, znajdujacych
˛ swój
wyraz w spostrzeganych znakach [5, 43].
146 D. Łukasiewicz
to the mental states of some eyewitnesses of p. But why should we believe that O indeed
referred to their mental states and that those eyewitnesses were reliable in their reports?
We are usually not in a position to check that. Arguably, we accept p not because we
have any epistemic access to the mental states of other agents, but because we have some
perceptual and inferential self-evidence that agent O is a competent and sincere person.
Therefore, finally we alone are responsible for justifying our belief that p, which is a
clearly reductionist and individualist stance.
Czeżowski admits that humans have a natural propensity to believe what they are
told and this propensity is rooted in their “drive to imitate” [5, p. 43]. However, beliefs
acquired and accepted by “the will to believe”, to use James’s terminology, are not
knowledge. Moreover, Czeżowski holds that the will to believe rooted in the imitation
drive is responsible for many negative phenomena, such as the spread of rumors and
superstitions or giving in to propaganda and other bad epistemic practices—in short,
for many false beliefs.10 That the will to believe and the drive to imitate are assessed
negatively is clear from Czeżowski’s critical remark describing these propensities as par-
ticularly common among young children and primitive tribes, or when mass psychology
is at play. This opinion seems a bit strange since the will to believe is grounded simply
in human nature and hence it can be attributed to every human, not only to children or
primitive tribes. Czeżowski does acknowledge that the will to believe may at times have
some positive effects; it may help in someone’s recovery, for example. However, there
is a huge gap between Czeżowski’s noticeably reserved approach to testimonial beliefs
and Reid’s optimistic epistemological project, in which the human inclination to believe
other people is conceived of as our basic cognitive capacity, on a par with perception (cf.
Reid’s principles of credulity and veracity). For Czeżowski, the natural fact that we tend
to rely on the word of others is a kind of epistemic imperfection in us, if not downright
impediment in our gaining knowledge—certainly not an epistemic asset. For that reason,
testimonial beliefs should be very critically assessed before they are accepted.
But, we may still wonder, if the will to believe others is rooted in human nature, how
can we expect an epistemic agent to be ready to asses critically his/her beliefs acquired via
testimony and, in this way, to act against his/her natural endowment? I suppose Czeżowski
might answer the above question in at least two ways. Firstly, following Aristotle and
many others, he might appeal to the great value of knowledge and truth, which surpasses
our natural inclinations and imposes some moral obligation on us to do our best in
striving to achieve this value [7, pp. 211–212).11 Secondly, Czeżowski always underlined
10 Czeżowski says: “Jeżeli przekazywanie komuś przekonań innej osoby nie dzieje si˛e w postaci
informacji czysto rzeczowej lub nauczania to nazywamy je poddawaniem lub sugerowaniem przekonań,
stosowanym lub nadużywanym w reklamie, propagandzie i wszelkiej agitacji, ułatwiajacym
˛ szerzenie si˛e
pogłosek i plotek; ma ono również jednak i dodatnie zastosowania, np. w lecznictwie chorób nerwowych,
gdzie poddanie choremu przekonania o możliwości wyzdrowienia jest nieraz warunkiem uleczenia.
U jego podstawy leży naturalna skłonność do uwierzenia w to, co ktoś drugi wyznaje szczerze lub
nieszczerze jako swe własne przekonanie; przejawia si˛e w niej właściwy naturze ludzkiej pop˛ed do
naśladownictwa, ten sam, który tłumaczy nam zjawiska mody. Ów pop˛ed jest w dziedzinie psychicznej
jakby zaraźliwościa˛ pewnych zjawisk psychicznych i wyst˛epuje wyraźnie zwłaszcza tam, gdzie nie
dochodza˛ do głosu indywidualne różnice w rozwoju psychicznym jednostek: u ludów pierwotnych, wśród
dzieci, w zjawiskach masowych tzw. psychologii tłumu.” [5, p. 43].
11 Czeżowski writes: “Sadz˛
˛ e bowiem, że także w dzisiejszych czasach zachował swa˛ wartość ideał
filozofa starożytnego-m˛edrca wcielajacego
˛ zasady teoretyczne” [7, p. 212].
the role of character-making and the importance of moral and cognitive training, which
contribute to the development of such intellectual virtues as criticism, exactness, open-
mindedness, conscientiousness or reliability [7, p. 212]. However, there is still some
tension in Czeżowski’s views on knowledge because, as said above, he is explicit that
we owe only a small part of what we know to our own experience and reasoning. If
knowledge is of such great value to us, which is rather unquestionable, and we owe only a
small portion of it to ourselves, this situation can be a bit disturbing for an epistemological
individualist for whom an individual search for truth is of great value too.
6 Czeżowski’s Internalism and Fallibilism
Regarding the internalism/externalism divide, Czeżowski’s position appears to be on the

internalist side. If a belief is justified by self-evidence of some kind, be it apodictic,
perceptual, introspective or inferential self-evidence, and all justified beliefs are self-
evident, then the epistemic agent holding a justified belief clearly has access to the relevant
justification [7, p. 56]. However, also here one could trace some tension in Czeżowski’s
views. Let us remember that when discussing in [5] the problem of what justifies our
second-hand knowledge, i.e. the knowledge that we owe to someone else’s reports and
teachings (significantly, the term knowledge is used in that context), Czeżowski points to
the mental states of our informers who said or wrote that p.12 The mental state of another
person, i.e. his/her perception or reasoning, may provide justification for our belief (“the
extendedness hypothesis”), but then it is clearly an externalist position.
It is difficult to speculate how Czeżowski would solve the problem of the criterion,
famously reintroduced into the epistemological debate by Roderick Chisholm [3], but
most probably he would subscribe to epistemological methodism rather than to particu-
larism. The attention Czeżowski paid to the notion of self-evidence, the types of it and the
criteria which beliefs must fulfill to count as knowledge, place him clearly in the Cartesian
camp. Let us repeat here the features of beliefs contributing to our knowledge. A belief
regarded as part of knowledge must be self-evident, obvious, firm and its object must be
comprehended in a clear and distinct way [7, p. 56].13
Another question is whether Czeżowski’s views on the justificatory role of self-
evidence could be regarded as a version of foundationalism. Certainly, it was not
Cartesian-style foundationalism, but I think Czeżowski might be understood as repre-
senting a bit reformulated version of moderate foundationalism, or fallibilism. He held
the view that there are basic beliefs but they are fallible, and even if belief is self-evident,
it can be refuted. Every true belief is self-evident but not every self-evident belief is true.
According to Czeżowski, every element of a scientific theory, be it an inductive law,
deductive axioms, historical hypotheses or a singular proposition based on experiential
data, can be rejected; “there are no infallible premises in empirical sciences” [6, p. 88].
When arguing against Cartesian foundationalism, he resorts to the history of science. In
12 See
footnote 9 above.
13 Amethod of an analytic description can be used to establish the characteristic of knowledge. This
method is very similar to Plato’s noeis or Aristotle’s induction.
148 D. Łukasiewicz
his numerous papers regarding the methodology and philosophy of science, he referred
to two—in his view particularly significant—scientific discoveries. The first was the
discovery of the non-Euclidean geometries. The non-Euclidean systems of geometry, as
Czeżowski argued, undermined Descartes’s and Kant’s belief in the apparently infallible
axioms of the classical geometry. He stressed that there can be various incompatible
deductive systems whose axioms are counted as true and, as David Hume had already
claimed, that there is no direct and only one relation between formal systems and
empirical facts. The second discovery to which Czeżowski referred was made by the
French mathematician and logician, Joseph Gergonne (1771–1859). Gergonne argued that
the axioms of deductive systems can be regarded as implicit definitions which—similarly
to the algebraic equations containing variables—consist of terms whose meaning is
constituted by the stipulation that the axioms are true.
Czeżowski observed that not only a priori beliefs and propositions are fallible, but
introspective, perceptual and moral beliefs can be fallible as well. But he was not a
skeptic. There seems to be no infallible knowledge, but some beliefs are more probable
than others and some beliefs are better justified than others—this allows for a progress
in science [5, 6]. That progress is possible because there are methods of inferential
justification of beliefs by calculations of probabilities. Czeżowski himself subscribed to
Hans Reichenbach’s logic of probability [7, p. 59].
But it must be stressed here that even despite his firm statements regarding the fallible
nature of all scientific knowledge, Czeżowski is clear that there are some infallible beliefs.
These are simple moral principles which are universally shared by all people in various
cultures and times. Therefore, we are also allowed to assume that they are fully reliable
and certain—or simply infallible [7, p. 44].
Another problem is how to interpret the epistemic status of ontological principles
such as the principle of non-contradiction or the principle of causality.14 Czeżowski
elaborated a rather sophisticated doctrine of how formal deductive systems are related to
the empirical world.15 From this point of view, ontological principles can be regarded
only as “interpretations” of deductive axioms and theorems. If a logical axiom or a
theorem is confirmed within the framework of a physical theory, be it the classical
Newtonian mechanic, relativistic cosmology or quantum mechanics, it is “materially
true”, as Czeżowski in a bit old-fashioned style would say. But since we will never be
able to provide such a confirmation—taking into account all possible interpretations and
all regions of the universe—we will never know whether and which formal and logical
truths are always materially true. In other words, we cannot say that a given principle
is infallible because we do not know whether every possible empirical and ontological
interpretation of a formal axiom or theorem (be it the logical law of non-contradiction
or the law of excluded middle) is materially true. On the other hand, Czeżowski, as
mentioned above, viewed the ontological principle of non-contradiction as necessary
and apodictically self-evident. To sum up Czeżowski’s stance concerning the problem
14 The principle of causality was strongly defended by the members of the Lvov-Warsaw School in face
of quantum indeterminacy [25, pp. 249–250].

15 Surely, Czeżowskian model was much more sophisticated than Russell’s model. It was Russell [22, p.
169] who said that “logic is concerned with the real world just as truly as zoology, though with more
abstract and general features”.
of foundationalism, I think that we could say that he was not a Cartesian foundationalist,
but his general adherence to what might be called moderate foundationalism/fallibilism is
not unproblematic either since he did assume some infallible moral and ontological basic
beliefs.
7 Czeżowski and Some Views on Belief-Formation Processes
Brentano’s and Twardowski’s heritage is clearly present in Czeżowski’s view on belief-

formation process, but I wanted to point to the fact that his views in that regard are
also a continuation of Descartes’s ideas. In his theory of belief formation, Czeżowski
directly refers to Brentano’s and Twardowski’s descriptive psychology.16 According to
the Brentanian approach, a belief (judgment) consists in our acceptance or rejection
of the existence of an object given to us in an act called presentation.17 Certain
technical distinctions aside, we can simply say that presentation is a neutral act of
comprehension (or understanding) related to a given content.18 The asymmetrical relation
between presentation and belief (judgment) is necessary; there can be no belief without
presentation, but presentation does not entail belief. As stated by Czeżowski, in order
to assert that “it rains”, one must know before what “rain” is, and in order to reject the
claim that “ghosts exist”, one must know before what the sentence means [5, p. 59].
This is exactly the Cartesian view of how the belief formation proceeds; an act of neutral
comprehension of a given content is prior to the act of acceptance or rejection. Though
apparently convincing, such a model of belief formation is from psychological perspective
rather inadequate, and, consequently, has been criticized by many philosophers and
scholars. According to developmental psychology, human beings have a strong natural
inclination to accept beliefs right away, without prior considering their content, which
view was also admitted by Czeżowski himself (cf. “the drive to imitate”). Thus, beliefs are
accepted as if simultaneously with their comprehension, without earlier “presentation”,
to put it in the Brentanian terminology. Acceptance is psychologically prior to critical
assessment and possible rejection. Whatever the evolutionary basis of this mechanism
of belief formation may be and its cognitive advantages or disadvantages, we naturally
and involuntarily tend to believe. As Elżbieta Łukasiewicz describes and summarizes that
problem:
First and foremost, in the mental development of children, the capacities of doubt and disbelief
are acquired much later and are deemed more complex than acceptance and trust, which are
ontogenetically prior. As Reid’s argument has it, if credulity were not a gift of nature and were
not primary in humans, but were developed by children on a par with reasoning and other mental
16 Twardowski’s theory of judgment is presented and discussed in a comprehensive way by Jacek Jadacki
[13, pp. 178–184].
17 On relations between presentation and judgments see also [18].
18 There is a distinction between an act (be it presentation or judgement), the content of an act and its
object. The distinction in question was forcefully defended by Kazimierz Twardowski and taken over by
all his disciples, including Czeżowski. Judgment is an assertive act which can be identified with belief.
The characteristic features of judgement (belief) are its propositional structure and being articulated in a
language.
150 D. Łukasiewicz
capacities, the attitude of trust should be greater in adolescents and adults than it is in children.
Very young children should reveal the most neutral attitude to what they are told compared to
other age groups. Since it is exactly the other way round, i.e. credulity is strongest in young
children and is lessened with age and experience, then it follows that trust must be the natural
endowment of human beings.19 So children’s minds are not Cartesian systems—were they such
systems, believing and disbelieving would be equally easy for them and these two abilities would
develop at about the same time and rate, which is obviously not the case. [19]20
Thus, one could say that mere comprehension of a proposition which is free of any
assessment is a myth [19]. Also a semantic argument could be raised against the Cartesian
and Brentanian models of belief formation. As argued by Johnson-Laird [15, p. 110], in
order to understand the meaning of a sentence like John is a bachelor one must know
the conditions under which the sentence is true (or false). How could we understand the
meaning of the above sentence without somehow representing it in the mind as the real
state of affairs, i.e. with its truth-value attached to it? It is not clear what this truth-neutral
representation would represent then [15].
Let us note here that it was Spinoza who was the first to decisively question and
reject Descartes’s model of belief formation.21 The objections are serious and cannot
be easily dismissed. Yet, I suggest that the Cartesian and Brentanian models of belief
formation could be defended by appealing to the distinction between descriptive and
normative epistemology. The difference between the two approaches may be expressed
in the following two questions. The first one is how we really form our beliefs—we know
that the Cartesian explanation of that process is at least wanting, if not simply false. The
second question is how we should form our beliefs, and the Cartesian model appears
much more promising when applied to this problem. If knowledge and truth are of great
value, then we have epistemic, if not moral, duties to do our best to acquire them. Thus,
even if we have a natural tendency to uncritically accept any incoming beliefs in order
to understand them, we still have the duty to subsequently evaluate them and we are free
to reaffirm or reject them if they do not meet our epistemic standards. By that critical
assessment, we preserve the Cartesian and Brentanian character of epistemology whose
essential part is an individualist search for truth and knowledge.
In conclusion, we could say that Czeżowski’s theory of knowledge and belief is, via
Brentano’s and Twardowski’s views, significantly Cartesian. The core idea of this project
is that one is obliged to strive for knowledge and one can do it solely by an individual
effort and individual cognitive faculties such as perception, introspection, memory and
inference. Czeżowski was clearly aware of the role played in our life by others’ testimony
and he did acknowledge our natural, psychologically grounded inclination to believe
others, nevertheless, he was explicit in his writings that beliefs based on authority,
emotions or grounded in “the will to believe” do not deserve to be considered part of
knowledge; they belong to mere “faith”. Contrary to the long medieval tradition, the Port
Royal School, David Hume, Thomas Reid and their followers, who considered authority-
based faith to be legitimate source of beliefs, Czeżowski showed sympathy with Cartesian
and Lockean epistemic individualism and evaluated faith-based beliefs rather negatively.
19 Cf.Reid Inquiry into the Human Mind, 1997.

20 See also Gilbert [11] and Lechniak [16].
21 See Gilbert [11].
In general, the epistemological program characteristic of almost all representatives of

the Lvov-Warsaw School, not only of Czeżowski, was deeply rooted in the Cartesian
individualist tradition. Furthermore, it was also under great influence of the philosophy
of ancient Greeks, who also greatly valued the ideal of individual search for knowledge.
It is true that by downgrading the role of authority and testimony, Czeżowski’s theory,
like many other epistemological projects, does encounter the problem of testimonial
knowledge and lack of proper justification for a huge part of what we claim to know.
But, on the other hand, one can find some merits in Czeżowski’s epistemic individualism.
Undoubtedly, one of its great advantages depends on a very straightforward connection—
which in many externalists accounts is missing—between the value of knowledge and its
ethical dimension, i.e. the credit we deserve for achieving knowledge if we gain it through
our own effort and our epistemic responsibility if we fail to meet sufficient epistemic
standards.
Acknowledgements I would like to express my gratitude to Elżbieta Łukasiewicz for both reading and
improving the paper.
References
1. Aquinas, T.: Faith, Reason and theology. Questions I–IV of his Commentary on the Trinitate of
Boethius. Pontifical Institute of Mediaeval Studies, Toronto (1987). Trans. M. Maurer
2. Arnauld, A., Nicole N.: Logic or the Art of Thinking. Cambridge University Press, Cambridge (1996).
Trans. Jill Vance Buroker
3. Chisholm, R.M.: The Problem of the Criterion. Marquette University Press, Milwaukee (1973)
4. Czeżowski, T.: O metafizyce, jej kierunkach i zagadnieniach, 1st edn. Wydawnictwo Uniwersytetu
Mikołaja Kopernika, T. Szcz˛esny, Toruń (1948)
5. Czeżowski, T.: Główne zasady nauk filozoficznych, 3rd edn. Zakład Narodowy im. Ossolińskich,
Wrocław (1959)
6. Czeżowski, T.: Filozofia na rozdrożu. PWN, Warszawa (1965)
7. Czeżowski, T.: Odczyty Filozoficzne, 2nd edn. Towarzystwo Naukowe w Toruniu, Toruń (1969)
8. Czeżowski, T.: O metafizyce, jej kierunkach i zagadnieniach, 2nd edn. Wydawnictwo Antyk, K˛ety
(2004)
9. Eliot, C.W. (ed.): Descartes: Discourse on the Method, of Rightly Conducting Reason and Seeking
the Truth in Science. P.F. Collier & Son, New York (1909–1914)
10. Fumerton, R.: The problem of the criterion. In: Greco, J. (ed.) Oxford Handbook of Scepticism, pp.
34–52. Oxford University Press, Oxford (2008)
11. Gilbert, D.T.: How mental systems believe. Am. Psychol. 46(2), 107–119 (1991)
12. Hume, D.: In: Beauchamp, T.L. (ed.) An Enquiry concerning Human Understanding. Oxford
University Press, Oxford/New York (1999)
13. Jadacki, J.: On Kazimierz Twardowski’s descriptive semiotics and its metaphysical basis. In: Jadacki,
J. (ed.) Polish Analytical Philosophy, pp. 175–199. Wydawnictwo Naukowe Semper, Warszawa
(2009)
14. Jadacki, J.: The Lvov-Warsaw School and its influence upon the Polisch Philosophy of the twentieth
century. In: Jadacki, J. (ed.) Polish Analytical Philosophy, pp. 77–131. Wydawnictwo Naukowe
Semper, Warszawa (2009)
15. Johnson-Laird, P.N.: Computer and the Mind: An Introduction to Cognitive Science. Harvard
University Press. Cambridge (1988)
16. Lechniak, M.: Zmiana przekonań. Wydawnictwo Katolickiego Uniwersytetu Lubelskiego, Lublin
(2012)
152 D. Łukasiewicz
17. Locke, J.: An Essay Concerning Human Understanding. Gateway Editions, Chicago (1956). Preface
by Russell Kirk
18. Łukasiewicz, D.: Tadeusz Czeżowski on Existence. In: Chrudzimski, A., Łukasiewicz D. (eds.)
Brentano and Polish Philosophy, pp. 183–215. Ontos, Frankfurt (2006)
19. Łukasiewicz, E.: Evidentiality and epistemic justification of beliefs. Kazimierz Wielki Univeristy
Press (2018)
20. Reid, T.: In: Brookes, D.R. (ed.) An Inquiry into the Human Mind on the Principles of Common
Sense. Pennsylvania State University Press, University Park (1764/1997)
21. Rojszczak, A.: Wahrheit und Urteilsevidenz bei F.Brentano. Brentano-Studien V, 187–218 (1994)
22. Russell, B.: Introduction to Mathematical Philosophy. George Allen and Unwin, London (1919)
23. Sellars, W.: Empiricism and abstract entities. In: Schilpp, P.A. (ed.) The Philosophy of Rudolph
Carnap, pp. 431–468. Open Court, LaSalle (1963)
24. Spinoza, B.: The Ethics. Princeton University Press, Princeton (1985). Trans. Edwin Curley
25. Woleński, J.: Logic and Philosophy in the Lvov-Warsaw School. Kluwer Academic, Dordrecht (1989)
D. Łukasiewicz ()
Kazimierz Wielki University, Bydgoszcz, Poland
What Is Reasoning?
Robert Kublikowski
Abstract The usage of language and cognition have perhaps been the oldest specific
features of a human mental activity. However, is there anything really exceptional in
the fact that we use language and that we cognise? Is there anything specific in how
we do it? It seems that reasoning is especially important among various linguistic and
cognitive human activities. But what is reasoning? What is human reasoning? Is there
anything specific about it? Which cases of reasoning are the correct ones and why? I take
as my starting point the views on reasoning presented by the logicians and philosophers
belonging to the analytic philosophy of language and epistemology, especially to the
Polish analytic philosophy (The Lvov-Warsaw School: Tadeusz Czeżowski, Kazimierz
Ajdukiewicz, Janina Kotarbińska and Alfred Tarski). In addressing the above questions
I develop an improved view according to which the formal conditions of correctness are
based on formal relationships (a consequence etc.). Such conditions depend on informal
(material) conditions of correctness which are based on informal (material) relationships
(reference, causality, spatial relationships etc.).
Keywords Language · Meaning · Truth · Assertion · Definition · Inference ·

Reasoning · Formal and informal conditions of correctness · Belief revision ·
Justification · Knowledge · Cognition
Mathematics Subject Classification (2000) Primary 03B65; Secondary 03B42
1 Introduction
Each of us belongs to the biological species called Homo Sapiens in terms of our
physical features, especially, in terms of the human nervous system and its main
element—the brain. We are, in this respect, a part of nature understood as the material
world. But we transcend this material dimension with our minds. In other words, the
human mind is something more than a complex of material (physical, chemical or
biological) elements and rules according to which such a complex of elements works.
Our human mind enables us: to feel emotions (feelings), to decide (the will) and to

154 R. Kublikowski
think (the practical and the theoretical reason, the intellect). The intellect is featured
by intelligence and an intellectual intuition. An activation of our mental, rational,
intellectual, intelligent, intuitional abilities (dispositions) enables us to use language and
to cognise.
Cognition may be understood as a process and its result—knowledge. Knowledge is a
true and justified belief. A belief is an asserted sentence with its meaning (sense) which
is a logical judgment (a proposition). Such a judgment can be justified in a direct way—
by perception or indirectly—by reasoning. However, strictly speaking, three types of a
cognitive process (and of knowledge) can be distinguished: the perceptual, the intuitive
and the discursive ones. I prefer to qualify these cognitive processes using the term “types”
rather than “kinds” as this distinction is not a product of classification but of typology. The
boundaries between these cognitive activities (and respectively, their concepts) are vague.
So, let us repeat, the following types can be distinguished:
• Perception is a cognitive process obtained by using senses: sight, hearing etc.
• Intuition is a momentous, quick, “without any steps” act of an intellectual seeing
(apprehending, grasping, understanding) that something has such and such a feature
or something is related to something else.
• A discursive process is “realised in steps”: e.g. an analysis, a classification, reasoning,
a discussion etc.
These cognitive activities are mutually and closely interrelated. Intuition is engaged in a
perceptual act during the process of getting a concept of something which is perceived
(conceptualisation). But intuition is also useful or perhaps necessary in the discursive
process of reasoning:
• to see the relationship between a premise and a conclusion (a premise is a sentence
which is a starting point and a conclusion—an ending point, a goal of reasoning),
• to grasp that a sentence is true or
• to acknowledge (accept) a sentence as a true premise or a true conclusion of reasoning.
Such a pragmatic activity is called assertion.
“A transition from some beliefs to a conclusion counts as inference only if the thinker
takes his conclusion to be supported by the presumed truth of those other beliefs. [. . . ]
It’s enough that we take our premises to be true, that is, judge them to be true.” [6, p.
4]. However, the question arises: “What is it to believe something because one takes
it to be supported by other things one judges to be true? What kind of taking are we
talking about?” [6, p. 6]. The notion of “taking” can be explicated in the following way:
“(Taking Condition): Inferring necessarily involves the thinker taking his premises to
support his conclusion and drawing his conclusion because of that fact. The intuition
behind the Taking Condition is that no causal process counts as inference, unless it
consists in an attempt to arrive at a belief by figuring out what, in some suitably broad
sense, is supported by other things one believes. In the relevant sense, reasoning is
something we do, not just something that happens to us. And it is something we do,
not just something that is done by sub-personal bits of us. And it is something that we
do with an aim—that of figuring out what follows or is supported by other things one
believes.” [6, p. 5], [10, 11, 22, 27]. However, what is reasoning? What makes something
reasoning?
What Is Reasoning? 155
2 The Definition of Reasoning
Reasoning is a process or an activity concerning sentences or judgments. The process

leads to a result: new sentences or judgments. Such a result is a logical structure consisting
of judgments which are linked as premises and a conclusion, reasons and a consequence,
a starting point and an ending point (a goal) of reasoning [13, p. 119].
Reasoning can be understood in various ways:
• as complex reasoning (something more complicated than inference) or
• as simple reasoning (inference).
Complex reasoning consists of inference and of another procedure (a method), for
instance, of questioning. A proof is a kind of complex reasoning.
A proof in a logical sense is a complex composed of a logical judgement to be
proved which is linked by a relationship of a consequence with other logical judgements
(premises). The conclusion can be deductively derived from the premises [12, pp. 91, 93].
At the starting point we know a conclusion which is the last element of the structure of
proof. We ask the guiding question: How do we get to know (prove) that the conclusion
is true? After that we try to find the correct premises among judgments we have already
accepted. However, the following question arises: “what is it to draw a conclusion from
a premise because you take the premise to provide support for the conclusion?” [22,
p. 389].
If we want to build a proof, we match a true premise with a judgment to be proved.
In a case when a proof is deductive reasoning, a premise is a reason and a judgment
to be proved (a conclusion) is a consequence. A premise must be connected with a
conclusion by a relationship of a logical consequence. In other words, a conclusion must
follow logically from a premise on the basis of a logical law. Such a proof is justifying
reasoning because we match a correct premise to support a conclusion. Such a proof is
also regressive reasoning because we “move” from a conclusion to a correct premise. So
the direction of finding a correct premise is opposite to the direction of the relationship of
a logical consequence [12, p. 93]. Yet, what is simple reasoning, that is, inference?
“Inferring is a movement of thought between propositions which may, in special
circumstances, result in the thinker coming to judge the proposition inferred to be true.”
[27, p. 28]. Inference is a way of thinking. But “thinking” is a broader concept than
“inference”. It means that every inference is a way of thinking, but not every way of
thinking is inference.
Inference “involves judging a conclusion to be true because one takes the (presumed)
truth of the premise to provide support for that conclusion.” [22, p. 389]. The word
“because” or “‘therefore’ is used to express or report an inference, it does so by virtue of
the contextually salient explanatory relation being precisely the relation that Frege spoke
of in describing inferring as believing a conclusion because one takes it to be justified
by something. It is, in other words, a relation that obtains between some conclusion that
is justified, and something else—a reason—that makes it justified. This is not the kind
of relation that some epistemologists call ‘propositional justification’, which is a relation
between a person and a proposition that the person is justified in believing, but rather a
generalization of the relation that epistemologists call ‘doxastic justification’, which is
156 R. Kublikowski
a relation between a person’s belief and whatever makes that person’s belief justified.”
[22, p. 400]. Thus, some relations can be distinguished:
(a) a relation between something and thinker (a person, a cognitive subject) S who
believes (judges, accepts, takes etc.) that conclusion (a logical judgment, a proposition
etc.) q is true.
This something is supposed to justify (support, warrant, entitle etc.) S believing that q is
true. But what is this something? Let us reformulate (a) to see it:
(b) a relation between premise p presumed to be true (a reason justifying q as true) and
S who believes that q is true because of p.
This is a pragmatic relation, that is, a relation between linguistic expressions and language
user S.
(c) This something can be understood not just as p, but as something which justifies p as
true, something which p refers to, something making p true (a truth-maker).
Such a relation is a semantic relation. But what are the conditions which justify S
believing that p is true? Or when is p true? Let us analyse this situation. At the starting
point of reasoning (a proof) S knows q and S searches p to justify q in a logically and
materially correct way. In other words, S is justified to judge q as true if q is justified
by p. And, thinking in a regressive way, that is, “going back”, S is justified to judge
p as a true belief and a right reason justifying q if S is justified to judge p as a true
belief and a right reason justifying q. And so on. There appears the infinite regress
(or a circularity) in reasoning. However, it is not identical with the infinite regress (or
a circularity) in justification. And perhaps a non-inferential justification is accessible.
One way of non-inferential justification is to claim that some beliefs—basic beliefs—
are justified by intuition (a reflection). Such beliefs must clearly, obviously, surely be
true etc. However, the problem with intuition is that a particular p is clear etc.—and thus
fundamental—for some people, but it is not for others. And it makes the notion of intuition
complicated. (I will analyse this problem in the following sections and I will also propose
a solution to this problem.)
Inference, or to be more specific inferring, is a mental, complex process consisted
of judgments. However, inference is not just an accidental collection of judgments. The
conclusion is accepted as true on the basis of the acceptance of premises. The crucial issue
is the relationship between the acceptance of premises and of a conclusion. Moreover,
weaker or stronger acceptance of premises leads to the acceptance of a conclusion which
has not been accepted yet or has been accepted with a lower degree of certainty. The
degree of acceptance of a conclusion is not higher than the degree of the acceptance
of premises [1, p. 107]. Inference is not just “a few sentences”. So, what is it that makes
inference? Inference consists of sentences. It seems that they are elements of an inferential
structure. And yet, what makes them a complex (a structure)? What unites such sentences?
What is the basis of the structure? Is it a chain (a sequence) of sentences? What is this
chain (this sequence)?
Premises accepted (acknowledged) as true, in other words, asserted premises are the
basis of asserting a conclusion. A sentence—a conclusion—is asserted on the basis
of a relationship which connects asserted sentences playing the role of premises or a
conclusion. But what is this relationship? What relationship justifies the fact that a premise
“sends” (“transmits”) its meaning—a logical judgment—to a conclusion? [21, p. 260].

The candidates for these relationships are formal and informal relationships [16].
Another question is what makes something a correct, good reasoning? Is incorrect, bad
reasoning still reasoning?
3 The Conditions of Correctness of Reasoning
The conditions (rules) allow to judge (evaluate) reasoning in terms of correctness

(rightness) or incorrectness (non-rightness). What makes reasoning right? Which premise
or conclusion is right? “The premise judgments need to have caused the conclusion
judgment ‘in the right way.”’ [6, p. 3]. It is important to emphasise that an act “might
seem right without being right: there is a distinction to be made between seeming right
and being right.” [11, p. 23]. What is the evidence that such a distinction is useful?
“An essential part of the attitude of seeming right is recognizing the possibility of
correction. You recognize that a certain sort of challenge to your act is appropriate
and may succeed. We may call the challenge ‘checking’. When an act seems right
to you, relative to a particular rule, you recognize that it might no longer seem
right to you if you were later to check what you did. This recognition appears as a
disposition. Having the attitude of seeming right involves a disposition to stop having
this attitude in particular circumstances, specifically if you were to check the act and
it were no longer to seem right. This is a counterfactual disposition, since you might
never check. You may not be disposed to check—perhaps because you are confident—
but you still have this counterfactual disposition. Checking may consist simply in
trying again in the same way to follow the rule, or it may involve something more.”
[11, p. 22].
Then let us consider that act of checking. Checking involves conditions (rules) of
evaluating, in our case, evaluating the correctness of reasoning. Some general conditions
of correctness have been distinguished: the formal (logical) and informal (material) ones:
see e.g. [1, pp. 97–181], [12, p. 93].
3.1 The Formal Conditions of Correctness
Formal logic concerns—as the term “formal” suggests—a form of used expressions. It
does not concern rather a content of such expressions. But the opinion that formal logic
totally passes over whatever contents would be slightly exaggerated for even shapes p, q,
r—representing simple sentences—express the content that p is identical with p, p is not
q, p differs from → (the symbol of implication) etc.
Sentences which play the roles of premises or a conclusion are linked by formal
relationships: in such cases a shape, a composition (an order) of expressions and their
logical value (truth, falsity) is taken under consideration. Such formal and inter-sentential
relationships are worked out by a propositional logic and a first-order logic. Sentences
158 R. Kublikowski
consist of names connected by intra-sentential relationships which are worked out by a

syllogistic logic and a first-order logic [21, p. 261].
It is worthy of distinguishing implicit and explicit formal conditions of correctness
of reasoning [20, p. 234]. Implicit conditions concern the correctness of procedures
linked with inference (questions guiding reasoning, definitions of terms used in premises
and a conclusion etc.). Explicit conditions fix the correctness of inference itself (e.g. a
consequence, truth).
Let us present and analyse some formal implicit conditions:
1. A question which guides inference has to be correct, that is:
• meaningful (linguistically well constructed),
• adequate (the question’s assumption has to be true),
• justified (there is a reason to pose such a question),
• decidable (it is possible to obtain an answer in a finite number of steps),
• creative (an answer for such a question delivers a new knowledge).
2. Terms used in premises and a conclusion have to be correctly defined:
• the word defined (definiendum) must not be used in the definiens in an explicit
definition,
• in the case of a lexical definition the connotation (intension) and the denotation
(extension) of the definiendum and the definiens must be identical, i.e. mutually
interchangeable (the condition of adequacy). In other words:
• the extensions of the definiendum and definiens must not be mutually exclusive,
• the extension of the definiens must not overlap with the extension of the definien-
dum,
• the extension of the definiens must not be superior to the extension of the
definiendum (i.e. the definition must not be too broad),
• the extension of the definiens must not be inferior to the extension of the definien-
dum (i.e. the definition must not be too narrow): see e.g. [1, pp. 68–70], [20, pp.
233–234].
There are also some formal explicit conditions of correctness:
3. It is against the logical law of non-contradiction: : ¬(p ∧ ¬p) to accept contradictory
premises.
Therefore, there are a number of formal conditions of correctness of inference. They are
based on many different formal relationships. And a logical consequence is not the only
formal relationship. However, it is the main one.
4. A logical consequence links the premises and a conclusion in the case of an infallible
reasoning: a deduction and a mathematical induction.
The literature concerning a logical consequence is very extensive: see e.g. [2, 14, 15,
23]. Nevertheless, it is not our major topic.
3.2 The Informal (Material) Conditions of Correctness
Not every consequence is a logical one. It is worthy of distinguishing a formal (logical)

and an informal (material) consequence. It seems that formal and informal consequences
are related to informal conditions of correctness of reasoning.
“We need not treat all correct inferences as correct in virtue of their form, supplying
implicit or suppressed premises involving logical vocabulary as needed. Instead, we
can treat inferences such as that from ‘Pittsburgh is to the west of Philadelphia’ to
‘Philadelphia is to the east of Pittsburgh,’ or from ‘It is raining’ to ‘The streets will be wet,’
as materially good inferences—that is, inferences that are good because of the content of
their nonlogical vocabulary.” [9, p. 85], [8, pp. 94–116].
Expressions are connected not only by formal relationships, but also by informal ones.
Such relationships come out when sentences and names are analysed in terms not only of
a logical form, but also of their contents [21, p. 262].
There are some informal conditions of correctness:
1. Sentences playing the roles of premises or a conclusion are presumed to be true.
2. Sentences consist of names: individual or general ones. Individual names (e.g.
“Christopher Columbus”) refer to unique objects. General names refer to classes of
objects: names of artefacts (e.g. “a building”) and natural kind terms (“gold”, “a
lemon”, “a tiger” etc.).
Such terms refer to natural kinds, i.e. kinds in nature: see e.g. [17]. An ostensive definition
is useful for introducing new terms into a language: see e.g. [19].
3. Sentences refer to states of affairs which are linked by many different informal
(material) relationships like causality, spatial relationships etc.
So far the following problems (questions) of the paper have been presented: What is
reasoning? What is human reasoning? Is there anything specific about it? Which cases of
reasoning are the correct ones and why? The definitions of reasoning have been given and
some formal and informal conditions of correctness have been shown. However, the main
questions have still not been answered in a satisfactory way. So, let us reformulate the
questions and consider how logicians can, for instance, know that ((p → q) ∧ p) → q.
In addition to that, let us express this logical rule (principle) called modus ponens (MP
or modus ponendo ponens—MPP) less symbolically as follows: (if p, then q) & p, so q.
And let us analyse the clear physical example of reasoning which fulfills the scheme: if
water is heated up to the temperature of approximately 100 ◦C at a standard atmospheric
pressure, then it boils. In fact, if water is heated up to such a temperature, then it boils.
How can logicians know that this scheme is correct?
It is worthy of noticing that “when you reason by modus ponens, you may do
some reasoning in identifying what is the antecedent and what is the consequent of the
conditional proposition you are reasoning with.” [11, pp. 20–21]. Moreover, if “you infer
q from p and p → q, this will seem right relative to the modus ponens rule, but not
relative to the rule of inferring a tautology. This means you must have a way of identifying
different rules to yourself, which means identifying different dispositions. You must be
able to identify some disposition as the modus ponens one, for instance. You can even
choose which rule to follow. If you choose the modus ponens rule, this explains why you
160 R. Kublikowski
infer q, rather than something else, from p and p → q. Choosing a rules mean choosing
a disposition. This is not as mysterious as it may sound, and it does not imply you can
map out in advance where the disposition will lead in all cases.” [11, p. 22].
Perhaps the machine-like (syntactic) model of inference is an adequate answer to
the question: What is reasoning? It may seem that formal conditions of correctness (a
consequence etc.) are more important than informal (semantic) conditions such as truth,
reference etc. According to such a model:
• if you apply true premises to a correct scheme of a deductive inference or of a
mathematical induction, you will get a true conclusion with an absolute degree of
certainty,
• if you apply true premises to a correct scheme of an enumerative induction, you will
get a true conclusion with a lower or higher degree of certainty which depends on a
quality and a quantity of evidence. Then you will be justified (entitled) to acknowledge
a conclusion as true with a lower or higher degree of certainty.
“It is tempting to think that there are two kinds of inference—deductive and inductive.
But in what could the difference between these two kinds of inference consist? Of course,
in some inferences the premises logically entail the conclusion and in others they merely
make the conclusion more probable than it might otherwise be. That means that there are
two sets of standards that we can apply to any given inference. But that only gives us two
standards that we can apply to an inference, not two different kinds of inference.” [6, p. 5].
Does the machine-like model of inference give adequate answers to the questions:
What is reasoning? What is human reasoning? Is there anything specific about it? Which
cases of reasoning are the correct ones and why? Not quite so.
Thus, let us now notice that “if rule-following is to explain what reasoning is,
eventually you must do some rule-following that does not involve reasoning. Moreover,
as Boghossian shows clearly, it must not depend on a belief (or any intentional attitude)
whose content is that you should act this way. As Boghossian earlier put it, following
Wittgenstein, this sort of rule-following must be done ‘blindly”’ [11, p. 21]. The point is
that the deepest basis of inference must be non-inferential. It is a blind rule-following.
How can such a sort of rule-following be blind? In what sense? Is the deepest basis of
reasoning blind, that is, accidental? It seems that it is not. Thus how can we explain
that an inference, for instance, according to modus ponens principle, is not just a matter
of good luck? What is the answer to the question: “how could MPP premises warrant
MPP conclusions while being blind? Answer: they do, because they are written into the
possession conditions for the conditional, and the conditional is a non-defective concept.
[. . . ] If we are to make sense of the justified employment of our basic logical methods of
inference, we must make sense of [. . . ] blind but blameless reasoning—a way of moving
between thoughts that is justified even in the absence of any reflectively appreciable
support for it.” [4, p. 248], [3, 5, 7, 18, 24–26].
To overcome an infinite regress (or a circularity) in reasoning, some conditions in logic
have been fixed to be met:
It is important to accept—as a formal foundation of the method of reasoning—a set of
rules (modus ponens, modus tollens etc.) and a set of axioms which are just assumed as
clearly, obviously, surely true etc. Such a set of axioms should be:
• independent (the axioms—sentences should not be mutually provable),
• non-contradictory (the axioms and their consequences should not be contradictory),

• complete (every correct sentence or its negation can be proved on the basis of the
axioms),
• decidable (there is a method of proving in a finite number of steps whether or not a
given sentence belongs to the system) etc.
If you do not assume anything, then you will not prove anything.
It is important also to accept—as an informal (material, empirical) foundation of
reasoning—the usefulness of an ostensive definition. If you do not assume it, then you
will not get empirical terms, including natural kind terms and empirical sentences which
are supposed to deliver an empirical content about some reality.
Perhaps the main claim of this article is also a good candidate for an explanation
of why or how we can reason. Namely, it seems that the formal (logical) conditions of
correctness of reasoning are based on formal (logical) relationships and these conditions
depend on informal (material) conditions of correctness based on informal (material)
relationships. The formal relationships are not sufficient for a correct inference and they
need to be supported by informal relationships. In other words, “the notion of formally
valid inferences is definable in a natural way from the notion of materially correct ones.
[. . . ] the notion of logically good inferences is explained in terms of a prior notion of
materially good ones.” [9, pp. 85–86], [8, pp. 104–105].
The formal relationship of the consequence between a sentence which is an antecedent
(reason) p and a sentence (the consequent) q in inference seems to depend on the
truthfulness of p and q. And the truthfulness of p and q is determined by the relationships
of references of p and q to the relevant states of affairs.
Let us analyse the above claims using the example of freezing water. An enumerative
induction is used in the context of discovery of regularities concerning such water.
Namely, if there is any water at a temperature of about 0 degree Celsius at a standard
atmospheric pressure, then the water freezes. It is a complex sentence consisting of
two sentences p and q linked by a logical connective of an implication expressing the
relationship of a consequence. The implication is true if p and q are true and they are
true if p and q—on the one hand—express the given logical judgments and—on the other
hand—p refers to the given temperature and atmosphere and q refers to the water which
freezes. Water in fact freezes at such temperature and such atmospheric pressure. So,
the basis of the whole case is the informal (material) relationship of causality between
the states of affairs: the temperature, atmosphere and freezing water. The implication of
sentences p and q depends on the (causal) laws of nature.
In the context of justification—however—the empirical claims (conclusions) are
supported by deductive inference (precisely speaking, hypothetico—deductive inference).
The problem is that deductive inference is infallible if the conditions of its correctness are
fulfilled, especially, if premises are true.
But in difficult cognitive cases, that is, if one cannot recognise (know) in a satisfactory,
certain way whether empirical premises are true, then one just believes the premises to
be true with a lower or higher degree of certainty depending on the quality and quantity
of evidence [21, p. 263]. Difficult cognitive situations “disrupt” infallibility of such a
deductive “machinery”. This is why our minds are able to cognise and express the laws
of nature just in a revisable, fallible (tentative) and changeable way.
162 R. Kublikowski
4 Conclusions
Reasoning is a mental, intellectual activity or its result. At the starting point there are
sentences and their meanings (logical judgments) or sentences accepted as true (beliefs).
Such sentences play the role of premises or a conclusion. Reasoning is a transition from
premises to a conclusion on the basis of formal and informal relationships.
It seems that a good candidate (hypothesis) for an adequate explanation of the human
capacity to cognise (to reason) is the compatibility of rational (intelligible) nature and
of us—humans. Namely, our mental “processes dovetail with the causal structure of the
world [. . . ] our processes of belief acquisition are indeed well adapted to providing us
with an accurate picture of the world” [17, p. 3]. Our human knowledge of the world is
possible on the basis of the assumed hypothetical fit between our mental cognitive abilities
and the reality, especially the assumed causal structure of the reality which, among others,
consists of natural kinds of objects.
This compatibility makes it possible to activate our mental, rational, intellectual,
intelligent, intuitive abilities to use language and to cognise (to reason). We are able
to reason in a rational way about the material world because—to a certain extent—we
are a part of the material world. It suggests the existence of a subtle mental, linguistic and
cognitive tuning, by analogy with a discovered subtle cosmological and biological tuning.
The explanation of why or how we can reason is a matter of a natural regularity and—to
a certain extent—naturally based ability to reason.
However, we humans transcend the material world with our minds as we have such
features and functions which do not exist in nature. Such a fundamental feature of
the human language and cognition is normativity. The conditions and evaluation of the
correctness of reasoning: correct (right, valid, sound, good, proper) or incorrect (non-
right, invalid, unsound, bad, improper) belong to normative vocabulary and are the
evidence of a specifically human normative aspect of reasoning.
References
1. Ajdukiewicz, K.: Pragmatic Logic. PWN – Polish Scientific Publishers, Warszawa and D. Reidel
Publishing Company, Dordrecht – Boston (1974) (Originally published in Polish as: Logika prag-
matyczna. Państwowe Wydawnictwo Naukowe, Warszawa (1965))
2. Blanchette, P.: Logical consequence. In: Goble, L. (ed.) Blackwell Guide to Philosophical Logic, pp.
115–135. Blackwell Publishers, Oxford (2001)
3. Boghossian, P.: Inference and insight. Philos. Phenomenol. Res. 63(3), 633–640 (2001)
4. Boghossian, P.: Blind reasoning. Proc. Aristot. Soc. Suppl. Vol. 77, 225–248 (2003)
5. Boghossian, P.: Inferentialism and the epistemology of logic: reflections on Casalegno and
Williamson. Dialectica 66(2), 221–236 (2012)
6. Boghossian, P.: What is inference? Philos. Stud. 169, 1–18 (2014)
7. Boghossian, P.: Reasoning and reflection: a reply to Kornblith. Analysis 76(1), 41–54 (2016)
8. Brandom, B.R.: Making It Explicit. Reasoning, Representing and Discursive Commitment. Harvard
University Press, Cambridge, MA, London (1998)
9. Brandom, B.R.: Articulating Reasons. An Introduction to Inferentialism. Harvard University Press,
Cambridge, MA, London (2001)
10. Broome, J.: Rationality Through Reasoning. Wiley-Blackwell, Chichester (2013)
11. Broome, J.: Comments on Boghossian. Philos. Stud. 169, 19–25 (2014)
12. Czeżowski, T.: Proof. In: Czeżowski, T., Knowledge, Science, and Values. A Program for Scientific
Philosophy, pp. 91–95. Rodopi, Amsterdam, Atlanta, GA (2000). Ed. L. Gumański (Originally
published in Polish as: Dowód. In: Czeżowski, T., Filozofia na rozdrożu (Analizy metodologiczne),
pp. 97–105. Państwowe Wydawnictwo Naukowe, Warszawa (1965))
13. Czeżowski, T.: The classification of reasonings and its consequences in the theory of science. In:
Czeżowski, T., Knowledge, Science, and Values. A Program for Scientific Philosophy, pp. 119–
133. Rodopi, Amsterdam, Atlanta, GA (2000). Ed. L. Gumański (Originally published in Polish as:
Klasyfikacja rozumowań i jej konsekwencje w teorii nauki. In: Czeżowski, T., Filozofia na rozdrożu
(Analizy metodologiczne), pp. 161–184. Państwowe Wydawnictwo Naukowe, Warszawa (1965))
14. Etchemendy, J.: The Concept of Logical Consequence. Harvard University Press, Cambridge, MA
(1990)
15. Haack, S.: Philosophy of Logics. Cambridge University Press, Cambridge (1978)
16. Kamiński, S.: On the classification of reasonings. In: Kamiński, S., Method and Language. Studies
in Semiotics and the Methodology of Sciences (in Polish), pp. 231–245. The Learned Society of the
Catholic University of Lublin, Lublin (1994). Ed. U. Żegleń
17. Kornblith, H.: Inductive Inference and Its Natural Ground. An Essay in Naturalistic Epistemology.
Bradford Book/MIT, Cambridge, MA, London (1995)
18. Kornblith, H.: On Reflection. Oxford University Press, Oxford (2012)
19. Kotarbińska, J.: On ostensive definition. Philos. Sci. 27(1), 1–22 (1960) (Originally published
in Polish as: Tak zwana definicja deiktyczna. In: Fragmenty filozoficzne II. Ksi˛ega pamiatkowa˛
ku uczczeniu czterdziestolecia pracy nauczycielskiej w Uniwersytecie Warszawskim profesora
Tadeusza Kotarbińskiego, pp. 44–74. Państwowe Wydawnictwo Naukowe, Warszawa (1959)). Eds. J.
Kotarbińska, M. Ossowska, J. Pelc, M. Przeł˛ecki, K. Szaniawski
20. Kublikowski, R.: Definition within the structure of argumentation. Stud. Log. Gramm. Rhetor. 16(29),
229–244 (2009)
21. Kublikowski, R.: Meaning, reasoning and normativity in the context of meaning inferentialism of
Robert B. Brandom (in Polish). Przeglad ˛ Filozoficzny 2(98), 259–269 (2016)
22. Neta, R.: What is an inference? Philosophical issues. Epistemic Agency 23, 388–407 (2013)
23. Tarski, A.: On the concept of logical consequence. In: Tarski, A., Logic, Semantics, Metamathematics.
Papers from 1923 to 1938, pp. 409–420. Hackett Publishing, Indianapolis, IN (1983). Ed. J. Corcoran
(Originally published in Polish as: O poj˛eciu wynikania logicznego. Przeglad˛ Filozoficzny 39, 58–68
(1936))
24. Williamson, T.: Understanding and inference. Proc. Aristot. Soc. Suppl. Vol. 77, 249–293 (2003)
25. Williamson, T.: Reply to Boghossian. Philos. Phenomenol. Res. 82(2), 498–506 (2011)
26. Williamson, T.: Boghossian and Casalegno on understanding and inference. Dialectica 66(2), 237–
247 (2012)
27. Wright, C.: Comment on Paul Boghossian, “What is inference”. Philos. Stud. 169, 27–37 (2014)
R. Kublikowski ()
The John Paul II Catholic University of Lublin, Faculty of Philosophy, Lublin, Poland
Tadeusz Kotarbiński: Socrates of Warsaw
Jacek Jadacki
The legacy of our millennial past includes treasures mixed with rubble.
Our and your Chopin (Kotarbiński [10, p. 391])
Abstract Tadeusz Kotarbiński was one of the main representatives of the Lvov-Warsaw
School. The paper presents his life and didactic activity, as well as the list of his main
works and the greatest achievements in philosophy. Kotarbiński supported concretism,
radical realism, and the directive of semantic reism. He created praxiology—the theory of
effective action. He advocated the system of ethics independent of particular worldview
(religious worldview including). Kotarbiński influenced, e.g., Stanisław Leśniewski’s
ideas of ontology (scil. his original logic of names) and of mereology (scil. his original
theory of collectives).
Keywords Concretism · Tadeusz Kotarbiński · Lvov-Warsaw School · Praxiology ·

Semantic reism
Mathematics Subject Classification (2000) Primary 01A70, 03A05; Secondary 00A99
1 Life
Tadeusz Kotarbiński was born on March 31, 1886 in Warsaw; he died there on October 3,
1981. In the Kotarbiński family there were many artists: a grandfather of Tadeusz’s uncle,
Wilhelm, his father, Miłosz, and his brother, Mieczysław—were painters; his uncle, Józef,
was an outstanding actor; his mother, Ewa née Koskowska, practiced music.
Even in his school years, in Warsaw, Kotarbiński listened to lectures on philosophy—
including the lectures of Stanisław Brzozowski and Adam Mahrburg. In the years 1907–
1912, he studied at the University of Lvov with Kazimierz Twardowski (philosophy)—
and with Twardowski’s most senior disciples: Jan Łukasiewicz (logic) and Władysław
Witwicki (psychology), as well as with Stanisław Witkowski (classical philology). After
receiving his doctorate in 1912—on the basis of the dissertation Utilitarianism in the
ethics of Mill and Spencer [1]—he was a teacher of Latin and Greek in Mikołaj Rej
Warsaw Grammar School.

166 J. Jadacki
After Poland regained independence in 1918 he was appointed lecturer in philosophy at

the University of Warsaw; one year later he became an associate professor, and in 1929—
a full professor at this university. He taught there until his retirement in 1961—including
from 1939–1944, when the university had to go underground, and in the first years of the
communist regime, which, in 1951, deprived him of the chair of philosophy, leaving him
only the less “ideological” chair of logic.
In the years 1945–1949, as the rector, he led the organization of the University of Łódź.
Throughout all his life, he took an active part in the philosophical world: in 1915,
he belonged to the founding members of the Institute of Philosophy; then he was the
president of the Warsaw Philosophical Society (1917–1939) and the Polish Philosophical
Society (until 1977). He was a member of the Warsaw Scientific Society (1921–1951) and
the Polish Academy of Sciences (1946–1951) until their termination by the communists.
He was also a member of the Polish Academy of Sciences since its inception (1952). He
received many awards, attained many honors (he was a doctor honoris causa of seven
universities, a foreign member of five academies, and many scientific societies), was
much respected (e.g. in the years 1957–1968 he was a president of the Polish Academy
of Sciences, and in the years 1960–1963—a chairman of the Institut International
de Philosophie)—maybe the most in the postwar period, when the authorities sought
to capitalize on his former left-wing sympathies and atheistic and materialistic (but
not “dialectical”) worldview. Kotarbiński was distinguished in different ways by the
authorities, but this did not go hand in hand with his real impact on the main direction
of scientific life in the country; however his great intellect and personal integrity and the
reliability of the scientific work positively affected the team of disposable “apparatchiks”
who had implanted in Poland the “only right” philosophy from the East.
2 Writings
The first book of Kotarbiński was Practical sketches [2], in which the main ideas
of praxiology appeared. His Elements of the theory of knowledge, formal logic and
methodology of sciences [3] was a foundation of education for several generations of
Polish intelligentsia. The textbook Logic for lawyers [4] was studied by many of future
Polish lawyers. Lectures on the history of logic [7] and Logic in Poland [9] constituted an
important contribution to the history of logic in general. Treaty on good work [5] initiated
praxiology as an independent scientific discipline. Meditations on fair life [11] gave a
summary of his ethical views.
In 1990, the first volume of his Collected Works [13] was published. The number of
Kotarbiński’s works is close to half a thousand.
3 Views
3.1 Semiotics and Methodology
In logic, he dealt creatively mainly with semiotics and methodology.

Tadeusz Kotarbiński: Socrates of Warsaw 167
The central problem of semiotics—the problem of sense—he solved in such a way,

that he recognized as making sense all and only sentences not containing onomatoids, or
those translatable to such sentences. In the first case, he spoke of basic (literal) sense, and
in the second—of a sense of briefly-substituted sense. Onomatoids—or apparent names—
he opposed to genuine names, i.e. names of concretes, or precisely speaking—of things,
and ultimately—bodies.
He was a supporter of semantic reism, i.e. of the directive to use only language whose
sentences are meaningful in the sense explained above. He believed that, in philosophy
in particular, using such language—reistic language—protects us against idle disputes,
especially ontological ones, consisting e.g. in hypostazing or treating sentences avoiding
(reistic) sense or used only in the brief-substitute sense—as sentences saying something
about existing objects. Problems, which are objects of such disputes, turn out to be badly
formulated.
He accepted the classical conception of truth: a given sentence is true, when it is just
as it says. He rejected the nihilistic idea, hence the idea that sentences predicting the
truthfulness of a given sentence are synonymous with this sentence itself. He believed
that this is so only if the word “true” is used verbally, i.e. if the sentence with truthfulness
assigned to it (e.g. the sentence claiming that a is P), is indicated by putting it in quotation
marks (and therefore by using the expression “a is P”). If, however, the word “truth” is
really used, i.e. if the sentence with truthfulness assigned to it is indicated in another way
(e.g. by using the expression “the first sentence of this paragraph”), the “truthfulness” can
no longer be thus eliminated.
He initially combined the classical conception of truth with logical indeterminism, i.e.
with rejecting the principle of the excluded middle, if it would have to proclaim that every
statement is either true or false. Although all truths are perpetual, not all of them—are
eternal. Truth is perpetual, i.e. if a sentence is now true, it will remain true forever. But
there are sentences that are not eternally true, and so being true now, were not always true
before. Not being also false, they were undefined in the past. If a sentence is undefined,
then it is possible that this sentence or its negation is true. If it is defined, then it is true or
false. In turn, if it is true, then it is necessary, and if it is false, it is impossible.
Of the methodological issues—he dealt with, among others, analysis of the reasoning
process, which eventually was considered by him as “justifying thinking”. Among
reasoning, he distinguished—on the one hand—deductive reasoning (i.e. inferring and
proving), in which logical consequences are (fully) justified by accepted reason, and
reductive reasoning (i.e. testing and explaining), in which logical reasons are justified
(only in part) by accepted consequences. On the other hand—he divided reasoning into
progressive reasonings (i.e. inferring and testing), i.e. those in which consequences of
given reasons are sought, and regressive reasonings (i.e. explaining and proving), i.e. those
in which reasons of given consequences are sought.
He indicated a new base of classification of the sciences into the theoretical and the
practical. He thought, namely, that both theoretical and practical sciences aim to establish
true and justified sentences about the objects of a certain domain, in particular, to establish
laws of this domain. Generally, he advocated for criticism, i.e. the postulate of recognizing
only sentences justified due to our knowledge. He distinguished practical and theoretical
sciences on the ground that practical sciences fixed some truth in order to provide a
description of the procedure leading to the formation of certain things, and that in these
sciences, handling actions (e.g. experiments) outweigh intellectual actions.
168 J. Jadacki
Although Kotarbiński was certainly one of the most eminent Polish philosophers of the
twentieth century, he initially proposed to eliminate the word “philosophy” because it does
not refer to any well-separated areas of knowledge. Not convincing anyone to abandon
this word, he began to introduce at least some conceptual order here. He considered
four concepts of philosophy to be the most important ones: practicing philosophy can
be identified either with creating a metaphysical worldview, or searching for practical
self-knowledge, or indicating an ethical pattern, or building a theory of knowledge. The
first understanding of philosophy coincided with what he called the “great philosophy”,
and more specifically to creating large systems; and the last one—with what he called
the “little philosophy”, i.e. with philosophical analysis. He himself basically practiced
philosophical analysis, but he thought that philosophy does not end on analysis: after
conducting analysis, it comes time to create a system.
3.2 Epistemology and Ontology
In the theory of knowledge, Kotarbiński claimed presentationism, called by him “radical

realism”. The world is directly knowable: objects of knowledge are given to us directly
and not through sensual data (contents). He combined presentationism with objectivism:
the world exists independently of whether it is or is not known.
He was an empiricist—considering experience to be the ultimate base of knowledge.
His empiricism was of an extraspectionistic character. The basis of knowledge is external
experience: sentences about introspection are reducible to sentences about extraspection.
He proposed that sentences of the type “A experiences that p” be paraphrased using
literally reistic sentences of the type “A experiences like this: p”. These sentences are
literally reistic as the onomatoid “that p” disappears from them: ‘A’ is a genuine name of
an “experiencing body”, and ‘p’ is a description of how—not what—A experiences (e.g.
seeing something). An introspective sentence can provide, among other things, knowledge
about how other people experience—not me alone. And a sentence claiming that someone
different from us experiences in such-and-such way, can by understood by us only as an
abbreviation of the sentence “A experiences as I do, when I would experience like this:
p”. This is therefore the imitationistic conception of introspection. Experiencing here is
made independent of the existence of the experienced object: we do not have to accept
the existence of that p, or some sensory data, in order to experience like someone who
experiences in the manner described by the phrase ‘p’.
In theory of objects, Kotarbiński advocated a certain version of nominalism, namely—
concretism. According to nominalism, there is exactly one ontological category: the
category of individuals. Concretism rolls this out: these individuals are identical with
concretes. In other words: every object is a thing. Yet another way: there are only
concretes—there are no abstracts. In particular, there are no general objects, i.e. no
universals. He argued as follows. Something that would be a universal object (e.g. a
man in general) with respect to some individual objects (here: to men) should have all
and only properties common to these objects. Suppose that a certain property (e.g. being
the author of The Spirit King) is the specificity of one of these objects (here: Juliusz
Słowacki). According to the (ontological) law of the excluded middle, our universal
has this specificity or it does not. It can not, however, have this specificity, because the
specificity is not common property of individual objects that fall under this universal. It
can not also not have this specificity, because then the universal would have the property
of not having this specificity, and this again is not a common property of these objects,
because does not reside in this specified object.
Kotarbiński’s concretism had a reistic interpretation: every concrete is a thing. The
reistic thesis—that there are only things—has become one of the most famous of his
views. Ontological reism, in turn, had in Kotarbiński a form of somatism: every thing is
a body. So there are only bodies. They are objects being spatio-temporally extended and
inert. If we agree, in turn, that the totality of bodies—the collection of all of them—is
matter, then somatism is a kind of materialism, because the totality of bodies is also a
body.
Initially, materialism was combined with indeterminism, or, strictly speaking, with
antipredeterminism. He was an opponent of the thesis that what took place at a certain
time was strictly determined by what had happened earlier. It seemed to him that if this
were so, as predeterminists proclaim, it would be impossible to act really creatively. Later
he departed from this view and concluded that—according to determinism—each event is
determined by a set of earlier facts, making together a sufficient condition of this event.
He identified the cause of the event with an important component of this condition—i.e.
such a component, without which the other components of the team no longer constituted
a sufficient condition of a given event.
3.3 Praxiology and Ethics
It is surprising that Kotarbiński, being a reist, devoted so much time to the analysis of
action, i.e. to something that did not exist from the point of view of the reistic doctrine,
not being a thing. In addition, he analyzed actions by no means in reistic terms. What
is more, praxiology—the theory of effective action—is usually tied with the name of
Kotarbiński as with reism, and he is rightly regarded as one of main creators of it.
What underlies praxiology is practical realism, i.e. the postulate of sanity in action:
reckoning with reality when taking any action—with what was, what is, and what can be.
The most important component of praxiology—as the theory of effective action—is a
set of practical directives, i.e. tangible normative sentences of the form: “To achieve this-
and-this in such-and-such time, it is good to do that-and-that”. The phrase “it is good”
is understood in these directives either as a sufficient condition (which is a synonym for
“it is enough”), or as a necessary condition (which is synonymous with “it should be”).
And to do something—is the same as—to take a sequence of simple actions, involving
deliberate and free (i.e. having its source in free impulse) pressure applied to something,
causing the creation of something.
Assume that the practical directive specifies that a condition for achieving a given
thing in certain circumstances is to do some other thing. The theoretical basis for such
a directive is the belief of an overlapping causal link between the occurrence of both of
these things. Causing an intended thing requires using specific materials and tools, that
is—in general—a specific technical base. On the other hand, doing a thing, which is a
condition of causing an intended thing, is a string of deliberate (intentional) actions.
170 J. Jadacki
Praxiology provides a theory of the organization of such activities. They are assessed
in terms of effectiveness—possibly of varying degrees. An action is effective when it
leads to the intended effect; an action blocking this effect is a counter-effective one. Other
activities are—from this point of view—ineffective (or in-counter-effective).
A praxiological estimation of actions applies also to their rationality. An action is
factually rational when it has a true theoretical basis; it is methodologically rational, when
its theoretical base is sufficiently justified.
The efficiency of actions and the rationality of both types are—as we can see—
mutually independent.
We should not mix norms (e.g. “Do not do this-and-this!”) with normative sentences.
Normative statements, regardless of whether they are of the type which has been already
mentioned—that is factual—or emotional (of the type “Doing this-and-this is a good/bad
thing”), are declarative sentences, so they assessed in terms of truthfulness and validity.
While norm—assessed in terms of validity—are devoid of logical value. We do not ask
whether they are true or false, but only whether they are or are not valid.
Kotarbiński advocated (meta)ethical absolutism: in spite of the volatility of norms,
there are absolute moral criteria which make it possible to decide which norms are
valid and which—are not. Ethics is independent of a particular worldview—including
a religious worldview. Kotarbiński himself was an atheist, but he distinguished atheism,
i.e. the refutation of the existence of God, from blasphemy, i.e. offending the feelings of
believers—and certainly he was not a blasphemer.
Justification of moral criteria is provided by elementary moral intuitions, which have
their source in conscience, assessing whether human behavior is “venerable” (i.e. morally
good), or—“shameful” (morally wrong).
Kotarbiński’s chief ethical norm was: Behave as a brave, trustworthy (or responsible)
guardian behaves! Be courageous, sacrificial, right and cornered in the fight against the
existing evil and prevent evil greater than the existing one from coming to be! This norm
also had its freer version—in the form of four recommendations: (1) like doing something,
(2) love someone, (3) dont’t be a scondrel, and (4) live seriously.
4 Resonance
4.1 Origin
The roots of semantic reism lie in what Kotarbiński heard from the mouth of one of his
teachers—Witwicki—and what he himself reported:
[Witwicki] called for the things themselves, demanding that all abstract ideas should be supported
by concretes, any generalization should be illustrated with examples and that these examples
should be expressiveliving. [6, p. 81]
A program closer to semantic reism was proclaimed at the time in the Vienna Circle:
this was the program of physicalism—especially in the version of Rudolf Carnap.
Physicalists conditioned the meaningfulness of sentences on their translatability to a
certain distinguished language: the language of things (just like Carnap) or the language of
impressions (as with some others neopositivists). There was here, however, an important
difference: from the point of view of neopositivism, the thesis of ontological resim must
be considered as lacking empirical sense, and so, briefly speaking, as senseless.
In the classification of reasoning, he initially followed Łukasiewicz, but ultimately he
accepted understanding of “deduction” and “reduction” offered by Rev. Jan Salamucha.
The direct source of ontological views was logical: namely the calculus of names
developed by Leśniewski and called by him “ontology.”
The function of the impetus for interest in praxiology was probably fulfilled by
Brzozowski’s philosophy of working, as well as the analysis of acts and products carried
out by Twardowski; without a doubt also, the organizing and pedagogical activity of
the latter—activity of rare regularity—became for Kotarbiński a practical verification of
his conception of efficient activity. Anticipations of independent ethics can be found in
Władysław Biegański.
4.2 Criticism
At first, Kotarbiński’s logical indeterminism became an object of criticism. Leśniewski

convinced him that relativizing the truthfulness to time is inadmissible. The sentences of
the form “The sentences “Something is such-and-such” is true in such-and-such time”
are meaningful when they can be translated into sentences of the form “The sentence
“Something is such-and-such in such-and-such time” is true”.
However, destructive criticism touched, above all, Kotarbiński’s ontological views.
After the Second World War, this came from the side of communist doctrinaires, who
came to the (legitimate!) conclusion that reism was not identical with their ideology. But
many years before, reism had undergone substantive criticism.
Firstly, it was indicated that reism had significant internal drawbacks.
Thus, Kotarbiński’s arguments against abstracts was proved inconclusive. Marian
Borowski and Roman Ingarden pointed out that Kotarbiński based the argument against
general objects on the mistaken assumption that universals have the same properties (or
properties of the same order) as properties which are possessed by individual object falling
under these universals.
And Kazimierz Ajdukiewicz showed that reism is an analytical thesis. It can be
expressed namely by the statement “Each object is a thing”, and this statement is
meaningful in reistic language, when the name “object” is a genuine name, that is—is a
name of things. Ajdukiewicz noted also that reists, in their own language, can not express
the belief in the nonexistence of abstracts, because the sentence of the form “There are no
abstracts” can not be considered as a meaningful sentence, even as a sentence in a briefly-
substituted sense. The case was put on a knife edge by Stanisław Ignacy Witkiewicz,
pointing out that since reism is a convention, it might as well take the view that, e.g., there
are only properties or—refuted by reists—contents (sense data).
Secondly, it turned out that reism is not an adequate ontology due to mathematical and
physical theories. One of the consequences of adopting the reistic thesis is recognizing
as “objectless” a large part of mathematics—including the part of set theory based on
the concept of an infinite set. This is because the last concept is in fact a lack of reistic
interpretation: no statement of those sets can be translated into an equivalent statement
about things.
172 J. Jadacki
Klemens Szaniawski and Zdzisław Augustynek raised further the objection that
reism—and in any case somatism—comes into conflict with modern physics, which
considers not only bodies (corpuscles) real, but also fields (waves). Meanwhile, fields—
from the point of view of reism—are abstracts, so they are hypostases.
Already Ajdukiewicz’s and Ingarden’s criticism caused Kotarbiński to abandon onto-
logical reism for semantic reism—in its liberal version, i.e. not refusing the meaningful-
ness of unrealistic sentences, but recommending only “as far as possible” avoiding such
sentences.
4.3 Continuation
The formulation of the classical conception of truth, made by Kotarbiński, was a reference
point of the semantic definition of truth given by Alfred Tarski [1933]. Considerations
on absolutism in regard to truthfulness contributed indirectly to Łukasiewicz discovering
three-valued logic: he attributed the third value—beyond truthfulness and falsity—just to
indeterminate sentences.
Marian Przeł˛ecki came to the conclusion that semantic reism can be defended, pro-
vided we find it to be a terminological convention concerning the term “meaningfulness”
or a criterion of a minimum ontological commitment of theories.
An interesting attempt to rescue semantic reism was taken by Bogusław Wolniewicz.
He suggested that reistic briefly-surrogate senses can be assigned not only to sentences
directly translatable to reistic sentences with literal meaning, but also to sentences
belonging to any theory for which it is possible to give a set of relevant axioms in reistic
language.
Psychologizing imitationism, i.e. supporting the reduction of introspection to ex-
traspection in structural psychology, was postulated by Zbigniew Jordan.
Modification of the original version of ontological reism was made by Janina Kotar-
bińska. According to her, two meanings of the word “to exist” should be distinguished:
the primary and the basic. In the primary meaning—only objects designated by names
(apparent names included) exist; in the basic meaning—only things exist. As a result of
this distinction, acceptance of the view that every sentence of the form “This-and-this
object is that-and-that” implies the existence of this-and-this object but does not force the
acceptance of the view that this object exists in the basic meaning of the word “to exist”
(i.e. as a thing).
On the other hand, Czesław Lejewski tried to defend the original thesis of ontological
reism against charges of tautologicality—considering that, contrary to Ajdukiewicz, the
sentence “Each object is a thing” is not an analytical sentence, because the name “object”
has a different (poorer) connotation than the name “thing”.
It was also attempted to “put into practice” some ideas of praxiology—the slogan
of good work, in particular. If these attempts did not bring the expected results, it was
probably because the attempts fell on the deaf ears of the statist economy, subordinated
entirely to dogmatic ideology.
5 Pupils
As a teacher, Kotarbiński taught, first of all, responsibility for words. As a debater he was
a master of what might be called the “idealizing recapitulation”. He could so interpret
someone else’s statement—even a statement of his opponent—that it became BOTH
clearer and more justified than in its original form.
Kotarbiński was the second great teacher within the Lvov-Warsaw School: in Warsaw
he played the role which was played in Lvov by Twardowski. The list of his pupils starts
with such eminent scholars as Maria and Stanisław Ossowskis—and ends with: Jerzy
Pelc, Tadeusz Pszczołowski, Szaniawski and Przeł˛ecki.
6 Summary
Karol Irzykowski called him “Socrates of Warsaw”—and this name characterizes Ko-
tarbiński’s personality in the most brief way. In logic—his postulate of dehipostazing
scientific language made history. In metaphysics—his reism was the most original, but
only his imitationistic theory of introspection survived in its original version.
In axiology—the slogan of reliable guardianship turned out to be the most catchy.
Tadeusz Kotarbiński’s Works Mentioned in the Paper
1. Utylitaryzm w etyce Milla i Spencera [Utilitarianism in the ethic of Mill and Spencer]. In: Wybór pism
[Selected Writings]. Vol. I. Myśli o działaniu [Thoughts on Acting], pp. 215–330. PWN, Warszawa
(1957) (Original work published 1912)
2. Szkice praktyczne. Zagadnienia z filozofii czynu [Practical Sketches. Problems from the Philosophy
of Act]. In: Wybór pism [Selected Writings]. Vol. I. Myśli o działaniu [Thoughts on Acting], pp. 13–
168. PWN, Warszawa (1957) (Original work published 1913)
3. Kotarbinski, T.: Gnosiology. The Scientific Approach to the Theory of Knowledge. Pergamon, Oxford
(1966). (Original work in Polish published 1929)
4. Kurs logiki dla prawników [Logic for Lawyers]. Gebethner i Wolff, Warszawa (1951)
5. Traktat o dobrej robocie. Ossolineum (1955) (English translation: Praxiology. An introduction to the
sciences of efficient action. PWN – Pergamon Press, Warszawa – Oxford (1965))
6. Nauczyciele sztuki nauczania [Teachers of the Art of Teaching]. In: Studia z zakresu filozofii, etyki
i nauk społecznych [Studies of the Domain of Philosophy, Ethics and Social Sciences], pp. 73–83.
Ossolineum, Wrocław (1970) (Original work published 1956)
7. Wykłady z dziejów logiki [Lectures on History of Logic]. Ossolineum, Łódź (1957) (French
translation: Leçons sur l’histoire de la logique. Presses Universitaires de France, Paris (1964))
8. Wybór pism [Selected Writings]. Vol. I. Myśli o działaniu [Thoughts on Acting]. Vol. II. Myśli o
myśleniu [Thoughts on Thinking]. PWN, Warszawa (1957)
9. La logique en Pologne. Son originalité et les influences étrangères. Angelo Signorelli Editore, Roma
(1959)
10. Nasz i wasz Szopen [Our and Your Chopin]. In: Studia z zakresu filozofii, etyki i nauk społecznych
[Studies of the Domain of Philosophy, Ethics and Social Sciences], pp. 391–395. Ossolineum,
Wrocław (1970) (Original work published 1960)
11. Medytacje o życiu godziwym [Meditation of Fair Life]. Wiedza Powszechna, Warszawa (1966)
174 J. Jadacki
12. Studia z zakresu filozofii, etyki i nauk społecznych [Studies of the Domain of Philosophy, Ethics and
Social Sciences]. Ossolineum, Wrocław (1970)
13. Dzieła wszystkie [Collected Works], vols. I–VI. Ossolineum, Wrocław (1990–2003)
J. Jadacki ()
Agency in a Praxiological Approach
Wojciech W. Gasparski
Abstract The paper presents issues of agency in the approach proposed by Tadeusz
Kotarbiński, the Polish philosopher from the Lvov-Warsaw School, author of the phi-
losophy of practicality with its dominant: praxiology. It also outlines a number of other
approaches to this notion as well as notions similar and related to the notion of an agent.
The conclusion points out that praxiology provides an insight into the reality it studies,
but the use to which this is put depends on the users themselves. That use is indirect more
often than direct, since it requires reflection on the agent’s own practicality and some
meta-skills that should characterize reflective practitioners.
Keywords Act · Action · Agent · Change · Goal · Kotarbiński · Practical situation ·

Praxiology
Mathematics Subject Classification (2000) Primary 00-02
1 Introduction
Agency is one of praxiology’s fundamental notions. It was introduced by Tadeusz

Kotarbiński, the author of the philosophy of practicality (Gasparski [7]). This philoso-
pher’s praxiology triptych consists of three volumes from his complete works, namely
Prakseologia - Cz˛eść pierwsza [Praxiology: Part One] [12, 462 pp.] containing papers
published before World War II and post-war works that featured the expression “dobra
robota” [“good work”], Traktat o dobrej robocie [literally A Treatise on Good Work,
published in English as Praxiology: An Introduction to the Sciences of Efficient Action,
Pergamon Press, New York 1965] [22, 220 pp.] and Prakseologia – Cz˛eść druga
[Praxiology: Part Two] [23, 699 pp.] containing papers published after World War II.
This volume opens with a work called “Abecadło praktyczności” [“ABC of Practicality”]
which in concise form presents the main issues of praxiology; the volume also includes
meta-praxiological works and papers on science studies. The indexes in the three volumes
show that the notions of “agent” and “agency” appear on 91 pages.

176 W. W. Gasparski
The present paper presents issues of agency in Kotarbiński’s approach and further
elaboration by his followers. A number of other approaches to this notion as well as
notions similar and related to the notion of an agent are also outlined.
2 Agency in Tadeusz Kotarbiński’s Approach
2.1 The Beginnings
The term “agent” first appears [13, p. 1] in an essay entitled “Cel czynu a zadanie
wykonawcy” [“An Act’s Goal versus the Doer’s Task”] which Kotarbiński delivered as
a lecture in 1910 in Lvov at a session of the Philosophy Club and which was published in
the volume Szkice praktyczne [Practical Sketches] in 1913 [12, pp. 5–85]. This essay
[14, pp. 6–19], inaugurating the general theory of action that was yet to be called
praxiology, listed its basic notions, among them the “agent, creator or doer” (op. cit., 7).
This notion served to define “action” as follows:
The entirety of work, but exclusively that performed by the doer of the deed, his activities, such as
kneading dough and putting it in the oven; walking to the station, buying a ticket and taking one’s
place in the carriage; manipulating the injection device and a child’s skin — let each such entirety
be called the ‘action’ of a given person (op. cit., 8).
Kotarbiński distinguishes action defined as above from an “act”, which is:

The whole process of forces operating, taken together and being the necessary condition of the
goal [of a given act] — let this be called an ‘act’; hence, for example, the entirety of the baker’s
muscular effort, the effect of high temperature on the dough, the effect of the yeast’s growth power;
the entirety of the efforts of the man going to the station, the work of buying a ticket, getting on
the train, the work of the train moving; the efforts of the doctor to inject the serum and the effect
of different forces of nature in the child’s body after the injection, all taken together (op. cit., 8).
The goal of a given act “is the production and existence of the given loaf of bread,
the given person’s getting to and staying in Kraków, the child’s recovery and continued
good health”, while the goal of an action is “the production, or coming into existence, the
lasting of the formed uncooked loaf placed in the oven in the first case, the sitting down
and staying in the carriage of the person in the second case, the serum getting under the
patient’s skin and staying there in the third” (op. cit., 8). Further in the essay the author
considers the relations between the two goals, namely: sameness, divergence, opposition,
the paradox of acting “in spite of”, concurrence, sequential succession, the paradox of
indirectness of effect and others. As a result, the goal of an action stops being set in
opposition to the goal of an act. Hence:
We separate [. . . ] from the entirety of the act, as our action [i.e. of us as agents — WWG],
part of it on the basis of any of its breakthrough features, meaning those that are materially or
methodologically important, which separate all of the activities, including our own, from the rest of
the activities which are independent of us. The notion of the goal of the action changes accordingly
(op. cit., 12).
An answer is sought to the question: “Where is this directness of our activity [as
agents—WG] exerted on external objects (of course as long as we are understood in
mental terms, as bodies wanting and acting, and not physically as our bodies)?” (op. cit.,
13). After discussing misunderstandings connected with mixing up the two kinds of goal,
Agency in a Praxiological Approach 177
of an act and of an action (superficial imitation, opposition, primacy and subordination,

collaborating counteraction), Kotarbiński proposes: “Let us now assume that there exists
such a direct activity, in external work on matter or in internal, purely mental work [. . . ]
and let us call it ‘absolute action”’ (op. cit., 14).
The final issue is the question: “What is the agent of an act morally responsible for?”
Kotarbiński offers the following supposition for consideration:
Perhaps the solution is found in the fact that we are morally responsible for a component of the
complex of forces we call an act, one which depends wholly on us and is equal to our effort; [we
are responsible] for the component that we sought above in vain, whose existence we assumed and
to which we gave the name ‘absolute action’ (op. cit., 18).
Kotarbiński adds “the shadow of the supposition”, as he writes,

that in each act of a moral nature, this absolute action is always ethical, and only the resultant
can be bad; that therefore there is no blame in human acts but only the advantage of alien forces
over an absolute action, which always constitutes, even in cases of glaring crime, our ethical merit
(ibid.).
2.2 Further Reflections
The above was supplemented with some detailed issues discussed in the next two chapters
of Szkice praktyczne. In the essay “Zagadnienie istnienia przyszłości” [“The Issue of the
Future’s Existence”] Kotarbiński points to the double-edged nature of being able to do
or not to do something. He illustrates this with a rather unpleasant but very instructive
example, especially for lawyers:
It is a certainty that every living being has to die; therefore a killer is not the agent, the author of its
death, he is at best only the agent of the kind of death this is and of its acceleration. But whoever
punishes a killer with death for a death also is not his death’s agent, for the same reason, that’s
that, if there are no other reasons that would justify condemning this form of retaliation [15, p. 70].
In the next essay, “O rozszerzaniu sfery czynu” [“On Expanding the Sphere of the
Act”], we read:
Who knows if the moment of capacity for performing an act is not always different from the
moment of the act itself, just as surely as the moment of that performance is always different
from the moment of its product’s presence; he is surely not in stark disagreement with common
understanding who thinks that before the agent performs the act, he can perform it sooner, and not
always just at the moment of the act [16, p. 73].
2.3 Analytical Definition of Agency
A rather short paper entitled “Poj˛ecie zewn˛etrznej możności działania” [“The Notion of
External Possibility of Activity”], published in 1923 in Przeglad ˛ Filozoficzny (vol. 26,
64–67), is important for the notion of agency. Kotarbiński attached great importance to
the analytical definition of agency provided in this paper, as its being quoted in full in [13]
testifies. The definition is as follows:
178 W. W. Gasparski
Due to impulse I of person S, belonging to moment k, S is the agent of fact D from the later
moment r and D is the work of S always and only when a set of facts from k containing I defines
D and no set of facts from k not containing I defines D [17, p. 104].
This definition is the result of abandoning the indeterminism of “the future’s unpre-
paredness prior to activity” due to the danger of “reduction to a contradiction” [13, p. 2]
and of adopting an assumption that is a “demand of determinism” [17, p. 105]. Further
on in the treatise Kotarbiński analytically defines the negation of agency and the external
possibility of agency, summarizing his thoughts as follows:
The above reasoning serves to show that it is possible (and how) to select definitions of agency
and external possibility of agency, with the help of certain terms, in such a way that they will
contain the common meaning of agency and the external possibility of agency and that a certain
common supposition will be expressed within determinism. This supposition is that if someone
does something, they could have done it and at the same time not have done it, and that abstaining
from an act that one could have performed is also an act [17, p. 107].
Kotarbiński points out the discrepancy between the theoretical approach to agency
and its common understanding, for example the assertion that a person who had the
possibility “to cause what happened later” at the same time “did not have the possibility to
prevent it” (op. cit. 107). This is because the common understanding of the possibility of
taking action assumes the motive of a “lack of obstacles” and the motive of “sufficient
competence”, whereas the presented theoretical approach only takes into account the
former (op. cit., 108). To conclude:
. . . even just the condition of the possibility of agency on account of a free behaviour from moment
k, concerning that very moment k, in combination with the assumption of that behaviour, results
in agency (ibid.).
This condition carries with it conditions concerning any moment, which enables the
conditions for moments g and n to be drawn as theorems.
In the same year1 the theory of action was given the name praxiology2; this was an
initiative that Kotarbiński put forward at the First Polish Philosophy Congress, presenting
Zasady teorii czynu [Principles of the Theory of Action] [18]. In this paper Kotarbiński
outlined the programme of praxiology, mentioning the notion of the agent as one of the
fundamental praxiological terms. As an example of “defining the meaning” of this notion,
he once again defined the relationship of agency, slightly modifying the style of his earlier
definitions. This was the new definition:
. . . person S is the agent of fact D, and this fact is that person’s doing, always and only if at a
moment earlier than the moment of fact D there occurs the fact of a free behaviour of person S,
namely impulse I, such that: (1) a certain set of facts contemporaneous with impulse I, containing
that impulse, determines fact D due to the inherent law of the sequence of events, (2) no set of
facts contemporaneous with impulse I and not containing that impulse determines fact D due to
that law (op. cit., 133).
1 The paper was not published until 1927.

2 Tadeusz Kotarbiński described praxiology as the science of efficient action. Praxiology according to
Ludwig von Mises (1987) is the science of means and not goals of action. Both founders of praxiology
schools—the Polish and the Austrian school—referred to the French initiator of praxiological research,
Alfred Victor Espinas, who focused on analysing the means of achieving goals (realizing intentions). Let
us add that in his sociological theory of action, Florian Znaniecki [28] favoured indicating the intention
of the acting subject rather than the goal of the activity. See also [3].
The author continued further on:

. . . if person S at the moment of his impulse I and due to that impulse has the external capacity to
cause fact D, then: (1) he also has the external capacity to cause the negation of fact D, (2) or he is
the agent of fact D (ibid.).
Two years later Kotarbiński published an essay on the agency relationship, illustrating
his thoughts with examples serving to highlight the qualities “in which one is usually
inclined to see constant traits of an agent” [19, p. 122]. This approach to the problem,
which Kotarbiński preferred to a formalized approach, served to prepare the ground, as
he said, for presenting an analytical definition of the notion of an agent, adjusted “to the
scope of the common understanding of an agent” (op. cit., 128). The definition goes like
this:
John is the agent of a given fact (and that fact is the work of John) means the same as: an
earlier free behaviour of John’s was an essential component of the all-encompassing complex of
contemporaneous facts that creates, in accordance with the laws of nature, the essential condition
of that fact (ibid.).
Compared to the earlier approach, the formalized approached is supplemented with the
notions of a system and the law of sequence of events. Here is the expanded analytical
definition of agency in a formalized version:
With respect to impulse I of person S, belonging to moment k, S is the agent of fact D from the
later moment r, and D is the work of S, always and only if a certain set of facts from k — in system
U that includes S — containing I, determines D due to some inherent law of the sequence of facts,
and if no set of facts from k, in the same system U, but without S, not containing I, determines D
due to such a law [18, p. 128].
The system has to be “sufficiently separate”, meaning one “whose parts are not subject
to the action of forces from outside it” (op. cit., 129). The law of the sequence of events is
meant to “take advantage of the intention contained in the colloquial word ‘must’ (‘. . . that
must have happened after this’)” (ibid.). This does not just mean the laws of nature but
also “laws” established by people as regulations, but only when acting in accordance
with them has become a regularity “of a psychological kind”, i.e. “such a disposition has
developed in people that they regularly follow this regulation in relevant cases, so a certain
regularity of events has developed from this particular structure of people’s preferences”
(op. cit., 129–130).
To supplement his definitions and disseminate them, the cited author published the
entry Czyn [Act] in volume one of an encyclopaedic outline of contemporary knowledge
and culture entitled Świat i Życie (The World and Life; 1933). In it we read that:
An acting person is called different things in different cases: an executor, perpetrator, author,
agent. We have chosen the last of these words as a technical term of the theory of action, i.e.
praxiology. [. . . ] every time one is an intentional agent of something, one is also an unintentional,
i.e. involuntary agent of a great many other events [20, p. 135].
The following year saw the publication of a treatise carrying the same title, Czyn3 [Act],
in which it is stated that “There is no act without an agent. But who do we call an agent?
It is the one who made an intentional effort to some aim. He is the agent of everything
that occurred as a result” [21, p. 141]. The notion of an agent was extended to include a
3 Subsequently published as Czyn (1934).

180 W. W. Gasparski
set of people and to distinguish “the kind of participation in a collective act when a given
participant is an agent of the work accomplished by the act, from participation when he is
only a co-agent but not an agent” (op. cit., 144).
2.4 Acting Subject
Kotarbiński also used the term “acting subject”; probably the first time he did so was in
the essay “O rozszerzaniu sfery czynu” [“On Expanding the Sphere of Acts”] from 1913,
where he indirectly expressed the notion’s meaning.
By the sphere of acts of a given subject I mean the totality of things that he can create, in other
words, those that remain in his power, whose existence at a given moment or whose indecision,
the beginning of nonexistence or indecision, are — as we say — dependent on that subject,
in the sense that the subject can cause an affirmative statement about any of those things to
become true at a given time, or can cause it to remain suspended, in that middle that is allegedly
excluded from logic, that in other cases the subject can cause that affirmative statement to become
false from a given moment, and he can also cause it to remain indecisive, and finally, in some
special cases, the subject can cause an affirmative statement about a given thing to become true
from a given moment, but can also cause it to become false from that moment [16, pp. 73–
74].
Kotarbiński distinguishes between different acting subjects, as follows: (1) an isolated

subject: such a creative individual “whose sphere of acts does not overlap with the spheres
of acts of others, and thus whose sphere of acts throughout the entire isolation time does
not contain objects that would belong to the entirety of things of another’s sphere of
acts in the time from the moment of isolation” (op. cit., 74); (2) a non-isolated subject:
one who “could at a later time have the possibility of action that he or she did not have
earlier” (op. cit., 82), but “the power of creating something is weaker in one who shares
it with others” (op. cit., 83). The term “acting subject” appears in Kotarbiński’s works
in the context of collaboration, positive cooperation (organization), negative cooperation
(fighting), creating culture, practical mistakes.
2.5 The Agent and the Issue of the Cause of an Effect
Chapter two of Traktat o dobrej robocie is devoted to simple acts, the agent and free
impulses [22, pp. 15–21]. Using examples, Kotarbiński explains the notion (law) of the
inherent sequence of events (e.g. death occurs after birth and not simply later) essential
in order to connect a cause with an effect of action, in other words to determine “a causal
link understood in the sense that we will have to consider when defining the notion of the
agent of a given outcome” (op. cit., 16).
Event B is the effect of earlier change A filling moment t, and change A — the cause of event B,
always and only if change A is a significant component of the sufficient condition of event B due
to moment t and due to the inherent rule of the sequence of events (ibid.).
In this, “any component event of that condition without which the system of the other
component events would not be a sufficient condition” (ibid.) is a significant part of the
sufficient condition. Thus formulated, this explains a cause in its ordinary, everyday sense,
Kotarbiński writes (e.g. “snow melting was the cause of flooding”). He also notes that
we usually see multiple causes “in the group of mutually contemporaneous components
of a given sufficient condition for a given effect, and multiple causes of a given effect
belonging to its different sufficient conditions, each of which belongs to a different
moment” (op. cit., 17).
After this groundwork we can move on to defining the relation of agency, which
is the relation between the agent and the work, i.e. the result of activity, meaning an
effect that constitutes a change or a state of affairs (being); a cause is always a change.
Hence:
The agent of a given event is the one whose free impulse is the cause of that event (op. cit., 18).
In this, an impulse is a generalized concept of “pressure” understood literally as

physical or metaphorically as mental, and “free” means intentional in accordance with
the will of the acting subject.
The acting subject is the agent of both what he intended and what he did not intend to
do but did do by mistake.
In the entirety of an event, we are the agent due to the given free impulse, and therefore each such
event is our work, its distinguishing feature being that the free impulse was its cause, even if we did
not bring about the event intentionally or consciously, and even if we were erroneously convinced
at the moment of the impulse that the event would not or could not come about (op. cit., 19).
As a formality, Kotarbiński reiterates that the acting subject “is a flesh-and-blood

living person, wanting one thing or another, moving in one way or another, or making
a mental effort to achieve what they want” (op. cit., 21). A simple action of the subject
thus understood is a single-impulse act (ibid.).
A further chapter of Traktat characterizes the notion and types of compound action. A
compound action is the fulfilment of the condition that
a relation of positive or negative cooperation occurs between its components” and when “two
acts are linked by such a relation always and only if one of them causes, enables, facilitates,
prevents or hinders the other, or when the two have the same impact on a third action. [. . . ] A
set of contemporaneous acts (i.e. such that each shares at least part of a moment with another)
that form part of a compound action shall be called a chord of actions; a set of consecutive
acts (even if parts of their moments overlap) that form a part of a compound action shall be
called a series of actions. A series of chords of actions shall be called a complex of actions (op.
cit., 48).
The notions of positive cooperation (developed in the chapter on the principles of

collaboration) and negative cooperation (developed in the chapter on the theory of
struggle) are discussed in the chapter of Traktat devoted to collective action. Kotarbiński
also discusses the subjectivity of institutions, as follows:
in our praxiological thinking we will sometimes approach institutions as if they were persons,
acting subjects with a special mental and physical structure, even though in our view, no institution
is ever, strictly speaking, an acting subject (op. cit., 73).
In the chapter on mental activity, this idea is presented as follows:

The area of intellectual work is full of reminders of the truth that a team, strictly speaking, is
never a subject but only a functional union of subjects, incapable of replacing the subject in acts
of learning or acts of decision (op. cit., 191).
182 W. W. Gasparski
2.6 Recapitulation
The notions of an agent and agency are also legal terms, so it is no wonder that Kotarbiński
published a recapitulation of the essence of the notion of agency in the law journal
Państwo i Prawo [23, pp. 42–46]. Let us cite some excerpts from this brief treatise to
add point to our report on the history of praxiological analysis of agency in Tadeusz
Kotarbiński’s approach.
To begin with, we have the notion of a free impulse:
If someone moved not involuntarily but voluntarily, we shall say that he performed a free
impulse; we extend this notion to the sphere of inner efforts by believing that a free impulse
was also performed by someone who, for example, curbed the desire to shout out loud or who
accomplished a short-term focusing of the attention to remember a forgotten name. [. . . ] Stating
the freedom of a given impulse does not mean being grounded in indeterminism, which accepts
freedom of will understood as intentional impulses being independent of preceding causes. That
an impulse was free only means that someone made the effort that they wanted to make (op.
cit., 42).
As for the notion of agent:

Whoever performed a free impulse at a given moment became the agent of a specific subsequent
event always and only if the impulse was a necessary component of a given set of circumstances
contemporaneous with it, and that set was a sufficient condition in that moment of the event due
to the law of the causal sequence of events. [. . . ] sufficient conditions are sets of events that
are mutually contemporaneous. An event is always a change of an object. [. . . ] We accept the
deterministic doctrine in strong form. It states that an event has a causal sufficient condition in
every earlier moment (op. cit., 43).
Then, characterizing agency in compound actions, Kotarbiński lists examples of

compound actions performed by the same acting subject—this is an individual compound
action. Next, he discusses a team compound action. He also draws attention to the paradox
of co-agency:
Is it possible [. . . ] that a team is an agent even when there is a lack of agency on the part of all of
its members? (op. cit., 45)
The paradox disappears when the free impulse of the other participants in a team
action was a part of the sufficient conditions of the compound action. This issue is
related to the earlier-discussed question of a team, i.e. two or more acting subjects, as
an agent.
. . . this occurs whenever no system of events from a [given] moment containing a free impulse of
one of those subjects but not containing the free impulses of all the other subjects, is a sufficient
condition of that result, but there exists such a sufficient condition of that result belonging to that
moment which contains the whole set of impulses of all those acting subjects as its necessary
component (op. cit., 46).
Tadeusz Kotarbiński believed that the problem of agency had not been exhausted yet
and thus deserved further consideration. Therefore let us look at some other sources.
3 Other Approaches
3.1 The Agent and His World According to Jakob Meløe
The title of this chapter is a reference to the ideas of Norwegian praxiologist Jakob Meløe
presented in the work The Agent and His World [24]. The notion of “our world” defines
the frame of studies on the agent in his world. Our world is the world of practices—the
totality of operations performed by one or many agents—in which use is made of what is
available, giving it the form of existence. According to Meløe, the basic form of a practical
operation is as follows:
x operates on y, where ‘x’ marks the place of the agent, or the subject of the operation, ‘y’ the
object of the operation, or its target, and where the verb ‘operate’, or ‘operate on’, is a stand-in for
some suitable verb of action [24, p. 15].
Meløe calls the thing towards which an action is directed a “tautologous object”, while
the “tautologous subject” is the one who performs a given operation, i.e. the agent.
To each operation in our world there corresponds a well-defined cut of our world, or a well-ordered
niche within it. Without that niche, the operation does not exist as that operation. The agent’s
necessary knowledge of his own operations, or of what he himself is doing, includes necessary
knowledge of that niche. That is, the agent necessarily has knowledge of the agent’s necessary
world.
The agent’s necessary world is also the smallest intelligible system within which his operations
are intelligible [24, p. 27].
3.2 The Approach of Evandro Agazzi
The notion of an operation as a type of action is also used by Evandro Agazzi, who
states that “man’s actions are always directed by an explicit or implicit confrontation with
an ‘ought”’ [1, p. 107]. An operation is “any human action aimed at the production of
a specific and concrete result (in general, an object)”, while those actions “for which
instead the ideal of perfection concerns the manner of execution”—are achievements
(examples: language, reasoning, dance etc.). These actions are evaluated on the basis of
how they follow the rules of performance. Activities that Agazzi calls pure actions, or
simply actions, are activities evaluated not according to their goal but according to an
ideal. They are considered right or wrong in themselves (op. cit., 108 onwards).
3.3 The Approach of Mario Bunge
Mario Bunge from McGill University in Montreal also defines action by pointing to
rationality and morality as dimensions typical of human change-inducing activity.
The action that one thing or its proxy (the agent) exerts upon another thing (the patient) may be
defined as the difference that the former makes to the history of the latter. [. . . ] In human action
theory the agent is a human being, or an animal or a machine under his control, and the patient may
184 W. W. Gasparski
be any concrete thing, whether human or not, that coexists with the agent at least during part of the
period under consideration. There is interaction if the patient reacts upon the agent, as in the case
of work, play, and conversation. And the action is social if both agent and patient are members of
the same animal species.
Humans are distinguished from other things not for being doers but for being capable of acting
rationally in a morally right or wrong way, i.e. for being able to use knowledge to do good or evil
[2, p. 323].
3.4 The Modern Praxiology Approach
Researchers dealing with praxiology4 propose different factors as the foundation of hu-
man activity. Some, as we remember, point to free impulses [12, p. 140], others—to a lack
of satisfaction with the state in which someone finds themselves [27, p. 27], others still—
to change5 as the primary goal of every transformation, including transformations caused
by humans, i.e. actions [8, p. 18]. What exactly do humans transform? They transform
their practical situations and/or their contexts from unsatisfactory to satisfactory ones.
What is characteristic of humans, according to modern praxiology [4], is that they
act—that each one of us acts—with respect to practical situations of which they—we—
are the subjects. Every such situation is an oikos of its subject, and the set of these is an
ecology (oikos and logos) of practical situations. The practical situation of a given subject
is determined by the facts that the subject distinguishes from among other facts due to that
subject’s values. Values give facts meaning on the basis of which the subject considers
them satisfactory or not. If a practical situation is unsatisfactory to the subject, then the
subject strives to change the facts in such a way as to achieve a satisfactory situation, But
even when the subject considers a situation to be satisfactory, change is still needed. In
this case, it is not the kind of “therapeutic” change described above, but a “prophylactic”
one serving to prevent any disturbance of the satisfactory situation by natural or artificial
(i.e. human-induced) processes. The former type of change applies to the inside of the
practical situation, while the latter type concerns the context of the situation—“the rest of
the world”.
Modern praxiology considers the “existence of action”, i.e. the reality connected with
activity, in terms—so to speak—of the ontology of practical situations. The practical
situation of a subject can be interpreted as a generalization of the concept of personal
space introduced by proxemics—a discipline dealing with individual and social space and
its perception by humans. Hence, on the one hand praxiology would be a generalization
of economics with respect to efficacy-focused behaviours (the “double E” of effectiveness
and efficiency), while on the other being a generalization of proxemics with respect to the
“bubbles” of practical situations in which each and every one of us is immersed [8].
4 Presented here are excerpts, with minor editing, from the author’s earlier works listed in the references
[5–11].
5 Change is also indicated by Mises, who writes that human activity is one of the factors that cause change,
it is an element of cosmic activity and becoming; it cannot be reduced to its causes but must be treated as
an ultimate given and studied as such [27, p. 32].
Humans behave actively because they have to (e.g. breathing) and because they want
to (e.g. driving a car). The former kind is called behaviour while praxiology calls the latter
“activity”, defining it as a free (i.e. compatible with the acting person’s will), conscious
human behaviour directed towards a chosen state of things called a target (making it a
targeted behaviour). Only a small range of relatively simple actions can be performed
by a person—the acting subject—by themselves. These are single-subject actions. All
other actions are multiple-subject actions, i.e. actions in which other people (other acting
subjects) participate.
Effective accomplishment of intended targets requires the use of appropriate means,
and doing this in a way that ensures a surplus of result over cost (economy or efficiency)
is the condition of the efficacy of actions. This applies to both single- and multiple-subject
actions.
Collaboration of people establishes society as the harmonized activity of many subjects
whose actions complement, overlap, support and compete with one another, and also
(consciously or not) impede one another, often in brutal and bloody ways. This last
element means that actions are considered not only with respect to their praxiological
core (subject/subjects and their goal/goals) but also with respect to their context. That
context is defined by social consensus subjects (which in itself is a collective activity)—
social contract subjects—setting down conditions regarding the targets and means of
action which should be met for an activity to win the consent of society. The axiological
character of consent is linked to values, i.e. things society treasures above all else. In this
sense, praxiology is situated between logic (core) and ethics (context).
Metaphorically speaking, human collaboration from the point of view of praxiology
is caused by factors similar to how communicating vessels function: a shortage in one
vessel is supplemented with the surplus from another. In activity, some people’s lesser
dispositional capacity for action is made up for by the greater dispositional capacity of
others. To this is added an external factor: unequal distribution of resources causing the
necessity for resources to flow from places of surplus to places of shortage. It is thanks to
people’s conscious efforts to improve unsatisfactory practical situations and/or maintain
satisfactory situations, where the conglomerate of these situations is uncountable, that
people collaborate. In such collaboration, they achieve the primary targets of organized
actions while also meeting their own targets, for which fulfilling the primary target is a
means and vice versa [9].
4 Conclusion
Praxiology offers an insight into the reality it studies, but the use that is made of this
depends on the users themselves. Such use is more often indirect than direct, since it
requires thinking about one’s own practicality as an agent, who acts the more efficiently
the more of a “reflective practitioner” he is [25, 26]. The meta-skills that reflective
practitioners should have for their actions to be suitably efficient for our times are: the
ability to gain new skills, the ability to obtain knowledge, the ability to design, and the
ability to perform multi-dimensional value judgments within the space defined by the
“triple E”: effectiveness, efficiency, ethicality [10, p. 35].
186 W. W. Gasparski
Acknowledgements I wish to thank the editors of Transformacje annual, in which the Polish version
[11] of the present paper is being published, for encouraging me to consider agency in a praxiological
approach.
References
1. Agazzi, E.: Dobro, zło i nauka: Etyczny wymiar działalności naukowo-technicznej. Oficyna Aka-
demicka, Warszawa, Poland (1997) (English trans.: Dilworth, C. (ed.) Right, Wrong and Science:
The Ethical Dimensions of the Techno-Scientific Enterprise. Rodopi, Amsterdam (2004))
2. Bunge, M.: Treatise on Basic Philosophy. Ethics: The Good and the Right, vol. 8. D. Reidel, Dordrecht
(1989)
3. Davidson, D.: Essays on Actions and Events. Clarendon, Oxford (1980)
4. Gasparski, W.: Projektowanie humanistyczne. Prakseologia 2(78), 27–42 (1981) (English trans.:
Humanistic design. In: Jaques, R., Powell, J.A. (eds.) Design: Science: Method, pp. 97–105.
Guildford, Westbury House (1981))
5. Gasparski, W.: O poj˛eciu praktyczności [On the concept of practicality]. Prakseologia 101, 57–70
(1987/1988)
6. Gasparski, W.: Prakseologia i GOMOPOP [Praxiology and GOMOPOP]. In: Szyszkowska, M. (ed.)
Polityka a prawo, gospodarka, moralność [Politics and Law, Economy and Morality], pp. 77–99.
Instytut Studiów Politycznych PAN, Warszawa (1992)
7. Gasparski, W.: A Philosophy of Practicality: A Treatise on the Philosophy of Tadeusz Kotarbiński.
Societas Philosophica Fennica, Helsinki (1993)
8. Gasparski, W.: Prakseologia [Praxiology]. Wydawnictwo Warszawskiej Szkoły Zarzadzania, ˛
Warszawa (1999)
9. Gasparski, W.: Działania zbiorowe — uj˛ecie prakseologiczne [Collective actions — praxiological
approach]. In: Matysiak, A. (ed.) Działania zbiorowe — teoria i praktyka [Collective Actions —
Theory and Practice], pp. 9–21. Prace Naukowe Akademii Ekonomicznej we Wrocławiu, No. 985,
Wydawnictwo Akademii Ekonomicznej we Wrocławiu, Wrocław (2003)
10. Gasparski, W.: Dlaczego ludzie tak działaja,˛ jak działaja˛ oraz jak działać powinni [Why people act
the way they act, and how they shoud act]. Master Bus. Adm. 1(90), 27–36 (2008)
11. Gasparski, W.: Sprawność w uj˛eciu prakseologicznym. In: Zacher, L.W. (ed.) Moc sprawcza ludzi i
organizacji, pp. 54–76. Poltekst, Warszawa (2016)
12. Kotarbiński, T.: Dzieła wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I].
Ossolineum, Wrocław (1999)
13. Kotarbiński, T.: Od tematu do Traktatu [From the theme to the treatise]. In: Kotarbiński, T. (ed.)
Dzieła wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I], pp. 1–4. Ossolineum,
Wrocław (1999)
14. Kotarbiński, T.: Cel czynu a zadanie wykonawcy [The goal of an act and the task of the agent]. In:
Kotarbiński, T. (ed.) Dzieła wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I], pp.
6–19. Ossolineum, Wrocław (1999)
15. Kotarbiński, T.: Zagadnienie istnienia przyszłości. In: Kotarbiński, T. (ed.) Dzieła wszystkie:
Prakseologia Cz˛eść I [Collected Works: Praxiology Part I], pp. 58–72. Ossolineum, Wrocław (1999)
(English transl.: The problem of the existence of the future. Pol. Rev. 13(3), 7–22 (1968))
16. Kotarbiński, T.: O rozszerzaniu sfery czynu [On the enlargement of the sphere of an act]. In:
Kotarbiński, T. (ed.) Dzieła wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I],
pp. 72–85. Ossolineum, Wrocław (1999)
17. Kotarbiński, T.: Poj˛ecie zewn˛etrznej możliwości działania [On the external possibility to act]. In:
Kotarbiński, T. (ed.) Dzieła wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I],
pp. 104–108. Ossolineum, Wrocław (1999)
18. Kotarbiński, T.: Zasady teorii czynu [Principles of action theory]. In: Kotarbiński, T. (ed.) Dzieła
wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I], pp. 132–133. Ossolineum,
Wrocław (1999)
19. Kotarbiński, T.: O stosunku sprawstwa [On the agency relationship]. In: Kotarbiński, T. (ed.) Dzieła
wszystkie: Prakseologia Cz˛eść I [Collected Works: Praxiology Part I], pp. 121–132. Ossolineum,
Wrocław (1999)
20. Kotarbiński, T.: Czyn [An act]. In: Kotarbiński, T. (ed.) Dzieła wszystkie: Prakseologia Cz˛eść I
[Collected Works: Praxiology Part I], pp. 134–138. Ossolineum, Wrocław (1999)
21. Kotarbiński, T.: Czyn [An act]. In: Kotarbiński, T. (ed.) Dzieła wszystkie: Prakseologia Cz˛eść I
[Collected Works: Praxiology Part I], pp. 138–152. Ossolineum, Wrocław (1999)
22. Kotarbiński, T.: Dzieła wszystkie: Traktat o dobrej robocie [Collected works, Treatise on good job].
Ossolineum, Wrocław (2000) (English transl. by O. Wojtasiewicz: Praxiology: An Introduction to the
Science of Efficient Action. Oxford, Warszawa/ Pergamon Press, PWN (1965))
23. Kotarbiński, T.: Dzieła wszystkie: Prakseologia Cz˛eść II [Collected Works: Praxiology Part II].
Ossolineum, Wrocław (2003)
24. Meløe, J.: The agent and his world. In: Skirbekk, G. (ed.) Praxeology: An Anthology, pp. 13–29.
Universitetsforlaget, Bergen (1983)
25. Schön, D.A.: Educating the Reflective Practitioner. Jossey-Bass, San Francisco (1987)
26. Schön, D.A.: The crisis of professional knowledge and the pursuit of an epistemology of practice.
In: Auspitz, J.L., Gasparski, W.W., et al. (eds.) Praxiologies and the Philosophy of Economics, pp.
163–186. Transaction Publishers, New Brunswick (1992)
27. von Mises, L.: Ludzkie działanie: Traktat o ekonomii (fragmenty). Prakseologia 3–4(128–129), 11–
234 (1987) (English trans.: Human Action: A Treatise on Economics. Ludwig von Mises Institute,
Auburn (1998))
28. Znaniecki, F.: Wst˛ep do socjologii [An Introduction to Sociology]. PWN, Warszawa (1988)
W. W. Gasparski ()
Koźmiński University, Warsaw, Poland
Zygmunt Zawirski: A Creator of New Ideas
in Methodology of Science and Scientific
Metaphysics
Abstract Zygmunt Zawirski was one of the greatest member of the Lvov-Warsaw
School. The paper presents his life and the list of his main works, the greatest achieve-
ments in philosophy of nature, methodology of science and metaphysics. He pointed at the
possibility of studying the scientific metaphysics with the use of the axiomatic method.
His important work was the philosophical issue of time: The evolution of the notion of
time. Zawirski looked for the justification of the scientific research, listing the problem of
induction and deduction, development of the scientific theories and relations of science
and metaphysics.
Keywords Methodology of science · Scientific metaphysics · Zygmunt Zawirski ·

Lvov-Warsaw School · Axiomatic method · Philosophy of nature
Mathematics Subject Classification (2000) Primary 01A70, 03A05; Secondary 00A30
1 Life
Zawirski Zygmunt Michał was born on September 29, 1882, in Podolia in Berezowica
Mała next to Zbaraż and died on April 2, 1948, in Końskie.
He has been studying under Kazimierz Twardowski’s direction at the Philosophical
Faculty of Jan Kazimierz University in Lvov from 1901 till 1906. During his studies he
also attended other lectures led, among others, by Mścisław Wartenberg (1868–1938)
on issues concerning metaphysics after Kant and Witold Rubczyński (1864–1938) on
history of Greek philosophy. He also studied mathematics, physics and philosophy in
Berlin (1910) attending lectures led by Carl Stumpf, Georg Simmel and Alois Adolf Riehl
and in Paris (1910).
In 1904 Zawirski became one of the founder members of The Polish Philosophical
Society, appointed by Kazimierz Twardowski in Lvov. He has achieved a PhD degree in
philosophy in July 1906 on the basis of his work, which was written under Twardowski’s
direction. Zawirski belonged to the first generation of Twardowski’s students.

190 K. Śleziński
In 1911 Zawirski started to cooperate with “The Philosophical Movement” and wrote
many reports on books and reviews of contents from the French and German philosophical
magazines such as “Revue Philosophique de la France et de L’étrange”, “Revue de
Métaphysique et de Moral” oraz “Archiv für Geschichte der Philosophie”. Apart from his
reporting activity, he has participated actively in the meetings of the Polish Philosophical
Society in Lvov, presenting many papers developed in the form of the scientific articles
or more advanced writings. Zawirski achieved the first prize in the 3rd competition
of “Philosophical Review” in 1912 on the basis of his work entitled Causality and
Functional Relation. An Epistemological Study [2]. In this work he demonstrated that
it is impossible to reduce completely the notion of causality to the notion of functionality.
Since 1915 the subject of his interest included the following problems: hypothesis of
constant returns all-matters, inductive metaphysics, relations between metaphysics and
science, detailed issues from logic and their significance in mathematical and natural
research. The research conducted by him had an impact on development of his opinions.
In 1923 Zawirski presented a paper entitled Modern Attempts for Axiomatization
of Mathematical Nature and their Philosophical Significance during the 1st Polish
Philosophical Convention to share the research conducted and results concerning the
implementation of axiomatic method used in philosophy of nature. In the same year he
wrote a thesis entitled Axiomatic Method and Natural Sciences [5] and presented it to
Władysław Heinrich (1869–1957). This work was the basis to initiate the proceedings for
the qualification as a university professor at the Jagiellonian University in 1924.
In the period between 1928 and 1936 Zawirski linked with the University of Poznań.
Władysław Mieczysław Kozłowski (1858–1935) had retired and Zawirski was appointed
as lecturer for theory and methodology of sciences courses at the Humanistic Faculty and
since the 1st of August 1929 as associate professor at the Mathematical and History of
Science Faculty. The classes led by Zawirski had a good reputation among students. He
combined his didactic and scientific activity. He was interested in students’ access to the
basic philosophical works which were the subjects of his lectures and seminars. Then,
he gathered valuable literature in his department, which were destroyed by the Nazis
during the World War II. At the time of his lectures, Zawirski focused on the philosophical
problems of history of nature, basic problems of mathematics, issues of epistemology and
theory of classes and relations.
The period of working for the University of Poznań is the most important stage in his
scientific life. He got in the close touch with his master Kazimierz Twardowski. Zawirski
sent him reports and reviews of books for “Philosophical Movement” edited by him.
He participated in the 7th International Congress of Philosophers in September 1930
in Oxford. Zawirski was awarded with the first and very prestigious prize in Rignan’s
competition in 1933 announced by the Italian magazine “Scientia” for his work entitled
L’évolution de la notion du temps [10].
Zawirski was nominated as a full professor in 1934. He stayed 2 years more at the
Faculty of Mathematics and Natural Sciences where he was a Dean and a Chair of
Theory and Methodological Sciences Department being an active member of scientific life
internationally. He participated in the 8th International Philosophical Congress in Prague
(1934) and in the 9th Congress in Paris (1934). In the same year he welcomed members
of the International Convention of Thomistic Philosophy held in Poznań instead of absent
Michał Sobecki, who was the President of Poznań Philosophical Society. Zawirski also
Zygmunt Zawirski: A Creator of New Ideas . . . 191
participated in the 1st (Paris 1935) and the 2nd (Kopenhaga 1936) Congress of Scientific
Philosophy and in the 3rd Polish Philosophical Convention in Cracow in 1936 where he
presented a paper entitled On the matter of Scientific Synthesis [11].
Near the end of his scientific work at the University in Poznań, he was awarded an
honorary doctorate by the University of Poznań and the Faculty of Mathematics and
Natural Sciences on the 12th of November in 1936 and accepted it on the 18th of
November in 1936 at the hands of president Ignacy Mościcki.
He was asked by Władysław Heinrich in 1935 to chair a faculty after Tadeusz
Grabowski (1869–1940). Zawirski accepted it and as a full professor has started his work
since the 1st of January in 1937 at the Philosophical Faculty. Later after its division, he
worked at the mathematics and Natural Sciences Faculty. In the period between 1938 and
1939 as well as between 1945 and 1946 he was a Dean of this Faculty.
He took over editorial office of “The Philosophical Quarterly” after Władysław
Heinrich in 1936. In the period between 1938 and 1945 he was the President of Cracow
Philosophical Society and gave papers entitled, among others On the Scientific Activity of
Professor Kazimierz Twardowski [22] and The Genesis and Development of Intuitionistic
Logic [21]. In the period between 1938 and 1941 he worked on Philosophical Dictionary
[23]. Unfortunately, the censorship stopped the printing of dictionary copies in 1948. A
manuscript of this dictionary survived in the Polish Academy of Sciences Archives and
only some terms were published in 1993.
The Nazis pacifist action “Sonderaktion Krakau” took place on the 6th of November
in 1939 against the Polish researchers and scientists. Zawirski was outside Cracow on this
very day and due to it he barely avoided a transportation to the Nazis concentration camp
in Sachsenhausen. During the World War II he participated in the clandestine academic
teaching. After the WW II he was a full professor at the Jagiellonian University. He has
been a chairperson of the Cracowian Philosophical Society since 1945.
Zawirski was very active in the scientific life in the period between 1945 and 1948. At
that time his works were published as the result of his long standing research. Travelling
to Zakopane to attend Philosophical Conference in winter 1947, a luggage with two
manuscripts was stolen from him including a methodology manual for natural sciences
On the Scientific Method [12] and manuscript of Patristic Monography.1
Zawirski prepared a written speech for the 10th International Philosophical Congress
in Amsterdam in 1948 but unfortunately he did not manage to present it. The work
undertaken at this year, and first of all the works of manuscript reconstruction overstrained
his organism moving the unexpected catastrophe. Zawirski died suddenly at his son
Kazimierz’s home in Końskie.
2 Writings
His leading papers include the following: Causality and Functional Relation [1], Philo-
sophical Relativism and the Physical Theory of Relativity [4], The Connection between
the Principle of Causality and the Principle of Relativity [6], Axiomatic Method and
1 Roman Ingarden in Wspomnienia o prof. Zawirskim (Recollection of Professor Zawirski [39, p. 261])
(1948) wrote that the author managed to reconstruct his stolen works.
Natural Sciences [5], The Eternal Return of the Worlds. Historical-critical Investigations
of the Idea of “Eternal Return” [7], On Indeterminism in Quantum Physics [9], Über das
Verhältnis der mehrwertigen Logik zur Wahrscheinlichkeitsrechnung [19], L’évolution de
la notion du temps [10], On the Problem of Scientific Synthesis [11], The Importance
of Logical and Semantical Investigations for the Theories of Contemporary [20], The
Genesis and Development of Intuitionistic Logic [21], Remarks on the Method for Natural
Sciences [12], Time: Selections from “The Evolutions of the Notion of Time” [13], Science
and Metaphysics [14, 15] and On the Relation between Metaphysics and Science [16].
3 Views
We present below only an outline of the whole academic and scientific achievements of
Zygmunt Zawirski, grouping it simply into three characteristic areas of research problems
originated from philosophy of nature, methodology of science and metaphilosophy. It also
includes a discussion on the mutual relations between science and philosophy as well as
problems of the scientific metaphysics. Philosophy of nature is one out of many research
areas interesting for representatives of Lvov-Warsaw School. Research results from
logic, methodology, philosophy of nature and metaphilosophy are commonly discussed.
However, little attention is devoted to research from philosophy of nature and scientific
metaphysics conducted within this School. The main representative of this trend was
Zawirski, who in much more degree used content of the investigated scientific theories
than it was observed in philosophy of science.
3.1 Philosophy of Nature
Zawirski was interested in development of natural sciences, first of all the theory of
relativity and quantum mechanics. He has left many original works linked with these
theories trying to answer the questions stated and philosophical matters discussed at that
time. Zawirski presented new philosophical implications derived from these theories in
his numerous works and their impact on wider understanding of reality.
One of the first matters out of the natural history raised by Zawirski was a hypothesis
on the so called “eternal returns of the worlds”. He published a series of three papers on
these matters in “The Philosophical Quarterly” in the period between 1927 and 1928. He
had already thought these problems over, studied and had been ready for 15 years. Its first
versions were presented on the 27th May and 24th June 1911 at the meetings of Polish
Philosophical Society in Lvov as well as during the Convention of Polish Doctors and
Naturalists in Cracow in the same year.
Zawirski came back to this motive of “eternal returns” in his main work entitled
L’évolution de la notion du temps, which was published in 1936. His work has been still
worth considering due to the clear and profound presentation of history linked with the
notion of time, to start with the Pythagoreans and to end with the modern philosophical
concepts of his time including H. Bergsona and E. Husserla and the latest relativist and
quantum physics. Zawirski, conducting research on nature of time, was interested in the
philosophical principles and implications linked with the general theory of relativity in
the 1920s of the twentieth century. Only 5 years after Albert Einstein had introduced his
theory, which means rather promptly, Zawirski published his paper entitled Philosophical
Reflection on the Theory of Relativity [3] in 1920. He noticed its crucial importance
and explained that it is required to eliminate a number of philosophical premises out
of science, which had been accepted earlier by the physical theories and expressed in
the notion of absolute time, space and movement. Zawirski paid close attention and
followed the development of research undertaken within Einstein’s theory. It is confirmed
by his numerous reviews of books on this theory published at that time (in total 10
reviews).
Zawirski also investigated problems of casual relation from various perspectives
during all his scientific research. He was assured about the importance of this principle
for scientific research. In his work entitled Causality and Functional Relation [1], he
demonstrated that the notion of “casual relation” and “functional relation” differ in terms
of content. As the casual relation is the real relation, considering the influences and time
relations between reasons and consequences, it cannot be replaced with the functional
relation. In his paper entitled Quantum Theory and the Causality principle [8] from 1930,
he polemicized with Heisenberg’s theory that uncertainty principle proves the falsity of
causality principle.
Zawirski was a pioneer of works about the implementation of many-valued logics
in quantum mechanics and works about the probabilistically based approach to many-
valued logics. He linked the system of logic, which can be used in the description
of quantum phenomena, with the system of infinite many-values corresponding to
the levels of probability. Zawirski designed the system of many-valued logic, which
possesses all the advantages of Reichenbach’s system developed in 1932. Simultaneously,
Zawirski avoids its disadvantages such as Reichenbach’s complications unnecessarily
introduced for counting implications and equivalence, which leads to an assignment
of these logic functions as many values as for the sum and product. Designing the
system of many-valued logic, Zawirski used also Jan Łukasiewicz’s and Emil Posta’s
research results. Both of them, independently of each other and almost simultaneously
around 1921, generalised of many-valued logic. However, none of them described the
relation between logic and theory of probability. Initially, it was achieved only by Hans
Reichenbach.
Zawirski pointed at two potential uses of many-valued logic in 1932 in the paper en-
titled Les logiques nouvelles et le champ de leur application in “Revue de Métaphysique
et de Morale” [17]. The first one focused on the link between the theory of probability
and many-valued logic. The second one focused on the use of Łukasiewicz’s three-valued
logic in the analysis of wave-particle duality. Nevertheless, Zawirski admitted that this
attempt was too early. In further works, our philosopher was for the use of the theory of
probability in the description of quantum phenomena. He stated that various degrees of
probability can be assigned to quantities of feedback such as time, energy, location and
momentum.
3.2 Methodology of Science
Zawirski paid the attention to any signal of philosophical thinking outside the bor-
ders of our country. He reacted violently to the methodological research included in
Karl R. Popper’s work entitled Logik der Forschung, published in Vienna in 1934.
Zawirski criticised Popper’s falsificationism stating that disagreement of some law
with one of the recognised and elementary opinions can be regarded as the sufficient
reason for the rejection of this law. Zawirski notices that a single empirical law
hardly ever is tested perceived separately from the other laws. The whole system of
opinions or theory is tested or invalidated. Moreover, a number of laws and indepen-
dent hypotheses constantly decreases during this test. Then, each general empirical
task “takes responsible for itself” for the whole system to which it belongs [20].
In this way development of real sciences constituents a continuation of the theory,
which fight with each other and modify constantly. Then their number always de-
creases.
A problem of testing hypotheses in empirical sciences was also undertaken by
Zawirski in the paper entitled Remarks on the Method of Natural Sciences [12]. In
a similar way to Popper, Zawirski assumes that we derive consequences out of the
accepted hypothesis. Then, we test if they agree with the facts concerned with them.
The hypothesis is proved correct when there is an agreement between directly observed
facts and conclusions derived out of it. If there is a disagreement, the hypothesis
should be rejected due to its invalidation. Zawirski in his paper, not mentioning the
name of Popper, stills criticised his concept of falsification and impossibility to achieve
experimentum crucis. Currently, it can be stated that he criticised the so called naive
Popper’s falsificationism.
Zawirski accepted the present asymmetry between positive and negative result in
testing a hypothesis. A question whether the negative result is more sonorous than
the positive one remained a problem for him. It is clear for Zawirski that the positive
result still does not prove a rightness of the particular hypothesis because it can be
changed by the minute. Whereas the negative result does not lead always towards a
complete withdrawal of the hypothesis. Invalidation of the hypothesis might be a decisive
moment only when none of the notions describing an experiment remained unchanged.
Every single change of meaning in the terms used leads to another undertaking of
the particular hypothesis, in spite of the fact that it was invalidated. The example
of such situation can be earlier rejected wave-particle theory of light after Foucault’s
experiments and later introduced by the theory of quantum. A similar situation occurs
in the following experiments experimentum crucis and instantia crucis when we select
the one which includes some newly-revealed fact out of two competing hypotheses or
theories.
Zawirski notices the analogy that is present between a verification of the particular
hypothesis and acceptance of the one out of two competing hypothesis. As long as we
expect the positive or negative answer in the first case, the positive answer linked with
the one hypothesis in the second case is simultaneously the negative answer for the
another one. In the situation of experimentum crucis it occurs rarely that two competing
hypothesis were the opposite opinions. There are usually more complicated hypotheses.
Therefore, logical conjunctions of sentences are often checked, whose negation is their
alternative. Popper does not notice it in his concept of falsification. Zawirski agrees with
Duhem for whom experimentum crucis have never invalidated one isolated hypothesis but
only the whole theory full of linked tasks.
Comments presented above refer only to the theoretical testing of hypothesis. In
practice, the situation does not look as such complicated. Each theory, apart form the
formulated laws, includes also numerous terms defined and agreements. The simplest
way of testing hypothesis is usually selected. However, more complicated cases occur
and may trigger “a revolution in science”. This situation occurred during the development
of theory of relativity and quantum mechanics when a number of negative experiences
increased constantly as well as supporting hypotheses which explained these experiences.
After some time, the edifice of classical physics has been reconstructed, which appeared
to be the best solution in this difficult situation.
3.3 Metaphilosophy
The scientific interests of Zawirski were link also with more general problems of truth
and being. The point of departure for his interests were the profound cognition of classical
metaphysics and natural sciences within which more and more often the problems were
undertaken previously having been reserved only for philosophers.
In the first decades of the twentieth century Zawirski witnessed an argument and a
dispute concerned with the role of natural sciences in the development of general outlook
on life. He noticed that both metaphysicians and opponents of metaphysics expressed a
need to develop a scientific outlook on life. This fact made him ponder over a possibility
to design metaphysics based on experience. He realized very quickly that metaphysics
understood in this way would not be able to replace completely classical metaphysics
perceived as scientia entis. It would not have been balanced only with the synthesis of
natural sciences admitted by positivists. However, considering the mutual straggle of the
most opponent reasoning movements, he undertook an attempt to develop a middle path
leading towards the formation of the scientific metaphysics—critical and open using the
results of the empirical experience. Zawirski presented his first ideas about the relation
between metaphysics and science on the 5th of May in 1917. It was his lecture entitled
On Relation between Metaphysics and Science [16] at the meeting of the Philosophical
Society in Lvov.
The issue of the relation between metaphysics and science and possibilities of
developing a general theory about the reality dominated in Zawirski’s research till
1923 when he published his post-doctoral thesis entitled Axiomatic Method and Natural
Sciences [5]. The work was the summary of the earlier conducted research focusing on the
possibility of axiomatization of metaphysical systems. Before 1923 Zawirski had written
also two more works at that time on the matter [14, 15] and [16]. Both works remained in
the manuscript and only in the period between 1995 and 1996 Science and Metaphysics
was published in “Philosophy of Science”. Then, the second work On Relation between
Metaphysics and Science was published in 2003.
Zawirski related once again to the problems presented in the post-doctoral thesis in
1936 during the session of the 3rd Polish Philosophical Convention in the paper entitled
On the matter of the scientific synthesis [11]. In his presentation he introduced a different
argumentation in the defence of scientific metaphysics than the one presented in the work
entitled Axiomatic Method and Natural Sciences. Following Gödel’s statement on the
incomplete systems, he argued that we should not refuse the sense of the metaphysical
issues only because they are not subject to empirical testing. He also pointed to the tasks
of philosophy, stating that it should not undertake the issues for which it is not possible to
find methods of their solution.
4 Resonance
Zawirski’s works contributed to the development of methodological, logical, ontological

and meta-philosophical research. Zawirski introduced the potentials for formalisation of
the scientific theories for natural history, the manner of investigating the philosophical
assumptions and implications derived from the modern physical theories, the critical
analysis of scientific terms including their role in the growth of scientific knowledge
as well as the implementation of methodological principles in scientific theories. He
showed an opportunity to develop scientific metaphysics using the axiomatic method and
opportunity to generalize scientific research philosophically to comprehend the world.
He looked for the grounds of scientific research, indicating the methodological problem
of induction and deduction, growth of scientific theories, relation between science and
philosophy. He pointed at the importance of ontological research linked with time and
space.
His suggestions within the methodology of science, which were studied under the
impact of critical perception of Popper’s work entitled Logik der Forschung, appeared
to be accurate. It was confirmed by Popper’s another work entitled Conjectures and
Refutations from 1963 where Zawirski’s objections were considered (without mentioning
his name). Zawirski suggested that induction and deduction are equally impotent in
the natural science. He said, that falsification, like verification, has never had the final
character of empirical theory, since one does not test one singular theory, but always a
group of theories or laws; and also, that the development and changes of science are of
the infinite nature. It should be exposed that Zawirski’s student named Zygmunt Spira
was also interested in Popper’s work and developed creatively the issues linked with
methodology of natural sciences in the work entitled Comments on the Methodology and
Popper’s Theory of Learning (1946). Zawirski developed a concept of three worlds in
the scope of his research such as the world of senses, the psychological world and the
world of eternal and ideal being. A concept of three worlds, similar to Zawirski’s one, has
been developed independently by Karl Popper (Objective Knowledge in 1972) and Roger
Penrose (Shadows of the Mind in 1994).
The research, previously conducted by Zawirski, on the use of the formal log-
ics in the investigation of the physical scientific theories were advanced by his stu-
dent named Romana Suszko. He developed the theory of models and used it in
the investigation of problems beyond mathematics (Remarks on Sentential Logics in
1958).
However, Zawirski’s main composition entitled L’évolution de la notion du temps

has been very important scientifically due to the detailed analysis of the time problem.
Many issues still discussed currently can be found there. They refer to thermodynamics of
irreversible processes, the second law of thermodynamics, theory of dissipative processes
(I. Prigogine, I. Stengers, Order out of Chaos—1984) and conformal cyclic cosmology
(R. Penrose, Cycles of Time—2010).
Zawirski’s works related to the critical and historical analysis of scientific terms
and philosophical reflation over the assumptions for the empirical theories and the
investigation of the traditionally philosophical problems entangled into these theories
constitute the model prototype of the scientific works from the field of philosophy of
nature, natural history and philosophy of nature. The works of this type are still conducted
in Cracow background of philosophers such as Józef Życiński (1948–2011) or Michał
Heller (born 1936), a director and founder of The Copernicus Centre for Interdisciplinary
Studies.
5 Pupils
Zygmunt Zawirski by his didactic and scientific work inspired many Polish logi-
cians and philosophers. The following students, among others wrote their diploma
works under his guidance or worked in these fields: Józef Maria Bocheński (1902–
1995), Andrzej Grzegorczyk (1922–2014), Zbigniew Jordan (1911–1977), Zygmunt Spira
(1911–1942?) and Roman Suszko (1919–1979). His close students are the last three
ones.
6 Summary
Zawirski’s works and presentations focused on the important research problems in the
first part of the twentieth century. In particular, they referred to the philosophy of nature,
formal logics, methodology of science and meta-philosophy. He started to cooperate with
philosophers, among others, from the Great Britain, France, Germany and Austria. He was
active in the scientific work at the international level. He gave lectures and participated in
discussions during the International Philosophical Congress including the 7th Congress
in Oxford, the 8th in Prague, the 9th in Paris; for the 10th Congress in Amsterdam
he had already sent his paper. He was interested in the development and consequences
of the general theory of relativity and the quantum theory. From the perspective of
developing natural history he investigated questions of the determinism, indeterminism
and causality, ontological problems of time and space, methodological problems of the
relations between the verification and the falsification, the relations between a hypothesis
and the empirical basis, but also the possibility of an experimentum crucis. He was
interested in the modality of judgements and epistemological principles of the intuitionist
logic—Brouwer-Heyting’s logic. He tried to applied to the new Łukasiewicz’s logic,
many-valued logic and quantum mechanics. His metaphilosophical works are also worth
paying attention. First of all, the works that focus on the investigation of the relations
between science and philosophy are important and the attempt to develop the scientific
metaphysics using the axiomatic method.
Zygmunt Zawirski’s Works Mentioned in the Paper
1. Przyczynowość a stosunek funkcjonalny [Causality and functional relation]. Przeglad ˛ Filozoficzny

(PF) [Philos. Rev.] 15(1), 1–66 (1912)
2. Przyczynowość a stosunek funkcjonalny. Studium z zakresu teorii poznania [Causality and Functional
Relation. An Epistemological Study]. Lwów (1912)
3. Refleksje filozoficzne nad teoria˛ wzgl˛edności [Philosophical reflections on the theory of relativity].
PF 23, 343–366 (1920)
4. Relatywizm filozoficzny a fizykalna teoria wzgl˛edności [Philosophical Relativism and the Theory of
Relativity]. Drukarnia Słowa Polskiego, Lwów (1921)
5. Metoda aksjomatyczna a przyrodoznawstwo [The axiomatic method and natural sciences]. Kwartal-
nik Filozoficzny (KF) [Philos. Q.] 1(4), 508–545 (1923/1924), 2(1), 1–58 (1923/1924), 2(2), 129–157
(1923/1924)
6. Zwiazek
˛ zasady przyczynowości z zasada˛ wzgl˛edności – odczyt habilitacyjny [The connection
between the principle of causality and the principle of relativity – postdoctoral thesis]. KF 2(4), 397–
419 (1924)
7. Wieczne powroty światów. Badania historyczno-krytyczne nad doktryna˛ “wiecznego powrotu” [The
Eternal Returns of the Worlds. Historical-Critical Investigations of the Idea of “Eternal Returns”].
Uniwersytet Jagielloński, Kraków (1927)
8. Teoria kwantów a zasada przyczynowości [The quantum theory and the principle of causality]. PF
33(4), 296–301 (1930)
9. W sprawie indeterminizmu fizyki kwantowej [Indeterminism in Quantum Physics]. Ksiażnica ˛ Atlas,
Lwów (1931)
10. L’évolution de la notion du tamps. Scientia 28(55), 4, 249–261 (1934)
11. W sprawie syntezy naukowej [On the matter of scientific synthesis]. PF 39(4), 347–352 (1936)
12. Uwagi o metodzie nauk przyrodniczych [Remarks on the method for natural sciences]. PF 44(4),
315–318 (1948)
13. Time: selections from “The Evolutions of the Notion of Time”. In: Szumilewicz-Lachman, I. (ed.)
Zygmunt Zawirski: His Life and Work. Kluwer Academic Publishers, Dordrecht (1994)
14. Nauka i metafizyka (I) [Science and metaphysics I]. Filozofia Nauki [Philos. Sci.] 3(3), 104–135
(1995)
15. Nauka i metafizyka (II) [Science and metaphysics II]. Filozofia Nauki 4(1), 131–143 (1996)
16. O stosunku metafizyki do nauki [On the Relation Between Metaphysics and Science]. Wydział
Filozofii i Socjologii Uniwersytetu Warszawskiego, Warszawa (2003)
17. Les logiques nouvelles et le champs de leur application. Revue de Métaphysique et de Morale 39(4),
503–579 (1931)
18. O stosunku logiki wielowartościowej do rachunku prawdopodobieństwa [On the relations between
many-valued logic and probability calculus]. In: Pami˛etnik XIV Zjazdu Lekarzy i Przyrodników
Polskich. Drukarnia P. Mitr˛egi, Cieszyn (1933)
19. Űber das Verhältnis der mehrwertigen Logik zur Wahrscheinlichkeitsrechnung. Stud. Philos. 1, 407–
442 (1935)
20. Doniosłość badań logicznych i semantycznych dla teorii fizyki współczesnej [The importance of
logical and semantical investigations for the theories of contemporary]. PF 41(1), 25–30 (1938)
21. Geneza i rozwój logiki intuicjonistycznej [The genesis and development of intuitionistic logic]. KF
16(2–4), 165–222 (1946)
22. O działalności naukowej prof. Kazimierza Twardowskiego [On the scientific activity of professor
Kazimierz Twardowski]. RF 15(3), 157 (1939)
23. Zmarli profesorowie, docenci i asystenci Oddziału Matematyczno-Przyrodniczego Wydziału Filo-

zoficznego [The departed professors, readers and associate professors at Mathematical and Natural
History Department at the Faculty of Philosophy]. In: Kronika Uniwersytetu Jagiellońskiego za
okres wojny 1939–45 oraz za rok akademicki 1945 [Jagiellonian University Chronicle of the Period
Between 1939 and 1945 Plus the Academic Year of 1945]. Kraków (1946)
K. Śleziński ()
Institute of Educational Science in Cieszyn, University of Silesia in Katowice, Cieszyn, Poland
Zygmunt Zawirski’s Concept of Scientific
Metaphysics
Abstract In the thesis Zygmunt Zawirski’s concept of scientific metaphysics two ques-
tions were posed as the means of presenting an explorative issue: Can axiomatic
method give rise to philosophy as a science, study of universal “order”? Does scientific
metaphysics constitute solid, mathematical representation of our intuitions and instinctive
strivings and development of previous attempts to demonstrate the existence of unity
among multiplicity of phenomena. The article aims to reconstruct Zygmunt Zawirski’s
metaphilosophy, among other things: relations between the sciences and philosophy.
Keywords Scientific metaphysics · Zygmunt Zawirski · Relations between the

sciences and philosophy · Metaphilosophy
Mathematics Subject Classification (2000) Primary 00A05; Secondary 00A30
1 Introduction
Polish philosophers from the beginning of the twentieth century apart from conducting the
detailed research on logics, semiotics, research methodology and ontology, were aware of
the general aims headed by philosophy [3, 4]. Their objective constituted the attempt to
define the whole view on reality with a particular focus on research method in philosophy,
which was discussed at the 2nd Polish Philosophical Meeting in 1927 in Warsaw. At that
meeting, Jan Łukaszewicz presented the paper entitled On Method in Philosophy [7, 8].
Florian Znaniecki presented his paper entitled Tasks for Philosophical Synthesis [25]. In
his speech he presented a pluralistic approach to philosophy perceived as the multitude
syntheses of knowledge accepted at the separate stages of the development linked with
civilisation.
A clear polemical tone is observable in the 1930s, among others in the speeches of
Jan Łukasiewicz, Maria Kokoszyńska, Zygmunt Zawirski, Roman Ingarden, Bolesław
Gawecki, Tadeusz Garbowski and Joachim Metallmann, on the views of members of
Vienna Circle linked with the synthesis of scientific knowledge. The issue has also been
undertaken at the 3rd Philosophical Convention in 1936 in Cracow. However, one mutual

agreement was achieved in terms of developing the general view of realities. Still, the
discussion undertaken at this meeting directed the trend of further research on the relations
between science and philosophy as well as a study on the scientific metaphysics, as Jan
Łukasiewicz appealed to, based on the axiomatic method [8]. It is worth noticing that
such attempts had been already undertaken by Benedykt Bornstein [2, 15] and Zygmunt
Zawirski, whose suggestions will be discussed below.
2 Relation Between Science and Metaphysics
The departure for Zygmunt Zawirski’s study on the scientific metaphysics was the
profound comprehension of the classical metaphysics and the scientific theories. In
particular, these were the general theory of relativity, thermodynamics and quantum
mechanics within which one can understand the problems previously reserved only to
philosophers.
Zawirski experienced a dispute related to the role of the natural sciences in the
study on the scientific metaphysics in the first decades of the twentieth century. He
noticed that both metaphysics and its opponents represent a need for the study on
the scientific view of the world. This detail made him started deliberations on the
possibility of studying metaphysics based on an experience. He realised very quickly
that metaphysics understood in this manner would not be able to replace the classical
metaphysics completely perceived as scientia entis. It would be an equivalent to it, only
to the synthesis of the natural sciences accepted by the Neo-positivists of the Vienna
Circle. Taking into consideration, the mutual aspirations for the most opposite directions
of thinking, he undertook an attempt to delimit the way in the middle leading towards to
the new scientific metaphysics being constituted. The new one was to be critical and open,
applying the results of the empirical experience. Zawirski presented the first reflections
about the relations between metaphysics and science on the 5th of May in 1917 in his
lecture entitled On the Relation between Metaphysics and Science [24] presented during
a meeting of Philosophical Society in Lvov.
The issues of the relations between metaphysics and science and the possible assump-
tions accepted in the study on the philosophical theory of reality dominated in Zawirski’s
research till 1923. This was the year when he published his postdoctoral thesis entitled
The Axiomatic Method and Natural Sciences [17]. This work has been the summary
of the previous research. Moreover, it included the possibility of axiomatization of the
metaphysical systems. Before 1923 Zawirski had written also two more works dedicated
to the matters undertaken at that time. They were entitled On the Relation between
Metaphysics and Science (1919), Science and Metaphysics—1920 [22, 23]. Both works
remained handwritten. It was only between 1995 and 1996 when Science and Metaphysics
was published in the pages of “Philosophy of Science”. Then, On the Relation between
Metaphysics and Science was published in 2003.
In these works, Zawirski distinguished two areas of empiriologia and meta-empirical
knowledge out of the field of metaphysical research heading towards the general knowl-
edge about the whole reality. Empiriologia is for him the science not undertaking the main
problems of classical metaphysics while meta-empirical knowledge is the science about of
Zygmunt Zawirski’s Concept of Scientific Metaphysics 203
the principle of being. Assigning and giving priority to empiriologia in the research on the
whole reality, he did not discredit the need of going in for meta-empirical knowledge. He
assumed that the science of empiriologia does not consist of the simple summing up of the
results from the detailed sciences and also is not their highest generalisation. Its paragon
is the transformation into the axiomatic and deductive system following mathematics.
However, each system of axioms and statements originated out of them tolerate
the possibility of many interpretations of their symbols. Then, many interpretations of
the results are possible in empiriologia, which occur in the process of meta-empirical
knowledge. Following Zawirski, although all interpretations of science in meta-empirical
knowledge, leading to various metaphysical systems, possess the same learning value,
he followed the idealistic concept criticising the individual metaphysical system. The
concept is worth attention and the most credible meta-empirical interpretation of the
scientific results defined by empiriologia [24, p. 96].
Empiriologia and meta-empirical knowledge were linked with the use of hypotheses
which went beyond experience. These hypotheses in empiriologia contribute to the
development of science and serve for the simpler depiction of content in experience.
These hypotheses in the meta-empirical knowledge are used in learning about objects
and metaphysical problems linked with them, which do not belong to the experience such
as the existence of absolute and transcendent being, being after death or deciding about
the issue of learning about the world.
Zawirski noticed that the fact that empiriologia existed, metaphysics referring to expe-
rience, would allow its many possible interpretations. He also noticed that empiriologia,
experiential metaphysics, would never replace the classical metaphysics completely. A
possibility of numerous interpretations originates from the fact that universal science
about reality is the inductive theory. The authenticity of such theory or any its type
cannot be proved and its absolute reliability cannot be shown [22, p. 113]. Symbols of
the inductive theory permit the infinite abundance of interpretations. Therefore, none of
these interpretations deserves the defined probability ranking. It indicates that inductive
metaphysics should not refer to the probability calculus. Taking it into consideration,
Zawirski postulates in the studies on the universal theory of reality that empiriologia
as the science should link the results of detailed sciences into the system freed from
contradictions and remain in such relation to these sciences in which detailed sciences
remain in relation towards their particular areas and fields [22, p. 135].
Empiriologia is the scientific metaphysics cannot replace the classical metaphysics.
Statements in empiriologia have got the nature of temporary hypothesis and constantly
are put into a test of arranging them into the previous experiences in the procedure of
experimentum crucis. These hypotheses are the important supplements of the scientific
picture of the world in empiriologia, which still does not give us its complete picture.
There are numerous and important issues for human being, such as ethical postulates,
outside the area of the scientific metaphysics.
In the definition of the universal opinion on reality within meta-empirical knowledge,
Zawirski achieved an interesting differentiation of entia rationis, the world of ideal beings
and entia naturae, the world of existence. Entia rationis exists independently of the
learning subject. The world of these beings include among others mathematical objects
and content of mathematical natural history. Entia rationis is not designed but it is learned
and discovered as everything else. The novelty constitutes only their life in our mind as
mental existences and as the defined content of our consciousness whereas not as eternal
existences beyond time. This world includes also figments of our thoughts, among others
metaphysical systems as well as all sorts of logical mistakes having the ideal being and
both true or false theories [24, p. 164]. However, should these systems or mistakes exist
only in our thoughts, they belong to the world of mental existence.
The world of senses is also the mental construction. Speaking about the mental
subjects and similar statements on reality are possible because there is absolute being,
over individual one, which following Zawirski’s opinion, belongs both to the world of
everlasting beings and to the world of existence as well.
The world of entia rationis is infinite and inexhaustible, beyond time and endless,
unlimited by human thoughts. Human mind should, in Zawirski’s opinion, following
Ockham’s principle, obtain out of this unlimited world of the eternal beings only as much
as it is necessary and indispensable to understand the world of existence [24, pp. 159–
164]. Moreover, it is forbidden to make a mistake and come down to a blind transfer of
the relations linked with into entia rationis into the relations linked with entia naturae.
The relations among entia rationis are ideal while the relations among existences are
dynamic. Considering an argument that the world of eternal beings incorporates the world
of existences, it indicates that all possible relations belong to entia rationis. The difference
between the one and another relation is that entia naturae relations refer to psychological
and temporal orders, which cannot be stated about the relations occurring between the
eternal beings. Then, the relations which can be implemented in the sense picture of the
world define interdependence and temporal order among the empirical data.
Zawirski’s research on the scientific metaphysic resulted in the study about the
ontological concept of three worlds—the world of senses, the world of psyche and
the infinity one. It should be notice that a similar to Zawirski’s concept of three
worlds was worked out independently by Karl Popper (Objective Knowledge, [12]) and
Roger Penrose (Shadows of the Mind, [10]). Undoubtedly, there are many similarities
among those concepts but also some differences. Each of them was still based on the
epistemological and ontological foundation in the analysis of learning and possessing
knowledge of the reality.
Zawirski referred to the problems listed in his postdoctoral thesis one more time in
1936 during the meeting of the 3rd Polish Philosophical Society in his paper entitled
On the Matter of Scientific Synthesis [19]. In his speech he presented a different
argumentation in a defence of the scientific metaphysics than the one which was described
in his work entitled The Axiomatic Method and Natural Sciences. Following Kurt Gödel’s
statement about the incomplete systems, he argued that the sense should not be refused in
terms of the metaphysical issues only because they are not subjected to the empirical test.
He also pointed at the tasks of philosophy, stating that it should not undertake the issues
for which a method of its solution could not be appointed.
3 Synthesis of Human Knowledge and Metaphysics
A synthesis of human knowledge was pointed often by the opponents of commonly

understood metaphysics in the 1930s of the twentieth century. Their synthetic views of
knowledge were anyway a sort of substitute for the acceptance of metaphysical resolu-
tions. The representatives of neo-positivism cited, on the one hand, the representatives
of the classical positivism in the process of knowledge unification, for example Comte or
Spencer. On the other hand, they tried to refer to Leibniz’s concept of mathesis universalis
[19, p. 347].
The second reference is unfortunate while it would introduce some metaphysical
elements into the constructive system of knowledge as it was done by the author of Mon-
adology. Leibniz solved the important issues, linking them more tightly with the elements
of his ontological-metaphysical system. The representative of neo-positivism followed
Encyclopaedism approach to the presentation of the whole knowledge rather than build a
synthesis where non-scientific and metaphysical elements should be introduced.
Zawirski, solving the problem related to the synthesis of human knowledge, was for
the impossibility of leading the demarcation line between the scientific synthesis and
metaphysics. In Zawirski’s opinion, the principal non-settlement of the metaphysical
hypotheses could not be accepted. None of the metaphysical or scientific hypotheses could
be tested directly. We can only discuss verifiability of the consequences of the accepted
hypotheses. The metaphysical and scientific hypotheses are the same on this account. This
situation does not change even having accepted Popper’s criteria for falsification of the
hypotheses. Zawirski agrees that falsification is an effective method of testing hypotheses.
However, as he adds, only a few and constructed incorrectly hypotheses are not subjected
to this procedure [19, pp. 347–348].
Development a synthesis of human knowledge without a part of science and meta-
physics is impossible. Both the scientific terms and the terms of the scientific metaphysics,
being under control of experience, are used in the construction of the synthetic view of
the knowledge following Zawirski. He defends the scientific metaphysics with terms
being under the control of experience and all its theses should fall within a category
of testing including their falsification. This range of metaphysics should include the
notion of time and space structure plus dynamic reality and the problem of determinism,
indeterminism or deliberation [19, p. 348]. In the development of the system for the
scientific metaphysics, Zawirski does not eliminate the intuitive learning. However,
should the doubts appear or even contradictions of the intuitive details, the intuitive
convictions should be consolidated in the axiomatic system. Only at that time one can be
sure what s/he wants to say plus what is said will be understood by the others. A depiction
of convictions into a system, which includes the appropriate order of sentences together
with their consequences, never releases us from a duty of searching for equivalences in an
experience for the particular conviction. However, the developed system must be reviewed
as well as falsification [19, p. 348]. This approach into developing a system is nothing
new. Zawirski notices that in the process of building even the simplest deductive system
we deal with the intuitive acceptance of its assumptions.
It is worth mentioning that Zawirski does not remove neither a priori or intuitive
elements in the attempt to understand the reality. He demands a depiction od all details
first of all into a system, which consequently should be confronted with an experience.
In Zawirski’s opinion many such systems can be constructed and each of them can be
convincing at the equally strong level of obviousness. Zawirski notices that the same set
of intuitive details can in the equal degree justify many of the accepted theories. The
final decision for which theoretical description of the intuitive details one should support
depends on an experience.
The scientific knowledge as well as the scientific metaphysics, in Zawirski’s opinion

is deprived of Leibniz’s illusions, relating to the possibility of settlement for a scientific
problem and simultaneously a reasonable one. It can be accepted only in the case of the
complete systems that a settlement and sense can coincide with each other. However, one
can experience more often the incomplete systems. On the account of this, investigating
this type of systems one should consider Gödel’s achievements. Still, it should be taken
into consideration that within the rich systems, which include arithmetic’s, it is possible
to construct such sentences that are based on this system and refer to its method legally
valid and accepted. These sentences can be neither proved or invalidated. A settlement
of the sentences will still take place in the case of enriching a language of the system
with the variables of the upper level for logical types. Gödel’s achievement reveals a
perfection of the modern logics and the results of the research from the field of semantics
in the development of language for the system which unifies our knowledge. Following
Zawirski, we reach a learning situation when the traditional epistemology must revive
through the consideration of research from the semantics and metalogics The Importance
of Logical and Semantical Investigations for the Theories of Contemporary Physics [21,
pp. 25–30]. A similar modernisation should be applied to the traditional metaphysics,
which is to form a tighter link with the achievements of the detailed sciences [19, p. 349].
Here one should pay attention to the scientific achievements of Alfred Tarski and
Kazimierz Ajdukiewicz, who attempted to bring the notions of the theory of cognition
to the sematic notions. In his work entitled The Notion of Truth in the Languages of
the Deductive Sciences [16], Alfred Tarski demonstrated a possibility of developing a
precise and formally correct definition of the right sentence on the basis of the deductive
languages, following a classical perception of the truth. As a result of this cognitive
detail it became clear that a scope of the term “true sentence” depends on a language,
which is the subject of the particular investigation. One can not discuss the meaning
of phrases separately from a language where these phrases occur while Ajdukiewicz
pointed at the existence of the untranslable. This situation relates to the closed ad coherent
languages. It does not refer to the fact that one of them could be too poor [1]. Ajdukiewicz
states that the language of the classical and relativistic physics is the example of the
untranslatable languages. Whereas in Zawirski’s previous theory of physics analysis,
following Reichenbach, he assumed that the statements of the classical physics are
the border case of the statements attributed to the modern physics. In the perspective
of Ajdukiewicz’s achievements, this view was taken as a completely wrong one. In
Zawirski’s opinion, the revealed untranslatability of languages can be a derivative from a
differentiation of logics founded by the respective languages. Systems with such qualities
of logics are known and one system cannot be translated into the other. Then, a system
should be found which could respond to reality [18, pp. 41–42].
3.1 Final Conclusions
Zawirski focused on the investigations of methodological and ontological assumptions in

the attempts of developing the scientific metaphysics In the first period of the scientific
work. The latter part, investigating the matter he worked on the issues of methodology
and semantics.
Zawirski listed the possibilities of formalising the scientific nature theories in the
area of the research on studying the scientific metaphysics, the manners of investigating
the assumptions and philosophical implications derived from the contemporary physical
theories, a critical analysis of these scientific terms including their role in the scientific
progress as well as the use of the methodological principles in the scientific theories.
He pointed at the possibility of studying the scientific metaphysics with the use of the
axiomatic method and the possibility of philosophical generalisation of the scientific
research in order to understand the world. He looked for the justification of the scientific
research, listing the methodological problem of induction and deduction, development of
the scientific theories and relations of science and philosophy.
Zawirski analysing the possibility of describing the scientific metaphysics achieved
many valuable results out of the investigations. He stressed the importance of the
ontological research related to time and space. On this account, one cannot omit
his important work entitled a L’évolution de la notion du temps, where we can find
many issues which still occur in the scientific discussions [20]. Namely they include
thermodynamics of irreversible processes, the second law of thermodynamics, theory of
dissipative processes. It should be underlined that the problems undertaken in L’évolution
de la notion du temps are discussed by I. Prigogin, I. Stengers (Order out of Chaos, [13])
and Roger Penrose (Cycles of Time, [11]).
Zawirski’s research are very important for the description of the scientific metaphysics
as they relate to the implementation of the formal logics into the natural analyses of
the scientific theories. The problems were developed particularly by his student Roman
Suszko (1918–1979), who developed the theory of models and implemented it for the
investigations of problems outside mathematics (Remarks on Sentential Logics—1958).
A possibility of defying the scientific metaphysics resulted also in the critical and
historical analyses of the scientific terms and philosophical reflection over the assump-
tions of empirical theories [5, 6, 9, 14]. It also led into the investigation of traditionally
philosophical problems entangled into natural and mathematical theories. It can be noticed
that the Cracow background to which Zawirski was linked in the middle of the twentieth
century, continues the research of the notions previously taken by him from the field of
natural philosophy, natural history and philosophy of nature. A special attention should
be paid to the works of such philosophers as Józef Życiński (1948–2011) or Michał
Heller (born 1936), a director and founder of The Copernicus Centre for Interdisciplinary
Studies.
References
1. Ajdukiewicz, K.: J˛ezyk i poznanie. Wybór pism z lat 1920–1939 [Language and Cognition. A
Selection of Papers from the Period Between 1920 and 1939], vol. I. PWN – Polish Scientific
2. Bornstein, B.: Architektonika świata [Architectonic of the World], vols. I–III. Gebethner & Wolff,
Warszawa (1934–1936)
3. Heinrich, W.: Filozofia i jej zadania [Philosophy and its tasks]. Kwartalnik Filozoficzny [Philos. Q.]
1, 1–18 (1922)
4. Heinrich, W.: O zagadnieniach podstawowych filozofji [On the basic philosophical terms]. Kwartalnik
Filozoficzny 2, 269–302 (1924)
5. Heinrich, W.: Teorya poznania [Theory of knowledge]. In: Maczka, J., Janusz, R., Senderecka, M.
(eds.). Biblos, Kraków-Tarnów (2005)
6. Kotarbiński, T.: W sprawie poj˛ecia prawdy [On the matter of the truth notion]. Przeglad ˛ Filozoficzny
[Philos. Rev.] 37, 85–91 (1934)
7. Łukasiewicz, J.: O metod˛e w filozofji [On method in philosophy]. Przeglad ˛ Filozoficzny 31, 3–4
(1928)
8. Łukasiewicz, J.: Co dała filozofii współczesna logika matematyczna? [What was given to philosophy
by the modern mathematical logic?]. Przeglad ˛ Filozoficzny 39, 325 (1936)
9. Mehlberg, H.: La théorie causale du temps. Lwów (1935)
10. Penrose, R.: Shadows of the Mind: A Search for the Missing Science of Consciousness. Oxford
University Press, Oxford (1994)
11. Penrose, R.: Cycles of Time. An Extraordynary New View of the Universe. Bodley Head, London
(2010)
12. Popper, K.R.: Objective Knowledge. Clarendon Press, Oxford (1972)
13. Prigogine, I., Stengers, I. (eds.): Order Out of Chaos: Man’s New Dialogue with Nature. Flamingo,
London (1984)
14. Śleziński, K.: Dyskusje nad popperyzmem w filozofii polskiej pierwszej połowy XX wieku [Discus-
sions about Popper’s ideas in Polish philosophy in the first part of the 20th century]. Studia z Filozofii
Polskiej 1, 243–259 (2006)
15. Śleziński, K.: Benedykta Bornsteina koncepja naukowej metafizyki i jej znaczenie dla badań
współczesnych [Benedykt Bornstein’s Concept of Scientific Metaphysics and Its Meaning for
Contemporary Research]. Wydawnictwo Uniwersytetu Ślaskiego,˛ Kraków-Katowice (2009)
16. Tarski, A.: Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych [The notion of the truth in the lan-
guages of deductive sciences]. Prace Towarzystwa Naukowego Warszawskiego, Wydział III Nauk
Matematyczno-Fizycznych 34, 1–116 (1933)
17. Zawirski, Z.: Metoda aksjomatyczna a przyrodoznawstwo [The Axiomatic Method and Natural
Sciences]. Kwartalnik Filozoficzny, Kraków (1923)
18. Zawirski, Z.: Próby zastosowania logiki wielowartościowej do współczesnego przyrodoznawstwa
[Attempts for using many-valued logics in the modern natural sciences]. Sprawozdania Poznańskiego
Towarzystwa Przyjaciół Nauk 5, 41–42 (1931)
19. Zawirski, Z.: W sprawie syntezy naukowej [On the matter of scientific synthesis]. Przeglad ˛ Filo-
zoficzny 39, 347–352 (1936)
20. Zawirski, Z.: Rozwój poj˛ecia czasu [The Evolution of the Notion of Time]. Uniwersytet Jagielloński,
Kraków (1936)
21. Zawirski, Z.: Doniosłość badań logicznych i semantycznych dla teorii fizyki współczesnej [The
importance of logical and semantical investigations for the theories of contemporary]. Przeglad ˛
22. Zawirski, Z.: Nauka i metafizyka (I) [Science and metaphysics I]. Filozofia Nauki [Philos. Sci.] 3,
104–135 (1995)
23. Zawirski, Z.: Nauka i metafizyka (II) [Science and metaphysics II]. Filozofia Nauki 4, 131–143 (1996)
24. Zawirski, Z.: O stosunku metafizyki do nauki [On the Relation Between Metaphysics and Science].
Wydział Filozofii i Socjologii Uniwersytetu Warszawskiego, Warszawa (2003)
25. Znaniecki, F.: Zadania syntezy filozoficznej [Tasks of philosophical synthesis]. Przeglad ˛ Filozoficzny
30, 103–116 (1927)
K. Śleziński ()
Institute of Educational Science in Cieszyn, University of Silesia in Katowice, Cieszyn, Poland
Stanisław Leśniewski: Original
and Uncompromising Logical Genius
Peter Simons
Leśniewski was notable for the degree of prolixity which he was willing
to admit in the interest of complete rigor and precision.
W.V. Quine, The Journal of Symbolic Logic 5, (1940), p. 83
Abstract Stanisław Leśniewski was one of the two originators and drivers of the
Warsaw School of logic. This article describes his work chronologically, from his early
philosophical work in Lvov to his highly original logical systems of protothetic, ontology
and mereology. His struggles to overcome logical antinomies, his absolute commitment
to logical clarity and precision, and his antipathy towards set theory made his nominalistic
approach to logic among the most original of the twentieth century, while his early death
and the loss of his papers meant his work was only gradually discovered and appreciated
outside Poland.
Keywords Stanisław Leśniewski · Lvov-Warsaw School · Protothetic · Ontology ·

Mereology · Antinomies
Mathematics Subject Classification (2000) Primary 01A70, 03A05; Secondary 03B05,

03B15, 03B99
1 Life
Stanisław Kazimierz Leśniewski was born in Serpukhov, near Moscow in Russia, on

28 March 1886. His father, Izydor Leśniewski, was an engineer who worked on the
construction of the Trans-Siberian Railway. His mother was Helena, née Palczewska.
Stanisław was baptised at St. Stanislav’s church in St. Petersburg. His mother died when
he was young and his father remarried, so Stanisław had several younger step-siblings.
He attended classical Gymnasium in Irkutsk in Siberia from 1899 to 1903. Between
1904 and 1910 he studied Philosophy and Mathematics at several universities: Leipzig,
Zurich, Heidelberg, St. Petersburg, and Munich. In Leipzig, he heard lectures by Wilhelm
Wundt, and in Munich by Hans Cornelius, Moritz Geiger and Alexander Pfänder. In
1910, already with a solid philosophical education behind him, he went as a doctoral

210 P. Simons
student to Lvov University, then in Austria-Hungary, completing his doctorate under

Kazimierz Twardowski in 1912 with a dissertation entitled Przyczynek do analizy zdań
egzystencjalnych [Contributions to the Analysis of Existential Propositions] [1]. This was
also published as an article the previous year in the leading Polish-language philosophy
journal Przeglad˛ Filozoficzny.
In 1913 he married Zofia Prewysz-Kwintów (1893–1958), from a Lithuanian landown-
ing family. They had no children. Four further publications on the philosophy of logic
followed until the outbreak of war, which Leśniewski spent largely in Moscow teaching
mathematics at Polish schools. During this period, he developed what he called his
theory of collections, which later became known as mereology. After the Bolshevik
Revolution in Russia, Leśniewski left Russia for Poland. During the 1919–1921 Polish-
Bolshevik War he worked as a codebreaker for the Polish General Staff’s Cipher Bureau.
He tried unsuccessfully to obtain his habilitation in Lvov, but got it instead in 1918
in the University of Warsaw, where in 1919 he became Extraordinary Professor of the
Foundations of Mathematics, a position especially created for him. From then until his
final illness Leśniewski lectured regularly on logical and mathematical topics, and built
up his logical systems. Leśniewski published nothing between 1916 and 1927. Between
1927 and 1931 he published two major pieces, one a series in Polish on mereology, the
other a long article in German on protothetic, his extended propositional logic. Leśniewski
was made a full professor in 1936. He contracted thyroid cancer and died on 13 May 1939
at the age of 53 after an unsuccessful operation. He was buried in the historic Old Powazki
˛
Cemetery in Warsaw. His papers, entrusted to his student Bolesław Sobociński, included
unfinished works on logical antinomies and on many-valued logic. All of these papers
were destroyed in Sobociński’s apartment during the Warsaw Rising of 1944.
2 Phases of Activity
It is common to divide Leśniewski’s work into two phases: an early one (up to 1916) in
which he worked informally on questions of the philosophy of logic, and a mature phase
(from roughly 1919 to his death) in which he formulated and perfected his formal logical
systems. Leśniewski himself gave his imprimatur to the idea that his early work should
be clearly separated from his later, since in 1927 he wrote that he formally repudiated all
his work before 1914, which he regarded as bankrupt: “philosophical”-grammatical, in
his words, with the exception of an argument against Twardowski’s conception of general
objects, which he still regarded as acceptable.
However, while from the point of view of content there is a clear difference between
the work before 1914 and that after, it is something of an oversimplification to divide his
activity into just two periods. True, the early papers between 1911 and 1913 are written
from the “philosophical”-grammatical point of view that Leśniewski came to despise.
Between 1914 and 1916 however, while Leśniewski continued to write in regimented
prose, his interests centred around finding a solution to the problem of Russell’s Paradox,
and mark a transition to his later interests and style. The long period without publication
is when Leśniewski was formulating and perfecting his systems. Between 1927 and 1931
there was a burst of publication. Then there was another pause in publishing, when
Stanisław Leśniewski: Original and Uncompromising Logical Genius 211
Leśniewski probably concentrated on writing a large systematic work on antinomies, and

on criticising many-valued logics. The year before Leśniewski died, a continuation of the
German paper appeared.
So it would be more accurate to say there were five phases: an initial, “philosophical”
phase, a transitional phase where Leśniewski was working out how to deal with Russell’s
Paradox, an intensive developmental phase where his mature systems were formulated
and results proved without publication, a concerted publishing phase, with this leading
into a second phase of consolidation without publication, the final paper being really an
artificially delayed postcript to the earlier burst of publishing. Had Leśniewski survived
longer there would doubtless have been another period of publication.
Continuity from 1919 to 1939 was provided both by Leśniewski’s efforts to formulate
and perfect his systems, and by his unbroken teaching in the University of Warsaw, which
centred firmly around the foundations of mathematics and how to approach them properly,
being the place where he could try out his ideas and perfect them with the collaboration
of students.
3 Early Writings
During his studies in Germany, Leśniewski encountered the philosophy of language of

Anton Marty, whose major work Untersuchungen zur Grundlegung der allgemeinen
Grammatik und Sprachphilosophie (1908), he contemplated translating into Polish.
Legend has it that the project foundered after he was unable to find an adequate
translation for the second word of the title. It is likely he also encountered Husserl’s
Logische Untersuchungen (1900/1901), and again thought of translating this (somewhat
later: 1915/1916), again without outcome. What he later called this “Austrian” approach
influenced his early work. This concentrated not on grammar or philosophy of language
per se but on aspects of the philosophy of logic. The dissertation [1] is dedicated to
showing, using notions of denotation and connotation derived from Mill, that existential
statements of the form S exists are tautological—and false. This barely believable thesis
is slightly softened by his claim that the seemingly equivalent sentence Some object is S
is both non-tautological and may be true. This kind of fine linguistic differentiation was
to become the hallmark of his work.
Other writings of this kind from this period were in good part reactions to work by
others, but always with an original slant. The paper ‘Próba dowodu ontologicznej zasady
sprzeczności’ [Attempt at a Proof of the Ontological Principle of Contradiction] (1913)
[2] is a reaction to aspects of Jan Łukasiewicz’s 1910 monograph O zasadzie sprzeczności
u Arystotelesa [On the Principle of Contradiction in Aristotle], a work that was to have far-
reaching consequences for Leśniewski’s own career. The article ‘Czy prawda jest tylko
wieczna czy też wieczna i odwieczna?’ [Is truth only eternal or both eternal and from
eternity?] (1913) [3] criticises an argument by Leśniewski’s friend and fellow-student
Tadeusz Kotarbiński that statements about future contingents are not true in advance of
the outcome, a view later taken up by Łukasiewicz as the motivation for his invention of
many-valued logic. Leśniewski upholds the view that statements about future contingents
are true or false in advance of the event, and his argument persuaded Kotarbiński to change
212 P. Simons
his mind. The most substantial of Leśniewski’s pre-war pieces was ‘Krytyka logicznej
zasady wyłaczonego
˛ środk[a]’ [Critique of the Logical Principle of Excluded Middle]
(1913) [4], again taking on ideas of Łukasiewicz and Twardowski. It contains, among
other things, an argument against Twardowski’s conception of a general object. Suppose
a general object—say the general tomato—is defined as the object having all and only
the properties common to all tomatoes. Suppose one tomato weighs 150 g and another
tomato weighs 100 g. Then the general tomato has neither the property of weighing 150
g nor the property of not weighing 150 g. Thus, the only way a general object can be
non-contradictory is if it is the sole object of its kind, which renders the notion useless.
This is the only argument of Leśniewski’s early work that he did not “repudiate” in 1927.
Another part of the paper discusses the Liar Paradox or Epimenides, the first indication of
Leśniewski’s long battle against antinomies.
In general, Leśniewski’s pre-war papers are characterized by extreme attention to
language and how he intends certain words and sentences to be understood, which is not
always in the same way as others, and a steely resolve to follow the argument wherever
it leads, even if the resulting theses are sometimes counterintuitive. At the same time, his
verbal commentary is often colourful and he makes abundant use of scare-quotes. These
characteristics were to remain with him throughout his writing career.
4 Antinomies, Classes, Parts
In Łukasiewicz’s 1910 monograph on the principle of contradiction in Aristotle, there

is an appendix discussing the treatment of contradiction in symbolic logic, employing
the notation of Louis Couturat. There is also a discussion of Russell’s Paradox of the
class of classes which are not members of themselves. Leśniewski read the book in 1911
and was so struck by trying to solve the paradox, which he initially thought was just a
trivial puzzle that could be solved in a couple of hours, that he is said to have missed
a railway connection from Poland to Russia and endured a long wait. In fact he worked
on the problem for 11 years, and never lost his fascination with antinomies. His first
published foray into the field was his 1914 paper ‘Czy klasa klas, nie podporzadkowanych
˛
sobie, jest podporzadkowana
˛ sobie?’ [Is the class of classes not subordinate to themselves
subordinate to itself?] [5]. The idiosyncratic formulation ‘subordinate to itself’ rather than
‘member of itself’ is taken from Łukasiewicz’s discussion. While it makes some relatively
unimportant assumptions that Leśniewski would later have criticised, the main thrust of
the article is that there is no antinomy, because every class (every collection having at least
one member) is a member of itself. The reason for this is that Leśniewski understands
‘member of A’ in effect to mean ‘part (or whole) of A’.
This “solution” was expounded in much greater detail in a forty-page monograph,
Podstawy ogólnej teoryi mnogości I [Foundations of the General Theory of Collections
I] (1916: no further parts appeared) [6] dedicated by Leśniewski to his wife. Consider a
line segment AB, and suppose it is composed of subsegments AC, CD, DE, EB, and that
the several segments AC, CD, DE and EB are the objects m. Then the class of (all the) m
just is the whole object AB, an object that we would now call the mereological sum of the
objects m. As is clear from Leśniewski’s explanations of his terminology, he considered
this to be the only sensible way to understand the notions of ‘class of m’ (the sum of all
the m), ‘set of m’ (a sum of some but not necessarily all the m), and ‘element of m’ and
‘subset of m’ (a proper or improper part of the class of the m). The monograph declares
this to give the true understanding of ‘set’ (Menge) in Cantor, and it is in effect a bid
to take over the terminology of the then exploding discipline of set theory and turn it in
Leśniewski’s direction. That his understanding is clearly not that of Cantor and other set
theorists is shown by the fact that there are four objects m but many more (including AB
and all its parts) that are elements of m, and that a single object such as AB can be the
class of many distinct groups of objects n, even such as have none in common with the m,
all of which runs counter to Cantor, and standard set theory.
The basic notion of the monograph is that of ‘part’ (cz˛eść), meaning a proper part,
not identical with the whole—Leśniewski’s term for a proper-or-improper part is here
‘ingredient’ (ingredyens), but he later preferred the term ‘element’ for this notion (they
are defined differently but are equivalent in the monograph). The development is based on
four axioms and three definitions, those for ingredient, set and class. The axioms say that
parthood is asymmetric and transitive, and that if there is at least object m then there is
exactly one class of m. Leśniewski later criticised this theory for including definitions at
the basic level, and would afterwards always attempt to base his theories on a single notion
governed by axioms, before introducing definitions and theorems. On the other hand,
the development in the monograph is extremely clearly done and there is very little to
criticise from Leśniewski’s later formal point of view. The theorems are stated and proved
in ordinary Polish augmented with variables, and employing a very restricted vocabulary.
They were thus very easy to formalize later.
Leśniewski’s concretistic, mereological understanding of the ideas of set theory was
destined not to catch on, and led in the 1920s and 1930s to increasingly critical and
ultimately personally bitter disagreements with the proponents of what Leśniewski
ironically called “official” set theory. To distinguish his views from theirs, in 1927 he
renamed his theory ‘mereology’ [mereologia], from the Greek meros, part. That is the
name that has become standard for formal part-whole theory ever since, and the 1916
monograph was the first formal theory of part and whole to appear in print.
5 Formalization
Upon his appointment to the professorship in Warsaw in 1919, Leśniewski embarked

on a 20-year teaching career, the bulk of which was concerned with various aspects
of the foundations of mathematics. It was Leśniewski’s aim and ambition to provide a
rigorous and antinomy-free foundation for mathematics, avoiding the inconsistency of
Frege’s logical foundation in Grundgesetze der Mathematik (1893/1903), the inexactness
of Whitehead and Russell’s foundation in Principia Mathematica (1910–1913), as well
as the—to Leśniewski—absurd and counterintuitive set theory developed by Zermelo and
others. This was now his life’s project.
The initial hurdle to this was the example of Whitehead and Russell, who were
notoriously careless about use and mention in the prose passages of Principia, a fact
which prevented the extremely literal and unsympathetic reader Leśniewski from even
understanding their work, and made him shy away from symbolic logic for several
years, on the assumption that symbolization itself was problematic. Two factors changed
214 P. Simons
his mind. One was his admiration for the metalogical care and clarity of Frege,
notwithstanding the inconsistency of his system, and the other was personal persuasion
by Leon Chwistek in 1920.
Leśniewski then set about putting his system into a formal and symbolic guise, starting
with the mereology from 1916, which he had continued augmenting and improving
in the intervening years, without publication. The way in which Leśniewski came to
formalization from already precisely formulated vernacular formulations was somewhat
unusual, and explains his attitude to formal logic. It is nowadays generally assumed,
in the vein of Hilbert and formalism, that a symbolic system consists of a collection
of symbols assembled into formulas by syntactic rules and conventions, given a proof
theory by inferences rules and axioms, and endowed with meaning by a formal semantics,
often though not always drawing on set theory. In the early twentieth century, several
prominent logicians, notable Frege, Whitehead, Russell and Brouwer, took a different
view, assuming that the symbols they employed in their logic had a determinate, intuitive
meaning, and that axioms and rules were teasing out the logical effects of these intuitive
meanings. Leśniewski’s sympathies lay with the latter group, since the symbols he
started employing from 1920 already inherited their meaning from his prior vernacular
formulations and did not need to have it conferred from outside. The point of the axioms
and rules was to capture as adequately as possible for logical purposes what these
meanings were. This is why, in later writing, Leśniewski ironically describes himself as
both a formalist and an intuitionist.
Having dealt with mereology, Leśniewski next turned to the logical apparatus under-
lying it, in the first place, the logic of names and predicates. Unlike Frege and Russell,
Leśniewski was quite happy to allow names to denote more than one individual, following
in this regard rather Aristotle, the tradition, and the algebraic logic of Ernst Schröder. He
also accepted that a name could be empty, that is, fail to denote anything. The basic notion
that he identified as requiring axiomatic determination was that of singular inclusion, in
the form ‘A is (a) b’, for example, ‘Socrates is a man’. In Polish, which lacks articles, this
is Sokrates jest człowiekiem, so logical attention focusses on the single word ‘jest’. He
then collected some logical truths he considered it to govern, for example ‘If A is b, then
A is A’, and ‘If A is B, and B is c, then A is c’. The use of lower and upper case variables
is not logically significant, but was merely an informal convention he adopted, taken over
from his early writings, that in a subformula which could only be true if the subject was
singular, he would use a capital letter for the subject term. For a symbol, Leśniewski took
over the Greek lower-case epsilon, ‘ε’, used by Peano and taken from the Greek esti, [is].
In 1920 he managed to come up with a single axiom governing ‘is’ or ‘ε’. Following
ideas from Russell’s theory of descriptions, according to which ‘(The) A is a b’ means
‘There is at least one A, and there is not more than one A, and every A is a b’, this was
the axiom
For all A and a: Aεa if and only if: for some B, BεA, and for all B and C, if BεA and
CεA, then BεC, and for all B, if BεA then Bεa.
Because this primitive enabled Leśniewski to define several other expressions involving
uses of the verb to be, such as those expressing existence, singular identity, inclusion, and
general identity, he came to call the resulting logical system ontology. The 1920 axiom
was subsequently replaced by shorter ones, and so came to be called the ‘long’ axiom of
ontology, but it remains the most intuitively evident, and embodies a kind of self-definition
of ‘ε’.
It remained for Leśniewski to formalize the logic of propositional connectives and

quantifiers presupposed in ontology and thus in mereology. This proved a little tougher.
Leśniewski’s view on definitions was that they should be expressed as equivalences, using
the connective ↔ of equivalence, rather than as metalogical abbreviations. He thus wished
to base his logic of propositions on material equivalence and universal quantification
alone, but could not see how to define conjunction. It was his genial young PhD student
Alfred Tarski who found the solution, defining conjunction by quantifying propositional
functors, as
p ∧ q ↔ ∀f (p ↔ (f (p) ↔ f (q)))
so that, by 1923, Leśniewski had a formal system of propositions, propositional functors

and quantifiers, which he dubbed protothetic. The full system of foundations now
comprised protothetic, ontology and mereology, in that order of logical precedence.
6 Consolidation and Publication
Through the early and mid-1920s, Leśniewski did not publish his results, preferring to
work on improving them and aspiring to publish a systematic treatise in the style of
Principia Mathematica. But while he worked on this and discussed his results with
others, he increasingly found that results obtained by himself or one or other of his
interlocutors were being held back for fear of disputes about priority and responsibility for
advances—both Leśniewski and several of his colleagues were notoriously sensitive about
attributions of priority. To relieve the difficulty, he decided on a different way of getting
his ideas into print, which was a quasi-autobiographical sequence of results in the order he
and others had found them. The first result of this was in effect a treatise in eleven sections,
‘O podstawach matematyki’ [On the foundations of mathematics], published in five parts
in Przeglad˛ Filozoficzny between 1927 and 1931 [7]. It was affectionately dedicated
to “My esteemed and beloved Professor of Philosophy, Dr. Kazimierz Twardowski”,
from “a philosophical apostate, but a grateful pupil.” The paper was a biography of
mereology, from 1916 onwards. Like the original, it was expressed in regimented Polish,
with a somewhat revised vocabulary, and contained detailed comparisons between the
various systems and proofs of over 250 theorems. It also contained a comparison with
and criticism of Whitehead’s theory of events, to which Tarski had drawn Leśniewski’s
attention. While Whitehead also used part-whole theory, and had probably developed his
ideas at about the same time as Leśniewski, his published formulation was flawed, and
Leśniewski mercilessly pointed out the defects.
The other major publication of this time was a long 1929 article, ‘Grundzüge eines
neuen Systems der Grundlagen der Mathematik’ [Fundamentals of a New System of
the Foundations of Mathematics] [8], published in Fundamenta Mathematicae, also
in eleven sections. After a further “autobiographical” account of the development of
protothetic, this continued with detailed metalogical specifications, but the development
was incomplete at the end of Sect. 11 and it was another 9 years before the series was
continued.
216 P. Simons
In this period Leśniewski also published four short articles, two on single-axiom
axiomatizations of group theory and Abelian group theory [9, 10], a short sketch of the
metalogic of ontology [11], and an account of principles of definition in propositional
logic [12].
7 Final Years
From 1931, Leśniewski’s publications faltered again, this time in part because he had
fallen out with the other editors of Fundamenta Mathematicae over their continued prac-
tice of promoting set theory through the journal. A chance to continue the ‘Grundzüge’
article only came towards the end of the decade, when a new journal, Collecteana
Logica, was founded, and Leśniewski wrote a long introduction to the continuation noting
developments in the intervening years [13], and following it with the completely formal
twelfth section, comprising no fewer than 422 theses. But the outbreak of war meant that
the journal never appeared. A preprint of the article survived in Harvard, perhaps sent to
Quine who then wrote a review for the Journal of Symbolic Logic in 1940. Leśniewski
was somewhat belatedly promoted to full professor in 1936. At the time of his death he
was working on a refutation of many-valued logic, as well as a treatise on antinomies.
Neither survived the Warsaw Rising, being lost with his papers and correspondence.
8 Metalogic
Because of his antipathy to set theory, which became the medium of choice for logical
semantics, Leśniewski never ventured into semantics, believing it unnecessary for systems
whose constants are already meaningful. It was also probably in good part because his
former student Tarski employed set theory in his famous paper on truth of 1933 that
Leśniewski did not accept Tarski’s results, though in the early part of the truth paper
Tarski paid generous tribute to Leśniewski for the analysis of the Liar Paradox and the
way quotation mark names are used. In his ontological views, Leśniewski was anti-
platonistic, and only declined to call himself a nominalist because he thought that some
mental phenomena such as after-images defied physicalistic explanation. On the other
hand, in his practice of logic, Leśniewski was uncompromisingly nominalistic. In his
view, a logical system is a concrete, spatiotemporal collection of marks or inscriptions
which can be added to over time by new marks constituting definitions and proofs of
theorems. This is in complete contrast to platonistic views, according to which a logic is
a system of eternal propositions standing in timeless logical relationships.
Because of this very down-to-earth view of logic, Leśniewski was unable to formulate
metalogical principles in what has become the standard way, by a recursive specification,
but gave detailed schematic directives on how to extend an existing logical system
with new theses. Because the theses are not set out in Plato’s heaven, the descriptions
of what may count as an acceptable continuation need to be self-adjusting, referring
back to expressions already introduced in previous theses. The directives for protothetic
include those for substitution, detachment, quantifier distribution, and very importantly,
for definitions. Definitions, being concocted like the rest of the system as one went along,
were regarded by Leśniewski as object-language equivalences introducing new symbols,
effectively new axioms, and they could be creative and non-conservative. Formulating
the principles governing what could count as a good definition gave Leśniewski much
trouble, and he regarded his directives for adding definitions to protothetic and ontology
as his finest work. In protothetic there was only one kind of definition, for sentences and
functors, while for ontology there were two kinds, one for adding new sentence-forming
functors (predicates of first and higher order), and one for adding names and nominal
functors of first and higher order. The directives were expressed in a highly precise way
employing a complex regimented vocabulary specified by a series of what Leśniewski
called terminological explanations, but which were in effect precise stipulations for the
meanings of his metalogical terms.
In the early 1920s, Leśniewski employed a version of Whitehead and Russell’s type
theory, but he was wary of its apparently inflated ontology and soon reformulated his
logical grammar as what was subsequently known as a categorial grammar, according
to which any expression is either a sentence or a name or a functor, with precise
argument input kinds and output kind, each such kind being what Leśniewski somehat
misleading called a semantic category. His inspiration for this was Husserl’s theory
of Bedeutungskategorien, though unlike Husserl he understood the categories to be
meaningful expressions rather than meanings, and the constraints to be syntactical.
The combinatory principles for such a grammar were first set out not by Leśniewski
but by Kazimierz Ajdukiewicz. Ajdukiewicz was unable to give satisfactory principles
for quantifiers or other variable-binding operators. This mirrored a difficulty faced by
Leśniewski, who treated the universal quantifier, binding any finite number of variables
from any available categories, as itself syncategorematic, and found himself unable to
give directives for adding new binding operators, despite offering students any degree they
needed if they could come up with a solution. Leśniewski found his theory of semantic
categories to be so natural that he considered a logical system should use it even if there
were no threatening antinomies. Tarski initially followed Leśniewski’s theory of semantic
categories but his later readiness to go beyond the system and allow transfinite types or
categories was no doubt another contributory factor to their disagreement.
When Quine met Leśniewski in Warsaw in 1933 they disagreed about whether
Leśniewski’s liberal practice of allowing quantifiers to bind variables of any category
committed him to a correspondingly complex ontology of abstract entities: Quine thought
it did; Leśniewski maintained it did not.
9 Teaching and Students
Leśniewski taught in Warsaw from 1919 to 1939, mainly on his own theories, occasionally
on work by others, such as Cantor, Zermelo, Peano and Łukasiewicz. Some of his courses
extended over 2 academic years, and he used them to trial his own ideas. Lejewski
reported that his advanced seminars on directives lasted for three semesters. His style
was to work from copious notes, writing theses on the board and asking students for
assistance in formulating examples and counterexamples. Quine found it easy to follow
218 P. Simons
despite knowing no Polish. By their nature, his classes did not attract a wide following,
and Leśniewski would quietly send away unknowing students, who were just making up
their hours, with a positive mark in their student book. Once when unexpectedly many
students turned up at the beginning of the class he expressed surprise at their numbers and
asked whether they had confused him with Bergson. Those who stayed were dedicated.
Leśniewski only ever supervised one doctoral student, Alfred Tarski, who obtained his
doctorate in 1923 at the age of 22. Leśniewski would proudly say he had 100% geniuses
as doctoral students.
In teaching, as in some of his publications, Leśniewski conducted proofs by a system of
natural deduction, which was much easier to follow than full axiomatic proofs. Although
one of the first logicians to carry out proofs in this way, he regarded such derivations as
mere sketches for “proper” proofs, comparing them with a lounge suit worn in preference
to formal evening dress with a stiff shirt and collar. He therefore never formulated exact
principles for these derivations, and the impetus for doing so, work pioneered by Stanisław
Jaśkowski, came not from Leśniewski but from Łukasiewicz.
10 Colleagues and Personality
Leśniewski remained in various forms of contact with his teacher Twardowski and study
colleagues from pre-war Lvov, including Jan Łukasiewicz, Kazimerz Ajdukiewicz and
Tadeusz Kotarbiński. Obviously Łukasiewicz was a close colleague in Warsaw, while
Tarski soon joined his two teachers as the third major logician of the Warsaw School.
There were also his mathematical colleagues in Warsaw, notably Wacław Sierpiński and
Kazimierz Kuratowski. His relationships with most of them eventually deteriorated, for a
variety of reasons. As the obverse of his obsession with exactness, Leśniewski was a fierce
and unsympathetic critic of sloppy thinking and his interventions at seminars in Warsaw
put people off from going to speak there. His written criticisms of the ideas of gifted
mathematicians such as Hausdorff and von Neumann were often blunt to the point of rude-
ness, and gained him few friends. His political views shifted rightwards over the years:
he supported the authoritarian Sanacja regime inspired by Józef Piłsudksi, and became
increasingly anti-semitic in his sentiments. At the time of his death his only remaining
close friend was the patient and forgiving Kotarbiński, who visited him in hospital in his
final days and was with him when he died. On the other hand, he inspired fierce loyalty
and admiration among the small group of his closest students, and even his critics had to
admit he was one of the most gifted and original of the brilliant interwar generation.
11 Legacy
From the 1930s onwards, a number of factors conspired to keep Leśniewski’s work out of
the logical mainstream. In part it was his own meticulous but often inconvenient logical
practices. In part it was his ideological disagreement over set theory and other platonistic
mathematical theories, which meant that he did not participate in the semantic revolution
in logic initiated by Tarski. Finally, his tragically early death and the wartime destruction
of his papers meant that what passed down was a torso of his work. After the Second
World War, his students, particularly Słupecki, Sobociński and Lejewski, made efforts to
reconstruct his ideas, recalled from lectures and papers, but most logicians, not least the
now famous and influential Tarski, regarded his work as old-fashioned and superseded.
Isolated aspects of his work were taken up more widely, in particular mereology, rendered
in more conventional logical guise by Henry Leonard and Nelson Goodman, but for the
most part his work was regarded as a quaint and ultimately unfruitful branch of modern
logic, and largely forgotten. It did not help that most of his work was in Polish, and
difficult to access.
From the 1980s things improved somewhat. Translations of rescued lecture notes [14]
and English [15] and French [16] translations of his articles brought his primary work
to a wider audience. Mereology was instated as a crucial tool of metaphysics. A Polish
collected edition, Pisma Zebrane [17], a biography by Jacek Jadacki and a commentary
monograph by Rafał Urbaniak rounded out the published picture. As the historiography
of the Lvov-Warsaw School was written, Leśniewski re-emerged not only as one of the
chief drivers of that remarkable school, but more widely as one of the most original and
inspirational logicians of the twentieth century.
Stanisław Leśniewski’s Works Mentioned in the Paper
1. Przyczynek do analizy zdań egzystencjalnych. Przeglad ˛ Filozoficzny 14, 329–345 (1911) (Cf.
[Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 15–31. Semper, Warsaw (2015)];
English translation: A contribution to the analysis of existential propositions. In: [Surma, S.J.,
Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 1–19. Kluwer/Polish
Scientific Publishers, Dordrecht/Warsaw (1992), with an annotated bibliography to 1978 by V.F.
Rickey])
2. Próba dowodu ontologicznej zasady sprzeczności. Przeglad ˛ Filozoficzny 15, 202–226 (1912) (Cf.
[Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 32–56. Semper, Warsaw (2015)];
English translation: An attempt at a proof of the ontological principle of contradiction. In: [Surma,
S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 20–46.
Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw (1992), with an annotated bibliography to
1978 by V.F. Rickey])
3. Czy prawda jest wieczna, czy też wieczna i odwieczna? Nowe Tory 18, 493–528 (1913) (Cf.
[Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 147–182. Semper, Warsaw
(2015)]; English translation: Is all truth only true eternally, or is it also true without a beginning?
In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 86–
114. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw (1992), with an annotated bibliography
to 1978 by V.F. Rickey])
4. Krytyka logicznej zasady wyłaczonego
˛ środka. Przeglad˛ Filozoficzny 16, 315–352 (1913) (Cf.
(2015)]; English translation: The critique of the logical principle of the excluded middle. In: [Surma,
S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 47–85.
1978 by V.F. Rickey])
5. Czy klasa klas nie podporzadkowanych
˛ sobie jest podporzadkowana
˛ sobie? Przeglad˛ Filozoficzny
17, 63–75 (1914) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 223–235.
Semper, Warsaw (2015)]; English translation: Is the classes of classes not subordinated to themselves,
subordinated to itself? In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected
220 P. Simons
Works, 2 vols, pp. 115–128. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw (1992), with an
annotated bibliography to 1978 by V.F. Rickey])
6. Podstawy ogólnej teorii mnogości I. Prace Polskiego Koła Naukowego w Moskwie. Sekcya
matematyczno-przyrodnicza, No.2. Popławski, Moscow (1916). No further parts appeared (Cf.
(2015)]; English translation: Foundations of the general theory of sets. I. In: [Surma, S.J., Srzednicki,
J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 129–173. Kluwer/Polish
Scientific Publishers, Dordrecht/Warsaw (1992), with an annotated bibliography to 1978 by V.F.
Rickey])
7. O podstawach matematyki, I–V. Przeglad ˛ Filozoficzny 30, 164–206 (1927); 31, 261–291 (1928); 32,
60–101 (1929); 33, 77–105 (1930); 34, 142–170 (1931) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane.
[Collected Writings], 2 vols, pp. 295–468. Semper, Warsaw (2015)]; English translation: On the
foundations of mathematics. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.):
Collected Works, 2 vols, pp. 174–382. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw
(1992), with an annotated bibliography to 1978 by V.F. Rickey]; French translation: [Sur les
fondements de la mathématique. Hermès, Paris (1989)]
8. Grundzüge eines neuen System der Grundlagen der Mathematik. Fundamenta Mathematicae 14,
1–81 (1929) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 489–569.
Semper, Warsaw (2015)]; English translation: Fundamentals of a new system of the foundations of
mathematics. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2
vols, pp. 410–488. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw (1992), with an annotated
bibliography to 1978 by V.F. Rickey])
9. Über Funktionen, deren Felder Gruppen mit Rücksicht auf diese Funktionen sind. Fundamenta
Mathematicae 13, 319–332 (1929) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2
vols, pp. 475–488. Semper, Warsaw (2015)]; English translation: On functions whose fields, with
respect to these functions are groups. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F.
(eds.): Collected Works, 2 vols, pp. 383–398. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw
(1992), with an annotated bibliography to 1978 by V.F. Rickey])
10. Über Funktionen, deren Felder Abelsche Gruppen in bezug auf diese Funktionen sind. Fundamenta
Mathematicae 14, 242–251 (1929) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2
vols, pp. 714–723. Semper, Warsaw (2015)]; English translation: On functions whose fields, with
respect to these functions are abelian groups. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D.,
Rickey, V.F. (eds.): Collected Works, 2 vols, pp. 399–409. Kluwer/Polish Scientific Publishers,
Dordrecht/Warsaw (1992), with an annotated bibliography to 1978 by V.F. Rickey])
11. Über die Grundlagen der Ontologie. Sprawozdania z posiedzeń Towarzystwa Naukowego Warsza-
wskiego, Wydział III [Comtes rendus des séances de la Société des Sciences et des Lettres de
Varsovie, Classe III ] 22, 111–132 (1930) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected
Writings], 2 vols, pp. 724–745. Semper, Warsaw (2015)]; English translation: On the foundations
of ontology. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2
12. Über Definitionen in der sogenannten Theorie der Deduktion. Sprawozdania z posiedzeń To-
warzystwa Naukowego Warszawskiego 23, 289–309 (1930) (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane.
[Collected Writings], 2 vols, pp. 746–766. Semper, Warsaw (2015)]; English translation: On defini-
tions in the so-called theory of deduction. In: McCall, S. (ed.) Polish Logic, 1920–1939, pp. 170–187.
Clarendon, Oxford (1967). Cf. [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.):
Collected Works, 2 vols, pp. 629–648. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw
13. Einleitende Bemerkungen zur Fortsetzung meiner Mitteilung u. d. T. ‘Grundzüge eines neuen System
der Grundlagen der Mathematik’ [Introductory remarks to the continuation of my article ‘Grundzüge
eines neuen Systems der Grundlagen der Mathematik’]. Collectanea Logica 1, 1–60 (1938) (Cf.
(2015)]; English translation: Introductory remarks to the continuation of my article ‘Grundzüge eines
neuen Systems der Grundlagen der Mathematik’. In: McCall, S. (ed.) Polish Logic, 1920–1939,
pp. 116–169. Clarendon, Oxford (1967). Cf. [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F.
(eds.): Collected Works, 2 vols, pp. 649–710. Kluwer/Polish Scientific Publishers, Dordrecht/Warsaw
14. Grundzüge eines neuen System der Grundlagen der Mathematik (Fortsetzung). Collectanea Logica
1, 61–144 (Cf. [Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols, pp. 630–713.
Semper, Warsaw (2015)]; English translation: Fundamentals of a new system of the foundations of
mathematics. In: [Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2
15. Srzednicki, J.T.J., Stachniak, Z. (eds.): S. Lesniewski’s Lecture Notes in Logic. Kluwer, Dordrecht
(1988)
16. Surma, S.J., Srzednicki, J.T.J., Barnett, J.D., Rickey, V.F. (eds.): Collected Works, 2 vols.
1978 by V.F. Rickey
17. Sur les fondements de la mathématique. Hermès, Paris (1989). Translation of [O podstawach
matematyki, I–V. Przeglad ˛ Filozoficzny 30, 164–206 (1927); 31, 261–291 (1928); 32, 60–101 (1929);
33, 77–105 (1930); 34, 142–170 (1931)] by G. Kalinowski
18. Jadacki, J.J. (ed.): Pisma Zebrane. [Collected Writings], 2 vols. Semper, Warsaw (2015)
P. Simons ()
Department of Philosophy, Trinity College Dublin, Dublin 2, Ireland
Izydora Dambska:
˛ The First Lady
of the Twentieth-Century Polish Philosophy
Anna Brożek and Jacek Jadacki
The main thing is to remain in harmony with ourselves and with the
truth, not caring about the rest.
Letter to Maria Obercowa. January 19, 1951 (Dambska
˛ [5])
Abstract Izydora Dambska

˛ was one of the most creative women-representatives of the
Lvov-Warsaw School. The paper presents her extraordinary personality, life, as well as
the list of her main works and the greatest scientific achievements. The main area of
her interest was the logic of natural language, methodology and the history of Greek
semiotics. She gave meticulous analysis of the relation between conventionalism on one
hand, and relativism, scepticism and agnosticism on the other. In semiotics, she proposed
new approach to the problem of empty names and material implication, of the correct
definition of truth, as well as of pragmatic functions of silence and namelessness. In
methodology, her reconstruction of the notion of scientific laws is of great importance.
Keywords Conventionalism · Izydora Dambska

˛ · Lvov-Warsaw School ·
Methodology · Pragmatics · Scientific laws · Semiotics
Mathematics Subject Classification (2000) Primary 01A70, 03B42; Secondary 03–03
1 Personality
She was an aristocrat by birth as well as an aristocrat of the spirit and intellect. During
World War 2—a soldier of the Polish Underground Army; after the War, she adamantly
defended the highest human values and did not enter into any compromise with the
communist regime which was ruling Poland. Being the last assistant of Kazimierz
Twardowski, the founder of the Lvov-Warsaw School, at the same time, she was one
of prominent representatives of the School.
Her philosophical interests were very vast, and her philosophical oeuvre was very
rich and creative. This oeuvre included long monographs as well as small—but deep—
analytical miniatures.

224 A. Brożek and J. Jadacki
Concepts of sign, truth and understanding were the main subjects of her research in
semiotics. In ontology, she analyzed both historical theses (irrationalism, conventional-
ism, relativism, skepticism and agnosticism, determinism and indeterminism) and purely
theoretical problems (like forms and the value of instrumental cognition). In methodology,
she was occupied by the question of the logical status of scientific laws and of the nature
of reasoning by analogy. In light of her origin and attitude to life, it is not surprising that
she devoted a lot of her attention to the axiological problem of freedom.
Her work in theoretical philosophy was underpinned by a thorough knowledge of the
history of philosophy—especially ancient Greek philosophy.
Dambska’s
˛ method of philosophizing reflected trends prevailing in the Lvov-Warsaw
School: she never limited the domain of analysed problems in advance; she was very
careful in expressing her opinions and very critical, especially with respect to her own
writings.
She accepted—together with the other representatives of the Lvov-Warsaw School—
the old principle: initium doctrinae sit consideratio nominis. She accepted this in theory—
and successfully applied it in practice.
2 Life
She was born on January 3, 1904, in Lvov—and died on June 18, 1983, in Cracow.
Dambska
˛ studied philosophy at the University of Lvov under Twardowski’s guidance
(1922–1927); then she was his assistant (1926–1930). Dambska ˛ complemented her
studies in Austria, Germany and France (1930–1931). During the German and later the
Russian occupation of Lvov, she was a lecturer of the secret Polish university. After World
War 2, threatened with arrest by the Soviet secret service, she moved to Gdańsk. In 1946,
Dambska
˛ received her habilitation at the University of Warsaw, presenting the dissertation
Irrationalism and the scientific cognition [4]. From the years 1950–1956, and again from
1964 on, she was removed from the university by the communist regime due to political
reasons. But she continuously—up to her death—led her privatissimum in Cracow.
She maintained close contacts with outstanding representatives of the Lvov-Warsaw
School (especially with Władysław Witwicki and Tadeusz Czeżowski) but also with great
philosophers from outside of the School, like Henryk Elzenberg and Roman Ingarden—as
well as with a great Polish poet, Zbigniew Herbert.
3 Work
The work of Dambska

˛ contains many original publications, translations of philosophical
texts and a huge number of notes on Polish publications in the Bibliographie de la
Philosophie.
The most important of her books are: La théorie du jugement de M. Edmond Goblot
[1], On laws in science [2], Outline of the history of Greek philosophy [3], Irrationalism
and the scientific cognition [4], French skepticism of the 16th and 17th centuries [6], Two
Izydora Dambska:
˛ The First Lady . . . 225
studies from the theory of scientific cognition [7], Instruments and object of cognition:
from the theory of instrumental cognition. On linguistic philosophy [8], Two studies on
Plato [9], Conventions and conventionalism [10].
Her selected writings are collected in her Znaki i mysli. Wybór pism z semiotyki, teorii
nauki i historii filozofii. (Signs and thoughts. Selected writings in semiotics, epistemology
and the history of philosophy) [11]. Recently, a volume of English translations of her
important work was published under the title Knowledge, language and silence [14].
4 The Lvov-Warsaw School Versus Neopositivism and Linguistic

Philosophy
Dambska’s
˛ philosophical research, which may be called “semiotic-logical analysis”, is
carried out in the spirit of Twardowski’s school. She characterized the scholar spirit of
this school in such a way:
Despite the fact that representatives of the Lvov-Warsaw School understood the scope of
philosophy differently, they however agreed to postulate that research should fulfil some definite
conditions characteristic for scientific cognition. The most important postulate, consequently
realized by members of the School in their works, ordered the application in philosophical research
of the method of semantic analysis and logical discourse by appreciating the role of broadly
understood intuition in the process of discovering statements; the postulate of clarity, precision and
logical correctness in formulating issues, theses and arguments and in defining concepts; finely —
the postulate of criticism and antidogmatism in estimation of theoretical assumptions [13, p. 29].
Dambska’s
˛ attitude towards the programs of other currents of analytical movement—
like neopositivism and linguistic philosophy—was ambivalent.
She accepted some views of the representatives of the Vienna Circle and the Oxford
School, but some their views were rejected by her—like the Carnapian opinion that a
system of knowledge is finally reducible to “what is empirically given” or the radical
opinion of post-Wittgensteinian philosophers that semiotic analysis is the only method
of scientific research in philosophy and, in consequence, that all the main questions of
philosophy are incorrectly posed, i.e. are pseudoproblems.
It seems that what unites all these currents of analytic movement is anti-irrationalism.
That is why Dambska
˛ paid considerable attention to the phenomenon of irrationalism.
According to Dambska,
˛ we should distinguish four main versions of irrationalism:
metaphysical, epistemological, logical and psychological.
Metaphysical irrationalism is an opinion that irrationality is an essential property of
reality itself; in consequence, using a rational conceptual apparatus to describe such
a reality is a kind of deformation; paradoxically, the rational attitude to the irrational
reality consists in . . . delighted silence. Epistemological irrationalism consists in accepting
irrational cognitive methods—like intuition, contemplation, empathy etc.—which are
to guarantee the scientific status of logically irrational sentences. It is interesting to
note that epistemological irrationalists consider these methods reliable—in opposition
to traditional rational methods, i.e. experience and reasoning based on it. In turn, it is
considered logically irrational for users of a given language. For all the sentences of this
language are either contradictory to laws known to these users, or essentially irresolvable;
it is clear that when somebody does not know a certain law, his acceptance of sentences
contradictory in fact to this unknown law could not be irrational. Finally, a person
who believes in logically irrational sentences—or is ready to use irrational methods of
cognition—is psychologically irrational.
˛ there is room for neither epistemological, logical, metaphys-
ical nor psychological irrationalism in science. Science should be rational, i.e. it should
use only rational or intersubjective methods of research.
5 Metaphysics
5.1 Ontology
Among ontological questions, the controversies concerning determinism–indeterminism

and causalism-acausalism were of special importance for Dambska.
˛
In these controversies, she was in favor of determinism and causalism. Accepting the
first, she appealed to the fact, that practicing science is rational only on the ground of
deterministic hypothesis. Accepting the second, she pointed out that the rule of limited
measurability (of the speed and location of physical bodies), supposed in quantum
physics, does not speak against the hypothesis of causalism—as some philosophers think;
the source of their mistake is confusion of indeterminism with indeterminacy, and the
principle of causality with the rule of predictability.
5.2 Epistemology
In the domain of epistemology, two problems especially occupied Dambska:

˛ the problem
of instrumental cognition and the problem of the relation between conventionalism on one
hand, and relativism, scepticism and agnosticism on the other.
The problem which arises in face of instrumental cognition, consists in the question of
how instruments of such a cognition or—as Dambska
˛ named them—“cognitive operators”
impact the other four elements of cognitive situation, i.e.: the subject of cognition, the
process of cognition, the object of cognition and the result of cognition (scil. picture
of cognized object). As it turns out, instruments affects all these factors, modifying
them, and vice versa. Traditional epistemology was not conscious of these complicated
interrelations.
In Dambska’s
˛ day, there was a widespread belief that conventionalism was dangerous
because of its relativistic, sceptic and agnostic consequences. In light of Dambska’s
˛
meticulous analysis, this belief was based on terminological misunderstandings. Let us
note that we have two versions of conventionalism: extreme and moderate. Extreme
conventionalism is the view that all scientific laws are conventions (in particular: they
are arbitrary definitions) and as such they are not empirically verifiable. Moderate
conventionalism simply states that conventions are present in various areas of human
life and in various domains of culture. According to Dambska,
˛ extreme conventionalism
is obviously false, whereas moderate conventionalism is obviously true, but both of
Izydora Dambska:
them imply neither relativism and scepticism nor agnosticism. For instance: relativism
assumes that reality itself is contradictory—and conventionalism assumes only that there
are different, sometimes contradictory, pictures of reality; scepticism assumes that norms
and definitions are accepted purely conventionally—and conventionalism assumes at most
that they are accepted with respect to their semantic function in the system of a given
theory; agnosticism (considered by Dambska
˛ to be a radicalized version of scepticism)
assumes that the classical conception of truth is binding—and conventionalism accept the
classical conception of truth.
6 Semiotics
6.1 Logical and Grammatical Categories
Dambska
˛ conducted deep criticism of the traditional (grammatical) categorizations of the
parts of speech, showing that in fact, it had serious shortcomings.
Here are two examples.
Firstly, she rejected the view that so-called empty names (e.g. “a parca”) feature a
lack of designata, while proper names (e.g. “Casimir’) feature a lack of connotation.
˛ all names signify something: possible objects of thought namely;
only such an approach is compatible with the common conviction that some sentences
containing empty names are true (like e.g. “Morta is a parca”), whereas some of them are
false (like e.g. “Morta is a siren”). On the other hand, all names have connotation, proper
names included, anyway taken contextually, when their connotation is identical with the
connotation of an appropriate description of the individual signified by the proper name
(whereas proper names taken acontextually remain variables).
Secondly, she stood in opposition to the idea that logical implications can be
considered adequate interpretations of conditional propositions of natural language of
the type “If p, then q”; according to Dambska
˛ the meaning of natural conditionals is the
judgement that between what is stated in the antecedent ‘p’ and what is stated in the
consequent ‘q’, there is a relation of sufficient conditioning. She also proposed a certain
modification of Twardowski’s conception of the meaning of the conditional proposition
in casus realis of the type “If it is the case that p, then q”. Twardowski identified this
meaning with the meaning of the complex of three judgments: (a) a judgement stating
the conditioning between ‘p’ and ‘q’; (b) a judgement stating the occurrence of the
condition and (c) a judgement stating the occurrence of the conditioned thing. According
to Dambska,
˛ there is only (a) because only (a) is explicitly expressed in this kind of
proposition. The remaining two components mentioned by Twardowski—namely (b) and
(c)—are what is expressed in this kind of proposition only implicitly.
6.2 Truth Relativized to Language
Dambska
˛ noted that definitions of truth are rooted in conceptions of language for which
the predicate “is true” is defined. She analysed in this respect three conceptions of
language which were in circulation in the twentieth century: the correspondent (Alfred
Tarski, Rudolf Carnap), the operational (late Ludwig Wittgenstein) and the immanent
(Kazimierz Ajdukiewicz). In the first conception, language is treated as a system of signs
which refer to a certain objective domain. In the second conception, language is a form
of biological and cultural behaviour of a person. In the third conception—language is
considered to be a set of signs and directives of creating signs and transforming one sign
into another.
Dambska
˛ shows that definitions of “truth” given in the frame of these conceptions are
relativized to them. The classical definition of truth harmonizes with the correspondence
conception of language (accepted, by the way, by Dambska).
˛ She formulates it as follows:
An affirmative sentence is true when the state of affairs corresponding to this sentence
occurs; a negative sentence is true when a state of affairs corresponding to it does not
occur. On the basis of the operational conception, a pragmatic definition of truth is natural.
Immanent conception is a basis of the syntactic definition.
According to Dambska—all
˛ these concepts are derivative with respect to the concept
of truthfulness as something that is a feature of judgements in the logical sense (or the
logical content of sentences) and what is correlated with some ontical categories—first of
all with the concept of existence.
6.3 Understanding
The expression “to understand” occurs, i.a., in contexts such as “A understands that p” and
“A understands X”. The context of the second type has several meanings—to understand
X is the same as: to know what X means, to know what X expresses, to know what X’s
structure is, and in the end—to know what idea is realized by X. Dambska
˛ was convinced
that, in all these cases, understanding: (a) concerns objects connected with man’s spiritual
life; (b) consists in becoming aware of relations which indicate the meanings of these
objects, but (c) this becoming aware is repeatable.
Dambska
˛ was convinced that conditions (a)–(c) may be considered as essential
conditions of understanding. With respect to condition (b)—understanding is a fallible
cognitive act.
6.4 Silence and Namelessness
Dambska’s
˛ analysis of the semiotic functions of silence has a multidimensional character.
Silence is either a simple lack of speech (not-speaking) or refraining from speaking
(signitive silence). Signitive silence analysed as a mark is either a symptom or a
signal. Considered as a communicative element of natural language, it is—leaving aside
expressive functions—a kind of indexical expression. Besides semantic functions, it
performs pragmatic ones, in particular, it is a means of a fight or a way of striving for
perfection.
In her semiotic-psychological-cultural research on the concept of namelessness,
Dambska
˛ starts from the ascertainment that, on the one hand, we hold our name in high
Izydora Dambska:
esteem and, on the other hand, sometimes we pretend to become nameless. Getting rid of
a name, changing or hiding it (namelessness) are not indifferent from the psychological
point of view (among motives of namelessness there are fear and a need of play) as well as
from a sociological point of view (namelessness in action is, first of all, a way of fighting).
7 Methodology
7.1 Justification
To justify a sentence ‘p’ is to show that if fulfils sufficient conditions to accept (state,
know, suppose, expect) that p.
Justification may by direct or indirect (scil. by reasoning).
7.1.1 Direct Justification
A sentence is justified directly if we accept it on the basis of our own experiences.

Scientific claims should be intersubjectively justified. But, how could we know that the
experience of someone else is the same or at least similar to ours—in the face of the same
facts?
According to Dambska
˛ we do not have to know this in order to do science. A
hypothesis that other people’s experiences are the same or similar to our experience
is not—contrary to appearances—a premise of scientific claims, but at the most a
metascientific hypothesis. The belief in the similarity of the content of human perception
has a similar status as the belief that scientists as such are not liars.
Moreover, if this hypothesis said that scientific theses are theses about intersubjective
objects, these objects would not be impressions as such but some relations between
impressions. These relations can be intersubjectively cognized. This is certified by the
fact that normal people in similar circumstances usually accept the same sentences.
7.1.2 Indirect Justification
Among types of indirect justification, Dambska

˛ was especially interested in reasoning by
analogy.
Dambska
˛ understood analogy to be “a structural similarity of some sets or systems, i.e.
a similarity of relations holding between elements of these sets of parts of these systems,
and between the properties determined by these relations”.
Reasoning by analogy may take one of two forms:
(a) [(A : B) :: (C : D)] ⇒ [(B : A) :: (D : C)].
This is a scheme of, for instance, the following reasoning: If God is for people what
a father is for children, then people are for God what children are for a father. Here,
the conclusion follows from the premise. This reasoning is deducible (and infallible)
and applied for instance in mathematics.
(b) [(A/B : C/D)) ∧ F (C/D)] ⇒ F (A/B).

This is a scheme of, for instance, the following reasoning: If God is for people what
a father is for children, and a father is the children’s just judge, then God is a just
judge for people. Here, the conclusion often does not follow from the premises and
the premises do not follow from conclusions.
The existence of reasoning of type (b) certifies the fact that the traditional division of
processes of reasoning into deductive (in which a conclusion follows from premises) and
reductive (in which premises follows from the conclusion) is inadequate.
Reasoning by analogy was considered by Dambska ˛ to be binding (and eo ipso
infallible), when the analogy occurring in them is essential, i.e. when the relation “fulfils
the conditions which satisfy the same rule or the same law”.
Reasoning by analogy often has an insightful character and may serve as a justification
of sentences about the future. The last point is based on the assumption of the isomor-
phism or homomorphism of future events with respect to already given ones.
Appreciating the cognitive value of reasoning by analogy and noting its insightful
character, Dambska
˛ at the same time emphasized the danger connected with making use
of analogies in which one element may not be cognized by the rational method in science.
Such analogies and reasoning based on them are—according to Dambska—irrational.
˛
7.2 Scientific Laws
There is no science without laws.

Dambska
˛ considered scientific laws to be general implications: (a) in which ranges of
variables are open classes (infinite or such that we may not decide whether they are finite);
(b) concerning a constant connection between phenomena; (c) without any absolute time
determination; (d) being an element of a certain science, (e) empirically verified.
Such a concept of law has applications in axiomatized formal disciplines in which
theses are tautologies. Here, axioms and some of their consequences—namely those
which are of a special importance, for instance, those which are applied in the reasoning
of other disciplines or in daily life, or which can simplify these kinds of reasoning—are
called “laws”. Criteria of this importance are too vague to distinguish laws so understood
from a set of tautologies.
This concept has—according to Dambska—applications
˛ in both the natural and human
sciences, especially in history. The peculiarity of history does not consist in the fact
that it is an idiographic science (or that it describes individual facts) in opposition to
natural sciences which are nomological (or which formulate laws). History differs from
natural sciences with respect to the level of complication of the examined facts, which are
relatively simple in the case of natural sciences but are very complicated combinations of
many phenomena—physical, psychical and sociological ones—in the case of history.
Izydora Dambska:
7.3 Truthfulness of Scientific Laws
Some scientists claim that truthfulness is neither a necessary nor sufficient condition of
scientific claims—in particular regarding physical laws. According to some of them—
this is because these laws are in fact arbitrary definitions or their analytical consequences.
According to others, laws are only provisory hypotheses, relativized to the changing state
of knowledge; if they are approximate hypotheses—they are simply false.
Dambska
˛ refuted this point of view. Laws of physics would not have any logical value
if they could be interpreted as functions whose degree of approximation is undefined (the
range of the unknown may be bigger or smaller); but usually in such laws it is (at least
provisionally) defined in certain orders. They are not, in Dambska’s
˛ opinion, propositional
functions. Even if in physics there are laws in which both antecedent and consequent are
false, we may only conclude that such laws are not verifiable. Unverifiability is not the
same as falsity.
According to Dambska—truthfulness
˛ is not a necessary condition of being a law. This
is of course classically understood truthfulness—not verifiability (since false sentences
may also be verified).
Another thing is that we sometimes simply do not know whether a given law is true.
We are inclined to believe in those sentences which are probable.
8 Axiology
In the domain of axiology—Dambska

˛ accepted absolutism and objectivism.
As opposed to the axiological nihilists, she believed that values exist in reality and
are not only purely intentional or fictional objects. As opposed to the relativists—she
was convinced that the changeability of conventions “does not testify to the relativism of
values, but that they be differently understood” or “that means of realizing them may be
chosen differently.” As opposed to subjectivists, who wanted to see the source of values
in “causative subjects”, she wrote:
The base of establishing [. . . ] the [ethical rules of legal laws] of decisions [. . . ] is created usually
by some judgements on values which pretend to be objectively justified; and decision is a choice
of a certain norm which — according to the opinion of the decision maker — states an obligation
to ways of conduct which aim at realizing or preserving these values [10, p. 114].
8.1 Man and the World of Values
The world of man is the world of values. Dambska

˛ wrote:
Each of our conscious actions is directed by a desire to realize values giving sense to this action
[10, p. 31].
Among sentences which concern the world of values as it is broadly understood—

Dambska
˛ distinguished, i.a., norms, evaluations and axiological sentences sensu stricto.
We have already mentioned norms above. Evaluations and axiological sentences sensu
stricto form a subclass of axiological sentences sensu largo. Evaluations—are sentences
stating that some objects are valuable (scil. that they possess value). Axiological sentences
sensu stricto are sentences which state what values are and what kinds of them exist.
Among these types of sentences—Dambska˛ formulated mostly normative and evalua-
tive sentences. However, she usually supplemented them with axiological sentences sensu
stricto.
We read in Dambska:
˛
A peculiar feature of metaphysical investigation in the Lvov-Warsaw School is emphasized which
is given to axiological moments: moral values which are assumed and produced by making
philosophy and to its peculiar ethos, which shapes the life of philosopher [13, p. 29].
There is no doubt that Dambska

˛ herself had undertaken analysis of “axiological
moments” in various domains of her research, first of all in research concerning the theory
of science.
She considered accuracy (scil. functionality) to be the most important cognitive value
of science. Accuracy—as opposed to truthfulness—in a gradable form. The better answers
to questions concerning its domain a science gives (the best that can be given under
specific conditions of cognition), the more accurate it is.
8.2 The Conception of Freedom
Among the values of human life, Dambska

˛ analysed i.a. freedom. It was a conscious
choice, motivated by the same factors as analyses devoted to silence and namelessness.
Freedom is not always an axiologically positive value. It is so only if freedom is a
necessary condition for realising some positive values.
One sometimes distinguishes freedom-from and freedom-to. Dambska ˛ [12] empha-
sized the fact that “freedom-from” and “freedom-to” are correlative terms. She wrote:
Science’s freedom from ideological and administrative pressure is freedom towards it inherent
function of searching for and delivering truth; freedom of speech is at the same time freedom to
publicly proclaim one’s views and convictions [12, p. 857].
The correlation occurs that freedom is a necessary (but not sufficient) condition of the
possibility of action. And so: freedom from mistake is a condition of morality etc.
8.3 Normative Ethics
In the domain of normative ethics, Dambska

˛ defended a certain version of perfectionism
and intellectualism (and she contrasted the latter to emotionalism). She formulated ethical
criterion as follows: Conduct is good if it aims at the perfection of a human; it is bad if it
discourages perfection.
Izydora Dambska:
9 Significance
The significance of the personality of Dambska—especially

˛ after World War 2—consisted
in the fact that she had the courage to bear witness to the truth in all circumstances of her
life.
The significance of the work of Dambska—lies
˛ in the fact that her work reflects laws
of reason and the axiological taste.
That is why she earned the title of the first lady of twentieth-century Polish philosophy.
Izydora Dambska’s
˛ Works Mentioned in the Paper
1. La théorie du jugement de M. Edmond Goblot. Towarzystwo Naukowe we Lwowie, Lwów (1930)

2. O prawach w nauce [On Laws in Science]. Gubrynowicz i Syn, Lwów (1933)
3. Zarys historii filozofii greckiej [Outline of the History of Greek Philosophy]. Filomata, Lwów (1935)
4. Irracjonalizm a poznanie naukowe [Irrationalism and the scientific cognition]. Kwartalnik Filo-
zoficzny XIV(i. 2), 83–118 (1937); (i. 3), 185–212 (1938)
5. List do Marii Obercowej [Letter to Maria Obercowa]. In: Perzanowski, J. (ed.) Izydora Dambska. ˛
1904–1983, p. 23. Wydawnictwo PAU, Kraków (2001) (Personal communication, 1951)
6. Sceptycyzm francuski XVI i XVII wieku [French Skepticism of 16th and 17th Centuries]. To-
warzystwo Naukowe w Toruniu, Torun (1958)
7. Dwa studia z teorii naukowego poznania [The Studies from the Theory of Scientific Cognition].
Towarzystwo Naukowe w Toruniu, Torun (1962)
8. O narz˛edziach i przedmiotach poznania: z teorii instrumentalnego poznania. O filozofii lingwistycznej
[Instruments and Objects of Congition: from the Theory of Instumental Congition. On Linguistic
Philosophy]. PWN, Warszawa (1967)
9. Dwa studia o Platonie [Two Studies on Plato]. Ossolineum & PAN, Wroclaw (1972)
10. O konwencjach i konwencjonalizmie [Conventions and Conventionalism]. Ossolineum & PAN,
Wroclaw (1975)
11. Znaki i myśli. Wybór pism z semiotyki, teorii nauki i historii filozofii [Signs and Thoughts. Selected
Writings in Semiotics, Epistemology and the History of Philosophy]. PWN, Warszawa (1975)
12. Gdy myśle o słowie “wolność” [When I think about the word “freedom”]. Znak XXXIII(6), 855–860
(1981)
13. O niektórych koncepcjach metafilozoficznych w Szkole Lwowsko-Warszawskiej [On some
metaphilosophical conceptions in the Lvov-Warsaw School]. In: Perzanowski, J. (ed.) Jak filozo-
fowac? Studia z metodologii filozofii [How to Philosophise? Studies from the Methodology of
Philosophy], pp. 22–29. PWN, Warszawa (1989)
14. Kowledge, Language and Silence. Selected Papers. Brill/Rodopi, Leiden & Boston (2016)
A. Brożek () • J. Jadacki

e-mail: [email protected]; [email protected]
Maria Kokoszyńska-Lutmanowa: The Great
Polemist
Filip Kawczyński
Abstract The paper concerns life and scientific achievements of Maria Kokoszyńska-
Lutmanowa (1906–1981) who was a significant logician, philosopher of language and
epistemologist belonging to the second generation of the Lvov-Warsaw School. She is
mostly known as the author of the important argumentation against neopositivism of the
Vienna Circle as well as one of the main critics of relativistic theories of truth. The article
is divided into five sections: 1. Life, 2. Main papers, 3. Views, 4. Pupils, 5. Summary, 6.
References.
Keywords Maria Kokoszyńska-Lutmanowa · Vienna Circle · Neopositivism ·

Justification · Relativism · Absolutism
1 Life1
Maria Kokoszyńska-Lutmanowa was one of the key figures among the so called second
generation of philosophical Lvov-Warsaw School, along with such prominent thinkers
as Alfred Tarski, Izydora Dambska,
˛ Janina Kotarbińska or Henryk Mehlberg. Although
the list of Kokoszyńska’s philosophical achievements is impressive, for what she is best
known is probably her substantial and influential criticism of ideas proposed by the main
philosophical force in that time—the Vienna Circle.
Kokoszyńska was born on 06.12.1906 in Bóbrka near Lvov in the family of Eugenia
Kokoszyńska (née Sidorowicz) who was a pianist and a music teacher and Stanisław
Kokoszyński, a civil servant. She undertook studies in philosophy and mathematics in
1923 at the Jan Kazimierz University in Lvov, where under the supervision of Kazimierz
Twardowski—founder of the Lvov-Warsaw School—she wrote a dissertation entitled
“Nazwy ogólne i wieloznaczne” (English translation of the title: “General and Ambiguous
Names”) and completed her PhD in 1928. Apart from Twardowski she has an opportunity
1 This section of the paper draws on Zygmunt [4].

236 F. Kawczyński
to be a student of several other great scholars like Roman Ingarden, Sefan Banach,
Hugo Steinhaus or Stanisław Ruziewicz. However, it is Kazimierz Ajdukiewicz who
should be considered as the teacher that had the major influence on Kokoszyńska’s
later development as a philosopher. Besides learning from the superb teachers, during
her university years she established long-standing friendships with other Twardowski’s
students and—as it turned out later—significant philosophers and logicians such as
Dambska,
˛ Seweryna Łuszczewska-Romahnowa or Daniela Gromska.
Between 1930 and 1934 she was an assistant of Ajdukiewicz in the Department of
Philosophy of the Jan Kazimierz University. In 1931 she got married to dr. Roman
Lutman, polish lawyer, historian and social activist (afterwards the director of Silesian
Library and Silesian Institute in Katowice). They had no children.
Between 1934 and 1939, due to being awarded with several scholarships, Kokoszyńska
was a visiting scholar in Prague, Vienna, Paris and Cambridge. During her visit in Vienna
she was a guest at the seminar organised by the Vienna Circle, where she presented a
paper that eventually has been published in the official journal of the Wiener Kreis, i.e. in
“Erkenntnis” [4]. That allowed her to get in contact with some of the most influential
philosophers of that era: Rudolf Carnap, Otto Neurath or Moritz Schlick. In Vienna
she also had an opportunity to meet Karl Popper and Kurt Gödel. In 1938 she went to
Cambridge where she was supposed to stay for 3 years, however, she got back to Poland in
June 1939 because of the war looming. Nevertheless, she managed to attend G.E. Moore’s
lectures and meet Ludwig Wittgenstein. During the years of travelling around Europe she
gave numerous talks at various philosophical conferences, most important of which are:
1st International Congress of Scientific Philosophy (Sorbonne, 1935), 9th International
Philosophical Congress (Paris, 1936), 3rd Polish Philosophical Congress (Cracow, 1936),
4th International Congress for the Unity of Science (Cambridge, 1938).
Kokoszyńska spent the war years in Lvov. In 1947 she moved with her husband
to Wrocław. The same year she obtained her postdoctoral degree on the basis of her
work “W sprawie wzgl˛edności i bezwzgl˛edności prawdy” (English translation of the
title: “On Relativity and Absoluteness of Truth”). In 1950 she occupied a position of
the head of newly-formed department of Logic and Methodology of Science at the
University of Wrocław. She governed the department for 26 years and during that period
she cooperated with many significant logicians, e.g. Jerzy Słupecki, Ludwik Borkowski,
Tadeusz Kubiński, Witold Pogorzelski, Ryszard Wójcicki or Bogusław Iwanuś.
Kokoszyńska became professor extraordinarius in 1951 and professor ordinarius in
1969. From 1951 to 1954 she was a dean of the Faculty of Philosophy and History of the
University of Wrocław and a vice-rector of the university from 01.01.1955 to 30.09.1956.
Kokoszyńska died in Wrocław on 30.06.1981 after struggling with a dread disease.
2 Main Papers
For comprehensive bibliography of Maria Kokoszyńska-Lutmanowa see: Jan Zygmunt,

“Bibliografia prac naukowych Marii Kokoszyńskiej-Lutmanowej” Filozofia Nauki 46
(2004).
Maria Kokoszyńska-Lutmanowa: The Great Polemist 237
[1] Z semantyki funkcyj zdaniowych [On the Semantics of the Propositional Func-
tions]. In: Ksi˛ega Pamiatkowa
˛ Polskiego Towarzystwa Filozoficznego we Lwowie
12.02.1904–12.02.1929, Lwów (1932)
[2] Syntax, Semantik und Wissenschaftslogik. Actes du Congrés International de
Philosophie Scientifique, Paris (1936)
[3] Logiczna składnia j˛ezyka, semantyka i logika wiedzy [Logical Syntax of Language,
Semantics and Logic of Knowledge]. Przeglad ˛ Filozoficzny 39(1), 38–49 (1936)
[4] Über den absoluten Wahrheitsbergiff und einige andere semantische Begriffe.
Erkenntnis 6, 143–165 (1936)
[5] Filozofia nauki w Kole Wiedeńskim [Philosophy of Science in the Vienna Circle].
Kwartalnik Filozoficzny 13(1), 151–165; (2), 181–194 (1937)
[6] Sur les éléments métaphysiques et empiriques dans la science. Travaus du IX
Congrés International de Philosophie, Congrés Descartes, Paris, 108–117 (1937)
[7] Bemerkungen über die Einheitswissenschaft. Ekrenntnis 7, 325–335 (1938)
[8] W sprawie walki z metafizyka˛ [On Fighting Metaphysics]. Przeglad ˛ Filozoficzny
41, 9–24 (1938)
[9] What Means “Relativity of Truth”? Studia Philosophica 3, 167–176 (1948)
[10] A Refutation of Relativism of Truth. Studia Philosophica 4, 93–149 (1951)
[11] (with T. Kubiński and J. Słupecki) Zastosowanie logiki matematycznej do wyjaś-
niania niektórych poj˛eć przyrodoznawstwa [Application of Mathematical Logic for
Explaining Some Concepts of Natural Sciences]. Studia Logica 4, 155–211 (1956).
Incl. summary in Russian and English
[12] O “dobrej” i “złej” indukcji [On “Good” and “Bad” Induction]. Studia Logica 5,
43–70 (1957). Incl. summary in English
[13] O stosowalności metody dedykcyjnej w naukach niededukcyjnych [On the Ap-
plicability of the Deductive Method in Non-deductive Sciences]. Sprawozdania
Wrocławskiego Towarzystwa Naukowego 15 A, 45–49 (1960)
[14] O dwojakim rozumieniu uzasadniania dedukcyjnego [Two Concepts of Deductive
Justification]. Studia Logica 13, 177–196 (1962). Incl. summary in English
[15] O dedukcji [On Deduction]. Studia Filozoficzne 1, 77–86 (1963). Incl. summary in
English
[16] W sprawie koncepcji nauk dedukcyjnych [About the Conception of Deductive
Sciences]. Studia Filozoficzne 1, 57–65 (1967)
[17] Złudzenia aprioryzmu [Illusions of Apriorism]. Ruch Filozoficzny 28(1–2), 72–78
(1970)
[18] Metodologiczne problemy integracji nauk [Methodological Problems of Integration
of Sciences]. Studia Filozoficzne 4(89), 47–56 (1973)
3 Views
When Kokoszyńska’s philosophical career began in the late 1920s, central problems
in analytic philosophy concerned language of science, its syntax and semantics. In
general, Kokoszyńska shared with other members of Lvov-Warsaw School the view
according to which science is thought to be a system of sentences with uniform logical
238 F. Kawczyński
structure and thus it is possible to verify and adjust it by using theoretical tools of logic,
methodology or (logical) semantics. Along with e.g. Zawirski, Czeżowski or Mehlberg
she paid special attention to the issue of logical form of sentences and their justification.
She was an adherent of the view that variables appearing within propositional functions
are meaningful and that their meaning is the integral component of the meaning of all
substitutions; in other words, it is meaning shared by all these substitutions (see [1]).
When it comes to dealing with the notion of justification of sentences, one of the crucial
tasks is defining what does it mean for the sentence to be true. According to logical
positivists (especially Carnap, but also Neurath or Hempel) truth—as well as some other
apparently semantic notions—can be (and should be) defined in terms of syntax and thus,
the whole metascientific reflection is reduced to syntactic analysis of the language of
science. Kokoszyńska (see [2, 3]) confronted such position with proposals outlined by
Tarski and Gödel, and came to the conclusion that positivists’ programme is too radical,
since—as it stems especially from Tarski’s works—semantics is inevitable and necessary
for understanding the notion of truth (and thus justification etc).
Kokoszyńska’s substantial criticism of the ideas proclaimed by the Vienna Circle—
although accompanied by respect and admiration of their contribution to the development
of philosophy—did not stop with the above disagreement. In her later works she got back
to the caution that the principle of primacy of pure syntax goes too far. One of the main
reasons for neopositivists to establish that principle was their idea of uniform scientific
language. They believed that such unity can be achieved on that very abstract level of
pure syntax. Kokoszyńska pointed (see [5]) that such an attitude is either: (a) futile—
as it is merely an assumption that all empirical scientific statements can be expressed
in one universal language and as a matter of fact there is no good evidence to accept
this assumption as correct; or (b) incorrect—because it has been conclusively shown that
logical knowledge cannot be expressed in a single language as semantic statements always
demand the other language—namely a metalanguage—to be expressed.
Kokoszyńska attacked also fundamental neopositivist postulate of abandoning meta-
physics or strictly speaking—the way in which it has been realised by them. She argued
that (see [6, 8]) logical positivists are too hasty in dismissing everything labeled as
“metaphysics” and she distinguished two concepts of metaphysics. In the first sense
metaphysics is considered as dogmatic and unjustified acceptance of some undetermined
propositions. Obviously, Kokoszyńska supports the idea that metaphysics so understood
has no entry to the area of serious scientific investigation. In the second sense, however,
metaphysics is a set of propositions of some kind, and there are three possible options
with regard to what exactly belongs to such a set: (a) determined analytic propositions; (b)
undetermined and testable propositions; (c) undetermined and non-testable propositions.
As may be expected, Kokoszyńska denies possibility of any scientific interest in the
propositions of the last type. However, contrary to members of the Vienna Circle, she
is not that certain that propositions of the kinds (a) and (b) also have no raison d’être.
As a matter of fact, she points out that such statements are indeed present in scientific
investigation and work there well as they are general yet still theoretically testable
propositions. Worth emphasising, Kokoszyńska believed that propositions of all above
types are meaningful; propositions of the type (c) are in fact useless for science, however,
there are no reasons to claim that they are senseless.
Maria Kokoszyńska-Lutmanowa: The Great Polemist 239
It may be said, that despite negative assessment of some ways in which neopositivists
realised their programme, Kokoszyńska in general shared their main philosophical ideas.
Her criticism was aimed not to sink logical positivism but rather to recognise its weak
points and thus make it possible to review this theory and eventually make it stronger.
Another field in which Kokoszyńska manifested her polemical edge was the debate
concerning relativity of truth. In general, members of Lvov-Warsaw School were ab-
solutists with regard to the issue of truth, and so was Kokoszyńska. Her in-depth and
comprehensive analysis of relativistic theories allowed her to provide clear explication of
relativity of truth:
(I) the term ‘true’ is an incomplete predicate; (II) ∃X{Y, Z[(X is true with respect to Y ) & Non-X
is true with respect to Z)]}. [10, p. 94]
She presented a series of arguments against arguments for relativism (see [10]). First,
she claimed that some of the arguments for relativism are in fact arguments for a very
weak version of relativism, according to which not all theorems are necessary, which is
of course the case, but does not entail any interesting theory. Second, she pointed out
that many relativists rely on theories from the area of psychology, sociology, pragmatics
etc. while in her opinion none of them is appropriate. Third, she showed that relativism
often stems from misinterpretation of how deictic expressions function in language. Apart
from criticising relativism, Kokoszyńska offered some ideas that should be taken into
account when absolutist theory of truth is to be developed. In such a theory it has to
be possible to distinguish truth from criteria of truth. The theory has to provide also a
method of interpreting all occasional expressions as well as involve the T -convention.
Worth mentioning here that Kokoszyńska—who thought that T -convention is a necessary
base for any adequate theory of truth—was one of the first philosophers who recognised
great value of Tarski’s proposal (in one of her letters to him she wrote that even he does
not fully appreciate enormous significance of his own theory).
Conflict between absoluteness and relativity affected also Kokoszyńska’s late works,
concerning deductive and non-deductive justification (see [12–16]). In one of her papers
[16] she enters the dispute with Quine over notorious analytic-synthetic distinction.
Notion of analyticity was required in her theory to support the distinction for absolute
deductive justification and relative deductive justification. One of the specific features of
the former is that it is applicable to analytic statements exclusively. Thus, Kokoszyńska
needed a criterion of identifying analytic (resp. synthetic) statements and she offered
very interesting criterion of that kind, based on the notion of denoting—she claimed
that analytic are the statements that are true if and only if appropriate rules of denoting
(expressed in metalanguage) fix existing objects as references of terms occurring in a
sentence. In other words, the fact that extension in question is not empty, guarantees
truth (in case of synthetic statements some further conditions besides existence have to be
fulfilled to make a given statement true).
4 Pupils
The most important students of Kokoszyńska are: Tadeusz Kubiński, Ryszard Wójcicki,
Witold Adam Pogorzelski, Waleska Rudek, Wanda Charczuk and Leon Gumański.
240 F. Kawczyński
Worth mentioning that investigations on analyticity conducted by Marian Przeł˛ecki were

strongly inspired by works of Kokoszyńska.
5 Summary
Kokoszyńska, although slightly outshined by philosophical giants like Ajdukiewicz or

Tarski, stays one of the most significant thinkers of Lvov-Warsaw School. Her works
on relativism or analyticity still can be considered as points of reference in a current
debates concerning these topics. Especially her theory of analytic-synthetic distinction
seems to deserve much more attention than it has received so far. Apart from philosophical
achievements Kokoszyńska is also a great source of inspiration as a scholar who was able
(and have enough courage) to compete with and argue against some of the most powerful
philosophers of her times. Last but not least—being the head of successful department
of logic of the University of Wrocław and taking part in numerous philosophical
enterprises—she was also a remarkable figure in Polish philosophical society.
References
1. Brożek, A.: O logice i filozofii w Szkole Lwowsko-Warszawskiej. W 30 lat po publikacji “Filozoficznej

szkoły lwowsko-warszawskiej”. Talk with Jan Woleński [On logic and philosophy in the Lvov-
Warsaws School. Thirty years after the publication of “Philosophical Lvov Warsaw School”. A talk
with Jan Woleński]. In: Brożek, A., Chybińska, A. (eds.) Fenomen Szkoły Lwowsko-Warszawskiej
[Phenomenon of the Lvov-Warsaw School]. Wydawnictwo Academicon, Lublin (2016)
2. Jadacki, J.J.: Orientacje i doktryny filozoficzne. Z dziejów myśli polskiej [Philosophical Directions and
Doctrines. From the History of Polish Thought]. Wydział Filozofii i Socjologii, Warszawa (1998)
3. Tałasiewicz, M.: Maria Kokoszyńska-Lutmanowa – Methodology, Semantics, Truth. In: Krajewski, W.
(ed.) Polish Philosophers of Science and Nature in the 20th Century. Poznań Studies in the Philosophy
of the Sciences and the Humanities, pp. 129–133. Rodopi, Amsterdam (2001)
4. Zygmunt, J.: Maria Kokoszyńska-Lutmanowa. In: Kolbuszewski, J. (ed.) Uczeni wrocławscy.
Członkowie Wrocławskiego Towarzystwa Naukowego 3 (1953–1996) [Scholars of Wrocław. Members
of the Wrocław Scientific Society], pp. 41–44. Wrocławskie Towarzystwo Naukowe, Wrocław (1996)
5. Zygmunt, J.: Bibliografia prac naukowych Marii Kokoszyńskiej-Lutmanowej [Bibliography of scien-
tific works of Maria Kokoszyńska-Lutmanowa]. Filozofia Nauki XII 2(46), 155–166 (2004)
F. Kawczyński ()
Seweryna Łuszczewska-Romahnowa
Roman Murawski and Jerzy Pogonowski
Abstract The paper is devoted to the description of life and scientific achievements as
well as the influence of Seweryna Łuszczewska-Romahnowa.
Keywords Łuszczewska-Romahnowa · Venn diagrams · Classification ·

Methodology · Logical pragmatics · Induction
Mathematics Subject Classification (2000) Primary 03-03; Secondary 01A60
1 Life
Seweryna Łuszczewska was born on August 10, 1904 in Mszana near Zborów (Tarnopol
Voivodeship, at that time in Kaisertum Österreich). She belonged to a noble aristocratic
family: her mother, Maria Antonina was a daughter of Count Wojciech Dzieduszycki
(1848–1909), a politician and philosopher. Her father, Konrad Łuszczewski (1876–1937)
was a grandson of Minister Jan Paweł Łuszczewski (1764–1812) and himself also a
politician. The genealogical tree of Seweryna Łuszczewska-Romahnowa is accessible at:
http://www.sejm-wielki.pl/b/psb.26376.1
She completed her primary and secondary education in Jazłowiec, Jarosław, Kraków
and Lvov. She studied at the Lvov University in the years 1922–1928 both: philosophy
(under guidance of Kazimierz Twardowski, Kazimierz Ajdukiewicz, and Roman Ingar-
den) and mathematics (under guidance of Hugon Steinhaus and Stefan Banach). She
obtained PhD in philosophy at the Lvov University on the basis of the dissertation O
wyrazach okazjonalnych (On occasional terms). The thesis was written under guidance of
Ajdukiewicz, but Twardowski was finally the supervisor. Seweryna Łuszczewska started
her academic work at the First Chair of Philosophy (headed by Ajdukiewicz) at the
University of Lvov in autumn 1932. Before the World War II she worked also as a teacher
in several secondary schools. In 1934 she married Dr Edmund Romahn, a teacher of
secondary schools in Lvov and also a philosopher.

242 R. Murawski and J. Pogonowski
The most tragic period of her life was the wartime 1939–1945. She was expelled from
the Lvov University by the authorities of Soviet Ukraine. In the years 1939–1941 (till
the seizure of Lvov by Germans) she worked as a teacher in secondary schools in Lvov
(lecturing on mathematics and astronomy in Ukrainian) and later as a private tutor. In May
1943 she, together with her husband was arrested by Gestapo and put to prison in Lvov.
Then she experienced a martyrdom of Nazi concentration camps: in Majdanek (where
her husband was killed in September 1943), Ravensbrück and Leipzig division of the
Buchenwald camp. After liberation, she worked shortly as a clerk for UNRRA.
In December 1946 Łuszczewska-Romahnowa came to Poznań and started (January
1, 1947) her work at the Chair of Theory and Methodology of Sciences headed by
Kazimierz Ajdukiewicz and converted in 1951 to the Chair of Logic at the Faculty of
Mathematics, Physics and Chemistry of the Poznań University. In 1954 she became an
associate professor and in 1962 an extraordinary professor. She took the post of the
chairman of the Chair of Logic in 1955, after Ajdukiewicz moved to Warsaw (the chair
was renamed the Department of Logic and included into the Institute of Mathematics in
1969). She was active at that post up to her retirement in 1974. Most of her academic
works were published in that period. Seweryna Łuszczewska-Romahnowa died June 27,
1978 in Poznań.
The main sources of biographical information about Seweryna Łuszczewska-Ro-
mahnowa are the papers by Batóg [13, 14] (reprinted also in [16], all in Polish, containing
very precise description of the curriculum vitae of Seweryna Łuszczewska-Romahnowa)
and [15] (in English).
2 Main Works
Publications by Seweryna Łuszczewska-Romahnowa are not numerous. Moreover, most

of them are in Polish. Still, they contain original results and reflections, which is witnessed
by the fact that some of them were reprinted in the influential collections of papers. One
can divide these works into three groups.
2.1 Logical and Mathematical Works
An extensive article [2] from 1953, published in the first volume of Studia Logica, is
devoted to a graphic method of verifying syllogistic inferences. Her generalization of the
method of Venn diagrams is original and makes it possible to avoid certain ambiguities
involved in this method. Moreover, the author proposes a discursive equivalent of the
method and obtains a decision procedure for the monadic predicate calculus.
Łuszczewska-Romahnowa introduced an important concept of a natural classification
(cf. [5]). First of all, by an n-level classification of a given set X one understands a
sequence of partitions (F1 , . . . , Fn ) of X forming a chain of ever finer partitions of X,
where F1 = {X}. This means that for each 1 < i n and every Y ∈ Fi there exists
Z ∈ Fi−1 such that Y ⊆ Z. Each n-level classification of X is thus a hierarchy of
classifications of this set. Any n-level classification F = (F1 , . . . , Fn ) of X generates
Seweryna Łuszczewska-Romahnowa 243
a measure of similarity (indistinguishability) between elements of X with respect to F.

Namely, let IF (x, y) (index of similarity) be defined as the number of the last level of F
at which objects x, y ∈ X still remain elements of the same set from the corresponding
partition. Let DF (x, y) = n − IF (x, y). Then DF is a quasi-distance function which
becomes a standard distance if the last member of F is a partition of X into one-element
sets. Now, it usually happens in empirical sciences that investigated objects are compared
in several different aspects and are considered as more or less similar with respect to the
criteria taken into account. That is to say, the Nature itself projects similarities between
objects. This can be formally expressed by a certain distance function d, established
empirically. If at the same time one classifies investigated objects, then one can ask
about a correspondence between empirical similarity characterized by d and similarity
based on the classifications characterized by DF . Łuszczewska-Romahnowa calls an n-
level classification F natural, if DF (x, y) < DF (u, v) implies d(x, y) < d(u, v), for all
x, y, u, v ∈ X.
Two papers written together with Tadeusz Batóg (cf. [9] and [10]) bring a generalized
theory of classifications. The essence of this generalization lies in the possibility of
constructing multi-level classifications of a given set such that the number of levels is
a transfinite ordinal number. The corresponding definition extends in a natural way the
concept of n-level classification mentioned above. The authors provide a few non-trivial
examples of such transfinite classificatory systems. Moreover, they obtain an algebraic
characterization of the class of all such systems: they prove that it is an arithmetic class
in a wider sense. Thus, the introduced concept is indeed a natural one. The authors show
also some connections between this concept and Boolean metric spaces.
2.2 Methodological Works
The paper [1] from 1948 deals with ambiguity of terms occurring in the scientific
language. According to it, the liberation of the scientific language from all ambiguities
does not take place in the research practice but is also in general impossible. Nevertheless,
this language functions fairly well as a tool of communication between academicians.
This is due to semantic stability of that part of the scientific language which is responsible
for logical consequence, truthfulness, empiricalness, etc. Polysemy creates no danger to
scientific argumentation as long as the scientists themselves are careful in conducting their
argumentation in a logically correct way.
The papers [6] and [8] may be classified as belonging to logical pragmatics. The author
tries to formulate anew the classic theory of argumentation errors (which included e.g.
petitio principii, non sequitur, material error). However, she obtains much more than
simply a new look at these phenomena. Her most important achievement is, at least in
our opinion, a proposal of a precise logical reconstruction of argumentation. It seems
likely that she was the first, who suggested an annotated representation of argumentation.
Łuszczewska-Romahnowa was aware of the limited scope of application of the proposed
formalism (e.g. it does not account for apagogic argumentation). Still, it is a nice piece of
work in formal pragmatics, a completely new discipline at that time.
Łuszczewska-Romahnowa criticized the probabilistic approach to the problem of
induction in [3] from 1957. After reconstructing the main assumption of that approach
she argues that they are not adequate. In her own opinion, even the most sophisticated
methods of inductive reasoning hardly enable us to obtain a true general knowledge
from true individual knowledge. Hence the probability of the conclusions of inductive
reasoning approach zero. This fact is testified by numerous examples from the history of
the natural sciences. Theories in such sciences are in a state of permanent change, they
undergo correction after correction, in the process of continuous confrontation with the
empirical data. The author claims that the real goal at which such sciences are oriented is
not the quest for absolute truth but rather to organize our scientific activity in such a way
that we can adjust our actions to the conditions imposed by the environment and we can
get a general orientation in that environment. Even false theories may have some cognitive
value: they pass away, but their life was not meaningless, so to speak. At least, they have
provoked us to change our world view, when their predictions collided with the empirical
data. To sum up: the goal of induction does not lie in the search of absolute truths, the
failure of induction should not be reduced to the fact that its conclusions do not follow
logically from its premisses. This view of induction bears thus a pragmatic stigma.
Among the methodological papers of the author one should also mention a short note
Czy filozofia obumiera? (cf. [7]) which may be considered as a defence of integrity of
philosophy. The author criticizes a popular view that philosophy remains unmatured and
is unable to solve its own fundamental problems. She points to a dynamic character
of philosophical reflection, recognizable for instance in clarification of philosophical
language due to the use of tools from logic. As an example, she mentions Ajdukiewicz’s
analyses of transcendental idealism.
Still another important paper is [11] from 1967 which contains a penetrating analysis
of the philosophical ideas of Kazimierz Twardowski. The author reports on Twardowski’s
program of scientific philosophy, discusses its main assumptions and shows how the
program in question influenced Polish philosophy.
2.3 Works on History of Logic and Methodology
Professor Łuszczewska-Romahnowa has translated into Polish the famous logical treatise
by Arnauld and Nicole La logique ou l’art de penser. The translation is accompanied by
her introduction containing an analysis of the philosophical thought of the seventeenth
century (cf. [4]). In particular, she stresses the differences between Pascal and Descartes
concerning the scientific method and the concept of knowledge itself. She has also
discussed the role played by the investigations more geometrico in European philosophy
of the seventeenth century.
Seweryna Łuszczewska-Romahnowa wrote also a short note [12] about the history of
logical investigations in Poznań (up to 1973). Our papers [17] and [18] are, in a sense, a
continuation of her work.
3 Views
Without any doubt the views of Professor Łuszczewska-Romahnowa were shaped under
the influence of the Lvov-Warsaw philosophical school. She was most close in her views
to Kazimierz Ajdukiewicz. This can be easily seen from the synthesis of analytical
philosophy and logic so characteristic for her style of writing.
According to Tadeusz Batóg, her political views (already before the World War II)
were decisively leftist, in spite of her own aristocratic background.
4 Resonance
Professor Łuszczewska-Romahnowa lectured on logic and the methodology of sciences

for a few generations of students in Poznań. She was responsible for all examinations in
philosophy obligatory for students applying for a PhD at the faculties of mathematics,
physics, chemistry, biology and geology of the Poznań University.
5 Pupils
Professor Łuszczewska-Romahnowa was a supervisor of three doctoral dissertations,

i.e. those of Tadeusz Batóg, Jerzy Czajsner and Mieczysław Jarosz. In turn, the first
of them supervised before his retirement four students (who have their own students)
so that Łuszczewska-Romahnowa has totally a dozen of scientific descendants (as of
2016). Tadeusz Batóg was her follower at the post of the chairman of the Department
of Mathematical Logic in the years 1974–1996, since 1996 the chairman is Roman
Murawski.
6 Summary
The life of Seweryna Łuszczewska-Romahnowa was heroic. Due to the horror of war and
Nazi persecution she experienced a tragedy of losing her husband, she was imprisoned
in the concentration camps, her health was ruined. Despite all this, she conscientiously
fulfilled all her academic duties during the three decades after the war. She is remembered
as a very modest person, always helpful and favorably disposed towards others. These
qualities of her personality, as well as her scientific achievements gained her a high esteem
in the academic community.
Acknowledgements Jerzy Pogonowski acknowledges the financial support of National Science Centre
[Narodowe Centrum Nauki] (grant No 2015/17/B/HS1/02232).
References
Selected Works by Seweryna Łuszczewska-Romahnowa
1. Wieloznaczność a j˛ezyk nauki [Polysemy and the language of science]. Kwartalnik Filozoficzny 17,
47–58 (1948) (Reprinted in: Jerzy Pelc (ed.) Semiotyka polska 1894–1969. Państwowe Wydawnictwo
Naukowe, Warszawa (1971))
2. Analiza i uogólnienie metody sprawdzania formuł logicznych przy pomocy diagramów Venna [An
Analysis and Generalization of Venn’s Diagrammatic Decision Procedure]. Stud. Logica 1, 185–213
(1953)
3. Indukcja a prawdopodobieństwo [Induction and probability]. Stud. Logica 5, 71–90 (1957)
4. Logika z Port Royal na tle historii logiki i metodologii XVII wieku [Port Royal logic from the
perspective of history of logic and methodology of the 17th century]. Introductory essay in: Arnauld
A., Nicole, P., Logika, czyli sztuka myślenia, pp. X–XXVI. Państwowe Wydawnictwo Naukowe,
Warszawa (1958)
5. Classification as a kind of distance function. Natural classifications. Stud. Logica 12, 41–81 (1961)
(Reprinted in: Przeł˛ecki, M., Wójcicki, R. (eds.) Twenty-Five Years of Logical Methodology in
Poland, pp. 341–373. D. Reidel, Warszawa-Dordrecht (1977))
6. Pewne poj˛ecie poprawnej inferencji i pragmatyczne poj˛ecie wynikania [On a certain concept of
correct inference and a pragmatic concept of entailment]. Stud. Logica 13, 203–208 (1962) (Reprinted
in: Pawłowski, T. (ed.) Logiczna teoria nauki, pp. 163–167. Państwowe Wydawnictwo Naukowe,
Warszawa (1966))
7. Czy filozofia obumiera? [Is philosophy withering?] Stud. Filozoficzne 1, 187–196 (1962)
8. Z teorii racjonalnej dyskusji [From the theory of rational discussion]. In: Rozprawy logiczne. Ksi˛ega
pamiatkowa
˛ ku czci profesora Kazimierza Ajdukiewicza, pp. 163–167. Państwowe Wydawnictwo
Naukowe, Warszawa (1964)
9. A generalized theory of classifications I. Stud. Logica 16, 53–74 (1965). Co-author: Tadeusz Batóg
10. A generalized theory of classifications II. Stud. Logica 17, 7–30 (1965). Co-author: Tadeusz Batóg
11. Program filozofii naukowej Kazimierza Twardowskiego [Kazimierz Twardowski’s program of scien-
tific philosophy]. Stud. Filozoficzne 4, 154–168 (1967)
12. Logika [Logic]. In: Labuda, G. (ed.) Nauka w Wielkopolsce, pp. 190–198. Wydawnictwo Poznańskie,
Poznań (1973)
Other Sources
13. Batóg, T.: Seweryna Łuszczewska-Romahnowa. Stud. Filozoficzne 1, 189–194 (1979) (Reprinted
in: Batóg, T.: Od Kanta do homofonów (Artykuły, autoreferaty, biografie, recenzje) [From Kant to
homophones. Articles, abstracts, biographies, reviews], pp. 232–237. Wydawnictwo i drukarnia Uni-
Druk s.j., Poznań (2015))
14. Batóg, T.: Romahnowa (Łuszczewska-Romahnowa). Polski Słownik Biograficzny Tom XXXI, z. 4,
575–576 (1989) (Reprinted in: Batóg, T.: Od Kanta do homofonów (Artykuły, autoreferaty, biografie,
recenzje) [From Kant to homophones. Articles, abstracts, biographies, reviews], pp. 237–239.
Wydawnictwo i drukarnia Uni-Druk s.j., Poznań (2015))
15. Batóg, T.: Seweryna Łuszczewska Romahnowa – logic and philosophy of science. In: Krajewski, W.
(ed.) Polish Philosophers of Science and Nature in the 20th Century. Poznań Studies in the Philosophy
of the Sciences and the Humanities, vol. 74, pp. 113–119. Rodopi, Amsterdam (2001)
16. Batóg, T.: Od Kanta do homofonów (Artykuły, autoreferaty, biografie, recenzje) [From Kant to
homophones. Articles, abstracts, biographies, reviews]. Wydawnictwo i drukarnia Uni-Druk s.j.,
Poznań (2015)
17. Murawski, R., Pogonowski, J.: Badania logiczne na Uniwersytecie Poznańskim w latach 1945–1955
[Logical investigations at the University of Poznań in 1945–1955]. Investigationes Linguisticae XIV,
68–77 (2006)
18. Murawski, R., Pogonowski, J.: Logical investigations at the University of Poznań in 1945–1955. In:
Bernhard, P., Peckhaus, V. (eds.) Methodisches Denken im Kontext, pp. 239–254. mentis Verlag,
Paderborn (2008)
R. Murawski ()
Adam Mickiewicz University in Poznań, Faculty of Mathematics and Computer Science, Department of
Mathematical Logic, Poznań, Poland
J. Pogonowski
Adam Mickiewicz University in Poznań, Institute of Psychology, Department of Logic and Cognitive
Science, Poznań, Poland
On Existential Dependence
and Independence in the World
of Thoughts and States of Affairs
(with Reference to Eugenia
Ginsberg-Blaustein’s and Roman
Ingarden’s Analyses)
Urszula M. Żegleń
Abstract The purpose of my paper, which has been inspired by Eugenia Ginsberg-
Blaustein’s analysis of the concepts of existential dependence and independence, is to
shed further light on the concept of existential dependence and connected concepts in
their applications to the mental sphere, namely to the analysis of mental content. To
realize this purpose, I shall extend Ginsberg-Blaustein’s analysis to Roman Ingarden’s
phenomenological approach with his characterization of the concepts of heteronomous
existence, existential dependence and other associated concepts.
Keywords Existential dependence · Thought · Mental content · Judgment · Act of

consciousness · Object of act · State of affairs
Mathematics Subject Classification (2000) Primary 03B40, 00A15; Secondary 81P05
1 Introduction
Eugenia Ginsberg-Blaustein’s paper On the Concepts of Existential Dependence and

Independence was originally published in Polish in 1931 in the anniversary collection of
papers presented in the Polish Philosophical Society in Lvov in the period 1904–1929 [2].
After half a century, her paper was translated into English and published in the collection
of papers edited by Barry Smith on logic and formal ontology [7]. Since then, the paper
has inspired further analyses, the example of which is the formal analysis given by
Marek Magdziak [23]. Ginsberg-Blaustein in her critical considerations of the concepts of
existential dependence and independence points out difficulties in their precise definitions
in the analyses given by Karl Stumpf, Alois Höfler, Kazimierz Twardowski, and especially
by Edmund Husserl, with reference to whom she raises serious objections concerning
inadequacy of his definitions. In her attempt avoid the difficulties, she gives her own

250 U. M. Żegleń
analysis, in which she distinguishes between two kinds of dependence and independence:
psychological and ontological. The first one is distinguished with regards to the possibility
or impossibility of separated presentation (Germ. Vorstellung), the second one—with
regards to the possibility or impossibility of separated existence. The main idea of her
analysis is reference to the concept of state of affairs and its subject and both concepts
are treated as primitive. She emphasizes the application of the analysed concepts in many
areas of philosophical inquiries, especially (1) in ontology—in the theory of objects (with
interesting reference to the theory of objects of higher order) and in the theory of parts
and wholes (to which she devotes special attention in her analysis), (2) in epistemology—
in the theory of acts and products, and (3) in the philosophy of language—in the theory
of meaning. Today the issue of existential dependence and independence belongs to the
fundamental topics of analytic metaphysics and phenomenological ontology.
The ontological analysis of dependence and independence and other connected con-
cepts was developed in detail by Roman Ingarden in his Controversy over the Existence
of the World (in the first two volumes originally published in Polish in 1947, vol. I and
in 1948, vol. II).1 Ingarden provides the most precise characterization of these concepts,
which are treated as existential moments and are crucial ontological categories in his
inquiries into the controversy between realism and idealism in question of the existence
of the world. In Ingarden’s analysis, the notions ‘dependence’ (Pol. ‘zależność’, Germ.
‘Abhändigkeit’) and ‘independence’ (Pol. ‘niezależność’, Germ. ‘Unabhändigkeit’) have
different meanings than the notions ‘samoistność’ and ‘niesamoistność’, which were
originally used by Ginsberg-Blaustein in her paper. The translator of her paper justifies
his choice of translation and makes some helpful remarks in which he stresses that
the Author followed the Husserlian rather than Ingardenian terminology, and applies
Ginsberg-Blaustein’s own convention of her interpretation of Ingarden’s notions.
In my paper I shall extend Ginsberg-Blaustein’s analysis to Ingarden’s definitions, and
use them in my analysis, applicable to the mental sphere. The purpose of my paper is the
application of these basic concepts to the analysis of mental content. Today the problem of
mental content belongs to the most topical questions in the philosophy of mind. There are
many various theories of mental content which are developed in a new context of research
provided by cognitive science. I want, however, to make use of these philosophical means
of analysis, especially those from Ingarden’s theory, in the characterization of mental
content considered from the epistemological-psychological point of view.
1 Inspite of the fact that Ingarden as phenomenologist was far from developing philosophy in the manner
of the Lvov-Warsaw School he was connected with the Lvov School. He started his philosophical
education under Kazimierz Twardowski (who, like Husserl was a student of Franz Brentano), whose
classes he attended in the period 1911–1912. His further academic career was associated with Lvov,
where in 1924 he received his Habilitation and in 1925 he was employed as Private Docent at Jan
Kazimierz University. In 1933 he received the position of Associate (‘Extraordinary’) Professor and
was appointed Head of the Chair of Philosophy, succeeding Kazimierz Twardowski. The Second World
War dramatically interrupted educational and research activity at Jan Kazimierz University. In 1940 the
university was renamed to Ivan Franko University and Roman Ingarden taught there until 1941. Eugenia
Ginsberg-Blaustein, together with her husband Leopold Blaustein, belonged to the second generation of
Twardowski’s pupils (who were also Kazimierz Ajdukiewicz’s pupils) and like her husband was under the
influence of Ingarden’s research. The Blausteins shared the tragic fate of many other Jews, and together
with their son, they were executed by the Nazis.
On Existential Dependence and Independence in the World of Thoughts. . . 251
The structure of my paper is as follows:

1. First, I shall give some basic definitions and theorems from Roman Ingarden’s
ontology.
2. Then, I shall use them in my analysis of mental content.
3. I shall finish my considerations with a brief conclusion, presenting the results which
should contribute to the current research in the philosophy of mind.
2 Ingarden’s Definitions and Theorems
I start my considerations with some chosen definitions and theorems from Ingarden’s
ontology. In some cases, I shall not present them in exactly the same formulations which
they have in Ingarden’s original text, nevertheless, the formulations, while sufficient for
the needs of my analysis, will retain the senses intended by Ingarden.
In his Controversy over the Existence of the World [20, p. 109] Ingarden distinguishes
four different pairs of opposite existential moments which are:
1. autonomy—heteronomy
2. originality—derivativeness
3. selfsufficiency—non-selfsufficiency
4. independence—dependence.
These existential moments of a given object always occur within the framework of
modes of existence of this object and are distinguished only in abstraction. Starting
his ontological considerations with the analysis of existential moments, Ingarden as a
phenomenologist makes no claim about the existence of the objects considered, but
merely attempts to discover which existential moments are proper for particular modes
of existence, such as: absolute, real, ideal and intentional.2
I shall start with the definition of a self-existing object.3 The notion of self-existence
is treated by Ingarden as synonymous to ‘existential autonomy’ and such a translation is
applied by the translator Arthur Szylewicz in the recent English edition of vol. I of The
Controversy over the Existence of the World [20].
A definition of self-existence, i.e. existential autonomy
D1. Something is selfexisting, i.e. existentially autonomous, if it has its existential
foundation within itself [20, p. 109].4
2 The philosophers who work on Ingarden’s ontology considers the problem of the interpretation of the
modes of existence (sometimes attempting to interpret them improperly in the category of relations or
properties). Ingarden was, however, clear in this issue, and his pupils and associates provided elucidative
responses to those erroneous interpretations.
3 This is an exact translation of the Polish notion ‘samoistny’. This translation was proposed by Helen
Michejda in her first English translation of some fragments of Ingarden’s Spór o Istnienie Świata. Later,
however, she changed it for ‘existential autonomy’ [18]. See the remark made by A. Szylewicz in footnote
247 of [20].
4 Ingarden writes down his definitions as conditional statements. I am, however, more inclined to write
them down in the form of equivalence using the sign ‘iff’ (if and only if) in place of ‘if’. It seems to
It has its existential foundation within itself, if it is immanently determined within itself.
This determination is thanks to its own essence (in Husserl’s terminology Eigenwesen)
[20, p. 115]. In Husserl’s view, the essence of an object is its constitutive nature.
Ingarden, taking this assumption from Husserl, adds—in agreement with his three-
aspectual definition of an object—that the essence of an object consists of its properties
together with formal and existential moments [17, p. 208]. Earlier in Essentiale Fragen
[15] he defined the essence of an individual object only as the material (qualitative)
equipment of this object. In consequence he propounds the following thesis:
T1. A selfexisting individual object is unambiguously completely determined in every
aspect of its qualitative equipment.
Thanks to this completeness it is comprehensively separated from other entities, and as
separated and total, it is formally perfectly closed in itself (in the particular case of a
real object, it is spatially closed). Nevertheless, some selfexisting objects (real objects,
for example) can enter into interaction with their environment thanks to their material
(qualitative) properties.
A definition of non-self-existence, i.e. existential heteronomy
D2. Something is non-selfexisting, i.e. exists heteronomously, if it has its existential
foundation ‘outside of itself’, i.e. in something else [20, p. 109f].
In consequence, it has such essential properties which have been given to it from outside.
It means that it does not have them within itself. The above definition allows us to
formulate the following thesis:
T2. Every property is existentially non-selfsufficient in regard to an object to which it is
ascribed.
A definition of an existentially original object
D3. An object is existentially original if, in accordance with its essence, it cannot be
produced by any other object [20, p. 118].
Thus, originality is ascribed only to an autonomous object the essence of which forces it
into existence [20, p. 118]. But an autonomous object does not have to be existentially
original (for example a real object is autonomous, but not original).
A definition of an existentially derivative object
D4. An object is existentially derivative if it can or does exist in virtue of having been
produced by another object [20, p. 118].
An existentially derivative object can be either existentially autonomous or heteronomous.
This is why there are different kinds of derivativeness. Each derivative object is existen-
tially imperfect which is seen not only in derivativeness of its existence, but also in its
me more adequate for definitions if only ‘something’ is not understood in the narrow sense of an object
as a thing. Ingarden, however, uses the notion ‘object’ where his translator Szylewicz uses the notion
‘entity’. In D1 Ingarden in fact, uses the word ‘something’ (Szylewicz uses here the word ‘entity’), but
his comment given in a footnote makes it clear that Ingarden is aware of danger of its antinomy and treats
‘something’ merely as a helpful abbreviation. On the analysis of Ingarden’s ontology see also [25].
fragility and brittleness of existence. As Ingarden says ‘no derivative object can exist
[. . . ], but once it has originated (for whatever reason) and does exist, it can always cease
to exist’ because its own essence does not sustain it in existence [20, p. 144f].
A definition of an existentially selfsufficient object
D5. An object is existentially selfsufficient, if in accordance with its essence, it requires
for its existence no other entity which would have to coexist with it within the unity
of some whole [20, p. 147].
A definition of existential non-selfsufficiency
D6. Something is existentially non-selfsufficient if, as implied by its essence, its exis-
tence involves a necessary coexistence with some other entity [20, p. 147].
Non-selfsufficiency has different degrees, for example, the colour red as a feature of
an individual red object (let’s say the red colour of the rose in the vase on my table) can
be distinguished from the moment ‘redness’ in the colour red. In first case, the red colour
coexists with an object (this rose, for instance) as its bearer, in the second one (where non-
sufficiency is of a higher degree) the moment ‘redness’ (beyond its bearer) coexists also
with colouration within the whole of ‘red colour’. Non-sufficiency—as is emphasized by
Ingarden—is characterized by peculiar relativity [20, p. 149]. There are different kinds
of existential non-sufficiency the variations of which are distinguished with regard to
different criteria [20, p. 152].
Within the framework of existential selfsufficient objects, Ingarden distinguishes
between existentially dependent and independent objects.
D7. An object is existentially dependent if in order to continue its existence it requires
another selfsufficient object [20, p. 153].
D8. An existentially selfsufficient object is independent if, in virtue of its essence it
requires no other selfsufficient object for its own existence. In this sense it is
absolutely independent [20, p. 153].
The above definitions have their application in distinguishing among different modes of
existence of objects. Because my further considerations will be focused on mental sphere
with its intentional objects I shall refer also to Ingarden’s definitions of pure intentional
objects. Speaking about pure intentional objects Ingarden distinguishes between primary
and secondary ones.
A definition of an intentional object
D9. An object is primary purely intentional if it has been directly or indirectly produced
by someone’s acts of consciousness in such a way that thanks to their immanent
intentionality it has its source of existence and complete equipment (as it is, for
example, in someone’s imagination).
D10. An object is secondary (as derived) purely intentional if its intentionality is given
to it (as for instance sense of utterances, linguistic expressions).
An intentional object has a two-sided formal structure, that is: (1) the content consists of
the non-intuitive content which belongs to the act of supposition, and (2) the intentional
structure. Ingarden (like Husserl as well as representatives of the Lvov School) makes also
distinctions between judgments and suppositions in regard to the function of assertion.
Judgments in strict sense are sentences which have the function of assertion, while
suppositions lack this function. This division has its grounds in Platonic epistemological
distinction between episteme and doksa (opinion).
I shall stop with these initial formulas without further developing Ingarden’s excep-
tionally rich analysis, and I shall go to on to analyse the mental sphere of thoughts.
3 Dependence in the Mental Sphere of Thoughts
The subject of my analysis is the content of someone’s beliefs, wishes, desires, etc.
simply someone’s thoughts, not in the Fregean sense, but rather in the epistemological-
psychological approach, when thoughts occur in someone’s process of thinking. At
the beginning, it is useful to make some terminological remarks concerning the basic
notions such as: judgment, proposition and basic assumptions. I define judgment in the
standard sense of epistemology as the product of someone’s cognitive process of judging,
presented in a propositional structure expressing its content which is affirmed or rejected
by a given cognitive subject. I mean here only assertoric judgements. Affirmation or
rejection is treated as a certain kind of objective operation of the cognitive subject (not
psychologically where it would be subjective). This is objective as a cognitive operation in
regards to reference to independently existing reality (when, for instance, someone asserts
that in 2016 Poland celebrates 1050th anniversary of its baptism, and this is a fact which
took place independently of this judgment). Reference to independently existing reality
allows the philosopher evaluate a judgment, which in this aspect is treated the same as
proposition in logical semantics. It can be useful here to assume the Kantian approach
according to which proposition (Germ. Satz) is semantic content of judgment (Germ.
Urteil).5 Although Kant’s treatment of judgment is not limited to assertoric (theoretical)
judgments I am close to his cognitivism according to which judgment is a specific kind of
cognition (Germ. Erkenntnis).
I make a distinction between judgment and belief. I use the notion ‘judgment’
appealing to this objective operation leading to objective knowledge according to usual
standard criteria.6 In consequence of this, each (assertoric) judgment is true or false in
the system of objective knowledge, and not only of someone’s knowledge. Beliefs are
typically elements of someone’s knowledge and as such I admit the case when someone
is convinced about the truth of someone’s belief while it is false. In logical analysis,
this is distinguished by the propositional attitude of having knowledge which is drawn
down by the operator ‘K’ (‘to know’) and by the attitude of having a belief which in
turn is drawn down by the operator ‘B’ (‘to believe’). The philosophers of mind who
espouse analytic philosophy respect this distinction without referring to the traditional
epistemological concept of ‘judgment’, only to the logical notion of ‘proposition’. The
notion of ‘judgment’ is used by analytic philosophers, but in epistemology and in the
philosophy of science, it is used mainly in reference to the issue of justification. As
5 There is an extraordinarily rich literature on Kant’s theory of judgments and numerous comments on his
treatment of the relation between judgment and proposition (here in the sense of its content).
6 I do not discuss here the problem of these criteria which—as it is known—belongs to the central topics
of analytic epistemology.
it is known from philosophy, the issue of judgment has a long history, and convey
lots of particular problems. At the beginning of the twentieth century, the concept
of judgment analysed in connection with the concept of proposition was especially
important in overcoming logical psychologism (which was observed in Husserl’s Logical
Investigations 1900/01 [11] and his Formal and Transcendental Logic 1929 [10]).
As my starting point, I assume simply that our thoughts have contents (which
can be referred to the Brentanian tradition) and thoughts require a thinking subject
(a Cartesian approach). However, my approach is not Cartesian because contrary to
Cartesian internalism I take the externalist position according to which contents (in the
sense of Germ. Inhalt) of our thoughts are acquired through interactions with external
environment. It is my answer to the question ‘what causes the content of our thoughts?’ If
I were asked whether the subject would have any thoughts without any interactions with
its external environment, including other persons, I would reply ‘no, it wouldn’t’. It does
not mean that I exclude any internal determination because in order to have thoughts the
subject must have a proper ‘cognitive equipment’ (which is described by disciplines of
cognitive science). However, I want to emphasize the distinction between searching for
the explanation of having thoughts and for the explanation of their contents. The first one
requires referring to the cognitive ‘equipment’, i.e. the faculties of the cognitive subject,
while the second one requires mainly the research concerning the character of mental
content (first of all its nature and structure). Hence, the explanation of having thoughts
can be advanced on the grounds of philosophical epistemology and it is the aim for
philosophy of mind and cognitive science. Especially important in this research both for
philosophers and scientists, is the Kantian question ‘how cognition is possible?’, which
can be reformulated here as ‘how having (conceptual) thoughts is possible?’, in other
words ‘what is required in order to have thoughts?’, ‘what conditions should be satisfied
by the subject?’.7 In the second issue concerning the mental content the explanation is
searched for today also in the philosophy of mind, but in connection with research in
cognitive semantics. In both kinds of issues I shall refer to Ingarden’s research.
At the level of epistemological considerations in which I refer to the Brentanian
tradition, we can say that thoughts are produced in acts of consciousness. In my
considerations I take into account only such processes of thinking in which someone
consciously holds certain beliefs, desires, makes some decisions, makes certain plans,
etc., although they are connected with other beliefs, desires, etc. of a given person who
does not need to be aware of them at present. I assume that beliefs, wishes, desires, etc.
have propositional (i.e. conceptual) character. This means that the content of the mental
state of believing, judging, desiring etc. is in a form of proposition (it is a propositional
content). Immediately the problem raises whether a propositional content is mental, as I
have already stated announcing the subject of my interest, which is the mental content, or
whether it is abstract as proposition itself. It is an ontological question. Today both issues:
the issue of mental content and the issue of proposition are controversial. In philosophical
considerations, even the concept of proposition is not so clear as it is in formal semantics
(where it is a primary bearer of truth-value) and its definition poses many problems and
7I purposefully use here the term ‘cognition’, not ‘knowledge’ in reference to the German word
‘Erkenntnis’ although Kant’s well-known question from his Critique of Pure Reason is usually quoted
as ‘What can I know?’, and further ‘What should I do?’ and ‘What may I hope?’.
provokes discussions even today (e.g. [13, 14]). In order to avoid confusion I reply that
in my approach the propositional content is mental, but only in the sense that it is the
content (just a mental content) of someone’s mental state. Because I take into account
only the structured content in a propositional form this means that also a process of
thinking which is taken here into account is structured and consists of linguistic units.8
Mental content is expressed in them in propositions. I do not prejudge here whether every
process of thinking is linguistic, although I am interested only in this kind of processes.
This is why I take into account conceptual thoughts which in the process of thinking
have propositional forms.9 I assume that propositions in their nature are primary logical
items (well-formed both syntactically and semantically), thus mind-independent. For my
analysis here, it is enough to say that propositionally structured mental content (i.e. in
proposition) is shareable by different subjects as well as by the same person at different
times. In virtue of their logical nature, propositions (as mind-independent) are treated as
constituents of objective knowledge (without any cognitive subject, as it might be said in
Popper’s terms).
However, in my considerations I deal with propositional thoughts of cognitive subjects.
A given cognitive subject entertains different attitudes to propositional content, such as:
affirmation or rejection (for judgments), doubt (for problems), duty (for decisions for
instance), etc. Someone’s thoughts, considered here in propositional forms, can also be
linguistically expressed by a given person in speech or in written language. A linguistic
form gives them the attribute of inter-subjectivity, and in this sense—objectivity. Thoughts
given in a propositional form (performed as judgments) develop certain states of affairs,
appropriate to their contents. I shall refer here to Ingarden’s theory, according to which
propositional content develops a state of affairs, proper for this propositional content,
which is a pure intentional equivalent, so-called objectum formale of a given judgment.10
Thus, the intentional state of affairs is never selfexisting (autonomous), but heteronomous
with regard to its judgment. If, however, a propositional content is also satisfied by a state
of affairs which occurs independently of the occurrence of that judgment, then we have to
do with an objective state of affairs, so-called objectum materiale of this judgment, which
has not only a function of predication as a sentence, but also a function of assertion.
Assertion is a proper function of judgment thanks to which a given judgment is true.11 I
8I do not refer to any phenomenal content which is beyond my interest here.

9 The issue of the relation between thought (or process of thinking) and language belongs to the crucial
topics of contemporary debates in analytic philosophy, cognitive linguistics and psychology. In the Polish
literature an outstanding monograph on this matter is the book by Arkadiusz Gut [8].
10 The issue of states of affairs has been developed under the influence of philosophical ideas of G.W.
Leibniz, B. Bolzano, G. Frege, H. Lotze, the Brentanists (as K. Stumpf, A. Marty), and especially of
Husserl, who dealt with it not only in epistemology, but also in formal ontology. The concept of the state
of affairs (facts) has played an important role in formal logic and also in the philosophy of logic, thanks to
L. Wittgenstein and B. Russell. In the philosophy of logic the state of affairs is treated as an abstract object
existing in possible worlds, identified with proposition or distinguished from it as its correlate, defined
also as a possible object of entertainment or assertion. In Logical Investigation [11] Husserl proposed the
concept of the state of affairs (Germ. Sachverhalte) as an objective correlate of judgment (Germ. Urteil).
In the rich literature on this topic see the overview papers by Chisholm [3], Smith [24], and a monographic
book by Armstrong [1].
11 Ingarden distinguishes between different kinds of judgments according to their formal structure, which
is most evident in his analysis concerning categorical judgments and conditional judgments. Only the
assume that also other kinds of thoughts (being someone’s desires, decisions, etc.), if only
they are given a propositional form, develop, thanks to their content, intentional states of
affairs, appropriately to their formal structure.
With reference to psychological considerations, I assume the occurrence of proposi-
tional content also in the case of a child who still does not know any ethnic language,
but if only it is able to say something, even only syllables (for example, ‘ma-ma’) in
reference to something in a meaningful way, then such a phrase can be treated as an
abbreviation of a sentence (‘this is the mother’) which expresses a proposition (as the
meaning of this sentence). I take into account such a level of the cognitive development
of the child at which it is able to have conceptual thoughts. I assume the same in the case
of some impairment in verbal communication, for instance in Broca’s (motor) aphasia
when a person is not able to communicate verbally in a proper way, but can communicate
by means of single words, which are meaningful for her and a receiver tries to recognize
their meaning.
Further, I assume that someone’s beliefs constitutes his (or her) knowledge. This
knowledge, enriched by the subject’s wishes, desires, decisions, etc., constitutes the
subject’s mental life. In our body of knowledge, we can have a great number of shared
beliefs which function in our common communication or belong to objective scientific
knowledge. We can also share some desires, wishes, have similar intentions, make the
same or similar decisions in a given situation. Some of them are common, others are
quite subjective. Each person has their own personal experience, their own mental life,
their own unique personality and identity, but can develop only by interactions with other
persons. What needs to be emphasized here is the difference between the psychological
approach (or the approach of psychophysical epistemology) and the approach of phe-
nomenological pure epistemology. In the latter (developing in the Cartesian and Kantian
tradition) the cognitive subject is a pure subject, namely pure ‘self’ being merely a
‘satisfier’ of his own conscious acts. Such an approach to the cognitive subject is, however,
controversial, even among phenomenologists themselves which is seen in discussions
between Ingarden and Husserl (when Husserl in his Ideas [12] was in favour of the
existence of pure ‘self’).
Now I turn to the basic ontological notions by means of which I shall further analyze
mental content, which is treated here as belonging to someone’s thought. In the approach
analyzed, someone’s thought has no guarantee of its existence in itself, but finally in acts
of consciousness of the cognitive subject. At the epistemological level in the structure of
consciousness, following Ingarden’s analysis (based on Twardowski’s distinction, made
also by Husserl in his Logical Investigations [11]), I distinguish between act, its content
and an external object, pointed by the content thanks to an intentional moment of a given
first ones satisfy the function of assertion. Beyond assertoric judgments Ingarden also intended to analyze
modal (problematic and necessary) judgments. This kind of analysis could be interesting when asking
a metaphilosophical question concerning the relationship between the epistemological and semantic
approaches (here possible world semantics). The difference between the epistemological notion of
‘judgment’ in a philosophical theory of modal judgments and the logical notion ‘proposition’ is blurred
when modality is ascribed to the linguistic unit or to the subject of judgments, i.e. to states of affairs, but
it remains relevant if modality is treated in the Kantian sense and then it is associated with the attitude of
judging subject. On the application of possible world semantics to Ingarden’s ontology see [22].
act.12 Thus, the mental content considered as the content of someone’s conscious act is
not selfexisting, but heteronomous with regard to this act. If we have to do with an act of
judging, then its mental content constitutes the content of judgment as the product of this
act. As it has been already said, the content of judgment develops an intentional state of
affairs. But as it has been also said, judgment in its function of assertion maintains a state
of affairs which exists independently in the domain in which a given judgment ‘locates’
it. It is the domain of real or ideal objects. According to Ingarden’s ontology, a domain is
‘natural’ (autonomous) plurality of autonomous objects. As it is known, a large body of
Ingarden’s research pertains to the theory of literary work, in which he analyzes the so-
called quasi-judgments (for example, ‘Oedipus loves Jocasta’, ‘Macbeth murders the king
Duncan’) with their reference to intentional objects. Intentional objects, however, do not
constitute any domain in the strict sense of Ingarden’s definition of this term (nevertheless,
quasi-judgments can be evaluated according to the world presented in fiction).
Further, it is worth to emphasize Ingarden’s radically realistic approach, according to
which no act of consciousness is able to ‘burst into’—as he says metaphorically—the
domain of real objects and to make any changes in them [16, p. 557], [19, p. 466]. There
is no existential connection between them, only a purely mental assignment. Thus, both
spheres are in the above ontic sense existentially selfsufficient (mutually separated). In this
way he defines real transcendency between the domain of real objects and consciousness.
But also the domain of ideal objects is selfsufficient, separated and independent from acts
of consciousness. I do not claim that there are any ideal objects in the sense of Ingarden’s
ontology (i.e. ideas and individual ideal objects). Nevertheless, I maintain that Ingarden’s
definition of objective correlates of judgments referring to the ideal domain can be applied
in the domain of science, where scientific language is used in mathematical modelling of
different aspects of reality (its physical processes or events). These models, being highly
abstract in their character, are idealizations of the examined reality.
In my considerations, as I have stressed, I take into account not only judgments (in the
strict sense of ‘judgment’), but all these thoughts which are meaningful and expressed in
a propositional form. As expressed in a propositional form they are products of certain
linguistic operations, roughly speaking—operations for sentence production (as used in
Ingarden’s terminology). As meaningful, they are in turn products of significant acts
which ascribe meaning to them. Following Ingarden’s theory, the meaning of a linguistic
unit (in which I also include propositional thoughts) is a purely intentional product of
object apprehension consisting of a direction indicator, existential moments of a given
object (whether it is real or ideal or only intentional, for example), its existential position
(i.e. affirmation or not affirmation in existence) and its formal and material content.
Formal content determines the formal structure of an object (whether it is an individual,
an event, a process or something else) as well as the logical and grammatical form of
a given linguistic unit, and material content determines a qualitative characterization
of a given object. Ingarden’s theory of meaning like traditional theories starts with
the characterization of nominal meaning (meaning of names) as selfsufficient (but not
autonomous, in the sense of semiotics—not selfinforming) linguistic units. Contrary to
12 Imake only very simplified remarks on intentionality in reference to Ingarden’s theory, which is much
richer than Brentano’s and even than Husserl’s theory of intentionality. On certain comparative analysis
of Ingarden’s and Husserl’s theories see: Artur Chrudzimski [4] among others.
Ingarden, my interest is more focused on the meaning of sentences (not names [21])
because in my view the basic structure of our thoughts is already propositional. In this
point, I direct my considerations more towards the Fregean compositional semantics
which in this aspect is closer to current approaches.
Looking, however, from the epistemological point of view, meaning is not only
considered semantically, but as intentional it is prescribed to linguistic units in significant
acts of consciousness. In this way the intentional correlates of meaning are heteronomous,
existentially dependent and derivative; primarily from the meaning itself, and secondarily
from given significant acts.
A reader of this paper can ask here whether ‘meaning’ (sense) is the same as ‘mental
content’? If it were the same, then meanings would be only something mental in our
heads [5]. Because I take advantage of Ingarden’s conception, again I refer to his view.
Ingarden attempts to grasp all aspects of meaning, that is, both its stable element and its
dynamicity which is an effect of all changes of language. This is why the meaning of
linguistic units is characterized as a complex structure (which has already been shown by
its constituents). This is not all, because the meaning of linguistic units also depends
on their context, both linguistic and situational. From a psychological point of view,
situations in which a cognitive subject finds themselves have influence on his (or her)
mental states, not only cognitive acts, but also emotional or sensitive ones. The content
of someone’s mental state be expressed linguistically in an adequate linguistic form and
choice of utterances. In a speech act also the way of uttering sounds, the timbre and
tone of someone’s speech and the behaviour accompanying the speech act have a role
in expressing the content of someone’s thoughts. In different situations someone can
manifest or hide his or her mental states. Thus, the meaning which is associated with
utterances is not the same as the mental content of someone’s thoughts although we
can take into account the same schema for their characterization. Nevertheless, in the
cognitive function, judgments and sentences which are used to perform the content of
thoughts are transcendent with regard to someone’s experiences. Then also the mental
content in its propositional form can remain the same in repeated states of judging at
different times or by different persons. As such it becomes a certain stable element of
meaning. Under Husserl’s influence, in his early works Ingarden assumes the occurrence
of the so-called ideal concepts. In spite of his later intentional conception of meaning, he
maintains a certain stable element of meaning without which no communication would
be possible. What is also very important, his intentionalism in reference to meaning does
not imply any psychological conception of meaning, but on the contrary, it is important in
argumentation against psychologism, as in the case of Husserl13 and late Twardowski.14
13 Husserl only in his early work, which is Philosophie der Arithmetik (1891) [9] manifested traditional
psychologistic tendencies, which immediately met objections from Frege. But according to Dan Zahavi’s
recent view, Husserl’s criticism of psychologism and his own antipsychologistic position was not
motivated by Frege’s objection, but was an effect of his studies on H. Lotze and A. Pfänder writings
[27]. Husserl’s conception of meaning as an ideal object, in turn, is interpreted in the sense of logic which
is close to Frege’s theory, and not in the ontological Platonic approach.
14 It is, however, not quite clear whether Twardowski overcame psychologism—as Jan Woleński [26, p.
41] pointed out referring to Izydora Dambska

˛ who observed that Twardowski’s approach to meaning is
not quite autonomous in regards to experience [6]. Ginsberg-Blaustein appeals to Twardowski’s analysis
of existential dependence and independence, but only in question of metaphysical parts and the whole.
From an externalist point of view, I assume that the mental content of our thoughts
is determined by objective states of affairs, situations, events or facts. But I also admit
inner determination. At the internal level, I take into account biological factors of our
organisms, which occur in various consciously experienced states of the organism and
psychological states. In the first ones, I include, for instance being tired, thirsty, etc. If a
person is aware of these states, and focuses on them, then she is in a state of sensitive inner
perception (as Husserl and Ingarden maintain). Thus these states can also determine the
mental content of such thoughts as ‘I feel tired’, ‘I am thirsty’ and so on. I do not use the
notion ‘determine’ in a strict sense of determination because the mechanism for sentence
formation belongs to cognitive operations and it is not the same mechanism, which is
biologically activated when the organism is in the state of being tired, thirsty and so on. I
rather want to say that the cognitive subject must have special faculties to perceive such
states, to be aware of them, to distinguish among them, and to express them in a linguistic
form. The inner state of the organism is here in a certain analogy to an external state of
affairs to which someone refers when, for example, he or she thinks about something in
their environment. In spite of the fact that the state of being thirsty, as a sensitive state, is
intuitive (in the sense of Twardowski, Husserl, Ingarden), the cognitive subject is able to
associate his or her conceptual thoughts with it. Thus, a cognitive subject who has among
his or her faculties the faculty of producing conceptual thoughts is able to refer to this
state of his or her organism in the propositional form of his or her thought.
In the second kind of states—as has already been said—I include the states of emotion,
feelings, as well as earlier experiences. In this case Husserl and Ingarden, talk about
non-sensitive inner perception, as for instance ‘I am happy’, ‘I am depressed’ and so
on, and not about sensitive inner perception. Also in these cases, I take into account
only the propositional mental content which, in fact, is contained in someone’s thoughts
associated with these states, although their content is non-intuitive (as it is stressed in
phenomenological epistemology and Twardowski’s works).
Beyond that, as in a standard semantic analysis, I distinguish between content and a
bearer of content. I say that the bearer of mental content is someone’s thought. But at the
neuronal level the vehicle for this meaningful thought is a certain configuration of neural
network. One of the most important aims of interdisciplinary research in cognitive science
is just to resolve the Kantian problem by showing how all these levels are integrated in
the cognitive subject in his or her mental life, if only restricted to having propositional
thoughts. But this is a topic for another discussion.
4 Concluding Remarks
In this paper, which was inspired by Ginsberg-Blaustein’s analysis of the concepts of

existential dependence and independence, I took advantage of Ingarden’s analysis in
application to my considerations on mental contents occurring in the states of judging,
believing, etc. For the purpose of my considerations, I established the basic terminology
(to make a distinction between judgment, belief, proposition on the one hand, and
conscious act, its mental content and its object—on the other hand), and I presented some
epistemological and ontological assumptions of Ingarden’s ontology. The results of my

considerations can be briefly summarized as follows:
1. Referring to the analytical tradition of the Lvov School and Ingarden’s phenomenolog-
ical approach allows us to see better the ontological assumptions which are important
in the analysis of mental content.
2. Mental content occurring in the states of judging, believing, desiring, etc., has a
propositional character and as such it can be sharable by different subjects or by the
same person at different times.
3. Following Ingarden’s view, I claim that in spite of the fact that mental content is
propositional, it is not the same thing as linguistic meaning.
4. Mental content is determined both externally (by external objects and states of affairs
which are cognitively accessible to the cognitive subject) and intrinsically thanks to
special cognitive faculties of the cognitive subject who, being in different states (not
only cognitive, but also sensitive) is able to produce conceptual thoughts.
Acknowledgements My special thanks are due to Dr Tomasz Fojt for his proofreading and helpful
suggestions and to Mateusz Klonowski for his technical support.
References
1. Armstrong, D.: A World of States of Affairs. Cambridge University Press, Cambridge (1997)
2. Blausteinowa z Ginsbergów, E.: W sprawie poj˛eć samoistności i niesamoistności. In: Ksi˛ega
Pamiatkowa
˛ Polskiego Towarzystwa Filozoficznego we Lwowie 12.II.1904–12.II.1929. Polskie To-
warzystwo Filozoficzne, Lwów (1931) (English translation: Ginsberg-Blaustein, E.: On the concepts
of existential dependence and independence. In: Smith, B. (ed.) Parts and Moments. Studies in
Logic and Formal Ontology, pp. 265–287. Philosophia Verlag, Munich (1982). English translation
P. Simons)
3. Chisholm, R.: States of affairs again. Noûs 5, 179–189 (1971)
4. Chrudzimski, A.: Teoria intencjonalności Romana Ingardena [Roman Ingarden’s theory of intention-
ality]. Edukacja Filozoficzna 25, 249–262 (1998)
5. Chrudzimski, A.: Are meanings in the head? Ingarden’s theory of meaning. J. Br. Soc. Phenomenol.
30(31), 306–326 (1999)
6. Dambska,
˛ I.: Semiotyczne koncepcje w filozofii Twardowskiego [Semiotical conceptions in Twar-
dowski’s philosophy]. Ruch Filozoficzny 27(1), 1–9 (1969)
7. Ginsberg-Blaustein, E.: On the concepts of existential dependence and independence. In: Smith, B.
(ed.) Parts and Moments. Studies in Logic and Formal Ontology, pp. 265–287. Philosophia Verlag,
Munich (1982). English translation P. Simons
8. Gut, A.: O Relacji Mi˛edzy Myśla a J˛ezykiem. Studium Krytyczne Stanowisk Utożsamiajacych
˛ Myśl
z J˛ezykiem [On the Relationship between Thought and Language. A Critical Study of Approaches
That Identifies Thought with Language]. KUL, Lublin (2009)
9. Husserl, E.: Philosophie der Arithmetik. Psychologische und Logische Untersuchungen. In: Eley, L.
(ed.) Gesammelte Werke. Husserliana, vol. 12. M. Nijhoff, The Hague (1891/1970)
10. Husserl, E.: Formal and Transcendental Logic. M. Nijhoff, The Hague (1969). Transl. D. Cairns
(original work published 1929)
11. Husserl, E.: Logical Investigations. Routledge, London (1970, repr. 2001). Trans. J.N. Findlay from
the second German edition 1913
12. Husserl, E.: Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy. First
Book: General Introduction to Pure Phenomenology. M. Nijhoff, The Hague (1982). Trans. F. Kersten
(original work published 1913)
13. Iacona, A.: Propositions. Name, Genoa (2002)

14. Iacona, A.: Are there propositions. Erkenntnis 58, 325–351 (2003)
15. Ingarden, R.: Essentiale Fragen. Ein Beitrag zum Problem des Wesens. Jahrbuch für Philosophie und
Phänomenologische Forschung 7, 125–304 (1925) (Polish translation in: Ingarden, R.: Z Teorii J˛ezyka
i Filozoficznych Podstaw Logiki [From the Theory of Language and Philosophical Foundations of
Logic], pp. 327–482. Państwowe Wydawnictwo Naukowe [Polish Scientific Publishers], Warszawa
(1972))
16. Ingarden, R.: Spór o Istnienie Świata [Controversy over the Existence of the World], vol. II, 2nd edn.
17. Ingarden, R.: Spór o Istnienie Świata [Controversy over the Existence of the World], vol. I, 2nd edn.
18. Ingarden, R.: Time and Modes of Being. Charles Thomas, Springfield, IL (1964). Partial transl. of
Ingarden, R.: Essentiale Fragen. Ein Beitrag zum Problem des Wesens. Jahrbuch für Philosophie und
Phänomenologische Forschung 7, 125–304 (1925) by H. Michejda
19. Ingarden, R.: Z Teorii J˛ezyka i Filozoficznych Podstaw Logiki [From the Theory of Language
and Philosophical Foundations of Logic]. Państwowe Wydawnictwo Naukowe [Polish Scientific
Publishers], Warszawa (1972)
20. Ingarden, R.: Controversy Over the Existence of the World, vol. I. Peter Lang Edition, Frankfurt a.
Main (2013). Translated and annotated by Arthur Szylewicz
21. Jadacki, J.J.: O pogladach
˛ Romana Ingardena na j˛ezyk. Stud. Semiotyczne 5, 22–54 (1974). English
edition: On Roman Ingarden’s semiotic views. Analecta Husserliana 27, 523–540 (1989)
22. Krysztofiak, W.: The platonic idealism controversy in light of possible worlds semantics. Axiomathes
17(1), 75–90 (2007)
23. Magdziak, M.: A logical analysis of existential dependence and some other ontological concepts.
A comments to some ideas of Eugenia Ginsberg-Blaustein. Axioms 5(3), 4–10 (2016). Ed. Urszula
Wybraniec-Skardowska
24. Smith, B.: Logic and the Sachverhalt. Monist 72(1), 52–69 (1989)
25. Wachter von, D.: Roman Ingarden’s ontology. Existential dependence, substances, ideas and other
things empirists do not like. In: Chrudzimski, A. (ed.) Existence, Culture and Persons. The Ontology
of Roman Ingarden, pp. 55–82. Ontos, Frankfurt a. Main (2005)
26. Woleński, J.: Filozoficzna Szkoła Lwowsko-Warszawska. Państwowe Wydawnictwo Naukowe,
Warszawa (1985). English edition: Logic and Philosophy in the Lvov-Warsaw School. Kluver,
Dordrecht (1989)
27. Zahavi, D.: Husserl’s Phenomenology. Stanford University Press, Stanford (2002)
U. M. Żegleń ()
Department of Cognitive Science and Epistemology, Institute of Philosophy, Nicolaus Copernicus
University, Toruń, Poland
Józef I.M. Bocheński
Korneliusz Policki
Abstract Bocheński’s philosophical biography consists in his original attempt to join

the method of the Lvov-Warsaw School with an open Thomism. Paradigm changes in
Bocheński’s philosophy were preceded by changes in science, politics, sociology as well
as ecclesiology.
Keywords History of logic · Philosophical methodology · Thomism · Sovietology ·

Logic of religion · Logic of authority · Logic of the business enterprise
Mathematics Subject Classification (2000) Primary 01A70
1 Life
J.I.M. Bocheński was born August 30, 1902 in Czuszów in the vicinity of Cracow. In 1920
he began studies in law and economics in Lwów (Lemberg, Lvov), completing them at the
University of Poznań. He entered the Dominican order in 1927, and a year later he began
to study philosophy at the University of Fribourg (CH) where he was awarded the doctoral
degree. Theological studies followed at the Angelicum in Rome and culminated as well
in the doctoral degree in 1935. In the following year he acceded to the ‘Habilitation’ in
philosophy at the University of Cracow and began teaching logic. Throughout the period
of the World War II he interrupted his scholarly pursuits in order to devote himself to the
struggle for freedom in his capacity as military chaplain.
Following the cessation of hostilities Bocheński is named to the chair of modern
and contemporary philosophy in the University of Fribourg, a position he held until his
retirement in 1972. In the period 1951–1952 he was dean of the Faculté des lettres and
in the years 1964–1966 rector of the University. In the latter capacity Bocheński was
active in expanding the university. Moreover, in 1958 Bocheński founded, in Fribourg,
the Institute of East European Studies, and 4 years later he became, thanks to Konrad
Adenauer’s support, cofounder and director of the Ost-Kolleg in Köln.
On November 14, 1992 Bocheński was granted the order ‘Polonia Restituta’ on the
recommendation of President Lech Wał˛esa and his scholarly achievements earned him
entry into the august society of the Polish Academy of Science. Bocheński’s long and

264 K. Policki
productive life came to end on February 8, 1995, in Fribourg. The urn containing his
ashes is found in the Dominican sepulcher of the Albertinum.
The following statistics, derived from Bocheński’s curriculum vitae, are revealing:
• he served in three military campaigns,
• he studied over a course of 13 years in four different university faculties,
• he earned two doctorates and the habilitation,
• he taught at least one semester in 12 universities in Europe, South America, North
America, and Africa,
• he was named to professorships in the Angelicum, Rome, and the University of
Fribourg,
• he was twice dean of the Faculté des lettres and later rector of the University of
Fribourg,
• he was awarded five honorary doctorates,
• he published 44 books and 191 scholarly articles,
• he completed a journey round the world as pilot in command.
2 Main Research Interests
The scope of Bocheński’s research and the range of his interests were very wide. His areas
of research can be divided into fundamental directions to which he attributed priority and
secondary interests that occupied him per accidens. The main areas are logic (the history
of logic, Hindu logic, applied logic, the logic of religion, the logic of authority, the logic
of the business enterprise) and philosophy (Thomism, analytic philosophy, sovietology).
Of the secondary interests mention should be made of political science, systems theory,
philosophical psychology, economics, Egyptology, as well as sapiential philosophy.
Bocheński was a polyglot. He lectured and wrote in all major European languages,
including Russian. He was at ease with classical languages: Greek, Latin, and Hebrew.
He knew Sanskrit as well as Chinese. Given his historical interests he was adept at
paleography.
2.1 The Main Writings
Bocheński’s most important publications

1932 Die Lehre vom Ding an sich bei M. von Straszewski
1933 Elementalogica Graecae
1938 Nove lezioni di logicasimbolica
1939 De virtuti militari
1947 Petri Hispani Summuas Logicales
1947 La logique de Théophraste
1947 Europäische Philosophie der Gegenwart
1949 Précis de logique mathématique
1951 Ancient Formal Logic
Józef I.M. Bocheński 265
1952 Der sowjetrussische dialektische Materialismus

1954 Die zeitgenōssischen Denkmethoden
1956 Formale Logik
1958 Handbuch des Weltkommunismus (together with G. Niemeyer)
1959 Wege zum philosophischen Denken
1965 The Logic of Religion
1974 Guide to Marxist Philosophy (together with T.J. Blakeley)
1974 Was ist Autorität?/Qu’est-ce que l’autorité ?
1985 Zur Philosophie der industriellen Unternehmung
1987 Sto zabobonów
1987 Über den Sinn des Lebens und über die Philosophie
1988 Mi˛edzy logika˛ a wiara˛ z J.M. Bocheńskim rozmawia J. Parys/«Entre la logique et
la foi»
1988 Autorität, Freiheit, Glaube
1992 Podr˛ecznik madrości
˛ tego świata
1993 Religia w Trylogii
1994 Sens życia
2003 Gottes Dasein und Wesen
3 Bocheński’s Views
3.1 Logic and the History of Logic
Bocheński was a passionate devotee of logic. Thanks to his style of writing and
thinking he is counted among the members of the Polish school of logic. He maintained
lively contacts with many Polish logicians, including S. Leśniewski, L. Chwistek, K.
Ajdukiewicz, T. Kotarbiński, A. Tarski, but in particular J. Łukasiewicz whom he counted
as a friend.
How did Bocheński conceive logic? How many parts of logic did he distinguish? The
following table renders Bocheński’s understanding of the parts of logic.
Logic (in general)
(a) in the strict sense
• pure logic (formal, mathematical)
• applied logic
– Methodology
– Semiotics
Syntactics
Semantics
Pragmatics
266 K. Policki
(b) in the broad sense

• history of logic
• philosophy of logic
Bocheński made substantive contributions to each of these areas of logic.
Logic can be understood from a subjective and objective perspective. From the
subjective perspective logic, according to Bocheński, “[. . . ] is the ability to move about, to
engage in thinking at a very high level of abstraction.” Understood objectively, logic is the
description of objects, the description of things, it investigates the general structure of the
world; it determines not only relations among sentences but first of all among things in
general. According to Bocheński, formal logic is but ontology carried out axiomatically.
After Aristotle’s death a controversy arose concerning the question whether logic is
mereos, that is, a part, or organon, that is, a tool for philosophy, or also pajdagogos,
that is, a means of education. As regards their material objects, logic and ontology do not
differ. As disciplines they are distinguished only by their methods, that is, their respective
formal objects.
Bocheński often cited an example from Albert the Great to illustrate the meaning in
which logic is a tool (organon): “It is certain that in his workshop the smith produces a
hammer. And when the hammer is ready it serves in the same workshop as a tool with
which to produce other tools.” The same is true of logic. Logic is an autonomous science,
but this special science is a necessary tool of the other sciences, in particular philosophy
and theology.
Logic qua pajdagogos is nothing other than the ethics of thinking and speaking. It has
major educational significance.
3.2 Philosophy
Bocheński ventured down the path of philosophy in the shadow of three professors
whose names all begin with the letter ‘Z’. All three exerted considerable influence on the
young student Bocheński, in particular in regard to his style and way of philosophizing.
His professor of mathematics at the gymnasium was Zygmunt Zawirski, an analytic
philosopher associated with the “Lvov-Warsaw School” whose founder was Kazimierz
Twardowski, a student of Franz Brentano. Zawirski would later be among the experts
who evaluated Bocheński’s habilitation presented to the Jagiellonian University in 1938.
The second professor whose influence played a formative role for Bocheński was
Florian Znaniecki, a student of Henri Bergson, later the president of the American
Association of Sociology. Znaniecki fascinated Bocheński; it was Znaniecki who at that
time awakened Bocheński’s enthusiasm for statistics and sociology.
Bocheński owed his passion for precision to the school of Czesław Znamierowski, the
creator of the analytic philosophy of law. Philosophy is not to be likened to the painterly
art where each artist creates a new work; quite to the contrary, the philosopher should
analyze already existing ‘paintings’ and in this way enrich science. Indeed, Bocheński
created no philosophical system, but he did enrich a number of philosophical domains
thanks to several important and innovative contributions.
Among the latter the following are the most important:

1. Metaphilosophy Bocheński pioneered the first broad concept of analytic philosophy.
He proposed as well the idea of the cyclical development of philosophy, the specific kind
of development to which it is subject. He set out a clear definition of the concept of
‘worldview’ and argued strongly for the claim that a philosopher should not attend to
issues having to do with worldviews.
2. Ontology Bocheński argued that formal logic is axiomatic ontology. He proposed a
new formulation of moderate Platonism with regard to universals according to which what
is at stake is basically a specific concept of identity. For example, “redness” refers to the
identity of red objects. Further, he systematically worked out the concept of the formal
system and its purpose. Early in his career, he carried out empirical research into the
concept of causality.
3. Metaphysics Bocheński has reservations with regard to over hasty constructions of
metaphysical systems. He published a program of investigation concerning God as well
as a critical study of Aquinas’ theodicy (I,2–11).
4. Epistemology Bocheński carried out a critical analysis of the argument in favor of
skepticism and concluded that skepticism is a philosophical superstition. He likewise
criticized contextualism. Further, he discussed certain aspects of medical methodology.
But above all he was the first to advance an analysis of analogy understood as isomor-
phism with the means of mathematical logic.
5. Philosophy of Nature He advanced a naturalist perspective, claiming that for science
there is no fundamental difference between man and other animals. He undertook as well
an analysis of the concept of the meaning of life.
6. Social Philosophy He was the first to propose an extensive logical analysis of the
concept of authority. He advanced a novel analysis of the commercial enterprise and
the free society. In addition, he provided a formal logical analysis of the concept of
responsibility.
7. Ethics Bocheński was most probably the first to write on the ethics of war. Further,
he insisted on the separation of morality, ethics, sapiential wisdom, and different kinds of
religious morality (fear of the Lord). He carried out a logical analysis of the controversy
concerning animal experimentation.
8. Philosophy of Religion Bocheński proposed a novel definition of the philosophy of
religion and he created a new scholarly domain, the logic of religion. He formulated
the theory of the religious hypothesis in metanoia/repentance. He advanced as well the
religious hypothesis distinguishing fundamental faith from religious systems as well as
from the history of saints and all souls.
9. The Philosophy of Spirituality Bocheński distinguished fundamental visions [of the
world] from worldviews. He carried out a still relevant analysis of the fundamental visions
of the Middle Ages and the Renaissance. Catholicism and Islam are the constituents of the
fundamental medieval vision. They are static and anthropocentric, inimical to the power
of progress. Anthropocentrism remained present within the Renaissance vision, however
the Renaissance believes in man’s unlimited potential, in progress, science, and rationality
268 K. Policki
generally. The criterion of these two visions should be “contemporary spiritual situation”
(a concept borrowed from Jaspers).
10. Sovietology It was common in the West to believe that, in order to understand thought
in the Soviet Union, it sufficed to read several key works of Karl Marx, a preconception
that Bocheński rejected as erroneous. He distinguished marxology, leninology, and
sovietology and argued that each of these areas of research is far too extensive for any
one person to be able to encompass all of them. To be sure, a sovietologist would have to
be well read in Marx and Lenin, however he need not and cannot be a specialist in these
areas. The converse holds equally: a specialist in Marx is not for that reason a specialist
in Leninism, and the latter specialist is not for all that a sovietologist. Each successive
sovietological sub-discipline incorporated the core of the earlier one completing it in its
specific manner all the way to Gorbachev’s perestrojka. Bocheński never ceased to be
irritated when he was identified as a marxologist, considering himself on the contrary to
be a sovietologist.
11. Anthropology We find in Bocheński reflections on philosophical and theological
anthropology, the latter particularly in his early career. His theological anthropology is
situated between theism and soteriology. In his later development, Bocheński gave up
hylomorphism in his conception of man in favor of systems theory and substituted meta-
anthropology for anthropology. He was an opponent of scientific humanism, allowing
however for religious and intuitive humanism. This went in pair with his naturalism and
minimalist anthropology. He rejected ontological as well as methodological humanism. In
the course of his development he gave up his early Thomist perspective on the concepts of
freedom and responsibility in favor of an analysis applying the theory of logical relations.
In like manner Bocheński set out a formalized proof of the existence of the human soul.
12. Axiology In his value theory Bocheński defended a moderate realism. The mode
of existence of values is like that of universals. The foundation of values is a two-
place logical relation between the subject and the object. In the set of theoretical values
the relation to the good is of a contemplative—passive—nature. These are values that
entail the ought of existence though not the ought of action. For example, aesthetic
values: beauty, ugliness, elegance, coarseness, delicateness, etc. On the other hand, in
the set of practical values the relation/attitude to the good is of an active nature. These
values comprise both the ought of existence and the ought of action. Their character is
appellative. The set of practical values includes two subsets: technical values and spiritual
values. For technical values the object of the action is external to the agent (facere).
The techniques of physical activity evince a conditional (hypothetical) character. As for
spiritual values, in keeping with his Thomism Bocheński saw them as primarily moral
values possessing a categorical, unconditional character. For these values the object of
action (agere) is internal to the agent. These are the so-called techniques of spiritual
activity. Thomist traditionally included religious values among the moral values. In his
later analytic period, however, Bocheński follows Max Scheler in distinguishing religious
values from moral values.
13. Therapeutic Philosophy Under the influence of Hindu philosophy with which he
had come in contact, Bocheński, while studying this civilization’s logic, took an interest
in subjective philosophy. It is well known that the basic question of Hindu philosophy
concerns freeing oneself from suffering. Even logic should be able to assist (16 categories
of Hindu logic help in freeing oneself from suffering). Bocheński took up the question
of suffering in several essays: “On suffering,” “The Sin of sadness,” “To the broken
man,” “The phenomenology of the abyss,” “With regard to evil people,” “The meaning
of life,” “A handbook of worldly wisdom.” Suffering is at times so great that even religion
cannot help the sufferer. Homo patiens is not entirely coextensive with animal rationale.
Bocheński postulated the creation of a distinct interdisciplinary science of suffering.
3.3 Theology
Bocheński was likewise a theologian, of a Thomist and equally an analytic stripe.

Wherever possible he applied logical means to theology and believed that the right logic
for theology is many-valued logic. In his Logic of Religion he constructed a semantics of
theological language. The logical definition of logic has the following form: “Theology
can be described as the domain of investigation in which, along with other axioms, at least
one sentence from the Credo is admitted and accepted exclusively by the faithful of the
corresponding religion.” Bocheński broadened this definition by means of a more liberal
conception of the Credo. For example, the “Our Father” is the “Credo” of the so-called
basic faith that is the common feature of all monotheistic religions.
Bocheński did not accept the fideist understanding of theology, credo qui absurdum,
but rather the thesis “credo quia ratio.”
Pope Jean Paul II drew on many of Bochenski’s ideas in his encyclical ‘Fides et ratio’.
With regard to theology Bocheński preferred the reductive model of theology.
Methodologically, theology is similar to physics. The role of protocol sentences is
played by those sentences of the faith which require clarification by means of theory and
theological hypotheses.
4 Reactions
4.1 Political Reaction
Of all the members of the Lvov-Warsaw School Bocheński exerted the greatest political
influence on the changes that occurred in Central Europe as well as on the rise of the
Perestrojka in the Soviet Union.
Bocheński educated 2/3 of the philosophical sovietologists working in the United
States.
Bocheński’s sovietological studies contributed to the rise of Solidarity, the fall of the
Berlin Wall and the overthrow of Communism.
He was a co-founder of the review ‘Kultura’ in Paris which exercised an enormous
influence on the Polish intelligentsia.
Despite repressive means applied by the communist authorities in Poland, Bocheński’s
works published in his second period were read and discussed in workers’ and academic
milieus.
270 K. Policki
Thanks to Dominican pastoral centers Bocheński’s works found their way into Catholic
circles and contributed to the renewal of the Catholic Church following Vatican Two.
4.2 Scholarly Reaction
So far five symposia have been devoted, in Poland, to Bochenski’s scholarly heritage and
two in Switzerland. Many directions in Bocheńki’s work continue to be discussed and
developed.
Bocheński’s Thomism is carried forward by Ignacy Dec, J. Krucina, K. Wolsza, and
Czeslaw Glombik.
With regard to Sovietology, Marxism-Leninism as well as Russian philosophy,
Bocheński’s work has influenced Edward Świderski in Switzerland and in Poland Jan
Parys.
Development of Bocheński’s logical ideas has been the work of logicians in Warsaw—
Edward Nieznański, K. Świ˛etorzecka, Marek Porwolik, R. Tomanek, and Anna Brożek.
In the areas of history of logic and metalogic mention should be made of J. Angelelli,
M. Brander, Hans Burckhard, Peter Rutz, and Dariusz Gabler.
The logic of religion has been pursued by Roger Pouivet, M. Kaemfert, E. Żabski, and
Jan Woleński.
Bocheński’s interest in phenomenology played a significant role in the work of Anna
Teresa Tymieniecka and the meeting of analytic philosophy and phenomenology was
pursued by Bocheński’s former assistant and later successor at the University of Fribourg,
Guido Küng.
Aspects of Bocheński’s ethics are deployed in studies by J. Weinsenbeck, Helmut
Fleischer, Czesław Por˛ebski, and Korneliusz Policki.
4.3 The Educative-Didactic Effect
Bocheński directed more than 50 doctoral dissertations and an astounding number of

master’s and licentiate theses.
He published four textbooks of logic in which he presents, in a most accessible manner,
recent research in logic. In the history of philosophy he published two textbooks that have
been translated into several languages. Bocheński enjoyed considerable celebrity thanks
to his popular introductions to philosophy, his guides to Marxism-Leninism as well as
Thomism. Many of these works were likewise translated throughout the world.
Bocheński was also an excellent preacher and retreat master. His talks and sermons
have been published following his death. Virtually up to his death he conducted a
philosophical seminar in Polish named by him “Sempol.”
5 Criticism
In Bocheński’s later works, in particular in his second phase, a certain change of

philosophical paradigm is noticeable, a certain ‘perestrojka’.
What is evident is a passage from a maximalist to a minimalist philosophy, from
metaphysics to ontology, from a synthetic to analytic philosophy, from humanism to
naturalism, from ethics to metaethics, from hylomorphism to systems theory.
Throughout his life Bocheński remained a realist, a defender of objectivity, an optimist
struggling with “philosophical superstitions.”
Unfortunately, Bocheński remained indifferent to developments in the philosophy of
culture, to questions of social ontology, and current studies in the philosophy of science
as well as in the methods of the natural and human sciences.
6 Summary
Stanisław Lem’s description best characterizes Bocheński’s life and activity:

He was a person out of the ordinary endowed with so many diverse talents. Born in 1902, a cavalry
man during the Polish-Bolshevik conflict, he spent the Italian campaign during World War II at the
side of Bishop Gawlina. He travelled all over the world, encircled the earth with all the kilometers
of his travels, he was personally acquainted with the leading logicians and philosophers — all the
while dressed in a monkish dress.
K. Policki ()
University of Fribourg, Fribourg, Switzerland
J.M. Bocheński’s Theory of Signs
Korneliusz Policki
Abstract The article concerns Bocheński’s thus far unpublished considerations per-
taining to signs that are among the materials in my archive. Of particular interest are
Bocheński’s reflections on the ontological, psychological, and epistemological founda-
tions of the theory of signs. In his semiotics the concept of the sign is broadened to include
application to the logic of authority, the concept of bi-directionality of human phenomena
as well the theory of ‘philosophical superstitions’.
Keywords Signs: material, fundamental, formal · The emitter and receiver of signs
Mathematics Subject Classification (2000) Primary 03B80
1 Introduction
In the domain of semiotics Bocheński succeeded in rendering a number of basic concepts

more precise. My task in this paper is to present some of the results from within this
domain of research found among thus far unpublished materials of Bocheński’s archive. In
constructing his sign theory Bocheński relies on Thomism as well as the work of ‘Lwów-
Warsaw School’ (the concept of the functor).
2 Ontological Assumptions
The world consists of things (substantiae). We name as ‘thing’ any object that exists
self-sufficiently, is not determined by some other object, and undergoes change. Things
include, e.g. people, planets, tables and chairs, etc. These things bear certain properties.
In contrast with a thing, a property does not exist self-sufficiently (it is an accident), it
appears on the basis of and is determined by the thing. Certain properties differ ontically
(phenomenologically) from the thing in relation to which they emerge. Independently
of the cognizing subject there exists a difference between a given property and the thing

274 K. Policki
whose property it is. These properties include, for example, a man’s anger, his knowledge,
the shape of a piece of turf, an electron’s spin. Other properties are but certain aspects
of the thing, ontically indistinguishable from the latter, even though the human intellect
separates them fictively (abstractly) from their basis. An example would be, for example,
a man’s humanity: humanity is abstractable in a man as his property. In the thing itself
humanity is a constitutive component of the man. It is a characteristic property and the
sign of being human.
A property of a specific kind is existence. We name as “existence” that which accounts
for the existence of a given object. Existence is a very specific property in that it does
not change the makeup of the object in any way, but only transfers it from the sphere of
possibility to that of reality. What remains in an object when we exclude its existence
we name its ‘makeup’ (content, or essentia in the ontological sense of the term). For
that reason let us speak of the content of the sign as well as of its existence. A sign
is characterized by its phenomenological, ontic, metaphysical, subjective and objective
foundations [3, p. 55].
However, existence is multifarious. First of all, there is real existence, designated
empirically. This is how, for example, living people, galaxies, street signs, and railway
semaphores, etc. exist. Secondly, there is the mode of existence of merely possible objects;
were possible objects to be deprived of a mode of existence it would be difficult even to
think about them. However, it is a fact that we do think about such objects, for instance
when we create new signs. Thirdly, another mode of existence characterizes objects that
cannot exist empirically but which we can, so to speak, “allow ourselves” in creating
fictional entities, e.g. the square circle, the Sphinx, the wooden stone, etc. The upshot is
that we need to distinguish at least four modes of existence: actual, potential, possible,
and purely putative existence.
Ontological assumptions are essential for understanding the objectively real and ideal
content of a sign. A sign refers to an objective content that can be material, real, and ideal.
A theory of signs will vary as the ontology varies. Thus Tadeusz Kotarbiński’s theory of
signs differs from that of Bocheński.
Whatever has a mode of existence (actual, potential, putative, possible, real) we tradi-
tionally call a ‘being’. Mt Blanc is a being as is the as yet non-existent but possible railway
line and the square circle; but whereas Mt Blanc is an actual being, the future railway line
is a potential being (though both Mt Blanc and the railway line are real beings) and the
square circle is an purely intentional being (ens rationis).
3 Psychological Assumptions
According to Bocheński the course of the cognitional process begins with a simple
perception (simplex apprehensio) that is followed secondly by the judgment. The
perception is a certain sort of intellectual assimilation of the cognized object; it differs
from a material assimilation in that there arises in the psyche of the knower a mental
picture of the object of knowledge which is a subjective concept (conceptus subiectivus).
But the simple perception never grasps the object as a whole; it is always the grasp of only
one aspect of the object. What the simple perception grasps is a property or properties of
the object. These are called the ‘ideal’ concept (conceptus obiectivus). Thus in the course
J.M. Bocheński’s Theory of Signs 275
of the cognitional process we attain the extra-mental object thanks to the operation of two
intermediate factors: first by means of the “subjective concept” and then by means of what
is immediately conceived in the “ideal concept”.
The subjective concept as well as the objective and ideal concepts play important roles
in our practices. For example, a road sign that has been willfully displaced or knocked
down by vandals retains its normative force as an ideal concept acquired during the driver-
training course despite the absence of an actual perception and a subjective concept.
The ideal concept is always an abstraction from actual existence; quite simply, it is
eternal. We can construct subjective and ideal concepts, that is to say, come to know
objects that do not exist in a given moment, and that even cannot exist in the external
world. Even when the object of knowledge is existence itself we construct the concept of
existence in thought.
The construction of judgments proceeds analogously. In a judgment the counterpart of
the subjective concept is the subjective, intended sentence. This sentence should not be
confused with a spoken or written sentence. The counterpart of the objective concept is
the objective sentence, the so-called “ideal sentence” (enuntiabile). It is by way of these
sentences that the judgment attains its object, the fact. The ideal sentence has the same
characteristics as the ideal concept: it is objective but is abstracted from real existence [1,
pp. 1–4].
Tadeusz Kotarbiński did not agree with Bocheński’s theory: according to Kotarbinski,
concepts and ideal sentences are hypostases. He held that only nouns exist with designata
in the world of persons and material things [7, p. 15]. These nouns are material signs. For
Bocheński, on the contrary, there exist (non-material) psychological constructs such as
concepts and subjective sentences. They are, in other words, so-called mental signs.
4 The Concept of the Sign
For Bocheński, a sign has three aspects: a material, fundamental, and formal aspect. The
material sign is the thing that serves as the means to know something. The fundamental
sign is some property (typically a certain shape) or the motion of the thing that is directly
significant. For example, in the cavalry whereas the raising of the sabre three times by
the commander means “gallop”, the sabre itself is a material sign, and its motion and rest
is the fundamental sign. Finally, the formal sign is the relation that holds between the
fundamental sign and what the sign expresses, means, and denotes. This relation is called
the ‘formal sign’ for the reason that the form determining that the material sign becomes
a sign in the first place is precisely the relation: pragmatic, semantic and connotative.
We distinguish between natural and conventional signs. The smoke from Bocheński’s
pipe is the natural sign that Bocheński is smoking a pipe. The aforementioned raising
of the sabre is a conventional sign in the cavalry. Whereas smoke leads anyone familiar
with the use of tobacco to conclude, without recourse to any convention, to the fact of
Bocheński’s pipe-smoking, the raising of the sabre by Bocheński’s cavalryman would
lead no one to the thought that major Bocheński wishes to transit to the gallop were there
no convention to this effect among the troopers. For Bocheński conventional signs are
especially interesting.
276 K. Policki
A conventional sign in the fundamental sense is a characteristic that, thanks to an

agreement between at least two persons, allows one of them (the recipient) to recognize
something distinct from the characteristic as a natural sign, and this on the basis of an
agreement between the recipient and another person who is the emitter of the sign.
5 The Theory of Signs and Authority
Bocheński created the logic of authority [5]. He distinguished two kinds of authority.
One kind is epistemic authority (based on knowledge), the other is deontic authority
(based on power). Bocheński divides deontic authority into the authority of solidarity and
punishment (sanction). All of these kinds of authority are applicable to the theory of signs,
in particular to the theory of conventional signs. Of the many emitters of conventional
signs authorities are an important case. Authorities ascribe the intentional as well as the
associative meaning to various objects that thereby become signs. However, this alone
does not suffice. In order for conventional signs to become effective deontic authority
is required; in order to institute the sign, there needs to be a power. Deontic authority
motivates, sanctions, and regulates the signs defined by epistemic authority. This deontic
power acts functionally to ensure solidarity among those who respect road signs and to
apply sanctions (mandates).
How are signs received by the user? Bocheński writes about the bi-directionality of
cognitive phenomena [8, pp. 141–145] that is present at all levels of man’s cognitive
activities. It is called the law of action and reaction. In the epistemology of signs bi-
directionality is especially evident. On one hand, we have the centripetal direction;
the sign is the stimulus, the call, it ‘imposes’ itself, as it were ‘calls to’, ‘enters into’
the knower. On the other hand, there is the centrifugal reaction of the receiver of the
sign ranging over not only imagination, judgment, and emotions, but likewise external
behavior and reactions, that is, centrifugal endeavors.
6 The Theory of Signs and ‘Philosophical Superstitions’
Bocheński created as well the theory of ‘philosophical superstitions’. A philosophical

superstition is a minor false world view, though regarded by some as true. An absolute
philosophical superstition is a view that is at odds with logic, methodology, science as
well as common sense. A relative philosophical superstition is a view that is relative to
some belief system, e.g. for Christians the pagan Roman religion is a superstition and
conversely; for Tadeusz Kotarbiński Bocheński’s theory of signs is a superstition since
it resorts to hypostases, mental signs, whereas for Bocheński Kotarbinski’s reism and
concretism are “philosophical errors”.
Bocheński’s “One hundred superstitions” has so far appeared only in Polish [6],
though a German translation is in preparation. Bocheński’s philosophical and logical
theories can be applied likewise to the theory of signs. An example of superstitions
related to the theory of signs is astrology (the theory of zodiac signs), numerology (the
J.M. Bocheński’s Theory of Signs 277
theory of numerical signs exerting effects on human life), and semantic anthropocentrism
(only humans employ signs), the confusion of sign-constitutive functors and arguments,
extreme positivism, nominalism in the theory of signs.
7 Conclusion
Bocheński’s studies in the theory of signs testify to the growth and development of the
philosophy of signs and semiotics. Progress in this domain of philosophy consists less
in attaining non-ambiguous and unquestionable results and more in discovering new
conditions and associations. For example, Bocheński’s theory of signs can be broadened
to include not only the logic of authority but also the logic of religion (“sacred signs”) [2],
something that Bocheński did in fact attempt in his unpublished “Was ich glaube?” in my
archive.
References
1. Bocheński, J.M.: A. Church, N. Goodmann (eds.) The Problem of Universals. Notre Dame Press, Notre
Dame (1956)
2. Bocheński, J.M.: The Logic of Religion. New York University Press, New York (1965)
3. Bocheński, J.M.: Die zeitgenösssichen Denkmethoden. UTB Francke, Bern (1973)
4. Bocheński, J.M. Grundriss der Logistik. Ferdinand Schönigh, Padeborn (1973)
5. Bocheński, J.M.: Was ist Autorität?, Einführung In die Logik der Autorität. Herder, Freiburg (1974)
6. Bocheński, J.M.: Sto zabobonów [One Hundred Superstitions]. Instytut Literacki, Paris (1987)
7. Kotarbiński, T.: Kurs logiki dla prawników [Logic Course for Jurists]. PWN, Warszawa (1961)
8. Policki, K: Filozofia człowieka we wczesnej twórczości J.I.M. Bocheńskiego [Philosophical Anthro-
pology in the Early Works of J.I.M. Bocheński]. PWT, Wrocław (2005)
K. Policki ()
University of Fribourg, Fribourg, Switzerland
Jan Salamucha (1903–1944)
Abstract Father Jan Salamucha (1903–1944) was one of the most prominent Polish
Catholic intellectuals of the first half of the twentieth century. He was a victim of the
World War II, killed in the Warsaw Uprising. We present his intensive academic research
and mention his didactic, pastoral, and pro-independence activities.
Keywords Jan Salamucha · Cracow circle · Analytical Christian philosophy
Mathematics Subject Classification (2000) Primary 01A60; Secondary 03-03
1 Beginnings
Jan Salamucha was born on 10 June 1903 in Warsaw to the family of Stanisława (née
Marciniak) and Andrzej Salamucha. He was the first of two children—he had a younger
sister, Genowefa (born in 1909). The Salamucha family had a working-class background
(the father was a worker at a foundry in Warsaw). From (approx.) 1914 to 1919, Jan
Salamucha studied at the Stefan Chrupczałowski humanistic junior high school, and later
attended the Metropolitan Seminary. During his studies at the Seminary, he took part in
the Polish-Soviet war of 1920 (August–October) as a volunteer paramedic.
2 Education
Upon graduating from the Seminary—in 1924—he commenced studies at the Faculty
of Theology at the University of Warsaw, the Philosophy Section, which he completed
in 1926, being awarded the master’s degree on the basis of his dissertation entitled
O kategorii πρoξ u Arystotelesa (About the πρoξ category according to Aristotle). A year
later, he defended his doctoral dissertation in the field of Christian philosophy, entitled
Teoria wynikania modalnego u Arystotelesa. Studium krytyczne (Aristotle’s theory of
modal inference. A critical study) Both dissertations were supervised by Rev. Stanisław

280 K. Świ˛etorzecka
Kobyłecki. From 1927 to 1929, he continued studies at the Gregorian University in Rome,
which resulted in his master’s degree dissertation entitled De deductione aqud Aristotelem
et S. Thomam. A Polish version of this dissertation, Poj˛ecie dedukcji u Arystotelesa i
św. Tomasza z Akwinu (The idea of deduction according to Aristotle and Saint Thomas
Aquinas) was the basis for Salamucha’s habilitation at the Jagiellonian University in
1933. However, the Ministry of Religion and Public Education approved the habilitation
as late as 1936. The first attempt at having the habilitation approved by the Ministry of
Religion and Public Education in 1933 ended unsuccessfully in unclear circumstances
of personal intrigue (associated with father Andrzej Krzesiński, who was conflicted with
Salamucha’s supervisor, Rev. Konstanty Michalski) and anonymous denunciation letters
regarding Salamucha (about his political views and close contact with people suspected
of involvement with freemasonry—Jan Łukasiewicz and Wacław Sierpiński). Although
the rejection of his habilitation seriously contributed to Salamucha’s inability to conduct
academic and educational work, luckily it did not stop him from further academic activity.
Salamucha was a student of numerous prominent figures of the Catholic Church
and the Polish academic circles in the pre-war period. At the Seminary, among his
teachers were: Rev. Mieczysław W˛eglewicz and Rev. Stanisław Mystkowski. In the
period of studies at the University of Warsaw, he was also taught by: Tadeusz Kotar-
biński, Władysław Tatarkiewicz, Wiktor Wasik ˛ (philosophy); Stanisław Leśniewski and
(probably) Jan Łukasiewicz (logic); Stefan Mazurkiewicz, Stefan Staszewicz, (probably)
Wacław Sierpiński (mathematics); Rev. Stanisław Kobyłecki (psychology); Wincenty
Kwiatkowski (theology). At the Gregorian University, Salamucha was taught, among
others, by Pierre Hoenen. Salamucha’s post-doctoral dissertation was written under the
auspices of Rev. Konstanty Michalski. Under the influence of such excellent teachers,
during the Third Polish Philosophical Congress in Cracow in 1936, father Salamucha,
together with Rev. Innocenty Maria Bocheński and Jan Franciszek Drewnowski, initiated
a discussion which resulted in the creation of the Catholic variant of the Lvov-Warsaw
School—the so-called Cracow Circle.
The academic and personal life of father Salamucha was also strongly influenced by
his patron of many years, the Metropolitan of Cracow, Archbishop Adam Sapieha.
3 Didactics
Salamucha started his didactic work in 1929 at the Warsaw Seminary, where he
taught philosophy until 1933 (e.g. in formal logic, methodology of sciences, theory of
cognition, and psychology). Due to the commencement of post-doctoral degree studies
at the Jagiellonian University, he vacated his position at the Seminary. Because of the
problems related to the approval of habilitation, the second Department of Philosophy
(planned previously by the Ministry of Religion and Public Education) at the Jagiellonian
University, to be headed by Salamucha, was not opened, and he was only offered the
position of freelance lecturer. In such circumstances, father Salamucha declined that
offer and gave no lectures for almost a year (his post at the Warsaw Seminary was
already taken). In the academic year 1934/1935, he was able to resume his lectures at the
Seminary, and decided to give freelance lectures at the Jagiellonian University. In 1938,
the authorities of the Jagiellonian University proposed that father Salamucha be appointed
Jan Salamucha (1903–1944) 281
Associate Professor of Christian Philosophy, a title which he was finally awarded in the
same year by the President of the Republic of Poland, Ignacy Mościcki.
In the entire period of his pre-war didactic activity at the Jagiellonian University,
Salamucha taught formal logic, the history of logic, the theory of cognition, theodicy,
cosmology, and the history of ancient philosophy. It should also be noted that in that
period, one of his students was the future Pope—John Paul II.
Immediately after the outbreak of war, Salamucha came to Warsaw, where he became
the chaplain for the forces defending one of the districts of Warsaw (Bem’s Forts). In
recognition of his valour and courage, the authorities of the occupied Poland awarded
him the Cross of Valour. In the same year, he travelled back to Cracow to continue
his didactic activities. On 6 November 1939, together with a group of 182 lecturers
of the Jagiellonian University, he was arrested and transported to the Sachsenhausen
concentration camp, and later to Dachau (in the so-called Sonderaktion Krakau). Thanks
to the intervention of Heinrich Sholz, the German authorities agreed to Salamucha’s
release and return to Cracow (1941). After a short period of convalescence, Salamucha
went to Warsaw. There, he continued his academic and activities. He participated in
clandestine seminars organised by Tatarkiewicz, Łukasiewicz and Adam Krokiewicz and
his close friend Bolesław Sobociński; he also taught philosophy (metaphysics, logic) at
the secret Warsaw Seminary and at the University of the Western Lands (operating until
1945). At that time, Salamucha taught e.g. Henryk Hiż and Andrzej Grzegorczyk; he also
reviewed the post-doctoral dissertation of Stefan Świeżawski.
4 Priesthood
Because of his spiritual formation, in addition to the didactic and academic activities,
Salamucha was simultaneously active in pastoral work. Immediately after the completion
of studies in Rome (in 1929), he took the post of vicar at the St. Wojciech Church in
Wiazowna
˛ in the Otwock region (near Warsaw), were he stayed for almost a year. During
the break in the academic work, resulting from his troubles with the habilitation (1933–
1934), he served as a vicar at the Parish of Our Lady of Loreto in Warsaw’s Praga district,
at the St. Florian’s Church. As already mentioned, he was the chaplain at Bem’s Forts,
as well as one of those priests who attracted numerous members of the intelligentsia in
Warsaw under occupation. The last place of his ministry was the parish of St. James in
Warsaw’s Ochota district.
5 Death
While serving as the vicar at the Church of St. James, Salamucha decided to become
involved in the Warsaw Uprising—he was assigned to the District of Warsaw Home Army,
and as a military chaplain for the units defending the Wawelska Redoubt in Warsaw,
participated in the fighting and helped with the evacuation of insurgents and residents of
the Staszic Colony (housed just near Wawelska Redoubt). He served as a priest until the
end—staying with the wounded and those who did not manage to escape before the end
of the insurgents’ struggle—and supporting them spiritually. On 11 August 1944, during
mass murder of the residents of the old Ochota district in Warsaw, he was killed by the SS
RONA brigade—a Ukrainian unit of the Russkaya Osvoboditelnaya Narodnaya Armiya.
6 Intellectual Formation
Salamucha’s intellectual output was undoubtedly conditioned by certain characteristics

of his personality. He was exceptionally talented and hard-working, with a scientific
mind and great sensitivity; courageous, steadfast, and able to make sacrifices for the
greater good. Therefore, his intellectual interests were broad—he was drawn to the
issues of classical metaphysics and theodicy, the history of philosophy (in particular:
the history of logic), formal logic, as well as matters of ethical and social nature.
On the other hand, Salamucha’s intellectual formation was influenced by the interests
and views of some of his teachers. Father Salamucha was influenced intellectually
by such prominent members of the Lvov-Warsaw school as: Łukasiewicz, Leśniewski,
Tatarkiewicz and Kotarbiński. As a Catholic philosopher, Salamucha implemented the
basic postulates of this school on the basis of Christian philosophy. (He also inspired
other members of the Cracow Circle: Bocheński, Drewnowski and their consultant on
logic: Sobociński.) Thus, in line with those postulates, he researched the issues of classical
philosophy and its history, but using precise tools of contemporary formal logic. The
strict method used by Salamucha (following the lead of Łukasiewicz and Leśniewski),
was to bring the philosophical deliberations closer to science, and to help capture the
rationalism, realism, impartiality, and maximalism of the Catholic philosophy understood
as a modern continuation of scholastic philosophy. The most impressive realisation of the
methodological and philosophical programme by Salamucha is the logical reconstruction
of the ex motu argument by St. Thomas Aquinas proving the existence of God [2].
In the history of human thought, Salamucha’s formalisation was the first attempt at
substantiating the basic thesis of theodicy on the existence of the Absolute, made with
the use of contemporary tools of formal logic and the set theory. It was also the starting
point for broader research based on formalised theodicy, conducted e.g. by: Bocheński,
Johannes Bendiek, Francesca Rivetti Barbo, Edward Nieznański, Korneliusz Policki.
During the occupation, Salamucha became more interested in ethical and social
matters. Consequently, he saw those issues from a philosopher’s and analyst’s point of
view, paying great attention to precision, accuracy, and clarity of his expressions. He
analysed concepts such as: evil, suffering, love, coercion; he also attempted to specify the
notion of Catholic ethics.
Father Jan Salamucha was one of the most prominent Polish Catholic intellectuals
of the first half of the twentieth century. He had little time to pursue the goals he set
for himself—he was a victim of one of the most dramatic periods in the contemporary
history of Poland—World War II. His intensive academic, didactic, pastoral, and pro-
independence activities allow us to recreate the image of an exceptional man—a talented
and creative philosopher, a wise teacher and a spiritual guide, a patriot and a hero. This
picture would be incomplete, if we failed to mention that Salamucha was a faithful and
honest friend, and a lover of nature (an enthusiastic mountaineer and rower) and of arts.
Jan Salamucha (1903–1944) 283
Selected Publications
Salamucha’s academic achievements are presented on a full list of publications in: Jan
Salamucha. Wiedza i wiara. Wybrane pisma filozoficzne, ed.: Jacek J. Jadacki, Kordula
Świ˛etorzecka. TN KUL, Lublin (1997); English edition: Jan Salamucha, Knowledge
and Faith. Collected Works, ed.: Jacek J. Jadacki, Kordula Świ˛etorzecka, A. Mickiewicz
University Press, Poznań (2003)
We list selected papers:
1. Poj˛ecie dedukcji u Arystotelesa i św. Tomasza z Akwinu. Studium historyczno-krytyczne [The idea of
deduction according to Aristotle and Saint Thomas Aquinas. A Historical and Critical Study]. Polish
Theological Society (1930)
2. Dowód ex motu na istnienie Boga. Analiza logiczna argumentacji św. Tomasza z Akwinu. Collectanea
Theologica, ch. 15, vol. 54, book 1–2, 53–92 (1934) (English translation by Tadeusz Gierymski,
Marian Heitzman: The Proof ex motu for the Existence of God. Logical Analysis of St. Thomas’
Arguments. The New Scholasicism 32, No 3, 327–372 (1958). French summary: Bolesław Sobo-
ciński, Le preuve ex motu de l’existence de Dieu. Analyse logique de l’argumentation de St. Thomas
d’Aquin. Studia Philosophica, vol. 1, 483–484 (1935))
3. Pojawienie si˛e zagadnień antynomialnych na gruncie logiki średniowiecznej [The appearance of
antinomial questions in medieval logic]. Przeglad ˛ Filozoficzny, ch. 40, book 1, 68–89; book 3, 320–
343 (1937)
4. Cardinale Thomaso de Vio Gaetano nel quatro centenario della sua morte [rev.]. Przeglad ˛ Filo-
zoficzny, ch. 40, book 4, 474–476 (1937)
5. Zestawienie scholastycznych narz˛edzi logicznych z narz˛edziami logistycznymi [A comparison of
scholastic logical tools with logistic tools]. In: Myśl katolicka wobec logiki współczesnej, pp. 35–
48 (1937) (French summary of professional accomplishments: ibid., 167–169)
6. O «mechanizacji» myślenia [On the «Mechanisation» of Thinking]. In: Myśl katolicka wobec
logiki współczesnej, pp. 112–121 (1937) (French summary of professional accomplishments: De la
mecanisation de la pensee. ibid., 182–186)
7. O możliwościach ścisłego formalizowania dziedziny poj˛eć analogicznych [Possibilities of a Precise
Formalisation of the Domain of Analogical Notions]. In: Myśl katolicka wobec logiki współczesnej,
pp. 122–153 (1937) (French summary of professional accomplishments: A propos de la formalisation
des domaines des concepts analogiques. ibid., 186–193)
8. Logika stara i nowa w wiekach średnich [Old and New Logic in the Middle Ages] (summary of
professional accomplishments). Ruch Filozoficzny 15(3), 156b (1938)
9. Z dziejów nominalizmu średniowiecznego (From the History of Medieval Nominalism]. In:
Czeżowski, T. (ed.) Christeria. Rozprawy filozoficzne złożone w darze Władysławowi Tatarkiewic-
zowi, pp. 207–238. PWN, Warsaw (1960) (Work completed in 1938)
10. Krytyka poznania [The Critic of Cognition]. Filozofia Nauki, ch. 3, No 1–2, 193–225 (1995) (Work
from 1935)
K. Świ˛etorzecka ()
Institute of Philosophy, Cardinal Stefan Wyszyński University in Warsaw, Warsaw, Poland
Struve and Biegański: Towards Modern
Approach to Logic
Roman Murawski
Abstract The paper is devoted to two Polish philosophers and logicians Henryk Struve
and Władysław Biegański. They represented the traditional pre-mathematical approach to
logic and stood on the threshold of the new paradigm of logic as formal mathematical
logic. Their views concerning logic and its philosophy as well as its relations to
philosophy and mathematics will be analyzed. Those views will be compared with views
of their contemporaries and with views of Jan Łukasiewicz—one of the main leaders of
Warsaw School of Logic who represented the new paradigm.
Keywords Traditional logic · Modern logic · Paradigm in logic · Struve ·

Biegański · Psychologism
Mathematics Subject Classification (2000) Primary 03-03; Secondary 01A55, 01A60
1 Henryk Struve
1.1 Life
Henryk Struve (alias Florian Gasiorowski)

˛ was born in Gasiorów
˛ near Koło on 27 June
1840. He studied theology and then philosophy at the University of Tübingen. In 1862,
he obtained his doctorate at the University of Jena. In the years 1863–1903, he was a
professor of the Main School [Szkoła Główna] in Warsaw, then the (Russian) Imperial
University of Warsaw. In order to maintain this post he had to defend his doctorate again
in Moscow (1870). In 1903, he decided to stay with his daughter in England. He died in
Eltham (Kent) on 16 May 1912. He dealt with philosophy and the history of philosophy.
He called his philosophical system ‘ideal realism’—it was an attempt at reconciling
realism with idealism. He popularized the history of Polish philosophy in Germany and
England. He wrote also many logic textbooks as well as a history of logic.

286 R. Murawski
1.2 Struve’s Conception of Logic
Henryk Struve is regarded as one of the most important figures of Polish logic in the
nineteenth century (cf. for example Woleński [37, p. 30]). In the interwar period he was
frequently referred to but his works were not analyzed and reprinted. We are interested
here in Struve’s views on logic as a science and his conception of logic.
Struve’s views on logic were presented in his fundamental work Historya logiki jako
teoryi poznania w Polsce [History of Logic as the Theory of Knowledge in Poland] [32]
from 1911 and in the textbook Logika elementarna [Elementary Logic] [31] from 1907
as well as in various papers (cf. for example [29]). There arise certain difficulties in
reconstructing his views—they result from the fact that he wanted to create a coherent
system of philosophy that would embrace all the traditional branches of philosophy.
Consequently the borders between particular branches were flexible and imprecise. The
principles of one division influence the foundations of the other and conversely. He
balanced between materialism and idealism aiming at the golden mean. This was reflected
also in his understanding of the object of logic.
At first he thought that the object of logic was principles and rules of thinking. In his
talk given in 1863, inaugurating his lectures at the Main School, he said:
Gentlemen! Logic is most generally the science of rational thinking, having thinking, its principles
and rules as its object.1 [28, Lecture 1]
However, he added that thinking is one of the powers of the soul; it is “an objective,
neutral consideration of this world by the soul”2 [28, p. 55]. Thus he introduces a
psychological element, and indirectly—an ontological one. In fact he claims that the soul
is the ideal embryo of the human being and “the limits of our being are the limits of our
correct thinking”3 [28, p. 35].
According to Struve logic concerns objective reality. However, it does not concern it
directly. The mediator between logic and the world is the thought. Nevertheless, this does
not lead to the thesis that thought reflects the logical structure of the world or to the thesis
that the world has some logical structure at all.
Struve’s earlier views were even more inclined towards psychologism. Initially, he
claimed—as indicated above—that thinking is “an objective, neutral consideration of
this world by the soul.” He upholds this thesis in Logika elementarna [31], but here he
separates logic from psychology, writing that logic deals with thinking as “an auxiliary
mean to get to know the truth” whereas psychology is interested in emotional and
volitional motives of cognition. Logic has both a descriptive and normative character and
is to oversee the application of the established norms and thus to evaluate the degree of
the truth of cognition.
The foundation of logic is philosophy, but also conversely: philosophy can be
developed only on the foundation of logical laws. The title of the main analyzed work of
Struve, Historya logiki jako teoryi poznania w Polsce [History of Logic as the Theory of
1 ‘Panowie! Logika jest to w najogólniejszem poj˛ eciu nauka myślenia majaca

˛ myślenie, jego zasady i
prawidła za przedmiot.’
2 ‘obiektywne, neutralne rozpatrywanie tego świata przez dusz˛
e.’
3 ‘granice naszego bytu sa˛ granicami naszego myślenia prawidłowego.’
Struve and Biegański: Towards Modern Approach to Logic 287
Knowledge in Poland], may suggest that he identified logic with the theory of knowledge.
In the first editions of Logika elementarna in Russian4 he made no clear distinction
between these two disciplines, but in the Polish version of the textbook [31] he wrote:
While it is true that thinking is the main co-factor of cognitive activity but not the only one; it
unites directly and constantly with the suitable expressions of emotion and will.5 [31, p. 3]
The examination of emotions and will as well as their relationships with thinking
belongs to the sphere of psychology whereas logic deals with thinking merely in one
aspect, namely:
As an auxiliary mean of getting to know the truth [. . . ]. Simultaneously, logic is not satisfied with
the real course of mental activity but seeks principles, i.e. laws and rules which one should follow
as norms if one wants to get to know the truth as exactly as possible. This separate view on thinking
gives logic the character of an independent science, which is strictly different from psychology,
namely this part of logic that investigates thinking as well.6 [31, p. 5]
In Historya logiki one finds the following words:

[. . . ] many separate the theory and criticism of cognition from logic as science dealing only with
thinking. Despite that, the connection between the development of correct thinking as well as
arriving at and getting to know the truth is so close from the psychological perspective that these
mental activities cannot be separated.7 [32, p. 1]
One can see here some traces of his discussions conducted with Kazimierz Twardowski
and the Lvov-Warsaw School. On the one hand, one can notice a certain readiness to
recognize the new understanding of logic and on the other hand, a desire to abide by his
current understanding of logic.
Struve differentiates between objective and subjective truth. The former is an ideal that
is independent of the human cognition and the latter is the reconstruction of the content of
being, of what exists in reality, in the mind—done through correct thinking. Logic controls
this reconstruction and thus through formal means it reaches the real being. Thus thinking
has a reconstructive and not a creative character. In Struve’s opinion three forms of logic
should be distinguished: (1) formal, (2) metaphysical and (3) logic treated as the theory
of knowledge. Formal logic considers the principles and laws of thinking regardless of its
object. Metaphysical logic (developed by Plato, Neo-Platonists, Spinoza and the German
idealists: Fichte, Schelling and Hegel) states that since thinking contains its object directly
in itself we get to know the very objective reality knowing the principles and laws of
4 Elemientarnaja logika was first published in 1874; there were altogether 14 Russian editions. It was the
obligatory manual of logic in classical junior high school from the year 1874. Its Polish version appeared
in 1907.
5 ‘Myślenie jest wprawdzie głównym, ale nie jedynym współczynnikiem czynności poznawczej; jednoczy
si˛e ono zarazem bezpośrednio i stale z odpowiednimi objawami uczucia i woli.’

6 ‘Jako środek pomocniczy poznania prawdy [. . . ]. Przytem nie zadowala si˛ e logika danym faktycznym
przebiegiem czynności myślowej, lecz odszukuje zasady, t.j. prawa i prawidła, któremi w myśleniu jako
normami kierować si˛e należy, chcac ˛ dojść do możliwie ścisłego poznania prawdy. Ten odr˛ebny poglad
˛ na
myślenie nadaje logice charakter samodzielnej nauki, ściśle różniacej ˛ si˛e od psychologii, a mianowicie
tej cz˛eści jej, która bada również myślenie.’
7 ‘[. . . ] wielu odróżnia zarówno teori˛e, jak i krytyk˛e poznania od logiki jako nauki samego tylko myślenia.
Pomimo to łaczność
˛ pomi˛edzy rozwojem prawidłowego myślenia a dochodzeniem i poznaniem prawdy
tak jest ścisła ze stanowiska psychologicznego, że tych czynności umysłowych rozerwać nie podobna.’
288 R. Murawski
thinking. Struve accepts neither the first nor the second conception of logic. He opts for
the third solution, treating it as the golden mean. Thus he understands logic as the method
of investigation and cognition of truth. Its task is to discover the principles according to
which man reconstructs the structure of the real world in his mind. Naturally, Struve sees
the difficulties connected with this view. In Logika elementarna he wrote:
The difficulties of examining the relation [. . . ] between thinking and the objective world are
obvious and can be reduced mainly to the fact that we are not able to compare directly our
images and concepts of objects and our views on them with the objects themselves. The question
concerning the objective knowledge of truth could be solved in favour of thinking only when it
turned out that the laws of our mind, and thus thinking, were fundamentally consistent with the
laws of the objective being which is independent of us. [. . . ] Nonetheless, showing the accordance
between the laws of the mind and the laws of the objective being requires a series of critical
investigations concerning the results of scientific studies.8 [31, pp. 6–7]
Struve calls this logic “logic of ideal realism”, uses to constitute the framework of a
coherent system of philosophy to give a general outlook on the world. We should add that
Struve is far from ascribing Messianic tendencies to logic (as Hoene-Wroński did—cf.
[24, 25]). He opts for a balance between the knower and the known, seeing the role of
emotions and will in cognition. He was interested in Leibniz’s view to which he referred
many times—the view that logic is abstracted from reality.
Struve begins his lecture on logic by giving images and concepts. Then he introduces
judgments. By “image” he means a kind of representation of the object through its
characteristics, and “concept” is a set of essential features. Moreover, images result from
certain mental processes. It is the object that makes the mind create images.
Struve attached great importance to the teaching of logic. He thought that teaching
how to think correctly is much more important than giving students concrete contents.
Consequently, he placed a strong emphasis on the teaching of logical culture.
We have already shown that Struve’s conception of logic places him between the old
and new paradigm or rather even in the old paradigm. We have mentioned that he did not
value the role and significance of symbolic and mathematized formal logic but he stressed
psychological questions. Consequently, one should ask what made him not see the
advantages of the new attitude. It seems that one of the reasons was the fact that Struve saw
no cognitive value in pure form devoid of content (cf. [33]). According to his conception
it is the object, i.e., external world that stimulates our thinking, that realizes its existence
and the characteristics of objects, and then using logical methods it creates notions which
in turn it uses, applying logical methods, to formulate judgments. Thus there can be no
cognition without content. Another reason may be that he set a low valuation on the role
and importance of mathematics. Trzcieniecka-Schneider even claims that Struve “did not
understand mathematics, reducing it only to the techniques of operations on numbers”
8 ‘Trudnościzbadania stosunku [. . . ] myślenia do świata przedmiotowego sa˛ oczywiste i sprowadzaja˛

si˛e głównie do tego, że nie jesteśmy w stanie porównywać bezpośrednio naszych wyobrażeń i poj˛eć o
przedmiotach ani pogladów˛ na nie z samymi przedmiotami. Kwestya przedmiotowego poznania prawdy
mogłaby być rozwiazan
˛ a˛ na jego korzyść dopiero wtedy, gdyby si˛e okazało, że prawa naszego umysłu,
a wi˛ec i myślenia, sa˛ zasadniczo zgodne z prawami niezależnego od nas bytu przedmiotowego. [. . . ]
Wykazanie atoli tej zgodności praw umysłu z prawami bytu przedmiotowego wymaga szeregu badań
krytycznych nad wynikami dociekań naukowych.’
[33, p. 93]. In “Filozofia i wykształcenie filozoficzne” [Philosophy and philosophical

education] he wrote:
Mathematics considers only quantitative factors: [. . . ] But in its activities the human mind is not
limited only to quantitative factors but everywhere supplements quantity with quality. [. . . ] One
cannot say that truth contains more or fewer thoughts than falsity: [. . . ] These are all qualitative
differences and they cannot be defined quantitatively; they cannot be understood rightly and
characterized closely from the mathematical point of view but they can only be understood from
a more general standpoint, going much beyond the scope of the quantitative factors alone.9 [30,
pp. 156–157]
Struve also opposed the introduction of quantifiers. He permitted quantitative elements

in logic only in the case of the conversion of judgments and in the square of opposition.
However, he was not consistent in his views when analyzing the relations between the
scopes of concepts using Euler diagrams he actually used arithmetic notation.
In his opinion formal logic “leads [. . . ] only to the development of one-sided formalism
without elevating the essential cognitive value of the relevant forms”10 [31, p. IX] and
“mathematical logic depends on the completely dogmatic transfer of the quantitative
and formal principles to the mental area where quality and content are of primary
importance”11 (ibid.). He also claims that “reducing judgment to equation and basing
conclusion on substance, i.e. substituting equivalents, does not correspond to the real
variety of judgments and conclusions”12 [31, p. X].
Struve’s aversion towards mathematics and mathematical methods in logic was
connected with his views on the function of language in logic and cognition as well as with
his conception of truth. Since if—in accordance with Aristotle—truth is the conformity of
thought to the content of propositions, conducting operations on propositions as symbols
means losing sight of this property to some extent. In fact, the characteristic of being true
does not refer to propositions but to their contents. Symbolically identical propositions
can differ with respect to their contents. Thus the operations conducted on the symbols
of propositions do not have much in common with establishing their truth. And yet logic
leads to the truth about the real world.
The above analysis shows that Struve stood on the threshold of the new way of
understanding and cultivating logic, combining the old and new paradigms. His position
and the meaning of his work was characterized by Kazimierz Twardowski in the following
way:
9 ‘Matematyka rozpatruje wyłacznie

˛ czynniki ilościowe: [. . . ] Tymczasem umysł ludzki w działalności
swej nie jest bynajmniej ograniczonym samemi tylko poj˛eciami ilościowymi, lecz uzupełnia wsz˛edzie
ilość jakościa.˛ [. . . ] Nie można powiedzieć, że prawda zawiera w sobie wi˛ecej lub mniej myśli niż
fałsz: [. . . ] To wszystko sa˛ różnice jakościowe, których ilościowo określić nie można, które należycie
zrozumiane i bliżej scharakteryzowane być nie moga˛ z punktu widzenia matematycznego, lecz poj˛ete być
moga˛ tylko ze stanowiska ogólniejszego, wynoszacego ˛ si˛e wysoko ponad zakres samych tylko czynników
ilościowych.’
10 ‘doprowadza [. . . ] tylko do rozwoju formalizmu jednostronnego bez podniesienia istotnej wartości
poznawczej odnośnych form.’

11 ‘logika matematyczna polega na zupełnie dogmatycznym przeniesieniu zasad ilościowych i formalnych
na pole umysłowe, gdzie jakość i treść maja˛ znaczenie pierwszorz˛edne.’

12 ‘sprowadzanie sadu ˛ do równania oraz oparcie wniosku na substytucyi, czyli podstawianiu
równoważników, nie odpowiada rzeczywistej rozmaitości ani sadów, ˛ ani wniosków.’
290 R. Murawski
So Struve was as if a link connecting this new period with the previous one. Between the
generations of the Cieszkowskis, the Gołuchowskis, the Kremers, the Libelts, the Trentowskis
and the contemporary generation there appears the distinguished figure of this thinker, writer,
who saved from the past what was of lasting value, and he showed the workers of today’s Polish
philosophy the direction through his prudent, and devoid of all prejudices, opinion.13 [34, p. 101]
2 Władysław Biegański
2.1 Life
Władysław Biegański was born in Grabów near Ostrzeszów on 28 April 1857. He studied
medicine in Warsaw in 1875–1880. After having completed his residency in Russia
(1881–1882) he continued studies in Berlin (1882–1883) and Prague (1883). Then he
settled in Cz˛estochowa where he practiced medicine (internal diseases, neurology) for
over 30 years working in a hospital and as a physician for a factory and the railways.
At the same time, he conducted research. His scientific interests included many medical
disciplines as well as the philosophy of medicine, and in particular the methodological and
ethical issues connected with it (cf. his works [4–6, 16]). Biegański represented the Polish
school of the philosophy of medicine (cf. [17]). He was interested also in epistemology
(cf. [9, 10, 14]). Biegański was involved also in social activities. He died in Cz˛estochowa
on 29 January 1917.
2.2 Biegański’s Views on Logic
Biegański’s true passion was in fact logic. As a student he listened to Henryk Struve’s
lectures. Besides his medical practice, he taught logic in local secondary schools for
some time. In 1914 there was even an initiative to appoint Biegański as the professor
of the Jagiellonian University Chair of Logic. It did not occur because of his poor health
condition and the outbreak of the First World War.
In discussing Biegański’s philosophical views on logic, we first should notice that
he was neither a formal nor a mathematical logician. It can be said that he was a
13 ‘Był tedy Henryk Struve jakby ogniwem, łacz˛ acym

˛ ten okres nowy z poprzednim. Mi˛edzy pokoleniem
Cieszkowskich, Gołuchowskich, Kremerów, Libeltów, Trentowskich (It is worth explaining who those
persons were. August Count Cieszkowski (1814–1894)—Polish philosopher, economist and social
and political activist; Józef Gołuchowski (1797–1858)—Polish philosopher, co-creator of the Polish
Romanticist „national philosophy”; Józef Kremer (1806–1875)—Polish historian of art, a philosopher,
an aesthetician and a psychologist; Karol Libelt (1807–1875)—Polish philosopher, writer, political and
social activist, social worker and liberal, nationalist politician; Bronisław Trentowski (1808–1869)—
Polish “Messianist” philosopher, pedagogist, journalist and Freemason, and the chief representative of
the Polish Messianist „national philosophy” [my remark—R.M.].) a pokoleniem współczesnem widnieje
czcigodna postać myśliciela, nauczyciela, pisarza, który z przeszłości ocalił to, co miało w niej wartość
trwała,˛ a dzisiejszym na polu filozofii polskiej pracownikom wskazał drog˛e rozważnym i dalekim od
wszelkiego uprzedzenia sadem.’
˛
philosophical logician from the standpoint formulated by Łukasiewicz (cf. [36]). The
latter characterized philosophical logic in the following way:
If we use here the term “philosophical logic” we mean the complex of problems included in
books written by philosophers, and the logic we were taught in secondary school. Philosophical
logic is not a homogenous science; it contains various issues; in particular, it enters the field
of psychology when it speaks not only about a proposition in a logical sense but also this
psychological phenomenon, which corresponds with a proposition and which is called “judgment”
or “conviction.” [. . . ] Philosophical logic also embraces some issues from the theory of knowledge,
for example, the problem of what truth is or whether any criterion of truth exists.14 [22, pp. 12–13]
It is immediately worth adding that Łukasiewicz himself did not value philosophical
logic—he was of the opinion that the scope of problems it considers is not homogenous,
and also that philosophical logic mixes logic with psychology. Moreover, both fields are
different and use different research methods.15
How did Biegański understand logic? Let us quote two definitions of logic given by
him. In Zasady logiki ogólnej [Principles of General Logic] he wrote:
Logic is the science of the ways or norms of true cognition.16 [7, p. 1]
In Podr˛ecznik logiki i metodologii ogólnej [Manual of Logic and General Methodol-

ogy] we find the following definition:
We call logic the science of the norms and rules of true cognition.17 [8, p. 3]
Therefore, according to Biegański the laws of logic concern the relationships of mental
phenomena because of its aim, which is true cognition. Consequently, logic aims at
investigating cognitive activities of the mind. At the same time Biegański claims that
one should separate and distinguish between logic and the theory of knowledge on the
one hand, and psychology on the other. The reason is that logic is a normative and applied
science whereas both the theory of knowledge and psychology are theoretical. However,
in practice Biegański—like other authors of his time—did not distinguish strictly between
logical and genetic questions, investigating logical constructions both from the precisely
logical and psychological points of view. Yet, it should be noted that in Zasady [7]
Biegański suggests that his conception of logic makes him reject the division into formal
and material truth whereas in Podr˛ecznik [8] he regards this distinction as correct. He also
adds that logic embraces the formal side of cognition.
Consider now his large (638 pages) monograph entitled Teoria logiki [Theory of
Logic] [12] from 1912 which was an attempt to consider the foundations of logic
comprehensively. This work presented general problems concerning logic, the study of
14 ‘Jeżeli
używamy terminu logika filozoficzna, to chodzi nam o ten kompleks zagadnień, które znajduja˛
si˛e w ksiażkach
˛ pisanych przez filozofów, o t˛e logik˛e, której uczyliśmy si˛e w szkole średniej. Logika
filozoficzna nie jest jednolita˛ nauka,˛ zawiera w sobie zagadnienia rozmaitej treści; w szczególności
wkracza w dziedzin˛e psychologii, gdy mówi nie tylko o zdaniu w sensie logicznym, ale także o tym
zjawisku psychicznym, które odpowiada zdaniu, a które nazywa si˛e „sadem” ˛ albo „przekonaniem”. [. . . ]
W logice filozoficznej zawieraja˛ si˛e również niektóre zagadnienia z teorii poznania, np. zagadnienie, co
to jest prawda lub czy istnieje jakieś kryterium prawdy.’
15 Cf. [35] or [26].
16 ‘Logika jest to nauka o sposobach albo normach poznania prawdziwego.’
17 ‘Logika˛ nazywamy nauk˛ e o normach i prawidłach poznania prawdziwego.’
292 R. Murawski
concepts, the study of judgments, the study of argumentation and the study of induction.
Every problem was considered in historical and comparative perspectives on the one hand
and a systematic perspective on the other hand. Although the author focused on the views
of the representatives of traditional logic, he also analyzed the algebra of logic. Biegański
wrote there:
The main aim of logic is to control argumentation. [. . . ] Logic, as the science and art of
argumentation, is an a priori science, i.e. science that draws its content not from experience and
not from the facts given in experience, but from certain a priori presumptions and constructions.18
[12, pp. 34–35]
One sees that logic appears here as a normative science. Therefore Biegański proposes
to use the name “pragmatic logic”. He separates logic from psychology, ontology and
epistemology. The basis of logic is formed by axioms: “the most general laws which are
directly obvious, i.e. requiring no proof” [12, p. 41]. The axioms are the laws of identity,
contradiction, excluded middle and sufficient reason.
This understanding of logic as the art of argumentation can be found in his earlier
treatise “Czem jest logika?” [What is Logic?] [11]. He wrote there:
[. . . ] logic does not reproduce the processes of thought and it does not aim at doing it at all.
Therefore, the definition of logic as the science or art of thinking is actually devoid of any basis.
[. . . ] But the origin of logic shows that this ability [i.e. logic—remark is mine] is neither a science
nor art of thinking, but was created by Plato and Aristotle as the art of argument. Such differences
in views cause serious consequences. If logic is a science or even an art of thinking, it is or should
be a branch of psychology; on the contrary, if it is only the art of argument, it becomes a separate
science that is independent from psychology. Logic as the art of argument does not describe the
ordinary course of thoughts, used in argumentation; it does not reproduce it; it does not find laws
for it, laws expressing the mutual causal relationship of thoughts, but uses ideal constructions
which serve to control the ways of argumentation and in this respect it is explicitly separated from
psychology.19 [11, p. 144]
As a consequence logic “must be of normative character” [11, p. 145]. Biegański

explains this in the following way:
The essence of argumentation consists in valuing. Looking for a proof of any proposition we
always follow the question about its cognitive value.20 [11, p. 145]
18 ’Logika ma na celu głównie kontrol˛ e dowodzenia. [. . . ] Logika, jako nauka i sztuka dowodzenia jest
nauka˛ aprioryczna,˛ tj. taka,˛ która swoja˛ treść czerpie nie z doświadczenia, nie z faktów w doświadczeniu
nam danych, lecz z pewnych naprzód powzi˛etych założeń i konstrukcji.
19 ‘[. . . ] logika nie odtwarza procesów myśli i nie ma wcale na celu tego zadania. To też określenie
logiki jako nauki lub sztuki myślenia jest pozbawione właściwie wszelkiej podstawy. [. . . ] Tymczasem
geneza logiki wykazuje, że umiej˛etność ta [tzn. logika—my remark, R.M.] nie jest ani nauka,˛ ani sztuka˛
myślenia, lecz utworzona została przez Platona i Arystotelesa jako sztuka dowodzenia. Takie różnice
w zapatrywaniach prowadza˛ za soba˛ poważne konsekwencje. Jeżeli logika jest nauka˛ lub nawet sztuka˛
myślenia, to w każdym razie jest lub powinna być działem psychologii, przeciwnie, jeżeli jest tylko
sztuka˛ dowodzenia, to staje si˛e nauka˛ odr˛ebna,˛ niezależna˛ od psychologii. Logika jako sztuka dowodzenia
nie opisuje zwykłego biegu myśli, stosowanego przy dowodzeniu, nie odtwarza go, nie wynajduje dla
niego praw, wyrażajacych˛ wzajemny zwiazek ˛ przyczynowy myśli, lecz posługuje si˛e konstrukcyami
idealnymi, które służa˛ dla kontroli sposobów dowodzenia i pod tym wzgl˛edem odgranicza si˛e wyraźnie
od psychologii.’
20 ‘Istota dowodzenia polega na wartościowaniu. Poszukujac ˛ dowodu dla jakiegokolwiek zdania, kieru-
jemy si˛e zawsze pytaniem o jego wartości poznawczej.’
What is important here is the veracity of a proposition and not its meaning and content.
“Every proof consists in stating the consistency between the content of the proposition and
the principles, which we recognize as true, and it is in this consistency that the essence of
truth lies” [11, p. 145].21
One should ask now what then is the relation between logic and psychology? It must
be noted that Biegański stressed their autonomy. He wrote:
Any direct [. . . ] dependence here is out of the question. Nonetheless, psychological investigations
are not completely meaningless to logic since they constitute an important control for logical
constructions [. . . ]. An ideal logical construction would be one that is the closest to the real
course of thoughts, that completely guarantees to distinguish truth and is easy to apply. [. . . ] Thus
psychological investigations are undoubtedly of great importance for the development of logic
because they can contribute to formulating new constructions which are the closest to the natural
course of thoughts.22 [11, pp. 147–148]
In this context the following question arises: what did Biegański mean by argumen-
tation? In fact, one do not find by him any clear idea of inference. He neither used
the concept of logical deduction nor distinguished between deductive and inductive
reasoning. He states only that inference is based on the idea of necessity, that the principles
of logic refer to the form and not the content of cognition. These ideas, however, are not
clear and they are mixed. In the work “Sposobność logiczna w świetle algebry logiki”
[Logical Modality in the Light of the Algebra of Logic] [13] he speaks about reliable and
possible deduction, which is a misunderstanding.
Discussing Biegański’s conception of logic one must add that his departure from
psychologism was not definitive. In fact in Podr˛ecznik logiki ogólnej [Manual of General
Logic] [15] from 1916 he returned to psychologism. He wrote there:
We call logic the science about the ways of controlling the truth of our cognitive thoughts.23
[15, p. 1]
Biegański’s conceptions concerning the foundations and philosophy of logic did not
evoke much interest and his work was criticized. Let us tell for example about the reaction
of Łukasiewicz. He published in the journal Ruch Filozoficzny the review of Biegański’s
work “Czem jest logika?” [11]. Łukasiewicz stressed there Biegański’s departure from
psychologism but noticed that it was not completely consistent. He also emphasized the
fact that Biegański’s conception of logic was too narrow—in fact he limited it to inference.
In Łukaszewicz’s opinion the object of logic should be reasoning in general, which should
include non-deductive reasoning as well. He wrote:
Logic does not only concern argumentation but reasoning in general, while using the term
‘reasoning’ as more general than ‘argumentation’ in accordance with Prof. Twardowski’s view
21 ‘Każdy dowód polega na stwierdzeniu zgodności treści zdania z zasadami, które uznajemy za
prawdziwe i w tej właśnie zgodności tkwi istota prawdy.’

22 ‘O bezpośredniej [. . . ] zależności nie może tu być mowy. Pomimo to badania psychologiczne nie sa˛
zupełnie bez znaczenia dla logiki, stanowia˛ bowiem bardzo ważna˛ kontrol˛e dla konstrukcji logicznych.
[. . . ] Ideałem konstrukcyi logicznej byłaby taka, któraby si˛e najbardziej zbliżała do rzeczywistego biegu
myśli, dawała zupełna˛ gwarancy˛e w odróżnianiu prawdy i była łatwa do stosowania. [. . . ] To też badania
psychologiczne maja˛ niewatpliwie
˛ duże znaczenie w rozwoju logiki, gdyż moga˛ si˛e przyczynić do
wynalezienia konstrukcyi nowych, najbardziej zbliżonych do naturalnego biegu myśli.’
23 ‘Logika˛ nazywamy nauk˛ e o sposobach kontrolowania prawdy naszych myśli poznawczych.’
294 R. Murawski
(cf. my dissertation O twórczości w nauce [About Creativity in Science], p. 8). Secondly,

argumentation or reasoning is also thinking, and thus psychologism returns. I would agree to
distinguish between logic as ‘science’ and ‘art’ but I would use different terms. Namely, I think
that logic as a theoretical science investigates relations which formal propositions (e.g. S is P)
maintain because of their truth or falsity, and makes laws of these relations (e.g. ‘if it is true that S
is M and M is P, it is also true that S is P’); as a practical science it applies these laws to solve tasks
in the area of reasoning in general, for instance to introduce some conclusion, like in induction, to
verify or prove some thesis, etc.24 (cf. [21])
So we have here the distinction between theoretical and practical logic, which
corresponds to the distinction between logica docens and logica utens. Łukasiewicz
defines both in an anti-psychologistic way. Briefly speaking, he understands logic as
argumentation theory, dividing reasoning into deductive and reductive; further dividing
deductive reasoning into concluding and verifying, and reductive reasoning: proving and
explicating (cf. [12]).25
We see that Biegański was involved only in the traditional paradigm of logic. However,
it should be added that he saw the advantages of the new approach, in particular the
values and advantages of the algebra of logic. In the introduction to the work “Sposobność
logiczna w świetle algebry logiki” [13], in which he attempted to apply the algebra of
logic to the theory of modal categories, he wrote:
Although logical calculus, called the algebra of logic or logistics, has not and cannot have a
large practical application, considering the logical evaluation of our judgments and conclusions,
it has undoubtedly important theoretical significance. [. . . ] Yet, algebraic symbols, which we use
in logical calculus, separate clearly the object of investigation from psychological factors and
objective relations, and bring to light all the properties of pure logical relations. Therefore, the
main value of the algebra of logic consists in the fact that using it we can explain more thoroughly
and mark strictly the relations that are explained variously in school logic.26 [13, p. 67]
It should be added however that those attempts of Biegański brought a poor result.
24 ‘Logika tyczy si˛e nie tylko dowodzenia, ale w ogóle rozumowania, przy czym zgodnie z prof.
Twardowskim używam terminu “rozumowanie” jako ogólniejszego od “dowodzenia” (cf. rozpraw˛e moja˛
O twórczości w nauce, str. 8). Po wtóre, dowodzenie czy rozumowanie jest także myśleniem, a wi˛ec
psychologizm powraca. Zgodziłbym si˛e natomiast na odróżnienie logiki jako “nauki” i “sztuki”, tylko
użyłbym innych terminów. Sadz˛
˛ e mianowicie, że logika jako nauka teoretyczna bada stosunki, w jakich
zdania formalne (np. S jest P) pozostaja˛ do siebie ze wzgl˛edu na swoja˛ prawdziwość lub fałszywość, i
ustanawia prawa tych stosunków (np. “jeśli prawda˛ jest, że S jest M i M jest P, to prawda˛ jest, że S jest P”);
jako nauka praktyczna stosuje te prawa do rozwiazywania
˛ zadań z zakresu rozumowania w ogóle, np. do
wyprowadzenia jakiejś konkluzji, jak we wnioskowaniu indukcyjnym, do sprawdzenia lub udowodnienia
jakiejś tezy itp.’
25 More on Łukasiewicz’s views on logic can be found in [35] as well as in [26] and [27].
26 ‘Rachunek logiczny, zwany algebra˛ logiki lub inaczej jeszcze logistyka, ˛ jakkolwiek nie ma i nie może
mieć rozległego zastosowania praktycznego przy ocenie wartości logicznej naszych sadów i wniosków,
posiada jednak niewatpliwie
˛ ważne teoretyczne znaczenie. [. . . ] Tymczasem symbole algebraiczne,
jakimi si˛e w rachunku logicznym posługujemy, odrywaja˛ wyraźnie przedmiot badania zarówno od
czynników psychicznych jako też od stosunków objektywnych i wydobywaja˛ na jaw wszystkie właś-
ciwości czystych stosunków logicznych. To też główna wartość algebry logiki polega na tem, że przy
jej pośrednictwie możemy dokładniej wyjaśnić i ściślej wyznaczyć stosunki, które w logice szkolnej
rozmaicie bywaja˛ tłomaczone.’
3 Conclusion
The above analyses lead to the conclusion that Struve and Biegański represented the
traditional “pre-mathematical” approach to logic. In fact the logical culture was decisively
“pre-mathematical” in Poland at the turn of the nineteenth and the twentieth centuries.
This opinion is supported, for example, by the first edition of Poradnik dla samouków
[A Guide for Autodidacts] which includes Adam Marburg’s paper “Logika i teoria
poznania” [Logic and the Theory of Knowledge] (cf. [23]), written in a rather old-
fashioned manner. It was so despite the fact that the first signs of interest in mathematical
logic appeared in Poland already in the 1880s. Let us mention here the treatise of
Stanisław Piatkiewicz
˛ Algebra w logice [Algebra in Logic] published in 1888—cf. [1]
and [2] as well as [3].
Struve’s and Biegański’s views on logic were shared by their contemporaries, for
example by Władysław Kozłowski (1832–1899) and Władysław Mieczysław Kozłowski
(1858–1935). The former characterized logic in his work Logika elementarna [Elemen-
tary Logic] in the following way:
logic is the science about mental activities with the aid of which we reach truth and prove it.27 [18,
p. 1]
In turn, Władysław M. Kozłowski wrote in Podstawy logiki [The Foundations of

Logic]:
Logic is the science about the activities of the mind which seeks truth.28 [19, p. 8]
The first chapter of his work was entitled “Thinking as Object of Logic” [19, p. 22].
This thought was repeated in the work Krótki zarys logiki [A Brief Outline of Logic]
where it was claimed that logic is a normative science and its task is “to examine the ways
leading the mind to truth” [20, p. 1].29 However, it was stressed that logic:
analyses mental operations conducted to reach the truth in a form that is so general that it could
be applied to any content. It investigates its form, separating it completely from the content. Logic
shares this property with mathematics [. . . ]. [. . . ] This formal character, common to logic and
mathematics, made these sciences close in their attempts, which were less or more developed, and
led to the creation of mathematical logic.30 [20, pp. 8–9]
27 Logika jest nauka˛ o czynnościach umysłowych, za pomoca˛ których dochodzimy prawdy i jej dowodz-
imy.
28 Logika jest nauka o czynnościach umysłu poszukujacego ˛ prawdy.
29 ‘badanie dróg, prowadzacych˛ rozum do prawdy.’
30 bada operacye umysłowe, wykonywane w celu osiagni˛ ˛ ecia prawdy w formie tak ogólnej, iżby mogły
zastosować si˛e do jakiejkolwiekbadź˛ treści. Bada je ze stanowiska ich formy, odrywajac ˛ si˛e zupełnie od
treści. Własność t˛e podziela z logika˛ matematyka [. . . ]. [. . . ] Ten formalny charakter, wspólny logice z
matematyka,˛ spowodował zbliżenie do siebie obu nauk w próbach mniej lub dalej posuni˛etych i znalazł
wyraz w utworzeniu logiki matematycznej.
296 R. Murawski
Finally, it was stated that logic can be defined “as the science about the forms of every
ordered field of real or imaginary objects” [20, p. 9].31
In Poland the road to the new paradigm in logic, to the new understanding of it, was
long and difficult. Only in the next generation of Polish scholars the new mathematical
approach to logic can be seen.
References
1. Batóg, T.: Stanisław Piatkiewicz

˛ — pionier logiki matematycznej w Polsce. Kwartalnik Historii Nauki
i Techniki 16, 553–563 (1971)
2. Batóg, T.: Stanisław Piatkiewicz
˛ — pionier logiki matematycznej w Polsce. Z Dziejów Kultury i
Literatury Ziemi Przemyskiej II, 325–330 (1973)
3. Batóg, T., Murawski, R.: Stanisław Piatkiewicz
˛ and the beginnings of mathematical logic in Poland.
Hist. Math. 23, 68–73 (1996)
4. Biegański, Wł.: Logika medycyny, czyli Zasady ogólnej metodologii nauk lekarskich. W. Drukarnia
W. Kowalewskiego, Warszawa (1894) (2nd edn. (1908); German translation: Medizinische Logik:
Kritik der ärztlichen Erkenntnis. C. Kabitzsch, Würzburg (1909))
5. Biegański, Wł.: Zagadnienia ogólne z teorii nauk lekarskich. Drukarnia Towarzystwa S. Orgelbranda
Synów, Warszawa (1897)
6. Biegański, Wł.: Myśli i aforyzmy o etyce lekarskiej. W.L. Anczyc, Warszawa (1899)
7. Biegański, Wł.: Zasady logiki ogólnej. Kasa Józefa Mianowskiego, Warszawa (1903)
8. Biegański, Wł.: Podr˛ecznik logiki i metodologii ogólnej dla szkół średnich i dla samouków. E. Wende,
Warszawa; W.L. Anczyc, Lwów; H. Altenberg, Kraków (1907)
9. Biegański, Wł.: Wnioskowanie z analogii. Polskie Towarzystwo Filozoficzne, Lwów (1909)
10. Biegański, Wł.: Traktat o poznaniu i prawdzie. Zapomoga Kasy dla Osób Pracujacych ˛ na Polu
Naukowem im. Józefa Mianowskiego, Warszawa (1910)
11. Biegański, Wł.: Czem jest logika? Sprawozdania z Posiedzeń Towarzystwa Naukowego Warsza-
wskiego, Wydział Nauk Antropologicznych, Społecznych, Historii i Filozofii, year III(6), 119–152
(1910)
12. Biegański, Wł.: Teoria logiki. E. Wende i S-ka, Warszawa (1912)
13. Biegański, Wł.: Sposobność logiczna w świetle algebry logiki. Sprawozdania z Posiedzeń To-
warzystwa Naukowego Warszawskiego, Wydział Nauk Antropologicznych, Społecznych, Historii i
Filozofii, year V(9), 67–79 (1912)
14. Biegański, Wł.: Teoria poznania ze stanowiska zasady celowości. Towarzystwo Naukowe Warsza-
wskie, Warszawa (1915)
15. Biegański, Wł.: Podr˛ecznik logiki ogólnej. E. Wende i spółka (Ludwik Fiszer), Warszawa; H.
Altenberg, G. Seyfarth, E. Wende i S-ka, Lwów; M. Niemierkiewicz, Poznań; Ludwik Fiszer, Łódź
(1916)
16. Biegański, Wł.: Etyka ogólna. Gebethner i Wolff, Warszawa (1918)
17. Borzym, S.: Poglady ˛ epistemologiczne Władysława Biegańskiego. Filozofia Nauki 6(3–4), 11–17
(1998)
18. Kozłowski, Wł.: Logika elementarna. Nakładem Towarzystwa Nauczycieli Szkół Wyższych, Lwów
(1891)
19. Kozłowski, Wł.M.: Podstawy logiki, czyli Zasady nauk: wykład systematyczny dla szkół wyższych
oraz dla samouctwa. Wydawnictwo M. Arcta, Warszawa (1916)
20. Kozłowski, Wł.M.: Krótki zarys logiki wraz z elementami ideografii logicznej. Wydawnictwo M.
Arcta, Warszawa (1918)
21. Łukasiewicz, J.: W. Biegański, Czym jest logika? [review]. Ruch Filozoficzny 2(8), 149a–149b (1912)
31 ‘jako nauk˛e o formach każdej uporzadkowanej

˛ dziedziny przedmiotów rzeczywistych lub urojonych.’
22. Łukasiewicz, J.: Elementy logiki matematycznej. Komisja Wydawnicza Koła Matematyczno-
Fizycznego Słuchaczów Uniwersytetu Warszawskiego, Warszawa (1929). Ed. by M. Presburger
(Reprint: Wydawnictwo Naukowe Uniwersytetu im. Adama Mickiewicza, Poznań (2008))
23. Marburg, A.: Logika i teoria poznania. In: Poradnik dla samouków. Wydawnictwo A. Heflera i St.
Michalskiego, Warszawa (1902)
24. Murawski, R.: Genius or Madman? On the Life and Work of J.M. Hoene-Wroński. In: Wi˛esław,
W. (ed.) European Mathematics in the Last Centuries, pp. 77–86. Stefan Banach International
Mathematical Center/Institute of Mathematics, Wrocław University, Wrocław (2005)
25. Murawski, R.: The Philosophy of Hoene-Wroński. Organon 35, 143–150 (2006)
26. Murawski, R.: Filozofia matematyki i logiki w Polsce miedzywojennej. Monografie Fundacji na rzecz
Nauki Polskiej, Wydawnictwo Naukowe Uniwersytetu Mikołaja Kopernika, Toruń (2011)
27. Murawski, R.: The Philosophy of Mathematics and Logic in the 1920s and 1930s in Poland.
Birkhäuser Verlag, Basel (2014)
28. Struve, H.: Logika poprzedzona wst˛epem psychologicznym. Odczyt Uniwersytecki Dra Henryka
Struve profesora Szkoły Głównej w Warszawie (1863). Lithograph
29. Struve, H.: Wykład systematyczny logiki czyli nauka dochodzenia i poznania prawdy, vol. 1: Cz˛eść
wst˛epna (Historya logiki u obcych i w Polsce). Warszawa (1868–1870)
30. Struve, H.: Filozofia i wykształcenie filozoficzne. In: Lubomirski, J.T., Stawiski, E., Plebański, J.K.
(eds.) Encyklopedia wychowawcza, vol. IV. J. Drukarnia J. Sikorskiego, Warszawa (1885)
31. Struve, H.: Logika elementarna. Podr˛ecznik dla szkół i samouków z dodaniem słownika terminów
logicznych. Wydawnictwo M. Arcta, Warszawa (1907)
32. Struve, H.: Historya logiki jako teoryi poznania w Polsce. Printed by the author, Warszawa (1911)
33. Trzcieniecka-Schneider, I.: Logika Henryka Struvego. U progu nowego paradygmatu. Wydawnictwo
Naukowe Uniwersytetu Pedagogicznego, Kraków (2010)
34. Twardowski, K.: Henryk Struve. Ruch Filozoficzny 2(6) (1912)
Dordrecht (1989)
36. Woleński, J.: Władysław Biegański jako logik. Filozofia Nauki 6(3–4), 19–26 (1998)
37. Woleński, J.: Mathematical logic in Warsaw: 1918–1939. In: Ehrenfeucht, A., Marek, V.W., Srebrny,
M. (eds.) Andrzej Mostowski and Foundational Studies, pp. 30–46. IOS Press, Amsterdam (2008)
R. Murawski ()
Adam Mickiewicz University in Poznań, Faculty of Mathematics and Computer Science, Poznań, Poland
Part II
Warsaw School of Logic, Its Main Figures
and Ideas: The Period of Prosperity
As it was mentioned in the previous part, the famous branch of LWS was Warsaw
School of Logic (WSL). Its founders were Jan Łukasiewicz and Stanisław Leśniewski,
who, after re-activation of the University of Warsaw, got professorships at the Faculty
of Mathematics and Natural Sciences. The biographies of both were introduced in Part
I. Here, in Part II, we present some papers connected with their achievements, as well
as biographies of some important disciples and representatives of WSL, followed by the
articles on their output.
Łukasiewicz and His Followers
in Many-Valued Logic
Grzegorz Malinowski
Abstract The aim of this work is a concise introduction to the Łukasiewicz logical world:
three and n-valued, n-natural or infinite, denumerable or of the power of continuum. We
present Łukasiewicz inventory work and its rationale, the elaboration of original ideas and
their technical complementation. Finally, we attempt to show the impact of Łukasiewicz
conceptions, their development and directions of resulting applications.
Keywords Logical value · Many-valuedness · Matrix consequence · Matrix

semantics · Functional completeness · Many-valued algebra · Probability · Fuzzy set
Mathematics Subject Classification (2000) Primary 03B50; Secondary 03A99
1 Łukasiewicz’s Way to Many-Valuedness
The assumption stating that to every proposition it may be ascribed exactly one of the
two logical values, truth or falsity, called the principle of bivalence, constitutes the basis
of classical logic. It determines both the subject matter and the scope of applicability
of the logic. While the roots of many-valued logics can be seen in Aristotle—with his
famous concern for future contingents and the sea-battle tomorrow—and traced through
the middle ages and the nineteenth century, the real ‘era of many-valuedness’ began in
1920 with the work of Łukasiewicz and Post. In this chapter we will look carefully at the
first approach, which since its history started as early as in 1918 may be considered as the
first modern approach to many-valuedness.
The actual introduction of a third logical value by Łukasiewicz [20], next to truth and
falsity, was preceded by thorough philosophical studies. Their crowning achievement was
a three-valued propositional calculus which, from the point of view of logic, represented
a standard line of approach. However, in view of the surprisingly rich motivation
substantiating the new logic and also the hopes it produced, its exceptional position has
been maintained.

302 G. Malinowski
1.1 Łukasiewicz and the Lvov–Warsaw School
Łukasiewicz’s work, especially that done in 1906–1922, resulted in a significant philo-

sophical context. The first and second decade of the 1920s witnessed the development
of the so-called Lvov–Warsaw philosophical school of which Łukasiewicz was a co-
originator and a prominent representative (see Woleński [49]).
It is an unenviable task to decide which of the debates carried out in the Lvov–Warsaw
school directly contributed to the chief logical discovery of Łukasiewicz. According to
several notes in his works one may maintain that the main source of Łukasiewicz’s views
on the logic may be subscribed to a discussion concerning the Brentano–Twardowski–
Meinong general theory of objects. Created by Meinong towards the end of the nineteenth
century, the theory of contradictory objects postulated the existence of objects having
contradictory properties, such as, for example, squaring the circle. Meinong claimed that
the non-existence of contradictory objects would result in the inability to utter any true
proposition, notably a proposition stating that they are not objects. Łukasiewicz advocated
the theory of contradictory objects. What is more, he even tended to think that non-
contradictory objects do not exist. It is noteworthy that the turn of the nineteenth century
witnessed Russell’s discovery of a paradox in set theory with the comprehension axiom.
This fact was acknowledged by Łukasiewicz who, consequently, attacked in [18] the
logical principle of contradiction (see Sect. 1.2).
The important part of Łukasiewicz’s studies was concerned with the problems of
induction and the theory of probability. Especially, while dealing with the latter, he
extricated himself from the ‘embarrassing’ principle of contradiction (see [19]) and
classified as undefinite the propositions with free nominal variables. Subsequently, he
assigned fractional ‘logical’ values to undefinite propositions indicating the proportions
between the number of actual variable values verifying a proposition and the number
of all possible values of that variable. Under this conception, ‘logical values’ exhibit
the feature of relativity and depend on the set of individuals actually evaluated. So, for
example, the value of the proposition ‘x 2 = 1’ amounts to 2/3 in the set {−1, 0, 1} and
to 2/5 in the set {−2, −1, 0, 1, 2}. Obviously enough, infinite sets of individuals are not
admitted and this immediately implies that Łukasiewicz’s suggestion cannot be taken
seriously within the theory of probability (some attempts to improve the idea are due to
Zawirski [54]). However, independently of the justification for combining the problems
of many-valuedness with probability the crucial fact is that already in 1913 Łukasiewicz
had employed the concept of logical value in an unorthodox manner.
The studies that finally led Łukasiewicz to the construction of three-valued logic
touched upon determinism, indeterminism and some related problems like the causality
principle and modality (i.e. possibility and necessity; see [17, 18]). Some historians of
logic suspect Łukasiewicz of being influenced by the discourse (in the Lvov–Warsaw
school) about freedom and creativity. Its main thesis was formulated by Kotarbiński who
in [15] suggested the need for revising the two-valued logic that seemed to interfere with
the freedom of human thinking. Łukasiewicz was a fierce follower of indeterminism,
which found its expression, among others, in the introduction of the third logical value,
next to truth and falsity, to be assigned to non-determined propositions; specifically, to
propositions describing casual future events, i.e. future contingents.
Łukasiewicz and His Followers in Many-Valued Logic 303
1.2 The Third Logical Value
The very earliest remarks about the three-valued propositional calculus can be found in
the Farewell Lecture given by Łukasiewicz in the Assembly Hall of Warsaw University on
the 7th of March, 1918. Next in the paper ‘O logice trójwartościowej’, [20], one may find
an outline of the three-valued logic and a brief motivation for the new logical construction
casting off the principle of bivalence. Łukasiewicz in [21] analyses the sentence ‘I shall
be in Warsaw at noon on 21 December of the next year’ and argues that at the time of its
utterance, it is neither true nor false, since otherwise fatalist conclusions about necessity
or impossibility of contingent future events would follow, its value (truth or falsity) is not
settled. Hence, the sentences of this kind pertain to a ‘third’ logical category.
Consequently, to the two classical values 0 and 1, Łukasiewicz added an intermediate
logical value 1/2 interpreted as ‘possibility’ or ‘indeterminacy’—the former of these
options was subsequently repudiated by him under the influence of the studies on
modality. First, Łukasiewicz extends the classical interpretation of implication (→) and
negation (¬) connectives, which resulted in the putting forward of the following tables1 :
α ¬α → 0 1/
2 1
0 1 0 1 1 1
1/ 1/ 1/ 1/ 1 1
2 2 2 2
1 0 1 0 1/ 1
2
The remaining connectives of disjunction, conjunction and equivalence were later

introduced through the sequence of the following definitions:
α ∨ β = (α → β) → β
α ∧ β = ¬(¬α ∨ ¬β)
α ≡ β = (α → β) ∧ (β → α).
Consequently, their tables are as follows:
∨ 0 1/
2 1 ∧ 0 1/
2 1
0 0 1/ 1 0 0 0 0
2
1/ 1/ 1/ 1 1/ 0 1/ 1/
2 2 2 2 2 2
1 1 1 1 1 0 1/ 1
2
≡ 0 1/
2 1
0 1 1/ 0
2
1/ 1/ 1 1/
2 2 2
1 0 1/ 1
2
1 The truth-tables of binary connectives ∗ are viewed as follows: the value of α is placed in the first vertical
line, the value of β in the first horizontal line and the value of α ∗ β at the intersection of the two lines.
304 G. Malinowski
A valuation of the set of formulas, For, in the three-valued logic is any function v :
For → {0,1/2 , 1} ‘compatible’ with the above tables; a three-valued tautology is a formula
which under any valuation v takes on the designated value 1. An instant reflection shows
that any such formula is the classical tautology.
System Ł3 = T aut3 of the three-valued propositional calculus differs radically from
CPC. On the one hand some important laws of the classical logic are not tautologies of
Ł3, as
(1) p ∨ ¬p (law of the excluded middle)
(2) ¬(p ∧ ¬p) (principle of contradiction).
but on the other hand some classically contradictory formulae2 are consistent in the logic
of Łukasiewicz. An important example of such formula is
(3) p ≡ ¬p.
All these properties can be checked with the help of any valuation which yields 1/2 for p.
Such a valuation associates 1/2 with (1), (2) and 1 with (3).
The thorough-going refutation of the law of the excluded middle and the principle of
contradiction was intended, in Łukasiewicz’s opinion, to codify the principles of inde-
terminism: ‘both disjunction and conjunction of two possible propositions are possible
propositions (and nothing else)’. Whereas the consistency of (3) supports the claim that
Łukasiewicz logic is adjusted to the formalization of reasonings about contradictory
objects. Also, as Łukasiewicz remarks, Russell’s paradox of ‘the set of all sets that are
not their own elements’ ceases to be an antinomy in Ł3. The Russell’s set Z is defined as
Z = {x : x ∈ x}, and therefore the equivalence
Z ∈ Z ≡ Z ∈ Z,
which is a substitution of (3) is not-contradictory, since putting for 1/2 p makes it true.
1.3 Modality, Axiomatization and Interpretation
Another Łukasiewicz’s intention was to get formalization of modal functors of possibility

M and necessity L. which would preserve the consistency of middle ages’ intuitive
theorems on modal propositions. Since expressing these functors within the truth-
functional classical logic is impossible, Łukasiewicz took the three-valued logic as a new
base. In 1921 Tarski gave simple definitions, using negation and implication, of the two
connectives which meeting the Łukasiewicz’s requirements3:
x Mx x Lx
Mα =df ¬α → α
0 0 0 0
1/ 1 1/ 0
2 2
Lα =df ¬M¬α = ¬(α → ¬α)
1 1 1 1
2 That is, formulae taking 0 at arbitrary logical valuation.

3 See Łukasiewicz [21].
Using M, L and other Łukasiewicz connectives we get third modal connective “it is
contingent that”or, “it is modally indifferent”:
α Iα
0 0
1/
I α = Mα ∧ ¬Lα
2 1
1 0
We may treat I as an operator distinguishing the intermediate logical value 1/2 . Further,
applying I allows the formulation within Ł3, of counterparts of the law of the excluded
middle and the principle of contradiction:
p ∨ Ip ∨ ¬p
¬(p ∧ ¬Ip ∧ ¬p)
rendering altogether that Łukasiewicz’s logic is three-valued.

In spite of the promising combination of trivalence and modality the full elaboration
of modal logic on the basis of the three-valued logic never succeeded (with the mere
exception of algebraic constructions of Moisil), which was the result of the further
Łukasiewicz’s investigations on modal sentences (see Łukasiewicz [21]). Many years
after Łukasiewicz comes back to the idea of construction of modal logic of possibility
and necessity founded of the four element Boolean matrix, Łukasiewicz [22].
Through the law of excluded fourth and the extended contradiction principle, Ł3
expresses its three-valuedness. However, contrary to the classical case this expression
it is limited since not all connectives described by {0,1/2 , 1}–tables are definable through
formulas. One important example is the constant connective T , such that T x = 1 /2 for
any x ∈ {0,1/2 , 1}.
The axiomatization of Ł3 due to Wajsberg [47] is the first known axiomatization
of a system of many-valued logic. Accepting the rules MP (detachment) and SUB
(substitution) Wajsberg axiom system for (¬, →)-version of Łukasiewicz’s three-valued
propositional calculus is as follows:
W1. p → (q → p)
W2. (p → q) → ((q → r) → (p → r))
W3. (¬p → ¬q) → (q → p)
W4. ((p → ¬p) → p) → p.
The result obviously applies to the whole Ł3 since the other Łukasiewicz connectives
are definable using those of negation and implication. Słupecki [39] enriched the set of
primitives by T and extending Wajsberg system with
W5. Tp → ¬Tp
W6. ¬Tp → Tp
got an axiom system for a functionally complete three-valued logic.
306 G. Malinowski
2 Matrix Semantics for n-Valued Logics
In 1922 Łukasiewicz generalized his three-valued logic and defined the family of many-
valued logics, both finite and infinite-valued.4 Łukasiewicz n-valued matrix has the form
Mn = (Ln , ¬, →, ∨, ∧, ≡, {1}),
where
⎧
⎨ {0, 1/n−1 , 2/n−1 , . . . , 1} if n ∈ N, n ≥ 2
Ln = {s/w : 0 ≤ s ≤ w, s, w ∈ N and w = 0} if n = ℵ0
⎩
[0, 1] if n = ℵ1 .
and the functions are defined on Ln as follows:

(i) ¬ x = 1 − x
x → y = min(1, 1 − x + y)
(ii) x ∨ y = (x → y) → y = max(x, y)
x ∧ y = ¬(¬x ∨ ¬y) = min(x, y)
x ≡ y = (x → y) ∧ (y → x) = 1 − |x − y|.
2.1 Basic Matrix Properties and Consequence
The introduction of new many-valued logics was not supported by any separate argu-
mentation. Łukasiewicz merely underlined, that the generalization was correct since for
n = 3 one gets exactly the matrix of his 1920’ three-valued logic. The future history will,
however, show that Łukasiewicz logics have nice properties, which locate them among
the most important logical constructions.
First, the Łukasiewicz matrix M2 coincides with the matrix of the classical logic.
And, since the set {0, 1} is closed with respect to all Łukasiewicz connectives A2 is a
subalgebra of any algebra (Ln , ¬, →, ∨, ∧, ≡) and M2 is a submatrix of Mn . Therefore
all tautologies of n-valued Łukasiewicz propositional calculi, T autn are included in T aut,
i.e. the set of tautologies of the classical logic.
T autn ⊆ T aut2 = T aut.
What is more, the family of all finite Łukasiewicz matrices has more submatrix properties,
between tautologies of finite matrices further inclusion relations hold. Actually, they are
established by the following Lindenbaum condition5:
For finite n, m ∈ N, T autn ⊆ T autm iff m − 1 is a divisor of n − 1.
4 See Łukasiewicz [23, 1970, p. 140].

5 See Łukasiewicz and Tarski [24].
It may be proved that the infinite-valued Łukasiewicz matrices produce the same set of
tautologies, which is an intersection of all T autn :

T autℵ0 = T autℵ1 = {T autn ) : n ≥ 2, n ∈ N}.
For any n as above, one may define the matrix consequence relation |n putting for a set of
formulas X and a formula α, X |n α iff for any valuation v of formulas in Mn , v(α) = 1
whenever v(X) ⊆ {1}). Then, using the same argument one may extend onto finite n and
m the Lindenbaum tautology inclusion result for matrix consequence relations |n :
|n ⊆ |m iff m − 1 is a divisor of n − 1.
Completing the view for n infinite, we should say that though the matrices Mℵ0 and
Mℵ1 define the same sets of tautologies, their consequence relations are different, more
precisely,
|Mℵ1 ⊆ |Mℵ0 , but |Mℵ1 = |Mℵ0 .
The first inclusion is easy: all valuations in C are valuations in Mℵ1 . The reverse, however,
due to the absence of irrational numbers in Lℵ0 , which obviously are in Lℵ1 , is not true,
see Wójcicki [51].
The last two formulas are but some of the properties of many-valued consequence
relation and operation. An ingenious general elaboration Tarski’s idea of logical conse-
quence [44] by Wójcicki and his collaborators, led to a general matrix approach in the
theory of sentential calculi, see [50]. Łukasiewicz matrices and logics played a major
role as an experimental background in the crucial period of this creation, see Wójcicki,
Malinowski [52].
Łukasiewicz n-valued logics Łn or, equivalently, their algebras are not functionally
complete. All what was established for n = 3 applies for each finite n. First, no constant
except 0 and 1 is definable in (Ln , ¬, →, ∨, ∧, ≡). Second, adding the constants to the
stock of connectives makes this algebra functionally complete. And, since Mn is one
generated, either by 1/n−1 or by n−2/n−1 , also adding only one of them do the job as
well. McNaughton [25] proved an ingenious definability criterion, both finite and infinite,
which shows the mathematical beauty of Łukasiewicz’s logic constructions.
2.2 Axiomatizability
A proof that finite matrices are axiomatizable was given in Łukasiewicz and Tarski [24].
However, the problem of formulation of a concrete axiom system for finite Łukasiewicz
logics for n > 3 remains open till 1952. Rosser and Turquette [37] are the authors
of a general method of axiomatization of n-valued logics with connectives satisfying
the so-called standard conditions. The method can be applied, among others, to Łn
since such connectives are either primitive or definable in Łukasiewicz finite matrices.
Hence, for every n an axiomatization of Łukasiewicz’s n-valued propositional calculus
308 G. Malinowski
can be obtained. The axiomatization, however, becomes very complicated due to the high
generality of the method given by Rosser and Turquette.
In 1930 Łukasiewicz conjectured that his ℵ0 -valued logic was axiomatizable (see [24])
and that the axiomatization of the infinite-valued propositional calculus together with MP
and SUB was the following:
L1. p → (q → p)
L2. (p → q) → ((q → r) → (p → r))
L3. ((p → q) → q) → ((q → p) → p)
L4. (¬p → ¬q) → (q → p)
L5. ((p → q) → (q → p)) → (q → p).
Due to Łukasiewicz6 this hypothesis was confirmed by Wajsberg in 1931. Next comes
the reduction of the axiom set: Meredith [26] and Chang [2] independently showed that
axiom L5 is dependent on the others. There are two main accessible completeness proofs
of L1–L4 (with MP and SU B): based on syntactic methods and linear inequalities by
Rose and Rosser [36], and purely algebraic—by Chang [4].
Chang’s proof is based on properties of MV algebras, algebraic counterparts of
the infinite-valued Łukasiewicz logic, defined and studied in [3].7 The key role in the
approach have additional binary connectives + and ·, which directly correspond to the
algebraic operations of MV algebras are defined by
α + β =df ¬α → β, and α · β =df ¬(α → ¬β).
Several axiomatizations for finite-valued Łukasiewicz logics (n > 3) were obtained by

way of extension of the axiom system L1–L4. Grigolia [10] employs multiplying use of
the connectives + and ·. Let kα will be a replacement of the formula α + α + . . . + α (k
times) and α k a replacement of the formula α · α · . . . · α (k times). Given a finite n > 3,
Grigolia’s axiom system for Łn consists of the schemes of L1–L4 and
Ln 5. nα → (n − 1)α
Ln 6. (n − 1)((¬α)j + (α · (j − 1)α)),
where 1 < j < n − 1 and j does not divide n − 1.
Tokarz [45] extension of L1–L4 is based on the characteristic functions of the set Ln in
[0, 1] and the properties of the consequence relation of Mℵ0 . The axiom set for a given n-
valued Łukasiewicz logic, including n = 2, results from L1–L4 by adding a single special
“disjunctive” axiom
p ∨ ¬p ∨ δn1 (p) ∨ . . . ∨ δnn−2 (p),
6 Łukasiewicz [23, 1970, p. 144]; no publication on the topic by Wajsberg exists.

7 MV algebras are presented in Sect. 4.2.
where, for any k, 1 ≤ k ≤ n−1, the algebraic operation δnk (x) corresponding to a formula
is the “characteristic function” of the logical value k/n−1 in the infinite Łukasiewicz matrix
Mℵ0 . v(δnk (x)) is 1 only when v(x) = k/n−1 , and is 0 otherwise. Such a formula is definable
in virtue of the McNaughton criterion in [25].
Another axiomatization of finite Łukasiewicz logics, offered by Tuziak [46], is for-
mulated in the standard propositional language with sequences of ascending implications
defined inductively by: p →0 q = q, p →k+1 q = p → (p →k q). The axiom set
for n-valued Łukasiewicz logic consists of ten formulas taken from the Hilbert-Bernays
axiomatization of CP C and two axioms:
T1. (p →n q) → (p →n−1 q)
T2. (p ≡ (p →s−2 ¬p)) →n−1 p for any 2 ≤ s ≤ n − 1 such that s is not a divisor of
n − 1.
3 Post Logics and Słupecki’s Functional Completeness

Programme
Post n-valued logical constructions in [32, 33] were mainly inspired by the well-known
formalization of the classical logic of Principia Mathematica [48] and by the truth-
table method of verification of logical laws. Contrary to Łukasiewicz logics all they
are functionally complete, i.e. they have the property that their any (finite-valued)
propositional connective can be represented or defined as a composition of their primitive
connectives.8
3.1 Post Matrices and Functional Completeness
The basic many-valued Post constructions are connected with two primitives: cyclic
negation (¬) and disjunction (∨). For any natural n ≥ 2 Post builds an n-valued logical
algebra on the linearly ordered set of objects
Pn = {t1 , t2 , . . . , tn }
(ti < tj iff i < j ) equipped with two operations: unary rotation ¬ (cyclic negation) and
binary disjunction ∨, defined in the following way:

t if i = n
¬ti = i+1 ti ∨ tj = tmax(i,j ).
t1 if i = n
8 The very property applies to the logic algebras i.e. matrices without designated set of elements. Nb. the
applications of Post construction are focused on algebras.
310 G. Malinowski
The disjunction function fixes on a natural and intuitive meaning of the disjunction
connective, typical for most known many-valued constructions. In plain terms, the logical
value of disjunctive proposition equals the greater of the values of its components. The
function of cyclic rotation permutes, in some specified manner, the set Pn and the negation
corresponding to it, the case n = 2 being excluded, is quite special: ¬t1 = t2 , ¬t2 =
t3 , . . . , ¬tn−1 = tn , ¬tn = t1 . It is just the fact of combining the latter with an appropriate
binary function of algebra on Pn that warrants the functional completeness of that algebra,
i.e.it ensures that by means of the primitive functions, every finite-argument function can
be defined on Pn , including constant functions and hence the objects t1 , t2 , . . . , tn . For a
given finite n ≥ 2 the algebraic structure:
Pn = ({t1 , t2 , . . . , tn }, ¬, ∨)
will be called an n-valued Post algebra.

The matrix Pn naturally associated with the algebra Pn :
Pn = ({t1 , t2 , . . . , tn }, ¬, ∨, {tn })
will, in what follows, be referred to as the (basic) n-valued Post matrix. It is easily seen
that the two-valued Post matrix is isomorphic to the negation–disjunction matrix for the
classical propositional calculus. To check it, one must replace t1 in P2 by the falsity (0)
and t2 by the symbol of truth (1). Simultaneously, however, the matrices Pn for n > 2
are totally incompatible with the mentioned classical matrix, which is the result of the
non-standard mode of the negation connective. Hence for n = 3, for instance, t3 could
be the only counterpart of ‘truth’ with respect to the adopted interpretation of disjunction
but then t1 would have to correspond to ‘falsity’ as ¬t3 = t1 , which should not take place
because ¬¬t3 = ¬t1 = t2 = t3 . A contradiction.
3.2 Słupecki Class of Functionally Complete Finite-Valued Logics
The pioneering Post construction of algebras of logic emerged scholar’s interest in

formulating handy criteria which might determine whether or not a given algebra (or,
logic) is functionally complete.
An early important result on the field a partial result on definability of one-argument
operations:
(Picard [31]). In Un all one-argument operations are definable whenever the functions
H, R, S are definable in it.9
One immediate application of this criterion led Słupecki to complete the three-element
Łukasiewicz algebra L3 = ({0, 1/2 , 1}, ¬, →, ∨, ∧, ≡) with the operations
determined by tables in Sect. 1.2 is an example of an incomplete algebra. Słupecki then
remarked that the one-argument constant function T : T x =1 /2 for any x = {0, 1/2 , 1}
9 See next page.

is not definable in terms of the basic operations. However, in virtue of Picard’s criterion,
adding T to the stock of functions of L3 leads to the (three-element) functionally complete
algebra (Słupecki [39]).
In the late 1930’s Słupecki undertook a serious investigation of the problem of
functional completeness. In the result, he provided one of the best appreciated criterion
for all finite logics [40], constructed the largest possible class of functionally complete
finite logics and gave a general method of their axiomatization, [41]:
(Słupecki [40]). An n-valued algebra Un (n ≥ 2, n finite) is functionally complete if

and only if in Un there are definable:
10 . All one-argument operations
20 . At least one two-argument operation f (x, y) whose range consists of all the values
i for 1 ≤ i ≤ n.
The Słupecki matrix Snk (1 ≤ k < n) with k-valued set of designated elements
{1, 2, . . . , k} is of the form
Snk = ({1, 2, . . . , n}, →, R, S, {1, 2, . . . , k}),
where → is a binary (implication), and R, S unary operations defined in the following

way:

y if 1 ≤ x ≤ k
x→y=
1 if k < x ≤ n

x + 1 if 1 ≤ x ≤ n − 1
R(x) =
1 if x = n
⎧
⎨ 2 if x = 1
S(x) = 1 if x = 2
⎩
x if 3 ≤ x ≤ n.
Functional completeness of each of these matrices results from the Picard’s criterion: R
and S are two necessary functions. To define the third, it suffices to put:

1 if x = 2
H x = (x → R(x → x)) → Sx for k = 1, then H x =
x if x = 2

1 if x = k
H x = R(x → x) → x for k > 1, then H x =
x if x = k.
Słupecki produced an effective proof of axiomatizability of T autSnk (any pair (n, k)

as above) giving a long list of axioms formulated in terms of implication and special
one-argument connectives defined through the superpositions of R, S, and H . The chief
line of approach here is to make capital of the character of implication, which can be
classically axiomatized using the detachment rule (MP). Słupecki extends MP onto the
whole language, taking the Łukasiewicz formula ((p → q) → r) → ((r → p)
312 G. Malinowski
→ (s → p)) as the only axiom for implication, and provides an inductive, combinatorial
proof of completeness.
The original (¬, ∨) systems of Post logic have not yet been axiomatized. However, the
problem of their axiomatizability has for years been taken for granted, so in [41] Słupecki
constructed the largest possible class of functionally complete finite logics and gave a
general method of axiomatization. From this it evidently follows that Post logics are also
axiomatizable although the problem of providing axioms for their original version still
remains open.
4 Algebraic Interpretations of Łukasiewicz Logics
The attempts to obtain algebras which would play the same role for Łukasiewicz calculi as
Boolean algebras do for the classical logic bore several constructions. The most important
of them are: Moisil and Chang algebras.
4.1 Moisil Algebras
Moisil’s constructions, dating back to the 1940s, are best adapted to the finite case;
see [28]. These are algebras with operations corresponding to ‘modal’ connectives
of Łukasiewicz logics10 ; more specifically, distributive lattices with Boolean-valued
endomorphisms.
A structure
(L, ∪, ∩, N, s1 , s2 , . . . , sn−1 , 0, 1)
is an n-valued Moisil algebra if the following conditions are satisfied:

M1. (L, ∪, ∩) is a distributive lattice with 0 and 1.
M2. N is an involution of L (i.e. NNx = x) also satisfying: N(x ∪ y) = N(x) ∩
N(y), N(x ∩ y) = N(x) ∪ N(y).
M3. The elements of {sk }1≤k<n are endomorphisms of (L, ∪, ∩): sk (x ∪ y) = sk (x) ∪
sk (y) and sk (x ∩ y) = sk (x) ∩ sk (y), such that
(i) sk (x) ≤ sk+1 (x)
(ii) sk (st (x)) = st (x)
(iii) sk (N(x)) = Nsn−k (x)
(iv) Nsk (x) ∪ sk (x) = 1, Nsk (x) ∩ sk (x) = 0
(v) If sk (x) = sk (y) for every k, then x = y.
The simplest example of n-valued Moisil algebra (given finite n) is the linearly ordered
n-element algebra constructed on the base of Łukasiewicz matrix Mn :
Mn = (Ln , ∪, ∩, N, s1 , s2 , . . . , sn−1 , 0, 1),
10 Compare 1.3 for n = 3.

where x ∪ y = x ∨ y, x ∩ y = x ∧ y, N(x) = ¬x and

0 when 1 ≤ k ≤ n−j −1
sk (j/n − 1) =
1 when n−j −1 < k ≤ n−1.
In virtue of McNaughton criterion [25] the operations sk are definable in Mn . The

formulae definining them by means of ¬ and → were given by Suchoń in [43].
It is interesting to notice, that s1 , s2 , . . . , sn−1 are the modal functors in n-valued
Łukasiewicz logic. Thus, for n = 3, s1 = L and s2 = M, compare Sect. 1.3.
Moisil called his structures Łukasiewicz algebras. In the course of time the term Moisil
algebras came to be used because, for n ≥ 5, Łukasiewicz implication was not definable
in n-valued structures of this kind. Despite that fault, however, Moisil algebras proved to
be the most fertile source of applications of Łukasiewicz logics.
4.2 Chang Algebras
The concept of MV algebra, tailored by Chang in [3], resorts to mathematical group

theory. Consider an algebra
A = (A, +, ·, −, 0, 1),
where + and · are binary operations, − is a unary operation and, finally, 0 and 1 are two
different constants in A, 0 = 1. Let us introduce, moreover, the following notations:
x ∪ y = x · y− + y x ∩ y = (x + y − ) · y.
The algebra A so defined is called an MV algebra if the following conditions are satisfied:
C1. x+y =y+x

C2. x + (y + z) = (x + y) + z C1∗. x·y =y·x
C3. x + x− = 1 C2∗. x · (y · z) = (x · y) · z
C4. x+1=1 C3∗. x · x− = 0
C5. x+0=x C4∗. x·0=0
C6. (x + y)− = x − · y − C5∗. x·1=x
C7. (x − )− = x C6∗. (x · y)− = x − + y −
C9. x∪y =y∪x C8. 0− = 1
C10. x ∪ (y ∪ z) = (x ∪ y) ∪ z C9∗. x∩y =y∩x
C11. x + (y ∩ z) = (x + y) ∩ (x + z) C10∗. x ∩ (y ∩ z) = (x ∩ y) ∩ z
C11∗. x · (y ∪ z) = (x · y) ∪ (x · z).
The simplest example of the MV algebra is an arbitrary (finite or not) Łukasiewicz

matrix—in this case x + y = ¬x → y, x · y = ¬(x → ¬y), and ∪, ∩, − are the
functions of disjunction, conjunction and negation, respectively.
314 G. Malinowski
4.3 MVn Algebras and Axiomatization of Ln
The problem of adapting MV algebras to finite cases (i.e. of introducing such modifi-
cations which would allow algebras characteristic for particular cases of Łukasiewicz
finite-valued logics to be obtained) was taken up in 1973 by Grigolia (see [10]) who
introduced the notion of MVn algebra. Adopting kα as the replacement of the formula
α + α + . . . + α (k times) and α k for the formula α · α · . . . · α (k times)), we say that an
MV algebra A is an MVn algebra provided that the following conditions are satisfied:
C12. (n − 1)x + x = (n − 1)x C12 ∗ . x n−1 · x = x n−1
and for n > 3:
C13. {(j x) · (x − + ((j − 1)x)− }n−1 = 0

C13 ∗ . (n − 1){x j + (x − · (x j −1 )− )} = 1
where 1 < j < n − 1 and j does not divide n − 1 .

The common property of the two algebraic interpretations of Łukasiewicz logics is
the way in which they are related to Boolean algebras. Every Moisil algebra as well as
each MV algebra contains some Boolean subalgebra. The set of Boolean elements of an
arbitrary n-valued Moisil algebra is equal to
{sk (x) : x ∈ L, 1 ≤ k < n},
and the set of Boolean elements of an MV algebra A to
{x ∈ A : x + x = x} (or, equivalently, to {x ∈ A : x · x = x}).
From among the purely algebraic results for Moisil and Chang algebras those of the
greatest importance for logic are the representation theorems which state that any algebra
is (isomorphic to) a subdirect product of a class of linear algebras, i.e.definitional variants
of Łukasiewicz matrices. Employing MV algebras, in [4] Chang gave a purely algebraic
proof of completeness of the ℵ0 -valued Łukasiewicz logic. Thus, L1–L4 together with MP
and SUB create the axiomatization of Łukasiewicz’s ℵ0 -valued propositional calculus.
Several axiomatizations of finite-valued Łukasiewicz logics (n > 3) were obtained
by way of extension of L1–L4 (MP and SUB being assumed) with the sets of ‘specific
axioms’ depending on n. Thus, for instance, Grigolia for the purpose of axiomatization
of n-valued Łukasiewicz logics uses in [10] two additional binary connectives + and ·
defined by α + β = ¬α → β, α · β = ¬(α → ¬β) (in what follows we write kα as the
replacement of the formula α + α + . . . + α (k times) and αk for the formula α · α · . . . · α
(k times)). Given a finite n > 3, Grigolia’s axiom system for Ln consists of the schemes
of L1–L4 and, additionally, the axioms
Ln 5. nα → (n − 1)α
Ln 6. (n − 1)((¬α)j + (α · (j − 1)α)),
where 1 < j < n − 1 and j does not divide n − 1. The completeness proof, received by
an adaptation of the Chang’s method, uses MVn algebras.
Tokarz in [45] worked out a method of axiomatization of finite Łukasiewicz logics
based on the characteristic functions of the set Ln (in [0,1]) and the properties of the
matrix consequence of Mℵ0 .
5 In Search for Interpretation of Łukasiewicz Logics
The source interpretation of the third logical value as ‘possibility’ or ‘indeterminacy’

was strongly criticized as soon as 1938 by Gonseth, who noticed (see [9]) that the
connectives in Łukasiewicz logic are incompatible with the suggested ways of interpreting
1/ . Accordingly, for two propositions α and ¬α if α is undetermined, so is ¬α, and then,
2
according to the table of conjunction, α ∧ ¬α is undetermined, which contradicts the
intuition since, independently of α’s content, α ∧ ¬α is false. A similar argumentation
applies to the disjunction: take the formula α ∨ ¬α and put 1/2 . We then get 1/2 as a result,
while intuitively α ∨ ¬α should be true. As it can be easily observed, the definitiveness
of the original interpretations of Łukasiewicz is caused by the fact that they neglect the
mutual dependence of some ‘possible’ propositions.
The history of attempts to give a satisfactory interpretation of Łukasiewicz logics
is fairly long. Below, we shall discuss some interesting undeterminacy and probability
approaches.
5.1 Three-Valued Logic as a Base for Non-Boolean Undetermined

Events
An interesting attempt to interpret Łukasiewicz logic intuitively was made by Słupecki

[42]. By an ‘interpretation’ Słupecki means pointing out some definite language which
describes the property of the determination of events in a three-valued manner reconciling
Łukasiewicz’s truth-tables.
The language considered by Słupecki is a set-theoretical union of a set S comprising
propositions about events and a set S ∗ of propositions that do not state events. Both sets
contain simple (atomic) propositions and compound ones formed by means of disjunction
(∨), conjunction (∧) and negation (¬) connectives. Since S ∗ does not play a significant
role in what follows, it is omitted. Słupecki supposes the set of events Z described by
propositions of S to be closed under the operations of union (∪), meet (or intersection)
(∩) and complementation (−) and, furthermore, the structure Z = (Z, ∪, ∩, −) to be a
Boolean algebra. There is a causality relation → on Z (‘f1 → f2 ’ means ‘the event f1 is
a cause of the event f2 ’) providing the assumption that
(P1) f → f1 ∪ f2 iff f → f1 or f → f2
(P2) f → f1 ∩ f2 iff f → f1 and f → f2
(P3) If f → f1 for some f , then f + → −f1 for no f +
(P4) If f1 → f , then f1 ∩ f2 → f
316 G. Malinowski
for any f, f1 , f2 ∈ Z. For the purpose of defining the property of determination, he then
singles out a set of past and present events hereafter denoted by symbols g, g1 , g2 , . . .,
and puts
D(f ) = df there is a g ∈ Z such that g → f,

D(f ) = df not D(f ) and not D(−f ).
The above expressions can be read as:
D(f ) − f is (at the present moment) determined,

D(f ) − f is not determined (at the present moment).
The relationship between expressions of the language S and events is established by

relation ∗ of describability (‘p ∗ f ’ means: (proposition) p describes (event) f ) satisfying
the following conditions
p ∨ p1 ∗ f ∪ f1 whenever p ∗ f and p1 ∗ f1
(∗) p ∧ p1 ∗ f ∩ f1 whenever p ∗ f and p1 ∗ f1
when p ∗ f, then ¬p ∗ −f
for any p, p1 ∈ S. Moreover, Słupecki assumes that every proposition of S refers to some
event of Z and he then defines the properties 1(p), 0(p) and 1/2 (p) : 1(p) = p is true,
0(p) = p is false, 1/2 (p) = p has the ‘third’ logical value, as follows:
if p ∗ f, then { 1(p) iff D(f ) }
(DT) if p ∗ f, then { 0(p) iff D(−f )}
if p ∗ f, then { 1/2 (p) iff D(f ) }.
Using (P1)–(P4) and (∗), it is easy to check that for x ∈ {0, 1/2 , 1}
x(p ∨ q) = x(p) ∨ x(q)

x(p ∧ q) = x(p) ∧ (q)
x(¬p) = ¬x(p),
where ∨, ∧ and ¬ appearing on the right-hand side are the connectives of the
Łukasiewicz three-valued logic. Therefore, what Łukasiewicz proposed, referring to the
property of determining propositions as a manner of interpretation of logical values, is,
by (DT), to some extent justified.
Słupecki’s interpretation, however, is not free of faults. First, it is partial, lacking the
implication connective. Admittedly, Słupecki extends it onto the language with modal
connectives M and L, and in this enriched language the implication of Łukasiewicz is
definable. However, it certainly does not change anything since the interpretation of the
implication obtained is fairly unintuitive. Secondly, more profound analysis of the whole
construction reveals that the assumption concerning Z has to be modified: Nowak in [30]
proved the formal correctness of the interpretation exclusively when Z is a de Morgan
lattice and not a Boolean algebra. This result, nevertheless, does not depreciate Słupecki’s
proposal but, on the contrary, makes it still more noteworthy; three-valued logic can
thus be interpreted as a set of propositions describing events which form a non-classical
algebra. If so, then (DT) implies that the third value of Łukasiewicz, 1/2 , is assigned to
propositions concerning non-Boolean, undetermined events.
5.2 Logical Probability
Mathematical probability is a measure of casual events and the probability calculus in

its simplest form resembles many-valued logic. Therefore, the question of a connection
between probability and many-valuedness emerges quite naturally. Łukasiewicz [19]
invented a theory of logical probability where undetermined propositions (i.e. proposi-
tional functions) are associated with a fraction which is proportional to the number of
variable values verifying the proposition and the number of all values of a given finite
domain. The differentiating feature of this comprehended probability in comparison to
mathematical probability is the fact that the former refers to propositions and not events.
The continuators of Łukasiewicz’s conception, Reichenbach and Zawirski among them,
exerted much effort to create a many-valued logic within which logical probability could
find a satisfactory interpretation (see e.g.[54, 55] and [34]).
The foundations of the Reichenbach–Zawirski conception are provided by the assump-
tion that there is a function P r ranging over the set of propositions of the considered
propositional language, with values from the interval [0,1]. The basic characteristics of
P r is embodied in the following rules:
(P1) 0 ≤ P r(p) ≤ 1
(P2) P r(p ∨ ¬p) = 1
(P3) P r(p ∨ q) = P r(p) + P r(q) if p and q are mutually exclusive (i.e. if p ∧ q is
an ‘impossible’ proposition or, equivalently, if P r(p ∧ q) = 0)
(P4) P r(p) = P r(q) when p and q are logically equivalent, i.e.if p∧¬q and ¬p∧q
are ‘impossible’.
From the above, other properties of P r known in the probability calculus can be inferred.
Among them are:
(1) P r(p ∧ ¬p) = 0
(2) P r(¬p) = 1 − P r(p)
(3) P r(p ∨ q) = P r(p) + P r(q) − P r(p ∧ q).
Therefore, if we identify the logical value w(p) of p with the measure of its probability
P r(p), w(p) = P r(p), then the function w should satisfy the conditions:
(¬) w(¬p) = 1 − P r(p)
(∧) w(p ∧ q) = P r(p) + P r(q) − P r(p ∨ q)
(∨) w(p ∨ q) = P r(p) + P r(q) when p and q are exclusive.
However, let us note that completion of (∨) through a direct reference to P r is impossible.
Whenever propositions p and q are not exclusive, w(p ∨ q) is a certain number from
[0,1] not less than P r(p), P r(q) and not greater than P r(p) + P r(q). This number
obviously depends on the ‘content’ of both propositions and not only on their logical
values. What follows is a practical inability to satisfy the extensionality principle in the
system whose semantics resorts to w valuations; values of some compound propositions
formed by means of the disjunction connective are not uniquely determined by the values
of the component propositions. As an illustration, consider the case when w(p) = 1/2 (p).
318 G. Malinowski
Then we would have

1
/2 (p) ∨ 1/2 (p) = P r(p ∨ ¬p) = 1 (according to (P2))
and
1
/2 (p) ∨ 1/2 (p) = P r(p ∨ p) = 2P r(p) − P r(p) = 1/2 (p). (due to (3))
Consequently, logical probability must not be identified with logical values in any
ordinary extensional many-valued logic.
It was, however, not until the 1970s that Giles worked out an operationalistic con-
ception of subjective probability interpreted unquestionably in denumerable Łukasiewicz
logic.
5.3 Operationalistic Conception of Subjective Probability
The researches of Giles directed towards finding a logic appropriate for the formalization
of physical theories, including those with undetermined propositions, resulted in a very
convincing philosophical interpretation of countable Łukasiewicz logic Lℵ0 combining
the conceptions of logical many-valuedness and subjective probability (see Giles [8]).
The main point of the approach consists in a dispersive physical interpretation of standard
logic language: it is assumed that each prime proposition in a physical theory is associated
through the rules of interpretation with a certain experimental procedure which ends in
one of two possible outcomes, ‘yes’ or ‘no’. The tangible meaning of a proposition of
the language is then related to the persons using it (i.e.speakers or observers). In the case
of prime propositions it is determined from the values of probability of success ascribed
by users in respective experiments, whereas in the case of compound propositions it is
determined from the rules of obligation incurred by whoever asserts them and formulated
in the nomenclature of dialogue logic (see Lorenz [16]).
The starting assumptions of the construction are stated in two principles:
(1) Everyone who asserts a prime proposition A commits themselves to paying a certain
sum of money, say $1, when the experiment associated with A results in ‘no’.
Secondly, to assert a proposition is not the same thing as to assert the same proposition
twice (in the same debate).
(2) Every speaker (i.e. language user) is able to ascribe to any prime A a real number
p(A) ∈ [0, 1], the so-called subjective probability, such that, given any ε < 0, they
are willing: (∗) to assert A in return for a payment of $(1 − p(A) + ε); (∗) to assert
¬A in return for a payment of $(p(A) + ε).
which yield that all prime propositions are probability definite for all speakers. The risk
value of A for a given person is < A > = 1 − p(A), denoting the expected loss if she
asserts it.
The meaning of compound propositions is appointed by the rules of debate of two
participants: a given person and their partner who can be a fate as well. The rules tend
to reduce all assertions of such propositions to the assertions of sequences (sets) of
prime propositions. Giles adopts the following rules (for any propositions p and q):
The rule concerning negation is less handy than the others but it can be ‘standardized’
Assertion Obligation (commitment)

p∨q Undertaking to assert either p or q at one’s own choice
p∧q Undertaking to assert either p or q at the opponent’s choice
p → q Agreement to assert q if the opponent will assert p
¬p Agreement to pay $1 to opponent if they will assert p.
under the assumption that language contains a constant F (a falsum) signifying any false
proposition; next define ¬p as p →F.
Every dialogue runs as follows: one of the participants asserts a (compound)
proposition, the other joins the game or not. Subsequently, the uttered propositions
that undergo no decomposition in a given step are repeated, which especially
concerns prime propositions. Every position of a dialogue game is of the form
W, Z being the names of the participants, while W1 , W2 , . . . , Wk and Z1 , Z2 ,
W : Z
W1 , W2 , . . . , Wk : Z 1 , Z 2 , . . . , Zm
. . . , Zm are sequences of propositions asserted by W and Z, or tenets of W and Z,

respectively. A position with prime propositions exclusively is final. Once it is reached,
the game counts as finished, appropriate experiments are carried out and the accounts are
settled. The two examples illustrate the course of a dialogue:
W Z W : Z
∅ A → (B → A) ∅ : A→ B ∨A
A B →A A : B∨A
A, B A
A B A A
In the first case it is W who takes the risk of paying $1 when the test proves falsity of
B. In the second case, two strategies are possible: one safe for both sides (A : A), the
other ending with A : B; according to the rules of dialogue the choice lies with Z and it
depends on Z’s evaluation of the risk values of A and B.
The risk value of a tenet (A1 , A2 , . . . , Am ) is the sum of risks of its elements
< A1 , A2 , . . . , Am > = < Aj >, and that of final position for Z is the difference:
< A1 , A2 , . . . , Am : B1 , B2 , . . . , Bn > = < Bi > − < Aj >.
From a fundamental result of game theory stating that every game with perfect
information has a ‘saddle point’, it follows (see Giles [8]) that each valuation of prime
propositions (i.e. assigning to them risk values or, equivalently, subjective probability
values) has a unique extension onto the whole language guaranteeing both participants
no increase in the risk value of the initial position—or to put it differently: that optimal
320 G. Malinowski
strategy exists. The extension in question is defined for any statements P and Q through
the schemes:
 = sup{0, < Q > − }

 = inf {, < Q >}
 = sup{, < Q >}
< ¬P > = 1− ;
sup and inf are the symbols of the operations of lower upper bound and greater lower
bound in [0,1], respectively.
The formulae of the considered language to which any valuation assigns non-positive
risk value are referred to as tautologies; there are propositions, the utterance of which
may lead only to (theoretically) not losing final positions. Using the equality pr(P ) =
inf {1, 1− } one can describe the property of being a tautology in terms of
subjective probability. A simple calculation shows that:
pr(P → Q) = inf {1, 1 − pr(P ) + pr(Q)}

pr(P ∨ Q) = sup{pr(P ), pr(Q)}
pr(P ∧ Q) = inf {pr(P ), pr(Q)}
pr(¬P ) = 1 − pr(P ).
Thus, pr is a valuation of Łukasiewicz logic Lℵ0 (compare Sect. 2). In consequence,

the set of tautologies of dialogue logic just constructed is equal to T autℵ0 : only those
formulae are tautologies whose probability amounts to 1 independently of the values
assigned to prime propositions as its components.
6 On Applications and Influence of Łukasiewicz Work
The philosophically well grounded and mathematically mastered Łukasiewicz work

brought to science important theoretical results and subscribed to the development of
important logical strategies, like the matrix approach. Independently of that, however,
some concrete, worthy applications of Łukasiewicz many-valued logics and algebras may
be indicated. The section contains a short presentation of some of them, like the use logical
matrices to demonstrate the independence of axioms, the way the Łukasiewicz infinite
logic grounds the fuzzy sets and truth functional fuzzy logic and, finally, the application
of Moisil’s algebras in switching theory.
6.1 Independence of Axioms
The logical method of testing axioms independence using algebras and matrices is
credited to Bernays and Łukasiewicz. To prove that an axiom system is independent one
singles out a property, mostly validity, which is common to all axioms besides of one
chosen and is inherited, via accepted rules of inference, by all theorems of the systems.
The procedure being repeated as many times as there is the number of axioms in the
system.
Consider (¬, →)–system of the classical propositional calculus originating with
Łukasiewicz. Its axioms are:
(A1) (¬p → p) → p
(A2) p → (¬p → q)
(A3) (p → q) → ((q → r) → (p → r)),
and the rules MP and SUB. Let now
M(A1) = ({0, 1}, ¬1 , →, {1}) , M(A2) = ({0, 1}, ¬2 , →, {1})
be matrices wherein the implication connective → is determined classically (by the well-
known truth table), ¬1 (0) = ¬1 (1) = 0 and ¬2 (0) = ¬2 (1) = 1. Moreover let
M(A3) = ({0, 1/2 , 1}, ¬, →, {1})
be a matrix with the connectives defined by the tables:
α ¬α → 0 1/
2 1
0 1 0 1 1 1
1/ 1/ 1/ 1 0 1
2 2 2
1 1 1 0 0 1
It is readily seen that the sets of Tautologies of the three matrices, T aut(A1) , T aut(A2) and
T aut(A3) , are closed under (MP) and (SUB). Since
(1) A1 ∈ T aut(A1) , A2 ∈ T aut(A1) and A3 ∈ T aut(A1) ,
(2) A1 ∈ T aut(A2) , A2 ∈ T aut(A2) and A3 ∈ T aut(A2) ,
(3) A1 ∈ T aut(A3) , A2 ∈ T aut(A3) and A3 ∈ T aut(A3) .
the axiom system (A1)–(A3) is independent.
The application of the method described is not limited to the logical calculi. The proofs
of independence in set theory through the use of matrices built on the base of Boolean
algebras were presented by Scott and Solovay, [38].
322 G. Malinowski
6.2 Fuzzy Sets and Truth Functional Fuzzy Logic
Everyday reasonings operate on imprecise concepts and are supported by approximate

inferences. That makes the possibility of applying the apparatus of the classical logic to
formalize them very limited. Among special tools extending the formalization power of
the standard approach are fuzzy sets and fuzzy logics. Below, we outline fuzzy concepts
since, due to their properties, the infinite-valued Łukasiewicz logics are the most often
used as their bases.
Fuzzy Sets Zadeh in [53] defines a fuzzy set A of a given domain U as an abstract object
characterized by generalized characteristic function UA in the real set [0,1]: UA : U →
[0, 1]. The values of UA are interpreted as degrees of membership of elements of U to
the fuzzy set A. The extreme values of this function, 0 and 1, denote respectively: not
belonging to A and entire membership to A. Limiting the scope of UA to {0, 1} results in
an “ordinary” characteristic function and in this each “classical” set is a special case of a
fuzzy set.
Fuzzy sets are an instrument of modelling inexact predicates appearing in natural
languages. Thus, for example the property of “being much greater than 1” defined on
the set of real positives R + can be assigned to a fuzzy set W with a non-decreasing
characteristic function R + → [0, 1] which meets the conditions like: RW (0) =
0, RW (1) = 0, RW (5) = 0.01, RW (100) = 0.95, RW (500) = 1, etc. Certainly, in
the above example only values RW (0), RW (1) unquestionable and the selection of other
values is somehow arbitrary.
In the family F (U ) of fuzzy (sub)sets of a given domain the relation of inclusion
reflexes the order between the reals:
A ⊆ B if and only if UA (x) ≤ UB (x) for any a ∈ U,
and the counterparts of the operations of complement (−), union (∪) and intersection (∩)
are set by:
U−A (x) = 1 − UA (x)

UA∪B (x) = max{UA (x), UB (x)}
UA∩B (x) = min{UA (x), UB (x)}.
Bellman and Giertz [1] showed that UA∪B and UA∩B are the unique non-decreasing
continuous functions warranting both the compatibility of the construction with the
standard algebra of sets and the fact that (F (U ), ∪, ∩, −) is a de Morgan lattice if and only
if U−A is defined as above. It is in order to notice that, in spite of the naturalness of the
proposal, several studies admit as (more) helpful, fuzzy sets algebras defined otherwise.
The values of generalized characteristic functions may be identified with logical values
of propositions of the form “x ∈ X”, where ∈ is a “generalized” set-theoretical predicate.
Subsequently, using logical constants of a base logic one may set the inclusion and the
operations of fuzzy set algebra as:
A ⊆ B =df ∀x(x ∈ A → x ∈ B)
− A = {x : ¬(x ∈ A)}
A ∪ B = {x : x ∈ A ∨ x ∈ B}
A ∩ B = {x : x ∈ A ∧ x ∈ B}.
For Zadeh’s algebra the choice of a base logic is to great extent prejudiced: it (this logic)
must be based on ℵ1 -element matrix, wherein negation is expressed by the function 1 −p,
disjunction and conjunction, respectively by: max{p, q} and min{p, q}, and the universal
quantifier as the greatest lower bound (inf ). The function of implication is not uniquely
determined. However, evidently it should meet the requirement:
(.) If p → q = 1, then p ≤ q.
Though Łukasiewicz’s implication and, consequently, his ℵ1 -valued logic, have been
the most intensely applied, similar connectives of other logics have been also taken
into account. The commonly shared belief among scholars working on fuzziness, both
theoreticians and practicians, is that only a concrete application of fuzzy sets algebra can
decide about the choice of the base logic (see Gaines [7]).
Truth Functional Fuzzy Logics Algebraic and metamathematical studies of the infinite-
valed Łukasiewicz logic, see Cignoli et al. [5], are among the most important issues of
recent investigations. Somewhat related to these studies are activities delineating a class of
propositional logics, called by Hájek [12] fuzzy logics (in a narrow) sense. The influence
of fuzzy set theory initiated the study of a class of systems of many-valued logics, whose
semantics is based on the real interval [0,1]. Several comparisons between the systems
serving as a base for particular constructions directed the scholar’s attention to, possibly
idempotent, strong conjunctions connectives, whose corresponding truth functions were
associative, commutative, non-decreasing and have 1 as its neutral (unit) element. Such
functions were called t-norms. Accordingly, a binary function ∗ on [0,1] is a t-norm
(triangular norm) if for any x, y, z ∈ [0,1]
x ∗ (y ∗ z) = (x ∗ y) ∗ z
x∗y =y∗x
if x ≤ y, then x ∗ z ≤ y ∗ z
x∗1=x
Connectives corresponding to t-norms are conjunctions. Further to this one may also
define t-conorms which serve as truth functions of disjunctions and, possibly, relate the
two functions using appropriate negation. In Sect. 2 we have examples of both, the t-norm
( min(x, y) i.e. Łukasiewicz conjunction) and t-conorm (max(x, y) i.e. Łukasiewicz
disjunction).
Hájek [12] is the main study of logics defined by continuous t-norms (a t-norm is
continuous is considered in the mathematical terms is continuous as a mapping). Among
324 G. Malinowski
the important continuous t-norms are the following:

Łukasiewicz t-norm x ∗ y = max(0, x + y − 1),
Gödel t-norm x ∗ y = min(x, y)
product t-norm x ∗ y = x · y ;
these functions have been used in numerous applications of fuzzy set theory as well as
fuzzy logics.
The connectives defined through the continuous t-norm conjunctions (continuity with
respect to the left argument is sufficient) are special since there is algebraically nice
procedure relating them with implications, having good metalogical properties. Any such
implication → is defined as residuum of a given continuous t-norm ∗, i.e.
x → y = max(z : x ∗ z ≤ y).
Hájek’s basic fuzzy propositional logic, BL-logic, is the logic of continuous t-norms
on [0,1]. The language of BL has the connectives of conjunction &, implication →
and the constant ⊥ of falsity. The semantics of BL is established by the function of t-
norm, all other functions corresponding to the connectives are derived. A formula is a BL
tautology if and only if under each valuation of propositional variables compatible with
the functions of connectives it takes the value 1. Hájek’s axiom system adequate for BL
logic is the following:
(H1) (α → β)→ ((β → γ ) → (α → γ )
(H2) (α&β) → α)
(H3) (α &β) →(β & α)
(H4) (α &(α → β))→(β & (β → α)
(H5a) (α →(β → γ )) → ((α&β)→ γ )
(H5b) ((α&β)→ γ )→ (α →(β → γ ))
(H6) (α →(β → γ ))→ (((β → α)→ γ )→ γ )
(H7) ⊥→α
Any system of propositional logic determined by a t-norm in the way indicated may be
received as a strengthening of BL. For instance, Łukasiewicz logic results from BL by
addition the single one axiom schema:
(Ł) ¬¬α → α
BL extends to the basic fuzzy predicate logic in a standard way. Hájek [12] shows
interesting features of t-norm based predicate calculi, see also Hájek et al. [13], Montagna
[29].
6.3 Moisil Algebras and Switching Theory
Soon after the successful applications of the classical logic, Boolean algebras and other
algebraic structures (e.g. groups) in switching theory, in the 1950s, the scholars centered
the interests on the possibility of the use of many-valued logic algebras for similar
purposes (see e.g. Epstein et al. [6]). These interests brought about the birth of several
Fig. 1 Two oppositely x1

oriented contacts
x2
Fig. 2 The “real” x1 x2

switch-function •((◦ (1) •◦
•◦ (1/2 ) •◦
•◦ (0) •((◦
techniques of the analysis and synthesis of electronic circuits and relays based (mainly)
on Moisil and Post algebras (see Rine [35]). Below, we confine ourselves to some
remarks justifying the purposefulness of using many-valued algebras in switching and
relay circuits theory. The most elementary composite of the traditional electronic circuit
is a mechanical contact opening and closing some fragment of an electrical network. The
switch over of contacts is affected mechanically or electromechanically (i.e. using relays).
And, among the contacts of a given network one may find such pairs of contacts which
according to the technical assumptions have to change their positions into complementary
simultaneously. The simplest example of such situation is the gear of two oppositely
oriented contacts x1 and x2 positioned in parallel branches of a circuit (see Fig. 1): x1 is
normally closed while x2 normally open. When considering the ideal model of the circuit
one assumes that both contacts react momentarily to an affection and thus stroking, as
shown at Fig. 2, from the state (1) to (0).
Practically, however, it may happen that x1 will open still before x2 would be closed
and, consequently, contrary to the technical presumptions the gear during a time moment
will be open. That is just the reason for such a modelling in which the third state (1/2 )
(see Fig. 2) is considered; the table beforehand characterizes the “real” switch-function
as a function of states and contact (1 inside the table denotes normal contact’s state and
0 its denial). On the other hand, one also may read table treating x1 and x2 as (one-
argument) functions of states i.e. of z, and their values as states as well putting: x1 =
s1 (z), x2 = s2 (z) .
z x1 x2
1 1 1
1/ 0 1
2
0 0 0
Let us notice that then s1 and s2 are Moisil’s operations on {0, 1/2 , 1} . Subsequently,
to describe any network built of the contacts x1 , x2 and their complements x1 , x2
one should define binary operations ∪ and ∩ corresponding to two possible types of
connections and unary operation N such that xi = Nxi (i = 1, 2) and that NNz = z
326 G. Malinowski
for z ∈ {0, 1/2 , 1} . It appears that the most accurate ways of introducing these operations
leads to the Moisil algebra ({0, 1/2 , 1}, ∪, ∩, N, s1 , s2 ).
A generalization of the outlined construction onto the case of any number of contacts
similarly results in n-valued algebras. The algebraic treatment of switching devices aims
at providing several techniques of the analysis, the synthesis and the minimalization
of multiplex networks. The most important advantage of the many-valued approach is
the possibility of elimination of possible switching disturbance through the algebraic
synthesis of the networks, see e.g. Moisil [27].
Application of many-valued algebras is not limited to binary contacts. Investigations
concerning multi-stable contacts and switches have also been undertaken. However,
according to difficulties with technical realizations of devices working on voltage-mode
and the progress of technology of binary highly integrated circuits these activities are not
very common. Still, however, many-valued constructions attract attention of engineers.
Thus, for instance, multiple values may be useful for describing transistors. Hayes [14],
Hähnle and Kernig [11] give an example of such modelling MOS transistors. Due to a
degradation of signals, a MOS transistor has different signal levels at source and terminals.
Thus, a natural modelling leads to a many-valued logic, the values of which are organized
in form of a lattice.
References
1. Bellman, R.E., Giertz, M.: On the analytic formalism of the theory of fuzzy sets. Inf. Sci. 5, 149–156
(1973)
2. Chang, C.C.: Proof of an axiom of Łukasiewicz. Trans. Am. Math. Soc. 87, 55–56 (1958)
3. Chang, C.C.: Algebraic analysis of many-valued logics. Trans. Am. Math. Soc. 88, 467–490 (1958)
4. Chang, C.C.: A new proof of the completeness of the Łukasiewicz axioms. Trans. Am. Math. Soc.
93, 74–80 (1959)
5. Cignoli, R.L.O., D’Ottaviano, I.M.L., Mundici, D.: Foundations of Many-Valued Reasoning. Trends
in Logic: Studia Logica Library, vol. 7. Kluwer Academic Publishers, Dordrecht (1999)
6. Epstein, G., Frieder, G., Rine, D.C.: The development of multiple-valued logic as related to Computer
Science. Computer 7(9), 20–32 (1974)
7. Gaines, B.R.: Foundations of fuzzy reasoning. Int. J. Man Mach. Stud. 8, 623–668 (1976)
8. Giles, R.: A non-classical logic for physics. Stud. Logica 33, 397–416 (1974)
9. Gonseth, F. (ed.): Les entretiens de Zurich sur les fondements et la méthode des sciences mathéma-
tiques 6–9 décembre 1938. Zurich (1941)
10. Grigolia, R.: Algebraic analysis of Łukasiewicz–Tarski’s n-valued logical systems. In: Wójcicki, R.,
Malinowski, G. (eds.) Selected papers on Łukasiewicz sentential calculi, pp. 81–92. Ossolineum,
Wrocław (1977)
11. Hähnle, R., Kernig, W.: Verification of switch level designs with many-valued logic. In: Voronkov,
A. (ed.) Proceedings LPAR’93, St. Petersburg, Russia. Lecture Notes on Computer Science, vol. 698,
pp. 158–169. Springer, Berlin (1993)
12. Hájek, P.: Metamathematics of Fuzzy Logics. Trends in Logic: Studia Logica Library, vol. 4. Kluwer
Academic Publishers, Dordrecht (1998)
13. Hájek, P., Paris J., Sheperdson, J.: The liar paradox and fuzzy logic. J. Symb. Log. 65, 339–346 (2000)
14. Hayes, J.P.: Pseudo-Boolean logic circuits. IEEE Trans. Comput. C-35(7), 602–612 (1986)
15. Kotarbiński, T.: Zagadnienie istnienia przyszłości [The problem of existence of the future]. Przeglad
˛
Filozoficzny VI.1 (1913)
16. Lorenz, K.: Arithmetik und Logik als Spiele. Kiel (1961)
17. Łukasiewicz, J.: Analiza i konstrukcja poj˛ecia przyczyny. Przeglad ˛ Filozoficzny 9, 105–179 (1906)
18. Łukasiewicz, J.: O zasadzie sprzeczności u Arystotelesa. Studium krytyczne. Kraków (1910) (English
tr.: On the principle of contradiction in Aristotle. Rev. Metaphys. XXIV, (1971))
19. Łukasiewicz, J.: Die logischen Grundlagen der Wahrscheinlichkeitsrechnung. Kraków (1913) (En-
glish tr.: Logical foundations of probability theory. In: Borkowski, L. (ed.) Selected Works, pp. 16–63.
North-Holland, Amsterdam (1970)
20. Łukasiewicz, J.: O logice trójwartościowej. Ruch Filozoficzny 5, 170–171 (1920) (English tr.: On
three-valued logic. In: Borkowski, L. (ed.) Selected works, pp. 87–88. North-Holland, Amsterdam
(1970))
21. Łukasiewicz, J.: Philosophische Bemerkungen zu mehrwertigen Systemen des Aussagenkalküls.
Comptes rendus des séances de la Société des Sciences et des Lettres de Varsovie Cl. III 23, 51–
77 (1930) (English tr.: Philosophical remarks on many-valued systems of propositional logic. In:
McCall, S. (ed.) Polish Logic 1920–1939, pp. 40–65. Clarendon Press, Oxford (1967))
22. Łukasiewicz, J.: A system of modal logic. J. Comput. Syst. 1, 111–149 (1953)
23. Łukasiewicz, J.: Z zagadnień logiki i filozofii. Pisma wybrane. PWN, Warszawa (1961) (English tr.:
Borkowski, L. (ed.): Selected Works. North-Holland, Amsterdam (1970))
24. Łukasiewicz, J., Tarski, A.: Untersuchungen über den Aussagenkalkül. Comptes rendus des séances
de la Société des Sciences et des Lettres de Varsovie Cl. III 23, 30–50 (1930)
25. McNaughton, R.: A theorem about infinite-valued sentential logic. J. Symb. Log. 16, 1–13 (1951)
26. Meredith, C.A.: The dependence of an axiom of Łukasiewicz. Trans. Am. Math. Soc. 87, 54 (1958)
27. Moisil, G.: Zastosowanie algebr Łukasiewicza do teorii układów przekaźnikowo-stykowych [Appli-
cation of Łukasiewicz algebras to the study of relay-contact networks], vol. II, 1967 edn. Ossolineum,
Wrocław (1958)
28. Moisil, G.: Essais sur les logiques non-chrisipiennes. Editions de l’Académie de la Republique
Socialiste de Roumanie, Bucharest (1972)
29. Montagna, F.: An algebraic approach to propositional fuzzy logic. J. Log. Lang. Inf. 9(1), 91–124
(2000). Special issue on many-valued Logics of Uncertainty. Mundici, D. (ed.)
30. Nowak, M.: O możliwości interpretowania trójwartościowej logiki Łukasiewicza metoda˛ Słupeckiego
[On the possibility of interpreting the three-valued Łukasiewicz logic using Słupecki’s method]. Acta
Universitatis Lodziensis, Folia Philosophica 5, 3–13 (1988)
31. Picard, S.: Sur les fonctions définies dans les ensembles finis quelconques. Fundam. Math. 24, 198–
302 (1935)
32. Post, E.L.: Introduction to a general theory of elementary propositions. Bull. Am. Math. Soc. 26, 437
(1920)
33. Post, E.L.: Introduction to a general theory of elementary propositions. Am. J. Math. 43, 163–185
(1921)
34. Reichenbach, H.: Wahrscheinlichkeitslehre. Leiden (1935) (English tr.: The Theory of Probability.
University of California Press, Berkeley (1949))
35. Rine, D.C. (ed.): Computer Science and Multiple-Valued Logic. Theory and Applications. North-
Holland, Amsterdam (1977)
36. Rose, A., Rosser, J.B.: Fragments of many-valued statement calculi. Trans. Am. Math. Soc. 87, 1–53
(1958)
37. Rosser, J.B.,Turquette, A.R.: Many-Valued Logics. North-Holland, Amsterdam (1952)
38. Scott, D., Solovay, R.: Boolean valued models for set theory. In: Proceedings of the American
Mathematical Society Summer Institute on Axiomatic Set Theory 1967. University of California,
Los Angeles. Proceedings of Symposia in Pure Mathematics, vol. 13 (1969)
39. Słupecki, J.: Der volle dreiwertige Aussagenkalkül. Comptes rendus des séances de la Société des
Sciences et des Lettres de Varsovie Cl. III 29, 9–11 (1936) (English tr.: The full three-valued
propositional calculus. In: McCall, S. (ed.) Polish Logic 1920–1939, pp. 335–337. Clarendon Press,
Oxford (1967))
40. Słupecki, J.: Kryterium pełności wielowartościowych systemów logiki zdań [A criterion of complete-
ness of many-valued systems of propositional logic]. Comptes rendus des séances de la Société des
Sciences et des Lettres de Varsovie Cl. III 32, 102–109 (1939)
41. Słupecki, J.: Dowód aksjomatyzowalności pełnych systemów wielowartościowych rachunku zdań
[Proof of the axiomatizability of full many-valued systems of propositional calculus]. Comptes rendus
des séances de la Société des Sciences et des Lettres de Varsovie Cl. III 32, 110–128 (1939)
328 G. Malinowski
42. Słupecki, J.: Próba intuicyjnej interpretacji logiki trójwartościowej Łukasiewicza [An attempt of
intuitionistic interpretation of three-valued Łukasiewicz logic]. In: Rozprawy Logiczne. PWN,
Warszawa (1964)
43. Suchoń, W.: Définition des founcteurs modaux de Moisil dans le calcul n-valent des propositions de
Łukasiewicz avec implication et négation. Rep. Math. Log. 2, 43–47 (1974)
44. Tarski, A.: O poj˛eciu wynikania logicznego [On the concept of logical consequence]. Przeglad ˛
Filozoficzny 39, 58–68 (1936) (English tr. in: Tarski, A.: Logic, Semantics, Metamathematics: Papers
from 1923 to 1938, pp. 409–420. Clarendon Press, Oxford (1956). Translated by J.H. Woodger)
45. Tokarz, M.: A method of axiomatization of Łukasiewicz logics. Bull. Sect. Log. 3(2), 21–24 (1974)
46. Tuziak, R.: An axiomatization of the finitely-valued Łukasiewicz calculus. Stud. Logica 48, 49–56
(1988)
47. Wajsberg, M.: Aksjomatyzacja trójwartościowego rachunku zdań. Comptes Rendus de la Société
des Sciences et des Lettres de Varsovie Cl. III 24, 126–148 (1931) (English tr.: Axiomatization of
the three-valued propositional calculus. In: McCall, S. (ed.) Polish Logic 1920–1939, pp. 264–284.
Clarendon Press, Oxford (1967))
48. Whitehead, A.N., Russell, B.: Principia Mathematica, vol. I. Cambridge University Press, Cambridge
(1910)
49. Woleński, J.: Filozoficzna Szkoła Lwowsko–Warszawska. PWN, Warszawa (1985) (English tr.: Logic
and philosophy in the Lvov–Warsaw School. Synthese Library, vol. 198. D. Reidel, Dordrecht (1989))
50. Wójcicki, R.: Theory of Logical Calculi. Basic Theory of Consequence Operations. Synthese Library,
vol. 199. Kluwer Academic Publishers, Dordrecht (1988)
51. Wójcicki, R.: On matrix representations of the consequence operations of Łukasie-wicz’s sentential
calculi. In: Wójcicki, R. (ed.) Theory of Logical Calculi. Basic Theory of Consequence Operations.
Synthese Library, vol. 199, pp. 101–111. Kluwer Academic Publishers, Dordrecht (1977/1988)
52. Wójcicki, R., Malinowski, G.: Selected papers on Łukasiewicz sentential calculi. Ossolineum,
Wrocław (1977)
53. Zadeh, L.A.: Fuzzy sets. Inf. Control. 8, 338–353 (1965)
54. Zawirski, Z.: Znaczenie logiki wielowartościowej i zwiazek˛ jej z rachunkiem prawdopodobieństwa
[Significance of many-valued logic for cognition and its connection with the calculus of probability].
Przeglad˛ Filozoficzny 37, 393–398 (1934)
55. Zawirski, Z.: Stosunek logiki wielowartościowej do rachunku prawdopodobieństwa [Relation of
many-valued logic to the calculus of probability]. Prace Komisji Filozoficznej Polskiego Towarzystwa
Przyjaciół Nauk 4, 155–240 (1934)
G. Malinowski ()
University of Łódź, Łódź, Poland
Tomorrow’s Sea-Battle and the Beginning
of Temporal Logic
Kazimierz Trz˛esicki
Abstract Jan Łukasiewicz’s consideration of the deterministic consequences of the law

of the excluded middle and the principle of causality gave the incentive to the development
of temporal logic. Formal reformulation of arguments in favour of determinism is possible
in the language of temporal logic. Following Jan Łukasiewicz both the arguments, the
argument from the law of the excluded middle and the argument from the principle of
causality, will be discussed concomitantly.
Keywords Tomorrow’s sea-battle · Temporal logic
Mathematics Subject Classification (2000) Primary 03B44; Secondary 03 A10
1 Introduction
Aristotle (384 BC–322 BC) in his famous tomorrow’s sea-battle passage of De Inter-
pretatione 9 19a 30 stated the problem of the logical value of statements about future
contingencies. The problem has been discussed since the age of Aristotle as an important
philosophical question in connection with determinism. There are many solutions to it
[1]. Diodorus Cronus from the Megarian school of philosophy is famous as the author
of a version of the problem in his notorious Master Argument. In this subject some
achievements are due to Avicenna (980–1037). His work influenced medieval logicians
Albertus Magnus (1193/1206–1280) and William of Ockham (c. 1288–c. 1348). In the
nineteenth century Charles Sanders Peirce (1839–1914) wrote that he did not share the
common opinion that time is an extralogical matter. The deterministic consequence of the
law of the excluded-middle (LEM) and the principle of causality (PC) were considered
by Jan Łukasiewicz (1878–1956). It was, as writes Słupecki: [8, Introduction, p. vii]
. . . the problem in which Łukasiewicz was most interested almost all his life and which he strove to
solve with extraordinary effort and passion was the problem of determinism. It inspired him with
the most brilliant idea, that of many-valued logics.
To avoid fatalism, the deterministic (PRE-DET) consequence of LEM, he proposed

abolishing the logical principle of bivalence (PB—every proposition is either true or

330 K. Trz˛esicki
false)—as Łukasiewicz called it [5]—and to introduce a new logical value [6, 7, 9].1
He laid the ground for the historically first elaboration of three-valued logic [4]. The
sentences about future contingencies would be neither true nor false; neither 1 nor 0.
The logical value of such sentences would be intermediate. The inventor of temporal
logic, Arthur Norman Prior (1914–1969), was concerned with the philosophical matters
of free will and predestination. He learned logic from Polish logicians, in particular
Jan Łukasiewicz (even using his parenthesis-free notation); in his early tense-logical
considerations he referred to many-valued logic of Łukasiewicz [11–18]. Many-Valued
approaches to temporal logic were also discussed by Rescher and Urquhart [19, pp. 213–
237]. Though attempts at formalizing future contingents using many-valued logics were
not successful, the idea of many valuedness in temporal logic is still fruitful in particular
in AI [1, Łukasiewicz’s contribution to temporal logic, 149–154] and model checking [3].
In Łukasiewicz’s opinion [6, p. 126]:
A trivalent system of logic . . . differs from ordinary bivalent logic, that only one known so far, as
much as non-Euclidean systems of geometry differ from Euclidean geometry.
For him [10, pp. 84–85] it contributes to the

struggle for the liberation of the human spirit [from the] logical coercion given by Aristotle’s
science as a system of principles and theorems connected by logical relationships. [. . . ] The
creative mind revolts against this concept of science [. . . ] A system of three-valued logic [. . . ]
destroys the former concept of science, based on necessity [. . . ] Logic is a free product of man,
like a work of art.
The thought that many-valuedness is connected with freedom is still vivid. For
Karpenko [2, pp. 143–144]
. . . Łukasiewicz finite-valued logic, are not restrictions (as they are usually viewed), but in a sense,
extensions of C2 [the classical propositional logic]. The repercussions of such an extension are
quite serious as well rather surprising. The extension of the very basic logical universe resulted in
the logics of continual nature; in the possibility to characterize, structure, and describe classes of
prime numbers. Are all of these required for the logical reasoning? On the other hand, the problem
of fatalism and free will is also of continual nature, which usually goes unnoticed. As we wrote in
this book for the refutation of the doctrine of logical fatalism, Łukasiewicz, without being aware
of it, abandoned discreteness for continuity.
Referring to his book Karpenko even ask [2, p. 4]:

Is there, however, any link between the doctrine of logical fatalism and prime numbers?
In contemporary temporal logic the discussion of PRE-DET is usually limited to

logical PRE-DET. Łukasiewicz also considered the argument from PC, i.e. the principle
that any temporal fact has its cause in another fact that occurs at an earlier moment.
Usually the argument from PC in favour of PRE-DET is assumed as evident. Łukasiewicz
questioned the argument and showed that both the arguments from LEM and PC are
1 The article in question is a revised version of the address that Łukasiewicz delivered as a Rector during
the inauguration of the academic year 1922–1923 of Warsaw University. Later on Łukasiewicz revised
the address giving it the form of an article, without changing the essential claims and arguments. It was
published for the first time in Polish in 1946 and also [6, pp. 114–126]. An English version of the paper is
available in: [7, pp. 19–39]. It is worth noting that when the author delivered the talk initially, the theories
and discoveries in the field of atomic physics that undermined determinism were still unknown.
Tomorrow’s Sea-Battle and the Beginning of Temporal Logic 331
independent of one another. Moreover, Łuksiewicz questioned what in temporal logic

discussions is treated almost as dogma, that the past is completely determined (POST-
DET): We should not treat the past differently from the future [8, p. 127]. He rejects
the Latin saying “facta infecta fieri non possunt” that is, what once has happened cannot
become not happened. In order to analyze POST-DET, the principle of effectivity (PE), as
a principle symmetrical to PC, should be considered.
2 Language and Semantics
To use a formal logic to solve a philosophical problem, we have to have:

1. a formal language in that the problem can be expressed in an intuitively satisfactory
way,
2. the logic should be neutral with respect to this problem, i.e. the formula that expresses
it should not be a thesis of the logic (an analytical truth of the language).
The thesis of determinism as consisting of two theses, the thesis of pre-determinism PRE-
DET and the thesis of post-determinism POST-DET, can be formulated as follows:
DET. If ϕ, then
• PRE-DET. at any earlier moment it was true that there would be ϕ
and
• POST-DET. at any later moment it will be true that there was ϕ
The principle of causality says:
• PC. If ϕ occurs at t, then at t1 , some moment earlier than t, and at any moment between
t and t1 it was true that there will be ϕ.
The principle of effectivity, as symmetrical to PC, may be formulated as follows:
• PE. If ϕ occurs at t, then at t1 , some moment later than t, and at any moment between
t and t1 it will be true that there was ϕ.
The relation of causality is transitive [9, p. 28]:
This means that for any facts φ, ψ, and χ, if φ is the cause of ψ and ψ is the cause of χ, then φ is
the cause of χ.
Both the theses of pre- and post-determinism, PRE-DET and POST-DET, and both the
principles of causality and effectivity, PC and PE, can be expressed in the language of
temporal logic.
Let the language consist of:
1. p1 , p2 , . . . —propositional letters,
2. a functionally complete set of classical propositional connectives,
3. temporal operators (past tense and future tense operators).
Let AP (Atomic Propositions) be the set of propositional letters. Formulas are defined in
the usual way and will be denoted by Greek letters: ϕ, ψ, . . . , if necessary with indices.
332 K. Trz˛esicki
Let time be T = T , <!, where T is a non-empty set (of moments) and < is a binary
(earlier-later) relation on T . No conditions on < are imposed.
The temporal world W consists of time T , <! and facts that occur at elements of T .
Let V be a function, valuation, that to each t (∈ T ) assigns a subset of AP , the set of
propositional letters that are true at this point, V : T → 2AP .
Let temporal operators be defined in the usual way, i.e. as follows:
Definition 2.1 (G) T , <, V !, t | Gϕ iff for any t1 , t < t1 : T , <, V !, t1 | ϕ.
Definition 2.2 (F ) T , <, V !, t | F ϕ iff there is t1 , t < t1 : T , <, V !, t1 | ϕ.
Definition 2.3 (H ) T , <, V !, t | H ϕ iff for any t1 , t1 < t : T , <, V !, t1 | ϕ.
Definition 2.4 (P ) T , <, V !, t | P ϕ iff there is t1 , t1 < t : T , <, V !, t1 | ϕ.
Operators G, H are dual to, respectively, F and P :
• Gϕ ↔ ¬F ¬ϕ,
• H ϕ ↔ ¬P ¬ϕ.
Let us call the language Priorean. Let Kt be the set of all formulas that are satisfied in any
model of whatever are the set of moments of time T , the relation < and the valuation V ,
ϕ ∈ Kt iff for any W : W | ϕ.
The operators are usually read as:
F — it will at some time be the case that . . .
P — it has at some time been the case that . . .
H — it has always been the case that . . .
G — it will always be the case that . . .
Formulas of the language are recursively defined by:
ϕ ::= p|¬ϕ|ϕ ∨ ϕ|ϕ ∧ ϕ|ϕ → ϕ|F ϕ|P ϕ|Gϕ|H ϕ
In Priorean language PRE-DET and POST-DET are expressed as follows:

• PRE-DET. ϕ → H F ϕ
• POST-DET. ϕ → GP ϕ
Both the classes of formulas PRE-DET and POST-DET are theses of Kt [19, Chapter VI],
the minimal logic of the Priorean language, i.e. they are propositions that are satisfied in
any model independently of the property of time. Thus the Priorean language does not
fulfil the condition of neutrality.
In the following two languages it will be considered in which:
• the rejection of both the theses PRE-DET and POST-DET will be possible, and
• the principles of PC and PE can be expressed, and
• LEM will hold.
3 Branching Time Logic
Since the thesis of determinism, DET, is a thesis of minimal Kt logic, a new language has
to be defined. Ockamist and Peircean languages are well known proposals. In any case it
is assumed that time is branching in the future with many possible time-lines (histories).
There are many courses of events possible. Since we are interested in neutral language,
i.e. a language such that no discussed thesis is a thesis of minimal logic of the language,
we do allow also branching in the past.
A branch of T is any maximal linearly ordered subset of T . Each branch represents a
possible course of events. For a point t and branch b, if t is a member of b, we say that t
lies on b or that b goes through t.
To say that at point t there will be ϕ means, that at any branch that goes through t there
will be ϕ. Analogously the past tense operator is understood. Let B(t) be the set of all
branches that go through t.
Definition 3.1 (F ) T , <, V !, t | F ϕ iff for any b ∈ B(t) there is t1 ∈ b, t < t1 :
T , <, V !, t1 | ϕ.
Definition 3.2 (P ) T , <, V !, t | P ϕ iff for any b ∈ B(t) there is t1 ∈ b, t1 < t :
T , <, V !, t1 | ϕ.
The formulations of definitions of G and H do not differ of respective definitions of the
Priorean language.
The theses PRE-DET and POST-DET are expressible as:
• POST-DET. ϕ → GP ϕ
Neither p → H F p nor p → GP p are theses of the minimal logic of the language of
branching time.
If PC is valid, then
• ϕ → P F ϕ
should be valid.
If PE is valid, then
• ϕ → F P ϕ
should be valid.
It could be remarked that neither formula expresses the fact that the cause/effect is in
any “between” moment.
Neither p → P F p nor p → F P p are theses of the minimal logic of the
language of branching time.
FS: for any t and for any t1 , t1 ≤ t :
1. there exists t2 such that t2 < t1 , and
2. for any t3 : if t2 < t3 , then t3 < t1 or t1 ≤ t3 ≤ t or t < t3 .
PS: for any t and for any t1 , t ≤ t1 :
1. there exists t2 such that t1 < t2 , and
2. for any t3 : if t3 < t2 , then t1 < t3 or t ≤ t3 ≤ t1 or t3 < t.
334 K. Trz˛esicki
The conditions FS and PS taken jointly characterize both directions open segment s.
Openness means that for any t ∈ s :
1. there is t1 ∈ s such that t1 < t, and
2. there is t1 ∈ s such that t < t1 .
If < is dense, s can be finite, i.e.:
1. there is t ∈ T such that for any t1 ∈ s : t < t1 , and
2. there is t ∈ T such that for any t1 ∈ s : t1 < t.
We maintain that PC holds if FS holds and PE holds if PS holds. It is true also when
the relation of causality is transitive. Moreover, even if both the conditions are fulfilled,
there are counter-models of PRE-DET and POST-DET. For any combination of PRE-
DET, POST-DET, PC and PE there is a frame in which this combination is valid and
for the other thesis or principle there is a counter-model. For example, for all PRE-DET,
POST-DET, PC and PE there is a counter-model if time is branching in the past and in the
future and the relation < is discrete and irreflexive.
There are some objections to proposed semantics of branching time logic as a tool of
formalizing arguments in favour of determinism. The relation of reachability of possible
states of events is defined as the relation < of earlier-later, a relation on the set of time
points. Maybe it is supported by the idea of time as a result of change offered by the
theory of the relativity of time. In the definition of the operators F and P any points
of the branches are allowed. Due to the fact the formula F ϕ rather expresses inevitability
of ϕ than its determination. Analogously it is true in the case of P . It seems that time
should be the same independently of branch.
4 Temporal Logic of Possible Worlds
Let us look for a solution such that the possibility of some event is referred to one and the
same point of time, not—as it was the case in branching time logic—different moments
are allowed. „There will be a sea-battle tomorrow” is true if it is determined that tomorrow
there will be a sea-battle, i.e. in any possible course of events it will be tomorrow but not,
e.g. in one possible course of events tomorrow, and another course the day after tomorrow.
The language of temporal logic of possible world not only will better expresses the
idea of determination but it seems to be, in some sense, more typical for modal logics.
The relation of accessibility will be a relation defined on sets of pairs consisting of
a possible world and points of time. Possible worlds will not differ in time T , <!. They
will differ only in valuation V .
Let W(T , <) = { T , <, i! : i ∈ I }. The relation between (W, t) and (W1 , t1 ) says
that the moment t1 ∈ T of the world W1 is reachable from a moment t of the world W .
Formally:
⊆ (W × T ) × (W1 × T ).
For intuitive reason it will be assumed that for any W, W1 ∈ W, and for any t, t1 ∈ T :
• (W, t) (W, t), i.e. is reflexive
• (W, t) (W1 , t1 ) only if t ≤ t1
Let us define the tense operators.
Definition 4.1 (G ) W, t | G ϕ iff for any t1 , t < t1 , and for any (W1 , t1 ), (W, t)
(W1 , t1 ) : W1 , t1 | ϕ.
Definition 4.2 (F ) W, t | F ϕ iff there is t1 , t < t1 , such that for any
(W1 , t1 ), (W, t) (W1 , t1 ) : W1 , t1 | ϕ.
Definition 4.3 (H ) W, t | H ϕ iff for any t1 , t1 < t, and for any (W1 , t1 ), (W, t)
(W1 , t1 ) : W1 , t1 | ϕ.
Definition 4.4 (P ) W, t | P ϕ iff there is t1 , t1 < t, such that for any
(W1 , t1 ), (W, t) (W1 , t) : W1 , t1 | ϕ.
The language is defined in the usual way. The theses of PRE-DET and POST-DET and
the principles of PC and PE are expressible as:
• POST-DET. ϕ → G P ϕ
• PC. ϕ → P F ϕ
• PE. ϕ → F P ϕ
PC and PE are not theses of the minimal logic of the language. They hold if some
conditions on the relation are imposed.
FS1: for any W, t, t1 : if (W, t1 ) (W, t), then there exists
1. t2 such that t2 < t1 , and
2. for any W1 , t3 : if (W, t2 ) (W1 , t3 ), and t3 ≤ t, then W1 = W
PS1: for any W, t, t1 : if (W, t) (W, t1 ), then there exists
1. t2 such that t1 < t2 , and
2. for any W1 , t3 : if (W, t3 ) (W1 , t2 ), and t ≤ t3 , then W1 = W
In the frame T , <, ! that fulfils the condition F S1, PC holds. If P S1 is fulfilled, PE is
valid. In the case the frame fulfils both the conditions F S1 and P S1, PC as well PE
are valid. In any case a construction of a counter-model for the theses PRE-DET or
POST-DET is possible. Using discussed semantics it is possible—as it was in the case
of branching time logic—to constructed frames in which any combination and only that
of PRE-DET, POST-DET, PC, PE is valid.
5 Conclusion
Abolishing arguments in favour of determinism does not mean that the universe, and we as
part of it, is not determined. Newtonian physics gives reasons for Laplacean determinism.
336 K. Trz˛esicki
Quantum physics says that everything happens with probability. Though not all of the
interpretations of quantum physics imply indeterministic ‘choices’ of events. A universe
governed by deterministic laws is preferably for the sake of making predictions. To be
free means to make decisions according to mind. To make rational decisions, knowledge
about a state of affairs is needed. According to quantum theory any attempt to know the
state of affairs is changing the state. Our acts have only probable effects. To be free we
have to grasp the reality. Logic is the fundamental part of our consciousness. Maybe a
candidate for a logic of free will is the quantum logic.
References
1. Hasle, P.F.V., Øhrstrøm, P.: Temporal Logic: From Ancient Ideas to Artificial Intelligence. Kluwer
2. Karpenko, A.S.: Łukasiewicz Logics and Prime Numbers. Luniver Press, Beckington (2006)
3. Lu, Z., Liu, J., Augusto, J.C., Wang, H.: A many-valued temporal logic and reasoning framework
for decision making. In: Atlantis Computational Intelligence Systems. Computational Intelligence in
Complex Decision Systems, vol. 2, pp. 125–146. Atlantis Press/World Scientific, Amsterdam/Paris
(2010)
4. Łukasiewicz, J.: On three-valued logic. Ruch Filozoficzny 5, 170–171 (1920). (English translation in
Łukasiewicz, J.: In: Borkowski, L. (ed.) Selected Works. Polish Scientific Publishers/North-Holland
Publishing Company, Warszawa/Amsterdam (1970))
5. Łukasiewicz, J.: Philosophische Bemerkungen zu mehrwertigen Systemem des Aussgankalkül.
Sprawozdania Towarzystwa Naukowego Warszawskiego Wydział III 23, 51–77 (1930)
6. Łukasiewicz, J.: O determiniźmie. In: Z zagadnień logiki i filozofii. Pisma wybrane, pp. 114–126.
7. Łukasiewicz, J.: On determinism. In: McCall, S. (ed.) Polish Logic 1920–1939, pp. 19–39. Oxford
University Press, Oxford (1967)
8. Łukasiewicz, J.: Selected Works, Borkowski, L. (ed.). Polish Scientific Publishers/North-Holland
Publishing Company, Warszawa/Amsterdam (1970)
9. Łukasiewicz, J.: On determinism. In: Borkowski, L. (ed.) Selected Work, pp. 110–128. North-
10. Łukasiewicz, J.: On three-valued logic. In: Borkowski, L. (ed.) Selected Work, pp. 87–88. North-
11. Prior, A.N.: The craft of formal logic. Unpublished manuscript, 806 pages (1951)
12. Prior, A.N.: In what sense is modal logic many-valued? Analysis 12, 138–143 (1952)
13. Prior, A.N.: Łukasiewicz’s symbolic logic. Australas. J. Philos. 30, 121–130 (1952)
14. Prior, A.N.: Three-valued logic and future contingents. Philos. Q. 3, 317–26 (1953)
15. Prior, A.N.: Many-valued modal systems. An intuitive approach. Philos. Rev. 63, 626–630 (1954)
16. Prior, A.N.: Time and Modality. Oxford University Press, Oxford (1957)
17. Prior, A.N.: Past, Present and Future. Oxford University Press, Oxford (1967)
18. Prior, A.N.: Papers on Time and Tense. Oxford University Press, Oxford (1968)
19. Rescher, N., Urquhart, A.: Temporal Logic. Library of Exact Philosophy, vol. 3. Springer, Wien
(1971)
K. Trz˛esicki ()
Higher Vocational School in Suwałki, University of Białystok, Białystok, Poland
Leśniewski and Mereology
Peter Simons
Abstract This paper surveys mereology, the theory of parts and wholes, focussing on its
origins in Leśniewski, and noting its intended employment as a surrogate for set theory.
We examine parallel and independent work by Whitehead, Leonard and Goodman, and
outline the subsequent adventures of mereology, both in its formal guises and in its now
intensive application within philosophical ontology.
Keywords Mereology · Part · Class · Stanisław Leśniewski · Calculus of individuals
AMS Subject Classification (2000) Primary: 03B99, Secondary: 03A05, 01A99
1 How Leśniewski Came to Mereology
Philosophers have used the notion of part since the beginning of their subject: it is far
too ubiquitous and important not to have entered their discussions. Plato employs it in the
Parmenides, Aristotle anatomizes various meaning of the term in the Metaphysics, Euclid
in the Elements defines a point as that which has no parts, and scholastic and modern
philosophers used it also. One who did so explicitly and extensively was Franz Brentano.
But the idea of a formal theory of part and whole did not occur to anyone until Brentano’s
student Edmund Husserl proposed it in his third Logical Investigation, “On the Theory
of Wholes and Parts”. Stanisław Leśniewski, who obtained his doctorate with another
Brentano student, Kazimierz Twardowski, probably read Husserl’s investigation. But the
motivation for Leśniewski’s mereology, to use the name he himself coined for the formal
theory of part, whole and cognate concepts, was a different one. It came from the theory
of sets or classes, and arose out of his endeavours to understand and neutralize Russell’s
Paradox.
Leśniewski first learnt about Russell’s Paradox in 1911 ([1], p. 169, [2], p. 181)
from the appendix to Jan Łukasiewicz’s 1910 book On the Principle of Contradiction
in Aristotle, where Łukasiewicz gives a short introduction to symbolic logic in Couturat’s
notation, and briefly outlines Russell’s Paradox. That paradox, it will be recalled, concerns

338 P. Simons
a certain class, the class of all classes that are not elements of themselves. Assuming there
is such a class, if it is an element of itself, by its definition it is not an element of itself,
but then by its definition it is an element of itself. Therefore it both is and is not an
element of itself, which is a contradiction. So clearly there cannot be such a class. But the
unreflective assumption about the existence of classes, namely that for any clearly statable
condition there is a class of just the things satisfying that condition, called the Principle of
Comprehension, seems to make it unproblematic that there is such a class: the condition
is just (using a variable ‘x’ to mark the place for insertion of constants):
x is a class and x is not an element of x
and the class is then (using the standard brace notation to bind the free variable)
{x: x is a class and x is not an element of x}
Since some classes clearly are not elements of themselves, for example the class of all
dogs is not a dog, there should be elements of such a class, so why not collect them all
together?
When Leśniewski first read about the paradox, he was convinced, as initially was
Russell and were others, that it was a silly minor problem that could easily be fixed. But as
time went by, nearly everyone who dealt with it realised that solving the difficulty would
take more work. Leśniewski worked for over a decade on it before he was satisfied that it
could be avoided in a way that was both intuitively satisfying and rigorous. It clearly con-
tinued to fascinate him long afterwards, for he lectured on antinomies in 1934–35 ([3], p.
179) and when he died in 1939 he left behind him a large manuscript on antinomies, which
was sadly destroyed in Bolesław Sobociński’s apartment during the 1944 Warsaw Rising.
The paradox was not just a little local difficulty, because class theory, or set theory, as
it came to be known, was widely held to be the most promising basis for a foundation of
mathematics. When in 1902 Russell communicated his surprising result to Frege, whose
logicist system of foundations was nearing completion, Frege recognised immediately
that the antinomy undermined his whole system, which had been constructed to that date
with unprecedented logical rigour. Frege quickly cobbled together a repair and published
this as an afterword to the second volume (1903) of his Grundgesetze der Arithmetik, but
at some point he must have realised that his repair blocked his derivation of arithmetic
from logic, and gave up. Some time later, in 1938, Leśniewski showed that Frege’s repair
had the unacceptable consequence that there could not be more than one individual (cf.
[4]). Russell, after much to-ing and fro-ing, came up with his theory of types as a way
of maintaining consistency, while Ernst Zermelo developed axiomatic set theory in a
way which presumptively avoided inconsistency while providing a strong (but no longer
logicist) foundation for mathematics. Whitehead and Russell’s three-volume Principia
Mathematica carried the Russell theory through, but at a price: to derive basic arithmetic
they had to assume the existence of an infinity of individuals, which certainly did not
seem like a logical truth. Mathematicians shunned the complexities of the theory of types
and, when they needed to, worked with Zermelo’s much simpler set theory or one of its
closely related descendants.
Leśniewski wanted to get to the bottom of the motivation driving Russell’s Paradox,
and to this end provided a first analysis in a paper published in 1914, “Is the Class of
Classes not Subordinated to Themselves Subordinated to Itself?” [5] His provisional
Leśniewski and Mereology 339
conclusion was that, if we understand the notion of class correctly, there is no paradox,
because every class is subordinated to (is an element of) itself, so there is no Russell Class
about which to pose the question. This paper used terms like ‘class’ and ‘subordinated
to’ understood in an intuitive way. Clearly this did not satisfy Leśniewski, because his
next publication, written in wartime Moscow and published in 1916, was “Foundations
of the General Theory of Sets I” [6]. In it, he set forth his first axiomatic treatment of the
concepts part, ingredient, set, class, element, subset, and others. This reaffirmed the 1914
conclusion on a much more satisfactory, axiomatic basis, and although he had yet to coin
the term, it may count as his first treatment of axiomatic mereology, the theory of part
and related concepts.
Over the next years, Leśniewski continued to develop and refine his views, modifying
his axiom systems in the direction of greater elegance, but in no way changing his views
on what classes or sets are. To avoid confusion of his system with those of the more
prevalent, what he called “official”, set theories, he renamed it ‘mereology’, from the
Greek word meros, ‘part’. He was able to show that there are several equivalent ways to
axiomatize mereology, using different notions as primitive. Later his energies went into
developing the logico-linguistic basis for mereology: first, the logic of names and predi-
cates which he called ‘ontology’, and subsequently the logic of propositions, propositive
functors and quantifiers that he called ‘protothetic’, which extended what Whitehead and
Russell had called ‘theory of deduction’ and what is now called ‘propositional logic’.
Appalled at the presentational sloppiness and use/mention conflations of Principia
Mathematica, Leśniewski somewhat naively assumed that all symbolic logic had to be
like that, and so preferred to develop mereology and ontology initially in a carefully
restricted natural language idiom, a small fragment of his native Polish, supplemented by
variables. In 1920 Leon Chwistek persuaded him that symbolization was advantageous
and so Leśniewski set about formulating with unprecedented precision the axioms and
rules of development of his systems, consistent with a nominalistic attitude to logical
systems. He never published his mereology as a symbolic system, but some notes from his
Warsaw lectures were published posthumously in 1988 in the collection S. Leśniewski’s
Lecture Notes on Logic under the title “Class Theory” and they give a flavour of how he
presented his mature mereology to students (cf. [3], pp. 59–125).
2 Leśniewski’s Understanding of Sets or Classes
Leśniewski took his understanding of sets or classes from the words of Georg Cantor, who
wrote, “By a ‘set’ we understand any aggregation M of determined well-distinguished
objects m of our intuition or thought (called the ‘elements’ of M), into a whole.” It will
be noted that Cantor speaks of aggregation into a whole (Zusammenfassung zu einem
Ganzen). Both the idea of aggregation and that of a whole are resonant of mereology,
so Leśniewski’s way of taking Cantor’s idea, which is admittedly expressed less than
wholly perspicuously, is understandable. That way is to treat a set or class as a composite
individual literally made up or composed of its elements. If m are my books, then their set
M, or {m}, as Cantor sometimes writes it, is my library, understood as that single individual
or whole which contains all my books as parts, and nothing else outside that whole.
Leśniewski referred to this as a class taken in the collective sense. There is another
sense of class, taken in the distributive sense. For example, when we speak of the class of
340 P. Simons
all my books, when understood in the distributive sense, this does not mean the composite
individual made up of my books but simply my books, of which there are many. So the
expression ‘This paperback is an element of the class of all my books’ can be understood
in two different ways. In the distributive sense, it is simply an elaborate way of saying
‘This paperback is one of my books’, where the reference to a class drops out. The logical
predicate ‘is one of’ is what Leśniewski came to symbolize by the Greek letter ‘ε’ and
refer to as the functor of singular inclusion. It became the primitive predicate of his system
of ontology. In many contexts, for example if we say ‘Fido is one of (all) the dogs’, it is
simpler and more idiomatic to say ‘Fido is a dog’, and represent it semi-symbolically
as ‘Fido ε dog’. In Leśniewski’s native Polish, which is not cluttered up by articles, the
functor is rendered simply as ‘jest’: Fido jest psem.
Taken in the collective sense however, we cannot say that Fido is one of the class of all
dogs, since there is only one such class and that would mean Fido was it and therefore the
only dog. (Unless Fido has another dog as a proper part, which is conceivable but not how
dogs generally are.) Rather we must say that he is an element, member, or, as Leśniewski
is happy to say, that he is a part (cz˛eść) of this class.
There are, it must be said right from the start, reasons to think that Leśniewski’s
treatment of classes is not exhaustive. Consider the following fallacious syllogisms:
Fido is one of the dogs
The dogs are a species
therefore
Fido is a species
or the following mathematical example
8 is one of the Pythagorean triple of numbers 6, 8, 10
The Pythagorean triple 6, 8, 10 is one of the Pythagorean triples containing 8
therefore
8 is one of the Pythagorean triples containing 8.
In each case, whether we understand the object of the sentence as a class in the
distributive or in the collective sense, we get nonsense. If we understand ‘the dogs’
distributively, they cannot be one thing, a species, of which there are others, cats, horses
etc., but equally if we understand ‘the dogs’ collectively, then not only is Fido not a
species, though he is a member of a species, things other than dogs would have to be
taken as members of the species, such as the class of all attached dogs’ tails. What we
have in each case is a class of several things being one of a class of several other classes:
species, or Pythagorean triples. Neither of Leśniewski’s ways of taking the notion of
a class captures this feature, and that is why, when set theory became established as a
framework for dealing with mathematical matters, his efforts were sidelined, since the
representative and expressive power of set theory rests primarily on this ability to have
classes of classes (of classes . . . ) which do not collapse to the lowest level as they do on
either of Leśniewski’s understandings. Technically, it turns on the non-transitivity of the
element relation. Whether Leśniewski understands ‘A is an element of the class of as’ to
mean ‘A is one of the as’ or ‘A is a part of the (collective) class of the as’, the relational
functor is transitive, whereas in standard set theory it is not, and that is why standard set
theory can build hierarchies that Leśniewski cannot.
3 Terminology
We need to say something here about terminology, both Leśniewski’s and ours, because
several of the terms have been used differently by different people, and we are going to
have to fix their use for later. We have seen that Leśniewski starts using terminology
associated with standard set theory. Initially he called a collective class a multitude,
multiplicity or set (mnogość). In this sense a set is one thing, an individual, something
which is enshrined later in his axioms. A set of ms is a collective class of some ms,
at least one, and perhaps, but not necessarily, all. The unique set composed of all the
ms he calls the class (klasa) of all ms. Later Leśniewski dropped the term mnogość in
favour of the more usual term for sets or collections in Polish, zbiór, but he continued to
use klasa for the complete collective class of all things of a certain kind. Lejewski, who
brought Leśniewski’s ideas to England after the Second World War, used ‘collection’ for
sets and ‘complete collection’ for classes. Later mereologists have used the terms ‘sum’
and ‘fusion’ for collective classes. This more modern usage derives from Leonard and
Goodman, stressing the individuality of the result and moving away from Leśniewski’s
ancestral link to the theory of sets or classes. For the next section, we adopt Leśniewski’s
own terminology, in order to give its flavour, but subsequently we shall adopt the
following terminology: a collective class of some but not necessarily all as will be called
a fusion of as; a (in fact: the) collective class of all the as will be called a (the) sum of as.
Now for terminology about parts etc. Leśniewski used cz˛eść (‘part’) for the asymmetric
relation, which entails that no object is a part of itself. In English terminology, this is
now usually called the proper part relation, and we will adopt this term. He used the
term ingredyens (‘ingredient’) for the related improper part notion, according to which
an object is an improper part of itself. Later he introduces the terms ‘element’ and
‘subset’ which turn out to be equivalent to ‘ingredient’, and in later expositions dropped
‘ingredient’ in favour of ‘element’. We will stay closer to modern terminology and call
this the part relation. Where we mean ‘proper part’ we will say so.
Other terms, such as ‘overlap’ and ‘exterior’ are less problematic and will be
introduced as we come to them. None of this terminology is perfect, but once we
understand closely enough what is meant, we can convert to symbols, where the meaning
gets more precisely fixed by the axioms.
4 The Early System
In his 1916 monograph, Leśniewski takes the notion of proper part as primitive, and
lays down four axioms, two of which require some definitions before their introduction.
We will give them in a translation of his prose, and then in the following section give
them a symbolic form. Our lettering does not exactly follow Leśniewski: he talks of
‘object P’, ‘object P1 ’ and the like, using a capital letter ‘P’ sometimes with numerical
subscripts, when talking about an individual object (przedmiot) and ‘objects m’, (‘m’ as in
‘mnogość’) using a lower-case letter, when talking about one or several individual objects.
We will follow this distinction between upper-case and lower-case letters though it is a
342 P. Simons
presentational device only and can be omitted without loss of meaning or precision. We
recall that in 1916 Leśniewski used semi-prose only, no logical symbols.
Axiom I If object A is a proper part of object B, then object B is not a proper part of
object A.
This is the asymmetry of proper parthood. Leśniewski’s use of singular letters and the
expression ‘object A’, ‘object B’ indicate that he intends that if A is a proper part of B then
not only is the part A an individual (as signalled by its appearing as subject before ‘is a’),
but that the whole B is also an individual.
In his ontology, Leśniewski later made clear the difference between two ways to
understand negative sentences of the form ‘B is not an m’. One is as a sentential negation,
‘it is not the case that B is an m’. The other has a nominal negation, ‘B is a non-m’. In the
latter case the presence of ‘B’ before ‘is a’ clearly indicates that B is an individual. With
sentential negation, the negative sentence is true if the subject term is empty or plural. The
wording does not make it clear which is meant here, but nominal negation fits better.
Axiom II If object A is a proper part of object B, and object B is a proper part of object
C, then object A is a proper part of object C.
This is the transitivity of proper parthood. Again it is clear that the three schematic
terms are intended as singular denoting terms.
Definition I A part of A is (Def.) either A or a proper part of A.
Leśniewski actually expresses this definition as a metalinguistic description of how he
intends to use the expression ‘part’ (ingredyens):
I use the expression ‘part of object A’ to denote the same object A as well as every proper part of
this object. ([1], p. 261; [2], p. 132).
Definition II A set of m is (Def.) an object A, all of whose parts have a part in common
with an m which is part of that object A.
Employing, in advance of its definition, the idea of two objects overlapping when they
have a common part, we can paraphrase this and make it a little easier to understand:
A set of m is an object A, all of whose parts overlap some m which is a part of A.
This rules out any object which has a part that is not part of any m, but it also means that
any m that overlaps A is itself wholly within the set A, in other words, the set is composed
of all parts of those m involved. But these need not be all the m. For example, if m is dogs,
then one set is the whole composed of just three of these dogs, say Fido, Spot and Rover.
Indeed, a limiting case of a set would be the set of just one dog, which is simply that dog
itself. To get the biggest possible set, what Leśniewski calls the class of all m, he needs
another definition:
Definition III A class of m is (Def.) an object A such that every m is a part of A and
every part of A has a part which is also a part of some m.
Again, this is easier to understand if we employ the concept of overlapping, thus:
A class of m is an object A of which every m is a part, and of which every part overlaps
some m.
Because of the requirement that every m be a part of the class, the class is a maximal
set of m. Now Leśniewski is ready to state his final two axioms.
Axiom III If some object is an m, then some object is a class of m.
Axiom IV If A is a class of m and B is a class of m, then A is B.
The third axiom guarantees the existence of a class (sum) of all m provided at least one
m exists, and the fourth axiom ensures the uniqueness of any such class. The uniqueness
allows us to speak of ‘the class of m’ when one exists. In fact, there fails to be a class of
m only when there are no m. The final clause, ‘A is B’, should be understood as meaning
that A and B are the same object, so the ‘is’ is the ‘is’ of singular identity.
Leśniewski goes on to give a further six definitions and he states and proves 57
theorems using the primitive and defined notions. We give the definitions as they show
something interesting.
Definition IV A is an element of B if and only if (Def.) for some x, A is an x and B is the
class of x.
Here Leśniewski is introducing some standard set-theoretic terminology, alongside the
terms ‘set’ and ‘class’. But the term ‘element’ is redundant: the two theorems he proves
after the definition show that object is an element of another if and only if it is a part of it.
Definition V A is a subset of B if and only if (Def.) every element of A is an element of
B.
Once again it turns out that being a subset is coextensive with being a part (or
an element). What I think Leśniewski is doing here is adopting standard set-theoretic
terminology in order to hijack it and then show that it is not needed, that the original
notions of part and class (sum) are sufficient. The aim here is then not logical elegance,
as it would be later for Leśniewski: it is rather the annexation of set theory by mereology.
Definition VI A is a proper subset of B if and only if (Def.) A is a subset of B and A is
not B.
Again this definition is unnecessary, because being a proper subset is coextensive with
being a proper part.
Definition VII The Universe is the class of all objects.
Leśniewski goes on to state as a theorem:
Theorem XLIII Some object is the class of non-contradictory objects.
This class is the Universe (wszechświat) and it is unique. Leśniewski’s result here
smuggles in an unstated assumption: that there is at least one non-contradictory object.
This is of course true, but it does not follow from his previous axioms, all of which
would be true even if there were no objects. In his later work, Leśniewski would reject the
unstated assumption here and ensure that none of his logical systems have as a theorem
that there exists at least one object.
Definition VIII A is exterior to B if and only if (Def.) no object is a part of both object
A and object B.
344 P. Simons
Note that this is not the logical contradictory of overlapping since either or both of
the variables ‘A’ and ‘B’ might be plural or empty. But among existing individuals, two
objects overlap if and only if they are not exterior to (disjoint from) one another.
Definition IX A is a complement of B with respect to C if and only if (Def.) A is the class
of all parts of C that are exterior to B.
It is a quick consequence of this that
Theorem XLVIII If A is a proper part of B, then some object is the complement of A
with respect to B.
This say that any proper part of a whole has a remainder, its complement, which
disjointly exhausts the whole together with it. It is an instance of what is now called a
supplementation principle ([7], p. 26).
5 Remarks About This System
While Leśniewski’s meaning is clear, from the point of view of logic, the system is
still imperfect. Apart from the fact that the underlying logics of connectives, quantifiers,
names and functors had still to be developed, there are inelegancies of presentation. The
definition of ‘set’ is not needed for the definition of ‘class’. It is no doubt interesting to
see that a class is a special case of a set, but the definition of ‘class’ does not use that of
‘set’, though it might have done. Perhaps in a first version, Leśniewski had defined ‘class’
in terms of ‘set’ but later found the definition of ‘class’ that he gave and used that but
without removing the definition of ‘set’. Leśniewski himself came to see the interpolation
of definitions, those of ‘part’ and ‘class’, between the first two and the last two axioms, as
an imperfection and preferred to look for an axiomatization in terms of the primitive alone,
which he subsequently achieved. Also the fact that some of the variables are “unofficially”
singular and occur in phrases in apposition like “object A” is not transparent logical
practice. We will remedy this below in our symbolic reconstruction, drawing on his later
work. Further, as we noted, Leśniewski deliberately introduces redundant terminology
from set theory in order to try and demonstrate that set theory is using a lot of unnecessary
terms to talk about something rather straightforward. So the 1916 paper sacrifices much
logical elegance for the sake of propaganda.
Philosophically, there are two things to note. The first is that the first pair of axioms,
governing the notion of a proper part, are unobjectionable. The second remark however
is that the remaining two axioms, on the existence and uniqueness of sums, are far
from anodyne. The third axiom entails the existence of arbitrary sums of objects, which
is and remains one of the more hotly disputed propositions of mereology. We will
not enter into the philosophical controversy here, but simply note that Leśniewski’s
strong commitment results principally from his keenness to use mereological concepts
in providing a foundation for mathematics. The final axiom, in the presence of the
others, ensures a form of mereological extensionality: objects with the same parts are
identical. Because of the strength of these principles, Leśniewski’s axiomatization does
not really test the limits of the everyday notion of part: it goes all the way to the maximal
extent. Theorem XLVIII above is much closer to analysing the essential characteristics of
parthood, the existence of supplements of proper parts, but it is a mere consequence of
the strong Axioms III and IV and so its conceptual importance gets lost in the big noise
they make.
6 Symbolic Reconstruction
Here we draw on the symbolic developments Leśniewski was to introduce in his full
logic comprising protothetic and ontology, in order to give symbolic equivalents of the
prose axioms and definitions of 1916. For full explicitness some additional definitions are
required beyond those Leśniewski gives.
In what follows Italic letters are used for variables, Roman letters for constants.
The standard modern symbols for the logical constants, connectives and quantifiers
are used, where it must be noted that quantification does not, for Leśniewski, carry
ontological import. That is carried rather by the basic predicate of singular inclusion. We
represent singular inclusion, ‘is a’ or ‘is one of’ by ‘ε’, the standard undefined constant
of Leśniewski’s ontology. A sentence of the form ‘A ε m’ is true if and only if A is an
existing individual object and it is one of the object or objects m. Universal quantifiers
having a whole formula as their scope are omitted but should be understood as tacitly
present: Leśniewski does not allow free variables into his stated theses. Quantifier scope
is marked, as in Leśniewski, by upper corners. Unless otherwise indicated by parentheses
or constrained by corners, the order of grouping by binary connectives, from loosest to
tightest is: ↔ → ∨ ∧. Otherwise, the symbols are straightforward. We shall assume the
axiom and rules of ontology in the background as to give them explicitly would bloat this
article. We start with some definitions, where the definiendum is always on the left of the
equivalence and the definiens on the right. As in Leśniewski these are to be understood as
object-language equivalences rather than metalinguistic abbreviations.
D= A=B↔AεB∧BεA Singular identity
DV AεV↔AεA Object
DN A ε N(m) ↔ A ε A ∧ ∼(A ε m) Nominal negation
The undefined primitive functor is ‘proper part’. We symbolize that A is a proper part
of B using the name-forming functor of one nominal argument ‘ppt’ as follows: A ε ppt(B).
Here now are Leśniewski’s axioms and definitions:
A1 A ε ppt(B) → B ε N(ppt(A))
A2 A ε ppt(B) ∧ B ε ppt(C) → A ε ppt(C)
D1 A ε pt(B) ↔ A = B ∨ A ε ppt(B) (Proper or improper) part
There is subtlety to this definition of ‘part’, as indeed of the two previous definitions,
that is easy to overlook. Any expression of the form ‘A ε m’ entails by the meaning of the
singular inclusion functor the existence and singularity of A. Leśniewski covers this by
talking of ‘the object A’ and the like. So a definition like this, which Leśniewski would
346 P. Simons
later call an ontological definition, uses a variable, in this case ‘A’, in such a way that
the left hand side (the definiendum) can only be true if this object A exists and is an
individual. For this to be a satisfactory definition, that requirement has to be fulfilled by
the right hand side (the definiens) as well. Now in this case it is, because the sentence
‘A = B’ is true if and only if both names are singular and denote the same individual, as
indicated in D=, and the second disjunct likewise requires A to be an existing individual.
But not all of Leśniewski’s verbal definitions make this equally clear, in particular that of
fusion (our name from now on for ‘set’):
D2 A ε fu(m) ↔ A ε V ∧ ∀B B ε pt(A) →
∃C C ε m ∧ C ε pt(A) ∧ ∃D D ε pt(B) ∧ D ε pt(C) Fusion
Notice the need for the first conjunct in the definiens, since without it the right
hand side could be vacuously true were the name ‘A’ to be empty or plural. The first
conjunct rules these cases out and ensures that the definition is adequate. The import of
this definition may not be immediately clear. To help make it clearer, let’s introduce an
auxiliary definition that Leśniewski was to add later, that of two objects overlapping. Two
objects overlap when they have a common part:
Dov A ε ov(B) ↔ ∃C C ε pt(A) ∧ C ε pt(B) Overlapping
We can then express the equivalence in D2 as
A ε fu(m) ↔ A ε V ∧ ∀B B ε pt(A) → ∃C C ε m ∧ C ε pt(A) ∧ B ε ov(C)
D3 A ε Sm(m) ↔ A ε V ∧ ∀B B ε m → B ε pt(A) ∧ ∀C C ε pt(A) →
∃D D ε pt(C) ∧ ∃E E ε m ∧ D ε pt(E) Sum
Again we need to state explicitly in the definiens that A is an object. This too is easier
to understand if we employ the concept of overlapping, so
A ε Sm(m) ↔ A ε V ∧ ∀B B ε m → B ε pt(A) ∧
∀C C ε pt(A) → ∃D D ε m ∧ D ε ov(C)
A3 ∃A A ε m → ∃B B ε Sm(m)
A4 A ε Sm(m) ∧ B ε Sm(m) → A = B
D4 A ε el(B) ↔ ∃x A ε x ∧ B ε Sm(x) Element

D5 A ε sub(B) ↔ A ε V ∧ ∀C C ε el(A) → C ε el(B) Subset
As noted above, the three functors ‘pt’, ‘el’ and ‘sub’ are coextensive.
We now give all the other definitions Leśniewski uses:
D6 A ε psub(B) ↔ A ε sub(B) ∧ ∼(A = B) Proper Subset
This is likewise coextensive with ‘ppt’.
D7 U = Sm(V) Universe
This is the form of the definition Leśniewski gives: note that it is an identity. In his
mature work, nominal terms are introduced using singular inclusion, not identity, partly
for symbolic parsimony, but mainly because existence conditions need to be given. Having
a definition does not always mean something satisfies it, as in the case of a canonical
empty term. The definition D7 would be false if nothing existed, unless the identity
symbol were reinterpreted to be without existential import. Here then is an equivalent

but unproblematic definition for the Universe, in that form:
D7* A ε U* ↔ A ε V ∧ ∀B B ε V → B ε pt(A)
It is readily proved from the axioms and definitions that U* is unique, and that if U or
U* exists, then U = U*. So if any object at all exists, the object U (= U*) exists.
D8 A ε ex(B) ↔ A ε V ∧ B ε V ∧ ∼∃C C ε pt(A) ∧ C ε pt(B) Exterior
Two objects are exterior to (disjoint from) one another if they both exist but have no
common part.
Leśniewski’s last definition is of the (mereological) complement of one object in
another. To formulate it we need a way to yoke two names together conjunctively. Here is
how:
D∩ Aεm∩n↔Aεm∧Aεn Nominal Conjunction
Now we can render
D9 A ε Cpl(B,C) ↔ B ε sub(C) ∧ A ε Sm(el(C) ∩ ex(B)) Complement
The complement of B in C is the sum of all parts of C that are disjoint from C’s part
B. The supplementation theorem then comes out as
T48 A ε ppt(B) → ∃C C ε Cpl(A,B)
Since the complement is what is left when “subtracting” one part from a whole, another
notation for ‘Cpl(A,B)’ could be ‘B − A’.
7 Later Improvements
We have dwelt at some length on Leśniewski’s 1916 system for three reasons. The first
is that it is his first published mereology, and so the first published mereology überhaupt.
The second is that it is relatively underexposed in the secondary literature. The third is
that its use of redundant definitions and set-theoretic terminology makes Leśniewski’s
motivation clear. He wishes not only to give a solid basis to his own views, but is interested
in converting set theorists to his view by adopting their terminology but then interpreting it
in his own way. It becomes clear from this that Leśniewski’s discontent with “official” set
theory, which later led to his falling out with several of his mathematical contemporaries,
had roots going right back to his earliest work on foundational matters.
Despite intensive work on mereology as well as on his other systems, protothetic
and ontology, the next time Leśniewski published on mereology was in 1927–1931, in a
series of articles entitled O podstawach matematyki (On the Foundations of Mathematics),
appearing in Przeglad ˛ Filozoficzny (cf. [1], pp. 295–468; [2], pp. 174–382). The articles
are written in a somewhat unusual autobiographical manner, detailing the development of
Leśniewski’s own ideas in more or less chronological sequence. They contain a critical,
destructive part, followed by a longer, constructive part, and finish with general remarks
on singular propositions of the form ‘A ε b’ which are relevant to ontology rather than
mereology. In the critical part, Leśniewski disapproves strongly of the logical inexactness
348 P. Simons
of Whitehead and Russell’s Principia Mathematica, which he holds in much lower esteem
than Frege’s Grundgesetze der Arithmetik, the latter’s inconsistency notwithstanding. But
his chief criticism is reserved for standard systems of set theory, among whose proponents
he singles out Zermelo, Hausdorff, Dedekind, Schröder, Sierpiński and Fraenkel. His
main objections are that such systems “create”, “invent” or “introduce” an empty set,
an object consisting of or comprising no objects at all, and that they distinguish between
a single object and the set whose sole element is that object, as we would now say, an
object x and its singleton {x}. His own theory, he notes, admits no empty set and identifies
an object with its singleton. His tone is scathingly ironic, and this no doubt made him no
friends. In his critical remarks on Schröder’s extensionalist conception of classes, Frege
had criticised the idea of an empty class, as well as the distinction between an object
and its singleton, as inappropriate for Schröder’s conception, and Leśniewski quotes
Frege with approval. But his own view is close to Schröder’s, and Leśniewski professes
inability to understand Frege’s treatment of classes as the extensions of concepts, and so is
thrown back on his own previous conception, which he claims to have held from the very
beginning. Zermelo’s principles of set construction are considered ad hoc by Leśniewski,
simply blocking the paradoxes but without a satisfactory analysis of their source. Since
many of Zermelo’s constructive axioms closely follow the ideas of Cantor, this is a partly
unfair accusation, but the Axiom of Separation, which is the principal block to paradox,
does look like an artificially constrained version of the naïve Principle of Comprehension.
Leśniewski first reprises the 1916 ideas in a somewhat streamlined and improved form.
He places the definition of sum (class) before that of fusion (set), which he now calls not
mnogość but zbiór (collection); his theorems are developed in a slightly more streamlined
way. He then adds further definitions and theorems going beyond the 1916 results until he
has a total of 198 theorems, all presented and proved in much the same way as in 1916,
without special symbols other than variables. These, he says, are results he obtained up to
1920.
Leśniewski soon found himself able to give axiomatic treatments of his theory, which
from 1927 he took to calling ‘mereology’, without interpolating definitions. The first such
treatment used ‘ppt’ as the primitive, but another used ‘part’, (now called ‘element’ rather
than ‘ingredient’) and as this is slightly simpler, we give it here as Leśniewski formulated
it, modulo changes of terminology and schematic variables.
Axiom (a) If A is a part of B and not (B is A) then B is not a part of A
Axiom (b) If A is a part of B and B is a part of C, then A is a part of C
Axiom (c) If every a is a part of A and a part of B, and for any C, if C is a part of A or
a part of B, then some part of C is a part of some a, then A is B
Axiom (d) If some object is a, then for a certain A, ((for any B, if B is an a, then B is a
part of A) and for any C, if C is a part of A, then some part of B is a part of some a).
The terms ‘proper part’ and ‘class’ (sum) are then defined as
Def (e) A is a proper part of B iff A is a part of B and not (A is the same object as B)
Def (f) A is the class of objects a iff (A is an object, for any B, if B is an a, then B is a
part of A, and for any C, if C is a part of A, then some part of B is a part of some a)
Axiom (c) effectively asserts the uniqueness of sums, while Axiom (d) effectively
asserts the existence of a sum of as if any as exist.
Leśniewski shows that this more logically elegant system is deductively equivalent
to the original, and he then goes on to prove a grand total of 264 numbered theorems.
But we notice here the beginnings of a trade-off between logical elegance on the one
hand and perspicuity on the other. The axioms (c) and (d) are less easily motivated in
the abstract than the original 1916 axioms which used intermediate definitions: they are
as they are just in order to give the same results. Similar remarks apply to other systems
with other primitives that Leśniewski developed, including one from 1921 based solely
on ‘exterior’. Leśniewski was able to state that quite a few terms could be taken as sole
primitives, including the disjoint sum of two individuals A + B and its generalization, the
disjoint sum (suma) of individuals a, where
DDsm A ε Dsm(a) ↔ A ε Sm(a) ∧ ∀BC (B ε a ∧ C ε a) → (B = C ∨ B ε ex(C))
By now, logical innovation had given way to logical perfectionism and the exploration
of every last corner of possibility. Increasingly, the focus both in Leśniewski’s own work
and in that of his pupils and followers was on incremental refinements of the basic idea of
the systems, rather than questioning, motivating or revising them.
8 Whitehead’s Alternative Mereology
At almost exactly the same time as Leśniewski was formulating his mereology, a similar
development was taking place in the work of Alfred North Whitehead. Whitehead needs
no introduction, but the reason why he was interested in mereology does. Whitehead
was of course Russell’s teacher and his collaborator in the three-volumed Principia
Mathematica (1910–13), the most extensive attempt to show that mathematics is derivable
from logic, and the biggest compendium of results in mathematical logic to that date. In
the three published volumes, Whitehead and Russell dealt with logical basics, including
propositional logic and type theory, and showed how to formulate the basic notions of
cardinal, ordinal and real-number arithmetic. Unlike Frege however, they thought the
logicist approach could be applied to geometry, and they planned a fourth volume of
Principia on geometry, which Whitehead was to author alone. Whitehead’s approach to
and work on geometry is highly complex, and underwent several transformations, which
it is beyond the remit of this paper to outline. Suffice it to say that Whitehead had long
been impressed by the geometric algebras of Hermann Günther Graßmann and William
Rowan Hamilton, had studied modern axiomatics of affine and metric geometries with
care, and written short textbooks on them. His inclinations to integrate geometry as the
mathematical theory of space with the physics of the occupants of space and time were
apparent as early as his 1906 Royal Society memoir ‘On Mathematical Concepts of the
Material World’. But it is clear that when working on the fourth volume of Principia
(which for various reasons was never completed) Whitehead became interested in the
part-whole relation. He was interested in providing a point-free account of geometry, and
to this end developed, building on the earlier memoir, the idea of extensive abstraction,
350 P. Simons
taking points, and other lower-dimensional elements, not to be existent items within his
geometry but as special kinds of abstraction arrived at through sequences of successively
finer approximations. In his 1916 paper “La théorie relationniste de l’espace” [8], based
on a lecture given in Paris in 1914, he notes that ‘part’ can mean subset, a component
(“heterogenous part”) of a more complex whole, or a spatial part (“homogeneous part”).
He defines a relation he calls ‘inclusion’ and shows it is reflexive and transitive, but his
sketchy account falls short of providing a full mereology.
After the war however, Whitehead was more forthcoming. In his 1919 book An Enquiry
Concerning the Principles of Natural Knowledge [9] he outlines, without pretence at
a properly developed formal system, some principles governing a relation he calls
‘extending over’, defined on events ([9], pp. 101 ff.) Extending over is simply his name
for the converse of the proper part relation. The term probably comes from the English
translation of Graßmann’s word Ausdehnung. Among the principles Whitehead supports
are the asymmetry and transitivity of the proper part relation, a strong supplementation
principle, and the notion of a disjoint sum, which he calls a partition of an object. But
his mereology differs from that of Leśniewski in several ways. It is atomless and lacks a
universe: every object both is and has a proper part. Further it is mereologically dense: if
A is a proper part of B, there is a third object C which is a proper part of B and of which
A is a proper part. The lack of a Universe is directly contrary to Leśniewski; atomlessness
and density are independent of Leśniewski’s system.
Whitehead’s motivation for developing mereological notions is quite different from
Leśniewski’s: he is interested in providing an account of the spatiotemporal and geo-
metrical characteristics of the universe in terms of the relations among its inhabitants,
which are events. His treatment of mereology in Enquiry is done en passant and is
formally offhand and imprecise. When in 1926 Tarski drew Leśniewski’s attention to
the similarities between their two theories, Leśniewski found little difficulty in detecting
formal defects in Whitehead’s exposition. He outlined his criticisms, along with copious
quotations from Whitehead’s English, in a very long note at the end of Chapter IV
of ‘Foundations’. Leśniewski is right to find fault, but the defects in Whitehead are
quite easily remediable (cf. [10]), and Leśniewski characteristically fails to penetrate
beyond the formal shortcomings and consider the interesting doctrinal differences and
their motivation.
Whitehead developed a revised mereology in his 1929 Process and Reality [11], but
it is there mixed with topological concepts, and deals directly with spacetime rather than
with the events therein, underlining the geometric motivation of Whitehead’s treatment.
It is of course a huge pity that the work for Volume IV of Principia failed to survive, as
it would have been developed to a much higher formal standard and would have provided
a more robust comparison with Leśniewski’s work. What we can say is that Whitehead
and Leśniewski developed mereological theories at almost the same time, around 1914,
though Leśniewski was first into print. Their motivations were very different, and that is
why the details and implications of their theories differ, despite sharing a conceptual core.
9 The Polish Continuation
Leśniewski died a few short months before the Nazi invasion of Poland and the beginning
of the Second World War. His widow gave his papers to Bolesław Sobociński to edit,
but they were destroyed in 1944. Any results that Leśniewski possessed that had not
been published either were lost or had to be reconstructed from the memory of those
who had worked under him or heard him lecture in Warsaw. There were two principal
contributors of this kind. The first was Sobociński, who emigrated to the United States
and became professor at Notre Dame University, founding and editing the Notre Dame
Journal of Formal Logic. Sobociński was a loyal adherent of Leśniewski and did not
amend the master’s overall conception. His principal contribution was to consider systems
of atomistic mereology. In mereology, an atom is an object with no proper parts, and
mereological atomism is the theory that every object is completely composed of atoms,
or is a sum of atoms. The atomistic hypothesis is usually expressed as follows:
Dat A ε at ↔ A ε V ∧ ∼∃B B ε ppt(A)
Atm ∀A A ε V → ∃B B ε at ∧ B ε pt(A)
Atomism, its contrary, atomlessness, and mixed intermediate positions are all logically
independent of general mereology. Atomistic mereology, unlike its contraries, has finite
models. If in a model there are n atoms, then the model has cardinality 2n − 1, since
any non-empty collection of atoms has a sum. Sobociński showed how to axiomatize
atomistic mereology in ways suited to its peculiarities. He also made several incremental
improvements to axiom systems for general mereology.
The other adherent of Leśniewski who contributed to mereology was Czesław
Lejewski, who though a student of classics had attended Leśniewski’s advanced classes
in the 1930s. After the war he settled in England, becoming Professor of Philosophy
at the University of Manchester. Like Sobociński, he explored various alternative
axiomatizations of mereology and made numerous incremental improvements. Among
his novel contributions, one is worth mentioning here: he provided a syntactic consistency
proof of mereology relative to protothetic (cf. [12]).
Tarski himself contributed to mereology in the 1920s. He showed that Axiom (a) of
Leśniewski’s Przeglad ˛ system is redundant: since an individual A is identical to the sum of
its parts, we can use the definition and uniqueness of sum to show that if A is part of B and
B is part of A, then A = B, which gives Axiom (a) by partial contraposition. He then, in an
article published in French in 1929 [13], employed a mereology using singular variables
only and a standard notion of class to provide a point-free formulation of the geometry
of solids. Readers of Tarski in the English translation of this paper might be forgiven
for thinking that the system of mereology Tarski there gives was in the original, but in
fact it was added in the English translation. Arianna Betti has conjectured, plausibly, that
Tarski might not have wished to become embroiled in a priority dispute with his former
teacher, whose mature work on mereology had not yet been published, and also that the
formulation of the system, using standard class theory, would have seriously displeased
Leśniewski (cf. [14]).
352 P. Simons
Be that as it may, in the expanded English version of 1956, Tarski formulates

mereology using two axioms and two definitions (of which the first is not used for the
mereology as such but only for the ensuing geometry):
Definition I An individual X is called a proper part of an individual Y if X is a part of Y
and X is not identical with Y.
Definition II An individual X is said to be disjoint from an individual Y if no individual
Z is a part of both X and Y.
Definition III An individual X is called a sum of all elements of a class α of individuals
if every element of α is a part of X and if no part of X is disjoint from all elements of α.
Postulate I If X is a part of Y and Y is a part of Z, then X is a part of Z.
Postulate II For every non-empty class α of individuals there exists exactly one
individual X which is a sum of all elements of α.
This system forms the basis for an extremely elegant mereologically-based axiomatic
treatment of a solid geometry without points, using the primitive geometric notion
of sphere. From a mereological point of view, several remarks are in order. Tarski’s
elimination of Leśniewski’s first axiom is indeed a simplification, as he stated, but the
second Postulate is simply a telescoping of Leśniewski’s third and fourth axioms into
one, as revealed by the use of the expression ‘exactly one’, which conjoins the notions
‘at least one’ and ‘at most one’. Further, the logical basis includes sets (classes) in the
standard sense, and not in either Leśniewskian sense, though it could easily be transposed
into Leśniewskian ontology. Finally, because Tarski is employing ‘class’ and ‘element’
in the standard sense, he cannot continue with Leśniewski’s acquisitive mereological
terminology, so ‘sum’ comes into play instead of ‘class’. His effort thus lies partly outside
Leśniewskian orthodoxy, unlike the contributions of Sobociński and Lejewski. Somewhat
later, in 1937, Tarski used a system of mereology closely equivalent to this one in order
to provide axiomatic backing for the English biologist Joseph Henry Woodger’s use of
mereology in his The Axiomatic Method in Biology, where it appears as Appendix E [15].
In a paper on extended Boolean algebras of 1935 [16], he noted the formal (we would now
say: model-theoretic) similarity between mereology and extended Boolean algebras, that
is, Boolean algebras with variable-binding meet and join (product and sum) operators.
This consisted solely in the absence of a null element in mereology.
10 The Harvard Variant
Outside Poland, and the associated use of mereology in Woodger, formal systems of part
and whole first emerged from Harvard, where Whitehead had been teaching since 1924.
It was in 1940 that Henry S. Leonard and Nelson Goodman, both of whom studied there,
published in The Journal of Symbolic Logic the paper that introduced formal part–whole
theory to the majority of anglophone readers: ‘The Calculus of Individuals and its Uses’
[17]. Its prehistory is however surprisingly convoluted (cf. [18]) and goes back to around
1928, when Nelson Goodman submitted a Harvard Honors Dissertation (now seemingly
lost) which made some use of part and whole. His friend Henry S. Leonard developed a
formal calculus using mereological concepts in his 1930 PhD dissertation Singular Terms,
written under Whitehead, and along with acknowledgement to his supervisor, Leonard
notes that he discussed the subject extensively with Goodman. In 1935 they were alerted
by Quine, who had met Leśniewski personally in 1933, that their work closely resembled
the latter’s mereology, of which at that time they had no knowledge. In 1936 they gave a
joint talk to the Association for Symbolic Logic which developed into the 1940 paper. A
year later, in 1941, Goodman defended his PhD thesis A Study of Qualities, written under
C. I. Lewis, which also used their joint system. When Goodman came to publish a heavily
revised version of the dissertation in 1951 under the title The Structure of Appearance, he
had eliminated the modest amount of set theory that he and Leonard had used and was,
like Leśniewski, resolutely opposed to set theory. We outline Leonard’s system (quoting
from [18], as it is unpublished) and that from the 1940 paper.
Leonard’s Singular Terms is conceived as a parallel for singular terms of several
sections of Principia Mathematica which deal with class theory: its formal sections are
numbered like those of PM, and its results track those of Whitehead and Russell. The
undefined primitive is the binary mereological sum operator, written ‘+’, and introduced
thus:
By “x + y” we mean to describe that individual which arises from the most general togetherness
of any two other individuals. (Quoted from [18].)
The nominal expression ‘x + y’ is taken as a descriptive function in the sense of

Whitehead and Russell, that is, a definite description of the form ιx φx. However, since
the symbol ‘+’ is primitive, the description cannot be analysed in the usual Russellian
way, so all this means is that it is meaningful, firstly, to predicate singular existence of
an expression ‘x + y’ in the form E!x + y, and secondly, to countenance the possibility
that compound expressions employing ‘+’ may sometimes be empty. In terms of ‘+’,
Leonard sets out numerous definitions, two of which are required for his statement of the
postulates:
Definitions
*16.01 x < y = Df. x + y = y
*16.02 x ◦ y = Df. ∃z z < x ∧ z < y
*16.03 xy = Df. ιz z < x ∧ z < y ∧ ∀t t < x ∧ t < y → t < z
The first of these defines the part relation, the second overlapping, and the third the
product or greatest common part of two individuals, which of course does not exist if they
do not overlap.
Postulates
*16.1 x + y = y + x
*16.12 E!x + x → x + x = x
*16.14 [E!x + y ∧ E!y + z ∧ E!(x + y) + z ∧ E!x + (y + z))] → x + (y + z) = (x + y) + z
*16.16 x ◦ y → E!xy
*16.17 x ◦ (y + z) → (x ◦ y ∨ x ◦ z)
*16.18 x < (y + z) → (x < y ∨ x < z ∨ x < xy + xz)
354 P. Simons
It follows from these that any two individuals have a binary sum: E!x + y, and
many other now familiar mereological theses also follow, but a general sum operator is
missing.
By the time of the 1940 joint paper, this lacuna had been plugged, helped by Quine’s
allusion to Leśniewski and the appearance of Woodger’s book with the Tarski appendix.
The 1940 system uses a modest amount of set theory—just sets of individuals, as in Tarski.
It is based on disjointness (discreteness, exteriority or non-overlapping) as primitive: that
x is disjoint from y we write x | y. The relevant definitions and axioms are
I.01 x < y = Df. ∀z z | y → z | x
I.02 x ◦ y = Df. ∃z z < x ∧ z < y
I.03 x Fu α = Df. ∀z z | x ↔ ∀y y ∈ α → z | y

I.1 ∃x x ∈ α → ∃y y Fu α
I.12 x<y∧y<x→x=z
I.13 x ◦ y ↔ ∼(x | y)
This is by no means a perspicuous axiom set, but the upshot is “ideologically” the same
as in Leśniewski. The two-place predicate ‘Fu’ is worth a brief eludication. It can be read
as ‘fuses’: the individual x fuses the class α when x is the sum of the elements of α. To
have a singular term for this fusion Leonard and Goodman can use a definite description,
ιxx Fu α
which denotes an individual just in case α is non-empty (by I.1) and whose uniqueness is
provable from the axioms and definitions.
11 Subsequent Developments
Here we have to be brief and incomplete, as the literature on mereology has exploded
in recent years. For a long time, during the hegemony of ordinary-language philosophy
through the 1940s–1970s, there was little interest in formal theories like mereology.
Things changed only gradually at first. Two pieces commenting on Goodman appeared in
the 1970s. Rolf Eberle’s Nominalistic Systems [19] was based on his UCLA dissertation
and looked at Goodman’s nominalism and its use of mereology. Eberle was interested in
building up to the full strength of Goodman’s mereology in stages. Alfred Breitkopf’s
1978 paper ‘Axiomatisierung einiger Begriffe aus Nelson Goodmans The Structure
of Appearance’ [20] gives a precise axiomatization of the system behind Goodman’s
use of mereology in that book, and applies it to constructive matters of interest to
Goodman: qualia, magnitudes of qualia, and categories. In 1979 David Bostock employed
a mereological basis slightly weaker than full Leśniewskian strength to underpin a logicist
analysis of the real numbers [21]. Peter Simons’s 1982 essay ‘The Formalisation of
Husserl’s Theory of Wholes and Parts’ [22] attempted for the first time to provide a
formal framework for discussing the non-extensional notions of founding, dependence

and the like which occur in combination with mereological concepts in Husserl’s 1901
investigation.
The first full monographic treatise on mereology was Simons’s Parts [7], which
appeared in 1987. Although the principal aim of that work was to show how mereology
could be used as a tool in a range of metaphysical problems, as a preamble to such
applications it contained a survey of mereology to that date and the first detailed
comparison of Leśniewskian systems with those of Whitehead and of Leonard and
Goodman. It developed a free logic-based mereology and also looked at extending
mereology into the temporal and modal domains, the latter allowing discussion of
mereological essentialism and of Husserl. In the light of criticisms of various mereological
theses, such as the existence of sums of any collection of objects, it proposed isolating a
core or minimal mereology which confined itself to analysing the concept of part and
extracting the analytic principles governing it, so that the strength of additional principles
may be more discerningly identified. To this end a set of core principles was presented
based on a free logic and the relation of proper part (") [7, p. 362]
Existence x " y → E!x ∧ E!y
Asymmetry x " y → ∼(y " x)
Transitivity x"y∧y"z→x"z
Supplementation x " y → ∃z z " y ∧ z | x
The suggestion was that anything less than this fails to capture formally the notion of
part, whereas anything more goes beyond this core and embodies substantive metaphysi-
cal hypotheses.
In the decades since then, mereology has developed in two directions. In formal
terms, much greater attention has been paid to the strengths of mereological postulates
lying between the analytic core and the full strength of systems of Leśniewski, Leonard
and Goodman. The most thorough formal treatment is in Lothar Ridder’s 2002 book
Mereologie [23], which is punctilious in its proofs and detailed in its critical examinations
of alternatives.
From a philosophical point of view, mereology has achieved hitherto unrivalled
penetration of analytic metaphysics, being itself a subject of metaphysical scrutiny as well
as a standard tool for use within ontology. Here we survey the principal areas of activity:
composition, class theory, persistence, mereotopology, and vagueness, but for reasons of
space cannot enter into associated controversies. A readable and comprehensive survey
of both the formal and philosophical issues may be found in Achille Varzi’s Stanford
Encyclopedia of Philosophy article ‘Mereology’ [24], which also contains an extensive
bibliography of recent writings.
The notion of composition is already present in mereology. It can take one of two
forms. An object X is simply composed of objects a if X is, or (in case of non-identical
objects with the same parts) mereologically coincides with, the mereological sum Sm(a).
An object X is disjointly composed of objects a if each a is disjoint from every other,
and they simply compose X, which is then Leśniewski’s disjoint sum Dsm(a). In his
book Material Beings of 1990 [25], Peter van Inwagen raised the question as to when
a collection of several objects a disjointly compose another object. He called this the
356 P. Simons
Special Composition Question. His surprising and controversial answer was: only when
X is an organism and a are all its atomic parts. This means that, strictly and philosophically
speaking, the only composite objects that exist are organisms, so artefacts and other
inanimate objects, and the non-atomic proper parts of organisms such as organs and
cells, do not exist. The proposal contradicts standard mereology following Leśniewski,
for which any collection of disjoint individuals has a sum which it accordingly disjointly
composes. This position is now often called compositional universalism. If that represents
one extreme, in the discussions following van Inwagen’s proposal another extreme
position has emerged, which is compositional nihilism, according to which disjoint
composition never occurs. That may take one of two forms. Either there are many
atoms, and nothing composed of them; or, the world is one huge atom, without proper
parts. The former assumes, controversially, the truth of atomism, the latter is a form of
monism. Attempts to find a stable and satisfying intermediate position more consonant
with common sense than any of these theories have to date not been crowned with
conspicuous success.
In 1991, David Lewis published his book Parts of Classes [26], in which he showed
that if standard mereology is applied to classes in the sense of von Neuman–Bernays–
Gödel set-and-class theory, with the subclass relation interpreted as part, then it yields
the same objects as that theory, provided one can find a way to generate singletons so
that a singleton {x} is always different from its sole element x. Allowing this, and adding
Lewis’s identification of the null class with the mereological sum of all urelements, the
construction works on those not unproblematic assumptions. In a sense it brings the
development of mereology as an alternative to set theory unexpectedly full circle back
to its origins, in a way which would no doubt have infuriated Leśniewski, since it leaves
in abeyance the ontological status of singletons and the null set, precisely the issues which
he identified as problematic in standard set theory.
In 1930, Leśniewski addressed the question whether the following three propositions are
consistent (cf. [1], pp. 465 ff.; [2], pp. 379 ff.):
The Warsaw of 1830 is smaller than the Warsaw of 1930
The Warsaw of 1830 is Warsaw
The Warsaw of 1930 is Warsaw.
He argued that, if ‘Warsaw’ is not an empty name, and these are all true, and ‘is’ is
governed by the axioms of ontology, then we can conclude the absurdity that
The Warsaw of 1930 is smaller than the Warsaw of 1930.
His way out of this was to treat Warsaw as a four-dimensional extended object,
Warsaw from the beginning to the end of its existence, of which the Warsaw of 1830
and the Warsaw of 1930 were different temporal parts of lesser temporal extent than
the whole, so that the second and third premises are false. Alternatively, if ‘Warsaw’
were a common name, the conclusion would not follow from the premises. Leśniewski
preferred the first alternative, and so espoused a view of objects in time that has come to be
called four-dimensionalism. His view is similar to those of other “hard-minded” logician–
metaphysicians such as Bolzano, Carnap, Quine, and Lewis. There is an extensive modern
debate about whether the account of change, as here adumbrated by Leśniewski, requires
ordinary objects to be considered extended in time as well as in space. The principles of
mereology are not in question in this debate, but the way in which mereological notions
are best applied to concrete objects certainly is.
Another area to which mereological concepts have been applied is the structure of
space and time and their relation to their occupants (cf. [27, 28]). One strand of this
discussion takes off from Whitehead’s treatment in Process and Reality of mereology
in conjunction with topological notions, a combination that has come to be called
‘mereotopology’. Another issue related to both this and the issue of the previous
paragraph is whether there is one or more than one way in which the occupants of
space and time occupy the regions they do occupy. Another is whether the parts of such
regions correspond to parts of their objects, or whether it some objects may lack proper
parts while occupying regions that do have proper parts, a view found in Whitehead
but interestingly presaged in the pre-critical Kant. Such objects are generally known as
extended simples.
An issue which for a long time remained in the background in ontology and in logic
is the question of vagueness, that is, predications which are neither clearly true nor
clearly false but which may occasion Sorites paradoxes. Standard bivalent logic rules
out such cases or regards them as arising from epistemological limitations. Leśniewski,
as a convinced adherent of bivalence, would no doubt have consigned the consideration
of vagueness to the extra-logical dustbin of inexact vernacular language. However, if
vagueness is taken seriously and not brushed under the carpet, it interacts in an interesting
and not well understood way with mereology. Of ordinary objects, whether things or
events, it is often unclear where they begin and end, or what their parts are, so there are
cases where, of the following three statements
A exists
B exists
A is a proper part of B
the first two are unproblematically true, while the third is neither clearly true nor clearly
false. This is despite the fact that the following mereological statement is clearly false:
B+A=B−A
whereas both
B=B+A
and
B=B−A
are neither clearly true nor clearly false. There are several competing theories of
vagueness, and here is not the place to review them. Suffice it is to say that the application
of these theories to mereology is an area which is at present relatively underdeveloped.
358 P. Simons
12 Prospects
Leśniewski’s own mereology was, barring simplifications and variations, essentially

completed no later than 1927, and he and his adherents did not deviate from it.
Whitehead’s mereology, developed for different purposes, was not published as a formal
system and has attracted less attention. The Leonard–Goodman development shared many
of Leśniewski’s philosophical views and their mereology is “ideologically” equivalent
to that of Leśniewski, differences of logical vehicle aside, so that this joint vision has
come to be called classical extensional mereology [7, p. 37]. More recent developments
have however questioned several of the assumptions behind this classical theory and
investigated systems which have been ever more finely distinguished. This has been
accompanied by hitherto unprecedented levels of use of mereology within philosophical
ontology, as well as its extra-philosophical employment in database representations. The
mereological breakdown of an object into its principal component parts is called, in
database parlance, a ‘meronomy’, by analogy with ‘taxonomy’. Classicists may rightly
wince at the term ‘partonomy’ sometimes used instead. Not least because of this increased
activity across several domains, and because of several unresolved questions in the
applications, it is to be expected that mereology will continue for some time to come
to provide material for philosophical controversy, and to remain an essential tool for
philosophers and an object of metalogical research by logicians and mathematicians.
References
1. Leśniewski, S.: Pisma zebrane [Collected writings]. In: Jadacki, J.J. (ed.) vol. 2. Semper, Warsaw
(2015)
2. Leśniewski, S.: Collected works. In: Surma, S.J. et al. (eds.) vol. 2. Kluwer/Polish Scientific
Publishers, Dordrecht/Warsaw (1992)
3. Leśniewski, S.: Lesniewski’s lecture notes in logic. In: Srzednicki, J.T.J., Stachniak, Z. (eds.) Kluwer,
Dordrecht (1988)
4. Sobociński, B.: L’analyse de l’antinomie Russellienne par Leśniewski. Methodos 1, 94–107, 220–
228, 308–316 (1949); 2, 237–257 (1950). English translation: Leśniewski’s Analysis of Russell’s
Paradox. In: Srzednicki, J.T.J., Rickey, V.F. (eds.) Leśniewski’s Systems: Ontology and Mereology,
pp. 11–44. Nijhoff/Ossolineum, The Hague/Wrocław (1984)
5. Leśniewski, S.: Czy klasa klas nie podporzadkowanych
˛ sobie jest podporzadkowana
˛ sobie? Przeglad
˛
Filozoficzny. 17, 63–75 (1914). Cf. [1, pp. 223–35, 2015]. Translation: Is the Class of Classes not
Subordinated to Themselves, Subordinated to Itself? in [2, pp. 115–28]
6. Leśniewski, S.: Podstawy ogólnej teorii mnogości. I [No further parts appeared]. Moscow: Popławski
(1916). (Prace Polskiego Koła Naukowego w Moskwie. Sekcya matematyczno-przyrodnicza, No. 2.)
Cf. [1, pp. 256–94]. Translation: Foundations of the General Theory of Sets. I, in [2, pp. 129–73]
7. Simons, P.: Parts. A Study in Ontology. Clarendon, Oxford (1987)
8. Whitehead, A.N.: La théorie relationniste de l’espace. Rev. Metaphys. Morale. 23, 423–454 (1916)
9. Whitehead, A.N.: An Enquiry Concerning the Principles of Natural Knowledge. Cambridge Univer-
sity Press, Cambridge (1919)
10. Simons, P.: Whitehead und die Mereologie. In: Hampe, M., Maaßen, H. (eds.) Die Gifford Lectures
und ihre Deutung. Materialien zu Whiteheads Prozess and Realität, vol. 2, pp. 369–388. Suhrkamp,
Frankfurt am Main (1991); English translation: Whitehead and mereology. In: Durand, G., Weber,
M. (eds.) Les principes de la connaissance naturelle d’Alfred North Whitehead, pp. 215–233. Ontos,
Frankfurt am Main (2007)
11. Whitehead, A.N.: Process and Reality. An Essay in Cosmology. Macmillan, New York/London
(1929). Corrected edition, Macmillan, New York/London (1978)
12. Lejewski, C.: Consistency of mereology. J. Symb. Log. 34, 321–328 (1969)
13. Tarski, A.: Les fondements de la géométrie du corps. Ksi˛ega Pamiatkowa˛ Pierwszego Polskiego
Zjazdu Matematycznego, pp. 29–33. Supplement to Annales de la Société Polonaise de Mathéma-
tique, Kraków (1929). Extended translation: Foundations of the geometry of solids, in [29, pp. 24–29]
14. Betti, A.: Leśniewski, tarski and the axioms of mereology. In: Mulligan, K. et al. (eds.) The History
and Philosophy of Polish Logic. Essays in Honour of Jan Woleński, pp. 242–258. Palgrave Macmillan,
Basingstoke (2013)
15. Tarski, A.: Appendix E. In: Woodger, J.H. (ed.) The Axiomatic Method in Biology, pp. 161–172.
Cambridge University Press, Cambridge (1937)
16. Tarski, A.: Zur Grundlegung der Booleschen Algebra I. Fund. Math. 24, 177–98 (1935). Translation:
On the Foundations of Boolean Algebra, in [14, pp. 320–41]
17. Leonard, H.S., Goodman, N.: The calculus of individuals and its uses. J. Symb. Log. 5, 45–54 (1940)
18. Rossberg, M.: Leonard, Goodman, and the development of the calculus of individuals. In: Ernst, G.,
et al. (eds.) From Logic to Art: Themes from Nelson Goodman, pp. 51–69. Ontos, Frankfurt am Main
(2009)
19. Eberle, R.: Nominalistic systems. Reidel, Dordrecht (1970)
20. Breitkopf, A.: Axiomatisierung einiger Begriffe aus Nelson Goodmans the structure of appearance.
Erkenntnis. 12, 229–247 (1978)
21. Bostock, D.: Logic and Arithmetic. Vol. 2. Rational and Irrational Numbers. Oxford, Clarendon
(1979)
22. Simons, P.: The formalisation of Husserl’s theory of wholes and parts. In: Smith, B. (ed.) Parts and
Moments, pp. 113–159. Philosophia, Munich (1982)
23. Ridder, L.: Mereologie. Ein Beitrag zur Ontologie und Erkenntnistheorie. Klostermann, Frankfurt am
Main (2002)
24. Varzi, A.: Mereology. In: Zalta, E.N. (ed.) The Stanford Encyclopedia of Philosophy. http://
plato.stanford.edu/archives/spr2016/entries/mereology (2016)
25. van Inwagen, P.: Material Beings. Cornell University Press, Ithaca (1990)
26. Lewis, D.K.: Parts of Classes. Blackwell, Oxford (1991)
27. Casati, R., Varzi, A.: Parts and Places. The Structures of Spatial Representation. MIT Press,
Cambridge (1999)
28. Kleinschmidt, S. (ed.): Mereology and Location. Oxford University Press, Oxford (2014)
29. Tarski, A.: Logic, Semantics, Metamathematics. Papers from 1923 to 1938. Translated by J. H.
Woodger, 2nd ed. edited and introduced by J. Corcoran. Hackett, Philadelphia (1983) (1st edn.
Oxford, Clarendon Press, 1956)
P. Simons ()
Department of Philosophy, Trinity College Dublin, Dublin, Ireland
Alfred Tarski (1901–1983)
Jan Woleński
Abstract This paper presents the life and work of Alfred Tarski, one of the most
distinguished and influential logicians in the entire history.
Keywords Logic · Semantics · Truth · Metamathematics
Alfred Tarski (AT hereafter) was born in Warszawa on January 14, 1901. His family
was Jewish—Teitelbaum (Polish spelling: Tajtelbaum) was the original name of AT;
Wacław, AT’s brother was born 2 years later. Their parents, Ignacy Teitelbaum and Rosa
Prussak, belonged to Jewish business families. Rosa’s family was rich and involved in
the textile industry in Łódź. The Teitelbaums were not particularly religious people,
although decisively keeping Jewish identity; especially, they celebrated Jewish holidays.
On the other hand, they wanted to be partially assimilated. The family lived outside
Jewish settlements in Warszawa and spoke Polish. Consequently, both boys attended
an elementary school with teaching in Polish or Russian (Warszawa was in the Russian
Empire at that time). However, Alfred and Wacław also went to cheder (a Jewish school)
where they studied Hebrew and Torah. Thus, AT was fairly familiar with Jewish culture
and tradition. After years, Czesław Miłosz, AT’s colleague in Berkeley, told the present
author that when he translated a fragment of Torah into Polish, he was convinced that he
did the first translation. However, AT explained to him that the entire Hebrew Bible was
much earlier (in the years 1883–1914) translated into Polish by Izaak Cylkow. This story
documents that AT was fairly versed in Jewish religious literature published in Poland.
On the other hand, his general attitude toward religion was quite flexible. As I already
noted, the Teitelbaums preserved Jewish tradition, but AT celebrated in his family home
Easter and Christmas (the main Catholic holidays) as well. It can be considered as his
early inclination toward Polish identification. This attitude did not save him against anti-
Semitism. He and his brother frequently heard offending anti-Jewish remarks of their
Tarski’s life is extensively described in [3] and [13]. According to the character of the present volume
I concentrate on the years 1901–1939, that is, Tarski’s life in Poland. His postwar curriculum vitae is
treated less extensively. I mention dates of birth and death in the case of Polish people only.

362 J. Woleński
contemporaries. These unpleasant experiences from the youth made persistent traces in
AT’s consciousness.
AT entered high school in Warszawa. He studied languages (Latin, German, French),
Polish literature, science, mathematics, religion and history. He ended high school with
very good results in 1918. The high school years of AT happened to be very stormy from
the political points of view. World War I began in 1914. German troops attacked Warszawa
very soon and Russian army had to step back. What was extremely important for young
Polish patriots (AT decisively belonged to them) consisted in great hopes that Poland
would recover its independence lost at the end of the eighteenth century. It happened in
the autumn of 1918, exactly on November 11. AT began his university study 4 weeks
before this date; his matriculation (a celebrated nomination as a student) occurred on
October 15. As far as the matter concerns his nationality, he considered himself as a
Pole, although he did not deny his Jewish origin. As many others Jews living in Poland,
AT considered full assimilation as the only possible defense against anti-Semitism. He
stressed his “Polonization” by preferring the form Tajtelbaum over Teitelbaum; the latter
looked too German for him. AT’s political views were close to socialism at that time.
AT began his university education as a student of biology. Due to the difficult
political situation in Poland (the problem of fighting for the borders of the country) the
University of Warsaw was closed just after AT’s matriculation. He was taken to army
and served in a unit doing military logistic work. AT returned to his studies in 1919,
but he decided to study mathematics. This change was related to his participation in a
course in logic conducted by Stanisław Leśniewski, who had just started his teaching
as the Professor of the Philosophy of Mathematics. Leśniewski informed his students
on one open problems in set theory, concerning the concept of ordered set, and AT
solved it. Leśniewski immediately suggested AT to switch from biology to mathematics,
particularly mathematical logic. The result achieved by AT was published in his first
scientific paper, published in Przeglad ˛ Filozoficzny (Philosophical Review) in 1921; the
subtitle of this work has the subtitle “From the seminar of prof. Stanisław Leśniewski in
the University of Warsaw”. It was the very beginning of AT’s spectacular scientific career
as one of the greatest logicians in the entire history of logic.
Entering into the territory of logic, AT was probably not aware that he would inscribe
himself to a fast growing superpower in mathematical logic and the foundations of
mathematics. How did it happen that a country without a specially strong tradition in
logic so quickly (during one generation in the years 1918–1939) became a real stronghold
in this field? In particular, this qualification concerns the Warsaw School of Logic. This
school was established by philosophers and mathematicians. In philosophy, everything
goes to Kazimierz Twardowski, a student of Brentano and the founder of the Lvov-
Warsaw School. Twardowski (1866–1938) was appointed professor of philosophy at
Lvov University in 1895. He wanted to introduce Brentano’s metaphilosophical program
in Poland. In particular, he demanded clarity language and thought and believed in
scientific philosophy. Following his teacher, Twardowski maintained that philosophical
method is (or can be) exactly the same as that executed in so-called special sciences.
Twardowski was not a logician and did not consider himself as such. On the other hand,
his metaphilosophical views formed a very friendly environment for logic sensu largo,
that is, covering formal logic (the term “mathematical logic” was rarely used at that
time), semantics and the methodology of science. Twardowski lectured on elementary
algebra of logic in the academic year 1899/1900; in fact, it was the first university course
Alfred Tarski (1901–1983) 363
on this topic in Poland (more precisely in the part of Poland belonging to the Austro-
Hungarian Empire) . Jan Łukasiewicz (1878–1956) participated in this class and very soon
became attracted by logic (originally, he studied law, but switched to philosophy under
Twardowski’s influence). Łukasiewicz began systematic courses in advanced algebra of
logic and other logical topics. He trained many young philosophers with explicit interests
in logic, including Kazimierz Ajdukiewicz (1890–1963), Tadeusz Czeżowski (1889–
1981), Tadeusz Kotarbiński (1886–1981) and Zygmunt Zawirski (1882–1948); all of them
also studied mathematics, mostly under Wacław Sierpiński (1882–1969), who acquainted
his students with set theory. Stanisław Leśniewski (1886–1939) joined this circle in 1910.
Although this group, the Lvov Collegium Philosophicum, as Leśniewski used to say,
cannot be regarded as a logical school, logic played a distinguished role in this circle
of scholars. Hence, Twardowski insisted that his students should know philosophical
novelties, for instance, logical works of Gottlob Frege and Bertrand Russell were well-
known in Lvov.
Warszawa appeared on the logical stage exactly in 1916, when the University of
Warsaw was reopened; it was closed in 1831 and functioned in 1862–1869 as the Warsaw
Main School. The academic staff was mainly imported from Lvov. Łukasiewicz was
appointed professor of philosophy. He began lectures in logic and attracted many young
mathematicians. When Poland recovered its independence in 1918, this also resulted in
a great debate about the tasks and prospects of Polish science and culture (in fact, these
discussions began about 1916). Scholars in every field discussed how to develop their dis-
ciplines in the new expected political situation and what should be done in order to catch
up with world science. Particularly important was the discussion among mathematicians.
In fact, it had already started in Lvov, but was rather as a private enterprise, involving
Sierpiński and Zygmunt Janiszewski (1888–1920). Both were disappointed by a lack of a
common language and interests among Polish mathematicians and both were convinced
that set theory and topology should play a fundamental role in mathematics. The national
discussion about science, its needs and perspectives, was a good occasion for manifesting
views about the future of mathematics in Poland. Janiszewski became the main exponent
of the project, later known as the Janiszewski program, which defined the ideological
fundament of the Polish Mathematical School.
Roughly speaking, according to Janiszewski, Polish mathematicians should concen-
trate on chosen mathematical fields and work in one strong circle. The second point was
very soon abandoned, but the first was adopted. Although Janiszewski did not mention
any concrete topic to be cultivated in Poland, most Polish mathematicians understood it
as favouring set theory, topology and their applications to other branches of mathematics.
Janiszewski also postulated that Poland should have a special mathematical journal
published in international languages. This idea found its realization in Fundamenta
Mathematicae (the first volume appeared in 1920). Janiszewski’s program attributed a
great role to mathematical logic and the foundations of mathematics. The placement of
logic and the foundations at the heart of mathematics required definite steps in the sphere
of organization. The University of Warsaw had the Faculty of Mathematical and Natural
Sciences. The Department of the Philosophy of Mathematics was very soon organized
and Leśniewski became its head; Łukasiewicz left the University in 1918 in order to
serve as the Minister of Religious Denominations and Education in the government
under Ignacy Paderewski. He returned to the academic staff in 1919 and the University
established for him a special position in philosophy at the Faculty of Mathematics and
364 J. Woleński
Natural Sciences. Both professors began intensively teaching mathematical logic, mostly
among mathematicians but also among philosophers; Ajdukiewicz taught at this faculty in
the years 1926–1927. The first project of Fundamenta Mathematicae divided the journal
into two series, one devoted to set theory, topology and their applications, and second
to logic and the foundations. This project was finally abandoned, but the significance of
mathematical logic in the eyes of the founders of the Polish mathematical school found
its impressive manifestation in the composition of the Editorial Board of Fundamenta:
Mazurkiewicz, Sierpiński, Leśniewski and Łukasiewicz. Logic was also popularized
among students of philosophy by Kotarbiński.
Generally speaking, the logical circle in Warszawa, known as the Warsaw School of
Logic, was a child of two movements, namely the Lvov-Warsaw Philosophical School
and the Polish Mathematical School. Both determined the scientific environment in
which AT grew as a logicians. In fact, he opened the list of young mathematicians
and philosophers attracted by logic in Warszawa. This group included (in alphabetical
order and covering the whole interwar period): Stanisław Jaśkowski (1906–1965), Adolf
Lindenbaum (1904–1941?), Andrzej Mostowski (1913–1975), Moses Presburger (1904?–
1943), Jerzy Słupecki (1904–1984), Bolesław Sobociński (1904–1980; a philosopher by
training)) and Mordechaj Wajsberg (1902–1942?). The names of three other of logicians
who graduated shortly before 1939 or studied during War World II and began their
academic work after 1945 should be added to this list, namely Jan Kalicki (1922–1953; a
mathematician), Czesław Lejewski (1913–2001; a classicist and philosopher) and Henryk
Hiż (1917; a philosopher).
AT studied mathematics at the University of Warsaw from 1919 to 1923. He attended
courses and seminars by (inter alia) Leśniewski (the foundations of mathematics), Ko-
tarbiński (logic), Sierpiński (set theory), Mazurkiewicz (analysis), Kazimierz Kuratowski
(topology) and Leon Petrażycki (sociology). AT also met his student fellows interested in
logic, namely Lindenbaum and Wajsberg. The former became his close friend and col-
laborator; Bronisław Knaster (1893–1990), a mathematician was another close friend of
AT. Working with Leśniewski, AT obtained important results concerning prothotetic (an
extended propositional calculus), one of three logical systems constructed by Leśniewski.
These results constituted AT’s doctoral dissertation supervised by Leśniewski, defended in
1924. AT was the only person who did a doctorate under Leśniewski and the latter used to
say that he had a 100% of genius doctoral students. AT’s PhD dissertation was published
in two papers which appeared in Fundamenta Mathematicae. In 1923, AT acted as the
secretary of the logic section of the 1st Polish Philosophical Congress in Lvov. On that
occasion he met Stefan Banach (1892–1945). At the same year they published (1892–
1945) a famous paper on the paradoxical decomposition of a ball. This result, called the
Banach-Tarski paradox shows some surprising consequences of the axiom of choice. In
1925, AT obtained his habilitation on the base of a dissertation on the concept of finite
set and became the youngest docent (a scholar who had veniam legendi and thereby could
lecture at university) in the entire history of mathematics in Poland. After his doctorate,
AT was mostly involved in set theory. His deep and numerous results (partly achieved
together with Lindenbaum) ensured him a distinguished place in the Polish Mathematical
School.
AT all the time was thinking how to strengthen his Polonization. He decided to change
his surname, following advices of Leśniewski and Łukasiewicz, AT and Wacław, his
brother (he studied law), adopted the name Tarski. Firstly, it was added to Tajtelbaum.
For instance, AT’s mentioned paper published in Przeglad ˛ Filozoficzny is signed by
Tajtelbaum-Tarski, but his PhD diploma is for Alfred Tarski. AT also converted to
Catholicism. These moves were motivated by his intention to be recognized as a Pole. In
1929, AT married Maria Witkowska; they had two children; Jan born in 1934, and Ina born
in 1938. Although changing name and converting to Catholicism helped to some extent,
AT (and Wacław as well) were stigmatized as Jews by Polish anti-Semitic activists. In the
late 1930s. a booklet appeared with a list of dangerous Jews, that is, such who changed
names and converted. Both Alfred and Wacław were included into this infamous register.
As far as the issue concerning AT’s style of life, he belonged to a group of friends who
met in cafes and used alcohol or even drugs; he was a heavy smoker until the end of his
life. One of AT’s friends deserves a special attention. It was Stanisław Ignacy Witkiewicz
(1885–1939), called Witkacy. He was a writer, painter, philosopher and a very eccentric
person. Witkacy made portraits of Alfred and Maria. The head of the former is presented
inside spikes. A very accurate picture. AT and Maria frequently went to Zakopane, a very
popular resort in the Tatra mountains, where they climbed.
The professional career of AT was not easy. Having veniam legendi; he could lecture
and conduct seminars at university. He delivered many advanced courses, for instance, on
set theory, methodology of deductive sciences, geometry or arithmetic of natural numbers,
in order to mention a few. In 1929, he officially became an assistant of Łukasiewicz
in the Department of Philosophy existing at the Faculty of Mathematics and Natural
Sciences (this chair was especially established for Łukasiewicz). AT, in order to earn
money needed for his family had to find a job outside the university. He worked as a
teacher of mathematics in the Żeromski Secondary School in Warszawa and the National
Pedagogical Institute, which organized training for teachers of mathematics; he had
to resign from the latter for his Jewish origin. Perhaps AT’s teaching of elementary
geometry should be especially mentioned, because it resulted in his involvement into
writing textbooks for schools and inventing problems. AT also published a textbook on
mathematical logic (see [27]) . This small book was written for students of secondary
schools, particularly interested in logic. It was translated into German in 1937 and, in
a revised and extended version, into English in 1941 (see [28]), and reprinted many
times (translations into other languages were published as well). When I told one of my
American colleagues that the book in question was written for secondary schools, he
replied with a surprise that it is too difficult for most students of American universities.
In a guide for students of mathematics in Warsaw University published in 1926 after a
list of textbooks of elementary logic, we find information that the 1st volume of Principia
Mathematica is recommended of advanced students. These facts illustrate how logic was
taught in Poland, particularly in Warszawa in the interwar period.
AT’s academic ambitions went further than to occupy the position of a docent or
assistant. He intended to be a university professor. The first opportunity appeared in
1928, when the Lvov University decided to establish the professorship in mathematical
logic at the Faculty of Mathematics and Natural Sciences. Leon Chwistek (1884–
1944) and AT were competitors. The former had a strong support of mathematicians,
also Banach, who was in a difficult personal situation due to his mentioned work
with AT. Philosophers, particularly Twardowski, acting in the name of Leśniewski and
Łukasiewicz, preferred AT. The University invited Luitzen Brouwer, David Hilbert and
366 J. Woleński
Bertrand Russell as referees. These names show how serious was this competition. It is
not surprising for the great prestige of logic in the interwar period. In fact, Poland had five
professorships in logic in the years 1918–1939. How many were there outside Poland?
The answer is surprising: just one, in Münster, in Germany.
Chwistek won this rivalry. AT was convinced that his Jewishness played the decisive
role in this issue. However, one should be very careful in evaluation of what was going
on in the competition in question. Doubtless, Tarski’s Jewish origin did not help. In fact,
not many Jews became full professors in the interwar Poland, but some succeeded; for
instance Hugo Steinhaus (1887–1972), a distinguished mathematician and very powerful
professor in Lvov. Since his sister married Chwistek, Steinhaus’ support for AT’s rival
was natural. On the other hand, Jewishness of Mrs. Chwistek did not do a favor to her
husband, because anti-Semites did not like persons with Jewish spouses. Importantly,
Chwistek was older, better known at that time and supported by Jagiellonian University
in Kraków. According to Polish academic rules, universities had a right to opt for
candidates for professorships. In his case, Warsaw University supported AT, but Cracow
University voted for Chwistek. Last but not least, Russell wrote a well-know letter in
which he recommended Chwistek (opinions of Brouwer and Hilbert are unknown) .
Russell explicitly said that since he knew Chwistek and his work, the choice of Chwistek
would be a good decision. However, he added that his recommendation was not based
on a comparison of both rivals, because “The work of Mr. Tarski I do not at the moment
remember and do not have access to at present”. Ironically, Russell quoted AT’s papers in
the second edition of Principia Mathematica. According to recollections of some people
(I heard this story from Hiż) , the second opportunity for AT to be appointed as professor
appeared in Poznań in 1937, when Zawirski moved to Kraków; but Poznan University
cancelled the professorship in logic, apparently to block Tarski as a Jew. This affair is not
testified by existing documents, but if it actually happened, anti-Semtic attitude, strong in
Poznan, played the decisive role in rejecting Tarski. Tarski as a docent could not supervise
doctorates. However, he was the doctor father of Mostowski (1938)—Kuratowski acted as
the official supervisor. There is also a very surprising story about Presburger. He proved
the completeness of so-called Presburger arithmetic (theory of natural numbers with
addition as the sole operation). Presburger asked AT whether this theorem is sufficient
for obtaining the doctoral degree. The answer was entirely negative—AT considered this
result as too trivial. He made a mistake, because Presburger’s result is presently considered
as a very serious achievement.
Tarski intensively worked in the 1930s. He continued his work in set theory, but was
more and more involved in logic and metamathematics. In a series of papers, he defined
several notions used by logicians (and mathematicians as well) rather in an intuitive
way. For instance, he axiomatized the concept of logical consequence and deductive
system. His most important work concerned the concept of truth. In 1930, AT delivered
a talk about the concept of truth in deductive science. His famous monograph Poj˛ecie
prawdy w j˛eykach nauk dedukcyjnych (The Concept of Truth in Languages of Deductive
Sciences) appeared in 1933 (see [25]). Its German translation Der Wahrheitsbegriff in den
formalisierte Sprachen (The Concept of Truth in Formalized Languages) was published
in 1935 (see [26]) and English version in 1956 (see [29]). Since this idea is extensively
presented in another paper in this volume, I only mention it in this place. At the moment
let me add that the semantic definition of truth (it is, so to speak, the official label for AT’s
approach to truth) is (a) the most important result of AT; (b) one of the most important
achievements in analytic philosophy; (c) the idea which originated model theory as one
of the most important parts of contemporary logic; (d) the most important achievement in
the entire history of Polish philosophy. Even if someone will say that there are no clear
criteria of what belong to the most important philosophical achievements, sociological
measures support the opinion expressed in (b) and (d).
In the interwar period, AT active participated in scientific life, mathematical and
philosophical, in Poland, and on the international scale, in particular in all Polish
Philosophical Congresses (Lvov—1923, Warszawa—1927, Kraków—1936), in many
mathematical congresses (for example, the eight Mathematical Congress in Bologna,
the one Congress of Mathematicians of Slavic countries), in the eight International
Philosophical Congresses (Prague 1934) and in the 1st Congress for Scientific Philosophy
(Paris 1935). AT’s participation in the last event was remarkable. He, invited by Karl
Menger, visited Vienna in 1930. Menger introduced AT to many philosophers, directly
or indirectly associated with the Vienna Circle. In particular, AT met Rudolf Carnap and
Kurt Gödel for the first time. In the next years, AT visited Vienna several times and had
discussions with Viennese philosophers, particularly with Karl Popper. AT’s reported his
semantic ideas in discussions in Vienna. As it is known, the Vienna Circle was skeptical
about semantics and its significance for philosophy. AT convinced Carnap and Popper (he
was not a member of the Circle) to semantics. Carnap insisted that AT should deliver a
talk on semantics at the Paris Congress in 1935. Tarski agreed, although he expected a
criticism of semantics, and delivered one lecture on the foundations of semantics and the
second on the concept of logical consequence. These talks were recognized as the most
important scientific events of the Congress. The other paper about semantics was read by
Maria Kokoszyńska (1905–1981), a close friend of AT. All these contributions caused a
very hot discussion. The Poles (or the Polish camp as it was called) and Carnap definitely
defended the role of semantics in philosophy, but Otto Neurath radically opposed to using
semantic tools in philosophical analysis. He was afraid that semantics could introduce bad
metaphysics into philosophy. The controversy over semantics was continued in the next
years. Although Neurath (he died in 1945) never accepted semantics as a legitimate part
of philosophy, the Polish camp finally won. Thus, AT became one of the most influential
thinkers in contemporary philosophy. Even if someone were to say that contemporary
anti-realistic semantics rejected the semantic theory of truth, it is still semantics.
In the 1930s several philosophers visited Poland, for instance, Carnap. Joseph
Woodger, Willard van Orman Quine, and Heinrich Scholz. They appreciated very
positively the philosophical situation in the country, particularly AT and his achievements.
Scholz said once that he was surprised that AT had not been promoted to the professor
position. A new possibility appeared after Leśniewski’s death in May 1939. AT hoped to
be his successor. Yet the political situation in Europe became more and more dangerous.
Quine urged that AT should leave Poland. The Congress for Scientific Philosophy to be
organized at the Harvard University, created a good opportunity. AT had mixed feelings.
On the one hand, he liked to participate in scientific events, also because his ambitions
were satisfied by his position as a great star in logic; but on the second hand, he wanted
to be present in Warszawa, when the succession after Leśniewski would be decided; it
was expected to happen in the autumn of 1939. Finally, he decided to go to USA. On
August 11, AT embarked on a ship sailing to America. He took only a small suitcase, as
368 J. Woleński
he planned to return very soon. He landed in New York on August 24. The war began
exactly one week later. Maria Tarski with two children remained in Poland. She left
Warszawa and survived. Maria was Aryan, but Jan and Ina, the children, satisfied the
condition of Jewishness introduced by the Nürnberg statutes. AT’s parents were murdered
by the Nazis in Auschwitz, and Wacław, his brother was killed in Warszawa just before
the end of the war. If AT were to remain in Poland, his fate would probably be tragic. One
of the chapters in the Feferman’s book (see [3]) has “How the “Unity of Science” Saved
Tarski’s Life” as its title. Very proper, indeed. Several Polish logicians of Jewish origin,
including Lindenbaum and Wajsberg, perished in the Holocaust.
AT participated in the Harvard Congress. However, the question “What then?” was
urgent. Quine arranged a research appointment at Harvard for AT. He temporarily lectured
at Columbia University for undergraduate students of Ernest Nagel. Then, he became
a visiting professor at the City University of New York. Russell tried to help AT in
getting a permanent position at Columbia, but these attempts were unsuccessful. Some
support came from the Young Men’s Hebrew Association in New York. The Guggenheim
Fund gave a fellowship for AT; he used it to stay in Princeton where he met Gödel once
again. Important things happened at Harvard in 1940–1941. A discussion group of logic
was formed, including Carnap. Quine, AT and Russell as main participants (see [4]).
Moreover, AT gave several talks in many places in the USA from the East Coast to the
Midwest. Besides troubles with getting a permanent job, AT was terribly worried about
his family in Poland. He was, of course, fully conscious of the danger for the life of
Maria and children. From time to time, he had indirect news obtained with the help of
Father Józef M. Bocheński (1902–1995) and Anders Wedberg. The problems with job
ended in 1942, when the University of California at Berkeley offered a position for AT
as a lecturer. AT became a USA citizen in 1945. In the early 1945, he received a letter
from Maria informing him that she and the children survived. They arrived to Berkeley
on January 6, 1946. This day ends AT’s Odyssey from Warsaw to Berkeley.
AT became the full professor at the age of 45 in 1946. Perhaps it should be noted
in the Guiness book as a world record in the category of eminent scientists waiting
for the professorship. He was elected to the National Academy of Sciences, the Royal
Netherlands Academy of Arts and Sciences and the British Academy, served as President
of the Association of Symbolic Logic (1944–1946) and the International Union for the
History and Philosophy of Science in 1956–1957; initiated the International Congresses of
Logic, Methodology and Philosophy of Science (the 1st Congress took place in Stanford
in 1960). He received honorary degrees from the Catholic University of Chile (1975),
the University of Marseilles (1977) and the University of California (1982). He travelled
all over the world and, as before the war, participated in many scientific congresses and
conferences. He visited once Poland; it happened in 1959. He had a permanent sentiment
to his first homeland, although he considered himself as American after 1945. When
a conference was organized to celebrate AT’s one hundredth anniversary in Warszawa,
Leon Henkin wrote to me that Alfred’s heart remained in this city for ever. The family
spoke Polish in the daily life and his house in Berkeley was always opened to guests
from Poland. AT, remembering his youth, was always very sensitive to all signs of anti-
Semitism. He rejected an invitation to Poland in 1968, protesting against anti-Semitic
campaign executed by Polish authorities at that time. On the other hand, AT supported
the “Solidarity” movement in 1981–1982 and offered a considerable amount of money
for the Kościuszko Fund, giving scholarships for Polish scientists. He maintained close
contacts with Polish friends. AT’s famous collection Logic, Semantics, Metamathematics
(see [29]), published in 1956 has the dedication “To His Teacher Tadeusz Kotarbiński—
the Author”. AT had many distinguished teachers, but he decided to celebrate just
Kotarbiński, as a pattern of humanity. And AT changed his earlier socialist political views
to more social-democratic. Alfred Tarski died in Berkeley on October 26, 1983.
AT was a great teacher. Although he was very demanding and not always nice to
his students, he attracted many people to logic. As a result, he supervised 24 doctoral
dissertation. The list of his students includes (in the chronological order): Bjarni Jonsson,
Louise Chin Lim, Julia Robinson, Wanda Szmielew, Frederick Thompson, Anne Morel,
Robert Vaught, Cheng-Chung Chang, Solomon Feferman, Richard Montague, Jerome
Keisler, Donald Monk, Haim Gaifman, William Hanf, John Doner, Robert Bratford,
Haragauri Gupta, Donald Pigozzi, George McNaulty, Charles Martin, Roger Maddux,
Benjamin Wells and Kan Ng. AT also directly influenced the work of such logicians as
John MacKinsey, Dana Scott and Steven Givant. He created the California School of
Logic, the most powerful logical circle in the USA. The number of mathematicians and
philosophers indirectly influenced by AT and his ideas is enormous. His writings appeared
at least in the following languages: Bulgarian, Czech, Dutch, English, French, German,
Georgian, Hebrew, Hungarian, Italian, Polish, Portuguese, Romanian, Russian, Serbian-
Croatian, Spanish and Swedish.
AT research in his American period is complex. Perhaps his contributions to the
model theory are most important. AT and his students developed so-called Western
(Californian) model theory. Its main idea consists in investigations of the relation between
languages and mathematical structures. The former are regarded as the starting point.
By contrast, the Eastern model theory, developed by Abraham Robinson, generalizes
algebraic concepts. Roughly speaking, the former approach is more logical, but the
latter—more mathematical, although these differences are rather vague at the present.
Generally speaking, AT continued the ideology of the Warsaw School of Logic. Although
he considered mathematical logic as a part of mathematics, he considered it as a relatively
autonomous. In particular, he did not agree that logic is servant of mathematics. The most
important feature of his approach to logic consisted in admission of all mathematically
accepted methods, regardless whether they were constructive or not. For instance, AT had
no scruples in using the axiom of choice. This attitude was very characteristic for the
Polish Mathematical School. More specifically, philosophical controversies around this
axiom are entirely independent of its mathematical applications. On the other hand, there
is tension between a free use of infinistic methods by AT and his explicit predilections
toward nominalism as a philosophical position rejecting abstract objects; he also accepted
empiricism in epistemology. AT, asked how he reconciled his private philosophy with
his mathematical practice, answered that he felt like a tortured nominalist. AT added that
there exist various tale-stories, set theory belongs to this variety, but it is very useful. In his
early years, AT understood logic relatively widely as covering set theory, but, according
to his later views, he favored first-order, logic as the logic. As far as the issue concerns
the plurality of logical systems, although AT obtained important results in non-classical
logics, particularly, many-valued and intuitionistic, he preferred the classical system as
the logic. On the other hand, he was ready to investigate every logical system provided
that such a research was interesting from the mathematical point of view.
370 J. Woleński
Bibliography
Selected Books on Tarski, His Background and His Ideas
1. Brożek, A., Chybińska, A., Jadacki, J.J., Woleński, J. (eds.): Tradition of the Lvov-Warsaw School.
Ideas and Continuations. Brill/Rodopi, Leiden/Boston (2015)
2. Coniglione, F., Poli, R., Woleński, J. (eds.): Polish Scientific Philosophy: The Lvov-Warsaw School.
Rodopi, Amsterdam (1993)
3. Feferman, A., Feferman, S.: Alfred Tarski: Life and Logic. Cambridge University Press, Cambridge
(2004)
4. Frost-Arnold, G.: Carnap, Tarski and Quine at Harvard. Conversations on Logic, Mathematics and
Science. Open Court, Chcicago (2013)
5. Gruber, M.: Alfred Tarski and the “Concept of Truth in Formalized Languages”. A Running
Commentary with Consideration of the Polish Original and the German Translation. Spinger,
Heidelberg (2016)
6. Horsten, L.: The Tarskian Turn. Deflationism and Axiomatic Truth. The MIT Press, Cambridge (2011)
7. Jadacki, J.J.: Polish Analytical Philosophy: Studeis on Its Heritage, with the Appendix Containing
the Bibliography of Polish Logic from the Second Half of the 14th Century to the 1st Half of the 20th
Century. Semper, Warszawa (2009)
8. Jadacki, J.J. (ed.): Alfred Tarski: dedukcja i semantyka [Déduction et sémantique] [Deduction and
Semantics]. Semper, Warszawa (2003)
9. Kijania-Placek, K., Woleński, J. (eds.): Lvov-Warsaw School and Contemporary Philosophy. Kluwer
10. Kuratowski, K.: Half Century of Polish Mathematics: Remembrances and Reflections. Pergamon
Press, London (1980)
11. Lapointe, S., Marion, M., Miśkiewicz, W., Woleński, J.: The Golden Age of Polish Philosophy:
Kazimierz Twardowski’s Philosophical Legacy. Springer, Berlin (2009)
12. López-Escobar, E.G.K., Miraglia, F.: Definitions: The Primitive Concept of Logics or the Leśniewski-
Tarski Legacy. Instytut Matematyczny PAN, Warszawa (2002)
13. McFarland, A., McFarland, J., Smith, T.J.: Alfred Tarski. Early Work in Poland – Geometry and
Teaching. Birkhäuser, New York (2014)
Birkhäuser, Basel (2014)
15. Pantsar, M.: Truth, Proof and Gödelian Arguments: A Defence of Tarskian Truth in Mathematics. The
University of Helsinki, Helsinki (2009)
16. Patterson, D.: Alfred Tarski. Philosophy of Language and Logic. Palgrave Macmillan, New York
(2012)
17. Patterson, D. (ed.): New Essays on Tarski and Philosophy. Oxford University Press, Oxford (2008)
18. Rojszczak, A.: From the Act of Judging to the Sentence. The Problem of Truth Bearers from Bolzano
to Tarski. Springer, Dordrecht (2005)
19. Skolimowski, J.: Polish Analytical Philosophy. A Survey and a Comparison with British Analytical
Philosophy. Routledge and Kegan Paul, London (1967)
20. Szaniawski, K. (ed.): The Lvov-Warsaw School and the Vienna Circle. Kluwer Academic Publishers,
Dordrecht (1989)
21. Twardowski, B., Woleński, J. (eds.): Sixty years of Tarski’s Definition of Truth. Philed, Kraków
(1994)
22. Wagon, S.: The Banach-Tarski Paradox. Cambridge Univerity Press, Cambridge (1985)
Dordrecht (1989)
24. Woleński, J., Köhler, E. (eds.): Aldred Tarski and the Vienna Circle. Austro-Polish Connections in
Logical Empiricism. Kluwer Academic Publishers, Dodrecht (1999)
(The most complete bibliography works on Tarski is in McFarland, A., McFarland, J., Smith T. J;
Alfred Tarski. Early Work in Poland – Geometry and Teaching (see above), pp. 398–471)
Selected Works of Tarski (Only Books)
25. Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych [The Concept of Truth in Languages of Deductive
Sciences]. Towarzystwo Naukowe Warszawskie, Warszawa (1933)
26. Der Wahrheitsbegriff in den formalisierten Sprachen. Studia Philosophica I, 261–405 (1935)
27. O logice matematycznej i metodzie dedukcyjnej [On Mathematical Logic and Deductive Method].
Ksiażnica-Atlas,
˛ Lwów (1936)
28. Introduction to Logic and to the Methodology of Deductive Sciences. Oxford University Press,
Oxford (1941). (4th ed. (1994))
29. Logic, Semantics, Metaamathematics. Papers from 1923 to 1938. Oxford University Press, Oxford
(1956). (2nd edn., Hackett Publishing Company, Indianapolis (1983). Includes The Concept of Truth
in Formalized Languages, 152–278)
30. Collected Papers, vols. 1–4. Birhäuser Verlag, Basel (1986)
(The most complete bibliography of Tarski is collected by S. Givant: Bibliography of Alfred Tarski.
The Journal of Symbolic Logic 41, 913–941 (1986))
J. Woleński ()
University of Information, Technology and Management Rzeszów, Rzeszów, Poland
Some Philosophical Aspects of Semantic
Theory of Truth
Jan Woleński
Abstract The semantic theory of truth, formulated by Alfred Tarski in the 1939s,
is primarily a mathematical theory. On the other hand, it also has a considerable
philosophical content. This paper presents the second aspect of this theory. It can be
shown that several traditional philosophical issues pertaining to the concept of truth
can be illuminated by Tarski’s account of truth. It concerns, for instance, the idea of
correspondence, the relation of truth and logic, the problem of the relativity/absoluteness
of truth, etc.
Keywords Satisfaction · Model · Truth-bearers · Truth-criterion · Correspondence
The semantic theory of truth (STT), developed by Alfred Tarski, has two separate but
interconnected aspects.1 Firstly, STT is a formal mathematical theory of a central concept
of model theory, one of the most important branch of mathematical logic. Secondly,
STT is also a philosophical doctrine that elaborates the notion of truth investigated by
philosophers since antiquity. As the title indicates, this paper focuses on the second issue,
that is, STT as a philosophical theory. Due to their significance for philosophical analysis
of the concept of truth, some formal questions must be taken into account too.
However, the fate of STT as a mathematical theory and as a philosophical doctrine is
different. Consider the following prophecy [12, p. 135]:
[. . . ] you should ask yourself what your grandsons and granddaughters are likely to study when
they settle down to their ‘Logic for computing class’ at 9.30 after school assembly. Will it be
syllogisms? Just possibly it could be the difference between saturated objects and unsaturated
concepts, though I doubt it. I put my money on Tarski’s definition of truth for formalized
languages. It has already reached the universal textbooks of logic programming, and another 10
years should see if safely into the sixth forms. This is a measure of how far Tarski has influenced
the whole framework of logic.
1I prefer the label “semantic theory of truth” over “semantic definition of truth” or “semantic conception
of truth”, though I will use, mostly for stylistic reasons, the second name too. First of all, STT cannot be
reduced to a definition, because it consists of a statement defining truth and many other assertion. On the
other hand, the name “semantic conception of truth” is too vague.

374 J. Woleński
Clearly, Wilfrid Hodges in the quoted fragment says about STT as a mathematical
theory. Independently whether he is right or not in his prophecy concerning the logical
education of our grandsons and granddaughters, Tarski’s truth-definition is en vogue
amongst mathematical logicians and almost nobody denies its importance as an idea
belonging to mathematical logic. If one wonders why I say “almost nobody”, I recall what
Alan Turing said once about STT: “Triviality can go no further” (see [31, p. 144]). Hao
Wang grounds on this fact the following opinion: “There is a great difference of opinion
on the importance of [Tarski’s] contribution to this area [that is, the theory of truth—J.
W.]” (ibidem, p. 144).2 However, it is rather Wang’s personal evaluation of the situation
(I do not enter into his motives), and the real measure of the importance of Tarski’s work
as a mathematical enterprise is closer to Hodges’ statement.
The fate of STT as a piece of philosophy is much more complex and there is really
“a great difference of opinion on the importance of [Tarski’s] contribution.” It is, of
course, no surprise, because being controversial is the mode of existence in the case of
all interesting philosophical proposals, Tarski himself was convinced (at least, when he
published his results about truth) that he contributed to philosophy [24, pp. 266–267]3:
Its [that is, 25—J. W.] central problem—the construction of the definition of true sentence and
establishing the scientific foundations of the theory of truth—belongs to the theory of knowledge
and forms one of the chief problems in this branch of philosophy. I therefore hope that this work
will interest the student of the theory of knowledge above all that he will be able to analyse the
results contained in it critically and to judge their value for further researches in this field, without
allowing himself to be discouraged by the appearance of concepts methods used here, which in
places have been difficult and have not hitherto been used in the field in which he works.
The moral to be derived from the end of the above passage is that according to
Tarski, formal (mathematical aspects) aspects of STT are indispensable for its proper
comprehension (see also later Tarski’s papers, namely [28] and [31]).
Tarski’s hopes about possible interests of philosophers in STT were correct. Most
philosophers, who oriented philosophy toward logic and used logical tools in philosoph-
ical investigations, welcomed his ideas; In fact, it is difficult to find today a serious
monograph concerning the concept of truth which would not refer to Tarski’s truth-
definition. The philosophical significance of the semantics theory of truth was recognized
very soon. According to Alfred Ayer [2, p. 116]
Philosophically the highlight of the Congress [of the Scientific Philosophy in Paris in 1935—J.
W.] was the presentation by Tarski of a paper which summarized his theory of truth.
Note that Ayer did not belong to the protagonists of the semantic definitions of truth.
Three important philosophers, namely Kazimierz Ajdukiewicz, Rudolf Carnap and
Karl Popper, radically changed their earlier views under Tarski’s influence (see [33]).
Ajdukiewicz abandoned radical conventionalism which was among others a theory of
language and meaning [1, p. 315]:
2 Jean-Yves Girard is another critic of STT as a mathematical theory. See [8], section 2.3 of this book
has “Tarksism” as its title and contains rather nasty comments on Tarski’s on pp. 36–37, 213, 491–496,
499–500. In fact, Girard’s criticism is also directed against philosophy behind STT.
3 Page-references are to translations or reprints, if they are mentioned in References at the end of the
present paper.
Some Philosophical Aspects of Semantic Theory of Truth 375
The objection [. . . ] communicated to me by Tarski in a conversation [. . . ] seems to show that the

concept of meaning is not definable in purely syntactical terms without the use of semantic terms
in the narrower sense.
A similar point was made by Carnap [4, p. X]:

Tarski, both through his book, and in conversation, first called my attention to the fact that the
formal method of syntax and semantics must be supplemented by semantical concepts, showing
at the same time that these concepts can be defined by means not least exact than those of syntax.
Thus the present book owes very much to Tarski, more indeed than to any other single influence.
Briefly, Carnap, under Tarski’s influence (or better, above all under Tarski’s influence)
passed from philosophy as logical syntax to philosophy as exact semantic analysis. There
is no exaggeration, if we say that Tarski essentially contributed to semantic revolution in
philosophy (see [34]).
Popper recalls [18, p. 322]:
[. . . ] I met Tarski in July 1934 in Prague. It was early in 1935 that I met him again in Vienna in
Karl Menger’s Colloquium [. . . ] It was in those days that I asked Tarski to explain me his theory
of truth, and he did so in a lecture of perhaps 20 min on a bench (un unforgotten bench) in the
Volksgarten in Vienna. He also allowed me to see the sequence of proofs sheets of the German
translation of his great paper on the concept of truth, which were than just begin sent to him from
[. . . ] Studia Philosophica. No words can describe how much I learned from all this, and no words
can express my gratitude for it. Although Tarski was only a little older than I, and although we
were, in those days, on terms of considerable intimacy, I looked upon him as the one man whom I
could truly regard as my teacher in philosophy, I have never learn so much from anybody else.
How Tarski’s ideas influenced Popper? Generally speaking, Popper began to defend
realism in his approach to science, because he came to the conclusion that STT renewed
the classical idea that truth consisted in the conformity between propositions and the
objective reality. Thus, Tarski essentially contributed to the development of scientific
realism.
These three examples together with Ayer’s general assessment are perhaps the most
spectacular single traces of the Tarski’s influence on philosophy consisting in the full
acceptance of his ideas. However, the philosophical role of STT is by no means
limited to such measures. Almost every book (introductory or advanced) in semantics,
philosophy of language or the history of analytic philosophy mentions this theory.
Almost every discussion of such topics as the definition of meaning, semantic realism
or scientific realism uses Tarski’s ideas or at least alludes to them. Several important
views in contemporary philosophy employ STT, for example, Donald Davidson’s theory
of meaning (see [6]) as based on truth-conditions or various semantic theories of induction
(Carnap and his followers). Tarski’s theory was more or less modified, like in [14] (Saul
Kripke) or [9] (Anil Gupta, Nuel Belnap or replaced by other constructions as in [11]
(Jaakko Hintikka); both modifications and replacements refer to STT as the solid starting
point. There is no exaggeration in the statement that every post-Tarskian theory of truth
(at least in analytic philosophy) is propter-Tarskian.
STT is also strongly criticized. Of course, it is not surprising that most non-analytic
philosophers simply ignore the semantic definition of truth. Others regard it as a typical
degeneration of the logical or computational (whatever it means) mind. I will not
comment on such criticisms. However, I would like to explain why a discussion between
philosophers belonging to various metaphilosophical camps is a very delicate matter. The
main problem is that metaphilosophical options largely decide about substantial solutions.
376 J. Woleński
Thus, if someone says that truth is entirely outside logic or semantics and its problem
must be located in philosophical anthropology, there is a very little chance for a fruitful
discussion between such controversies and a philosopher who believes in philosophy
based on logical analysis. As a logical philosopher I do not say that other philosophy
is wrong and has no value; I only indicate that, perhaps except explaining fundamental
misunderstandings and disagreements, there is not very much to discuss. A consequence
of this attitude which I regard as rational is this: it sounds as a restriction of criticism of
STT to those? which arose within the analytical camp.
Returning to the past, Carnap with an astonishment noted [5, p. 61]:
When I met Tarski again in Vienna in the spring of 1935, I urged him to deliver a paper on
semantics and on his definition of truth at the International Congress for Scientific Philosophy
to be held in Paris in September. I told him that all those interested in scientific philosophy and
the analysis of language would welcome this new instrument with enthusiasm, and would be eager
to apply it in their own philosophical work. But Tarski was very sceptical. He thought that most
philosophers, even those working in modern logic, would be not only indifferent, but hostile to the
explication of the concept of truth. I promised to emphasize the importance of semantics in my
paper and in the discussion at the Congress and he agreed to present the suggested paper.
At the Congress it became clear [. . . ] that Tarski’s sceptical predictions had been right. To my
surprise, there was vehement opposition to even on the side of our philosophical friends. [. . . ].
Neurath believed that the semantical concept of truth could not be reconciled with a strictly
empirical and anti-metaphysical point of view. Similar objections were raised in later publications
by Felix Kaufmann and Hans Reichenbach.
The point is that objections raised by the enemies of STT belonging to logical
empiricism denied any philosophical significance of it. The criticism noted by Carnap
assumed a very concrete philosophical basis, namely rather radical logical empiricism as
far as the matter concerned the empirical basis of knowledge. Moreover, Neurath argued
that the semantic account of truth reintroduces a very bad metaphysics into philosophy.
However, several authors argued later for the philosophical sterility of STT from a
quite general perspective. For instance, Max Black (see [3]) tried to show that Tarski’s
theory of truth, although correct from the purely logical point of view, is neutral in fact
with respect to old philosophical controversies over the concept of truth. A very radical
criticism against STT was raised by Hilary Putnam (see Putnam [19–21]) who maintained
that this theory is completely mistaken, which—although does not cause troubles for
mathematical logic, yet fatally deceives philosophers (see [35] for defending STT against
Putnam’s objections). Another criticism was advanced by from the point of view of anti-
realism based on constructive (intuitionistic) logic, for instance. Michael Dummett (see
[7]). According to objections forwarded by anti-realists of Dummett’s brand, meaning-
conditions should be defined as related to assertibility, but not as associated with truth in
a sense of Tarskian semantics.
The above brief and selected survey focuses on positive as well as negative influences
of Tarski’s ideas. Both axes reportedly show that STT belong to the contemporary
philosophical equipment, at least in the camp of analytic philosophy and those currents,
even inside the continental school, provided that they are ready to discuss various issues
with colleagues working with use of logical devices. Some philosophers try to continue
Tarski’s ideas and develop them according to new challenges, for example, advanced
by anti-realism, whereas other are at least stimulated by STT in their philosophical
investigations, even resulted in alternative semantic accounts. One should remember
that, for instance, anti-realistic semantics is still a semantic theory. In fact, Tarski’s
talks (see [26, 27]) saved philosophical semantics in general, not only his version of
it. Although in philosophy many unexpected happenings took place, the return to pre-
semantic ear, for instance, to the dominance of the syntactic approach in the style of
early logical empiricism seems very unlike. This general assessment of the role of STT in
contemporary philosophy should be illustrated by more specified data. However, before
focusing on a philosophical examination of STT, I will outline its formal shape.4
At the beginning of the present paper (see also footnote 1) I distinguished STT as a
mathematical (logical) theory and a philosophical doctrine. I deliberately used the terms
“theory” and “doctrine” as contrasts in their meaning. However, I abandoned this way
speaking when I passed to a closer examination of tasks that any philosophical account
(it is another convenient label) of truth has to confront (I also employ the expression
“truth-theory” (“theory of truth”) as referring to philosophical theories of truth. Now
there appears a question of the ambiguity of the word “theory”. Typically (at least in
metamathematics) a theory is a set of sentences, in particular, definitions closed under a
selected consequence operation. STT as a logical construction is a theory in this sense.
It is based on definitions and lemmas which enable us to prove in the exact mathematical
manner various properties of the set of true sentences, for instance, that this set forms
maximally consistent set. It is also possible to embed STT into the weak second-order
arithmetic with the axiom of arithmetic comprehension and assess its logical complexity
(see [10]). STT as a philosophical theory belongs to different order of things. By a
philosophical theory I understand a body of interconnected statements related to a set
of philosophical and metaphilosophical constraints. For instance, materialism, idealism,
rationalism or empiricism are such theories.
Some basic preliminary intuitions are as follows. Consider two stocks of ideas (for
simplicity I limit informal as well as formal explanations for monadic formulas, that is,
of the type P(. . . ), where an individual variable or an individual name occurs in the place
of dots occurring in the symbol (. . . )); the letter U represents the assumed universe of
discourse; it is convenient to claim that U is infinite):
(I) (General case): open formulas, satisfaction by an object from a given set U, non-
satisfaction by an object from the complement of U;
(II) (Special case): closed formulas (sentences), a special case of satisfaction relatively
to a given set of objects belonging to a given set U, satisfaction by no object from a
given set U.
Informally speaking, open formulas are neither true nor false, but satisfied or non-satisfied
by some objects. For instance, the number 2 satisfies the condition “x is a prime number”,
but the number 4 does not satisfy this condition. Yet we have an intimate connection
of truth and satisfaction. Some substitutions convert open formulas into true sentences,
but other ones—into falsehoods. This heuristics suggests to treat truth, resp. falsity, as
4 My presentation of STT uses rather contemporary settings than original Tarski’s version given in [24, 25]
and [29]). The main issue concerns the explicit role of the concept of model and working with first-order
languages.
378 J. Woleński
a special case of satisfaction, resp. non-satisfaction. Since a (declarative) sentence is

traditionally defined as a sentence which is true or false, we have a hint for defining
truth.
Let A be a sentence (closed formula, that is not having free variables). Now, it is
convenient to use sequences of objects, not objects as such. One can prove that A is either
satisfied by all sequences of objects from U or is not satisfied by any sequences of objects
(satisfied by no such sequence object). Tarski proposed to define truth as satisfaction by
all sequences of objects and falsity as satisfaction by no sequence of objects. The above
condition for truth is equivalent to saying that a sentence is true, if it is satisfied by at
least one infinite sequence of objects or by the empty sequence of objects. Technically,
since formulas can be of an arbitrary finite length, it is convenient to introduce infinite
sequences of objects in order to obtain a general scheme for all possible syntactic cases.
Moreover, because U is always associated with a structure of the type M = U, P!, where
P is collection of predicates (M is a model), we have the following definition :
(TrDef) A sentence A is true in a model M if and only if it is satisfied by all infinite
sequences of objects from U (or at least by one such sequence or the empty
sequence; otherwise A is false in M).
Thus, truth is defined in the above way is an outcome of an elegant analogy displayed
by (II). In particular, (TrDef) satisfies Tarski’s claim that a satisfactory definition of truth
should logically entail the equivalence (at the moment I use its naive form)
(TE) A sentence A is true if and only if A (in symbols, TA ⇔ A),
for any A belonging to a language L. This requirement is called the convention T (CT).
(TE) is frequently called the T-scheme. (CT) establishes the condition of the material
adequacy for a truth-definition, also for (TrDef).
The letter L in CT serves a new parameter in the entire construction. Hence (TrDef)
should be completed as
(TrDef*) A sentence A of a language L is true in a model M if and only if A is satisfied
by all infinite sequences of objects from U (or at least one such sequence or the
empty sequence; otherwise A is false in M).
The relativisation to L is associated with the semantic paradoxes, especially with the Liar
antinomy. It is generated by (TE) by substituting “this sentence is false” for A. In order to
solve the difficulty, (TE) is transformed into
(TE*) Tn(A) ⇔ [A],
where n(A) is a name of the sentence and [A] refers to a retranslation of A into a
metalanguage. Let L be German, but English serves as a metalanguage. We consider a
German sentence “Schnee ist weiss” and say (it is a special case of (TE*) that the sentence
“Schnee is weiss” of German is true if and only if snow is white. The expression “Schnee
ist weiss” is a name of the German sentence occurring inside quotes, but the right hand
of the equivalence in question is a translation of the German sentence. Consequently,
if A belongs to an object-language L, (TE*) and (TrDef*) must be formulated in the
metalanguage ML. Generally speaking, a semantic theory for L should be formulated in
ML in order to avoid semantic paradoxes. It is very important to see that (TE) is not a
truth-definition. CT states that every instance of the T-scheme must be derivable from
(TrDef*), but, except the case, when L is finite (languages are sets of sentences), we have
no equivalence between (TrDef*) and (TE*).
To proceed more formally, consider a formalized language L for which truth (more
precisely a set of true sentences) is defined. Due to arithmetization (or other similar tech-
nique), the syntax of L can be represented in L itself. However, the Tarski undefinability
theorem (the set of true sentences of arithmetic of natural numbers in not arithmetically
definable; UT for brevity)) shows that semantics of L is not fully expressible in L
itself. In order to define semantic relations, we need to use ML which has a greater
expressive power than L. Perhaps the most important observation is that ML remains
partly informal. Even if we formalize ML, we must use MML and the story reappears.
Generally speaking, the hierarchy L1 , L2 (= ML1 ), L3 (=ML2 ), . . . , Ln (=MLn-1 ), . . . of
languages has the following property: if Lk is formalized, Lk+1 (1 ≤ k ≤ n) has some
informal features with respect to Lk+1 (= MLk ). Thus, there occurs a necessary connection
between formal and informal aspects of STT. Yet one point requires an explanation.
According to Tarski, the truth-definition for L generated by STT can be given in the
syntax of ML. Does it mean that semantics is reducible to syntax? Not at all. Tarski
worked in the paradigm of logic on which set theory belonged to logic. In fact, the
definition of truth via satisfaction proceeds in the set-theoretical framework. Thus, we
can say at most that defining semantic relation for L assumes a set theory. Putting this
fact into the contemporary fashion, if T is a first-order theory (principally every deductive
theory can be expressed in the first-order language), its semantics can be constructed
in weak second-arithmetic with the arithmetic axiom of comprehension, employed as a
metatheory MT of T. However, this does not mean that we reduce semantics of T to
T-syntax. Otherwise speaking, if T is completely formalized, MT does not admit such
a treatment. As it was pointed out above, it is possible to stay with first-order object
languages and the weak-second order arithmetic as the metatheory.
What about constraints of a successful theory of truth? Omitting earlier proposals
answering this question, I recall conditions stated by Bertrand Russell. According to him
(see [23]), any theory pretending to be the satisfactory account of the concept of truth
must conform to:
(A) the theory of truth must also explain the nature of falsehood;
(B) truth must be taken as a property of beliefs;
(C) truth is an external relation of belief to something existing outside them.
However, these conditions say too much on the one hand, but too little on the other. Too
much, because (B) and (C) exclude some important ideas. The requirement (B) ignores
other accounts of truth-bearers, but (C), selecting the correspondence theory as proper,
rejects theories which consider truth as consisting in relations between judgments without
making any reference to the external world; in particular, (C) excludes the coherence
theory. On the other hand, Russell’s constraints are too weak, if he intended to favour the
correspondence theory. For example, if one says that truth consists in evidence of beliefs,
one also offers an account which satisfies (C). In general, the Russellian conditions are
unfair to the richness of problems usually investigated by truth-theories that are known
from a very history of the subject.
Thus we need a more complex scheme of any philosophical truth-theory which intends
to be historically faithful. Without ambition to completeness (the list is filtered by STT),
I mention the following problems which should be touched by any philosophically
380 J. Woleński
reasonable. By being philosophically I do not mean ‘correct’, but ‘deserving an attention

in the world of philosophy’) truth-theory in philosophy:
1. What are the bearers of truth?
2. What are initial intuitions associated with a given truth-definition?
3. How to define truth?
4. How is truth related to logic?
5. If we classify truth-bearers into true and false, is this division exhaustive and disjoint
(are there values apart from truth and falsehood or perhaps are there truth-falsehood-
value gaps)?
6. Is this division stable, that is, do at least some truth-bearers sometimes change their
truth-values (briefly: is truth relative or absolute)?
7. What is a truth-criterion and what about the relations of truth-criteria and truth-
definition?
8. What is the relation of a particular truth-theory to its rivals?
9. How a given truth-theory can be defended against various objections?
10. What is the relation of truth to other philosophical problems?
As we see, there is a lot to do for a theory of truth. I will try to show how STT of truth is
related to these questions, or at least to some of them.
(Ad1) STT assumes that truth-bearers are sentences in the syntactic sense. We have
several other possibilities (see [22] for a survey). Sentences? Propositions? Statements?
Judgments? These entities can be either linguistic units or objects expressed by linguistic
utterances. By contrast, concepts are not truth-bearers, contrary to Hegelians. To have a
convenient label, we can say that according to STT entities qualified as true or false are
of the propositional syntactic category. This ways of speaking has nothing to do with the
question of the ontological nature of propositions, for instance, as abstract objects.
Tarski himself chose sentences as entities on which truth is predicated. But there is an
additional very important point [25, pp. 166–167]:
It remains perhaps to add that we are not interested here in ‘formal’ languages in sciences in one
special sense of the word ‘formal’, namely sciences to the signs and expressions of which no
material sense is attached. For such sciences the problem here discussed [the problem of truth]
has no relevance, it is not even meaningful. We shall always ascribe quite concrete and, for us,
intelligible meanings to the signs which occur in the language we shall consider. The expressions
which we call sentences still remain sentences after the signs which they occur in have been
translated into colloquial language.
Thus a language L for which STT applies is always interpreted, even if it is formalized.
Consequently an interpretation of L always precedes definitions of semantic concepts
including truth. Thus we arrived at the problem of how ‘formal’ is related to ‘formalized’.
The answer is that formal languages do not need to be equipped, contrary to formalized
languages.
A common misunderstanding of Tarski’s views consists in attributing to him the
opinion that STT applies to formal languages only. This mistake neglects what Tarski
explicitly explained (see essays in [16] for a discussion), that truth-bearers are correct
syntactic units of the propositional category having meaning. It does not mean that
Tarski’s views about language and meaning have no weak points. In particular, he did
not define the concept of meaning. In fact, he intentionally avoided this question and
deliberately preferred to speak about interpreted languages as semantic items. For Tarski,
the concept of language was clearer than the concept of meaning (see [17] for an
extensive presentation of Tarski’s philosophy of language). Yet STT does not seem to
be especially dependent on a particular theory of meaning. Another controversial point
concerns STT and natural language. Tarski regarded natural language as universal, and
thereby generating semantic paradoxes. Roughly speaking, natural languages do not block
self-referential use of semantic predicates. The Liar sentence “This sentence is false”
illustrates this fact, as it attributes the property of being false to itself. In Tarski’s view,
dividing natural language into strata (levels), like the object language, the meta-language,
the meta-meta-language, etc. is inconsistent with its naturalness and universality. On
the other hand, it is quite legitimate to define truth for fragments of natural language
(see [28]). Finally, Tarski’s view on the relation of truth and meaning differs from that
of Davidson. Whereas the latter maintained that truth-conditions define the meaning of
sentences, the former considered meaning as prior to truth (see [38]).
(Ad2) Tarski always stressed that his definition follows intuitions of Aristotle. He
was influenced by Stagirite himself and by his teachers (see [16, 17]). However, Tarski’s
statement in his particular writings on truth differ. In [24, p. 152 ] he says:
[. . . ] in this work I shall be concerned exclusively with grasping the intentions which are true,
contained in the so-called classical conception of truth (‘true—corresponding with reality), in
contrast, for example, with the utilitarian conception (‘true—in a certain respects useful’).
Further (p. 155), he adds that

true sentence is one which says that the state of affairs so and so, and the state of affairs is indeed
so and so.
However, the Polish original text has no exact counterpart of the expression “state of
affairs”. On the other hand, English (but not German) translation quotes famous passage
from the Aristotle’s Metaphysics:
To say of what is that it is not, or what is not that it is, is false; while to say that of what is that it
is, or what is not that it is not, is true.
This quotation also appears in [28, p. 667] and is explained by statements “The truth of
a sentence consists in the agreement with (or corresponding to) reality” and “A sentence
is true if it designates an existing state of affairs”. Tarski’s comment (p. 267) is as follows:
However all these formulations can lead to various misunderstandings, for none of them is suffi-
ciently precise and clear (though this applies? much less to the original Aristotelian formulation);
at any rate, none of them can be considered a satisfactory definition of truth.
In [30, pp. 402–403] the above formulations are repeated together with similar critical
remarks. Tarski subsequently says (p. 403):
The conception of truth which is found in its expression in the Aristotelian formulation (and in
related formulations of some recent origin) is usually referred to as the classical, or semantic,
conception of truth. By semantic, we mean the part of logic which, loosely speaking, discusses
the relations between linguistic objects, e.g. sentences and what is expressed by these objects; the
semantic character of the term “true” is clearly revealed by the explanation offered by Aristotle
and by some formulations which will be given in our further discussion. One speaks sometimes of
the correspondence theory of truth as the theory based upon the classical conception.
To sum up, Tarski, at the beginning, identified the classical and correspondence theory
of truth, but later he expressed greater reservations with respect to explanations via ex-
pressions, like “agreement” or “correspondence” than to Aristotle’s original formulation.
382 J. Woleński
If we assume that STT follows Aristotle’s intuitions, are they captured by (TE*) of
(TrDef*)? The answer that the former seems fairly correct and justified by Tarski’s own
explanations. So the example ‘the sentence “snow is white” is true if and only if snow is
white’, points out that because the sentence in question says that snow is white and it is so
and so as this sentence says, it (the sentence) is true. What about the intuitive content of
(TrDef*)? We have two options; first, having some justifications in Tarski’s explanations
that is a mathematical trick, and second, that the official definition brings some intuitions.
First of all, sequences of objects cannot be identified with facts. Moreover, the satisfaction
by the empty sequence appears as an artificial construction (see [24, p. 195]). On the other
hand, if the semantic truth-definition is a special case of the definition of satisfaction and
the latter is based on explicit intuitions, it suggests that perhaps some intuitions are behind
(TrDef*) as well. I am inclined to take the last option; that whether an open formula
is satisfied or not by an object depends of valuation of free variables. Such valuations
are irrelevant in the case of sentences. Consequently every infinite sequence of objects
can be ascribed to bound variables (note that individual constants can be eliminated by
identity and existential quantification). The same can be expressed by saying that the
empty sequence satisfies a sentence. What remains? The answer is that being true depends
on how L is interpreted and, metaphorically speaking, how things are in M associated with
L. And it precisely expresses what is established by the T-scheme. Informally speaking,
truth depends on the domain which sentences to say about. (see [43]) to use the name
“weak correspondence” or “semantic correspondence” in the case of (TrDef*) and (TE*)
as something different from “strong correspondence”, used, for example, by Russell in
his definition of truth (the structure of a judgement or a proposition) which corresponds
with the structure of a corresponding fact provided that this judgment is true. Thus I will
consider STT as the classical truth-theory based on the weak concept of correspondence.
(Ad. 3) Tarski defined truth by a single formula (the definition satisfaction is recursive).
He considered (see [26]) introducing truth by axioms, but rejected this possibility for
philosophical reasons. More specifically, he was afraid of a criticism on the side of
physicalism. This worry was associated with his scepticism mentioned by Carnap (see
above). This motivation is presently completely historical. Tarski himself mentioned that
taking all instances of T-scheme as axioms could be regarded as an axiomatization of the
set of true sentences. Yet he was fully aware that such procedure would be trivial and
leading to the infinite collection of axioms. Today, the axiomatization of the concept of
truth is commonly applied (see [10, 13]) and also proposed in order to avoid semantic
paradoxes. I will not enter into this issue.
Tarski’s way has an important consequence because of his undefinability theorem.
Assume, what is natural, that the collection TRUTH of all truths is infinite. By UT (see
below), TRUTH is not definable by resources conceptually available within it. Yet saying
that TRUTH exists appears to be philosophically tempted. The only way out admissible
by set theory consists in conserving TRUTH as too big set (Zermelo-Fraenkel system),
a class as distinct from sets (Bernays-Gödel-von Neumann) or a category. All these
outcomes are formally correct but lead to not quite pleasant consequences, at least for
philosophers who like having something to say on the set of all truths. However, set theory
and UT seriously limit such theoretical ambitions. We can eventually say that TRUTH
constitutes consistent deductive systems, which has no finite axiomatization, even by
schemes. On the other hand, TRUTH is not compact, which means that although its
every finite subset has a model, the collection of all truths has no model. Consequently, the
collection MOD of all models is not a set as well. The medieval theory of transcendentals
assumed that truth coincides with being and that ens (and other transcendentalia, for
instance, verum, that is truth), omnia genera transcendit. In a sense, considerations about
TRUTH and MOD (it can represent the being) justify the medieval intuition about the
transcendentality of truth and being (see [40] for a more closer analysis). It is rather an
unexpected application of STT.
Ad (4 and 5) Tarski proved that STT implies bivalence, which is the conjunction of
metalogical principles of excluded middle and (non)contradiction. This means that this
theory is inherently associated with classical logic. However, Tarski’s proof is sometimes
criticized contested as circular, as assuming classical logic in metatheory. It is possible to
demonstrate that the above result can be constructively achieved (see [36]); the argument
employs the fact that truth can be defined as satisfaction by the empty sequence). Now,
the problem arises whether construction a la Tarski can be realized in the case of non-
classical logic, in particularly, one rejecting the presumption that every sentence is either
true of false. Many-valued logics and logics with truth-value gaps provide standard
examples. Paraconsistent logic, intuitionistic logic or quantum logic are further examples.
Clearly, more or less modified Tarskian ideas have applications for non-classical logics
but according to my knowledge no general results are available. For instance, some
constructions use partial models also for excluding paradoxes, but only special cases are
available.
Consider T as a modality. We read TA as “it is true that A”. We have the following
formulas TA, T¬A, ¬T¬A, ¬TA, TA ∨ T¬A, ¬T¬A ∧ ¬TA. These formulas satisfy
logical modal principles organized by a generalized logical square, for instance TA ⇒
¬T¬A. The conjunction ¬T¬A ∧ ¬TA is logically possible (consistent) and opens room
for other (than truth and falsity) logical values or truth-value gaps. We have also the
principle TA ⇒ A, but its converse does not hold as a purely logical rule. Thus, we
can add the formula A ⇒ TA and obtain the T-scheme as a new theorem, but it is
not forced by logic. Adding the T-scheme results in cancelling ¬T¬A ∧ ¬TA as a
possibility. Moreover, TA, ¬T¬A, A as well as T¬A, ¬TA, ¬A become equivalent, and
the same concerns TA ∨ T¬A, TA ∨ ¬TA, A ∨ ¬A. Since we can now interpret T¬A,
¬TA, ¬A as expressing “A is false”, the principle of bivalence is valid in the logic of
truth with the T-scheme. This reasoning shows that T-scheme favours classical logic.
Since ¬(TE*) ⇒ ¬(TrDef*), rejecting T-scheme results in rejecting the semantic truth-
definition. Yet, this conclusion does not preclude a revision of Tarski’s definition for needs
of particular logical systems. For instance, TA ⇒ A is a theorem of classical modal theory
of truth, but a paraconsistent logician can accept its negation, that is the formula TA ∧ ¬A,
but its truth-condition requires a modification of (TRDef*). Incidentally, modal analysis
of truth shows that the T-scheme is not a logical tautology, because the formula A ⇒ TA
is not logically valid. If one were to say that modal logic is not a logic in the proper sense,
we could still point out that there is a difference in the status of both components of the
T-scheme: the formula TA ⇒ A is a formal modal theorem, but its converse is not.
(Ad 6) The classical concept of truth is commonly considered as absolute, that is,
if A is true such eternally (for ever) and sempiternally (since ever). On the other hand,
(TrDef*) indexes truth by L and M. Does this relativisation deprive truth of its absolute
character? This question is connected with such issues as bivalence, logical determinism
384 J. Woleński
or many-valued logic. Without entering into details concerning this fairly complex stock
of ideas, let me somehow dogmatically suggest (details are in [40, 43]) that we can
model-theoretically prove that truth is eternal if and only if it is sempiternal. If so, the
classical theory of truth in the semantic setting can be considered as associated with the
absolute concept of truth. Even if this conclusion encounters reservations, the possibility
of analysing the absolutism/relativism controversy within the philosophical theory of truth
via (TrDef*) is a remarkable fact.
(Ad 7) Clearly (TrDef*) is a-criterial. This means that the definition in question
does not generate any truth-criterion, though it says what truth is. If mathematics is
taken into account, proof can be regarded as a measure of truth. However, there arises
a problem. Let the symbol Pr denote the provability operator. By the Löb theorem, we
have PrA ⇒ A, a theorem very similar to TA ⇒ A. But, due to the first incompleteness
theorem, the formula A ⇒ PrA cannot be consistently added to the provability logic.
Hence, there is no counterpart of (TE*) with Pr instead T, that is, the scheme PrA ⇔ A
and we must conclude that proof is not a complete truth-criterion even in mathematics.
This fact can motivate various ways out, for instance, modifying the concept of proof
(every true mathematical assertion can be proved in a formal system; this assertion
does not contradict the incompleteness theorem) or replacing truth by proof, eventually
with additional constraints, for instance, that proofs must be constructive. However, such
proposals are restricted to mathematics. Another suggestion is like this. Consider an open
formula P x. It can be transformed to the true sentence P a via substituting x by the
term a denoting an object which has a property P . Another way is to go through from
P x to ∃x P x. Both strategies require some empirical or deductive steps based on some
criteria. And these criteria are truth-criteria. Generally speaking, truth-criteria consist of
procedures which justify satisfaction of open formulas by some objects.5 Note that the
proposed criterion does not work for satisfaction by all sequence of objects.
(Ad 8) Tarski grew up in the tradition of division of truth-theories into the classical
theory and so-called non-classical theories of truth, namely the evidence theory (A is true,
if is evident), the coherence theory (A is true, if it can be embedded into a coherent system
without destroying its coherence), the common agreement theory (A is true, if specialists
agree about its correctness) and the utilitarian theory (A is true, if A is useful). The non-
classical theories are criteria, because they appeal to procedures assuring that something
is true.
Tarski himself mentioned the last definition (see above) and the coherence account
(see [30, p. 403]). He considered them as lacking of precision and did not discuss them
as serious alternatives for STT. However, it seems that the coherence theory can be
discussed with help of some logical ideas. Let us agree that consistency is a component
of coherence. Thus, we defined a system S as coherent if and only if S is consistent and
satisfies some additional requirements, for instance is comprehensive, has an empirical
support etc. Anyway, we have S ∈ COH ⇒ S ∈ CONS. It means that consistency is a
necessary condition for coherence. The provability operator Pr satisfies the condition
(*) Pr(A ⇒ B) ⇒ (PrA ⇒ PrB). By definition, Pr(S ∈ COH ⇒ S ∈ CONS). This and the
condition (*) gives Pr(S ∈ COH) ⇒ Pr(S ∈ CONS). By the rule contraposition, we obtain
¬Pr(S ∈ CONS) ⇒ Pr(S ∈ COH). Finally, the second incompleteness theorem, asserting
5 This suggestion was made by Anna Kanik, a former student of mine.

that consistency is not generally provable, suggests that the coherence is not a general
criterion of truth. Thus, at least under classical logic, there are truths, which satisfy
(TrDef*) but cannot be tested by coherence. On the other hand, the above argument
does not suggest that there are truths not subjected to any justification. Yet, no absolutely
universal criterion of truth based on exclusively deductive resources seems to be possible.
By analogy, one can argue that any criterion of truth appealing to this or that kind of
justification, works in the concrete circumstances and cannot define the concept of truth
in its full generality. If so, STT (or the classical theory) is the only known account of
truth, which is actually universal.
Another issue involving the relation between various truth-theories concerns substan-
tial and minimalist account (see [15] for an analysis of conceptions of truth from the point
of view of this axis). The latter approach (the redundancy theory, the deflationary theory,
etc.) reduces the truth-definition to the T-scheme. Under this view, STT is a minimalist
theory. Tarski himself (see [27, p. 682–683]) discussed this question. His counterexample
was the sentence “All consequences of true sentences are true”. To continue Tarski’s
analysis let us assume that the minimalist translation of the above assertion runs “for
every B, if A, and B is a logical consequence of A, then B”, we still need to explain
the ground of truth in the case of universal closure (that is, sentences with the universal
quantifier in front) and the meaning of the predicate “is a logical consequence”. Thus,
T-scheme does not justify to assert that all consequences of true sentences are true. There
are much more complicated cases, for instance, the sentence “There exist true but not
provable sentences”, which seems to be not subjected to a minimalist translation. If so,
STT is essentially richer than any minimalist theory of truth.
(Ad 9) I will address in this section three objections stated by Franz Brentano against
the classical theory and try to show that STT meets them successfully (see [32]). Firstly,
the concept of correspondence is obscure and cannot be explained satisfactorily. More
precisely, in order to establish what a truth-bearer corresponds to reality, one must
compare the former with the latter. But it is impossible, due to relata of such a comparison.
However, this objection applies to the strong notion of correspondence, not to its weak
form. The second objection is more serious. Assume that we define truth by a definition D.
Yet D is a sentence. In order to have a good definition, D must be true. Now, the definition
is either circular (if it uses itself) or falls into the regressum ad infinitum, because in
order to formulate D, we must appeal to D’ related to D, etc. Thirdly, the concept of
correspondence does not explain truth of negative sentences.
How things are in STT, relatively to the second objection (see also [28, pp. 680–
681])? Obviously, the answer depends on the relation of L (for which truth is defined) to
ML (in which truth is defined). Tarski observed that the latter must be essentially richer
than the former. Using a more contemporary way of speaking, it holds for semantics
of L, which is not fully expressible in its syntax. Consequently, the expressive power
of ML must be greater if the concept of truth is to be defined. In particular, ML has
to have resources to define the concept of satisfaction. According to Tarski, (TrDef*)
is expressible in the syntax of ML. This explanation is slightly misleading, because the
satisfaction is defined in set theory. The crucial point is that (TrDef*) for the concepts
of truth and of satisfaction is defined recursively. Although one should assume that the
metatheory of truth is consistent, consistency is a syntactic property. Thus circularity
does not occur in the entire procedure. The same concerns the regressum ad infinitum.
386 J. Woleński
However, another interesting point appears. Clearly, if L is formalized, ML is not, at least

not entirely. Consequently, the content of ML exceeds the content of L. One can observe
that this phenomenon leads to the situation that semantic properties of poorer and thereby
less problematic theories are defined in richer and thereby more problematic conceptual
systems. In the case of STT this circumstance is somehow limited by the mentioned fact
that the weak second-order arithmetic is sufficient for (TrDef*), but the phenomenon in
question is very intriguing from the philosophical point of view. The problem of negative
sentences has a simple solution in STT, because they are true (or false) under the same
definition as positive ones.
(Ad 10) According to Tarski [28, p. 686]:
[. . . ] We may accept the semantic theory of truth without giving up any epistemological attitude
we may have had; we may remain naive realists, critical realists or idealists, empiricists or
metaphysicians — whatever we were before. The semantic conception of truth is completely
neutral toward all these issues.
These words seem to block any serious involvement of STT into traditional philo-
sophical debates and controversies. Two remarks are in a row here. Firstly, if we look at
the stock of terms used in the above passage, we can ask whether Tarski’s evaluation
might be generalized. My impression is that terms “naive realists”, “critical realists”,
“idealists” or “empiricists” refer to views concerning the philosophy of perception; even
“metaphysicians” can be taken as referring to philosophers speaking about the reality of
what is perceived. Secondly, even if one claims that Tarski employed the labels in question
as exemplifications of his general and somehow negative attitude to philosophy, we should
ask whether he was right.
Independently of Tarski’s intentions, it is easy to give an example of a philosophical
problem closely related to STT, namely the semantic realism/semantic anti-realism
debate. Generally speaking, (semantic) realists use STT but (semantic) anti-realists reject
this account to truth (see [37]). This issue concerns the mutual relation of the condition of
truth and condition of assertibility. Generally speaking, the realist says that the meaning
of a sentence (MS) is given by its truth-conditions (TC), whereas the anti-realist argues
that MS is given by its assertibility-conditions (AC). Thus we have two equalities:
(i) MS = TC;
(ii) MS = AC.
However, (i) and (ii) are still too vague. In fact, we should transform (i) and (ii) into
(iii) (MS = TC) ∧ (TC > AC);
(iv) (MS = AC) ∧ (TC = AC),
respectively. In fact, the realist says that truth-conditions exceed assertibility-conditions
but the anti-realist identifies truth-conditions with the assertibility conditions.6 How does
STT work here? It justifies (iii), but refuses (iv). If, as Dummett maintains, the conditions
of assertibility are governed by intuitionistic logic, it does not generate sufficient and
necessary conditions for asserting any mathematical sentence. The point is that the
6I model these formulas on the earlier mentioned debate between Davidson (realism) and Dummett (anti-
realism). My own view (also expressed in [38]) is that meaning of a sentence exceeds its truth-conditions.
However, the points (iii)–(iv) suffice for further analysis.
incompleteness theorem constructively holds for the Heyting arithmetic, that is, the Peano
arithmetic based on intuitionistic logic. If so, the anti-realist cannot say that there are true,
but unprovable sentences, but the realist can do so by appealing to STT.
Although I accept the semantic theory of truth as the correct account of the concept of a
true sentence, I am very far from saying that its philosophical uses are unproblematic and
the only correct. For instance, my analysis of the issue of realism/anti-realism should be
taken as an analysis of what follows, if STT is assumed. My main intention in the present
paper consists in demonstrating that Tarski’s semantic ideas are not philosophically
sterile.
References
1. Ajdukiewicz, K.: Zagadnienie empiryzmu a koncepcja znaczenia [Empiricism and the Problem of
Meaning]. Studia Filozoficzne 1(36), 3–14 (1964) (Eng. tr. in: Ajdukiewicz, K., The Scientific World-
Perspective and Other Essays, 1930–1963, pp. 306–319. D. Reidel, Dordrecht (1978))
2. Ayer, A.J.: A Part of My Life. Collins, London (1977)
3. Black, M.: The semantic definition of truth. Analysis 9, 49–63 (1948) (Reprinted in: Black, M.,
Language and Philosophy, pp. 89–107. Cornell University Press, Ithaca (1949))
4. Carnap, R.: Introduction to Semantics. Harvard University Press, Cambridge (1942)
5. Carnap, R.: Intellectual autobiography. In: Schilpp, P. (ed.) The Philosophy of Rudolf Carnap, pp.
1–84. Open Court, LaSalle (1963)
6. Davidson, D.: Truth and Meaning. Synthese 17, 304–323 (1967) (Repr. in: Davidson, D., Inquiries
into Truth and Interpretation, pp. 65–75. Oxford University Press, Oxford (1984))
7. Dummett, M.: The Logical Basis of Metaphysics. Duckworth, London (1991)
8. Girard, J-Y.: The Blind Spot. Essays on Logic. European Mathematical Society, Zürich (2011)
9. Gupta, A., Belnap, N.: The Revision Theory of Truth. The MIT Press, Cambridge (1993)
10. Halbach, V.: Axiomatic Truth Theories. Cambridge University Press, Cambridge (2011)
11. Hintikka, J.: The Principles of Mathematics Revisited. Cambridge University Press, Cambridge
(1996)
12. Hodges, W.: Truth in a structure. Proc. Aristot. Soc. 86, 135–151 (1985–1986)
13. Horsten, L.: The Tarskian Turn: Deflationism and Axiomatic Truth. The MIT Press, Cambridge
(2011)
14. Kripke, S.: An outline of a theory of truth. J. Philos. 72, 690–716 (1975) (Repr. in: Kripke, S.,
Philosophical Troubles. Collected Papers, v. 1, pp. 75–98. Oxford University Press, Oxford (2011))
15. Künne, W.: Conceptions of Truth. Oxford University Press, Oxford (2005)
16. Murawski, R., Woleński, J.: Tarski and his precedessors on truth. In: Patterson, D. (ed.) New Essays
on Tarski and Philosophy, pp. 21–43. Oxford University Press, Oxford (2008)
17. Patterson, D.: Alfred Tarski Philosophy of Language and Logic. Palgrave Macmillan, Hampshire
(2018)
18. Popper, K.: Objective Knowledge: An Evolutionary Approach. Clarendon Press, Oxford (1972)
19. Putnam, H. (ed.): Do true assertions correspond to reality? In: Mind, Language and Reality.
Philosophical Papers, vol. 2, pp. 70–84. Cambridge University Press, Cambridge (1975)
20. Putnam, H.: On truth. In: Cauman, L., Parsons, C, Schwartz, R. (eds.) How Many Questions? Essays
in Honour of Sidney Morgenbesser, pp. 35–56. Hackett, Indianapolis (1983) (Repr. in: Putnam, H.,
World and Life, pp. 315–329. Harvard University Press, Cambridge (1994))
21. Putnam, H.: On comparison of something with something else. New Lit. Hist. 17, 61–79 (1985) (Repr.
in: Putnam, H., World and Life, pp. 330–350. Harvard University Press, Cambridge (1994))
22. Rojszczak, A.: From the Act of Judging to the Sentence. The Problem of Truth Bearers from Bolzano
to Tarski. Springer, Dordrecht (2005)
23. Russell, B.: The Problems of Philosophy. Wiliam and Norgate, London (1912)
388 J. Woleński
24. Tarski, A.: Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych [The Concept of Truth in Languages of
Deductive Sciences]. Towarzystwo Naukowe Warszawskie, Warszawa (1933)
25. Tarski, A.: Der Wahrheitsbegriff in den formalisierten Sprachen. Stud. Philos. I, 53–198 (1935)
26. Tarski, A.: Grundlegung der wissenschaftlichen Semantik. In: Actes du Congrès international de
philosophie scientifique, Paris 1935, fasc. 3: Semantique, pp. 1–14. Herman, Paris (1936) (Eng.
tr.: The Establishment of Scientific Semantics. In: Tarski, A., Logic, Semantics, Metamathematics.
Papers of 1923 to 1938, pp. 401–408. Clarendon Press, Oxford (1956); 2nd ed., pp. 401–408. Hackett
Publishing Company, Indianapolis (1983))
27. Tarski, A.: Über den Begriff der logischen Folgerung. In: Actes du Congrès international de
philosophie scientifique, Paris 1935, fasc. 7: Logique, pp. 1–11. Herman, Paris (1936) (Eng. tr.: On the
Concept of Logical Consequence. In: Tarski, A., Logic, Semantics, Metamathematics. Papers of 1923
to 1938, pp. 409–420. Clarendon Press, Oxford (1956); 2nd ed., pp. 409–420. Hackett Publishing
Company, Indianapolis (1983)). See also [39, 41, 42]
28. Tarski, A.: The semantic conception of truth and the foundations of semantics. Philos. Phenomenol.
Res. 4, 341–395 (1944) (Repr. in: Tarski, A., Collected Papers, vol. 2, pp. 665–699. Birkhäuser, Basel
(1986))
29. Tarski, A.: The concept of truth in formalized languages. In: Tarski, A. (ed.) Logic, Semantics,
Metamathematics. Papers of 1923 to 1938, pp. 152–278. Clarendon Press, Oxford (1956) (2nd edn.,
pp. 152–278. Hackett Publishing Company, Indianapolis (1983))
30. Tarski, A.: Truth and proof. L’age de la Sci. 1, 279–301 (1969) (Repr. in: Tarski, A., Collected Papers,
v. 4, pp. 399–422. Basel: Birkhäuser (1986))
31. Wang, H.: Beyond Analytic Philosophy: Doing Justice to What We Know. The MIT Press, Cambridge
(1986)
32. Woleński, J.: Brentano’s criticism of the correspondence theory of truth and Tarski’s semantic
theory. Topoi 6, 105–110 (1989) (Repr. in: Woleński, J., Essays in the History of Logic and Logical
Philosophy, pp. 103–109. Jagiellonian University Press, Kraków (1999))
33. Woleński, J.: Semantic Revolution: Rudolf Carnap, Kurt Gödel, Alfred Tarski. In: Woleński, J.,
Köhler, E. (eds.) Alfred Tarski and the Vienna Circle, pp. 1–15. Kluwer Academic Publishers,
Dordrecht (1999)
34. Woleński, J.: Semantic conception of truth as a philosophical theory. In: Peregrin, J. (ed.) Truth and
Its Nature (If Any), pp. 51–65. Kluwer Academic Publishers, Dodrecht (1999)
35. Woleński, J.: In defence of the semantic definition of truth. Synthese 126(1–2), 67–90 (2001)
36. Woleński, J.: Truth and satisfaction by the empty sequence. In: Rojszczak, A., Cachro, J., Kurczewski,
G. (eds.) Philosophical Dimensions of Logic and Science. Selected Contributed Papers from the 11th
International Congress of Logic, Methodology, and Philosophy of Science, Kraków, 1999, pp. 267–
276. Kluwer Academic Publishers, Dordrecht (2003) (Repr. in: Woleński, J., Essays on Logic and Its
Applications in Philosophy, pp. 17–24. Peter Lang, Frankfurt am Main (2011))
37. Woleński, J.: Logic, semantics and realism. In: Benmakhlouf, A. (ed.) Sémantique et épistemologie,
pp. 135–148. Editions Le Fennec, Casablanca (2004) (Repr. in: Woleński, J., Essays on Logic and Its
Applications in Philosophy, pp. 51–60. Peter Lang, Frankfurt am Main (2011))
38. Woleński, J.: Two critical contributions to the problem of truth and meaning. Polish J. Philos. 1, 137–
141 (2007) (Repr. in: Woleński, J., Essays on Logic and Its Applications in Philosophy, pp. 183–186.
Peter Lang, Frankfurt am Main (2011))
39. Woleński, J.: Truth is eternal if and only if it is sempiternal. In: Tegtmeier, E. (ed.) Studies in the
Philosophy of Herbert Hochberg, pp. 223–230. Ontos Verlag, Frankfurt am Main (2012)
40. Woleński, J.: Ens et Verum Concertuntur (Are Being and Truth Convertible)? A Contemporary
Pesrpective. In: Dumitriu, M., Sandu, G. (eds.) Truth, pp. 75–83. Editura Universitãtii din Bucureşti,
Bucureşsti (2013)
41. Woleński, J.: An analysis of logical determinism. In: Lapointe, S. (ed.) Themes from Ontology, Mind
and Logic. Present and Past. Essays in Honor of Peter Simons, pp. 423–442. Brill, Leiden (2015)
42. Woleński, J.: Formal and informal aspects of the semantic theory of truth. In: Kuźniar, A., Odroważ-
˛
Sypniewska, J. (eds.) Uncovering Facts and Values. Studies in Contemporary Epistemology and
Political Philosophy, pp. 56–66. Brill/Rodopi, Leiden (2016)
43. Woleński, J., Simons, P.: De Veritate: Austro-Polish contributions to the theory of truth from Brentano
to Tarski. In: Szaniawski, K. (ed.) The Vienna Circle and the Lvov-Warsaw School, pp. 391–442.
Kluwer Academic Publishers, Dordrecht (1989)
J. Woleński ()
University of Information, Technology and Management Rzeszów, Rzeszów, Poland
Tarski’s Influence on Computer Science
Solomon Feferman
Abstract Alfred Tarski’s influence on computer science was indirect but significant in a
number of directions and was in certain respects fundamental. Here surveyed is Tarski’s
work on the decision procedure for algebra and geometry, the method of elimination of
quantifiers, the semantics of formal languages, model-theoretic preservation theorems,
and algebraic logic; various connections of each with computer science are taken up.
Keywords Tarski · Decision procedures · Quantifier elimination · Cylindrical

algebraic decomposition (CAD) · Time complexity · Space complexity · Semantics for
formal languages · Fixed point theorem · Finite model theory · Preservation
theorems · Relational database theory · Algebraic logic · Relation algebras · Cylindric
algebras
Mathematics Subject Classification (2000) Primary F.4.1; Secondary F.4.3
The following is the text of an invited lecture for the LICS 2005 meeting held in Chicago
June 26–29, 2005.1
Almost exactly 8 years ago today, Anita Feferman gave a lecture for LICS 1997 at
the University of Warsaw with the title, “The saga of Alfred Tarski: From Warsaw to
Berkeley.” Anita used the opportunity to tell various things we had learned about Tarski
while working on our biography of him. We had no idea then how long it would take
to finish that work; it was finally completed in 2004 and appeared in the fall of that
The author “Solomon Feferman” is deceased.

Reprinted from Logical Methods in Computer Science, vol. 2 (3:6), pp. 1–13, https://doi.org/10.2168/
LMCS-2(3:6)2006.
This reprint for Studies in Universal Logic style in LATEX is prepared by Zbigniew Bonikowski.
1 I want to thank the organizers of LICS 2005 for inviting me to give this lecture and for suggesting the
topic of Tarski’s influence on computer science, a timely suggestion for several reasons. I appreciate the
assistance of Deian Tabakov and Shawn Standefer in preparing the LATEX version of this text. Except
for the addition of references, footnotes, corrections of a few points and stylistic changes, the text is
essentially as delivered. Subsequent to the lecture I received interesting comments from several colleagues
that would have led me to expand on some of the topics as well as the list of references, had I had the
time to do so.

392 S. Feferman
year under the title, Alfred Tarski: Life and Logic [14]. The saga that Anita recounted
took Tarski from the beginning of the twentieth century with his birth to a middle-class
Jewish family and upbringing in Warsaw, through his university studies and Ph.D. at the
ripe young age of 23 and on to his rise as the premier logician in Poland in the 1930s
and increasing visibility on the international scene—despite which he never succeeded in
obtaining a chair as professor to match his achievements. The saga continued with Tarski
coming to Harvard for a meeting in early September, 1939 when the Nazis invaded Poland
on September 1st, at which point he was, in effect, stranded. Then, during the next few
years he went from one temporary research or teaching position to another on the East
Coast. He was finally offered a 1 year position in 1942 as Lecturer in Mathematics at the
then far off University of California in Berkeley, with the suggestion that it might stretch
into something longer. In fact, he not only succeeded in staying, but rose to the rank of
Associate Professor by the end of the war and a year later was made Full Professor, thus
finally obtaining the position he deserved. At Berkeley, Tarski built from scratch one of the
world’s leading centers in mathematical logic, and he remained there, working intensively
with students, colleagues and visitors until his death in 1983.
Tarski became recognized as one of the most important logicians of the twentieth cen-
tury through his many contributions to the areas of set theory, model theory, the semantics
of formal languages, decidable theories and decision procedures, undecidable theories,
universal algebra, axiomatics of geometry, and algebraic logic. What, in all that, are the
connections with computer science? When Anita started working on the biography-which
only later became a joint project-she asked me and some of my colleagues exactly that
question, and my response was: none. In contrast to that-as she said at the conclusion
of her Warsaw lecture-John Etchemendy (my colleague in Philosophy at Stanford, and
now the Provost of the University) responded: “You see those big shiny Oracle towers on
Highway 101? They would never have been built without Tarski’s work on the recursive
definitions of satisfaction and truth.”2 It took me a while to see in what sense that was
right. Indeed-as I was to learn-there is much, much more to say about his influence on
computer science, and that’s the subject of my talk today. I owe a lot to a number of
colleagues in the logic and computer science areas for pointing me in the right directions
in which to pursue this and also for providing me with very helpful specific information.3
Alfred Tarski: Life and Logic was written for a general audience; the biographical
material is interspersed with interludes that try to give a substantive yet accessible idea of
Tarski’s main accomplishments. Still, given the kind of book it is, we could not go into
great detail about his achievements, and in particular could only touch on the relationship
of his work to computer science. Before enlarging on that subject now, I want to tell a
2 For those who may not know what the “big shiny Oracle towers” are, the reference is to the
headquarters of Oracle Corporation on the Redwood Shores area of the San Francisco Peninsula. A
duly shiny photograph of a few of these towers may be found at http://en.wikipedia.org/wiki/Image:
OracleCorporationHQ.png.
3 I am most indebted in this respect to Phokion Kolaitis. Besides him I have also received useful comments
from Michael Beeson, Bruno Buchberger, George Collins, John Etchemendy, Donald Knuth, Janos
Makowsky, Victor Marek, Ursula Martin, John Mitchell, Vaughan Pratt, Natarajan Shankar, and Adam
Strzebonski. And finally, I would like to thank the two anonymous referees for a number of helpful
corrections.
Tarski’s Influence on Computer Science 393
story that is in our biography [14, pp. 220–230], and is in many respects revelatory of his
own attitude towards the connection.
I had the good fortune to be Tarski’s student in the 1950s when he was beginning the
systematic development of model theory and algebras of logic. In 1957, the year that I
finished my Ph.D., a month-long Summer Institute in Symbolic Logic was held at Cornell
University. That proved to be a legendary meeting; in the words of Anil Nerode: “There
has been nothing else in logic remotely comparable.” What the Cornell conference did
was to bring together for the first time, leaders, up-and-coming researchers, and students
in all the main areas of logic, namely model theory, set theory, recursion theory, and proof
theory. Besides Tarski, the top people there—along with their coteries—were Alonzo
Church, Stephen Kleene, Willard Quine, Barkley Rosser, and, in the next generation,
Abraham Robinson and George Kreisel. The organization of the meeting itself had been
inspired by the mathematician Paul Halmos, who, independently of Tarski, had developed
another approach to the algebra of first-order logic. As Halmos wrote about it in his
Automathography [23, p. 215]:
There weren’t many conferences, jamborees, colloquia in those days and the few that existed were
treasured. . . . I decided it would be nice to have one in logic, particularly if it were at least partly
algebraic.
And, “nice” it was.

In addition to the four main areas, the Cornell logic conference was the first to include
many speakers from the emerging field of computer science, the theoretical foundations
of which had been laid in the 1930s by Gödel, Church, Turing, Post, and Kleene. The
connections between the theory and application of computation began toward the end
of World War II when the first large scale electronic digital computers were built. At
that point, for each kind of application, the hardware had to be programmed by hand,
a long and arduous task. John von Neumann was instrumental in demonstrating how to
circumvent that process by introducing the first form of software.
By 1957, companies such as IBM and Remington Rand were producing the first
generation of commercial electronic computers, and the high-level programming lan-
guage FORTRAN had become established as an industry standard. Some-but by no
means all-logicians were quick to grasp the implications of these developments. At the
Cornell meeting, Rosser gave a talk on the relation between Turing machines and actual
computers; Church gave a series of talks on the logical synthesis of switching circuits
for computer hardware; and Abraham Robinson spoke on theorem proving as done by
man and machine. Among the younger contributors, Michael Rabin and Dana Scott
spoke about finite automata, and Martin Davis talked about his implementation on the
“Johnniac” computer (at the Institute for Advanced Study) of a decision procedure for the
arithmetic of the integers under addition-a procedure that had been discovered in 1930 by
Tarski’s student Mojżesz Presburger in his Warsaw seminar.
On the industry side, IBM and some of the other companies employed a number of
researchers with backgrounds in mathematics and logic, and these people turned out
in large numbers at Cornell, both to listen and to speak. There were fifteen talks given
by researchers from IBM, many of them demonstrating the utility of FORTRAN-like
programs for solving problems of potential interest to logicians. In particular, the talk by
George Collins—a former student of Rosser’s—on the implementation of parts of Tarski’s
decision procedure for the algebra of real numbers on an IBM 704 should have caught
394 S. Feferman
Tarski’s attention because it suggested possible practical applications of his procedure.

But a few years ago, when I asked Collins about Tarski’s reaction to that, he said: “He
didn’t show any appreciation for my work, either then or later. I was somewhat surprised
and disappointed.” It is indeed surprising that-despite Tarski’s own recognition of the
importance and systematic pursuit of the decision problem for various algebraic theories,
he did not evince the least bit of interest in the practical computational applications of
those problems for which a decision procedure had been found. And, he didn’t even seem
to be interested either in the work of Rabin and Scott, both of whom were high on his list
of favorites, Scott as a former student and collaborator, and Rabin as someone he wanted
to bring to Berkeley.
Let’s look at what Tarski and Collins were up to in more detail. At heart, Tarski’s
decision procedure rests on the solution of an algebraic problem for the reals, i.e., the
ordered field R, +, ×, <, 0, 1!. Tarski’s procedure uses the method of elimination of
quantifiers to associate with each first-order formula of the language of the reals an
equivalent quantifier-free formula; those without free variables are then easily decided.
The procedure reduces to determining for each system P of polynomial equations and
inequalities and one of its variables x, whether or not there exists a common real solution
x; the answer is to be expressed in terms of the coefficients of the polynomials involved
and the remaining variables. Algorithms for special cases of this problem go back through
the history of algebra. Tarski’s procedure generalizes one due to Sturm for computing
the number of roots of a real polynomial in a given interval. On the face of it, Tarski’s
procedure is non-elementary in time complexity, i.e., greater than all finite towers of
powers of 2, and so it was important for Collins to find a more feasible procedure than
the one that he had talked about at the 1957 Cornell meeting. Further improvements
on Tarski’s procedure by Abraham Seidenberg and later by Paul Cohen didn’t really
help much in that respect. In the meantime, Collins was working on various aspects
of computer algebra and in 1974 and 1975 he published [9, 10] a new method of
doubly exponential upper bound complexity called Cylindrical Algebraic Decomposition
(CAD). Incidentally, this was during a year that he was visiting Stanford University from
Wisconsin.4
The most comprehensive source of information on the development of the CAD
procedure and related work is [7].5 Here are a few of the things I learned from that
invaluable volume. In the first stage pursuant to quantifier elimination, the CAD algorithm
takes all the polynomials in the matrix of a prenex formula ϕ with a total of m free and
bound variables, and outputs a cell decomposition in Rm , on each cell of which each of the
given polynomials is sign invariant; furthermore, the cells are arranged in cylinders. The
QE part of the algorithm uses the output of the CAD algorithm to determine which cells
of the decomposition satisfy the matrix of ϕ in order to eliminate the bound variables.
The first implementation of the CAD method was made in 1979–1980 by Collins’ student
Dennis S. Arnon. In 1991, Collins and another of his students, Hoon Hong, published a
4 Collins reports [7, p. 86] a communication from Leonard Monk in 1974 stating that he and Bob Solovay
had obtained a triply exponential upper bound decision procedure for real algebra, though not a quantifier
elimination procedure. Fischer and Rabin say (op. cit., p. 124) that Solovay found a doubly exponential
upper bound, based on Monk’s work.
5 This includes a reprint of Tarski’s “A decision method for elementary algebra and geometry” [39].
substantial improvement for various examples in practice, though not in complexity upper
bound, requiring only partial CAD [11]. This was subsequently implemented by Hong
under the acronym QEPCAD. The Caviness and Johnson volume presents a number of
applications, including polynomial optimization, polynomial best approximation in lower
degree (by n−2 degree polynomials), the topology of semi-algebraic sets, algebraic curve
display, and robot motion planning. By the way, the system Mathematica implements
another form of CAD, according to Adam Strzebonski of Wolfram Research, Inc.
To round out the complexity picture, Fischer and Rabin [17] gave an EXPTIME lower
bound of the form 2cn for deciding for sentences of length n whether or not they are
true in the reals, no matter what algorithm is used; the same applies even with non-
deterministic algorithms, such as via proof systems. They also showed that the cut-point
by which EXPTIME sets in, i.e., the least n0 such that for all inputs of length n ≥ n0 , at
least 2cn steps are needed, is not larger than the length of the given algorithm or axiom
system for proofs. Thus real algebra is definitely infeasible on sufficiently large, though
not exceptionally large inputs. The applications mentioned above are in a gray area with
relatively small numbers of variables, where feasibility in practice depends on the specific
nature of the problems dealt with. As for space complexity, there is a PSPACE lower
bound on the theory of the reals, as a consequence of a result of Stockmeyer’s. Ben-
Or, Kozen and Reif [4] established an EXPSPACE upper bound and conjectured that the
set of true first-order sentences of the reals is EXPSPACE-complete. The exact time and
space complexities of this set are to this date an open problem (Phokion Kolaitis, personal
communication).6
Tarski’s own route to the decision problem for the reals began in the mid-1920s with
his development of an elegant first-order axiomatization of geometry [44]. One of his
main goals was to prove the completeness of this axiomatization, and that led him to
consider its interpretation in the first-order theory of the reals. Tarski recognized that
the method of eliminating quantifiers that had been initiated by Leopold Löwenheim
and then applied by Thoralf Skolem and C.H. Langford was—when it succeeded—a
way of determining all the complete extensions of a first-order axiom system—and in
particular of proving the completeness of complete systems. In the latter part of the
twenties Tarski ran the “exercise sessions” for the seminar at Warsaw University led by
the logic professor, Jan Łukasiewicz, and he used the opportunity to systematically pursue
the method of elimination of quantifiers. As an “exercise”, Tarski suggested to one of the
students, Mojżesz Presburger, that he find an elimination-of-quantifiers procedure for the
additive theory of natural numbers, i.e., for the structure N, +, <, 0, 1!. In that case, full
quantifier-elimination is not possible, but can be carried out in a definitional extension of
its language, obtained by adding as atomic formulas all those of the form x ≡ y(mod m)
for each m = 2, 3, 4, . . .. Mathematically, the procedure comes down to solving a system
of simultaneous congruences and thus the Chinese remainder theorem. Presburger’s result
served as his master’s thesis in 1928 and it was published a year later [33]. This slim paper
of nine pages was to be his sole work in logic; after that he went to work in the insurance
6 Just
minutes before my lecture, I learned from Prakash Panangaden that John Canny (U.C. Berkeley
School of Engineering) proved [6] that the existential theory of the reals is in PSPACE.
396 S. Feferman
industry. Some people think Presburger should have received the Ph.D. for that work, but
it has to be admitted that its significance was not realized until much later.7
The set of first-order truths of the additive structure of natural numbers is called
Presburger Arithmetic. As I mentioned earlier, Martin Davis presented his work on
programming the Presburger procedure on the Johnniac at the Cornell conference in
1957. That was long before Fischer and Rabin [17] showed that there is a doubly-
exponential time lower bound on any algorithmic procedure for Presburger Arithmetic,
including non-deterministic ones. If Martin had known that, he might not even have tried,
even with today’s computers.8 On the other hand, such lower bounds tell us little about
the feasibility in practice of deciding relatively short statements. As to upper bounds,
Presburger’s own procedure is non-elementary; this was improved to triply-exponential
by Derek Oppen [32]. A search on “Google Scholar” came up with a number of references
to Presburger Arithmetic. Near the top are applications to the symbolic model checking
of infinite state systems [5] and proving safety properties of infinite state systems [20];
further applications via combination decision procedures are indicated in [37].
Let’s return to Tarski’s own work on elimination of quantifiers for the elementary (i.e.,
first-order) theory of real numbers: although it was obtained by 1930 and he considered
it to be one of his two most important results (the other being his theory of truth), it’s
surprising he didn’t get around to preparing it for publication until 1939. That was under
the title, “The completeness of elementary algebra and geometry”9 for a new series on
metamathematics planned by a Parisian publisher, but the actual publication was disrupted
by the German invasion of France in 1940. As Tarski later wrote: “Two sets of page proofs
which are in my possession seem to be the only material remainders of that venture.”
The next time he got around to working on its publication was in 1948 when his friend
and colleague J.C.C. McKinsey was at the RAND Corporation in Santa Monica. My
guess is that McKinsey suggested to his superiors that there would be potential value
to applying Tarski’s procedure to the computer calculation of optimal strategies in certain
games. (Game theory was in those years a very popular subject at RAND.) However,
any implementation would first require writing up its theoretical details in full. Working
under Tarski’s supervision, McKinsey took on the job, revising the 1939 manuscript in
its entirety. That came out as a RAND Report under the new title “A decision procedure
for elementary algebra and geometry” in 1948; it was finally brought out publicly 3 years
later by UC Press as a second edition [39]. The change in title from 1939 to 1948/1951
corresponds to a change in aims, from completeness to decidability. (By the way, a lightly
edited version of the 1939 page proofs eventually appeared under the original title in 1967
in France).
Though Tarski may not have been interested in actual computation at any time in this
entire history, he was interested in mathematical applications of his procedure. In fact,
7A sad coda to this story is that Presburger, a Jew, perished in the Holocaust in 1943.
8 Shankar [37] takes as an epigram a quote from Davis [12] re his experiment with Presburger Arithmetic:
“Its great triumph was to prove that the sum of two even numbers is even.” A second epigram from the
same source, quoting Hao Wang, is that: “The most interesting lesson from these results is perhaps that
even in a fairly rich domain, the theorems actually proved are mostly ones which call on a very small
portion of the available resources . . . ”
9 By the elementary theory of a structure, Tarski means the set of its first-order truths.
one of Tarski’s strongest motivations throughout his career was to attract mathematicians
to the results of work in logic, and he often did this by reformulating the results in a
way that he thought would be more digestible by mathematicians. One side result he
noticed about his elimination-of-quantifiers argument for the first-order theory of the real
numbers is that every definable set has the form of a union of a finite number of intervals
(not necessarily proper) with algebraic end-points. He used this to illustrate the general
concept of definable set of elements in a structure. At the outset of his 1931 paper on
definable sets of real numbers [42] he said that mathematicians in general don’t like to
deal with the notion of definability. One reason is that used informally it can lead to
contradictions, like the paradox of Richard; that uses an enumeration in English (say)
of all the real numbers definable in English, to define (in English) a real number not
in that enumeration, by diagonalization. Another reason for mathematicians’ aversion
mentioned by Tarski is that mathematicians think the notion of definability is not really
part of mathematics. In a way, he agrees, for he says that
The problems of making [the meaning of definability] more precise, of removing the confusions
and misunderstandings connected with it, and of establishing its fundamental properties belong to
another branch of science-metamathematics. [Italics mine]
In fact, he says, he has “found a general method which allows us to construct a rigorous
metamathematical definition of this notion”.
But then
by analyzing the definition thus obtained it proves to be possible . . . to replace it by [one]
formulated exclusively in mathematical terms. Under this new definition the notion of definability
does not differ from other mathematical notions and need not arouse either fears or doubts; it can
be discussed entirely within the domain of normal mathematical reasoning. [Italics mine]
The metamathematical explanation of definability in Tarski’s 1931 paper is given in

terms of the notion of satisfaction, whose definition is only indicated there. Under the
mathematical definition, on the other hand, the definable sets and relations are simply
those generated from certain primitive sets of finite sequences corresponding to the
atomic formulas, by means of the Boolean operations and the operation of projection.
Although the 1931 paper concentrates on the concept of definability in a structure, in a
footnote to the metamathematical explanation it is stated that “an analogous method can
be successfully applied to define other concepts in the field of metamathematics, e.g., that
of true sentence . . . ”
Tarski later spelled this out in his famous 1935 paper “Der Wahrheitsbegriff in den
formalisierten Sprachen” (The concept of truth in formalized languages, [41]).10 Some
regard this work as one of the most important instances of conceptual analysis in
twentieth century logic, while others think he was merely belaboring the obvious. After
all, logicians like Löwenheim, Skolem and Gödel had been confidently using the notions
of satisfaction and truth in a structure in an informal sense for years before Tarski’s
work and an explicit definition was not deemed to be necessary, unlike, for example,
the conceptual analysis of computability by Turing. I have to agree that there is some
10 Itwas not until 1957, in a paper with Robert L. Vaught [45], that Tarski explicitly presented these
notions as those of satisfaction and truth in a structure. See the discussion by Hodges [26] and Feferman
[15] of the relationship of that to the “Wahrheitsbegriff” paper.
398 S. Feferman
justice to the criticism since the definitions are practically forced on us, once one attends
to providing them at all. But even if that’s granted, Tarski’s explication of these concepts
has proved to be important as a paradigm for all the work in recent years on the semantics
of a great variety of formal languages.
In particular, the influence of Tarski on the semantics of programming languages is so
pervasive that to detail it would require an entire presentation in itself. Let me mention just
one example, namely that of the semantics of the lambda calculus and its extensions via
domain theory, as developed by Dana Scott and his followers. This happens to connect
with the item in Tarski’s list of publications that is most cited in the computer science
literature, namely his lattice-theoretic fixpoint theorem [40], which is an elegant abstract
formulation of the essential characteristic of definition by recursion.11 There is also a
significant personal connection: Scott began his studies in logic at Berkeley in the early
50s while still an undergraduate. His unusual abilities were soon recognized and he
quickly moved on to graduate classes and seminars with Tarski and became part of the
group that surrounded him, including me and Richard Montague; so it was at that time
that we became friends. Scott was clearly in line to do a Ph. D. with Tarski, but they had a
falling out for reasons explained in our biography of Tarski [14]. Upset by that, Scott left
for Princeton where he finished with a Ph.D. under Alonzo Church. But it was not long
before the relationship between them was mended to the point that Tarski could say to him,
“I hope I can call you my student,” and rightly so: not only did Scott’s thesis deal with a
problem that had been proposed by Tarski, but all of Scott’s work is in the best Tarskian
tradition of breadth, rigor, clarity of exposition and clarity of purpose. And, like Tarski,
he prefers set-theoretic and algebraic methods, of which the domain-theoretic approach
to the semantics of type-free functional programming languages is a perfect example. So
Tarski’s influence on computer science manifests itself here at just one remove, though of
course Scott’s contribution, beginning in 1976 [36] with the construction of a domain D
isomorphic to D → D, is completely novel.12
Satisfaction and truth in a structure are the basic notions of model theory, whose
systematic development in the 1950s is initially largely due to Tarski and his school.
The notions are relative to a formal language, which is usually taken to be first-order
(FO), because of the many happy properties of FO logic such as that of compactness
(cf. the texts by Chang and Keisler or Hodges). But other kinds of logics in which, e.g.,
compactness fails, turned out to be partially susceptible to useful model-theoretic methods
as shown by the greatly varied contributions to the collection Model-Theoretic Logics
[3]. Among these are the model-theory of infinitary languages as well as second-order
and higher-order languages. For computer science, a great variety of finite structures,
such as various classes of graphs, arise naturally, and it was discovered that a number of
11 Tarski proved that every monotonic function over a complete lattice has a complete lattice of fixed
points, and hence a least fixed point. This is a generalization of a much earlier joint result of Knaster and
Tarski and so is sometimes referred to as the Knaster-Tarski theorem. A related result used in applications
is that every continuous function on a complete lattice has a least fixed point; credit for it is unclear, and
thus it is considered a “folk theorem”. The history of these and other fixed point theorems relevant to
computer science is surveyed in [29].
12 Scott informed me that his use of lattice fixed points was initiated in the fall of 1969 in work
with Christopher Strachey and exposed in many lectures in Oxford while on leave there. For further
developments and a large bibliography see [21].
questions in complexity theory may be framed as questions in finite model theory (cf.,
e.g. [13]). In addition to first-order logic (FOL) and its finite variable fragments, other
logics that have proved to be useful in finite model theory are finite-variable infinitary
logics, monadic second-order logic (MSOL) and its fragments, and certain fixed point
logics such as Datalog and least fixed-point logic LFP.
Recently there has been a surge of very interesting work on analogues in finite FO
model theory to a class of general results called preservation theorems in classical FO
model theory. The newest and most exciting of these is due to Ben Rossman [35]. So I
can limit myself to explaining the general nature of the main results.
Given a relation R between structures and a sentence ϕ, we say that ϕ is preserved
under R if whenever M satisfies ϕ and N is in the relation R to M, then N satisfies ϕ.
The results from classical FO model theory characterize up to logical equivalence, the
form of sentences preserved under various R. The most famous ones are the following,
all from the 1950s.
• EPT (Łoś-Tarski). ϕ is preserved under extensions iff ϕ is equivalent to an existential
sentence.
• OHPT (Lyndon). ϕ is preserved under onto-homomorphisms iff ϕ is equivalent to a
positive sentence.
• HPT (Łoś-Tarski-Lyndon). ϕ is preserved under (into-)homomorphisms iff ϕ is
equivalent to an existential positive sentence.
The finite analogues of these results are obtained by restricting to finite M and N.
The ‘if’ directions of course hold in all the finite versions, but the ‘only if’ analogues of
the Extension Preservation Theorem (EPT) and the Onto-Homomorphism Preservation
Theorem (OHPT) are known to fail. In particular, the failure of the finite analogue of
EPT is due to Bill Tait in 1959, who thereby disproved a conjecture of Scott and Suppes;
Tait’s result was rediscovered by Gurevich and Shelah in 1984. The failure of OHPT in
the finite is due to Rosen [34].13 What Rossman [35] has proved, surprisingly, is that HPT
holds in the finite. There are interesting relations to Datalog programs, which are given by
existential positive FO inductive definitions. A Datalog formula may thus be considered as
an infinitary disjunction of existential positive FO formulas. Using a simple compactness
argument, Rossman’s result about HPT in the finite also implies the theorem of Ajtai and
Gurevich [1] that on finite structures, if a Datalog sentence is equivalent to a FO sentence
then it is equivalent to a single existential positive sentence, exactly those preserved by
homomorphisms.14 The failure of the Łoś-Tarski characterization in the finite shows that
preservation theorems do not in general relativize from a class to a subclass. Thus it is
also of interest to ask for which classes of finite structures HPT holds. This had already
been investigated by Atserias, Dawar and Kolaitis [2] in which one of the main results is
that HPT holds for every class of finite structures of bounded tree width; another result is
that HPT holds for the class of all planar graphs.
13 Lyndon’s famous positivity theorem implies OHPT. Ajtai and Gurevich, and then Stolboushkin in a
simpler way, proved failure in the finite of the positivity theorem, but their constructions did not prove
failure in the finite of OHPT.
14 According to Rossman, the implication was known to hold prior to his discovery.
400 S. Feferman
At the Tarski Centenary Conference held in Warsaw in 2001, Johann A. Makowsky

presented a survey of applications of another kind of preservation result that goes under
the heading of the Feferman-Vaught Theorem [31]. What Vaught and I had shown in our
joint 1959 paper [16] was that for a great variety of sum and product operations O on
structures Mi (i ∈ I ), the first-order properties of
M = O Mi |i ∈ I !
are determined by the first order properties of each Mi together with the monadic
second-order properties of a structure on the index set I . It follows that elementary
equivalence between structures is preserved under such operations O. In later work,
Läuchli, Gurevich and Shelah extended our reduction-to-factors theorem to monadic
second-order properties. In his paper, Makowsky gives a unified presentation of this work
with emphasis on its algorithmic applications, in particular to splitting theorems for graph
polynomials. I’ll have to leave it at that, since it would take too much time to try to go
further into that here.
The final thing I want to tell something about is the connection of Tarski’s ideas
and work with database theory. Here it is not a matter of direct influence but rather of
the pervasiveness of his approach to things, since the development of database theory
apparently proceeded quite independently. Jan Van den Bussche has written an excellent
survey [46] of the connections, which I urge you to read; here are a few of the high
points. Codd [8] introduced a relational algebra for expressing a class of generic (i.e.,
isomorphism invariant) queries on databases; he also proved that the queries expressible
in his relational algebra are exactly those that are domain independent and definable
in FO logic. For those who know Tarski’s work on relation algebra, cylindric algebras
and algebraic logic more generally, the immediate question to raise is the nature of
the connection.15 (You can find a quick introduction to Tarski’s work in this respect in
Interlude VI of our biography [14].16 ) I view Tarski’s work on algebraic logic as part of
his general effort to reformulate logic in mathematical as opposed to metamathematical
terms, in the hopes of thus making logic of greater mathematical interest. Tarski had done
much work in the 1930s on Boolean algebras, of which algebras of sets and algebras of
propositions (up to equivalence) are specific cases. Stone’s representation theorem for
Boolean algebras showed that every abstract BA is isomorphic to a concrete one in the
sense of fields of sets, and in that sense the equational axioms of BA are complete. With
Tarski’s 1941 paper “On the calculus of relations” [38] he single-handedly revived and
advanced the nineteenth century work on binary relations by Peirce and Schröder, and
introduced an elegant finite equational axiom system for relation algebra, from which
all known special cases of valid relational identities could be deduced. However, it was
shown by Roger Lyndon in 1950 that there are non-representable relation algebras, so
15 According to Van den Bussche (personal communication) the first people from the database community
to recognize the connection between Codd’s relational algebra and Tarski’s cylindric algebras were Witold
Lipski and his student Tomasz Imielinski, in a talk given at the very first edition of PODS (the ACM
Symposium on Principles of Database Systems), held in Los Angeles, March 29–31, 1982. Their work
was later published in Imielinski and Lipski [27].
16 Some other applications to computer science — not discussed here — of Tarski’s work on relation
algebra are indicated on p. 339 and its footnote 4 of that interlude.

Tarski’s axioms are not complete; later, Donald Monk proved (in 1964) that there is no
finite axiomatization of the valid equations in the language of these algebras. This is
related to the fact that what can be expressed in relation algebra is exactly what can be
expressed about binary predicates in 3-variable FOL.17 See [30] for more on the history
of relation algebras.
Given the weakness of relation algebra, in the early 1950s Tarski introduced the idea of
cylindric algebras (CAs) of dimension k for any k ≥ 2, finite or infinite. (NB: Cylindric
Algebras have nothing to do with Cylindrical Algebraic Decomposition.) In addition to
the Boolean operations, these algebras use operations Cn of cylindrification for each n <
k and diagonal constants dn,m for each n, m < k. The concrete interpretations are given
by fields of subsets of a k-ary space U k , with the Cn interpreted as cylindrification along
the nth axis, and the dn,m as the set of k-tuples in U for which the nth and mth terms
are equal. Thus k-dimensional CAs abstract k-variable FOL with identity. The theory of
CAs was extensively developed by students and colleagues of Tarski and the results are
exposited in the volumes by Henkin, Monk and Tarski [24, 25]. It turns out that there are
non-representable CAs for every k ≥ 2, finite or infinite, but Henkin and Tarski showed
that “locally finite” CAs are representable for every infinite k. A CA is called locally
finite if for each element a of the algebra, Cn (a) = a for all but a finite number of n < k.
The local-finiteness condition corresponds to each formula in FOL having at most a finite
number of free variables, and the representation theorem for infinite dimensional locally
finite CAs corresponds to the completeness theorem for FOL with identity.
As Van den Bussche points out in [46], the language of ω-CAs provides an alternative
to Codd’s relational query language, and that of k-CAs for k finite is an alternative to
queries definable in FOL with at most k distinct variables. But how does Codd’s language
match up with that of Relation Algebras (RAs)? In [43] it is shown that adjunction
of a suitable “pairing axiom” to RA makes it as strong as FOL. It turns out that the
corresponding idea has been developed in the case of database theory by Gyssens, Saxton
and Van Gucht [22] using “tagging” operations, giving a form RA= that simulates Codd’s
relational algebra.18
So, does that justify John Etchemendy’s statement that the shiny Oracle towers on
Hwy 101 wouldn’t be there without Tarski’s recursive definition of satisfaction and truth?
It would be more accurate to say that the Oracle towers wouldn’t be there without the
theoretical development of database theory, and that wouldn’t be there without rethinking
the model theory of first-order logic in relation-algebraic and/or cylindric-algebraic terms,
and that wouldn’t be there without Tarski’s promotion of both model theory and algebraic
logic. Does Larry Ellison know who Tarski is or anything about his work? At the time of
my lecture, I wondered whether Ellison even knew who Codd was or the whole theoretical
development of database theory, without which the Oracle towers would indeed not be
there. I learned subsequently from Jan Van den Bussche that not only did Ellison know
about Codd’s work but he marks the reading of Codd’s seminal paper as the starting
point leading to the Oracle Corporation; cf. his biography given by the “Academy of
Achievement” at http://www.achievement.org/autodoc/page/ell0bio-1/. Actually, Codd
17 Cf.
the papers [18, 19] dealing with expressibility/inexpressibility in Tarski’s algebraic framework.
18 Van den Bussche’s article concludes with a survey of some interesting connections to constraint
databases and geometric databases.
402 S. Feferman
himself didn’t refer to Tarski in his fundamental papers on database theory. But other
workers in the subject, such as Imielinski and Lipski, and later, Kanellakis, did; they were
well aware of the connection and brought explicit attention to it (cf., e.g. [28, p. 1085]).
In whatever way the claim is formulated, I think it is fair to say that Tarski’s ideas and the
approaches he promoted are so pervasive that even if his influence in this and the various
other areas of computer science about which I spoke was not direct it was there at the
base, and—to mix a metaphor—it was there in the air, and so the nature and importance
of his influence eminently deserves to be recognized.
References
1. Ajtai, M., Gurevich, Y.: Datalog vs. first-order logic. J. Comput. Syst. Sci. 49, 562–588 (1994)
2. Atserias, A., Dawar, A., Kolaitis, P.G.: On preservation under homomorphisms and unions of con-
junctive queries. In: 23rd ACM SIGMOD-SIGACT-SIGART Symposium on Principles of Database
Systems (PODS), pp. 319–329. ACM (2004)
3. Barwise, J., Feferman, S. (eds.): Model-Theoretic Logics. Springer, Berlin (1985)
4. Ben-Or, M., Kozen, D., Reif, J.H.: The complexity of elementary algebra and geometry. J. Comput.
Syst. Sci. 32(2), 251–264 (1986)
5. Bultan, T., Gerber, R., Pugh, W.: Symbolic model checking of infinite state systems using Presburger
arithmetic. In: CAV’97: Proceedings of the 9th International Conference on Computer Aided
Verification, pp. 400–411. Springer, London (1997)
6. Canny, J.F.: Some algebraic and geometric computations in PSPACE. In: ACM Symposium on Theory
of Computing, pp. 460–467. ACM, New York (1988)
7. Caviness, B.F., Johnson, J.R. (eds.): Quantifier Elimination and Cylindrical Algebraic Decomposition.
Springer Series in Symbolic Computation. Springer, Berlin (1998)
8. Codd, E.F.: A relational model of data for large shared data banks. Commun. ACM 13(6), 377–387
(1970)
9. Collins, G.E.: Quantifier elimination for real closed fields by cylindrical algebraic decomposition –
preliminary report. SIGSAM Bull. 8(3), 80–90 (1974)
10. Collins, G.E.: Quantifier elimination for real closed fields by cylindrical algebraic decomposition.
In: Proceedings Second GI Conference on Automata Theory and Formal Languages. Lecture Notes
in Computer Science, vol. 33, pp. 134–183. Springer, Berlin (1975) (Reprinted in: Caviness, B.F.,
Johnson, J.R. (eds.): Quantifier Elimination and Cylindrical Algebraic Decomposition. Springer
Series in Symbolic Computation, pp. 85–121. Springer, Berlin (1998))
11. Collins, G.E., Hong, H.: Partial cylindrical algebraic decomposition for quantifier elimination.
J. Symb. Comput. 12(3), 299–328 (1991) (Reprinted in: (Reprinted in: Caviness, B.F., Johnson, J.R.
(eds.): Quantifier Elimination and Cylindrical Algebraic Decomposition. Springer Series in Symbolic
Computation, pp. 174–200. Springer, Berlin (1998))
12. Davis, M.: The prehistory and early history of automated deduction. In: Siekmann, J., Wrightson, G.
(eds.) Automation of Reasoning 1: Classical Papers on Computational Logic 1957–1966, pp. 1–28.
Springer, Berlin (1983)
13. Ebbinghaus, H.D., Flum, J.: Finite Model Theory, 2nd edn. Springer, Berlin (1991)
14. Feferman, A.B., Feferman, S.: Alfred Tarski: Life and Logic. Cambridge University Press, New York
(2004)
15. Feferman, S.: Tarski’s conceptual analysis of semantical notions. In: Benmakhlouf, A. (ed.) Séman-
tique et épistémologie, pp. 79–108. Editions Le Fennec, Casablanca [distrib. J. Vrin, Paris] (2004)
16. Feferman, S., Vaught, R.L.: The first order properties of products of algebraic systems. Fundam. Math.
47, 57–103 (1959)
17. Fischer, M.J., Rabin, M.O.: Super-exponential complexity of Presburger arithmetic. In: Karp,
R. (ed.) Complexity of Computation. SIAM-AMS Proceedings, vol. 7 , pp. 27–42. American
Mathematical Society, Providence (1974) (Reprinted in: (Reprinted in: Caviness, B.F., Johnson, J.R.
(eds.): Quantifier Elimination and Cylindrical Algebraic Decomposition. Springer Series in Symbolic
Computation, pp. 122–135. Springer, Berlin (1998))
18. Formisano, A., Omodeo, E.G., Policriti, A.: Three-variable statements of set-pairing. Theor. Comput.
Sci. 322(1), 147–173 (2004)
19. Formisano, A., Omodeo, E.G., Policriti, A.: The axiom of elementary sets on the edge of Peircean
expressibility. J. Symb. Log. 70, 953–968 (2005)
20. Fribourg, L., Olsén, H.: Proving safety properties of infinite state systems by compilation into
Presburger arithmetic. In: Mazurkiewicz, A.W., Winkowski, J. (eds.) CONCUR. Lecture Notes in
Computer Science, vol. 1243, pp. 213–227. Springer, Berlin (1997)
21. Gierz, G., Hoffman, K.H., Keimel, K., Lawson, J.D., Mislove, M., Scott, D.S.: Continuous Lattices
and Domains. Encyclopedia of Mathematics and its Applications, vol. 93. Cambridge University
Press, Cambridge (2003)
22. Gyssens, M., Saxton, L.V., Van Gucht, D.: Tagging as an alternative to object creation. In: Freytag,
J.C., Maier, D., Vossen, G. (eds.) Query Processing for Advanced Database Systems, pp. 201–242.
Kaufmann, San Mateo (1994)
23. Halmos, P.: I Want to be a Mathematician: An Automathography. Springer, Berlin (1985)
24. Henkin, L., Monk, D., Tarski, A.: Cylindric Algebras, vol. 1. North-Holland, Amsterdam (1971)
25. Henkin, L., Monk, D., Tarski, A.: Cylindric Algebras, vol. 2. North-Holland, Amsterdam (1971)
26. Hodges, W.: Truth in a structure. Proc. Aristot. Soc. 86, 131–151 (1985–1986)
27. Imieliński, T., Lipski, W. Jr.: The relational model of data and cylindric algebras. J. Comput. Syst.
Sci. 28(1), 80–102 (1984)
28. Kanellakis, P.C.: Elements of relational database theory. In: van Leeuwen, J. (ed.) Handbook of
Theoretical Computer Science, Volume B: Formal Models and Semantics (B), pp. 1073–1156. MIT
Press, Cambridge (1990)
29. Lassez, J.L., Nguyen, V.L., Sonnenberg, E.A.: Fixed point theorems and semantics: a folk tale. Inf.
Process. Lett. 14, 112–116 (1982)
30. Maddux, R.D.: The origin of relation algebras in the development and axiomatization of the calculus
of relations. Stud. Logica 50(3–4), 421–455 (1991)
31. Makowsky, J.A.: Algorithmic uses of the Feferman-Vaught theorem. Ann. Pure Appl. Logic 126,
159–213 (2004)
2pn
32. Oppen, D.C.: A 22 upper bound on the complexity of Presburger arithmetic. J. Comput. Syst. Sci.
16, 323–332 (1978)
33. Presburger, M.: Über die Vollständigkeit eines gewissen Systems der Arithmetik ganzer Zahlen,
in welchem die Addition als einzige Operation hervortritt. Comptes Rendus du I congrs des
Mathmaticiens des Pays Slaves, 92–101 (1929)
34. Rosen, E.: Some aspects of model theory and finite structures. Bull. Symb. Log. 8, 380–403 (2002)
35. Rossman, B.: Existential positive types and preservation under homomorphisisms. In: Panangaden, P.
(ed.) Proceedings of the Twentieth Annual IEEE Symposium on Logic in Computer Science. LICS
2005, pp. 467–476. IEEE Computer Society Press, Washington (2005)
36. Scott, D.S.: Data types as lattices. SIAM J. Comput. 5, 522–587 (1976)
37. Shankar, N.: Little engines of proof. In: Eriksson, L.-H., Lindsay, P. (eds.) FME 2002: Formal
Methods – Getting IT Right, Copenhagen, pp. 1–20. Springer, Berlin (2002)
38. Tarski, A.: On the calculus of relations. J. Symb. Logic 6, 73–89 (1941)
39. Tarski, A.: A Decision Method for Elementary Algebra and Geometry, 2nd edn. University of
California Press, Berkeley (1951) (Reprinted in: Caviness, B.F., Johnson, J.R. (eds.): Quantifier
Elimination and Cylindrical Algebraic Decomposition. Springer Series in Symbolic Computation,
pp. 24–84. Springer, Berlin (1998))
40. Tarski, A.: A lattice-theoretical fixpoint theorem and its applications. Pac. J. Math. 5, 285–309 (1955)
41. Tarski, A.: The concept of truth in formalized languages. In: Corcoran, J. (ed.) Logic, Semantics,
Metamathematics. Papers from 1923 to 1938, 2nd edn., pp. 182–278. Oxford University Press, Oxford
(1983) (original date (1935))
42. Tarski, A.: On definable sets of real numbers. In: Corcoran, J. (ed.) Logic, Semantics, Metamathe-
matics. Papers from 1923 to 1938, 2nd edn., pp. 110–142. Oxford University Press, Oxford (1983)
(Original date (1931))
43. Tarski, A., Givant, S.: A Formalization of Set Theory Without Variables. Number 41 in Colloquium
Publications. American Mathematical Society, Providence (1987)
404 S. Feferman
44. Tarski, A., Givant, S.: Tarski’s system of geometry. Bull. Symb. Log. 5, 175–214 (1999)
45. Tarski, A., Vaught, R.L.: Arithmetical extensions of relational systems. Compos. Math. 13, 81–102
(1957)
46. Van den Bussche, J.: Applications of Alfred Tarski’s Ideas in Database Theory. In: International
Workshop on Computer Science Logic. Lecture Notes in Computer Science, vol. 2142, pp. 20–37.
The Absence of Multiple Universes
of Discourse in the 1936 Tarski
Consequence-Definition Paper
John Corcoran and José Miguel Sagüillo
Dedicated to Professor Roberto Torretti, philosopher of science,

historian of mathematics, teacher, friend, collaborator—on his
eightieth birthday.
Abstract This paper discusses the history of the confusion and controversies over
whether the definition of consequence presented in the 11-page Tarski consequence-
definition paper is based on a monistic fixed-universe framework—like Begriffsschrift
and Principia Mathematica. Monistic fixed-universe frameworks, common in pre-WWII
logic, keep the range of the individual variables fixed as ‘the class of all individuals’. The
contrary alternative is that the definition is predicated on a pluralistic multiple-universe
framework—like the Gödel incompleteness paper. A pluralistic multiple-universe frame-
work recognizes multiple universes of discourse serving as different ranges of the
individual variables in different interpretations—as in post-WWII model theory. In the
early 1960s, many logicians—mistakenly, as we show—held the ‘contrary alternative’
that Tarski had already adopted a Gödel-type, pluralistic, multiple-universe framework.
We explain that Tarski had not yet shifted out of the monistic, Frege-Russell, fixed-
universe paradigm. We further argue that between his Principia-influenced pre-WWII
Warsaw period and his model-theoretic post-WWII Berkeley period, Tarski’s philosophy
underwent many other radical changes.
Keywords Tarski · Logical consequence · Universe of discourse · Monistic

fixed-universe · Pluralistic multiple-universe
Mathematics Subject Classification (2000) Primary 03A05, 01A60; Secondary 03-03

406 J. Corcoran and J. M. Sagüillo
1 Introduction
In the long history of this text even what is obvious has often been overlooked. —Norman
Kretzmann on a passage in the Organon, Buffalo (1972).
In his now-famous 11-page consequence-definition paper [63],1 Tarski broke with

various established traditions in several ways. He focused on a ‘consequence’ relation
instead of a converse, i.e., an ‘implication’ relation: ‘is a consequence of’ replaces
‘implies’. He further separated himself by taking consequence to be a relation of an
individual to a ‘class’ instead of a relation of an individual to an individual.2 His
consequence cannot be expressed with a sentential connective. Even more, he chose his
consequence relation to concern sentences of a given fully interpreted language3 instead
of ‘propositional functions’, uninterpreted ‘formal sentences’, or ‘logical forms’—on one
hand—or abstract ‘propositions’, ‘judgments’, ‘statements’, or ‘thoughts’—on the other.
Moreover, his consequence was metalinguistic relative to the object-language sentences
it related. Thus, it could not be ‘iterated’; it was impossible to say that one consequence
sentence was a consequence of a class of object-language sentences—or conversely, that
one object-language sentence was a consequence of a class of consequence sentences.
Some of the ‘paradoxes’ were thus defused. For Tarski, one given sentence of a fully
interpreted language is, or is not, a consequence of a given class of sentences of the
same language.4 Each language has its own consequence relation—just as in earlier Tarski
works each language had its own rules of inference.
In addition for Tarski, the respective truth-values of the given sentence and those in
the given class are irrelevant to determining consequence, except of course in case the
given sentence is false and all those in the given class are true. For example, in some
cases a given true sentence is not a consequence of a given class of true sentences;
Tarski’s consequence is not a material-consequence relation. And for Tarski, [sc. logical]
consequence is formal in the traditional sense: if the conclusion of a given argument is
a consequence of its premises, the same relation holds between the conclusion and the
premises of any other argument in the same form.5 Finally, Tarski avoided traditional
logic’s reference to premise-conclusion arguments: he left it entirely up to traditional
1 Unless explicitly said otherwise, ‘consequence’ is used in a ‘logical’ sense so that the expression ‘logical
consequence’ is redundant. Thus, we are not discussing the ‘consequences’ of events, of actions, or of
inactions, etc.
2 ‘Class’ is the word Tarski used [63, pp. 414–419]. But, although others had distinguished ‘classes’ from
multiplicities such as ‘sets’, ‘aggregates’, ‘collections’, ‘extensions’, and the like, no distinction is implied
in [63] or elsewhere in his collection [69].
3 One of the History and Philosophy of Logic referees required us to give more emphasis to this point
than it had received in the submitted version. We are grateful that this needed improvement was brought
to light; it helps us to highlight a major philosophical and technical change between Tarski’s pre-WWII
attitude and his post-WWII thinking. After the war uninterpreted constants became legitimate and were
assigned an important role. Contrast [57].
4 Tarski did not dwell on the philosophical ramifications of this ‘relevance’ or ‘pertinence’ requirement,
which might have been foreshadowed by Aristotle. But others have noted its importance (Corcoran [8]).
5 Tarski is not as explicit as one might wish. This point is not made in any single sentence though it can
be gleaned from the paragraph that begins on p. 414. Incidentally, this is the only place in [69] that uses
‘form’ in the required sense of logical form. See Corcoran’s [14] piece ‘Logical form’ in [1].
The Absence of Multiple Universes of Discourse . . . 407
logicians to discern that an argument’s being valid in their sense is its conclusion being
a consequence of its premises in Tarski’s terminology. They were left to infer that
Tarski was defining argument validity: Tarski was silent on the connection between
defining consequence and defining validity. Of course, breaking with one tradition is
frequently keeping with some other tradition. In a sense our paper is about whether in his
consequence-definition paper he had already broken with the Frege-Russell fixed-universe
paradigm.
Although we discuss the details, background, and ramifications of Tarski’s definition;
as stated in the abstract, our central concern is the issue of whether his definition has a
monistic framework that keeps the range of the individual variables fixed as ‘the class
of all individuals’, or ‘the actual universe of things’ as in Principia Mathematica, for
every interpretation, or whether—to the contrary—it has a pluralistic framework which,
like that of the Gödel incompleteness paper [27, translation in 78], recognizes multiple
universes of discourse serving as different ranges of the individual variables in different
interpretations (‘sequences’ or ‘models’).
The 1936 consequence-definition paper did not become widely available until 1956
when J.H. Woodger translated it into English [64, 74]. Previously it had appeared in Polish
and German versions [72, p. 409]. Church [6, p. 325], who refers only to the Polish and
German versions, implied that Tarski [63] presupposed a Gödel-type changeable-universe
framework. In the early 1960s, after the English had been available for years, nevertheless
Corcoran and many other logicians still concurred with the Church interpretation. Now to
the contrary, it is widely held that Tarski had not yet fully shifted out of the monistic
Frege-Russell paradigm (Mancosu [41]). We too support the monistic, one-universe
interpretation of the Tarski consequence-definition paper [63]. In other words, we argue
for the absence of multiple universes of discourse in the Tarski consequence-definition
paper [63].
After the issue of interpreting the 1936 paper came to Corcoran’s attention in 1964, he
reread the paper and felt forced to recognize his mistake. From then on his understanding
has been that, contrary to Church and to his own former belief, Tarski [63] presupposes a
monistic Principia-type fixed-universe framework within which the individual variables
have a range which is fixed throughout and which does not change from interpretation
to interpretation. The contrast between the two approaches to explicating ‘logical truth’
and ‘logical consequence’ is explained in Chapter 4 of Quine’s Philosophy of Logic [47],
where the older fixed-universe framework is preferred.
The issue of whether Tarski’s paper [63] employs a fixed or a changeable range of
individual variables is entirely separate from the question of whether Tarski in 1936 or
before acknowledged the fact, already noted by Aristotle in Posterior Analytics (Bk. A,
Ch. 10, esp. 76b10 ff.) that each science has its own separate domain of investigation,
sometimes called its subject matter, or its genus (e.g., Sagüillo [51, p. 268]). As Sagüillo
has amply established, there are many well-known formalizations of sciences in which
the individual variables range over a universe of discourse that is wider than the domain
of investigation of the science. One example is Tarski’s own 1929 geometry of solids,
where—as Tarski says—the term ‘individual’ is used in the same way as in Principia
Mathematica. But the domain of investigation is not the range of the individual variables;
rather it is ‘the class of all [sc. geometrical] solids’ [72, pp. 24–29].
However, the issue is related to, but not determined by, the question of whether Tarski
was aware before 1936 of the De Morgan-Boole concept of the universe of a given
discourse (Boole [3, p. 42]; Corcoran [15, p. 242], [16, p. 941]), and if aware, whether he
accepted its theoretical importance in logic.
Underlying the historical and hermeneutic issue of how the 1936 paper is faithfully
interpreted, there is a list of broader and more important historical and philosophical
issues. Near the top of this list, we would put the question of whether, as De Morgan
and Boole held, we have the capacity to discuss a limited domain of investigation without
referring to things outside of the domain and in particular without referring to the entire
universe of individuals. For example, is it possible to say that every person is mortal using
a sentence such as ‘for every person x, x is mortal’ with the common noun ‘person’
indicating the range of the variable and without in any way referring to objects other than
persons? Or to the contrary, in order to say that every person is mortal is it necessary to
use a sentence such as ‘for every [object] x, if x is a person, then x is mortal’ with the
‘universe of individuals’ as the range of the variable? Do we really have to talk about
every individual in order to make a statement about every person?6
The view held by Frege and Russell is that individual variables necessarily range over
the entire universe of individuals and any restriction of subject is to be accomplished by
means of a qualifying condition. For example, to be more explicit, on their view, to say
that every person is mortal we would be required to use a sentence such as ‘for every
individual x, if x is a person, then x is mortal’, where the common noun ‘individual’
indicates that the range of the variable is taken to be the class of all individuals. In practice,
the quantifier phrase ‘for every individual x’ is often written elliptically as ‘for every x’
with the range indicator ‘individual’ to be ‘understood’. Moreover, the entire quantifier
phrase ‘for every individual x as Tarski [66, p. 7] says, ‘[. . . ] is often omitted and has to be
inserted mentally’: ‘if x is a person, then x is mortal’ is used to express the proposition ‘for
every individual x, if x is a person, then x is mortal’.7 On this view, Aristotle’s universal
affirmative propositions cannot be expressed in a logically perfect language without using
the truth-functional connective ‘if-then’ and using a variable that ranges over the entire
universe of individuals.
Boole espoused the Principle of Wholistic Reference (PWR): every proposition refers
to its entire universe of discourse, regardless of how limited the number of objects
explicitly referred to by its non-logical concepts Corcoran [20]). Although within a given
discourse, it is impossible to refer to objects outside of the universe of that discourse,
as said, there are formalizations of sciences in which the individual variables range over
a universe of discourse that is wider than the domain of investigation of the science.
However, frequently but—as already said—not always, the universe of discourse of a
proposition is not wider; it coincides with the domain of investigation of the science
in which it occurs (Sagüillo [51]; Corcoran [15, 16]). In such cases, there need not be
two different non-logical constants for the two different but coextensive concepts. But
when there are two terms, the formalization often contains a ‘theorem’ to the effect
6 According to Leonard Jacuzzo [36], whose 2005 dissertation research involved comparisons of dozens
of introductory logic texts, most books he studied teach the affirmative answer to this question (personal
communication). Our own less extensive experience confirms his sad finding.
7 For more on these points see Gupta [31]; Tarski [66, Section 3]; Corcoran [17].
that every object in the universe of discourse is in the domain of investigation. Tarski
[61, pp. 310–311] mentions a geometry containing a sentence ‘which asserts that every
individual [sc. in the universe of discourse] is a point [sc. member of the domain of
investigation]’—emphasis added. Thus, the universe of discourse coincides with the
domain of investigation. He also mentions an ‘axiom system of arithmetic’ that ‘contains
a sentence to the effect that every individual [sc. in the universe of discourse] is a number
[sc. member of the domain of investigation]’—emphasis added.
Frege and Russell carried Boole’s PWR one step further: by putting all of the domains
into one all-encompassing cosmic universe of individuals over which every individual
variable ranged. They and many of their followers replaced PWR by the Principle of
Cosmic Reference (PCR)—although they did not, indeed could not, put it this way. The
PCR is that every proposition refers to the entire, cosmic universe of all individuals. The
hermeneutic question mentioned above is whether Tarski [63] subscribed to the Principle
of Cosmic Reference.8 Our opinion is that he did: in some passages he seems to think
that the individual variables range over ‘all possible objects’ [63, p. 416]. He never made
use of the De Morgan-Boole concept of universe of discourse, i.e., discourse universes,
in the 400-odd pages of Logic, Semantics, Metamathematics. The expression ‘universe of
discourse’ occurs there about a dozen times: never in the consequence-definition paper,
never in the plural, and never in the sense of the range of the individual variables of the
object language. Sometimes, it is used for the domain of investigation of a science [72,
p. 28, pp. 135–140] and sometimes it is used for the carrier9 of a structure such as a
Boolean algebra [62, pp. 320, 322, 335, 347, 350, and 373]. Tarski’s use of ‘universe of
discourse’ is discussed further below.
8 Space limitations preclude discussion of how Frege arrived at PCR (or a similar monistic universal
variable-range view), why it was so widely adopted, how its conflict with pluralistic views escaped notice
for so long, and when its soundness came under scrutiny.
9 Today the set of elements of a Boolean algebra, or of any other algebraic structure such as a group or
a ring, is called its carrier. The maximal element of the Boolean algebra is usually denoted by the digit
‘1’ and is called its unity. If a given Boolean algebra is ‘formalized’ using a first-order language whose
individual variables range over the carrier, then the carrier is the universe of discourse of the ‘theory of
the algebra’. But this can distort Boole’s [3] viewpoint. Take Boole’s ‘universe of men (sc. humans)’.
Boole used ‘1’ to denote this class and ‘0’ to denote the null class: two elements of the carrier of the
corresponding Boolean algebra. In such cases, where a Boolean algebra of classes is under discussion,
there are two things competing for the names ‘universe’, ‘universe of discourse’, ‘universal class’ and
the like: the carrier and the carrier’s unity. The carrier is often the powerset of the carrier’s unity. In such
cases, the carrier’s unity is the union of the carrier. The Boolean tradition would incline towards using
such terms for the unity. The modern abstract-algebra viewpoint that abstracts from the nature of the
elements of an abstract algebra would incline toward using such terms for the carrier. Tarski called the
carrier or the set of elements of a given Boolean algebra its universe of discourse on the first page of the
article beginning on p. 347 of [72].
Even in the late 1970s and early 1980s, the issue of single versus multiple universes
in Tarski [63] was still completely (and wrongly) settled in favor of the pluralistic
interpretation in the minds of all logicians Corcoran was in contact with and who cared
about it, except for the few he succeeded in convincing otherwise.10 In fact, as far as we
know, there was little or no discussion of this point until Corcoran brought it up in the
winter of 1964 or the spring of 1965.
2 Corcoran’s Awareness of the Issue in the 1960s and 1970s
People who comb or concoct obscure sources in order to support usually lame or self-serving
priority claims concerning well-known discoveries, theorems, ‘theses’, or conjectures have their
priorities reversed. It is not the first person who discovers something that counts; it is the last
person—the person who discovers it so that it never needs rediscovery. —Peter Freyd, 1968
Philadelphia Logic Colloquium.
For years, without having studied Tarski [63] very carefully, Corcoran was under the
impression that it contained the pluralistic, multiple-universe conception of consequence,
probably to some extent on the strength of Church’s assertion [6, p. 325, fn 533]. In the
fall of 1964, Corcoran reread the paper carefully and was surprised that his expectations
were not fulfilled: he found that Tarski had a monistic fixed-universe framework. Still,
Corcoran was not very interested in the issue; he thought that Church’s mistake must have
been caught by many others. He let the issue drop.
Corcoran traces the awakening of his interest in the issue to a conversation that
semester with William Craig11 in Craig’s Dwinelle Hall office on the Berkeley campus
of the University of California late one afternoon. He recalled casually mentioning to
Craig that Church [6, p. 325, fn 533] was not entirely accurate in crediting Tarski [63]
with explication of the concept of logical consequence Church used and that had been
used by the American Postulate Theorists12 including Church’s teacher Oswald Veblen.
10 To the best of Corcoran’s knowledge, there is no chance that any unpublished writing by Tarski, either
a passage in the Tarski-Corcoran correspondence (preserved at the Bancroft Library at U.C. Berkeley),
or anywhere else, would suggest that Tarski [63] conceives of various models having various universes
of discourse. In fact, according to Paolo Mancosu, evidence to the contrary is to be found not only in
the Tarski-Corcoran correspondence (Mancosu [26 and 39, p. 451]) but also in Tarski’s unpublished
1940 lecture ‘On the completeness and categoricity of deductive systems’ also in the Bancroft Library
(Mancosu [41, pp. 754–756]).
11 Even at that time, William Craig was a distinguished mathematical logician, former doctoral student of
Quine at Harvard, Full Professor of Philosophy at Berkeley. He was a member of the UC Berkeley Group
for Logic and Methodology, in which Tarski was still active. He was internationally known for what was
then called the Craig Interpolation Lemma. The Craig Interpolation Theorem, as it is sometimes known
today, is ‘one of the basic results of the theory of models [2], almost on a par with, say, the compactness
theorem’ (Boolos et al. [4, p. 260]). An entire chapter of [4] is devoted to this theorem, which is to be
distinguished from a less deep but equally famous result then called Craig’s theorem, now sometimes the
Craig Axiomatizability Lemma or the Craig Reaxiomatization Lemma (Boolos et al. [4, p. 198]).
12 For more on the American Postulate Theorists see Scanlan’s article ‘Who were the American Postulate
Theorists?’ [53] and for the influence they had on Tarski see the same author’s ‘American Postulate
Theorists and Alfred Tarski’ [54].
The discrepancy was that the Church concept, contrary to Tarski’s, involved changing
universes of discourse. Corcoran was under the impression that it was common knowledge
that the Tarski [63] definition did not Vary universes’. Corcoran was surprised that Craig
had not heard of this rare inaccuracy in Church [6] and even more surprised to learn that
Craig thought that Corcoran must be wrong.
Corcoran remembers excusing himself for a moment, going next-door to his office, and
returning with a copy of Logic, Semantics, Metamathematics, which he handed to Craig
asking him to show where Tarski says that different interpretations can have different
universes of discourse, i.e., that the individual variables can have different ranges relative
to different interpretations. As befits his dignity, Craig asked permission to read at his
leisure and to continue the discussion the next day. The next day he came to Corcoran’s
office reporting he had reread the entire article and that, to his surprise, he found no
mention of alternative universes of discourse in Tarski [63].
What Tarski [63] presents is a kind of updated and transformed version of the
Russellian formal implication. In Russell’s sense, one proposition formally implies a
second if and only if the ‘generalized conditional’ of the former with the latter is true.
By the generalized conditional of one proposition with another is meant the universal
closure of the result of replacing all of the non-logical constants in the conditional by
suitable variables (Russell [49, pp. 5, 11, 14, and 36–41]).13 To use an adaptation of one
of Russell’s examples, the single proposition ‘Socrates is human and everything which is
human is mortal’ formally implies the single proposition ‘Socrates is mortal’ if and only
if the following single proposition is true: ‘Given any individual x, any property P , and
any property Q, if x is P and everything which is P is Q, then x is Q’. There is no place
here for changing universes.14
As already noted, one of Tarski’s innovations was his broadening the situation by
taking the ‘implicant’ to be a class of propositions (interpreted sentences), possibly
infinite, and thereby avoiding the conditional altogether in favor of a metalinguistic
condition.
A person scrutinizing [63] for signs of interpretations having different variable ranges
might first look for the words ‘domain’ and ‘universe’; neither of them occur there.
Moreover, the first full sentence on p. 416 criticizes the no-expressible-countermodels
definition15 on the ground that it implies ‘the designations of all possible objects occurred
13 We never use the expression ‘formally implies’ in Russell’s sense without explicitly adding ‘in Russell’s
sense’ or an equivalent. In fact, as Nabrasa pointed out, few if any logicians do either. He reminded us
that the fact that a person defines an expression in a certain sense, in and of itself, is no evidence that the
person uses the expression in that sense (Frango Nabrasa, personal communication).
14 This is yet another example of Russell’s habit of using previously established terminology in a sense
never before employed and without explaining or even alluding to the previous senses. His friend and
colleague G. E. Moore criticized him for this in connection with using ‘implies’ in the sense of the truth-
functional conditional. Previously no traditional logician would have said that ‘Some animal is not a dog’
logically implies ‘Every oak is a tree’.
15 See p. xxii of Corcoran’s Editor’s Introduction to Tarski [72]. A no-expressible-countermodels
definition of consequence defines a premise-conclusion argument expressed in a given language to be

valid if no argument in the same form expressible in the same language has all true premises and false
conclusion. Such a definition is featured in Quine’s Philosophy of Logic [47], where great care is taken to
ensure that the language has the required ‘richness’, to use Tarski’s expression [64, p. 416]. Even today,
some authors claim, of course without giving any justification, that the no-expressible-countermodels
in the language in question’, suggesting that he is thinking of languages whose individual

variables are unrestricted in range, i.e., which have the cosmic universe of individuals as
their range.16 The fact that Corcoran’s observation about the Church inaccuracy was a
surprise to Craig highlighted the issue in Corcoran’s mind. He recalls resolving to say
something about it in print when appropriate opportunities arose.
Corcoran recalls discussing this with some of his friends, students, and teaching
assistants then and over the next couple of years. He remembers Robert Barnes, Herbert
Bohnert, Oswaldo Chateaubriand, William Frank, Edward Keenan, Ray Lucas, John
Pollock, David Sherr, George Weaver, but there were probably some others. At the time
he was in regular correspondence with Lucas (then at the University of Wisconsin) and
with Bohnert (then at IBM Research). Chateaubriand, Pollock, and Barnes were in his
1964 ‘Philosophy of Mathematics’ seminar. Frank, Keenan, Sherr, and Weaver attended
Corcoran’s 1966 graduate course ‘Mathematical Logic’, which had Church [6] as its only
required text. In the late 1960s Corcoran was in constant contact with his friend and
colleague Henry (or Henryk) Hiż, whose ‘Reexamination of Tarski’s semantics’ [34] did
not even touch this issue.17
Corcoran also mentioned Tarski’s fixed-universe viewpoint, as it has come to be
known,18 in lectures at Georgia, Michigan, Pennsylvania, New York, Ontario, Quebec
and Mexico over the next few years in connection with his work on argumentation in
philosophy of logic and with his work on Aristotle. From 1971 to 1973, he circulated a
series of typescripts that appeared as Corcoran [11] containing the following passage on
pp. 126 and 127, bracketed matter added.
Church [6, p. 325] attributes this [concept] [. . . ] of logical consequence to Tarski [72, pp. 409–
420], but Tarski’s notion of true interpretation (model) seems too narrow [. . . ] in that no mention
of alternative universes of discourse is made or implied. In fact, the limited Tarskian notion seems
to have been already known by Lewis and Langford [37, p. 342], to whom [. . . ] I am indebted
for the terms ‘interpretation’ and ‘true interpretation’, which seem heuristically superior to the
Tarskian terms ‘sequence’ and ‘model’, the latter of which has engendered category mistakes—a
‘model of a set of sentences’ in the Tarskian sense is by no means a model, in any ordinary sense,
of a set of sentences.
conception is the classical definition of consequence. E.g. in [4, p. 101] where ‘argument’ is elliptical
for ‘expressed argument’, we find: ‘Logic teaches that the premisses [. . . ] (logically) imply or have as
a (logical) consequence the conclusion [. . . ], because [sic] in any argument of the same form, if the
premisses are true, then the conclusion is true’.
16 Tarski should not be construed as referring to possible objects as opposed to actual objects. By ‘possible
objects’ Tarski means ‘objects’; the use of the modal adjective ‘possible’ is entirely empty—it is what is
sometimes called redundant rhetoric, filler, or expletive. Tarski’s usage of modal words is almost always,
if not absolutely always, expletive, like putting ‘absolutely’ before ‘always’, ‘entirely’ before ‘empty’,
‘no matter how small’ after ‘every real number’ or ‘if any’ after ‘all odd perfect numbers’. See Corcoran
[18] and [19, p. 266]. Incidentally, Tarski may not be speaking very strictly in saying that designations of
all objects are needed.
17 Henry Hiż had been a fellow Quine PhD student at Harvard with William Craig and Robert
McNaughton, supervisor of Corcoran’s 1963 dissertation. McNaughton introduced Corcoran to Hiż

in 1961 or 1962 and to Craig in 1963. Hiż helped with the second edition of Logic, Semantics,
Metamathematics and with the Editor’s Introduction [72, pp. viii, xxvii].
18 See Sagüillo [52], but compare with Sagüillo [50, 51].
These points had been made in print earlier in ‘Conceptual Structure of Classical
Logic’, (Corcoran [8, p. 43] quoted in Sagüillo [50, p. 238]), a paper written in 1969,
and in ‘A Mathematical Model of Aristotle’s Syllogistic’ (Corcoran [9]). Both papers
were published in main-stream international journals with large circulations. To the best
of our knowledge these points were never disputed by anyone at the time, or even in the
1980s. Scanlan and Shapiro [55, pp. 149–150] mention related events.
3 The Place of the Monistic-Pluralistic Distinction
Here and elsewhere we shall not obtain the best insight into things until we actually see them
growing from the beginning. —Aristotle, Politics.
It would be interesting to know when the explicit observation that Tarski [63] holds
to a monistic and not to a pluralistic multiple-universe viewpoint was first made in print.
More generally, we can wonder when, after say 1972, it was first noted that Tarski [63]
employed a fixed-universe framework. In the 1972 article Corcoran had more pressing
issues to deal with and accordingly understated his observations19 about [63]: ‘There is
room to doubt whether [63] permits changes in domain (universe of discourse)’ (Corcoran
[8, p. 43]). The question is not one of priority: the issue is so obvious that no one
would want to claim credit for it. The question is when logicians came to feel that this
obvious point needed to be made. Our evidence seems to indicate that Chapter 8 of the
1990 Etchemendy book made this explicit in reconsidering the status of pure cardinality
sentences. Likewise, Chapter 2 of Simons [58] brought the issue up in reconsidering the
criterion for logicality sketched in the still monistic setting of Tarski [73].
It would also be interesting to know when the true statement that Tarski [63] holds a
monistic viewpoint was first explicitly denied. In our opinion, Church [6] makes no such
denial. In fact, he gives no indication that he specifically considered the issue. He does not
quote one sentence from the Tarski paper. It is likely that he did not deliberate on it and it
seems certain that he would have retracted his statement given the chance. Our evidence
seems to indicate that Gómez-Torrente [29 and 30] was in fact the first to explicitly deny
that Tarski held a monistic view and to claim that Tarski held a pluralistic view in the 1936
consequence-definition paper (cf. Mancosu [40, pp. 463–468], and [41, pp. 751–752]).
As mentioned, Tarski published two versions of the consequence-definition paper in
the year 1936, one in German and one in Polish.20 The above remarks apply directly to the
19 We should also point out that Corcoran is also to blame for a related misinterpretation of Tarski [63]. He
said that Tarski substituted new non-logical constants in the manner of the American Postulate Theorists.
Of course, Tarski substituted variables like Russell [49], not constants.
20 The following background has been generously supplied by David Hitchcock (personal communi-
cation): ‘Tarski wrote the paper in 1935. He delivered the German version at a conference in Paris
in September 1935, and appears to have left a copy of the paper with the conference organizers for
publication in the conference proceedings, which came out in 1936. The Polish version appeared in the
first (January) issue of the 1936 volume of Przeglad ˛ filozoficzny (Polish for ‘Philosophical Review’, the
leading Polish philosophy journal), and so must have been submitted by the end of 1935, allowing time for
typesetting and correcting proofs. Thus, Tarski wrote the paper no later than 1935. It is likely that he did
not write it earlier than 1935, since Carnap reports in his autobiography that Tarski visited Vienna in June
1956 English translation by Woodger of the 1936 German form of the Tarski consequence
definition paper. From the 1960s though the 1990s, Corcoran had no motivation to
consider the 1936 Polish form. In the intervening period Magda Stroińska and David
Hitchcock of McMaster University have translated the Polish form into English and
they have written extensive commentary accompanying their translation (Hitchcock and
Stroińska [33]).
In comparing the 1936 Tarski conception of consequence as it appears in the Polish
paper with modern conceptions, they wrote [33, p. 167]: ‘Tarski [. . . ] worked with
formalized languages, in which [. . . ] the domain is fixed.’
It is hard to believe that anyone who had read Tarski [63] and who was versed in the
history of modern logic going back to Frege and Whitehead-Russell could find ‘variable
universes’ in it. Most authors we know of who address the issue agree that the pluralistic
multiple-universe viewpoint is nowhere to be found in Tarski [63]. For example, in his
1992 Journal of Symbolic Logic review of John Etchemendy’s The Concept of Logical
Consequence [25], Vann McGee wrote [42, p. 254]:
Etchemendy emphasizes the divergence between Tarski’s [1936] analysis of logical truth and what
we may call the modified Tarski thesis, which identifies logical truth with truth in every model.
In particular, Tarski’s original analysis makes no provision for the special role of the universe
of discourse [. . . ]. It is unfortunate that Etchemendy focuses all his fire upon the original [1936]
Tarski thesis, for it is the modified thesis which Tarski himself employed in his mature work that
has won nearly universal acceptance.
By omitting acknowledgement of the De Morgan-Boole theory of discourse universes,

Frege—and then Russell—might have virtually ruled it out for their more dedicated
disciples, as is especially clear in the writings of van Heijenoort, Chateaubriand [5,
pp. 174f], and others. In 1936, Tarski had not yet developed many of the aspects of
his post-WWII, Berkeley-period, multiple-universe view that rejected features of the
Frege-Russell philosophy, e.g., higher-order logic. As indicated in Sagüillo [50, p. 233],
our present evidence suggests that Tarski’s earliest explicit written endorsement of the
pluralistic viewpoint is in his 1953 undecidability paper [68], published in [76]. How-
ever, according to Tarski’s student Roger Maddux (personal communication), the locus
classicus most often referred to in this connection is the 1957 Tarski-Vaught paper [77].
It is not the case, however, that Tarski [63] explicitly rejected the multiple-universe
viewpoint. In fact, it shows no awareness of the issue. As said, the expression ‘universe of
discourse’ does occur several times in Tarski [69]: but never in the consequence-definition
paper, never in the plural, and never in a relevant sense. After discussing formalization of
‘the arithmetic of real numbers’ wherein the first-order variables range over Rl the set of
real numbers, he ‘generalized’ his remarks by taking an arbitrary set instead. He wrote
1935 and that Carnap persuaded Tarski at that time to present his ideas on semantics at the September
1935 conference. Tarski had just finished translating his truth monograph into German (at the end of the
historical note which Tarski added to his German translation of his truth monograph, there appears in
Latin in italics, centered two lines below the end of the text the sentence
“Nachwort” allatum est die 13. Aprilis 1935.
—i.e. The ‘afterword’ was produced on the 13 April 1935.). The other paper that Tarski presented at
the Paris conference is a kind of summary of the ideas in semantics in that work. Both papers in fact
presuppose the concepts of the truth monograph.’ For further information see Hitchcock-Stroińska [33,
especially, pp. 155–158].
[69, p. 135]: ‘The set Rl is now replaced by an arbitrary set V (the so-called universe of
discourse or universal set)’. There is another passage in the 400-odd page Tarski [69]—
footnote 2 of p. 310 of Article X (Tarski [61])—where Tarski implicitly recognizes the
possibility of a formalized science whose domain is a proper subclass of what he calls
‘the class of all individuals’. But, the expression ‘universe of discourse’ does not occur
in this article and there is no special symbol for arbitrary universes of discourse. In fact,
the symbol V is used for ‘the class of all individuals’. Moreover, the first sentence of
the footnote implies that Tarski thought it was ‘customary’ to discuss categoricity only in
contexts where the individual variables were considered to range over all individuals and
not merely over the genus or domain of objects relevant to the science being formalized.
We are not aware of any hint in any of Tarski’s pre-WWII papers of appreciation of
the De Morgan-Boole concept of the universe of a discourse. He never mentioned the fact
that in a discourse the participants agree tacitly or explicitly to limit the subject-matter
of that discourse, i.e., in modern terms, to restrict the range of the individual variables
(as opposed to affixing restrictive relative clauses to the universal sentences leaving the
variables to have unrestricted range).21 It is true that Tarski uses the expression ‘universe
of discourse’, e.g., in the 1935 Boolean algebras paper [62] written about the same time.
However, here the ‘universes of discourse’, indicated by a non-logical constant B used in
restrictive relative clauses, are the carriers of the algebras, not the range of the individual
variables. On p. 199 of the 1933 truth-definition paper, Tarski shows acquaintance with a
relativized notion of truth when he said the following.
In the investigations which are in progress at the present day in the methodology of the deductive
sciences (in particular in the work of the Göttingen school grouped around Hilbert) another concept
of relative character plays a much greater part than the absolute concept of truth and includes it as
a special case. This is the concept of correct or true sentence in an individual domain a.
Tarski never makes the point (made repeatedly by De Morgan and by Boole) that
each discourse22 or discussion has its limited ‘universe’ as its ultimate subject matter,
a point that was central to Boole’s theory of propositions (Corcoran [18] and [19,
p. 275]). Nevertheless, by his repeated use of ‘domain-dedicated’ variables, Tarski’s work
exemplified Boole’s theory that different discourses can have different domains and that it
is not necessary to make statements about every thing. Among the domain-dedicated first-
order, individual variables used in Tarski’s pre-WWII papers, we find sentential variables,
21 The De Morgan and Boole work discovered the concept of the universe of a discourse in the conceptual
framework of the mathematics and science of their time. The role of universes of discourse persisted into
the conceptual framework of Tarski’s time. Tarski’s work shows this role but it does not show awareness
of that role nor does it show any appreciation of the De Morgan-Boole achievement. The fact that blood
circulated in Plato’s veins is no reason to credit him with discovering or knowing of blood circulation.
22 The wisdom of Boole’s choice of the word ‘discourse’ for a certain sort of extended exposition or
discussion may be questioned and it is open to doubt whether the word had ever been used in his precise
sense before. Others may have observed how much Boole enriched the English language by coining the
phrase ‘universe of discourse’, but no one seems to have suggested that his use of ‘discourse’ may have
been equally creative and meritorious. By the way, the word ‘discourse’ is more often used for a stretch of
speech involving typically more than one sentence-like expression, e.g., a paragraph or an argumentation.
It was used in this sense in Zellig Harris’s Discourse Analysis Project, which is discussed in Corcoran’s
paper ‘Discourse grammars and the structure of mathematical reasoning, Part I: Mathematical reasoning
and the stratification of language’ [7]. See also Corcoran [12].
string variables, and number variables not to mention the first-order object languages
whose variables range over a universe of ‘classes’. The most prominent of the latter, of
course, is the object language whose truths are defined in the 1933 truth-definition paper
[72, pp. 168ff.].
A no-countermodel view of logical consequence is implicit in the classic Padoa paper
[43] on definitional and implicational independence. Some readers might be inclined
to say that Padoa actually subscribed to the no-countermodels view of consequence.
It is tempting to speculate that he would have stated it clearly except that he had not
distinguished—on one hand—the epistemic relation of ‘is-a-deduction-from’ (intrinsi-
cally connected with deduction, the human capacity to reason logically) and—on the
other—the ontic relation ‘is-a-consequence-of’ (extrinsically serving as the objective
ground or standard of deduction). This distinction is not made despite the twin facts
that Padoa refers to deduction as a human activity and that he takes the existence
of a countermodel to be entirely objective, not dependent on human capabilities. The
distinction is overlooked because a kind of absolute soundness and a kind of absolute
completeness are presupposed [43, pp. 122–123]. Padoa’s explicit statements imply,
using the vocabulary of the English translation by van Heijenoort, that in order for
one given proposition to be a logical consequence of a given set of propositions it is
necessary and sufficient for no interpretation satisfying the set to satisfy the negation
of the given proposition. Tarski was well acquainted with Padoa’s thought. Moreover,
although Padoa mentions individuals and variables, he never mentions change in the range
of the individual variables. To all appearances, Padoa holds to a monistic fixed-universe
viewpoint.
The topic of precursors of Tarski’s 1936 consequence definition points to a remarkable
difference between his attitude toward precursors taken in the 1936 consequence-
definition paper and that taken by him in the earlier and more well-known 1933 truth-
definition paper. In the earlier paper, Tarski emphasized an ‘essential’ similarity—perhaps
identity—between the conception of truth Tarski characterized and that characterized in
Aristotle’s truth-definition. He quoted Aristotle: To say of what is that it is not, or what is
not that it is, is false, while to say of what is that it is, or of what is not that it is that it is not,
is true’ [69, p. 155]. In the later paper, there is no reference to Aristotle; no comparison
is made between Tarski’s relation of consequence and anything Aristotle says about
‘syllogistic consequence’: the relation of the conclusion of a syllogism and its premises.
There are two interesting points. First, in a course Tarski probably attended, Tarski’s
teacher Jan Łukasiewicz interpreted the syllogistic consequence to be something very
close to a special case of Russellian formal implication (Łukasiewicz [38, pp. 103–112,
especially p. 112]). Second, in Prior Analytics there is no succinct passage characterizing
syllogistic consequence; there is nothing analogous to the pithy passage in Metaphysics
characterizing truth. In fact, Aristotle seems to have deliberately avoided any verbal
characterization of syllogistic consequence (Corcoran [13 and 21, pp. 151–153]).
4 The Origin of the Modern form of the Monistic Framework
Every science which is ratiocinative or at all involves reasoning deals with causes and principles,
exact or indeterminate; but all these sciences mark off some particular being—some genus, and
inquire into this, but not into being simply or qua being. —Aristotle, Metaphysics, 1025b5–10.
Frege, in apparent willful ignorance of facts made clear by De Morgan and dissem-
inated by Boole (Corcoran [16, 18, 19]), created the fictional framework of monistic
fixed-universe logic. Cf. Peirce 1880 [45, p. 206f.]. Goldfarb [28, p. 351] says that for
Frege as well as for Russell the ranges of the variables ‘are fixed in advance once and for
all’ and the ‘universe of discourse is always the universe, appropriately striated’. Peano
(e.g., [44]) and then Whitehead-Russell (1910) follow Frege in this regard. They also
follow Frege in showing no awareness of an alternative—even though all four of them
repeatedly refer to [3], where we find the shift from—at first—using ‘ 1’ for the universe
to—later—using it as an indexical23 that denotes in a given discourse the universe of
that discourse (Boole [3, p. 54], Corcoran [18, p. 254]). To the best of our knowledge,
the monistic fixed-universe viewpoint has never been accepted by any logicians who
knew of the alternatives, except followers of Frege such as Jean van Heijenoort, Willard
Quine [47], and those who find Quine’s brand of wholistic naturalism attractive. Even
the post-WWII, Berkeley-period, Tarski, to his credit, adopted the multiple-universe
viewpoint—although the issue was never clear to him.
In pre-WWII logic and foundations, there were two philosophies living in tense but
peaceful co-existence. On one hand, there were the monists who recognized one single
fixed universe, who worked in a framework in which the universe of discourse, or range
of individual variables, remained fixed as the class of individuals, i.e., who subscribed
to the fixed-universe viewpoint. Many of the monists pursued what van Heijenoort
has somewhat mysteriously called ‘logic as language’. A few of them, also somewhat
mysteriously, regarded metalanguage statements as a kind of inspired and revealing
incoherence, as literally meaningless, or studiously avoided saying anything about their
own metalanguage, which belonged to a different discourse having a different range for
its individual variables. They included Boole (at least as late as 1848), the early Frege,
Peano, Padoa, Russell (in many of his writings), the early Wittgenstein, Lewis, Carnap,
pre-WWII Tarski, and others. As is to be expected, the monists misunderstood, ignored,
or even denigrated the De Morgan-Boole doctrine of universe of discourse. For example,
Lewis-Langford [37] uses the expression ‘universe of discourse’ twice in its 500-odd
pages. Once it is used on p. 28 in connection with the unity in the Boole-Schroeder algebra
(not the carrier) and once on p. 353, in quotes preceded by ‘as it were’, in connection with
interpreting a monadic letter used in restrictive relative clauses—not for the range of the
individual variables (which is never changed). Quine’1940 [46] never uses ‘universe of
discourse’ although ‘universe’ and ‘discourse’ occur sparingly.
In addition, monists tended to be reductionists who were not comfortable with the idea
of a plurality of separate ontologically independent mathematical sciences. Many of them
23 Byan indexical we mean a word such as ‘I’, ‘you’, ‘here’, ‘this’, and ‘now’ whose denotation changes
according to the context of the speech act it is used in. See Corcoran [20, pp. 159–160] for a discussion
of Boole’s innovations involving the symbol ‘1’.
stretched reductionism to its logicist extreme of ‘reducing’ all mathematical sciences to

logic. In some cases this meant exchanging an ontology of numbers, points, lines, solids,
classes, vectors, etc. for an ontology that restricted literal existence to the individual
entities over which the individual variables ranged, entities of the most unquestionably
substantial—yet abstract—and thus the most mysterious status, knowledge of which was
an article of faith. In this framework, it was natural to accept what has unfortunately
come to be known as ‘axioms of infinity’, albeit with bad conscience in some cases.
On the other hand, there were the pluralists who recognized with Boole a plurality of
universes of discourse, who tended to construct formalizations of sciences having in each
case its universe of discourse identified with the range of its individual variables, and
who tended to avoid having a variable range over all individuals without exception [71,
reprinted in 35]. These thinkers tended to accept with Aristotle a plurality of ontologically
independent mathematical sciences [10]. Among those showing pluralistic tendencies,
we find Dedekind, Poincaré, Hilbert, Veblen, Huntington, Gödel, and even Church—
somewhat surprisingly given his deep appreciation of many aspects of monistic thinking
in the work of Frege and Russell.24 Hilbert’s famous Foundations of Geometry [32],
which was probably read by every logician that came later, was a stunning endorsement
of pluralistic thinking: it took geometry to be an autonomous science and it had three
sorts of domain-dedicated individual variables—one sort ranging over points, one ranging
over lines, and one ranging over planes. Moreover, from Hilbert’s consistency and
independence results it is clear that his work presupposes a no-countermodels conception
of consequence based on a pluralistic multiple-universe framework.
Within a monistic framework one natural way of construing logical consequence is a
no-countermodels fixed-universe concept, one explication of which is given in the 1936
Tarski consequence-definition paper. Within a pluralistic framework one natural way of
construing logical consequence25 is a no-countermodels multiple-universe concept, one
explication of which is given in the Church Introduction to Mathematical Logic [6]. In the
pre-WWII period, the expression ‘logical consequence’ was ambiguous. Tarski [63] was
mistaken when he repeatedly writes ‘the concept of logical consequence’, ‘the concept
of consequence’, ‘the common concept of consequence’, ‘the consequence relation’, and
so on, suggesting that, aside from a little vagueness, there is essentially only one concept
expressed in the then-extant literature by the noun phrase ‘logical consequence’. It is
probably significant that Tarski does not mention Hilbert’s Foundations of Geometry
[32] in the 1936 consequence-definition paper, nor, for that matter, in the entire Logic,
Semantics, Metamathematics: Papers from 1923 to 1938, nor even in the Introduction
to Logic and to the Methodology of Deductive Sciences [66].26 Nevertheless, as noted
above, there is ample evidence that Tarski was aware of the pluralistic framework and
24 Of course, Church’s pluralistic tendencies might have been formed when he was studying with his
teacher and dissertation supervisor Oswald Veblen, the American Postulate Theorist [53, 54], before
studying Frege.
25 The idea that logicians down through the history of logic were expressing a Tarskian no-countermodels
concept by the phrase ‘is a consequence of’ is implausible to say the least. Tarski’s stated goal was not to
characterize the traditional concept but merely to define ‘a new concept which coincided in extent with
the common one’ [63, p. 409].
26 However, Hilbert’s Foundations of Geometry [32] is mentioned in the second edition of Logic,
Semantics, Metamathematics [72], but not by Tarski—it is on p. xvii of the Editor’s Introduction.
its conception of multiple ranges for individual variables ([59], [72, pp. 199–209] to
cite one prominent passage). Perhaps surprisingly, he nowhere shows awareness that
in presenting the monistic fixed-universe conception he might be taken as rejecting the
pluralistic multiple-universe conception. But, as far as we know there in not a scintilla of
direct evidence relevant to determining his process of deliberation.
What difference does it make? If we define a tautology as a proposition that is a logical
consequence of any and every set of propositions, then a tautology is a consequence both
of a given proposition and of its negation, and it is a consequence of the null set. If we de-
fine a contradiction as a proposition that has as a consequence any and every proposition,
then a contradiction has as a consequence the negation of each of its consequences, indeed
its own negation. Under these two definitions, a large chasm between the two definitions
of ‘logical consequence’ opens. In the case of the multiple-universe view the only pure
cardinality propositions that are contradictory are those to the effect that the universe is
empty and the only ones that are tautological are those to the effect that the universe
is non-empty. By a pure cardinality proposition is meant one to the effect that, for a
certain cardinal number, the universe does (or does not) have at least, exactly, or at most
that number of members. In the case of the fixed-universe view, every pure cardinality
proposition is contradictory or tautological, a result that is hard to swallow.
5 Concluding Remarks
If you by your rules would measure what doth not with your rules agree, forgetting all your learning
seek ye first what its rules may be. —Wagner, Die Meistersinger.
Why would Tarski base his consequence definition on the monistic fixed-universe
framework? Before attempting to answer this question it is important to note that we
are not asking why Tarski would make the monistic fixed-universe choice. We have
no evidence that any such choice was ever made. The question we ask is how the
monistic view came to play a foundation role in Tarski’s consequence-definition paper
[63]. Probably, the most important consideration is that in his earliest logic training and
in his dissertation Principia Mathematica was authoritative if not scriptural.
Once the monistic view was in place what obstacles may have kept Tarski from
questioning it? For one thing, Tarski was never bothered by the limiting cases of
tautologies and contradictions, or by the question of what human faculty is needed
to determine their truth or falsity. Moreover, Tarski never seemed to have appreciated
the De Morgan-Boole discovery of independent discourses each with its own limited
universe of discourse—a discovery that would come to demote what others had seen
as ‘the universe of individuals’—from an exalted and unique place, a sui generis, to
being just one of many universes of discourse (and a most questionable one at that).
Some scholars have argued that Tarski had more deliberate reasons for leaning toward
a monistic view (Rodriguez-Consuegra [48], Corcoran [22], and Tarski [60]). Although
we and many others believe Tarski did not question the monistic, fixed individual-variable
range view, for completeness it should be said that some well-informed scholars think that
Tarski had not only questioned it but actually rejected it. They think that its presence in
the consequence-definition paper is to be attributed to space limitation, style, rhetoric,
pedagogy, and other contingent considerations. Thus, they allow such contingencies
to outweigh truth in Tarski’s deliberations. For example, David Hitchcock wrote the
following (personal communication).
It seems quite puzzling that Tarski did not allow variation of the range of variables in his 1936
logical consequence paper, given (1) the extensive treatment in his truth monograph ([59], [72, pp.
199–209]) of the concept of a truth in a domain, (2) the fact that the logical consequence paper
is an application of ideas from the truth monograph, and (3) the fact that the fixity of the domain
overgenerates consequences by (for example) making the existence of at least two individuals
a consequence of the existence of at least one individual. The following possible explanation
occurs to us. The paragraph in brackets on p. 415 of Tarski [72] suggests that Tarski simplified
the exposition of his ideas, as would be appropriate for an audience of philosophers who were
not necessarily mathematically sophisticated. He may have consciously avoided exploration of the
variable-domain alternative in his 1936 logical consequence paper [65], in order to keep things
simple for his audience. As far as I can see, there is no mention of a domain-relative conception of
truth in the other paper that Tarski presented at the September 1935 conference, The establishment
of scientific semantics’, evidently for the same reason of keeping things simple; see the remark at
the bottom of page 406 of Tarski [72].
This is in keeping with what Hitchcock and Stroińska published [67 and 33, p. 170].
It is hard to imagine a motivation for promulgating revisionist history that would
make Tarski’s pre-WWII, Warsaw-period thinking agree with his (and the dominant)
post-WWII thinking about logical consequence, unless it is simply an inability to accept
the fact that even Tarski can change his mind. One dramatic example of the evolution
of his thought is his change from—at first—working in higher-order frameworks before
WWII to—then later—regarding them as wrong-headed: in the late 1970s he bluntly told
Corcoran that the definitional equivalence results in Corcoran, Frank, and Maloney [23]
were ‘meaningless’ because they concerned theories with second-order underlying logics
(personal communication).27 By 1936 Tarski had not yet fully shifted out of the monistic,
Frege-Russell, fixed-universe paradigm that had been presupposed in his Warsaw training.
Between his Warsaw period and his Berkeley period, Tarski’s philosophy underwent many
other radical changes [2]. Not only was higher-order logic banned and replaced by first-
order logic, but also type theory gave way to set theory, categoricity was de-emphasized
in favor of decidability and deductive completeness, and the prominence of propositional
logic was weakened while that of equational logic strengthened (e.g., [56], [70] and
Tarski-Givant [75]).
In one of his last sessions with Tarski, Corcoran asked him whether his research had
caused him to change his mind over the years on fundamental issues. He said in immediate
response that it would be sad if there were no cases of this—but then, after a long silence,
he asked whether Corcoran had had anything in particular in mind. When Corcoran said
no, Tarski said he would give it some thought. The topic never came up again (personal
communication).
27 Tarski’s oral evaluation contrasts with the evaluation published by Haskell Curry in Mathematical
Reviews [24].
Acknowledgements
We call in others to aid us in deliberating on important questions—distrusting ourselves as not
being equal to deciding. —Aristotle.
We started planning this paper in 2003 on a trip the two of us made to Lisbon from Santiago de
Compostela for Ricardo Santos’s Tarski Symposium. Earlier versions have been circulated (Mancosu
[40, p. 576]). For bringing errors and omissions to our attention, for useful suggestions, and for other
help, it is a pleasure to acknowledge the following scholars: O. Chateaubriand (Brazil), D. Hitchcock
(Canada), R. Torretti (Chile), I. Grattan-Guinness (England), S. Nambiar (India), H. Masoud (Iran),
R. Santos (Portugal), S. Read (Scotland), F. Rodriguez-Consuegra (Spain), J. Gasser (Switzerland), and,
from USA, M. Brown, W. Goldfarb, R. Grandy, I. Hamid, L. Jacuzzo, C. Jongsma, E. Keenan, M. La Vine,
R. Maddux, P. Mancosu, D. Merrill, J. Miller, M. Scanlan, S. Shapiro, J. Smith, J. Tarski, and G. Weaver.
During the entire period in which this article was being written we have been in almost constant contact
with our long-time friend Frango Nabrasa, whose incisive sarcasm and disarming skepticism dampened
any excessive enthusiasm we might have had. David Hitchcock has been especially helpful: not only has
he shared insights and given valuable criticisms, his published scholarship on this topic makes crucial,
actually essential, contributions. We are also grateful to Volker Peckhaus and to two anonymous History
and Philosophy of Logic referees for their useful suggestions and perceptive corrections.
References
1. Audi, R. (ed.): The Cambridge Dictionary of Philosophy. Cambridge University Press, Cambridge
(1999)
2. Bays, T.: On Tarski on models. J. Symb. Log. 66, 1701–1726 (2001)
3. Boole, G.: Laws of Thought. Macmillan, Cambridge (1854). Reprinted with introduction by
J. Corcoran. Prometheus Books, Buffalo (2003)
4. Boolos, G., Burgess, J., Jeffrey, R.: Computability and Logic. Cambridge University Press, Cambridge
(2002)
5. Chateaubriand, O.: Boole on reference and universes of discourse: reply to John Corcoran. Manuscrito
27, 173–182 (2004)
6. Church, A.: Introduction to Mathematical Logic. Princeton University Press, Princeton (1956)
7. Corcoran, J.: Discourse grammars and the structure of mathematical reasoning, part I: mathematical
reasoning and the stratification of language. J. Struct. Learn. 3, 55–74 (1971)
8. Corcoran, J.: Conceptual structure of classical logic. Philos. Phenomenol. Res. 33, 25–47 (1972)
9. Corcoran, J.: A mathematical model of Aristotle’s Syllogistic. Archiv für Geschichte der Philosophie
55, 191–219 (1973)
10. Corcoran, J. (ed.): Ancient Logic and Its Modern Interpretations. Kluwer, Dordrecht (1974)
11. Corcoran, J.: Aristotle’s Natural Deduction System. In: Corcoran, J. (ed.) Ancient Logic and its
Modern Interpretations, pp. 85–131. Kluwer, Dordrecht (1974)
12. Corcoran, J.: Argumentations and Logic. Argumentation 3, 17–43 (1989)
13. Corcoran, J.: Review of Tarski 1986(2). Mathematical Reviews. 91h:01101-4 (1991)
14. Corcoran, J.: Logical form. In: Audi, R. (ed.) The Cambridge Dictionary of Philosophy, pp. 511–512.
Cambridge University Press, Cambridge (1999)
15. Corcoran, J.: Domain. In: Audi, R. (ed.) The Cambridge Dictionary of Philosophy, p. 242. Cambridge
University Press, Cambridge (1999)
16. Corcoran, J.: Universe of discourse. In: Audi, R. (ed.) The Cambridge Dictionary of Philosophy,
p. 941. Cambridge University Press, Cambridge (1999)
17. Corcoran, J.: The logical form of quantifier phrases: quantifier-sortal-variable. Bull. Symb. Log. 5,
418–419 (1999)
18. Corcoran, J.: Introduction. In: Boole, G. (ed.) Laws of Thought. Macmillan, Cambridge (1854).
Reprinted with introduction by J. Corcoran. Prometheus Books, Buffalo (2003)
19. Corcoran, J.: Aristotle’s prior analytics and Boole’s laws of thought. Hist. Philos. Log. 24, 261–288
(2003)
20. Corcoran, J.: The principle of wholistic reference. Manuscrito 27, 159–171 (2004)
21. Corcoran, J.: Notes on the founding of logics and metalogic: Aristotle, Boole, and Tarski. In:
Martínez, C., et al. (eds.) Current Issues in Logic/Temas Actuates de Lógica, pp. 143–176 (2007).
Imprenta Univeridade Santiago de Compostela, Galicia (University of Santiago de Compostela Press)
22. Corcoran, J.: Review of: Sinaceur, H. B.: Tarski’s practice and philosophy: between formalism and
pragmatism. In: Lindström, S., Palmgren, E., Segerberg, K., Stoltenberg-Hansen, V. (eds.) Logicism,
Intuitionism, and Formalism: What Has Become of Them? pp. 357–396. Springer, Dordrecht (2009).
Mathematical Reviews. MR2509665 (2011b:03006)
23. Corcoran, J., Frank, W., Maloney, M.: String theory. J. Symb. Log. 39, 625–637 (1974)
24. Curry, H.: Review of: Corcoran, J., Frank, W., Maloney, M.: String theory. J. Symb. Log. 39, 625–637
(1974). Mathematical Reviews. MR0398771 (53#2622) (1977)
25. Etchemendy, J.: The Concept of Logical Consequence. Harvard University Press, Cambridge (1990)
26. Ferreiros, J., Gray, J.: The Architecture of Modern Mathematics. Oxford University Press, Oxford
(2006)
27. Gödel, K.: On Formally Undecidable Propositions. In: van Heijenoort, J. (ed.) From Frege to Gödel,
pp. 596–616. Harvard University Press, Cambridge (1931/1967)
28. Goldfarb, W.: Logic in the twenties: the nature of the quantifier. J. Symb. Log. 44, 351–368 (1979)
29. Gómez-Torrente, M.: Tarski on logical consequence. Notre Dame J. Formal Log. 37, 125–151 (1996)
30. Gómez-Torrente, M.: Rereading Tarski on logical consequence. Rev. Symb. Log. 2, 249–297 (2009)
31. Gupta, A.: The Logic of Common Nouns. Yale University Press, New Haven (1980)
32. Hilbert, D.: Foundations of Geometry. Open Court, La Salle (1899/1971)
33. Hitchcock, D., Stroińska, M.: Translators’ introduction to Alfred Tarski’s on the concept of following
logically. Hist. Philo. Log. 23, 155–196 (2002). See Tarski, A.: Über den Begriff der logischen Fol-
gerung. Actes du Congrès International de Philosophie Scientifique, vol. 7 (Actualités Scientifiques
et Industrielles, vol. 394), 1–11. Paris (1936)
34. Hiż, H.: Reexamination of Tarski’s semantics. Ann. Pure Appl. Logic 126, 39–48 (2004)
35. Hughes, R.: Philosophical Companion to First-order Logic. Hackett, Indianapolis (1993)
36. Jacuzzo, L.: The Pedagogy of Logic. PhD dissertation in Philosophy. University at Buffalo, Buffalo
(2005)
37. Lewis, C. I., Langford, C. H.: Symbolic Logic. Century, New York (1932)
38. Łukasiewicz, J.: Elements of Mathematical Logic (1929). Wojtasiewicz, O. (trans.). Macmillan, New
York (1963)
39. Mancosu, P.: Tarski on models and logical consequence. In: Ferreiros, J., Gray, J. (eds.) The
Architecture of Modern Mathematics, pp. 209–237. Oxford University Press, Oxford (2006)
40. Mancosu, P.: The Adventure of Reason. Interplay between mathematical logic andphilosophy of
mathematics: 1900–1940. Oxford University Press, Oxford (2010)
41. Mancosu, P.: Fixed- vs. variable-domain interpretations of Tarski’s account of logical consequence.
Philos. Compass. 5, 745–759 (2010)
42. McGee, V.: Review of Etchemendy, J. The Concept of Logical Consequence. Harvard University
Press, Cambridge (1990). J. Symb. Log. 51, 254–255 (1992)
43. Padoa, A.: Essai d’une théorie algébrique des nombres entiers, précede d’une introduction logique
à une théorie déductive quelconque [Logical introduction to any deductive theory]. Bibliothèque du
Congrès international de philosophie, Paris. In: van Heijenoort, J. (ed.) From Frege to Gödel, pp. 118–
123. Harvard University Press, Cambridge (1900/1967)
44. Peano, G.: Principles of mathematics presented by a new method. In: van Heijenoort, J. (ed.) From
Frege to Gödel, pp. 83–97. Harvard University Press, Cambridge (1967/1889)
45. Peirce, C. S.: The Essential Peirce: Selected Philosophical Writings (1867–1893), vol. I. Houser, N.,
Kloesel, C. (eds.). Indiana University Press, Bloomington (1992)
46. Quine, W.: Mathematical Logic. Harvard University Press, Cambridge MA (1940). Reprinted (1951)
47. Quine, W.: Philosophy of Logic. Harvard University Press, Cambridge MA (1970/1986)
48. Rodriguez-Consuegra, F.: Tarski’s intuitive notion of set. In: Sica, G. (ed.) Essays on the Foundations
of Mathematics and Logic, pp. 227–266. Polimetrica, Monza (2005)
49. Russell, B.: The Principles of Mathematics. Cambridge University Press, Cambridge (1903)
50. Sagüillo, J.: Logical consequence revisited. Bull. Symb. Log. 3, 216–241 (1997)
51. Sagüillo, J.: Domains of sciences, universe of discourse, and omega arguments. Hist. Philos. Log. 20,
267–280 (1999)
52. Sagüillo, J.: Methodological practice and complementary concepts of logical consequence: Tarski’s
model-theoretic consequence and Corcoran’s information-theoretic consequence. Hist. Philos. Log.
30, 21–48 (2009)
53. Scanlan, M.: Who were the American postulate theorists? J. Symb. Log. 56, 981–1002 (1991)
54. Scanlan, M.: American postulate theorists and Alfred Tarski. Hist. Philos. Log. 24, 307–325 (2003)
55. Scanlan, M., Shapiro, S.: The work of John Corcoran. Hist. Philos. Log. 20, 149–158 (1999)
56. Schmidt, H. A., et al. (eds.): Contributions to Mathematical Logic (Collog., Hannover, 1966). North-
57. Shapiro, S. (ed.): The Limits of Logic. Dartmouth Publishing company, Aldershot (1996)
58. Simons, P.: Philosophy of Logic in Central Europe from Bolzano to Tarski. Kluwer, Dordrecht (1992)
59. Tarski, A.: The concept of truth in formalized languages. In: Tarski, A. (ed.) Logic, Semantics,
Metamathematics, 2nd edn., pp. 152–278. Introduced and edited by J. Corcoran. Hackett, Indianapolis
(1933/1983)
60. Tarski, A.: Some observations on the concepts of ω-consistency and ω-completeness. In: Tarski,
A. (ed.) Logic, Semantics, Metamathematics, 2nd edn. pp. 279–295. Introduced and edited by J.
Corcoran. Hackett, Indianapolis (1933/1983)
61. Tarski, A.: Some Methodological Investigations on the Definability of Concepts. In: Tarski, A. (ed.)
Logic, Semantics, Metamathematics, 2nd edn., pp. 296–319. Introduced and edited by J. Corcoran.
Hackett, Indianapolis (1934/1983)
62. Tarski, A.: On the Foundations of Boolean algebra. In: Tarski, A. (ed.) Logic, Semantics, Meta-
mathematics, 2nd edn., pp. 320–341. Introduced and edited by J. Corcoran. Hackett, Indianapolis
(1935/1983)
63. Tarski, A.: Über den Begriff der logischen Folgerung. Actes du Congrès International de Philosophie
Scientifique, vol. 7 (Actualités Scientifiques et Industrielles, vol. 394), 1–11. Paris (1936). Cited from
English trans. in: Tarski, A.: Logic, Semantics, Metamathematics, 2nd edn., pp. 409–420. Introduced
and edited by J. Corcoran. Hackett, Indianapolis (1983)
64. Tarski, A.: On the Concept of Logical Consequence (1936). English trans. (Woodger, J. H. (trans.))
in: Tarski, A.: Logic, Semantics, Metamathematics, 2nd edn., Introduced and edited by J. Corcoran.
Hackett, Indianapolis (1983)
65. Tarski, A.: On the Concept of Following Logically (1936). English trans. (Stroińska, M., Hitchcock,
D. (trans.)) in: History and Philosophy of Logic 23, pp. 155–196
66. Tarski, A.: Introduction to Logic and to the Methodology of Deductive Sciences (1941). Helmer, O.
(trans.). Oxford University Press, New York (1946)
67. Tarski, A.: Introduction to Logic and to the Methodology of Deductive Sciences (1941). Trans., ed.
with preface and biographical sketch of the author by J. Tarski. Oxford University Press, New York
(1994)
68. Tarski, A.: A general method in proofs of undecidability. In: Tarski, A., Mostowski, A., Robinson, R.
(eds.), Undecidable Theories, pp. 3–35. North Holland, Amsterdam (1953)
69. Tarski, A.: Logic, Semantics, Metamathematics. Woodger, J. H. (trans.). Oxford University Press,
Oxford (1956)
70. Tarski, A.: Equational Logic and Equational Theories of Algebras. In: Schmidt, H. A., et al.
(eds.): Contributions to Mathematical Logic (Collog., Hannover, 1966), pp. 275–288. North-Holland,
Amsterdam (1966/1968). Reprinted in: Tarski, A.: Collected Papers, Givant, S., McKenzie, R. (eds.),
Vol. 4, pp. 347–362. Birkhäuser, Basel (1986)
71. Tarski, A.: Truth and proof. Sci. Am. (1969). Reprinted in: Hughes, R.: Philosophical Companion to
First-order Logic. Hackett, Indianapolis (1993)
72. Tarski, A.: Logic, Semantics, Metamathematics, 2nd edn., Introduced and edited by J. Corcoran.
Hackett, Indianapolis (1983)
73. Tarski, A.: What are logical notions? Hist. Philos. Log. 7, 143–154 (1986). Reprinted in: Shapiro, S.
(ed.): The Limits of Logic. Dartmouth Publishing company, Aldershot (1996)
74. Tarski, A.: On the Concept of Logical Consequence (1936). In Givant, S., McKenzie, R. (eds.)
Collected Papers. Birkhäuser, Basel (1986)
75. Tarski, A., Givant, S.: A Formalization of Set Theory without Variables. American Mathematical
Society, Providence (1987)
76. Tarski, A., Mostowski, A., Robinson, R.: Undecidable Theories. North Holland, Amsterdam (1953)
77. Tarski, A., Vaught, R.: Arithmetical extensions of relational systems. Compos. Math. 13, 81–102
(1957)
78. van Heijenoort, J.: From Frege to Gödel. Harvard University Press, Cambridge (1967)
J. Corcoran ()
Department of Philosophy, University at Buffalo, Buffalo, NY, USA
J. M. Sagüillo
Department of Logic, University of Santiago de Compostela, Santiago, Spain
Alfred Tarski: Auxiliary Notes on His Legacy
Jan Zygmunt
Abstract The purpose of this article is to highlight a selected few of Alfred Tarski’s
career achievements. The choice of these achievements is subjective. Section 1 is a general
sketch of his life and work, emphasizing his role as researcher, teacher, organizer and
founder of a scientific school. Section 2 discusses his contributions to set theory. Section 3
discusses his contributions to the foundations of geometry and to measure theory. Section
4 looks at his metamathematical work, and especially the decision problem for formalized
theories. Section 5 is a selected bibliography to illustrate Sects. 1–4.
Keywords Alfred Tarski · Logic · Metamathematics · Set theory · Foundations of

geometry
AMS Subject Classification (2000) Primary: 01A60, Secondary: 01A70, 01A72, 51-03,
54-03, 54A05
1 Life
Alfred Tarski was a logician and a mathematician. He exerted a significant influence

on the whole of the twentieth century development of logic and the foundations of
mathematics, and through his work in formal semantics and the foundations of logic
also on epistemology, the methodology of science, and the philosophy of language. He
was a famous representative of the Lwów-Warsaw school and founder of the Berkeley
school of logic and the methodology of sciences. He left behind him a rich and wide
Translated and edited by Robert Purdy.

A major part of this chapter is an extension of the first four sections of: J. Zygmunt, Alfred
Tarski—logik i metamatematyk (Alfred Tarski—Logician and Metamathematician), in the volume: O
przyrodzie i kulturze, Studium Generale, Seminaria Interdyscyplinarne, vol. XIII (2009), pp. 305–
327, Wydawnictwo Uniwersytetu Wrocławskiego: Wrocław 2009. See also: (1) J. Zygmunt, Szkic
biograficzny (A Biographical Sketch), in [95m], pp. vii-xx; and (2) J. Zygmunt, Alfred Tarski, in Polska
filozofia powojenna, edited by Witold Mackiewicz, vol. 2, pp. 342–375, Agencja Wydawnicza Witmark:
Warszawa 2001.

426 J. Zygmunt
scientific legacy in logic, metamathematics, semantics, set theory, the foundations of

geometry, general algebra and algebraic logic (Boolean algebra, Boolean algebras with
operators, relation algebras, cylindric algebras); nearly all of Tarski’s output is influenced
by the ideas of algebra. Sensitive to the idiomaticity of language, he wrote precisely
and clearly, persuasively setting issues in historical and intuitive perspective. He was a
charismatic teacher and lecturer. He recognized the role of aesthetics in science, and he
drew attention to the “intrinsic charm and prettiness” of theoretical constructions. He was
a fierce champion of academic freedom.
Tarski was born in Warsaw on January 14th, 1901, the eldest son of Ignacy aka Izaak
Tajtelbaum (often spelled Teitelbaum) and Róża (née Prussak). He had a younger brother
Wacław (1903–1944), who was a lawyer. Alfred died on October 27th, 1983, age 82,
in Berkeley, California, U.S.A. His creative working life spanned more than 60 years,
starting in 1921 and continuing in Warsaw up to August, 1939, then in the U.S.A., first on
the East Coast, and then, from 1942 to the end of his life, at the University of California,
Berkeley.
1.1 The Warsaw Years
From 1918 to 1924 Tarski was a student in the Faculty of Philosophy at the University
of Warsaw. He studied logic, mathematics and philosophy. Despite the interruptions of
the Soviet westward offensive of 1918–19 and the Polish–Soviet War of 1919–1921,
which intermittently obliged the University to suspend its operations,1 he managed to
fit all his undergraduate and graduate course work, sitting his doctoral exams, and
researching, writing and defending his dissertation into those six tumultuous years—
while at the same time publishing four journal articles, and, from January of 1924,
taking high school teacher training. His thesis advisor was Stanisław Leśniewski, and
his Ph.D. thesis was titled O wyrazie pierwotnym logistyki <On the Primitive Term of
Logistic>.
One year later he successfully defended his “habilitation” thesis, securing a “veniam
legendi” in the philosophy of mathematics and becoming a “docent” of the University
of Warsaw—licensed to solicit, accept and fulfill commissions to lecture in its name.2
He gave “ćwiczenia” (tutorials) and “wykłady zlecone” (commissioned lectures) on set
theory, the methodology of mathematics, the foundations of “school” geometry, and the
1 Tarski’sfirst university year, 1918–19, was a write-off. Classes were cancelled. Students and faculty
signed up for military service. Stanisław Leśniewski, Stefan Mazurkiewicz and Wacław Sierpiński
worked with the military on decoding Soviet communications. Tarski performed community service in
lieu of military service.
In June, 1920, lectures were again cancelled and students and faculty again volunteered, Tadeusz
Kotarbiński and Jan Łukasiewicz among them. This time Tarski served with a military supply and medical
unit.
2 Jacek Jadacki speculates that Tarski’s habilitation thesis was the 50-page paper Sur les ensembles
finis—i.e., [24c]. See J.J. Jadacki (ed.), Alfred Tarski: dedukcja i semantyka (déduction et semantique),
Wydawnictwo Naukowe Semper: Warszawa 2003, p. 117. If true, Tarski must have researched and written
two dissertations simultaneously.
Alfred Tarski: Auxiliary Notes on His Legacy 427
arithmetic of the natural numbers and the reals. On October 1st, 1929, he joined the
University’s payroll as a “starszy asystent” (senior assistant), and on October 1st, 1934, he
was promoted to “adiunkt” (no exact English equivalent; higher than senior assistant) in
Jan Łukasiewicz’s Philosophy Seminar. He tried for the position of “katedra” (department
chair) at the University Jana Kazimierza in Lwów in 1930, and at the University of Poznań
in 1937, but in both cases without success.
From 1925 on, he also taught high school mathematics at Gimnazjum im. Stefana
Żeromskiego in Warsaw. Throughout this period he kept up a frenetic pace of investigation
into mathematical logic, semantics, set theory, measure theory, the foundations of
geometry, and the teaching of logic and geometry. Arguably, the whole of his life-long
output can be traced to these roots.
His work in sentential calculi, methodology of the deductive sciences, cardinal
arithmetic, the Axiom of Choice, the definition of truth and more generally semantics
brought him recognition and acclaim. During these years in Warsaw he published 16
abstracts and short notes, 62 longer papers, 3 reviews, 14 exercises3 and 2 problems, 11
contributions to discussions, and 3 books . . .
• Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych <The Concept of Truth in the
Languages of the Deductive Sciences> (1933).
• O logice matematycznej i metodzie dedukcyjnej <On Mathematical Logic and the
Deductive Method> (1936). This has been translated into 12 other languages. Over the
years it has had four English-language editions.
• A high school geometry textbook.4
By 1930 he had already become a leading figure of the Warsaw school of logic and
mathematics, collaborating with nearly all of its preeminent members, as well as members
of the Lwów school. He co-authored publications with . . .
• Stefan Banach (on measure theory, decompositions of point sets in metric spaces, and
the paradoxical decomposition of the solid sphere).
• Adolf Lindenbaum (on set theory, and the theory of definability).
• Jan Łukasiewicz (the landmark paper Untersuchungen über den Aussagenkalkül
<Investigations into the Sentential Calculus>, 1930).
• Kazimierz Kuratowski (on projective sets—the Tarski–Kuratowski algorithm).
• Wacław Sierpiński (on inaccessible cardinal numbers).
• Andrzej Mostowski (on Boolean rings/algebras with ordered bases).
He participated actively in academic life both at home and abroad. He was a frequent
speaker at meetings of the Warsaw Philosophic Society, the Lwów chapter of the Polish
Philosophic Society, the Warsaw Mathematics Society, and the Warsaw Scientific Society.
He took part in the Polish Philosophy Congresses of 1923, 1927 and 1936, and the Polish
3 For English translations see Chapter 12, Exercises Posed by Tarski, in A. McFarland, J. McFarland,
James T. Smith (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching. With a
Bibliographic Supplement, Birkhäuser: New York 2014, pp. 243–272.
4 Geometria dla trzeciej klasy gimnazjalnej, co-authored with Z. Chwiałkowski and W. Schayer. For an
English translation see McFarland–McFarland–Smith [2014], pp. 273–318.

428 J. Zygmunt
Mathematics Congresses of 1927, 1931 and 1937. He also took part in international
mathematics and philosophy conferences in Bologna, Italy (1928), Warsaw (1929),
Prague (1934), Paris (1935, 1937), and Amersfoort, the Netherlands (1938). He played
a principal role in establishing cooperation and working contacts between the Viennese
and Warsaw schools. Thanks to his visits to Vienna in 1930 and 1935, and his lectures
Grundlegung der wissenschaftlichen Semantik <On the Semantics of Science> and Über
den Begriff der logischen Folgerung <On the Concept of Logical Consequence> given at
the first Paris Congress on Scientific Philosophy in 1935, Polish logic exerted a formative
influence on the Vienna Circle.
1.2 America
On August 21st, 1939 Tarski traveled to the United States to address the 5th International
Congress for the Unity of Science, being held September 3–9 at Harvard University in
Cambridge, Massachusetts, and to give a lecture tour at several American universities.
Owing to the German invasion of Poland on September 1st, 1939, the massive aerial
bombardment of Warsaw that began on the same day, and the course of events that
immediately followed, leading to the land siege of Warsaw from September 8th through
the 28th, Tarski was obliged for his own good to remain in America.5
Until the summer of 1942 he stayed on the east coast, with temporary appointments
at Harvard, CUNY, and the Institute for Advanced Study at Princeton. During this period
he forged relationships with Paul Erdös (at Princeton) and J.C.C. McKinsey (in New
York City, though not at CUNY—McKinsey was at NYU.) With Erdös he co-authored a
paper on fields of sets and large cardinal numbers, published in 1943. With McKinsey he
collaborated on algebraic aspects of general topology, and applications of topological
methods in intuitionistic and modal logics; their results appeared much later in three
jointly authored papers published in 1944–1948. He also renewed contacts with Kurt
Gödel (at Princeton) and Rudolf Carnap and W.V.O. Quine (both at Harvard at that time).
In 1941 he published what later turned out to be a groundbreaking work in algebraic logic,
a paper titled On the Calculus of Relations.
In October, 1942, upon securing a lecturer position at the University of California,
Berkeley, he relocated to the west coast. In 1948 he was promoted to full professor in
the Department of Mathematics at Berkeley, a position which he held to the end of his
life.6
His first 14 years at Berkeley, up to 1956, witnessed the realization/achievement of
several major scholarly and organizational undertakings which together cemented Tarski’s
scientific reputation on a world scale. First, he completed and published the results of
some important investigations he had begun before the war. These included . . .
5 . . . leaving his wife and children in Warsaw. In fact he had little choice, as his ship’s return sailing was
cancelled.
6 Although emeritus from 1968, he continued teaching until 1973, and continued supervising Ph.D.
candidates right up until his death in 1983.

1. The paper The Semantic Conception of Truth and the Foundation of Semantics
(1944)—a relatively non-technical, or “technical-lite”, exposition of his thinking on
semantics, addressed to the philosophical community.
2. The monograph A Decision Method for Elementary Algebra and Geometry (1st
edition 1948, 2nd edition 1951), received to great acclaim by the mathematics
community.
3. The twin books Cardinal Algebras (1949) and Ordinal Algebras (1956). Both books
developed results that had been announced without proof in the 1926 omnibus paper
Communication sur les recherches de la théorie des ensembles, co-authored by Tarski
and Adolf Lindenbaum. J.C.C. McKinsey and two of Tarski’s very first doctoral
students at Berkeley, Louise H. Chin7 and Bjarni Jónsson, played major roles in
editing Cardinal Algebras. Ordinal Algebras was the collective effort of Tarski, Bjarni
Jónsson, and C.C. Chang.
4. The anthology Logic, Semantics, Metamathematics: Papers from 1928 to 1938
(1956), presenting J.H. Woodger’s English translations of 17 of Tarski’s works from
the 1920s and 1930s on logic, the theory of truth, and the methodology of the deductive
sciences. This anthology made Tarski’s pre-war output available to the wider world for
the first time.
Second, he marked out new lines of inquiry to do with algebraic logic, decidability, and
model theory. Algebraic logic quickly gave rise to Tarski and McKinsey’s The Algebra
of Topology (1944) and On Closed Elements in Closure Algebras (1946); Tarski and
Jónsson’s Boolean Algebras with Operators (Part l, 1951; Part 2, 1952); and saw work
begin on the algebraization of quantificational logic with the help of cylindric algebras.
The foundations of the theory of cylindric algebras were worked out by Tarski and his
students Louise H. Chin and Frederick B. Thompson in the years 1948–1952. Cylindric
algebras were extensively researched at Berkeley, and indeed in centers of logical and
algebraic research around the world, up to the 1990s.8
Tarski summarized his work on decidability in the short book Undecidable Theories
(1953), written in collaboration with Andrzej Mostowski and Raphael M. Robinson.
This book is widely considered a masterpiece of the literature on mathematical logic.
In the articles Some Notions and Methods on the Borderline of Algebra and Meta-
mathematics (1952) and Contributions to the Theory of Models: I, II, III (1954–1955),
Tarski introduced the basic conceptual apparatus of model theory—a specialized area
of set-theoretic semantics for formalized languages and theories—and he showed that
model-theoretic techniques are productive in mathematics. He proved, specifically, that
the class of representable relation algebras is an equational class, axiomatizable by a set
of equations alone—a finding which spurred further investigation of the logical aspects of
relation algebras. More generally, one of the tasks model theory was supposed to fulfill
was to provide mathematically tractable definitions of metamathematical and metalogical
7 The Mathematics Genealogy Project lists her as Louise Hoy Chin Lim.
8 Fora beautifully written exposition of the first stages of these investigations, and their prehistory, see
[61a]. For later developments, see [71m ] and [85m ].
430 J. Zygmunt
concepts, such as, for instance, “arithmetical class”, definability, elementary equivalence,
etc. Tarski also played an important role in the formulation of a general definition of
“reduced product”, and he demonstrated a range of interesting applications of reduced
products. In the years 1957–1962 his students and colleagues C.C. Chang, J. Howard
Keisler, Dana Stewart Scott, Thomas E. Frayne and Anne C. Morel obtained what are now
recognized as classic results in model theory, by applying the methods and techniques of
reduced products.
Tarski also invented predicate logic with infinitely long expressions (1957) and model
theory for infinitary languages. This theory and its application to the study of inaccessible
cardinals were developed and elaborated by Carol Ruth Karp, William Porter Hanf, and
Dana Stewart Scott. Of these, Hanf was a student of Tarski, Scott a frequent collaborator,
and Karp an avid disciple.
Third, he founded a strong center of logic and foundations of mathematics at Berkeley,
which by the mid-1950s already enjoyed international renown. From 1952 to 1970 Tarski
was director of a program called Basic Research in the Foundations of Mathematics. The
program covered all the main areas of the foundations of mathematics, and the number of
people taking part in it was impressive. The proceedings for the period July 1st, 1959 to
June 30th, 1961 list the following topics: model theory, proof theory, infinitary logics, set
theory and its foundations, general theory of algebraic and relational structures, algebraic
structures in logic, and the foundations of geometry. At least 20 people were listed as
participating in the program as researchers, among them the Poles Jerzy Łoś, Wanda
Szmielew, Andrzej Ehrenfeucht, and Jan Mycielski.
In 1958 Tarski and his colleagues established The Group in Logic and the Methodology
of Science, whose mission was, and remains to this day, to foster and promote interdis-
ciplinary doctoral studies leading to a Ph.D. in logic and the philosophy of science. Its
idea was (and is) above all about research in mathematical logic in the broadest sense,
and its applications to information theory, computability theory, artificial intelligence,
methodology of science, philosophy of science, and philosophy of language. Since 1989
the Group has sponsored The Annual Alfred Tarski Lectures. Eminent scholars are invited
to give lectures on their current areas of interest.
With the aim of strengthening and broadening international collaboration Tarski
organized two international symposia at Berkeley: the first, in 1957/58, devoted to the
axiomatic method, with special emphasis on its applications to geometry and physics,
and the second, in 1963, devoted to model theory. The fruits of these two conferences,
beyond of course their stated aim of fostering international cooperation, which they
roundly succeeded in achieving, were two substantial volumes of proceedings, co-edited
by Tarski: The Axiomatic Method (1959) and The Theory of Models (1965). The papers
they contain highlight multifarious, subtle and often surprising connections between the
axiomatic method and model theory.
In subsequent years Tarski continued working in almost every area that had occupied
him previously, but especially universal algebra, equational logic and cylindric algebras,
and the foundations of geometry (see §3 below). His chapter zero to Cylindric Algebras.
Part I (1971) is a beautiful monograph on general algebra. He devoted the last years
of his life, with the help of Steven Givant, to the book A Formalization of Set Theory
without Variables, a work which at last brought to fruition Tarski’s life-long program of
algebraizing logic and the foundations of set theory.
In 1956–1957 he was president of the International Union of the History and
Philosophy of Science. Within the framework of the Union he set up a Division
of Logic, Methodology, and Philosophy of Science, whose purpose was to organize
international congresses on logic, methodology and philosophy of science. In setting
it up Tarski was partly carrying out one of the pre-war aims of the Unity of Science
movement.
Tarski guest lectured at numerous universities in America and internationally. He
participated in a huge number of conferences, symposia and congresses. He loved
traveling, and was curious about the world. He was a scholar who viewed the meaning
of his work not as the mere achieving of results in the form of theorems or theories
but as part of a communal effort toward the discovery of scientific truth. It was for this
reason that he chose to announce so many of his results jointly with his colleagues and
students. The Mathematics Genealogy Project shows him as having had 26 students,9
among them five women. Almost all of them went on to occupy prominent positions
in the world of learning (C.C. Chang, Solomon Feferman, Bjarni Jónsson, H. Jerome
Keisler, Richard Montague, Andrzej Mostowski, Julia Robinson, Wanda Szmielew,
Robert Vaught).
Tarski took a keen interest in cultural and academic life in post-war Poland.
Many a Polish logician, philosopher and mathematician benefited from his hospitality
and watchful professional care at Berkeley. He visited Poland several times,
including . . .
• For the symposium Metody infinitystyczne <Infinitistic Methods>, held in Warsaw in
September, 1959, at which he gave a talk On Predicative Set Theory (which was
never published). On that occasion he also visited Wrocław and gave a lecture at
a meeting of the Wrocław chapter of the Polish Philosophic Society titled Czym sa˛
poj˛ecia logiczne? <What are Logical Notions?> which was not published until after
his death.
• For a Methodology Colloquium on the justification of assertions and decisions (in
1961).
• For a conference on general algebra (in September, 1964).
He was awarded three honorary doctorates: by Pontificia Universidad Católica de Chile
(1974), Université d’Aix-Marseille II (1978), and the University of Calgary (1982). In
1966 he was recipient of the Alfred Jurzykowski Foundation’s Millennium Award. In
1981 the University of California at Berkeley awarded him its Berkeley Citation.
He was a member of the United States National Academy of Science, the British
Academy, and the Royal Netherlands Academy of Arts and Sciences. In 1945 he was
appointed correspondent member of the Polish Academy of Learning.10
9 The Mathematics Genealogy Project defines a “student of Tarski” as someone who was awarded a Ph.D.
and whose dissertation listed Tarski as “Advisor 1” or “Advisor 2.”
10 In March, 1949, he was stripped of this distinction, as were other Polish scholars living abroad at that
time, for failing to repatriate.

432 J. Zygmunt
2 Set Theory
Set theory was one of Tarski’s main research interests almost all his professional life,11
and it was one of his favourite tools for obtaining results in metamathematics, universal
algebra and infinitary logic. His first published work, [21], was a term paper he wrote
for Leśniewski’s seminar while just a third-year student at the University of Warsaw,
analysing Cantor’s notion of a well-ordered set. One of his last published works, a
monograph written jointly with John E. Doner and Andrzej Mostowski which was
published in 1978 3 years after Mostowski’s death, was a metamathematical study of the
elementary theory of well-ordering.12 Among other things it established that this theory
admits elimination of quantifiers, is axiomatizable by an infinite recursive axiom set, and
consequently is decidable. Tarski also co-authored, with Richard M. Montague and Dana
S. Scott, a manuscript for a planned book, to have been titled An Axiomatic Approach to
Set Theory.13
Tarski’s set-theoretic works focused on three major areas: (a) general or “pure” set
theory, (b) connections between set theory, measure theory and Boolean algebras, and (c)
presenting aspects of set theory in abstract algebraic calculi. We cite five examples:
2.1 Cardinal Arithmetic
He developed the arithmetic of cardinal numbers, and he established recursive formulas

for the exponentiation of alephs.14
2.2 Axiom of Choice
He studied the Axiom of Choice by looking for equivalents of it, weaker forms of it, and
equivalents of its weaker forms, and mapping out logical relations between the Axiom of
Choice and other sentences of set theory, such as the Generalized Continuum Hypothesis,
11 He seemed to tire of it for a spell in 1936 when he wrote to Karl Popper, “Ich arbeite an einer
Monographie aus der Mengenlehre, aber es interessiert mich wenig: alte Sachen, mit denen ich mich
schon seit Jahren nicht beschäftigt habe.” It is not clear if “schon seit Jahren nicht” was truth or posturing.
Or possibly he just meant he had lost interest in writing a survey of established results; he preferred to
work on getting new results. Sadly, the monograph he was referring to, Theorie der eineinendeutigen
Abbildungen, which he was writing jointly with Adolf Lindenbaum and which was to have been “ein
großes mathematisches Buch”, paid with its life. No working drafts ever surfaced, as far as anyone today
knows.
12 Where “elementary theory of well-ordering” is understood as the set of all formulas of first-order
predicate calculus that are true in every structure <U, R>, where U is a non-empty set and R is a (binary)
relation which well-orders the set U.
13 Montague died in 1971. For reasons which remain unclear, Scott and Tarski ceased work on the
manuscript in 1972. It remains unpublished to this day.

14 See [25] and [30f].
and sentences postulating the existence of very large cardinal numbers. Wacław Sierpiński
had proposed such a research program in 1918, in his survey paper L’axiome de M.
Zermelo et son rôle dans la Théorie des Ensembles et Analyse. From the late 1920s until
the mid 1950s Tarski and Sierpiński collaborated and competed to populate the program
with results.
Tarski established about 30 equivalents of the Axiom of Choice, many of which
concerned operations on or relations between cardinal numbers.15 One of the simplest
he found was this proposition:
m = m · m for every infinite cardinal number m.
From time to time he posed open questions. In 1924 he asked whether “m = 2 · m”
was equivalent to the Axiom of Choice.16 This was finally answered—in the negative, by
Gershon Sageev—only in 1975. The solution had to wait for a new technique of model
construction to be invented.
In [26a] Lindenbaum and Tarski stated without proof that the Generalized Continuum
Hypothesis (in the version that speaks of “transfinite numbers” rather than alephs) implied
the Axiom of Choice.17 Proofs of this remarkable result were found by others only in 1947
(by Wacław Sierpiński) and 1954 (by Ernst Specker).
In 1939 he established that the Axiom of Choice is a consequence of [one version of]
the assertion that inaccessible sets exist18 :
For every set A there exists a set M with the following properties:
(i) A is equipollent (equinumerous) to a subset of M;
(ii) the family of subsets of M which are not equipollent to M is equipollent to M;
(iii) there exists no set B such that the family of all subsets of B is equipollent to M.
As mentioned above, he also studied weaker versions of the Axiom of Choice, such as
the principle of dependent choice,19 and the Boolean prime ideal theorem.20
2.3 Finite Sets
He gave both a philosophical and a formal analysis of the notion of “finite set” that
avoided making any use of the Axiom of Choice or the axiom of infinity, and he showed
how the arithmetic of natural numbers could be defined in purely set-theoretic terms if
finiteness was taken to mean:
15 See [24a], [26], [38d], [39b], [48b], [49], [54], [64a b].
16 In[49m ] cardinals satisfying “m = 2 · m” were introduced on an abstract level as “idem-multiple”
(a + a = a) elements of a cardinal algebra.
17 Adolf Lindenbaum first posed it as an open question. In 1925 Lindenbaum and Tarski jointly proved it,
and asserted it without proof in [26a], §1, page 314, theorem 94.
18 See [39b].
19 See [48b].
20 See [54a f], [54a g], [54a h].
434 J. Zygmunt
The set A is finite iff every non-empty family of subsets of A contains a minimal element
with respect to inclusion.21
He framed sentences respectively equivalent to the Axiom of Choice and to the
Generalized Continuum Hypothesis in terms of finite sets as so defined.22 In [65a] he
announced several results obtainable in “weak” set theories—i.e., set theories without the
Axiom of Choice—having to do with D-finite infinite sets and D-finite infinite cardinals.23
The existence of such sets and cardinals had been conjectured by Henri Lebesgue
(1904) and by Russell and Whitehead (“mediate cardinals” in the terminology of
Principia Mathematica, *124·61). Tarski proved there was a set S of D-finite infinite
cardinals that was isomorphic to the set R of real numbers, under the “natural” ordering
of cardinals and reals. Issues of this kind could only have occurred to someone who cared
about fundamental things; for whom the notion of a set still required critical analysis. His
work on finiteness inspired many other authors. Andrzej Mostowski was the first, with
a metamathematical treatise titled On the Independence of Definitions of Finiteness in a
System of Logic (1938), followed by Azriel Lévy (1958), Arthur L. Rubin, Jean E. Rubin,
Erik Ellentuck (1962, 1965, 1968), John K. Truss (1972, 1984) and Agatha C. Walczak-
Typke (2005).
2.4 Theory of Large Cardinals
In a series of papers spanning a period of almost 35 years, from 1930 to 1964, he laid the
foundations for the theory of large cardinal numbers, including . . .
• a new definition of strong inaccessibility, and characteristic properties of inaccessible
cardinals24;
• several formulations of axioms asserting the existence of inaccessible sets (one such
axiom cited above)25;
• characteristic properties of strongly compact, measurable, and weakly compact cardi-
nals, and open questions surrounding them26 ;
21 See [24c]. The reader should bear in mind that “minimal element” and “least element” are different
concepts. Notice that this definition is independent of the notion of a finite natural number.
22 See [38c], page 163.
23 A set is said to be Dedekind-finite, or D-finite, iff there is no bijection of the set onto a proper subset of
itself (or equivalently, iff every one-to-one mapping of the set into itself is surjective.) A cardinal number
is said to be a Dedekind-finite cardinal, or a D-finite cardinal, iff it is the cardinality of a D-finite set.
In weak set theory—without the Axiom of Choice—it can be proved that if a set is finite by Tarski’s
definition then it is also D-finite, but it cannot be proved that if a set is D-finite then it is also finite by
Tarski’s definition.
In [49m ] D-finite cardinals were introduced on an abstract level as finite elements of a cardinal algebra.
24 See the joint paper with Wacław Sierpinski [30a].
25 See [38a] and [39b].
26 See [62]; the joint papers [43] and [61b] with Paul Erdös; and the joint paper [64] with H. Jerome
Keisler.
• a research program to investigate interconnections between large cardinals and infini-

tary logic (some of Tarski’s students—William Porter Hanf and H. Jerome Keisler
among them—obtained important results in this area).
2.5 Algebraization of Set Theory
He algebraized key aspects of the general theory of sets, having to do with order types,
relation types, cardinal and ordinal arithemetics and operations on infinite sequences.
He summarized his work in these areas in the books: Cardinal Algebras [49m ], and
Ordinal Algebras [56m a]. Both books could trace their origins to the 1926 Lindenbaum–
Tarski paper Communication sur les recherches de la théorie des ensembles, in which
Lindenbaum’s role had been immense. Tarski conceded that many results “were originally
established by Lindenbaum” and were first stated without proof in [26]. He added that
it was impossible to convey “an adequate idea of the extent of my indebtedness” to
Lindenbaum.
Substantial parts of several classic texts on set theory are based on Tarski’s results:
• W. Sierpiński’s Zarys teorii mnogości, third edition 1928; and Leçons sur les nombres
transfinis, 1928, second edition 1950 (cardinal arithmetic; and equivalents of the
Axiom of Choice);
• Sierpiński’s Cardinal and Ordinal Numbers, 1958, second edition 1965 (equivalents
of the Axiom of Choice and proofs of theorems from Tarski–Lindenbaum [26]);
• H. Bachmann’s Transfinite Zahlen, 1955, second edition 1967 (equivalents of the
Axiom of Choice and inaccessible numbers);
• K. Kuratowski and A. Mostowski’s Set Theory, second edition 1976 (generalizations
of Banach’s formulation of the Cantor-Bernstein theorem, Tarski’s recursive formulae
for exponentiation of alephs; the number of prime ideals in the power-set algebra of
an arbitrary set; basic cardinal equivalents of the Axiom of Choice; the exposition of
higher types of inaccessible numbers).
Tarski almost singlehandedly steered the development of set theory: dictating what
counted as set theory; what were its important questions, results, applications, methods
and tools; where were its frontiers; what was worth working on, and why. In the words of
Azriel Lévy27:
Alfred Tarski started contributing to set theory at a time when the Zermelo-Fraenkel axiom system
was not yet fully formulated, and as simple a concept as that of the inaccessible cardinal was not
yet fully defined. At the end of Tarski’s career the basic concepts of the three major areas and
tools of modern axiomatic set theory, namely constructibility, large cardinals and forcing, were
already clearly defined and were in the midst of rapid successful development. The role of Tarski
in this development was somewhat similar to the role of Moses showing his people the way to the
Promised Land and leading them along the way, while the actual entry into the Promised Land was
done mostly by the next generation. The theory of large cardinals was started mostly by Tarski,
and developed mostly by his school. The mathematical logicians of Tarski’s school contributed
27 A. Lévy, Alfred Tarski’s Work in Set Theory, The Journal of Symbolic Logic, vol. 53 (1988), pp. 2–6;
p. 2.
436 J. Zygmunt
much to the development of forcing, after its discovery by Paul Cohen, and to a lesser extent also
to the development of the theory of constructibility, discovered by Kurt Gödel. As in other areas of
logic and mathematics Tarski’s contribution to set theory cannot be measured by his own results
only; Tarski was a source of energy and inspiration to his pupils and collaborators, of which I was
fortunate to be one, always confronting them with new problems and pushing them to gain new
ground.
3 Geometry and Measure Theory
Tarski’s work in geometry showcases his fascination with metamathematics, axiom

systems and the axiomatic method, and it reveals his genius for developing innovative
tools and techniques to extract new ore from old seams. He gave serious thought to
the choice of primitive concepts, axiom sets and logical bases for various geometries—
affine, hyperbolic, elliptical, Euclidean, projective—and, for each geometry, to the
independence, economy and strength of its axioms, the independence, definability and
predicativity of its notions, and the decidability and completeness of the system as a
whole.28 A volume he co-edited with Leon Henkin and Patrick Suppes, [59e ], The
Axiomatic Method, with Special Reference to Geometry and Physics, underscores
geometry’s pride of place by its title alone.
Tarski’s earliest paper on geometry, [24b], O równoważności wielokatów
˛ <On the
Equivalence of Polygons>, was of a different character. It was primarily a study of
measure-theoretical properties of point sets in the Euclidean plane, R2 . The paper is
important from a historical perspective as it contains:
• One of the first formulations of the definition of equivalence by finite decompositions:
“Two geometric figures (thus in particular, two polygons) are called equivalent when
it is possible to divide them into the same finite number of respectively congruent
arbitrary geometric figures not having any common points.”
• A proof of the Wallace–Bolyai–Gerwien theorem: two polygons are equivalent iff they
have equal areas. Tarski’s proof relied essentially on Banach’s theorem on additive
measures on the plane.
• The first announcement of the result known as the Banach–Tarski Paradox: “Any two
polyhedra are equivalent” or, in an alternative formulation: “Any cube can be divided
into a finite number of parts (without common points), which can then be reassembled
to form a cube with an edge twice as long”.29
• The question, “Czy koło i wielokat ˛ o równych polach sa˛ równoważne?” (Are a circle
and a polygon with equal areas equivalent [by finite decompositions]?). This question
has since come to be known as “the Tarski Circle Squaring Problem”, . . . a name with
echoes of antiquity.30
28 See [34], [56c], the joint paper [56b] with E.W. Beth, and the joint papers [65a] and [79] with L.W.
Szczerba.
29 The Banach–Tarski Paradox was fully articulated, and its proof elaborated, in [24d], a subsequent paper
co-authored with Banach.

30 Tarski’s circle-squaring problem was finally answered in the affirmative in 1990 by Miklós Laczkovich.
See: M. Laczkovich, Equidecomposability and Discrepancy: a Solution to Tarski’s Circle-squaring

Jan Mycielski, an authority on paradoxical decompositions, writes:

The Banach-Tarski paradox is not an inconsistency of mathematics, although it shows that 2
equals 1 in a certain sense. It is proved within the usual system of set theory (using the Axiom
of Choice), and we have no reason to doubt the consistency of that system. But it destroys certain
naive intuitions about point sets in three-dimensional space R3 . . . .
This theorem is a striking demonstration that the unrestricted concept of a set of points has little to
do with the idea of a physical body, and also that, to develop a reasonable theory of areas, volumes,
etc., one must limit oneself to more special sets (e.g., Borel sets, or Lebesgue measurable sets).31
In [30] Tarski gave an effective proof—i.e., without using the Axiom of Choice—that
for any infinite set E the following two statements were equivalent:
1. There exists a finitely additive two-valued measure m on the family of all subsets of E
such that m(E) = 1 and for every finite subset F ⊆ E, m(F) = 0.
2. There exists a maximal ideal in the field of all subsets of E.
Then he proved that 2 was a theorem of general set theory with the Axiom of Choice
added—i.e., that the Axiom of Choice implied 2. His demonstration of a maximal ideal
relied on the well-ordering principle and induction on transfinite ordinals. He developed
this same technique and line of reasoning further in the papers [39, 45] on ideals of various
types—m-additive, prime, p-saturated—in complete fields of sets.32
In [29b] and [38g] Tarski explored logical connections between the existence of
paradoxical decompositions and the non-existence of some invariant measures. Following
von Neumann’s terminology, for any arbitrary set E, any subset I of E, and any group G
of transformations of E, a function m defined on the power set of E with values in the
set of all nonnegative real numbers is said to be an [E, I, G]-measure iff m is additive,
m is invariant under transformations in G, and m(I) = 1. Tarski’s remarkable result
was:
In order for an [E, I, G]-measure to exist, it is necessary and sufficient that there be no
paradoxical decompositions of I relative to G.
For an algebraic treatment of this and related issues see [49m ], Cardinal Algebras,
theorems 14.13, 16.8, 16.12 and 16.13.
Tarski’s particular achievement in the foundations of geometry was conceiving the
idea of a system of geometry based only on first-order logic with identity (i.e., elementary
logic) and completely free from any set theoretical assumptions (set theoretical notions,
primitive terms, axioms, rules of inference), and showing that such a system was plausible.
He called this system elementary geometry.
Problem, Journal für die Reine und Angewandte Mathematik, vol. 404 (1990), pp. 77–117. Laczkovich
proved that the circle could be decomposed into no more than 1050 different pieces, which could
be rearranged to compose a square of equal area. He needed the Axiom of Choice to obtain his
decomposition, which was highly non-constructive.
31 J. Mycielski, Review of The Banach–Tarski Paradox by Stan Wagon, The American Mathematical
Monthly, vol. 94, no. 7, pp. 698–700. The quoted passages are from page 698.
32 Though published 6 years apart, [39] and [45] were nominally parts I and II of the same two-part paper,
and appeared in consecutive issues of Fundamenta Mathematicae, vol. 32, pp. 45–63, and vol. 33, pp.
51–65. The journal’s operations were interrupted by the Second World War.
438 J. Zygmunt
Tarski’s axiomatic system of two-dimensional elementary geometry, E 2 , has only two

specific (or primitive) notions: a ternary predicate constant β for the relation of between-
ness and a quaternary predicate constant δ for the equidistance relation. The formula
β(x,y,z) is read: y lies between x and z, while δ(x,y,z,u) is read: x is as distant from y
as z is from u. Individual variables represent only points. By contrast, Hilbert’s system
of geometry has other primitive geometrical notions in addition to points, for instance
lines and planes. Tarski limited himself to points by exploiting the fact that figures
(configurations of points) dealt with in traditional geometry are uniquely determined by
finite numbers of points.
His system E 2 is based on twelve axioms and one axiom schema, each of which is
framed solely in primitive and logical terms. Despite avoiding using defined terms, the
axioms are relatively short and their meanings are intuitively clear. The axiom schema
defines an infinite recursive set of axioms, and plays the role of Dedekind’s continuity
axiom—or more precisely, it corresponds to Dedekind’s continuity axiom restricted to
[all and only] sets which are first-order definable in the language of E 2 . Relaxing this
restriction and going to a second-order language leads directly to the whole of Euclidean
geometry. This means that Tarski did successfully separate elementary geometry from
full geometry.
To define a model for E 2 Tarski applied the apparatus of his formal semantics: in
particular, his definition of truth. Namely, a relational system M = <A, B, D> is a
model of E 2 iff . . .
(i) A is an arbitrary non-empty set, and B and D are respectively a ternary and a
quaternary relation among elements of A; and
(ii) All the axioms of E 2 hold in M if all the variables are assumed to range over
elements of A, and the primitive constants β and δ are understood to denote the
relations B and D, respectively.
The representation theorem reads:
A relational system M is a model of E 2 iff M is isomorphic with the Cartesian space
C 2 (F ) over some real closed field F .
Thus the axiom set for E 2 is sound and adequate. Moreover, since the set of its
theorems is identical with the set of all true sentences in the Cartesian space C 2 (R) over
the field of real numbers, the system E 2 is complete, and consequently decidable. And the
so-called “Non-finitisability theorem” says that E 2 is not finitely axiomatizable.
Tarski [59], What is elementary geometry?, was first published in [59e ]: L. Henkin, P.
Suppes and A. Tarski (eds.), The Axiomatic Method, with Special Reference to Geometry
and Physics, pp. 16–29. It is one of that volume’s most highly polished gems. Perhaps it
owes its lapidary sheen to 33 years of being turned over and over again in Tarski’s great
tumbler of a head, starting all the way back in 1926.
For a detailed account of the long and fascinating history of Tarski’s thinking on his
system of elementary geometry see [99].
4 Decidable and Undecidable Theories
Let’s begin with a quotation:

By a decision procedure for a given formalized theory T we understand a method which permits
us to decide in each particular case whether a given sentence formulated in the symbolism of T can
be proved by means of the devices available in T (or, more generally, can be recognized as valid in
T). The decision problem for T is the problem of determining whether a decision procedure for T
exists (and possibly of exhibiting such a procedure). A theory T is called decidable or undecidable
according as the solution of the decision problem is positive or negative. As is well known, the
decision problem is one of the central problems of contemporary metamathematics. Since only
few theories turn out to be decidable, most endeavors are directed toward a negative solution.33
Although the search for particular decision procedures was plainly evident in Ernst
Schröder’s focus on the “Lösungsprobleme” (solution problems) for his relation algebras,
an awareness that the decision problem as defined in [53m ] was a metamathematical
issue in its own right emerged later, in the Hilbert school, with its work on the
“Entscheidungsproblem” (decision-making problem) to do with the decidability of the
“restricted functional calculus”—a particular version of first-order predicate logic. In
1936 Alonzo Church and Alan Turing published independent papers showing that, in
general, first-order theories were undecidable. Some special cases of first-order theories,
such as Presburger arithmetic, were shown to be decidable, but these were more the
exception than the rule.
In his 1946 Princeton address34 Tarski expressed the view that the decision problem
was one of the central issues—if not the central issue—of metamathematics of the
day, noting that, “Hilbert considered the main task of logic to be the construction of a
symbolism for use in solving the general decision problem: this was the raison d’être
of metamathematics.” He presented various open questions in logic and mathematics, at
that time including Cantor’s continuum hypothesis and Hilbert’s tenth problem. Then he
surveyed contemporary work and known results on the decision problem and proposed a
research program on it.
Broadly speaking, Tarski’s contributions to the decision problem were:
1. Proving that certain important theories formalized in classical first-order predicate
logic were decidable.
2. Laying the foundations for a general method for proofs of undecidability.
3. Proving the undecidability of certain important first-order and non-first-order theories.
There are several ways of proving theories decidable—model-theoretic techniques,
syntactic methods, and others. One way is simply to reduce the decidability of the
theory under question to that of a theory already known to be decidable, either by an
33 Undecidable Theories by Alfred Tarski, in collaboration with Andrzej Mostowski and Raphael M.
Robinson. North-Holland Publishing Co., Amsterdam, 1953. The quoted paragraph is from Chapter I, A
General Method in Proofs of Undecidability, §I.1. Introduction, page 3.
34 See H. Sinaceur (ed., with introduction), Address at the Princeton University Bicentennial Conference
on Problems of Mathematics, December 17–19, 1946, by Alfred Tarski, The Bulletin of Symbolic Logic,
vol. 6 (2000), pp. 1–44. See also Odczyt Alfreda Tarskiego na Konferencji o Problemach Matematyki w
Princeton, 17 grudnia 1946, in [01m ], pp. 396–413.
440 J. Zygmunt
embedding, an extension, or a representation. In proofs of decidability, the full-blown

theory of recursive functions is often not required. It is enough just to recognize that
certain procedures are algorithmic or effectively calculable.35
Historically one of the earliest techniques ever used for proving a theory decidable was
the method of eliminating quantifiers . . . a syntactical “process” whose first appearance
in the literature was in a paper by Leopold Löwenheim (1915),36 and which seems to
have been inspired by Schröder’s work on the Eliminationsproblem. The scare quotes
around “process” are intended as a warning that quantifier elimination is not a mechanical
procedure. Doner and Hodges (1988) gives a beautifully readable précis of how it
works37 :
Let T be a first-order theory. We say that a set Φ of formulas is a set of basic

formulas for T (or more briefly a basic set) if, using the axioms of T, one can prove
that every formula φ of the language is equivalent to a Boolean combination φ* of
formulas in Φ which has only the same free variables as φ.
To analyze T by the method of quantifier elimination, one would look for a
basic set for T. Every axiom system has at least one basic set, namely the set of
all formulas of the language. But one would try to find a better basic set than
that. There is no exact criterion for a “good” basic set, but one would hope for a
basic set Φ with at least the following three properties. (1) It should be reasonably
small and irredundant. (2) Every formula in Φ should have some straightforward
mathematical meaning. (3) There should be an algorithm for reducing every
formula φ to its corresponding φ*.
(3) is precise. When we have it, we can claim to have effective quantifier
elimination for T. (1) and (2) are more a matter of judgment.
In the best cases we have one thing more: (4) an algorithm which tells us, given
any basic sentence , either that is provable or that it is refutable from T. Given
(3) and (4), we have both a completeness proof and a decision procedure for the
theory T.
To quote Chang and Keisler, “This method applies only to very special theories.
Moreover, each time the method is applied to a new theory we must start from scratch . . .
On the other hand, the method is extremely valuable when we want to beat a particular
theory into the ground. When it can be carried out, the method of elimination of quantifiers
35 Formore comments on this see Tarski [48m ], note 10.

36 See: L. Löwenheim, Über Möglichkeiten im Relativkalkül <On Possibilities in the Calculus of
Relatives>, Mathematische Annalen, vol. 76 (1915), pp. 447–470.
37 The four paragraphs in the box are an extended quote from: John Doner and Wilfrid Hodges, Alfred
Tarski and Decidable Theories, The Journal of Symbolic Logic, Vol. 53, No. 1, (March, 1988), pp. 20–
35. The text reproduced here is from page 24 of their article, where they present it in three paragraphs,
not four.
gives a tremendous amount of information about a theory . . . [it] may be thought of as a

direct attack on a theory.”38
Three things should be emphasized. First, quantifier elimination is not just, or even
primarily, about proving a theory decidable. It is about discovery generally. Second, it
can be applied to all formulas, not just to sentences. Free variables are welcome. Third, it
is more art than science. But when it works, it works a treat.
Tarski encountered the method as a student and applied it in his seminar exercises when
he was a docent. One of the exercises he set Mojżesz Presburger was to use the method
of quantifier elimination to investigate the completeness of the arithmetic of integers with
addition.39
Using the method of elimination of quantifiers, Tarski contributed to a deeper
understanding of the first-order theories of dense orders and linear orders.40 He proved,
among other things, that every sentence in the elementary theory of dense order, TDO , was
deductively equivalent to a Boolean combination of the following two (basic) sentences:
“There is no first element”, and “There is no last element”. Since the elimination of
quantifiers was “effective”, in Doner and Hodges’s precise sense (3) above, TDO turned
out to be decidable. As a corollary, Tarski obtained a classification of all axiomatic
extensions of TDO , as well as a semantic characterization of all complete extensions of
TDO . There are four such complete extensions—the theories of the order types η, η + 1,
1 + η, and 1 + η + 1.41
Together with Andrzej Mostowski and John Elliott Doner, Tarski proved that the
elementary theory of well-ordering was decidable, described the ordinals definable by
first-order formulas, and gave a description and classification of all models of the theory.42
Again using quantifier elimination he obtained a series of decidability results on
“elementary algebra and geometry” and “related systems”.43 It is worth noting a few
38 See: C.C. Chang and H.J. Keisler, Model Theory, North-Holland: Amsterdam 1973 (3rd edition, 1990),
pages 49–60. This quotation is from page 49 in the 3rd edition.
39 See: M. Presburger, Über die Vollständigkeit eines gewissen Systems der Arithmetik ganzer Zahlen,
in welchem die Addition als einzige Operation hervortritt, Sprawozdanie z I Kongresu Matematików
Krajów Słowiańskich, Warszawa 1929 (=Comptes-rendus du I Congrès des Mathématiciens des Pays
Slaves), Ksi˛eżnica Atlas: Warszawa 1930, pp. 92–101 & p. 395; see p. 97, footnote 1.
See also: W. Hodges, A Visit to Tarski’s Seminar on Elimination of Quantifiers, in J. van Benthem et
al. (eds.), Proof, Computation and Agency, Synthese Library 352, Springer Science+Business Media
B.V. 2011; pp. 53–66.
40 In [36d], §5, Tarski axiomatized these two theories, calling them, respectively, “die elementare Theorie
der dichten Anordnung” (p. 290), and “die elementare Theorie der isolierten Anordnung” (p. 294). In
footnote 1 on page 293 he asserted that his results on the theory of dense order supplemented those
reported in Langford’s 1927 paper. (Literally: “Die unten angegebenen Tatsachen, die diese Theorie
betreffen, bilden die Ergänzung der Ergebnisse Langfords”). He used the phrase “sukzessiven Elimination
der Operatoren” (successive elimination of quantifiers) in print for the first time on pp. 293 and 295 . . .
somewhat odd considering he had been promoting the method since 1926 or earlier.
41 By “a theory of an order type α” is meant, the set of all first-order sentences true in any structure <X,R>,
where the relation R orders the set X according to the type α.

42 See [78]. Precise attribution of specific results is a long and complicated story. See especially the
historical comments on page 1.

43 These results were announced in a variety of published works, scattered over time, and in varying
notations and terminologies. See: [31], [48m ], [48m ](1), [49a d], and [67m ]. The last was originally
intended for publication in 1939 but was interrupted by the Second World War.
442 J. Zygmunt
of these results that are pertinent to the ordered field of real numbers. (For those related
to geometry, see §3 above.)
Let R = R, +, ·, >, 0, 1, −1! be this field, where +, ·, >, 0, 1 and −1 have their usual
meanings. The system of elementary algebra is by definition the set Th(R) of all first-
order sentences formulable in the vocabulary of R, +, ·, >, 0, 1, −1! which are true in
R. Tarski proved that the theory Th(R) admits effective elimination of quantifiers, and he
described a decision procedure for it. Moreover, he showed that Th(R) is axiomatized by
the first-order axioms for real closed fields and, in consequence, any two real closed fields
are arithmetically indistinguishable (or, using Tarski’s later terminology, elementarily
equivalent). As a by-product, one gets the following important theorem:
• a set of real numbers is first-order definable in the field R iff it is a sum of a finite
number of intervals (open or closed, bounded or unbounded) with algebraic end
points.
In [48m ](1), p. 45, Tarski asked about the decidability of the elementary theory of
the real field expanded by the “exponential with a fixed base, for example base 2”, i.e.,
the theory Th ( R, +, ·, >, 0, 1, −1, Exp!), where Exp is the unary operation given by
the formula y = 2x for all x ∈ R. Commenting on this problem Tarski wrote, “The
decision problem for the system just mentioned is of great theoretical and practical
interest. But its solution seems to present considerable difficulties. These difficulties
appear, however, to be of purely mathematical (not logical) nature: they arise from
the fact that our knowledge of conditions for solvability of equations and inequalities
in the enlarged system is far from adequate.” In [67m a], p. 38—an earlier version of
[48m ], originally to have appeared in 1940, with an emphasis on completeness in its
title—Tarski posed the problem of finding a complete set of axioms for the theory
Th ( R, +, ·, >, 0, 1, −1, Exp!), commenting on it in a similar vein: “The attempt to
carry out [this task] is confronted by difficulties of a purely mathematical nature which
nevertheless do not appear to raise any question of principle”.44
Another question raised by Tarski in [48m ] was about the decidability of the theory
of the structure Th R, +, ·, >, 0, 1, −1, Al! where the relation Al(x) means “x is an
algebraic number”. Abraham Robinson proved (1959) that one obtains a complete theory
44 Tarski was perfectly right that difficulties resided in mathematics, and this has been confirmed by the
further development of algebraic geometry. The importance of the two problems is discussed in: D.E.
Marker, Model Theory and Exponentiation, Notices of the American Mathematical Society, vol. 43
(1996), pp. 753–759.
Tarski’s theorem that the ordered field of real numbers (a first-order theory) admits quantifier
elimination triggered a stream of research. Important contributions were made by: Abraham Seidenberg
(1954), Abraham Robinson (1959 and 1971), Stanisław Łojasiewicz (1964–65), Paul Joseph Cohen
(1969), Joseph R. Shoenfield (1971), George Edwin Collins (1982), and Helmut Wolter (1986). They
are summarized in an expert presentation by Lou van den Dries: Alfred Tarski’s Elimination Theory for
Real Closed Fields, The Journal of Symbolic Logic, vol. 53 (1988), pp. 7–19.
A comprehensive review by Charles I. Steinhorn of Alex J. Wilkie’s paper Model Completeness
Results for Expansions of the Ordered Field of Real Numbers in The Journal of Symbolic Logic, vol.
64 (1999), pp. 910–913, contains a survey of research up to the end of the 1990s stimulated by Tarski’s
beautiful, influential, and far-reaching monograph [48m ], A Decision Method for Elementary Algebra
and Geometry. See also Bob F. Caviness and Jeremy Russell Johnson (eds.), Quantifier Elimination and
Cylindrical Algebraic Decomposition. Wien, New York: Springer 1998.
by adding to the axioms for the real closed ordered fields sentences which state that the
elements of the model which satisfy Al(x) constitute a real closed proper subfield which
is dense in the entire model.
Tarski once opined, “In fact, I am rather inclined to agree with those who maintain
that the moments of greatest creative advancement in science frequently coincide with
the introduction of new notions by means of definition”.45 As Feferman and Feferman
observed, a touch swellheaded. But he knew he could get away with it. Tarski himself
defined many notions that contributed enormously to advances in logic, metalogic and
metamathematics. A general theory of undecidability could never have been created if
Tarski had not given precise mathematical definitions of two concepts: (1) that of an
essentially undecidable theory, and (2) of the interpretability of one theory in another.
The two concepts were introduced, and a theory of undecidability expounded, in the
paper A General Method in Proofs of Undecidability, which constituted Chapter 1 of
the book [53m ], Undecidable Theories. Though authorship of the book as a whole was
credited to Tarski in collaboration with Andrzej Mostowski and Raphael M. Robinson
(and with some later contributions by Julia Robinson), this first part was penned by Tarski
alone.
Theories considered in [53m ] were all formalized in a standard way within first-
order predicate logic with identity (without predicate variables). The set of syntactically
articulable formulas of the language of any theory was general recursive. A set of valid
sentences was identified for each theory T, subject to the condition that the set be closed
under consequence operations of predicate logic. T was said to be axiomatizable if every
valid sentence could be logically derived from a fixed recursive set of axioms, and finitely
axiomatizable if the set of axioms could be finite. If the set of valid sentences was
recursive, T was said to be decidable; otherwise T was called undecidable.
A theory T2 was said to be an extension of a theory T1 (and T1 a subtheory of T2 )
if every valid sentence of T1 was valid in T2 . Two theories T1 and T2 were said to be
compatible if they had a common consistent extension—equivalent to saying that the
union of T1 and T2 was consistent. And finally, a theory T was said to be essentially
undecidable, if it was not only undecidable, but also every consistent extension of T
with the same constants as T was undecidable.46,47 Key properties of undecidability
and essential undecidability were then presented in several theorems, among them the
following:
• Let T1 and T2 be two compatible theories such that every constant of T2 is also a
constant of T1 . If T2 is finitely axiomatizable and essentially undecidable, then T1
is undecidable, and so is every subtheory of T1 which has the same constants as T1
(Theorem 6, page 18).
45 See [44a], p. 359.

46 Tarski’suse of the word “essentially” here came from a distinction he drew between two different ways
of extending a theory. An extension T2 of T1 was called inessential, if every constant of T2 which did not
occur in T1 was an individual constant, and every valid sentence of T2 was derivable from a set of valid
sentences of T1 .
47 By the work of Church and Rosser from the year 1936, elementary Peano arithmetic is essentially
undecidable.
444 J. Zygmunt
To exploit all these mechanisms, Tarski needed two more notions: (1) interpretability,
and (2) relativization of quantifiers.
First, interpretability. Let T1 and T2 be two theories. Assume, first, that they have no
non-logical (or “specific”) constants in common. In this case, T2 is said to be interpretable
in T1 , if T1 can be extended by adding to its set of valid sentences “possible” definitions
of the non-logical constants of T2 in such a way, that the resulting extension of T1 turns
out to be an extension of T2 as well. Then, in the case when T1 and T2 do have some
non-logical constants in common, first rewrite T2 with new non-logical constants that do
not occur in T1 , but without changing the syntactical structure of T2 in any other respect.
If the resulting theory T2 is interpretable in T1 , then T2 may be said to be interpretable
in T1 as well. And last, a theory T2 is said to be weakly interpretable in T1 , if T2 is
interpretable in some consistent extension of T1 with all the same constants (both logical
and non-logical).
Now, relativization of quantifiers.48 Let P be any unary predicate constant not in
the language of T. For every formula of T in which P does not occur, replace all
occurrences of ∀x with ∀x(Px→), and replace all occurrences of ∃x with ∃x(Px∧).
Call the result T(P ) . Notice that this construction creates a one-to-one mapping from the
formulas of T to the formulas of T(P ) . Apply this mapping to the set of valid sentences
in T, then apply standard consequence operations to the image-set to derive its logical
closure in T(P ) , and let that closure be the set of valid sentences of T(P ) . Then T(P ) is said
to be obtained from T by relativization of the quantifiers of T to P.
The requirement that P initially not occur in the language of T was, as things turned
out, not logically necessary. But Tarski could not prove his main meta-theorem about T(P )
without making this assumption.
The transformation of T into T(P ) preserves some important properties of T in T(P ) .
For instance . . .
(1) T(P ) is axiomatizable iff T is axiomatizable.
(2) When only finitely many individual constants and operator symbols occur in T, T(P )
is finitely axiomatizable iff T is finitely axiomatizable.
(3) T(P ) is essentially undecidable iff T is essentially undecidable (Theorems 9 & 10).
By merging interpretability with relativization one obtains relative interpretability.
A theory T2 is said to be relatively interpretable [/relatively weakly interpretable] in a
theory T1 iff the correlated theory T2 (P ) (obtained by relativizing the quantifiers of T2 to
a predicate P) is interpretable [/weakly interpretable] in T1 in the senses defined above.
To apply this apparatus, one needs a simple, finitely axiomatizable, essentially unde-
cidable theory. The second paper in [53m ], Undecidability and Essential Undecidability
in Arithmetic, written jointly by A. Mostowski, R.M. Robinson, and A. Tarski, constructs
just such a theory. This is the famous theory Q, a subtheory of Peano arithmetic.
48 The general idea of relativization was introduced by Tarski in 1930. The idea of relativization of
quantifiers was due to Adolf Lindenbaum and Tarski, and dated from 1935. See [56m ], p. 69, p. 314
(footnote 1) and p. 396. With Tarski as one of his doctoral advisors, Andrzej Mostowski used the method
of relativization of quantifiers in his Ph.D. dissertation, which he published under the title O niezależności
definicji skończoności w systemie logiki <On the Independence of Definitions of Finiteness in a System of
Logic>, in Dodatek do Rocznika Polskiego Towarzystwa Matematycznego (Supplement to the Annales
de la Société Polonaise de Mathématique), Vol. 11 (1938), pp. 1–54.
It has seven simple axioms describing the successor function (a unary operation),
multiplication and addition (binary operations), and the number zero. The authors show
that Q is essentially undecidable, since any recursive function of a single variable is
metamathematically definable in Q. The proof relies on Julia Robinson’s characterization
of recursive functions of a single variable, and a version of Tarski’s theorem on
undefinability of arithmetical truth, which together make it possible to avoid having to
construct a special provability predicate.
In the third paper of [53m ], Undecidability of the Elementary Theory of Groups, written
by Tarski alone, the headline result is proved by demonstrating the following . . .
(i) The complete theory Th( I,+,·!) of integer arithmetic is undecidable, since there
exists a natural relativization of Q that is interpretable in Th( I,+,·!).
(ii) The theory Th( I,+,·!) is itself interpretable in the elementary theory of groups in
such a way, that a decision procedure for the theory of groups would yield a decision
procedure for Th( I,+,·!).
Other results obtained by the method of interpretation or relative interpretation include
the undecidability of the elementary theory of lattices, of modular lattices, of modular
lattices with complementation, and of rings. The method was also used in proving a weak
system of set theory undecidable (see [52b], by Szmielew and Tarski). In the 1960s Tarski
and Szczerba used the method of interpretation to investigate the decision problem in
geometry.
Tarski’s talk at the 1946 Princeton Logic Conference on Problems of Mathematics
began with a short reflexion of a semantic nature:
Now the word “problem” has two distinct senses: in one sense, a problem is a definite question
like “Is such-and-such the case?”; in another sense, we mean by a problem something of a less
determined nature—which could perhaps more properly be characterized as a task—such as
“Construct something with such-and-such properties.” It is in this second—more general, if you
prefer—sense that I will call the attention of this assembly to some important unsolved problems
in mathematical logic.
Tarski’s later publications treated the decision problem as a problem in the second,
more general sense, recasting it as a challenge to explore what decidability is really all
about, what are the different kinds of decidability, and which kinds are worth investigating
in more detail. He identified what he called restricted decision problems, and second-
degree decision problems.49
By restricted decision problems Tarski meant, “problems of determining whether a set
S of all valid sentences of a theory T satisfying certain additional conditions is recursive”.
He cited Hilbert’s tenth problem, the word problem for groups, and other word problems
as prime examples of problems of this type.50 In [87m ], §5.5 & §8.5, he gave three more
examples, and proved they were answerable in the negative51:
49 Both terms were coined in [53m ] (see pp. 34–35).

50 See [53m ], p. 35, and [68], p. 287.
51 To obtain these proofs he applied a method that he had earlier set out, in [53m ], p. 22, footnote 17,
which could be termed “generalized interpretation”.

446 J. Zygmunt
• Certain subsystems of classical propositional logic, finitely axiomatizable under the

rules of substitution and detachment, are undecidable.
• The equational theory of relation algebras and the equational theory of representable
relation algebras are undecidable.52
• The equational theory of omega-relation algebras is finitely based53 and essentially
undecidable.
By second-degree decision problems, Tarski understood a catch-all category of meta-
questions about questions—a category more easily limned by “e.g.” than by “=def. ”. He
gave his first example at the Princeton address, citing the (at that time open) question “of
finding a procedure to tell whether a given set of formulas is adequate as a set of axioms
for the sentential calculus”.54 In [53m], pp. 34–35, he considered “the problem of the
existence of a method which would permit us in each particular case to decide whether
or not a given theory is decidable”. Narrowing this down to something more tractable,
he asked if the family of all finite sets of sentences which are axiom sets for decidable
first-order theories with standard formalizations is recursive.
Based on the provable existence of an essentially undecidable theory, and on specific
properties of deducibility in theories with standard formalizations, Tarski concluded that
these questions were answerable in the negative. Analogous reasoning gave him a negative
answer to the question: Is the family of all finite consistent sets of sentences recursive?55
Abraham A. Fraenkel, Yehoshua Bar-Hillel, and Azriel Lévy surveyed a sweeping cast
of second-degree decision problems in their Foundations of Set Theory. They wrote56:
The reader might have already asked himself whether there might not exist a decision method by
which it could be effectively determined for every given formalized theory, whether it is decidable
or not. However, a rather simple argument shows that, at any rate for finitely axiomatizable first-
order theories, no such general method could possibly exist, hence that this second-degree decision
problem, so to speak, is unsolvable. [The authors cite [53m ], p. 35]. Other higher-degree decision
problems deal, for instance, with the existence of an effective procedure of deciding, for every
given presentation of a certain algebraic structure, such as of a semi-group, of a semi-group with
cancellation, or of a group, whether the structures defined by these presentations have certain
algebraic properties such as cyclicity, finiteness, simplicity, decomposability into a finite product,
etc.
In [68], on page 287, Tarski wrote, “Various notions, problems, and results discussed
so far in this paper suggest in a natural way corresponding decision problems. These are
52 For a class K of similar algebras (of a fixed similarity type τ), the equational theory of this class is
defined as the set of all identities (in type τ) which are true or hold in every algebra belonging to the
class K.
53 In equational logic finitely based means axiomatizable by a finite number of equations (identities).
54 That such a procedure does not exist was communicated in 1949 by Samuel Linial (Gulden) and Emil
Leon Post. For a detailed presentation of their result, see Mary Katherine Yntema, A detailed argument
for the Post-Linial theorems, Notre Dame Journal of Formal Logic, vol. 5 (1964), pp. 37–51.
55 See: George F. McNulty, Alfred Tarski and Undecidable Theories, The Journal of Symbolic Logic, vol.
51 (1986), pp. 890–898.

56 A. Fraenkel, Y. Bar-Hillel, A. Lévy, Foundations of Set Theory, Studies in Logic and the Foundations
of Mathematics, Vol. 67, Elsevier, Amsterdam 1958, (2nd edition 1973). The quotation is from the
second edition, Chapter 5, §7, The limitative theorems of Gödel, Tarski, Church and their generalizations,
page 320.
problems of the type: is a given set of equations, or of finite sets of equations, or of finite
algebras, recursive? The meaning of the term recursive in these contexts is clear; finite
algebras can be regarded as algebras whose universe consists of finitely many integers.”
He set out examples of such decision problems on pp. 287–288. Let range over finite
subsets of equations of a fixed finite similarity type—i.e., equations in which only a finite
number of operation symbols occur. Then the following six condition-schemata, with free
variable , can be postulated:
c1. is a base for a given finite algebra A.
c2. There is a finite algebra A for which is a basis.
c3. For a given positive integer k, the equational theory generated by has an
independent base consisting of k equations.
c4. The equational theory generated by is consistent.
c5. The equational theory generated by is complete.
c6. The equational theory generated by is decidable.
To each of the above six condition-schemata (ci ) there corresponds the decision
problem (Pi ): Is the family of all such that has property (ci ) recursive? Six parallel
decision problems can be obtained when is restricted to range over one-element sets of
equations.
Also consider the following three condition-schemata on finite algebras, with free
variable A, and their related decision problems:
e1. A is finitely based (that is, the set of all equations true in A is axiomatizable by a
finite set of equations).
e2. A is equationally complete.
e3. A has an independent base consisting of k elements, where k ≤ ℵ0 .
Of these, the problem associated with condition (e1), known as Tarski’s finite basis
problem, turned out to be most difficult, and attempts to solve it profoundly influenced
the development of universal algebra and computability theory for 30 years. It was only
in 1996 that Ralph McKenzie announced the result that the class of all finite algebras
which are finitely based is not recursive.57
Flashback and flashforward can be revealing devices not just in cinematography. Here
is Tarski speaking in 196858:
There is much affinity between the formalisms of sentential calculus and equational logic; as
a consequence various metalogical results established for sentential calculus can frequently be
carried over to equational logic with appropriate changes in formulations and proofs.
Here are Łukasiewicz and Tarski speaking in 193059:

In conclusion we should like to add that, as the simplest deductive discipline, the sentential
calculus is particularly suitable for metamathematical investigations. It is to be regarded as a
laboratory in which metamathematical methods can be discovered and metamathematical concepts
constructed which can then be carried over to more complicated mathematical systems.
57 R. McKenzie, Tarski’s Finite Basis Problem is Undecidable, International Journal of Algebra and
Computation, Vol. 6 (1996), pp. 49–104.
58 See [68], p. 288.
59 See [56m ], p. 59.
448 J. Zygmunt
And here, through the medium of Steven Givant, is Tarski speaking from beyond the
grave in 198760:
The above observations may seem somewhat paradoxical. The formalism T of two-valued
sentential logic is usually regarded as the simplest and most trivial logical formalism, with an
almost empty mathematical content. Nevertheless, the formalism L× r , so closely related to T in
its syntactical part, presents an adequate basis for the development of set theory, which is, in a
sense, the richest mathematical discipline; and even in its logical part L×r embodies an interesting
and far from trivial mathematical theory, namely the equational theory of relation algebras.
One conclusion emerges from our discussion: the connection between the formal structure of the
language and its intended semantical interpretation is much looser than we might be inclined to
believe.
It may be interesting to observe that the logic of T is well known to be decidable, while the logic
of L× r is undecidable . . . Thus, we have obtained an example of an undecidable subtheory of the
two-valued sentential logic (in fact a subtheory based upon a finite set of axiom schemata). Also
. . . this subtheory can be supplemented by means of finitely many axioms to form an essentially
undecidable theory ; this will be a theory in the same formalism (i.e., L× r ) as sentential logic,
but clearly not a subtheory of that logic.
5 Selected Works of Tarski
Listed below are all of Tarski’s works cited or referred to in the present paper. Also listed
are a few other works which, while not cited above, are central to his canon. The method
and style of citing Tarski’s works adhere as closely as possible to the conventions adopted
in Steven R. Givant’s Bibliography of Alfred Tarski, in The Journal of Symbolic Logic,
Vol. 51, No. 4 (Dec., 1986), pp. 913–941.
Givant’s bibliography was updated, corrected, and translated into Polish by Jan
Zygmunt in [95m], pp. 333–372. Both were divided into the same ten sections: Papers,
Abstracts, Monographs, Exercises and problems, Contributions to discussions, Reviews,
Publication as editor, Project reports, Letters, and Appendix. Both give full information
on later re-editions and translations of an item, as well as references to reviews in
Mathematical Reviews and The Journal of Symbolic Logic. The bibliography below
is divided into only four sections: (A) Monographs, (B) Papers, (C) Abstracts, and (D)
Publications as editor.
A more recently updated bibliography has been supplied by Andrew McFarland,
Joanna McFarland and James T. Smith in [14m ]. See that volume’s Ch. 16: Posthumous
Publications, Ch. 17: Biographical Studies, Ch. 18: Research surveys, and especially its
excellent Bibliography.
60 See [87m ], pp. 167–168.

5.1 Monographs
[33m ] Poj˛ecie prawdy w j˛ezykach nauk dedukcyjnych. Prace Towarzystwa

Naukowego Warszawskiego, Wydział III—Nauk Matematyczno-
fizycznych, nr. 34, Warszawa 1933, vii + 116 pp. + errata. (see [35b] for
German translation.)
[36m ] O logice matematycznej i metodzie dedukcyjnej. Bibljoteczka Matematy-
czna, vol. 3–5, Ksiażnica-Atlas,
˛ Lwów and Warszawa 1936, 167 pp.
(1) Einfürung in die mathematiche Logik und in die Methodologie der
Mathematik Julius Springer Verlag, Vienna, 1937, x + 166 pp. (German
translation of [36m ].)
[41m ] Introduction to Logic and to the Methodology of Deductive Sciences. Oxford
University Press, Oxford and New York 1941, xviii + 239 pp. [Fourth, enlarged
edition of [41m ], edited by Jan Tarski, Oxford Logic Guides, vol. 24, Oxford
University Press, New York – Oxford 1994, xxiv + 229 pp.]
[48m ] A Decision Method for Elementary Algebra and Geometry (prepared for
publication by J.C.C. McKinsey) U.S. Air Force Project RAND, R-109, the
RAND Corporation, Santa Monica, California 1948, iv + 60 pp.
(1) Second, revised edition of [48m ] (prepared for publication with the assis-
tance of J.C.C. McKinsey), University of California Press, Berkeley and Los
Angeles, California 1951, iii + 63 pp.
[49m ] Cardinal Algebras, With an Appendix: Cardinal Products of Isomorphism
Types (by B. Jónsson and A. Tarski), Oxford University Press, Oxford and
New York 1949, xii + 327 pp.
[53m ] Undecidable Theories (with A. Mostowski and R. M. Robinson), North-
Holland Publishing Company, Amsterdam 1953, xii + 98 pp.
[56m ] Logic, Semantics, Metamathematics. Papers from 1923 to 1938 (translated by
J.H. Woodger), Clarendon Press, Oxford 1956, xiv + 471 pp.
(1) Second, revised edition, with editor’s introduction and an analytic index (J.
Corcoran, editor), Hackett Publishing Company, Indianapolis, Indiana 1983,
xxx + 506 pp.
[56m a] Ordinal Algebras, with appendices: Some Additional Theorems on Ordinal Al-
gebras (by C.C. Chang) and A Unique Decomposition Theorem for Relational
Addition (by B. Jónsson), North-Holland Publishing Co., Amsterdam 1956,
133 pp.
[67m a] The Completeness of Elementary Algebra and Geometry, Institut Blaise
Pascal, Paris 1967, iv + 50 pp. (A reprint from page proofs of the work
scheduled to appear in 1940 in Actualités Scientifiques et Industrielles,
Hermann & Cie , Paris. See [48m ].)
450 J. Zygmunt
[71m ] Cylindric Algebras. Part I. With an Introductory Chapter: General Theory

of Algebras (with L. Henkin and J.D. Monk), North-Holland Publishing
Company, Amsterdam 1971, vi + 508 pp.
[83m ] Metamathematische Methoden in der Geometrie (with W. Schwabhäuser and
W. Szmielew), Springer-Verlag, Berlin 1983, viii + 482 pp.
[85m ] Cylindric Algebras. Part II (with L. Henkin and J.D. Monk), North-Holland,
Amsterdam 1985, vii + 302 pp.
[86m ] The Collected Works of Alfred Tarski, vol. 1, 1921–1934; vol. 2, 1935–
1944; vol. 3, 1945–1957; vol. 4, 1958–1979 (S.R. Givant and R.N. McKenzie,
editors), Birkhäuser, Basel, Boston and Stuttgart 1986, xii + 658 pp. (vol. 1);
xii + 699 pp. (vol. 2); xii + 682 pp. (vol. 3); xii + 757 pp. (vol. 4). [Volume
4 (pp. 739–757) contains a reprint of S. Givant, Bibliography of Alfred Tarski,
JSL, vol. 45 (1986), pp. 913–941]
[87m ] A Formalization of Set Theory without Variables. (with S.R. Givant), Col-
loquium Publications, vol. 41, American Mathematical Society, Providence,
Rhode Island 1987, xxi + 318 pp.
[95m ] Pisma logiczno-filozoficzne, vol. 1: Prawda. (Translated and annotated, with
an introduction by Jan Zygmunt), Biblioteka Współczesnych Filozofów,
Wydawnictwo Naukowe PWN, Warszawa 1995, xxiv + 390 pp. + errata.
[01m ] Pisma logiczno-filozoficzne, vol. 2: Metalogika. (Translated and annotated,
with an intoduction by Jan Zygmunt), Biblioteka Współczesnych Filozofów,
Wydawnictwo Naukowe PWN, Warszawa 2001, xiv + 516 pp.
[14m ] Alfred Tarski: Early Work in Poland—Geometry and Teaching. With a
Bibliographic Supplement. (A. McFarland, J. McFarland, James T. Smith,
editors), Birkhäuser: New York 2014, xxiii + 499 pp.
5.2 Papers
[21] Przyczynek do aksjomatyki zbioru dobrze uporzadkowanego

˛ <A Contribution to
the Axiomatics of Well-Ordered Sets>, Przeglad
˛ Filozoficzny, vol. 24 (1921), pp.
85–94.
[24a] Sur quelques théorèmes qui équivalent à l’axiome du choix, Fundamenta
Mathematicae, vol. 5 (1924), pp. 147–154.
[24b] O równoważności wielokatów
˛ <On the Equivalence of Polygons>, Przeglad
˛
Matematyczno-fizyczny, vol. 2 (1924), pp. 47–60.
[24c] Sur les ensembles finis. Fundamenta Mathematicae, vol. 6 (1924), pp. 45–95.
[24d] Sur la décomposition des ensembles de points en parties respectivement congru-
entes (with S. Banach), Fundamenta Mathematicae, vol. 6 (1924), pp. 244–277.
[25] Quelques théorèmes sur les alephs, Fundamenta Mathematicae, vol. 7 (1925),
pp. 1–14.
[26] Communication sur les recherches de la théorie des ensembles (with A. Lin-
denbaum), Sprawozdania z Posiedzeń Towarzystwa Naukowego Warszawskiego,
Wydział III Nauk Matematycznych i Przyrodniczych (= Comptes Rendus des
Séances de la Société des Sciences et des Lettres de Varsovie, Classe III), vol.
19 (1926), pp. 299–330.
[29a] Geschichtliche Entwicklung und gegenwärtiger Zustand der Gleichmächtigkeits-
theorie und der Kardinalzahlarithmetik, Ksi˛ega Pamiatkowa
˛ Pierwszego
Polskiego Zjazdu Matematyczniego (Supplement to Rocznik Polskiego
Towarzystwa Matematycznego), Kraków, 1929, pp. 48–54.
[29b] Sur les fonctions additives dans les classes abstraites et leur application au prob-
lème de la mesure. Sprawozdania z Posiedzeń Towarzystwa Naukowego Warsza-
wskiego, Wydział III Nauk Matematycznych i Przyrodniczych (= Comptes
Rendus des Séances de la Société des Sciences et des Lettres de Varsovie, Classe
III), vol. 22 (1929 – published 1930), pp. 114–117.
[30] Une contribution à la théorie de la mesure, Fundamenta Mathematicae, vol. 15
(1930), pp. 42–50.
[30a] Sur une propriété caractéristique des nombres inaccessibles (with W. Sierpiński),
Fundamenta Mathematicae, vol. 5 (1930), pp. 292–300.
[30b] Über Äquivalenz der Mengen in Bezug auf eine beliebige Klasse von Abbil-
dungen, Atti del Congresso Internazionale dei Matematici, Bologna, 3–10
settembre 1928, vol. 6, Nicola Zanichelli, Bologna 1930, pp. 243–252.
[30c] Über einige fundamentalen Begriffe der Metamathematik, Sprawozdania
z Posiedzeń Towarzystwa Naukowego Warszawskiego, Wydział III Nauk
Matematyczno-fizycznych (=Comptes Rendus des Séances de la Société des
Sciences et des Lettres de Varsovie, Classe III), vol. 23 (1930), pp. 22–29.
[30d] Untersuchungen über den Aussagenkalkül (with J. Łukasiewicz), Sprawozda-
nia z Posiedzeń Towarzystwa Naukowego Warszawskiego, Wydział III Nauk
Matematyczno-fizycznych (=Comptes Rendus des Séances de la Société des
Sciences et des Lettres de Varsovie, Classe III), vol. 23 (1930), pp. 30–50.
[30e] Fundamentale Begriffe der Methodologie der deduktiven Wissenschaften. I,
Monatshefte für Mathematik und Physik, vol. 37 (1930), pp. 361–404.
[30f] Sur les classes d’ensembles closes par rapport à certaines opérations élémen-
taires, Fundamenta Mathematicae, vol. 16 (1930), pp. 181–304.
[31] Sur les ensembles définissables de nombres réels. I, Fundamenta Mathematicae,
vol. 17 (1931), pp. 210–239.
[31a] Les opérations logiques et les ensembles projectifs (with C. Kuratowski), Fun-
damenta Mathematicae, vol. 17 (1931), pp. 240–248.
452 J. Zygmunt
[31b] O stopniu równoważności wielokatów

˛ <On the Degree of Equivalence of Poly-
gons>, Młody Matematyk, vol. 1 (Supplement to Parametr, vol. 2) (1931), pp.
37–44.
[34] Z badań metodologicznych nad definjowalnościa˛ terminów <Methodological
Investigations into the Definability of Terms>, Przeglad
˛ Filozoficzny, vol. 37
(1934), pp. 438–460.
[35a] Grundzüge des Systemenkalküls. Erster Teil, Fundamenta Mathematicae, vol.
25 (1935), pp. 503–526.
[35b] Der Wahrheitsbegriff in den formalisierten Sprachen, Studia Philosophica, vol.
1 (1935), pp. 261–405. (German translation of [33m ] by L. Blaustein, with an
appendix. Also printed separately in Lwów, 1935, 145 pp.)
[36] O ugruntowaniu naukowej semantyki <On Foundations for a Semantics of
Science>, Przeglqd Filozoficzny, vol. 39 (1936), pp. 50–57.
[36a] O poj˛eciu wynikania logicznego <On the Concept of Logical Consequence>,
Przeglqd Filozoficzny, vol. 39 (1936), pp. 58–68.
[36b] Über die Beschränktheit der Ausdrucksmittel deduktiver Theorien (with A.
Lindenbaum), Ergebnisse eines Mathematischen Kolloquiums, vol. 7 (1936),
pp. 15–22.
[36d] Grundzüge des Systemenkalküls. Zweiter Teil, Fundamenta Mathematicae, vol.
26 (1936), pp. 283–301.
[38a] Über unerreichbare Kardinalzahlen, Fundamenta Mathematicae, vol. 30
(1938), pp. 68–89.
[38c] Ein Überdeckungsästz für endliche Mengen, Fundamenta Mathematicae, vol.
30 (1938), pp. 156–163.
[38d] Eine äquivalente Formulierung des Auswahlaxioms, Fundamenta Mathemati-
cae, vol. 30 (1938), pp. 197–201.
[38g] Algebraische Fassung des Maßproblems, Fundamenta Mathematicae, vol. 31
(1938), pp. 47–66.
[38h] Der Aussagenkalkül und die Topologie, Fundamenta Mathematicae, vol. 31
(1938), pp. 103–134.
[39] Ideale in vollständigen Mengenkörpern. I, Fundamenta Mathematicae, vol. 32
(1939), pp. 45–63.
[39a] Boolesche Ringe mit geordneter Basis (with A. Mostowski), Fundamenta Math-
ematicae, vol. 32 (1939), pp. 69–86.
[39b] On well-ordered subsets of any set, Fundamenta Mathematicae, vol. 32 (1939),
pp. 176–183.
[41] On the calculus of relations, The Journal of Symbolic Logic, vol. 6 (1941), pp.
73–89.
[43] On families of mutually exclusive sets (with Paul Erdös), Annals of Mathematics,
vol. 44 (1943), pp. 315–329.
[44a] The semantic conception of truth and the foundations of semantics, Philosophy
and Phenomenological Research, vol. 4 (1944), pp. 341–375.
[48b] Axiomatic and algebraic aspects of two theorems on sums of cardinals, Funda-
menta Mathematicae, vol. 35 (1948), pp. 79–104.
[49] Cancellation laws in the arithmetic of cardinals, Fundamenta Mathematicae,
vol. 36 (1949), pp. 77–92.
[52b] Mutual interpretability of some essentially undecidable theories (with W.
Szmielew), Proceedings of the International Congress of Mathematicians.
Cambridge, Massachusetts, U.S.A., August 30 – September 6, 1950, vol. 1,
American Mathematical Society, Providence, Rhode Island 1952, p. 734.
[54] Theorems on the existence of successors of cardinals, and the axiom of choice,
Indagationes Mathematicae, vol. 16 (1954), pp. 26–32.
[54a] Contributions to the theory of models. I, Indagationes Mathematicae, vol. 16
(1954), pp. 572–581.
[54b] Contributions to the theory of models. II, Indagationes Mathematicae, vol. 16
(1954), pp. 582–588.
[55] Contributions to the theory of models. III, Indagationes Mathematicae, vol. 17
(1955), pp. 56–64.
[56b] Equilaterality as the only primitive notion of Euclidean geometry (with E.W.
Beth), Indagationes Mathematicae, vol. 18 (1956), pp. 462–467.
[56c] A general theorem concerning primitive notions of Euclidean geometry, Indaga-
tiones Mathematicae, vol. 18 (1956), pp. 468–474.
[57a] Arithmetical extensions of relational systems (with R.L. Vaught), Compositio
Mathematica, vol. 13 (1957), pp. 81–102.
[59] What is elementary geometry? in [59e ], pp. 16–29.
[61a] Cylindric algebras (with L. Henkin), Lattice theory, Proceedings of Symposia
in Pure Mathematics, vol. 2 (R.P. Dilworth, editor), American Mathematical
Society, Providence, Rhode Island, 1961, pp. 83–113.
[61b] On some problems involving inaccessible cardinals (with Paul Erdös), Essays on
the Foundations of Mathematics (Y. Bar-Hillel, E.I.J. Poznański, M.O. Rabin
and A. Robinson, editors), Magnus Press, Jerusalem 1961, pp. 50–82.
[62] Some problems and results relevant to the foundations of set theory, in [62e ], pp.
125–135.
[64] From accessible to inaccessible cardinals. Results holding for all accessible
cardinal numbers and the problem of their extension to inaccessible ones (with
H.J. Keisler), Fundamenta Mathematicae, vol. 53 (1964), pp. 225–308.
454 J. Zygmunt
[65] A simplified formalization of predicate logic with identity, Archiv für mathema-
tische Logik und Grundlagenforschung, vol. 7 (1965), pp. 61–79.
[65a] Metamathematical properties of some affine geometries (with L.W. Szczerba),
Logic, Methodology and Philosophy of Science. Proceedings of the 1964 Inter-
national Congress (Y. Bar-Hillel, editor), North-Holland Publishing Company,
Amsterdam 1965, pp. 166–178.
[68] Equational logic and equational theories of algebras, Contributions to Math-
ematical Logic. Proceedings of the Logic Colloquium, Hannover 1966 (H.A.
Schmidt, K. Schütte and H.J. Thiele, editors), North-Holland Publishing Com-
pany, Amsterdam 1968, pp. 275–288.
[78] The elementary theory of well-ordering—A metamathematical study (with J.
Doner and A. Mostowski), Logic Colloquium ’77 (A. MacIntyre, L. Pacholski
and J. Paris, editors), North-Holland Publishing Company, Amsterdam 1978, pp.
1–54.
[79] Metamathematical discussion of some affine geometries (with L.W. Szczerba),
Fundamenta Mathematicae, vol. 104 (1979), pp. 155–192.
[86] What are logical notions? (S. Givant, editor), History and Philosophy of Logic,
vol. 7 (1986), pp. 143–154.
[99] Tarski’s system of geometry (with S. Givant), Bulletin of Symbolic Logic, vol. 5
(1999), pp. 175–214.
5.3 Abstracts
[49a c] Arithmetical classes and types of Boolean algebras, Bulletin of the American
Mathematical Society, vol. 55 (1949), pp. 64, 1192.
[54a f] Prime ideal theorems for Boolean algebras and the axiom of choice, Bulletin
of the American Mathematical Society, vol. 60 (1954), pp. 390–391.
[54a g] Prime ideal theorems for set algebras and ordering principles, Bulletin of the
American Mathematical Society, vol. 60 (1954), p. 391.
[54a h] Prime ideal theorems for set algebras and the axiom of choice, Bulletin of the
[64a b] The comparability of cardinals and the axiom of choice, Bulletin of the
[65a ] On the existence of large sets of Dedekind cardinals, Notices of the American
Mathematical Society, vol. 12 (1965), p. 719.
5.4 Publications as Editor
[59e ] The Axiomatic Method, with Special Reference to Geometry and Physics (edited
with L. Henkin and P. Suppes), North-Holland Publishing Company, Amsterdam
1959, xi + 488 pp.
[62e ] Logic, Methodology and Philosophy of Science. Proceedings of the 1960
International Congress (edited with E. Nagel and P. Suppes), Stanford University
Press, Stanford, California 1962, ix + 661 pp.
[65e ] The Theory of Models. Proceedings of the 1963 International Symposium at
Berkeley (edited with J.W. Addison and L. Henkin), North-Holland Publishing
Company, Amsterdam 1965, xv + 494 pp.
J. Zygmunt ()
University of Wrocław, Wrocław, Polska
Stanisław Jaśkowski: Life and Work
Andrzej Indrzejczak
Abstract In this brief note we would like to outline the main events of life and the main
achievements of Stanisław Jaśkowski (1906–1965) one of the important Polish logician
and mathematician of the first half of twentieth century.
Keywords Jaśkowski · Natural deduction · Nonclassical logics · Decidability
Mathematics Subject Classification (2000) Primary 03F03; Secondary 01A60, 01A70
1 Life
Stanisław Jaśkowski was born on 22.04.1906 in Warsaw as a son of landowner Feliks

Jaśkowski and Kazimiera Dzierzbicka. His family represented high standards of hu-
manistic culture. In particular, his grandfather—Jan Nepomucen Jaśkowski—was a poet
and a writer; his father was a musician. It was not surprising that his parents expected
him to study on humanistic faculty. Despite family’s expectations in 1924 Jaśkowski
started studies in mathematic faculty of Warsaw University. As a result he became one
of the representatives of Warsaw School of Logic. Among his teachers were Leśniewski,
Tarski and Łukasiewicz, who had the strongest influence on his early development. In
particular, Łukasiewicz had a direct impact on his first, and one of the most important
logical contribution—the invention of natural deduction. It was Łukasiewicz who posed
on his seminar a problem of providing a fully formal account of methods of proof
applied by mathematicians in their practice. The first results of Jaśkowski concerning this
problem were presented in 1926 on the seminar and then announced on the First Polish
Mathematical Conference in Lwów in 1927 in: Ksiȩga Pamia̧tkowa I Polskiego Zjazdu
Matematycznego, Uniwersytet Jagielloński, Kraków 1929 [8].
Because of the serious health problems with lungs Jaśkowski had a lengthy break in
his education and scientific career. In 1929–1930 he was cured in Davos in Switzerland.
After recovery he participated in the Second Polish Mathematical Conference in Vilnius
in 1931 where he presented his first axiomatization of the geometry of solids based on
the primitive notion of semispace. This work shows for the early origins of his second
passion—foundations of mathematics.

458 A. Indrzejczak
In 1932 Jaśkowski gained his doctor’s degree under the supervision of Łukasiewicz.
The thesis [9], devoted to the presentation of the first system of natural deduction,
was printed in 1934 as the first volume of “Studia Logica”, the new journal edited by
Łukasiewicz. Strongly delayed publication of his first, and one of the most important,
discoveries was very unfortunate since in the same year Gerhard Gentzen published the
first part of his own version of natural deduction [7] and Jaśkowski lost the priority.
Moreover, in contrast to Jaśkowski’s paper, Gentzen’s work was soon widely known to
logical community and made Jaśkowski’s contribution relatively unknown.
In 1935 Jaśkowski took part in the International Congress of Scientific Philosophy in
Paris where he presented important results concerning adequate matrix characterization
for intuitionistic logic. This event was very important for his scientific development but
1937 was particularly important for his private life. Jaśkowski married Aniela Holewińska
(1905–1976) a student of mathematics in Warsaw University. In 1939 their daughter Anna
was born.
Till the September 1939 Jaśkowski continued scientific work, mainly on modal
functions and logical systems based on the notion of dependent variable. This last
contribution was very often applied in his later publications as a kind of methodological
basis for studies on paraconsistent and causal logics. The Second World War interrupted
his work on habilitation. Due to his health problems he was not admitted to the regular
army. Still he served as a volunteer in the defence of Warsaw; he gave his car to the
disposal of the 151st Column of Heavy Trucks. During the Nazi occupation of Poland, he
was living in his estate in Wolka near Rawa Mazowiecka and in Warsaw where he worked
as a bookkeeper. He was also arrested for a few weeks in 1942. Almost all of his scientific
manuscripts were lost during the Warsaw Uprising in 1944.
After the War Jaśkowski came back to scientific work. He was working for half a year
as a lecturer in the newly founded Łódź University, then he moved to Toruń where he
lived and worked from 1.10.1945 until his death in 1965. Even when he was proposed
to move to Kraków to became the Chair of Logic Department in Jagiellonian University
after Professor Zawirski’s death in 1949, he decided to stay in Toruń. Over the last 20
years of life he played a very important role as a scientist, a teacher, and an organizer
in the development of the new Nicolas Copernicus University in Toruń. Concerning the
last issue we list some of the most important facts. In 1945 Jaśkowski organised the
Department of Mathematical Logic and was its first Chair. Until 1965 he was the director
of mathematical departments in Toruń University. In 1952–53 he organized the Faculty
of Mathematics, Physics and Chemistry and was its first dean in 1953–54. In 1956–59 he
was a deputy prorector for science and in 1959–62 the rector of Toruń University. One
may also mention that he was a co-founder and the first president of Toruń division of
the Polish Mathematical Society. Despite the numerous occupations in Toruń from 1950
he was also a member of the National Institute of Mathematics at the Polish Academy of
Sciences.
He was also very active as a teacher. In the early years of its existence Toruń University
had serious problems with completing qualified staff sufficient for providing all necessary
courses. As a result, in addition to teaching mathematical logic Jaśkowski was forced to
provide also courses in analysis, set theory, geometry, probability theory. In the last years
of his life he was also strongly engaged in organization of computer laboratory; his last
seminar was devoted to the theory of automated deduction.
Stanisław Jaśkowski: Life and Work 459
In the meantime Jaśkowski finished his habilitation concerned with a new definition of
real numbers under the supervision of Zygmunt Zawirski. The habilitation colloquium
was conducted on 1.10.1945 and confirmed on 7.04.1946. He obtained the title of
associate professor in July 1946 and in 1957 he was nominated a full professor.
One should also mention his efforts in preparation of a new modern programme of
teaching mathematics in secondary school. The new syllabus was under a great influence
of Jaśkowski’s ideas of how to teach mathematics and was finally introduced in 1960s.
Despite all these organizational and educational duties he was still active on the field of
scientific research in logic and mathematics. Some of the most important achievements
will be sketched in the next two sections.
Unexpectedly he fell ill with infectious jaundice in 1962 and this caused a lot of
health complications. In consequence of postjaundice complications Jaśkowski died on
16.11.1965.
2 Works
The scientific activity of Stanisław Jaśkowski may be roughly divided into two over-
lapping fields: logic and mathematics. In what follows we briefly characterize his most
important achievements. A deeper presentation of his scientific ideas may be found in
[32, 34] and [33].
2.1 Logic
The invention of natural deduction systems, presented in [9], is the first of his logical
achievements and it is often claimed as the most important contribution of Jaśkowski to
logic. We do not attempt to describe his approach here since this is the subject of the
special paper presented in this volume.
Jaśkowski was particularly active on the field of investigation on nonclassical logics.
Not only he provided some results for already known logics like intuitionistic logic
or modal logics. He was also the inventor of many new and important systems, in
particular:
1. In his study on natural deduction [9] he constructed the first system of inclusive
logic. This is a kind of first-order logic which admits models with empty domains
in the semantics. Syntactically it is weaker than classical logic since some theses
(valid for nonempty domains only) are excluded. Logics which are inclusive and
additionally free, in the sense of having terms which do not denote existing things,
are called universally free (see Bencivenga [1]). Such systems are very often treated
as philosophically more neutral basis than classical logic and commonly applied for
constructing intuitively sound modal first-order logics (see e.g. Garson [6]). It should
be emphasized that the first recognized systems of inclusive and free logics were
proposed much later in 1950s by Mostowski, Leblanc, Hintikka, to mention just a few
scholars. The fact that Jaśkowski constructed the first inclusive logic (and implicitly
460 A. Indrzejczak
also the first universally free logic—see Bencivenga [2]) has gone unnoticed and,
surprisingly, is still not very well known even for authors writing on Jaśkowski’s
achievements. We characterize briefly his approach to first order logic in the paper
on natural deduction in this volume.
2. In the field of studies on intuitionistic propositional logic Jaśkowski provided not only
a natural deduction system but also an adequate matrix characterization. In his famous
result Gödel has shown that there is no adequate matrix for intuitionistic logic with
finitely many values. Jaśkowski’s study from 1930s shed a new light on this result by
providing a recipe for construction of adequate matrix characterization of intuitionistic
logic. The construction consists of the infinite sequence of finite matrices. It was
presented on the International Congress of Scientific Philosophy in Paris and published
in [10].
3. In [13] Jaśkowski provided a philosophical justification and a formal construction of
discursive logic which was the first system of paraconsistent logic. In such logical
systems the presence of contradictory statements does not lead to trivialization of
the system, in contrast to classical logic where, by Duns’ law, all propositions follow
logically from contradiction. In his system D2, material implication is replaced with
discursive implication which may be read: if the antecedent is possibly true, then the
succedent is true (with possibility understood as in modal logic S5). This work was
then extended in [21] where the notion of discursive conjunction was introduced.
Jaśkowski developed discursive logic mainly as a tool for analysing the situation of
contradictory views represented in discussion. Soon it appeared that many other serious
motivations for developing such systems may be provided and it gave the impact for
development of several paraconsistent logics.
4. Jaśkowski provided also the basis for the development of causal logic by means of
dependent sentential variables, i.e. variables representing propositions whose truth
depends on some arguments. The theory of dependent variables was first developed
in [14] and applied to modal functions. In [24, 25] a theory of causal functions is
introduced where three notions of causal implication are defined: factorial, efficient
and definitive. This work of Jaśkowski is not widely known but the study on causal
logic was continued by August Pieczkowski and Max Urchs.
His work on logic includes also some papers on classical logic and their fragments as
well as on traditional logic. Jaśkowski [15] is devoted to axiomatization of classical logic,
definition of ternary connectives and reversible substitution. References [30, 31] provide
a decision procedure for some fragmentary propositional calculi. Finally [23] contains
a new interpretation of Aristotelian syllogistics, where non-Aristotelian terms (empty
and universal names) are admitted. In this study he proposed a way of interpretation of
categorical statements in the first order-logic which keeps all classical laws as valid even
when non-Aristotelian terms are substituted for variables.
2.2 Mathematics
As a mathematician Jaśkowski was mainly interested in investigations on the foundations

of mathematics. In fact, the border between his logical and mathematical works is very
light since he always insisted on using logical tools in mathematics. In particular, he was
working on the notion of number, foundations of geometry and decidability problems
where he obtained both positive and negative results. Namely, he proved in [22] a
decidability of the elementary theory of Boolean rings and generalised the method to
elementary additive Boolean algebra. Above we also mentioned his decidability result for
parts of propositional logic [30, 31]. As for negative results, in [17] he proved that some
interesting classes of formulas of theory of groups and topology are undecidable. He has
shown that the theory of free groupoids is undecidable [28]. Also the abovementioned
paper [22] contains a theorem of the undecidability of some class of equalities of
Boolean algebras. Interesting results on the undecidability of some existential problems
in certain system of differential equations are provided in [27]. All these studies show that
Jaśkowski was able to apply logical tools to sophisticated mathematical problems of great
importance.
The investigation on the notion of number was the basis for Jaśkowski’s habilitation
and was summarised in [16]. He has shown that integers and real numbers may be defined
in terms of some operations on the classes of sets.
Jaśkowski was also engaged in developing the geometry of solids which avoids ideal
objects like points and straight lines. He was strongly convinced that such an approach
is better for applications in quantum physics. In [19, 20] Jaśkowski modified Tarski’s
approach to geometry of solids and introduced the axiomatics based on the notion of
“semispace” as primitive term in [12, 18]. He was preparing a book on the foundations of
geometries but his premature death interrupted this work.
Last but not least, Jaśkowski was very active in educational sphere. He wrote two
popular books on the geometry of ornament [26, 29] and prepared a textbook which may
be seen as the first course in logic based on the application of natural deduction [11].
The latter will be treated in detail in the next paper in this volume. As for his books on
geometry it must be emphasized that they show a great educational talent of Jaśkowski.
In a popular way, not demanding any mathematical knowledge from the readers, he
presented many interesting issues from geometry illustrated with examples taken from
science, history and art. He also shows how to apply an abstract mathematical theory of
groups to a concrete problem like classification of ornaments.
He actively participated in the work on modernization of the program of teaching
mathematics in secondary schools. The problem was widely discussed in several commis-
sions organized by the Ministry of Education, the Institute of Pedagogics and the Polish
Mathematical Society. Jaśkowski was very critical with respect to the present state of the
art. In his opinion the content of the present programme corresponds to the level which
was obtained in seventeenth century mathematics before the achievements of Newton
and Leibniz. The outcome of his activity in this field is a series of papers and notes where
Jaśkowski insisted that teaching mathematics should be based on the application of logical
tools and closer to the application in science and technique. His postulates concerning the
changes in teaching mathematics were used in the construction of the new syllabus which
was put to work in 1960s.
462 A. Indrzejczak
3 Influence
Although Jaśkowski passed away so early and his list of publications is not particularly
long1 he had significant achievements in many fields: logic, mathematics, education and
organization of science. The problem of his influence on the development of logic and
mathematics is a complicated matter. In Poland Jaśkowski had a number of students (e.g.
Lech Dubikajtis, Jerzy Kotas, August Pieczkowski) who continued his work and made at
least some of his achievements better known. However, it should be underlined that his
contribution to logic is very often not recognized. In addition to his premature death there
were some other unfortunate circumstances connected with his scientific achievements.
Almost all of his ideas and results become known to logical community in the World much
later after they were introduced, sometimes long after his death. We briefly comment on
three selected issues.
His invention of natural deduction was not only significantly delayed in print but
even after 1934 it was not as widely known as Gentzen’s approach. In fact nowadays
many logicians writing on, or presenting some systems of natural deduction, even do not
mention Jaśkowski. But in fact, his style of presenting proofs, as a series of subordinate
derivations with several bookkeeping devices applied to separate the scope of assumptions
is widely applied. We discuss these matters in the second paper, here we only mention
that Jaśkowski introduced a graphical method of separating subproofs and a technique
of prefixes. The former was then popularised by Fitch [4] and is nowadays commonly
called Fitch-style natural deduction.2 The latter may be treated as the first application of
labels which is now widely used in logic to construct several kinds of proof systems.3
Thus Jaśkowski, even if not explicitly mentioned, is the true father of numerous natural
deduction systems presented in textbooks and applied in practice. Gentzen’s approach,
based on the application of trees as proof-structures, is rather connected with theoretical
investigations in proof theory.
Similarly, his system of inclusive logic, also presented in [9], was recognized as the
first system of this type much later after the advent of studies on free logics.4
The work on paraconsistent logics was also developed in early stages without any
knowledge of Jaśkowski’s work; his contribution was recognized much later. Still it seems
that his invention of discursive logic is better known than any other of his achievements.
In fact, both his papers on discursive logic were written in Polish and published in
the journal of local character, so there was no chance to make them known to logical
community immediately after publication. However, soon after the works of Jaśkowski,
a serious investigations on paraconsistent logics was undertaken by a group of logicians
from Latin America, in particular by Newton da Costa in Brazil and Florentio Asenjo
from Argentina. Later, in 1960s, a similar research started in United States and in
Australia (Michael Dunn, Rober Meyer, Richard Routley (Sylvan), Graham Priest and
others), in strong connection with the investigation on relevant implication. Nowadays,
1 The full list of Jaśkowski’s publications has 48 items and may be found in [3].
2 Infact Fitch mentioned Jaśkowski in the Preface to his textbook as a source of inspiration.
3 See e.g. Gabbay’s theory of labelled systems in [5].
4 See historical remarks in [1].
paraconsistent logics form one of the most well known group of nonclassical logics
with wide spectrum of applications. Fortunately enough Jaśkowski work in this field was
continued by Jerzy Kotas who cooperated with Newton da Costa. It seems that this joint
work made Jaśkowski’s achievements widely known to the community of researchers in
paraconsistent logic.
In general, the problem with the lack of propagation of Jaśkowski’s achievements was
mainly connected with the fact that his papers were published very often in the journal
of local character, sometimes written in Polish. After the II World War Polish scholars
had in general difficulties with dissemination of their ideas in the World due to the new
geopolitical situation. It seems that Jaśkowski’s case was not exceptional. Summing up,
we can say that he had a strong influence on some domains of research in logic but usually
it was not a direct impact. In most cases we may rather say about later recognition of his
ideas.
References
1. Bencivenga, E.: Free logics. In: Gabbay, D., Guenthner, F. (eds.) Handbook of Philosophical Logic,
vol. III, pp. 373–426. Reidel Publishing Company, Dordrecht (1986)
2. Bencivenga, E.: Jaśkowski’s universally free logic. Stud. Logica 102(6), 1095–1102 (2014)
3. Dubikajtis, L.: The life and works of Stanisław Jaśkowski. Stud. Logica 34(2), 109–116 (1975)
4. Fitch, F.: Symbolic Logic. Ronald Press Co, New York (1952)
5. Gabbay, D.: LDS – Labelled Deductive Systems. Clarendon Press, Oxford (1996)
6. Garson, J.W.: Modal Logic for Philosophers. Cambridge University Press, Cambridge (2006)
7. Gentzen, G.: Untersuchungen über das Logische Schliessen. Mathematische Zeitschrift 39, 176–210,
405–431 (1934)
8. Jaśkowski, S.: Teoria dedukcji oparta na dyrektywach zał ożeniowych [Theory of deduction based on
suppositional directives]. In: Ksiȩga Pamia̧tkowa I Polskiego Zjazdu Matematycznego. Uniwersytet
Jagielloński, Kraków (1929)
9. Jaśkowski, S.: On the rules of suppositions in formal logic. Stud. Logica 1, 5–32 (1934) (Reprinted
in: Storrs McCall (ed.) Polish Logic 1920–1939, pp. 232–258. Oxford (1967))
10. Jaśkowski, S.: Recherches sur le systéme de la logique intuitioniste. In: Actes du Congrés Interna-
tional de Philosophie Scientifique, pp. 58–61. Sorbonne, Paris (1936) (English translation in: Studia
Logica 34(2), pp. 117–120 (1975))
11. Jaśkowski, S.: Elementy logiki matematycznej i metodologii nauk ścisł ych [Elements of mathemati-
cal logic and methodology of formal sciences]. Lecture notes, Toruń (1947)
12. Jaśkowski, S.: O aksjomatyce geometrii brył [On the axiomatization of the geometry of solids]. The
Report of the VI-th Congress of the Polish Mathematicians (1948)
13. Jaśkowski, S.: Rachunek zdań dla systemów dedukcyjnych sprzecznych [A propositional calculus for
inconsistent deductive systems]. Studia Societatis Scientiarum Torunensis, Sec. A 1, 57–77 (1948)
(Reprinted in: Studia Logica 24, pp. 143–157 (1969))
14. Jaśkowski, S.: Sur les variables propositionnelles dépendantes. Studia Societatis Scientiarum
Torunensis, Sec. A 24, 17–22 (1948)
15. Jaśkowski, S.: Trois contributions au calcul des propositions bivalent. Studia Societatis Scientiarum
Torunensis, Sec. A 1, 3–15 (1948) (English translation in: Studia Logica 34(2), pp. 121–132 (1975))
16. Jaśkowski, S.: Sur certains groupes formés de classes d’ensambles et leur applications aux définitions
des nombres. Studia Societatis Scientiarum Torunensis, Sec. A 1, 23–35 (1948) (English translation
in: Studia Logica 34(2), pp. 132–144 (1975))
17. Jaśkowski, S.: Sur le probléme de decision de la topologie et de la theorie des groups. Colloq. Math.
1, 176–178 (1948)
464 A. Indrzejczak
18. Jaśkowski, S.: Sur certains axiomes de la géométrie élémentaire. Ann. Polish Math. Soc. 21, 349–350
(1948)
19. Jaśkowski, S.: Une modification des définitions fondamentales de la géométrie des corps de A. Tarski.
Ann. Polish Math. Soc. 21, 298–301 (1948)
20. Jaśkowski, S.: Geometria brył [The geometry of solids]. Matematyka 1, 1–7 (1949)
21. Jaśkowski, S.: O koniunkcji dyskusyjnej w rachunku zdań dla systemów dedukcyjnych sprzecznych
[On the discussive conjunction in the propositional calculus for inconsistent deductive systems].
Studia Societatis Scientiarum Torunensis 1, 171–172 (1949) (Reprinted in: Logic and Philosophy
7, pp. 57–59 (1999))
22. Jaśkowski, S.: Z badań nad rozstrzygalnościa̧ rozszerzonej algebry Boole’a [Investigations on
decidability of extended Boolean algebra]. Casopis pro pestovani matematiky a fizyky, roc. 74, 136–
137 (1949)
23. Jaśkowski, S.: O interpretacji zdań kategorycznych Arystotelesa w rachunku predykatów [On inter-
pretations of Aristotle’s categorical sentences in predicate calculus]. Studia Societatis Scientiarum
Torunensis, Sec. A 2, 77–90 (1950)
24. Jaśkowski, S.: Interpretacje funkcji przyczynowych w rachunku zmiennej zdaniowej zależnej [In-
terpretations of causal functions in the calculus of dependent propositional variable]. Sprawozdania
Toruńskiego Towarzystwa Naukowego 1–4, 123–124 (1950)
25. Jaśkowski, S.: On the modal and causal functions in symbolic logic. Stud. Philos. 4, 71–92 (1951)
26. Jaśkowski, S.: O symetrii w zdobnictwie i przyrodzie [On symmetry in decorative art and in nature].
Warszawa (1952)
27. Jaśkowski, S.: Example of class of systems of ordinary differential equations having no decision
method for existence problems. Bulletin de l’Academie Polonaise des Sciences 2(4), 153–155 (1954)
28. Jaśkowski, S.: Undecidability of first order sentences in the theory of free groupoids. Fundam. Math.
43, 36–45 (1956)
29. Jaśkowski, S.: Matematyka ornamentu [The mathematics of ornament]. Warszawa (1957)
30. Jaśkowski, S.: Über Tautologien, in welchen keine Variable mehr als zweimal vorkommt. Zeitschrift
für mathematische Logik und Grundlagen der Mathematik, Bd. 9, 219–228 (1963)
31. Jaśkowski, S.: On formulas in which no individual variable occurs more than twice. J. Symb. Log.
31(1), 1–6 (1966)
32. Kotas, J., Pieczkowski, A.: Scientific works of Stanisław Jaśkowski. Stud. Log. 21, 7–16 (1967)
33. Pietka, D.: Stanisław Jaśkowski’s logical investigations. Organon 37(40), 39–69 (2008)
34. Woleński J.: Logic and Philosophy in the Lvov-Warsaw School. Dordrecht, Kluwer (1989)
A. Indrzejczak ()
Department of Logic, University of Łódź, Łódź, Poland
Stanisław Jaśkowski and Natural Deduction
Systems
Andrzej Indrzejczak
Abstract In 1934 Stanisław Jaśkowski published his groundbreaking work on natural

deduction. At the same year Gerhard Gentzen also published a work on the same topic.
We aim at presenting (three versions) of Jaśkowski’s system and provide a comparison
with Gentzen’s approach. We also try to outline the influence of Jaśkowski’s approach on
the later development of natural deduction systems.
Keywords Jaśkowski · Gentzen · Natural deduction · Inclusive logic
Mathematics Subject Classification (2000) Primary 03F03; Secondary 01A60, 01A70
1 Introduction
Stanisław Jaśkowski is one of the founders of modern systems of natural deduction (ND).
He presented his system in 1934 as the first volume of the series STUDIA LOGICA
initiated by J. Łukasiewicz.1 In fact, ND systems were constructed independently by
two logicians; the second was Gerhard Gentzen. It is a matter of coincidence that
at the same year Gentzen started to publish his Habilitationschrift which appeared in
two parts in Mathematische Zeitschrift [13]. The name ND is due to Gentzen—he has
called his system Natürliche Kalkül. Jaśkowski used the term “composite system” in
contrast to Hilbert axiomatic “simple system”; below we explain the sense of this term
in Sect. 3.1.2. Despite the differences in both approaches ND systems were conceived
as formal realizations of traditional means of proving theorems in mathematics, science
and ordinary discourse. Since then, several variants of ND were devised and presented in
hundreds of logic textbooks, giving an evidence that ND systems are commonly accepted
as the most efficient way of teaching logic. Still, simplifying a bit, but truly indeed, we
may say that everything so far constructed in the field of ND and the related systems is
based, more or less directly, on the ideas developed by these two researchers.
1 This series after the War was revitalised as the well known logical journal.

466 A. Indrzejczak
However, Jaśkowski’s role in this enterprise and influence of his results on later
developments in the field are not very well recognized. It is a common practice that
authors presenting natural deduction systems of one sort or another are mentioning only
Gentzen as the inventor of this kind of proof systems. It is not at all surprising that
Gentzen’s work is widely known since it provides much better developed body of research
of a great generality. He did not present just an ND system but also a sequent calculus
and provided important theoretical results. His famous normalization theorem for ND
was indirectly proved on the basis of equally famous cut elimination theorem holding
for sequent calculus. As a byproduct of cut elimination he obtained consistency and
decidability results for propositional classical and intuitionistic logic and a version of
Herbrand theorem. These profound results of Gentzen are the cornerstones of modern
proof theory and this is the main reason that rather modestly looking, relatively short
paper of Jaśkowski seems to have gone unnoticed.
But is that true that Jaśkowski’s work on ND was really unnoticed and has no impact on
the development of research on ND? It is the aim of this paper to show that despite of the
absence of Jaśkowski’s name in many textbooks presenting ND systems, his solutions had
real and strong influence on further research. Roughly, we can say that whereas Gentzen
had an enormous impact on the development of theoretical investigations on proof
theory, Jaśkowski greatly influenced the practical side of the story. Most of ND systems
popularised by hundreds of logic textbooks are directly based on Jaśkowski’s proposal,
usually without the authors’ awareness of the roots of their solutions. We will point out
also some other contributions of Jaśkowski’s paper, like e.g. introduction of inclusion
logic which also passed unnoticed at the time of publication, and were rediscovered
by other scholars in later years. Surprisingly the same situation was connected with
his invention of the first systems of paraconsistent logics (discursive logics) in 1948.
Systematic research on this kind of non-classical logics started in 1960s without the
knowledge of Jaśkowski’s results.
We start with a general remarks on ND,2 then we provide a detailed description of
Jaśkowski’s work on ND and compare his solutions with Gentzen’s approach. Finally we
describe some types of ND which were based on Jaśkowski’s solution.
2 Natural Deduction in General
It is a lot of proof systems in use which are called ND systems and sometimes they differ
greatly at first sight. Accordingly it is hard to provide a precise definition of ND-systems
that would be generally accepted. Some authors tend to use this term in a broad sense,
so that it covers also Gentzen’s sequent calculus and various forms of tableau systems.
In fact, all these systems are in close relationship to each other but we prefer to use this
notion in a narrower sense. There are at least three reasons to make such a choice. First, for
Gentzen his sequent calculus was meant as a technical tool to prove some metatheorems
on natural deduction, not as a kind of ND. Secondly, both for him and for Jaśkowski, ND
was supposed to reconstruct, in a formally proper way, traditional ways of reasoning. It
2 This part is an excerpt from more detailed considerations contained in my [16] and [17].
Stanisław Jaśkowski and Natural Deduction Systems 467
may be a matter of discussion if existing ND systems realize this task in a satisfying way,
but certainly systems like tableaux or resolution are worse in this respect. Finally, taking
a term ND in a wide sense would be a classifying operation of doubtful usefulness.
So what is ND in a narrow sense? In Pelletier [24] and Pelletier and Hazen [25] one
may find a discussion of several definitions of ND and their inadequacy. Instead of precise
definition we provide three criteria which should be satisfied for genuine ND systems:
• ND system allows for entering assumptions into a proof and also for eliminating them.
• ND system consists of rules; there are no (or, at least, very limited set of) axioms.
• ND system admits a lot of freedom in proof construction and possibility of applying
several proof search strategies.
Some authors (c.f. [3] or [24, 25]) formulated additional conditions characterising ND,
but in our opinion these three are essential. According to this loose characteristics ND
system should be open for reasoning from arbitrary assumptions and for the application
of different proof constructions. The user is free in constructing direct, indirect, or
conditional proofs. He may build more complex formulae or decompose them, as
respective introduction/elimination rules allow. Instead of using axioms, or already proved
theses, he is rather encouraged to introduce assumptions and derive consequences from
them. The presence of axioms is permitted but not essential since their role is taken over
by the set of primitive rules. This flexibility of proof construction in ND is in striking
contrast to other types of deductive systems usually based on one form of proof.
The above characteristics of ND is still very broad and it allows a lot of freedom in the
selection of primitive rules, design of a proof or graphical devices used as bookkeeping
devices for indicating the scope of an assumption. These are also important features
which make a variety of ND systems presented in textbooks apparently different but
are of no real importance in delimiting this class of proof systems. In particular, both
Jaśkowski’s and Gentzen’s approaches were similar in the three points we mentioned
although different in many other respects, and this should be treated as a decisive
argument for such a characterization of ND.
Additionally, in the first ND systems proposed by Jaśkowski and Gentzen one can
identify two types of rules which we will call rules of inference and proof construction
rules. The former have the form / ϕ; we read them as follows: if we have all formulae
from (premises) present in the derivation we can add ϕ (conclusion) to this derivation.
By derivation we mean an attempted, i.e. unfinished proof. Proof construction rules are
more complex. In general they allow us to build a proof, enter additional assumptions
opening nested subderivations, and show under which conditions we may discharge these
assumptions and close the respective subderivations. Typical proof construction rules
are meant to formalize the old and well known proof techniques like conditional proof,
indirect proof, proof by cases, etc.
Although much can be said about the prehistory of ND, 1934 is commonly accepted
as the first year in the official history of such systems. In this year two groundbreaking
papers of Jaśkowski [19] and Gentzen [13] were published. It should be of no surprise that
the two logicians with no knowledge of each other’s work, independently proposed quite
different solutions to the same problem. The need for deduction systems of this sort was
in the air. Hilbert’s proof theory already offered high standards of precise formalization
in terms of axiom systems but the process of actual deduction in Hilbert calculi is usually
complicated and needs a lot of invention. Moreover, axiomatic proofs are lengthy, difficult
468 A. Indrzejczak
to decipher, and far from informal proofs provided by mathematicians. In consequence,

axiom systems, although theoretically satisfying, were considered by many researchers as
practically inadequate and artificial. Hence, two goals were involved in this enterprise: a
theoretical justification of traditional proof methods on the ground of modern logic, and a
formally correct and practically useful system of deduction.
A closer look at the circumstances of Jaśkowski’s discovery shows that he may be
rightly treated as the first inventor of ND. He was influenced by Łukasiewicz, who posed
on his Warsaw seminar in 1926 a problem: how to describe, in a formally proper way,
proof methods applied in practice by mathematicians (cf. Woleński [32]). In response to
Łukasiewicz problem, Jaśkowski, as a young student3 presented a first solution in the
same year to his tutor. Officially, his first results on ND were announced in 1927, at the
First Polish Mathematical Congress in Lvov, mentioned in [18]. Unfortunately, Jaśkowski
had a lengthy break in his research due to serious health problems. After recovery in 1932
Jaśkowski gained his doctor’s degree under the supervision of Łukasiewicz on the basis
of his work on ND. The thesis was eventually published as [19].
3 Jaśkowski’s Research on ND
Usually two versions of ND are attributed to Jaśkowski, the first called by Pelletier [24]
a graphical method and the second a bookkeeping method. We are going to show that
it is reasonable to say that Jaśkowski provided three versions of ND, quite similar yet
different in a significant way. His first version of ND system (graphical) was not published
in 1920s, and we do not know exactly for what logics, in what languages, and by means
of what rules, it was conducted. The only thing we know is the format of proof applied by
Jaśkowski in the original version since he provided examples in the footnote to [19]. Yet
this feature is important enough to treat this proposal as different from the one officially
presented in [19]. The latter, called by Pelletier a bookkeeping method differs significantly
at least with respect to proof layout. We will call it the second ND (or the official) system
of Jaśkowski.
After the War, Jaśkowski published his lecture notes [20] on mathematical logic in
1947. His presentation of classical logic in the script is not axiomatic but based on
the application of ND. It seems that it is the first educational application of ND in the
World where adequate system of ND is consequently applied as a form of presentation
of classical logic in a textbook. His treatment of ND in [20] is different in some respects
from [19] so we feel justified in saying about the third version of Jaśkowski’s ND. In what
follows we describe in separate sections the second and the third version. Remarks on the
first version will be added to the presentation of the second one, because of the lack of
knowledge mentioned above.
3 In 1926 he was 20 years old.

3.1 Jaśkowski’s Official ND
Jaśkowski’s dissertation is not very long. However, on the 27 pages he provided ND

systems for the following logics:
1. positive propositional logic;
2. intuitionistic propositional logic in the version of Kolmogoroff [22];
3. classical propositional logic (CPL);
4. (classical) propositional logic with quantifiers;
5. (inclusive) first-order logic.
Jaśkowski is using a language with →, ¬ and ∀ as primitives, and applies so called Polish
notation (parentheses-free) due to Łukasiewicz. In what follows we will be using standard
notation for better readability. We do not repeat also the original formulation of rules since
it is strongly connected with Jaśkowski’s way of displaying proofs in the system and we
explain this issue below. Instead we apply some neutral (to the proof format) way of
description and additionally apply ⊥ (not used as primitive by Jaśkowski, but introduced
for illustration) as as a metalinguistic sign of inconsistency.
3.1.1 Rules
He started with CPL, then he just get rid with negation and a suitable rule for it,
corresponding to indirect proof technique. Next, intuitionistic logic is obtained by a slight
modification of this rule. So what are the rules for CPL? There are four such rules:
Rule I allows for introduction of an assumption prefixed with the letter ‘S’ (for
supposition) in any place of the proof, hence it is neither inference nor proof construction
rule. Rule II–IV formalize (in that order) Conditional Proof, Modus Ponens and (the
strong form of) Indirect Proof. So rules II and IV are proof construction rules and III
is the only rule of inference in the system. In proof-theoretic formulation (and without
specific Jaśkowski’s devices) the rules may be described in the following way:
Rule II If , ϕ # ψ, then # ϕ → ψ
Rule III ϕ, ϕ → ψ / ψ
Rule IV If , ¬ϕ # ⊥, then # ϕ
where denotes a set (possibly empty) of other active assumptions. As we mentioned

Jaśkowski proposed also modifications of his calculus leading to weaker (he call them
incomplete) propositional logics, namely, he has observed that the last rule may be
weakened:
Rule IVa If , ϕ # ⊥, then # ¬ϕ
This form yields ND formalization of Kolmogoroff’s version of intuitionistic logic,

whereas deletion of any rule for ¬ captures positive logic of Hilbert. It should be
underlined that the version of intuitionistic logic considered by Jaśkowski is weaker than
the well known Heyting’s formalization. In particular, the intuitionistic thesis ¬p →
(p → q) is not provable in his system (although p → (¬p → ¬q) is provable). Note that
in Gentzen’s system this weaker form of indirect proof is sufficient for obtaining Heyting’s
470 A. Indrzejczak
intuitionistic logic but Gentzen is using ⊥ as a primitive constant (¬ is definable) and

a rule of trivialization ⊥ / ϕ, so deduction of q from ¬p and p is not a problem.
Incidentally, Jaśkowski is also mentioning a theory obtained by addition of and ⊥;
we will describe it in connection with the proof format. Also, by the end of his paper he
considered proper rules for conjunction and proposed the obvious ones:
(∧I ) ϕ, ψ / ϕ ∧ ψ
(∧E) ϕ ∧ ψ / ϕ and ϕ ∧ ψ / ψ
Jaśkowski formulated also ND system for propositional logic with universal quantifier,
called by him the extended theory of deduction. In such a system we can define ⊥ as
∀p, p which is in fact shown by Jaśkowski. He defines in an obvious way the notions
of free (real) and bound (apparent) propositional variable and add to CPL two new rules
which may be formally displayed as follows:
Rule V ∀pϕ / ϕ[p/ψ]

Rule VI If # ϕ, then # ∀pϕ
In rule V ϕ[p/ψ] denotes the operation of proper substitution of ψ for p in ϕ, which

means that all occurrences of p which were bound in ∀pϕ are substituted by ψ and no
propositional variable in ψ is bound in ϕ[p/ψ]. Rule VI has a side condition that p
is not free in any active assumption in . Although we have formulated it as a proof
construction rule it may be also described as inference rule with side conditions since
there is no subtraction from the set of active assumptions.
Finally Jaśkowski developed ND for first order logic (calculus of functions) but with
explicit remark that it is weaker than classical version. Jaśkowski states that “whether
individuals exist or not, it is better to solve this problem through other theories. We
shall present therefore a system [. . . ] where all theses will be satisfied in the null field
of individuals” (p. 28). Thus he introduced the first system of inclusive logic, whereas the
first recognized system of this type was presented in 1950s.4
Jaśkowski realised that in such a system free variables are in fact not variables but
rather (nonlogical) constants whose existence we assume in the proof. Such variables are
called valid in the part of a proof where we declared their existence. He did not introduce
different kind of letters for denoting free variables (as Gentzen did) but instead he applied
a special technique of explicit signification that some variable is held constant for the sake
of proof. It is an additional rule which parallels the rule I. This rule VII allows to introduce
a term supposition T x for any variable not valid (so far) in this part of proof. Also rule
I must be restricted; we can introduce as assumptions only such formulae which do not
4 Cf. historical remarks

in Bencivenga [4]; in [5] he also pointed out that Jaśkowski’s system may be easily
changed into a system of free logic.
contain free variables not valid in the respective part of proof. The remaining rules are
variants of rules V and VI:
Rule Va T y, ∀xϕ / ϕ[x/y]

Rule VIa If , T x # ϕ, then # ∀pϕ
Note that ϕ[x/y] denotes the proper substitution of y for x in ϕ but only if y is valid in
respective part of a proof (T y was introduced earlier by rule VII), hence in contrast to
rule V this is a two-premise rule. Also VIa is a “real” proof construction rule since the
term-assumption T x ceased to be valid in the result of its application and it is deleted
from the set of active assumptions (a respective subproof is discharged).
3.1.2 Proof Format
In the case of ND it is very important how we define a proof since uncontrolled

introduction of assumptions without clear indication of their scope may lead to troubles
very often met when students are taught to use ND.
Let us analyse the following “proof” apparently performed with the help of
Jaśkowski’s rules (with added rule for disjunction introduction (∨I ) in line 3):
1 Sp ∨ r → q R.I
2 Sp R.I
3p∨r 2, ∨I
4q 1, 3, R.I I I
5p→q 2, 4, R.I I
6 (p → q) ∧ p 2, 5, ∧I
But (p → q) ∧ p does not follow logically from p ∨ r → q. We cannot apply (∧I )

correctly to formulae from lines 2 and 5 since assumption Sp is not active in this place.
The application of rule II in line 5 discharged the assumption 2 and this formula cannot
be further used in the proof. This problem is generally connected with the application of
such rules as II or IV (and VIa) which are not simply inference rules but proof construction
rules showing that if some proof is being constructed on the basis of some assumption,
then, in the effect, we obtain another proof in which this assumption is not in force. Thus
linear proof admitting additional assumptions and means for closing dependant parts of a
proof is in fact not a simple sequence of formulae but rather a richer structure containing
nested subderivations (subordinate proofs). To avoid scoping difficulties some devices
must be used for separating the parts of proof which are in the scope of discharged
assumption, hence not available.
All that we know about the first version of Jaśkowski’s ND is that he provided a clear
indication of the scope of every assumption introduced into a proof. His first original
solution to the problem consisted in making boxes for each assumption and dependent
part of a proof. Every introduction of an assumption was connected with starting a new
box, and this assumption was always put as the first formula in it (he did not applied
the prefix “S” for that). An application of any proof construction rule like II or IV was
472 A. Indrzejczak
connected with closing a current box, and inferred formula was immediately written down
as the next element of outer derivation. He also used an additional rule of repetition to
shift a formula from outer open box to the inner; transition in the other direction was of
course forbidden. Schematically, the application of both proof construction rules in his
propositional system5 looks like this:

ϕ ¬ϕ

.. ..
. .
ψ ⊥
ϕ→ψ ϕ
On the diagrams possibly empty ⊆ refers to formulae obtained by repetition.

In his official version Jaśkowski applied an apparatus of numeric prefixes, instead of
boxes. These are finite sequences of natural numbers separated with dots and written
before each formula in a proof (except of a thesis). Each time we enter an assumption we
extend prefix with additional number (and a prefix S in front of an assumption.). Each
application of rules II, IV (IVa) and VIa is connected with subtraction of the last number
in the prefix. Application of inference rules, like II or V is admitted only if prefixes of
their premises are initial parts, or are identical, to the prefix of the last formula in a proof
(prefix of a conclusion must be identical to it). This way Jaśkowski avoided introduction
of repetition as a rule.
Application of prefixes instead of boxes may seem as an editorial simplification but
in fact it is connected with some philosophical motivations taken from Leśniewski,
concerning dynamic nature of deductive system. Jaśkowski thought of prefixes as
indicators of domains in which formulae with this prefix are valid. Thus formulae with
empty prefixes were ordinary theses (valid in every domain), and those with nonempty
prefixes are theses relative to some domain in which some suppositions are postulated as
valid. Prefixes are then records of dependency of a formula on assumptions in the context
of a proof or on the existence of some objects named by term-suppositions. Domain-
validity is hereditary with respect to nested subdomains. Thus a formula e.g. 3.2.1.ϕ is
valid not only in the domain 3.2.1. but also in 3.2.1.1., 3.2.1.5.2 but not in 3.2. or 3.3.1.
Jaśkowski directly pointed out that “Every domain can be considered as a system having
its own axioms and constants, though not every domain gives a complete system, much
less an interesting one.” (p. 14). As an example of “interesting” system he considered the
one with added suppositions St (representing ) and S¬u (interpreted as ¬⊥). As the
first is introduced with prefix 7. and the second with prefix 7.1. we can consider every
formula valid in the domain prefixed with 7.1. as a thesis of a system with and ¬⊥
being additional axioms. Such a rationale behind using prefixes is also a reason for using
a name “composite system” for ND, as it is “composed of many systems”.6
5 Thetwo examples provided in the footnote in [19] show only propositional proofs.
6 By the way, an innovation introduced by Jaśkowski (i.e. prefixes) may be classified in a different way; we
may treat his second version as the first example of ND defined not on formulae but on labelled formulae.
Below we illustrate both versions with an example of a proof.
1 p R.I 1 1.Sp R.I

2 ¬¬¬p R.I 2 1.1.S¬¬¬p R.I
3 p (1, rep.) 3 1.1.1.S¬¬p R.I
4 ¬¬p R.I. 4 1.1.¬p 2, 3, R.I V
5 ¬¬¬p (2, rep.) 5 1.¬¬p 1, 4, R.I V
6 ¬p 4, 5, R.I V 6 p → ¬¬p 1, 5, R.I I
7 ¬¬p 3, 6, R.I V
8 p → ¬¬p 1, 7R.I I
Additionally we provide an example of proof of the first two theses in his system with
quantifiers:
cf 1 1.S∀xyAxy R.1
cf 2 1.1.T z R.V I I
cf 3 1.1.∀yAzy 1, R.V a
cf 4 1.1.Azz 3, R.V a
cf 5 1.∀zAzz 4, R.V I a
cf 6 ∀xyAxy → ∀zAzz 1, 5, R.I I
cf 7 1.1.1.T v R.V I 1
cf 8 1.1.1.Azv 3, 7, R.V a
cf 9 1.1.∀vAzv 7, 8, R.V I a
cf 10 1.∀zvAzv 2, 9, R.V I a
cf 11 1.2.T x R.V I I
cf 12 1.2.1.T y R.V I I
cf 13 1.2.1.∀vAyv 10, 12, R.V a
cf 14 1.2.1.Ayx 11, 13, R.V a
cf 15 1.2.∀yAyx 12, 14, R.V I a
cf 16 1.∀xyAyx 11, 15, R.V I a
cf 17 ∀xyAxy → ∀xyAyx 1, 16, R.I I
Each line contains a successive thesis of a calculus of function (hence ‘cf’) valid in
respective domain. Absolute theses are formulae cf 6 and cf 17. The example illustrates
not only the application of rules for inclusive quantifiers but also a dynamics of the system.
For example in line 7 a term assumption is introduced not with a prefix 2 but with a prefix
1.1.1, as a continuation of domain 1.1. In fact, a thesis cf 6 is not in itself very interesting
and may be seen as an auxiliary result required for proving cf 17 (to be more precise lines
1–3 are necessary for a proof of cf 17.).
474 A. Indrzejczak
3.1.3 Adequacy
Demonstration of completeness for ND systems is not demanding. If we have at our

disposal adequate axiomatic system it is enough to show that all axioms are provable
and primitive rules may by simulated in ND. We can also directly prove completeness for
ND with respect to semantics in the same way as it is done for axiomatic systems. Indirect
results of the first kind are provided in [19] for all ND systems except his inclusive logics
since there were neither axiomatic formulation of such a logic nor semantic one.
Usually the problem for ND systems with additional bookkeeping devices, is to prove
their soundness, because all this additional machinery must be somehow “translated”
either into semantics of suitable logic or into a simpler syntax of axiomatic system.
Jaśkowski [19] established some standard form of soundness proof extensively used by
many logicians in respective proofs for ND-systems. Shortly, for each prefixed formula
we build its development, which is a descending implication with suppositions for each
number in the prefix as antecedents and formula itself as the consequent. For example,
the development of prefixed formula i1 . . . .in .ϕ is ψ1 → (ψ2 → . . . (ψn → ϕ) . . .),
where each ψk , 1 ≤ k ≤ n is an assumption introduced with addition of successive ik to
already existing prefix, i.e. we have i1 . . . .ik .Sψk above i1 . . . .in .ϕ in the proof. Now we
may either directly prove that the development of a formula in each line is semantically
valid or that it is provable in respective axiomatic system. In the first case we proceed
by showing that the first line of a proof is valid (ϕ → ϕ) and that all rules expressed
in terms of developments are validity preserving. In the second case we must prove that
some formulae are theses of axiomatic systems. Specifically, Jaśkowski has proved for
his rules I, II, III (and additionally IVa) that the development of a formula in each line is
a thesis of an axiomatic system of positive (and intuitionistic logic). For CPL he proved
a soundness of his ND directly whereas for the system with quantifiers the result is only
pointed out. In case of his inclusive logic no such result was possible of course.
This manner of showing soundness for ND systems is commonly applied. There are
many variants of this technique but essentially we proceed in such a demonstration by
turning formulae of any proof into formulae or a kind of sequents (we add a record
of active assumptions), and then by showing that (such modified) rules are validity
preserving.7
3.2 Natural Deduction in Jaśkowski’s Lecture Notes
In 1947 Jaśkowski published in mimeographed form his lecture notes “Elements of

Mathematical Logic and Methodology of Deductive Sciences” in polish. The book
consists of 105 pages and is of great importance for us since Jaśkowski decided to apply in
it his natural deduction rules. It is perhaps the first logic textbook where natural deduction
is uniformly used as a method for presentation of logic. It is used from the beginning
for proving theorems of logic without any reference to axiomatic systems. Moreover, it
7 Although this approach is by no means the only one possible. For example, in [16] we proposed a
different general strategy of proving soundness for any ND system in Jaśkowski format.
is applied also in proofs of metalogical results and even truth-functional semantics is

introduced via analysis of ND proofs of selected theses.
A system described in [20] deserves the separate treatment since it contains significant
differences with the version from [19]. In [20] Jaśkowski presented:
1. classical propositional logic;
2. propositional logic with quantifiers;
3. classical first-order logic;
4. theory of identity in the second order language.
The most important changes in his later approach to ND are the following:
1. Richer language. Jaśkowski introduced ∧, ∨, ↔ and ∃ as primitive constants and
defined introduction and elimination rules for all of them.
2. Omission of nonclassical logics. Only classical logic is presented in lecture notes.
In particular, instead of inclusive logic there is a set of rules characterising classical
quantifiers. Moreover, the rules are generalised for second order variables and theory
of identity is expressed in the extended language.
3. Different style of layout for proofs.
3.2.1 Rules
Again the list of rules is opened by the rule for introducing assumptions: any formula may
be added with a horizontal bracket above it as an assumption (no prefix ‘S’ is attached in
front of). The rules for → and ¬ are without changes and the names for them are: C1
(implication introduction), C2 (MP) and N1 (negation elimination). For the remaining
connectives we have the following inference rules:
(K1) ϕ, ψ / ϕ ∧ ψ
(K2) ϕ ∧ ψ / ϕ and ϕ ∧ ψ / ψ
(A1) ϕ / ϕ ∨ ψ and ψ / ϕ ∨ ψ
(A2) ϕ ∨ ψ, ϕ → χ, ψ → χ / χ
(E1) ϕ → ψ, ψ → ϕ / ϕ ↔ ψ
(E2) ϕ ↔ ψ, ϕ / ψ and ϕ ↔ ψ, ψ / ϕ
In case of propositional logic with quantifiers two rules for ∀ are the same as in [19] (now
called (∀1) and (∀2)) but he added three more rules for ∃:
(∃1) ϕ[p/ψ] / ∃pϕ

(∃2) ∀p(ϕ → ψ), ∃pϕ / ∃pψ
(∃3) ∃pϕ / ϕ
The last one is the rule of eliminating vacuous quantification since it has a side condition
that p does not occur in ϕ. In (∃1) we have of course a proper substitution of ψ for p in
476 A. Indrzejczak
the premiss. Note that there is not one elimination rule for ∃ in the system. The possible
effect is divided into two rule with (∃2) being of rather mixed character.
The rules for quantifiers in first order logic are identical, the only difference is that
individual variables are bounded instead of propositional ones. Hence in particular,
introduction of ∀ is not a “real” proof construction rule in this system; we add ∀x to
some ϕ only after checking that x is not free in any active assumption. In contrast to rule
VIa from [19] there is no closure of a subproof and the rule exactly parallels rule VI. Of
course there is also no rule of introduction of term-assumptions in this system.
In both logics with quantifiers there is some innelegancy in the treatment of ∃.
However, it works and we avoid the problems usually generated in other ND systems
where some rule for ∃ elimination is postulated.8 In order to show that the set of rules is
complete it is enough to demonstrate as a thesis ∃xAx ↔ ¬∀¬Ax. We will show it in the
next section after characterising proof format.
Jaśkowski extended the application of his rules for quantifiers to second order logic,
mainly to develop the theory of identity. In an informal way he describe first the conditions
of proper substitution for predicate variables which is necessary for rules (∀1) and (∃1).
Identity is first characterised by Leibniz condition ∀A(Ax ↔ Ay) without explicit
introduction of a new constant. Then, in the section on definitions, he formulates its
definition ∀xy(x = y ↔ ∀A(Ax ↔ Ay)) and shows that by addition of it as a
new assumption we can deduce all characteristic properties of identity. Finally, when
discussing axiomatic systems, he provides also axiomatic characterization:
A1 ∀x, x = x
A2 ∀A∀xy(x = x → (Ax → Ay))
In all these formulae A is a predicate variable.
3.2.2 Proof Format
Proofs in [20] are written in linear form but the system of prefixes is absent in this
presentation. Instead Jaśkowski is using horizontal brackets over the assumption and
under the last formula in a subproof. This solution looks like a simplified version of his
first idea of using boxes. In fact, he tends to simplify other things as well; there is no rule
of repetition, no prefixes indicating suppositions and even no numeration of lines. Perhaps
resignation from prefixes is connected with resignation from Leśniewski’s like treatment
of deductive system as a dynamic body of theses valid in different domains. Jaśkowski in
[20] just provided a series of separate proofs of theses.
8 The history of successive versions of Copi’s ND with numerous mistaken formulation of this rule is
particularly instructive—see e.g. Annellis [1].
Below we provide a proof of the same propositional thesis which was displayed in
Sect. 3.1.2. for illustration.

1 p ass.

2 ¬¬¬p ass.

3 ¬¬p ass.

4 ¬p 2, 3, N1

5 ¬¬p 1, 4, N1

6 p → ¬¬p 1, 5, C1
In order to show the difference between the inclusive rules from [19] and classical rules we
provide two proofs of a thesis which is also valid in inclusive logic. For easier comparison
we settle the first proof also in Jaśkowski’s new (bracketing) style.

1 ∀x(Ax → Bx) R.I

2 ∀xAx R.I

3 Ty R.V I I.
4 Ay → By 1, R.V a
5 Ay 2, R.V a
6 By 4, 5, R.I I I

7 ∀xBx

3, 6, R.V I a
8 ∀xAx

→
∀xBx 2, 7, R.I I
9 ∀x(Ax → Bx) → (∀xAx → ∀xBx) 1, 8, R.I I

1 ∀x(Ax → Bx) ass.

2 ∀xAx ass.
3 Ax → Bx 1, ∀1
4 Ax 2, ∀1
5 Bx 3, 4, C2
6 ∀xBx

5, ∀2
7 ∀xAx

→
∀xBx 2, 6, C1
8 ∀x(Ax → Bx) → (∀xAx → ∀xBx) 1, 7, C1
478 A. Indrzejczak
The selection of rules for ∃ may seem doubtful at first but, in contrast to some other
solutions, it has some advantages. All the rules are simple in their form and normal in
the sense that premisses logically imply conclusions. In Gentzen’s approach the rule for
elimination of ∃ is a proof construction rule introducing additional subproof. In systems
where some inference rule of this kind is provided it is connected with some (sometimes
considerably complicated) side conditions (like in Quine’s natural deduction [31] or
Słupecki and Borkowski’s [6] solution).
In order to show that Jaśkowski’s characterization of ∃ is sufficient it is enough to
demonstrate that ∃xAx ↔ ¬∀x¬Ax is derivable (normality of rules yields soundness).

1 Ax ass.

2 ¬¬∀x¬Ax ass.
3 ∀x¬Ax 2, CP L
4 ¬Ax

3, ∀1
5 ¬∀x¬Ax

1, 4, N1
6 Ax → ¬∀x¬Ax 1, 5, C1
7 ∀x(Ax → ¬∀x¬Ax) 6, ∀2

8 ∃xAx ass.
9 ∃x¬∀x¬Ax 7, 8, ∃2
10 ¬∀x¬Ax

9, ∃3
11 ∃xAx → ¬∀x¬Ax 8, 10, C1
We first prove an auxiliary thesis (in line 7) which is then used as one of the premisses for
the application of (∃2). Also (∃3) is used in line 10 to eliminate vacuous quantification.

1 ¬∃xAx ass.

2 ¬¬Ax ass.
3 Ax 2, CP L
4 ∃xAx

3, ∃1
5 ¬Ax 1, 4, N1
6 ∀x¬Ax

5, ∀2
7 ¬∃xAx → ∀x¬Ax 2, 6, C1
8 ¬∀x¬Ax → ∃xAx 7, CP L
In the proof of the converse we applied (∃1) in line 4 thus showing that all three rules for
∃ yield a complete characterization of ∃.
4 Other Approaches to ND
ND was not also independently proposed by Gentzen but his proposal is widely known,
in contrast to Jaśkowski. Before we try to explain why Gentzen is better recognized as a
father of ND we briefly describe the most important similarities and differences.9
When we consider the rules both approaches are very similar. Gentzen also considered
classical and intuitionistic logic; the former also in first order case, the latter in Heyting’s
version (and not restricted to propositional part). In contrast to Jaśkowski he prefers to
work with richer language, in particular because he was interested in the philosophical
project of syntactic characterization of logical constants by means of rules later developed
in inferentialist program. The main difference lies in the format of proof chosen by
both authors. Gentzen defined proofs as trees labelled with formulae, where leafs are
assumptions and the conclusion is put in the root. Transitions between nodes correspond
to elementary inferences.
The distinction between ND-systems using tree- or linear-format of proof seems to be
not very serious from the theoretical point of view. After all we can redefine every binary
tree as a sequence. But in practice the difference is very important because in linear proof
we are dealing with formulae, whereas in tree-proofs we are dealing with their concrete
occurrences. Since we may use the same formula many times in linear proof, we are
forced to introduce some devices for cancelling the part of a proof which is in the scope
of an assumption already discharged. Otherwise we could “prove” everything, as was
illustrated in Sect. 3.1.2.
Scoping difficulties do not occur in tree-proofs because we are operating not on
formulae but on their single occurrences. Thus premises of any application of a rule
must always be displayed directly over the conclusion. Consequently, we cannot use in
a proof something which depends on discharged assumptions, because a part of a proof
responsible for deduction of a formula must be reproduced above. So Gentzen did not
need to bother about technical devices to block nonvalid deductions. Tree format requires
less complicated machinery and is very good in representing ready proofs, because the
structure of inferential dependencies is readably represented. Moreover, tree proofs are
better for proving theoretical results in proof theory. Prawitz [26] proved the profound
result on the existence of proofs in normal forms for many ND systems with tree-proofs
(Gentzen did it indirectly by showing equivalence with cut-free sequent calculus). No
wonder that in works concerned with theoretical investigations this format is very popular
(good witness is Negri and von Plato [23].).
But the features of tree proofs that make them so attractive are also the source of some
problems. Tree nicely shows the structure of a finished proof but it is hardly suitable for
actual derivation. Mental process of proof construction has rather linear structure; we start
with assumptions and deduce conclusions until we get the desired goal. Gentzen himself
was well aware of this fact, when he wrote that “we are deviating somewhat from the
analogy with actual reasoning. This is so, since in actual reasoning we necessarily have
(1) a linear sequence of propositions due to the linear ordering of our utterances, and (2)
we are accustomed to applying repeatedly a result once it has been obtained, whereas the
9 More detailed comparison of both approaches may be found in Hazen and Pelletier [15].
480 A. Indrzejczak
tree form permits only of a single use of a derived formula.” ([13] citation from [30, page
76]) According to Gentzen however, this form of representation is simpler and resulting
deviations are inessential.
The choice of linear proof format has also some computational advantages; we can
show that for each proof D in tree format we can provide a linear proof D such that the
length of D is the same or smaller than the length of D. The converse does not hold
because in tree proofs we work with occurrences of formulae when in linear proofs we
work with formulae themselves. It forces us to repeat many times the same proof-trees if
their starting assumptions are used several times. This also shows that linear proofs are
better from the practical point of view.
It seems that Jaśkowski decided to use linear format as much closer to actual reasoning,
and much more useful for actual proof-search. The applications of ND in logic textbooks
are good witnesses of this choice; there is only a few such books using tree format.
Majority of them use linear proofs and basically all of them are some variations on the
first two solutions introduced by Jaśkowski.
Despite the apparent differences, one thing is common to all variants of Jaśkowski’s
ND—an essential idea of dividing a proof into separated and partially ordered subproofs.
It appears as the most popular solution in hundreds of textbooks where ND-techniques are
applied. His first version (graphical), although abandoned by the author himself, is much
more popular nowadays. It has many variants but there is always some graphical device
added to linear sequence of formulae in a proof. The original format of boxes was used
by Kalish and Montague [21], but with some adjustments which make their system one of
the most flexible in practise. In their system each box is preceded with so called ‘show-
line’ which indicates the goal of deduction to be performed inside the box. After closing
a box such a show-line is treated as a new formula in the proof. Simplified account, where
each assumption is entered with the vertical line which continues until this subproof is in
force, is due to Fitch [10], whereas popular system of Copi [8] applies vertical bracketing
to closed subproofs. These solutions were repeated in hundreds of textbooks.
It should be noted also that this approach proved especially useful with respect to
many nonclassical logics formalized via ND systems. Because parts of proof are separated
graphically it is easy to distinguish between different types of subproofs and formulate
several kinds of repetition rules with restrictions on the form of formulae which may be
shifted to subproofs. One may find ND systems of this kind for modal logics (c.f. Fitting
[11], Garson [12], Indrzejczak [16]), relevant logics (Anderson and Belnap [2]) and many
others.
The second solution of Jaśkowski is not so popular in ND setting. Borkowski
and Słupecki in their ND system from [6] followed this route but with significant
simplifications. First of all they treat prefixes as just line-numbers of the proof. They
also avoid a proliferation of prefixes since they do not introduce a new prefix for every
assumption. Each thesis is analysed in terms of descending implication and all antecedents
are written in the same proof level. The proof is ready if the succedent is deduced. For
example, if we want to prove a thesis of the form ϕ1 → (ϕ2 → ψ) we construct a proof

looking like this:
1. ϕ1 ass.
2. ϕ2 ass.
..
.
n. ψ
instead of Jaśkowski’s more complicated form:
1 1.Sϕ1 R.I
2 1.1.Sϕ2 R.I
..
.
n 1.k.ψ
n + 1 1.ϕ2 → ψ 2, n, R.I I
n + 2 ϕ1 → (ϕ2 → ψ) 1, n + 1, R.I I
Of course, if a thesis to be proved is not an implication we must start with indirect

assumption and proceed with indirect proof, hence a rule for indirect proof is a primitive
one. In fact it is also the only indispensable proof construction rule in the system
since in the definition of a proof they allow for introduction of previously proved
theses. In consequence, such rules like introduction of implication, or other based on
the introduction of additional assumptions, are admissible in their system but in fact we
can dispense with subproofs and additional column of numerals for their indication. The
problem of elimination rule for ∃ is also solved in the original way in their system. They
apply the inference rule which implicitly uses skolemization. One may find an extensive
applications of their system to logic and set theory in many textbooks written in Polish as
well as in English translation [28].
The system of Słupecki and Borkowski is rather not known outside Poland but it was
also interestingly applied in the field of automated deduction. In 1970s Andrzej Trybulec
started to develop an integrated framework for deduction of theorems in mathematical
theories called MIZAR. It is basically a computer environment allowing formalization and
proof-checking on the basis of rich library. In 1980s Professor Marciszewski initiated
a research program concerned with the applicability of MIZAR to construct and to
check formal proofs in Słupecki and Borkowski system. The program was developed
by numerous scholars in many centers, for example in Opole (Wybraniec-Skardowska,
Bryniarski) and Łódź (Malinowski, Nowak, Łukowski). It shows a great potential of
MIZAR and natural deduction system in formalization of logic and formal theories. Proofs
in MIZAR are constructed similarly as in Słupecki and Borkowski system although some
additional devices for users-friendly presentation are added. The present library is based
on axioms of set theory in the version due to Tarski-Grothendieck and includes over 1300
articles written by nearly three hundreds of researchers (see www.mizar.org).
482 A. Indrzejczak
There is a kind of ND systems which at first sight may be seen also as a simplification
of Jaśkowski’s second variant. I mean here a system of Suppes [29] where in each line
of a proof we have added a set of numerals of all assumptions active for the formula in
question. But the similarity to Jaśkowski’s prefixes is apparent in this kind of ND. Lines in
Suppes’ ND correspond rather to sequents; a set of numerals is a shortcut for antecedent
of a sequent. Such a simplification is possible if all rules operate only on succedents of
sequents. We do not enter into the details of such solution but it should be noted that such
kind of ND is rather a by-product of Gentzen’s later paper [14].
Jaśkowski’s third system was not known and it is hard to find similar solutions, except
perhaps a system presented by Corcoran and Weaver [9]. Here proofs are written down
horizontally with subproofs put in brackets. Thus our example proof in Corcoran’s style
looks like that:
[p[¬¬¬p, p[¬¬p, ¬¬¬p]¬p]¬¬p]p → ¬¬p
Our claim that Jaśkowski’s lecture notes are perhaps the first consequent textbook
application of ND requires some justification. Quine [31] claims that the first textbook
applying ND is due to Cooley [7] and was printed in 1942, then reprinted in 1946.
In fact, Cooley applies numerous inference rules throughout the book, however it may
be disputable if it is ND system satisfying our three criteria. Conditional proofs based
on additional assumptions are only described on pp. 126–140 but not used as the main
form of presentation of logic. Moreover, Cooley did not apply any devices for separating
subproofs and a rule for elimination of existential quantifier is stated without sufficient
restrictions. Hence in our opinion it cannot be treated as a correct system of ND. It seems
that the first textbook which consequently applies ND is that of Fitch [10] published
in 1952. In Quine’s [31] from 1950, ND is also introduced only in three sections as
an illustration rather, not as the main proof system. Quine mentioned also some earlier
mimeographed notes of himself and of Rosser which applied ND but I had no possibility
to check them. A well known textbook of Rosser [27] is using axiomatic system and
introduces additional ND-like rules only as a metalogical devices for simplification of
axiomatic proofs.
We can conclude our considerations with the following remark concerning Jaśkowski
and Gentzen. Both authors laid down the foundations for further investigations on ND
but in a slightly different fashion. Jaśkowski seemed to be more concerned with practical
aspects of deduction and his general approach, as well as his technical solutions, are of
common classroom and textbook use. On the other hand, Gentzen was more theoretically
oriented; his investigations led him to profound results in general proof theory.
This is my own evaluation of Jaśkowski’s real influence on ND. It is based on the
analysis of his texts and easily verifiable. But it should be contrasted with the real
knowledge of his achievements and impact on ND. In the earliest applications of ND,
like in Quine’s or Fitch’s book, the origins of the method are known and confirmed.
For example, Fitch in foreword claimed that he is using the method of subordinate
proofs since 1941 but both Gentzen and Jaśkowski are mentioned as the source of
inspiration. Unfortunately, later authors often tend to say about Fitch’s ND and forget
about Jaśkowski.
References
1. Anellis, I.H.: Forty years of “unnatural” natural deduction and quantification. A history of first-order
systems of natural deduction from Gentzen to Copi. Mod. Log. 2(2), 113–152 (1991)
2. Anderson, A.R., Belnap, N.D.: Entailment: The Logic of Relevance and Necessity, vol. I. Princeton
University Press, Princeton (1975)
3. Barth, E.M., Krabbe, E.C.: From Axiom to Dialogue. Walter de Gruyter, Berlin (1982)
4. Bencivenga, E.: Free logics. In: Gabbay, D., Guenthner, F. (eds.) Handbook of Philosophical Logic,
vol. III, pp. 373–426. Reidel Publishing Company, Dordrecht (1986)
5. Bencivenga, E.: Jaśkowski’s universally free logic. Stud. Logica 102(6), 1095–1102 (2014)
6. Borkowski, L., Słupecki, J.: A Logical System based on rules and its applications in teaching
Mathematical Logic. Stud. Logica 7, 71–113 (1958)
7. Cooley, J.C.: A Primer of Formal Logic. Macmillan, New York (1942)
8. Copi, I.M.: Symbolic Logic. The Macmillan Company, New York (1954)
9. Corcoran, J., Weaver, G.: Logical consequence in modal logic: Natural Deduction in S5. Notre Dame
J. Formal Log. 10, 370–384 (1969)
10. Fitch, F.: Symbolic Logic. Ronald Press Company, New York (1952)
11. Fitting, M.: Proof Methods for Modal and Intuitionistic Logics. Reidel, Dordrecht (1983)
12. Garson, J.W.: Modal Logic for Philosophers. Cambridge University Press, Cambridge (2006)
13. Gentzen, G.: Untersuchungen über das Logische Schliessen. Math. Z. 39, 176–210, 405–431 (1934)
14. Gentzen, G.: Die Widerspruchsfreiheit der reinen Zahlentheorie. Math. Ann. 112, 493–565 (1936)
15. Hazen, A.P., Pelletier, F.J.: Gentzen and Jaśkowski Natural Deduction: Fundamentally Similar but
Importantly Different. Stud. Logica 102(6), 1103–1142 (2014)
16. Indrzejczak, A.: Natural Deduction, Hybrid Systems and Modal Logics. Springer, Berlin (2010)
17. Indrzejczak, A.: Natural deduction. In: Internet Encyclopedia of Philosophy. http://www.iep.utm.edu/
nat-ded/
18. Jaśkowski, S.: Teoria dedukcji oparta na dyrektywach założeniowych [Theory of deduction based on
suppositional directives]. In: Ksi˛ega Pamiatkowa
˛ I Polskiego Zjazdu Matematycznego. Uniwersytet
Jagielloński, Kraków (1929)
19. Jaśkowski, S.: On the rules of suppositions in formal logic. Stud. Logica 1, 5–32 (1934) (Reprinted
in: Storrs McCall (ed.) Polish Logic 1920–1939, pp. 232–258. Oxford (1967))
20. Jaśkowski, S.: Elementy logiki matematycznej i metodologii nauk ścisłych [Elements of mathematical
logic and methodology of formal sciences]. Lecture notes, Toruń (1947)
21. Kalish, D., Montague, R.: Logic, Techniques of Formal Reasoning. Harcourt, Brace and World, New
York (1964)
22. Kolmogoroff, A.: O principie tertium non datur. Matematicesi Sbornik 32, 651
23. Negri, S., von Plato, J.: Structural Proof Theory. Cambridge University Press, Cambridge (2001)
24. Pelletier, F.J.: A Brief History of Natural Deduction. Hist. Philos. Log 20, 1–31 (1999)
25. Pelletier, F.J., Hazen, A.P.: A history of natural deduction. In: Gabbay, D., Pelletier, F.J., Woods, E.
(eds.) Handbook of the History of Logic, vol. 11, pp. 341–414 (2012)
26. Prawitz, D.: Natural Deduction. Almqvist and Wiksell, Stockholm (1965)
27. Rosser, J.B.: Logic for Mathematicians. McGraw Hill, New York (1953)
28. Słupecki, J., Borkowski, L.: Elements of Mathematical Logic and Set Theory. Pergamon Press,
Oxford (1967)
29. Suppes, P.: Introduction to Logic. Van Nostrand, Princeton (1957)
30. Szabo, M.E.: The Collected Papers of Gerhard Gentzen. North-Holland, Amsterdam (1969)
31. Van Orman Quine, W.: Methods of Logic. Holt, Rinehart and Winston, New York (1950)
32. Woleński, J.: Logic and Philosophy in the Lvov-Warsaw School. Kluwer, Dordrecht (1989)
A. Indrzejczak ()
Department of Logic, University of Lodz, Lodz, Poland
Variations on Jaśkowski’s Discursive Logic
Barbara Dunin-K˛eplicz, Alina Powała, and Andrzej Szałas
Abstract Stanisław Jaśkowski, in his 1948–1949 papers on propositional calculus for

contradictory deductive systems, proposed discursive logic D2 . The main motivation
behind D2 is the need to properly deal with contradictions that naturally appear in
many areas of philosophy and discourse. The intuitive justification of this logic reflects
knowledge fusion occurring when “the theses advanced by several participants in a
discourse are combined into a single system.” This point of view was seminal in the mid
twentieth century and remains visionary nowadays.
In contemporary autonomous systems operating in dynamic, unpredictable
information-rich environments, distributed reasoning routinely takes place. This
explains the key role of knowledge fusion, among others, in Distributed Artificial
Intelligence. Therefore, different types of modern knowledge and belief bases become
primarily concerned with inconsistent or lacking information. This requirement leads to
recent approaches to paraconsistent and paracomplete reasoning, where nonmonotonic
techniques for disambiguating inconsistencies and completing missing knowledge can be
applied.
In this chapter we remind Jaśkowski’s seminal, pioneering work on paraconsistent
reasoning and indicate some of its relations to contemporary research on reasoning in
Distributed AI.
Keywords Discussive logic · Paraconsistent reasoning · Argumentation · Belief

structures
Mathematics Subject Classification (2000) Primary 03B53; Secondary 03B50, 03B42,

68T27, 68T30
1 Prelude
Stanisław Jaśkowski introduced discursive logic D2 (called also discussive logic) in his
visionary papers [24, 25] (for their English versions see [26, 27]). It has been the first
formal paraconsistent logic proposed in the literature and has opened a wide area of
paraconsistent reasoning (for surveys see [2, 4, 34]). It also inspired many researchers

486 B. Dunin-K˛eplicz et al.
who published many papers focusing solely on or directly motivated by D2 (like,

e.g., [3, 8, 9, 11, 31, 39]).
In defining D2 , Jaśkowski used S5 worlds to model sets of beliefs of the discussing
participants. We say that a statement is a consequence of a discussion if it follows from at
least one belief set (i.e., at least one S5 world). That way different participants may express
contradictory statements α and ¬α while the conjunction α ∧ ¬α cannot be derived.
In the current chapter we recall D2 and indicate its connections to contemporary
research on reasoning in many subareas of artificial intelligence. In particular, we provide
a new formalization of Jaśkowski ideas in terms of belief structures introduced by
Dunin-K˛eplicz and Szałas in [17–19]. Belief structures are built over a four-valued
logic of [40] with truth values t (true), f (false), i (inconsistent) and u (unknown). This
new paraconsistent and paracomplete formalization provides a shift from the deductive
perspective to belief bases perspective. While, in the former, reasoning depends on
deriving conclusions valid in all models of premises—in the later, one derives conclusions
valid in a single model representing the current state of the world. Of course, the
formalization in belief structures is not equivalent to D2 as it is well-known that there
is no characterization of S5 in any finitely-valued logic [13].
In the formalism of belief structures, belief bases are understood as sets of worlds.
However, these worlds can contain contradictory claims what makes them incompatible
with modal worlds. Also, there is no need to use Kripke-like accessibility relation on
worlds. Instead, we focus on epistemic profiles designed for reflecting the dynamics of
belief formation and revision. The concept of epistemic profile embodies an individual’s
(alternatively called an agent) or group of individuals reasoning capabilities encompassing
techniques suitable for different aspects of activities.
Arguably, Jaśkowski with his ideas addressing paraconsistent reasoning, especially
in the context of discursive logics, has been much ahead of his times. To show the
bridge between D2 and contemporary research on belief bases, argumentation, knowledge
representation, artificial intelligence, autonomous systems, etc., we define a new logic D4 .
While formalizing Jaśkowski’s ideas behind D2 , it also enjoys the following features:
• the formalization allows to distinguish among statements supplied by different partic-
ipants of discussion;
• it provides tools for both paraconsistent and paracomplete reasoning, allowing for dis-
ambiguating of inconsistencies and completing missing knowledge in a nonmonotonic
manner;
• it is computationally feasible: for implementation one can use 4QL, a rule language
developed in [28, 29, 37].1
Our variations on Jaśkowski’s ideas start with the current prelude. Next, in Sect. 2 the
main theme, Jaśkowski’s Discursive Logic D2 , is reminded. The “movement” (Sect. 3)
presents the main ideas and definitions related to belief structures. Then, in Sects. 4–6
three variations on D2 are composed. The first one (Sect. 4) presents D4 , a new four-
valued formalization of Jaśkowski’s intuitions behind D2 . The second variation (Sect. 5)
relates D2 and D4 to contemporary research on dialogues. The last variation (Sect. 6)
elaborates on connections of discursive logics to selected work on argumentation. Finally,
the coda (Sect. 7) concludes our variations.
1 For open-source interpreters of 4QL, see 4ql.org.

Variations on Jaśkowski’s Discursive Logic 487
2 Theme: Jaśkowski’s Discursive Logic D2
In his papers on D2 [24–27], Jaśkowski addressed the following problem:

[. . . ] the problem of the logic of contradictory systems [inconsistent systems] is formulated here
in the following manner: the task is to find a system of the sentential calculus which: 1) when
applied to contradictory systems would not always entail their over-completeness; 2) would be
rich enough to enable practical inference; 3) would have an intuitive justification.
For simplicity, as the underlying logic Jaśkowski has chosen propositional modal logic
S5 with usual classical connectives ¬, ∧, ∨, →, ≡ together with modalities , ♦, and
considered additional connectives:
def
• discussive implication: p →d q ≡ (♦p → q);
def
• discussive equivalence: p ≡d q ≡ (p →d q) ∧d (q →d p);
def
• discussive conjunction: p ∧d q ≡ (p ∧ ♦q).
As summarized in [34],
we think of each participant’s belief set as the set of sentences true at a world in a S5 model M.
Thus, a sentence α asserted by a participant in a discourse is interpreted as “it is possible that α”
(♦α).
Let us now define the discursive consequence relation. For a similar formulation see,
e.g., Example 24 of [7]. We shall need the following translation function from D2 formulas
into S5 formulas:
def
T r(p) = p for p being a propositional variable;
def
T r(α ∧d β) = T r(α) ∧ ♦T r(β);
def
T r(α →d β) = ♦T r(α) → T r(β);
def
T r(α ≡d β) = (♦T r(α) → T r(β)) ∧ (♦T r(β) → T r(α)).
We assume that T r preserves all other connectives and, for a set of formulas F ,
def def
T r(F ) = {T r(α) | α ∈ F } and ♦F = {♦α | α ∈ F }.
Definition 2.1 The discursive consequence relation, D2 is defined by:
F D2 α iff ♦T r(F ) S5 ♦T r(α), (2.1)
where F is a set of formulas and α is a formula.
We are now in position to recall Jaśkowski’s motivations concerning discussive connec-

tives →d , ∧d and ≡d .
First, the motivation behind →d , as stated by Jaśkowski (see [26]), is the failure of
modus ponens-based reasoning when traditional implication is used:
If implication is interpreted so as it is done in two-valued logic, then out of the two theses one of
which is p → q and thus states “it is possible that if p then q”, and the other is p, and thus states
“it is possible that p”, it does not follow that “it is possible that q”, so that the thesis q does not
follow intuitively, as the rule of modus ponens requires.
[. . . ] This is why in the search for a “logic of discourse” the prime task is to choose such a function
which, when applied to discursive theses, would play the role analogous to that which in ordinary
systems is played by implication.
Indeed, from p (i.e., ♦p) together with p →d q (i.e. ♦p → q) we can deduce q

(so ♦q, too).
The discussive conjunction and equivalence are motivated by the following important
theorem (see [27]):
Each thesis α of the two-valued classical calculus containing no other symbols than →, ≡, ∨ or
∧ is transformed into thesis of the discussive calculus D2 by replacing in α functors → by →d , ≡
by ≡d , and ∧ by ∧d , respectively.
Additionally, discussive conjunction maintains the adjunction principle according to

which p, ¬p | p ∧ ¬p. Namely, for ∧d adjunction holds [25], since from p, ¬p one
can deduce p ∧d ¬p.
3 Movement: Belief Bases and Belief Structures
This section is based on [19]. However, for clarity, we restrict the presentation to
propositional logic. We use the classical propositional syntax but the presented semantics
substantially differs from the classical one. Namely,
• truth values t, i, u, f (true, inconsistent, unknown, false) are explicitly present;
• the semantics is based on sets of literals rather than on valuations of propositional
variables.
This allows one to deal with the lack of information as well as inconsistencies. The
underlying semantics of propositional connectives is the one of [40]. It is summarized
in Table 1. Observe that definitions of ∧ and ∨ reflect minimum and maximum w.r.t. the
ordering:
f < u < i < t, (3.1)
as advocated, e.g., in [10, 28, 37, 40]. Such a truth ordering appears to be natural and
reflecting intuitions of the classical two-valued logic. For example, a conjunction is true
if all its operands are true, etc.
Table 1 Truth tables for ∧, ∨, → and ¬ (see [28, 29, 40])

∧ f u i t ∨ f u i t → f u i t ¬
f f f f f f f u i t f t t t t f t
u f u u u u u u i t u t t t t u u
i f u i i i i i i t i f f t f i i
t f u i t t t t t t t f f t t t f
A positive literal is a propositional variable and a negative literal is a negated proposi-

tional variable.
Definition 3.1 The truth value of a literal w.r.t. a set of literals L, denoted by (L), is
defined as follows:
⎧
⎪t
⎪
⎨
if ∈ L and (¬) ∈
/ L;
def i if ∈ L and (¬) ∈ L;
(L) =
⎪
⎪ u if ∈
/ L and (¬) ∈
/ L;
⎩
f if ∈
/ L and (¬) ∈ L.
Definition 3.1 is extended to all propositional formulas in the standard way, using the
semantics provided in Table 1.
If S is a set then by FIN (S) we understand the set of all finite subsets of S. By C we
denote the set of all finite sets of literals.
Definition 3.2 By a belief base we understand any finite set of finite sets of literals,
i.e., any finite set ⊆ C.
Note that such belief bases can be tractably implemented using the 4QL rule language [28,
29, 37]. They serve as basis for belief structures. Indeed, constituents and consequents
being basic building blocks of belief structures are, in fact, belief bases in the sense of
Definition 3.2.
By information ordering we understand the ordering on truth values shown in Fig. 1.
This ordering reflects the process of gathering and fusing information. Starting from the
lack of information, in the course of belief acquisition, evidence supporting or denying
investigated hypotheses are collected. This finally permits one to decide about the truth
value of the hypotheses.
Definition 3.3 Let be a belief base and α be a formula. We define the belief operator
def
by: Bel α ≡ LUB{α(D) | D ∈ }, where LUB denotes the least upper bound w.r.t.
the ordering shown in Fig. 1.
For clarity let us indicate that:

Bel t = t when t ∈ {t, i, f, u}. (3.2)
Fig. 1 Information ordering

on truth values
Note that sets D ∈ appearing in Definition 3.3 can be considered as four-valued worlds.
Comparing to Kripke-like semantics for beliefs (see, e.g., [20]), at this point the main
differences are:
• we do not require fixed, rigid structure connecting worlds via accessibility relations;
• we use four rather than two truth values.
We are now ready to define (indeterministic) belief structures, as in [19].2 Belief
structures consist of constituents and consequents: an agent starts with constituents,
which are further transformed into consequents via the agent’s or group’s epistemic
profile. While constituents contain initial, “raw” beliefs acquired by perception, expert-
supplied knowledge, communication, discussion and other ways, consequents contain
final, “mature” beliefs. In short, an epistemic profile encapsulates agents’ or groups’
reasoning capabilities, including methods of both disambiguation of inconsistencies and
completing missing information.
Definition 3.4
• By a constituent we understand any set C ∈ C;
• by an indeterministic epistemic profile we understand any function E of the sort
FIN(C) −→ FIN(C);
• by an indeterministic belief structure over an indeterministic epistemic profile E we
mean B E = C, F !, where:
– C ⊆ C is a nonempty set of constituents;
– F = E(C) is the set of consequents of B E .
def

A formula is Bel -free if it does not contain belief operators. Let us emphasize that Bel -
free formulas reflect properties of initial beliefs, being evaluated in constituents while the
belief operator Bel refers to consequents, so allows us to express properties of final
beliefs, as stated in the following definition.
Definition 3.5 Let B1E = C1 , F1 ! and B2P = C2 , F2 ! be indeterministic belief
structures. The semantics of formulas is defined by:

BelC1 α when α is Bel -free;
α(B1E ) =
def
BelF2 β when α is of the formBelBP β and β is Bel -free,
2

where BelC1 ,v α and BelF2 ,v β are defined in Definition 3.3.3
The above definition can be extended for all formulas by defining the semantics of
connectives as in Sect. 3 and nested Bel operators starting from the innermost ones.
2 Note that epistemic profiles of [17, 18] are functions of the sort FIN(C) −→ C. That is, they basically
are deterministic epistemic profiles with F consisting of one consequent.
3 Note that, in the simplest case, B E and B E can be identical.
1 2
Recall after [19] that typical requirements as to belief operators are satisfied, where α
is any formula and B E is any belief structure4 :

¬BelBE f B E = t (consistency of beliefs)
E
BelBE α → BelBE Bel B =t
BE α (positive introspection)
¬BelBE α → BelBE ¬BelBE α BE = t (negative introspection)
4 Variations Part I: D4 —A New Framework for Discursive

Logics
D2 has a potential to be extended in many directions. In particular, the following aspects

can be addressed.
• A participant in a discussion should be allowed to submit inconsistent statements, as
advocated, among others, in [23]. Therefore, the relevant worlds should not exclude
contradictory statements, as it happens in S5, so in D2 , too.
• In contemporary systems it is often important to distinguish among statements
supplied by various participants of distributed reasoning and knowledge fusion. This
aspect might be essential in formulating adequate strategies of disambiguation of
inconsistencies.
• As the accessible information may be incomplete as well, to reflect this property not
only paraconsistent but also paracomplete reasoning is often needed.
In order to formalize Jaśkowski’s intuitions behind discursive logic while addressing the
above aspects, one can use the framework of belief structures. Technically speaking, be-
liefs are represented as sets of literals constituting paraconsistent belief bases. Epistemic
profiles are represented as specific rules operating on possibly complex belief structures in
order to draw individual conclusions. Discursive reasoning can be used to define epistemic
profiles of individuals and groups.
In order to define a logic D4 , let us first assume that discussion
participants have, as
a group, an associated belief structure, say B E . Since Bel corresponds to modal , we
define ♦ as usually:
def
♦α ≡ ¬BelBE ¬α , (4.1)
modify translation T r, in such a way that wherever ♦ occurs, it is

and, consequently,
replaced by ¬Bel ¬. We denote this modified translation by T rm .
To compute the consequences according to Definition 2.1, we have to evaluate the
formula ♦T r(F ) S5 ♦T r(α). Since formulas involved are completely modalized, we
4 Observe that the property of consistency of beliefs requires beliefs to exclude only falsity f. On the other
hand, beliefs can contain contradictory claims.
use deduction theorem for S5 [43] and obtain that ♦T r(F ) S5 ♦T r(α) is equivalent to:
S5 ♦T r(F ) → ♦T r(α). (4.2)
Now, rather than using S5, we use our formalization by evaluating the implication:

¬BelBE ¬T rm (φ) → ¬BelBE ¬T rm (α) . (4.3)
φ∈F
To distinguish among different discussion participants, we consider operators ♦A , where

A is a discussion participant, rather than just ♦ as in the original D2 . This is a rather
immediate extension of the method outlined above. Namely, the translation T rm should
be applicable to modal operators ♦A , so we replace such operators by ¬BelBE ¬ . . . ,
A
where BAE is a belief structure associated with participant A. Now one can use (4.3) with
such modified translation T rm .
5 Variations Part II: Relation to Dialogue
Complex communication patterns are essential in intelligent systems. Nowadays, rather

than rigid communication protocols, more relaxed communication forms are devel-
oped. Indeed, communicative actions are “actions that change your mind” [38]. Taking
a commonsense reasoning perspective calls for defeasible reasoning.
Contemporary approaches to communication in intelligent systems draw upon Walton
and Krabbe’s semi-formal theory of dialogue [42], adapting the normative models of
human communication, including paradigmatic dialogue types like inquiry, information
seeking, deliberation, persuasion and negotiation. See [5, 6, 12, 20, 30, 32, 33, 35]
for investigations in argumentation-based dialogue, and [42] for the definitions of
dialogue types. Each model of dialogue is defined by its initial situation, the participants’
individual goals, and the aim of the dialogue as a whole (see Table 2).
Complex dialogues are composed with the use of speech acts—the basic building
blocks of communication. Contemporary understanding of speech acts comes form the
works of Austin and Searle [1, 36] including the most popular taxonomy of speech acts,
identifying:
• assertives, committing to the truth of a proposition, e.g., stating;
• directives, which get the hearer to do something, e.g., asking;
• commissives, committing the speaker to some future action, e.g., promising;
• expressives, expressing a psychological state, e.g., thanking;
• declaratives, changing reality according to the proposition e.g., baptizing
Recently we developed a paraconsistent, paracomplete, dynamic and tractable formal
model of communication including:
• a formal model of speech-acts and reasoning schemes [16, 21];
• formalization of inquiry as a dialogue type for knowledge acquisition [15];
• formalization of persuasion as a dialogue type for conflict resolution [14].
Table 2 Types of dialogue recalled from [41]

Type of dialogue Initial situation Participants’ goal Goal of dialogue
Persuasion Conflict of opinions Persuade other party Resolve or clarify
issue
Inquiry Need to have proof Find and verify Prove (disprove)
evidence hypothesis
Negotiation Conflict of interests Get what you most Reasonable
want settlement both can
live with
Information seeking Need information Acquire or give Exchange
information information
Deliberation Dilemma or practical Coordinate goals and Decide best available
choice actions course of actions
Eristics Personal conflict Verbally hit out at Reveal deeper basis
opponent of conflict
Such a model of communication can be used to enrich D2 by developing discussion

patterns and related schemes.
The nature of multi-party inquiry and persuasion dialogues resembles distributed
defeasible reasoning processes, especially collective problem solving. The complex
logical architecture of both dialogue types permits to associate specific belief structures
with each of them. Namely, the specific rules governing each dialogue type are included
in the epistemic profile of a discussing group. Also specific methods for disambiguation
of inconsistencies and information completion, specific for inquiry and persuasion are
included in the involved epistemic profiles. Such an encapsulation of methods in epistemic
profiles permits to effectively model and investigate different dialogue types indicated in
Table 2. Technically, with each dialogue D, terminated or in progress, a specific epistemic

profile and a belief structure B D is associated and one can use belief operators BelBD
to formalize Jaśkowski’s discursive connectives, as outlined in Sect. 4.
Using this framework, one can obtain a rich formalism, adjustable to a variety of
dialogue types indicated in Table 2. Such a broader scope can still be rooted in D2 or
D4 , and deserves further investigations.5
6 Variations Part III: Relation to Argumentation
In realistic environments, heterogeneity of argumentation participants w.r.t. reasoning

manifests itself in different conclusions drawn by participants even facing the same
evidence. The notion of epistemic profile directly exposes this concept. In its abstract
form, epistemic profile, being arbitrary function, conveys all reasoning capabilities of an
argumentation participant. Due to this generic definition, also non-deductive reasoning
methods like argumentation schemes, can be included as a part of epistemic profiles.
5 Of course, one should take into considerations rich theories developed outside of logical formalisms, in
particular in the case of negotiations.
Argumentation schemes, originating from legal argumentation, attempt to classify

different types of everyday arguments, utilizing the ideas underlying nonmonotonic
formalisms. Each scheme is accompanied by a set of critical questions, used to evaluate
the argument. Although particular schemes may represent different types of reasoning
(e.g., deduction, induction, abduction, presumption), in general they aim to model
plausible, thus defeasible, reasoning.
In [16], paraconsistent argumentation schemes are modeled with the use of two
dedicated sets of premises and exceptions. Intuitively, when all premises are present and
none of the exceptions is present, the conclusion of the scheme can be drawn. To model
such schemes, we consider three sets of ground literals: premises (P ), exceptions (E) and
conclusions (Con), together with a function PAS({P , E}) = Con, which represents the
paraconsistent argumentation scheme. The set P contains candidates for conclusion of
the scheme. They are obtained by means specific to every argumentation scheme. The
elements of E are triggers that, when present, prevent drawing the respective candidate
conclusion. Intuitively, a conclusion c cannot be obtained when the exceptions indicate
¬c. Ultimately, the conclusion of the scheme is obtained as follows. If there exists
a candidate for a conclusion c ∈ P (value of c is not u), check whether there exists
a trigger ¬ c ∈ E blocking this candidate (value of ¬c is t). If the trigger:
• does not exist, the candidate conclusion becomes the final scheme conclusion,
• exists, the scheme cannot be applied causing the value of c ∈ Con to be u.
In short, a conclusion c is established based on the supporting arguments given by
the set P (i.e., c(P , v) = u) and (lack of) rebutting triggers provided by the set E
(i.e. ¬c(E, v) = t).
The definition below presents the paraconsistent argumentation scheme as a partial
function: a fragment of agent’s epistemic profile that expresses agent’s or group’s
argumentative skills.
Definition 6.1 Let P and E be two constituents, representing the set of premises and
exceptions, respectively, and let S = {P , E} ⊆ C be a nonempty set of constituents.
Then, a paraconsistent argumentation scheme (over S and Con) is a partial function:
def
PAS : FIN(C) → C such that for Con = PAS({P , E}) and c being a literal, we have:
⎧
⎪
⎪ t iff c(P) = t and ¬c(E) = t;
⎨
def i iff c(P) = i and ¬c(E) = t;
c(Con) =
⎪
⎪ u iff c(P) = u or ¬c(E) = t;
⎩
f iff c(P) = f and ¬c(E) = t.
By a belief structure associated with PAS we mean B PAS = S, {Con}!.
Note that the belief structure B PAS in Definition 6.1 is, in fact, deterministic as
the set of consequents contains only Con. This reflects the intuition that conclusions
are determined, e.g., by applying belief operators. A more comprehensive theory of
argumentation and communication founded on belief structures and 4QL, TALKLOG, is
presented in [14–16, 21]. Observe that 4QL as the implementation tool guarantees the
tractability of approach [28, 29, 37].
All and all, as in the case of dialogues, paraconsistent (and paracomplete) argumenta-
tion schemes can be viewed as a part of an agent’s epistemic profile utilizing the notions of
belief structures that can be directly translated into 4QL. Since with every paraconsistent
argumentation schema PAS there PAS , one can
is an associated belief structure B
consider belief operators BelBPAS and other operators of Jaśkowski’s discursive logic,
as indicated in the end of Sect. 6. This framework, as in the case of dialogues, opens
a wide spectrum of applications of D2 and D4 in modeling argumentation schemes and
reasoning about them.
7 Coda
Jaśkowski;s discursive logic occupies a meaningful place in philosophical logic from the
moment of its inauguration. Importantly, nowadays we observe an increased demand
for paraconsistent logics, which is stimulated by the needs of complex, real world
applications. As Dov Gabbay [22] noticed, “New logic areas have become established
and the old areas were enriched and expanded”. D2 fits in perfectly with this current
trend.
As expressed in Jaśkowski’s motivations behind discursive logic, inconsistency should
not immediately trivialize reasoning. This approach opens up the opportunity to continue
inference even when some information sources deliver contradictory information. In real-
world complex applications such a situation might be common for many practical reasons.
Ultimately, the inconsistencies are typically being resolved according to a chosen strategy
as to timing which, again, depends on the application in question. Apparently, various
forms of defeasible reasoning are applicable in this context.
In the current paper, when defining D4 we indicate a shift from modal perspective, with
reasoning over arbitrary theories, to reasoning from knowledge bases. While modeling
the world and reasoning usually ends up in models of high complexity, we generally have
more humble expectations from contemporary intelligent systems. We, therefore, often
lean to tailor the reasoning to rule-based approaches. Long investigations on complexity
of reasoning, in particular in the field of descriptive complexity, provide us with a very
good picture of what is and what is not tractable and supports this shift. Therefore, a
knowledge base perspective on reasoning presented in this chapter is beneficial also from
the complexity point of view.
Taking into account highly complex nature of environments real-world intelligent
systems are embedded in, the use of paracomplete and paraconsistent reasoning methods
proves invaluable. Also within that picture, Jaśkowski’s ideas are viable and inspiring.
Acknowledgements Supported by the Polish National Science Centre grants 2015/19/B/ST6/02589 and
2015/17/N/ST6/03642.
References
1. Austin, J.L.: In: Urmson, J.O., Sbisa, M. (eds.) How to Do Things with Words, 2nd edn. Clarendon
2. Béziau, J.-Y.: What is paraconsistent logic? In: Priest, G., Batens, D., Mortensen, C., vanBendegem,
J.P. (eds.) Frontiers in Paraconsistent Logic, pp. 95–111. Research Studies Press, Philadelphia (2000)
3. Béziau, J.-Y.: The paraconsistent logic Z. A possible solution to Jaśkowski’s problem. Log. Logical
Philos. 15(2), 99–111 (2006)
4. Béziau, J.-Y., Carnielli, W., Gabbay, D.M.: Handbook of Paraconsistency. College Publications,
London (2007)
5. Black, E.: A generative framework for argumentation-based inquiry dialogues. PhD thesis, University
College London (2007)
6. Bonzon, E., Maudet, N.: On the outcomes of multiparty persuasion. In: McBurney, P., Parsons, S.,
Rahwan, I. (eds.) Argumentation in Multi-Agent Systems. LNCS, vol. 7543, pp. 86–101. Springer,
Berlin (2012)
7. Carnielli, W., Marcelo, E.C., Marcos, J.: Logics of formal inconsistency. In: Béziau, J.-Y., Carnielli,
W., Gabbay, D.M. (eds.) Handbook of Paraconsistency. College Publications, London (2007)
8. Ciuciura, J.: Algebraization of Jaśkowski’s paraconsistent logic D2. Stud. Log. Grammar Rhetoric
42(1), 173–193 (2015)
9. da Costa, N.C.A., Francisco, D.A.: On Jaśkowski’s discussive logics. Stud. Logica 54(1), 33–60
(1995)
10. de Amo, S., Pais, M.S.: A paraconsistent logic approach for querying inconsistent databases. Int. J.
Approx. Reason. 46, 366–386 (2007)
11. de Moraes, L., Abe, J.: Some results on Jaśkowski’s discursive logic. Log. Logical Philos. 9, 25–33
(2004)
12. Dignum, F., Dunin-Kȩplicz, B., Verbrugge, R.: Creating collective intention through dialogue. Log.
J. IGPL 9, 145–158 (2001)
13. Dugundji, J.: Note on a property of matrices for Lewis and Langford’s calculi of propositions. J.
Symb. Log. 12(4), 150–151 (1940)
14. Dunin-Kȩplicz, B., Strachocka, A.: Paraconsistent multi-party persuasion in TalkLOG. In: Pro-
ceedings of the PRIMA’15: Principles and Practice of Multi-Agent Systems – 18th International
Conference. LNCS, vol. 9387, pp. 265–283. Springer, Cham (2015)
15. Dunin-Kȩplicz, B., Strachocka, A.: Tractable inquiry in information-rich environments. In: Proceed-
ings of the 24th International Joint Conference on Artificial Intelligence, IJCAI’15, pp. 53–60. AAAI
Press, Menlo Park (2015)
16. Dunin-Kȩplicz, B., Strachocka, A.: Paraconsistent argumentation schemes. Web Intell. 14, 43–65
(2016)
17. Dunin-K˛eplicz, B., Szałas, A.: Epistemic profiles and belief structures. In: Proc. KES-AMSTA 2012:
Agents and Multi-agent Systems: Technologies and Applications. LNCS, vol. 7327, pp. 360–369.
18. Dunin-K˛eplicz, B., Szałas, A.: Taming complex beliefs. Transactions on Computational Collective
Intelligence XI. LNCS, vol. 8065, pp. 1–21. Springer, Berlin (2013)
19. Dunin-K˛eplicz, B., Szałas, A.: Indeterministic belief structures. In: Jezic, G., Kusek, M., Lovrek, I.,
Howlett, J.R.J., Lakhmi, J.C. (eds.) Agent and Multi-Agent Systems: Technologies and Applications:
Proc. 8th Int. Conf. KES-AMSTA, pp. 57–66. Springer, Berlin (2014)
20. Dunin-K˛eplicz, B., Verbrugge, R.: Teamwork in Multi-Agent Systems. A Formal Approach. Wiley,
Hoboken (2010)
21. Dunin-Kȩplicz, B., Strachocka, A., Szałas, A., Verbrugge, R.: Paraconsistent semantics of speech
acts. Neurocomputing 151, 943–952 (2015)
22. Gabbay, D.M.: Preface to the second edition. In: Handbook of Philosophical Logic, vol. 4, pp. vii–ix.
Kluwer Academic, Dordrecht (2002)
23. Gabbay, D.M., Hunter, A.: Making inconsistency respectable: a logical framework for inconsistency
in reasoning, part I—a position paper. In: Jorrand, Ph., Kelemen, J. (eds.) Fundamentals of Artificial
Intelligence Research: International Workshop FAIR’91, pp. 19–32. Springer, Berlin (1991)
24. Jaśkowski, S.: Rachunek zdań dla systemów dedukcyjnych sprzecznych. Stud. Soc. Sci. Torun. 5,
55–77 (1948)
25. Jaśkowski, S.: O koniunkcji dyskusyjnej w rachunku zdań dla systemów dedukcyjnych sprzecznych.
Stud. Soc. Sci. Torun. 8, 171–172 (1949)
26. Jaśkowski, S.: Propositional calculus for contradictory deductive systems. Stud. Logica 24, 143–157
(1969)
27. Jaśkowski, S.: On the discussive conjunction in the propositional calculus for inconsistent deductive
systems. Log. Logical Philos. 7, 57–59 (1999)
28. Małuszyński, J., Szałas, A.: Living with inconsistency and taming nonmonotonicity. In: de Moor, O.,
Gottlob, G., Furche, T., Sellers, A. (eds.) Datalog 2.0. LNCS, vol. 6702, pp. 384–398. Springer, Berlin
(2011)
29. Małuszyński, J., Szałas, A.: Partiality and inconsistency in agents’ belief bases. In: Barbucha, D.,
Le, M.T., Howlett, R.J., Jain, L.C. (eds.) KES-AMSTA. Frontiers in Artificial Intelligence and
Applications, vol. 252, pp. 3–17. IOS Press, Amsterdam (2013)
30. McBurney, P., Hitchcock, D., Parsons, S.: The eightfold way of deliberation dialogue. Int. J. Intell.
Syst. 22(1), 95–132 (2007)
31. Nasieniewski, M., Pietruszczak, A.: On modal logics defining Jaśkowski-like discussive logics. In:
Beziau, J.-Y., Chakraborty, M., Dutta, S. (eds.) New Directions in Paraconsistent Logic, pp. 213–228.
32. Parsons, S., McBurney, P.: Argumentation-based dialogues for agent coordination. Group Decis.
Negot. 12, 415–439 (2003)
33. Prakken, H.: Models of persuasion dialogue. In: Argumentation in Artificial Intelligence, pp. 281–
300. Springer, Dordrecht (2009)
34. Priest, G., Tanaka, K., Weber, Z.: Paraconsistent logic. In: Zalta, E.N. (ed.) The Stanford Encyclopedia
of Philosophy. Metaphysics Research Lab, Center for the Study of Language and Information,
Stanford University (Spring 2015 Edition). http://plato.stanford.edu/archives/spr2015/entries/logic-
paraconsistent/
35. Reed, C., Walton, D.: Towards a formal and implemented model of argumentation schemes in agent
communication. Auton. Agent. Multi-Agent Syst. 11(2), 173–188 (2005)
36. Searle, J.R., Vanderveken, D.: Foundations of Illocutionary Logic. Cambridge University Press,
Cambridge (1985)
37. Szałas, A.: How an agent might think. Log. J. IGPL 21(3), 515–535 (2013)
38. van Linder, B., van der Hoek, W., Meyer, J.-J.Ch.: Actions that make you change your mind. In:
KI-95: Advances in Artificial Intelligence. LNCS, vol. 981, pp. 185–196. Springer, Berlin (1995)
39. Vasyukov, V.: A new axiomatization of Jaśkowski’s discussive logic. Log. Logical Philos. 9, 35–46
(2001)
40. Vitória, A., Małuszyński, J., Szałas, A.: Modeling and reasoning with paraconsistent rough sets.
Fundam. Inform. 97(4), 405–438 (2009)
41. Walton, D.: Types of dialogue and burdens of proof. In: Proceedings of COMA’10: Conference on
Computational Models of Argument, pp. 13–24. IOS Press (2010)
42. Walton, D.N., Krabbe, E.C.W.: Commitment in Dialogue: Basic Concepts of Interpersonal Reasoning.
State University of New York Press, Albany (1995)
43. Zeman, J.J.: The deduction theorem in S4, S4.2, and S5. Notre Dame J. Formal Log. 8(1–2), 56–60
(1967)
B. Dunin-K˛eplicz
Institute of Informatics, University of Warsaw, Warsaw, Poland
A. Powała
(Formerly Alina Strachocka) Institute of Informatics, University of Warsaw, Warsaw, Poland
A. Szałas ()
Institute of Informatics, University of Warsaw, Warsaw, Poland
Department of Computer and Information Science, Linköping University, Linköping, Sweden
Czesław Lejewski: Propagator
of Lvov-Warsaw Ideas Abroad
Peter Simons
Abstract Czesław Lejewski studied in Warsaw before the Second World War, after
which he settled in England and resumed an academic career, becoming Professor
of Philosophy in Manchester. His writings, all articles, continue and extend the ideas
of his teachers, especially Stanisław Leśniewski in logic and Tadeusz Kotarbiński in
metaphysics.
Keywords Czesław Lejewski · Stanisław Leśniewski · Protothetic · Ontology ·

Mereology · Reism
Mathematics Subject Classification (2000) Primary 01A70; Secondary 03B05, 03B15,

03B65
1 Life
Czesław Lejewski was born in Minsk in the Russian Empire, on 14 April 1913. In 1920 his
family moved to Lublin, where he attended Gimnazjum. From 1931 he studied Classics
at Warsaw University, where he obtained a master’s degree in 1936 with a dissertation
on tropes in the sceptics. After military service he returned to the university in 1937 to
study for a doctorate in Classics, concentrating on ancient logic. This interest drew him
to courses and seminars on logic given by Jan Łukasiewicz and Stanislaw Leśniewski,
and philosophy with Tadeusz Kotarbiński. His dissertation De Aenesidemi Studiis Logicis
was examined and passed by the examiners, including Łukasiewicz, but he was unable
to take his degree due to the outbreak of war. He was taken prisoner by the invading
Soviets and spent 2 years in terrible conditions in Soviet labour camps, before joining
the 2nd Polish Army Corps under General Władysław Anders after the Nazi invasion
of the USSR. In 1942 he made his way by land and sea from Russia to Britain via Iran,
Iraq, Palestine, Egypt, South Africa, South America, and the USA. Lejewski joined Polish
military intelligence in London. After the war he taught English to Polish ex-servicemen
in London, marrying an English woman in 1949, and remarrying in 1973 after the death
of his first wife. He became a British citizen in 1955.

500 P. Simons
The communist takeover in Poland decided Lejewski to make his home in England,
and he took up doctoral studies again, under the supervision of Karl Popper at the London
School of Economics, passing (with Łukasiewicz as examiner again, flown specially from
Dublin) with a dissertation Studies in the Logic of Propositions in 1954. In 1956 he
joined the Philosophy Department at the University of Manchester, where in 1966 he
succeeded Arthur Prior as professor, and he remained there until his retirement in 1980.
He was visiting professor at Notre Dame University (USA) in 1960–1961 and at Salzburg
(Austria) in 1984. He never returned to his native land, even after the end of communist
rule. After his retirement he lived quietly in Manchester, and died after a long illness on 9
July 2001. His scientific books and posthumous post-war papers are housed in the Special
Collections Department of the library at the University of Leeds.
2 Main Works
Lejewski published only papers, of which probably the most influential was his 1954
article “Logic and Existence” [1], in which he opposed the standard understanding of
quantification, coming from Tarski and Quine, with the Leśniewskian understanding
in which he had been trained. The difference turns on the fact that in the standard
interpretation, which Lejewski called restricted, only denoting names (terms) may be
substituted for bound variables in the rules of inference, whereas in the unrestricted
interpretation also non-denoting names could be so substituted. Another quite widely
quoted piece is his 1958 article “On Leśniewski’s Ontology” [2], which is the clearest
exposition of this, Leśniewski’s central logical system. To aid understanding of the
system, which in contrast to standard predicate logic includes not only singular and empty
terms but also plurally referring terms, Lejewski introduced modified Euler diagrams for
representing logical relations among the extensions of terms. He also defined numerous
ontological functors and showing that partial inclusion (“Some a are b”) can serve as the
sole primitive functor. A particular tour de force among Lejewski’s historical articles is the
long encyclopedia entry “History of Logic”, written for the 15th edition of Encyclopedia
Britannica (1975) [3], which gives evidence of Lejewski’s remarkable knowledge of the
history of logic.
3 Views
Lejewski’s views are based on and extend the logical and philosophical views of
his principal teachers: in logic, Stanisław Leśniewski; and in philosophy, Tadeusz
Kotarbiński. His logical systems are all either reformulations, adaptations, or extensions
of Leśniewski’s ideas, while his fewer philosophical articles are either defences of
Leśniewski against criticism, or comparisons of Leśniewski with other approaches, and
in one case an elaboration of Kotarbiński’s materialistic reism.
Lejewski’s published papers fall into several thematically connected groups. One group
expound or simplify Leśniewski’s major logical systems of mereology, ontology and
protothetic. A second group offer alternatives and extensions to Leśniewski’s systems,
often prompted by criticisms of the artificiality of the latter. A third group of articles reflect
Czesław Lejewski: Propagator of Lvov-Warsaw Ideas Abroad 501
on the philosophical implications of logic from a Leśniewskian standpoint. A fourth group

examine miscellaneous aspects of logic, while a fifth group cover topics in the history of
ancient and contemporary logic. We treat the last two groups together.
Lejewski did not start to publish until he was forty, and despite his Polish origins, most
of his published papers are in English or are translations from his English.
3.1 Exposition and Simplification of Leśniewski
Lejewski’s early articles on mereology propose minor simplifications or new axiomatiza-

tions of Lesniewski’s theory, based on single axioms. One article discusses atomless and
atomistic extensions to mereology [4]; another shows that mereology is consistent relative
to protothetic [5]. Lejewski modified Leśniewski’s mereological terminology to make
it more understandable for English speakers, replacing ‘ingredient’, ‘part’ and ‘class’
respectively by ‘part’, proper part’ and ‘complete collection’. In ontology, in addition to
the expository 1958 article, Lejewski showed that Boolean Algebra in the axiomatization
of Ernst Schröder could be understood as elementary ontology (quantifying only nominal
variables), and with nominal definitions can be based on a single axiom for weak inclusion
(“Any a are b”) [6]. Lejewski also showed, in work parallel to that of Jerzy Słupecki, how
to bridge the gap between Aristotle’s syllogistic in Łukasiewicz’s modernized form and
the elementary Ontology of Leśniewski [7].
3.2 Alternatives and Extensions to Leśniewski
Whereas Leśniewski’s Ontology treats all names, whether singular, empty or plural, as
belonging to a single category, natural languages tend to use only singular proper names,
which may be empty. To accommodate this, Lejewski proposed a system of what he called
non-reflexive identity, employing only singular or empty names, and based on identity
sentences a=b which are only true if both a and b exist and are the same individual
[8]. Another and more radical innovation by Lejewski follows an idea of Kazimierz
Ajdukiewicz and introduces a special category of names for abstract classes, resulting
in what Lejewski called a bicategorial ontology [9]. Lejewski offered this system in an
ecumenical and ontologically neutral spirit, as offering a medium for those who believe
in classes (sets) to be able to talk with a Leśniewskian accent about such classes, but he
himself denied that there are abstract classes. In the l982 article “Ontology: What Next?”
[10], Lejewski goes beyond mereology to chronology, an ontological theory of temporal
objects, based on two primitive notions, one topological, that of an object’s being wholly
earlier than another object, and one metrical, that of an object’s having a shorter duration
than another object. He envisaged further extensions beyond chronology, to stereology, a
theory of objects extended in space as well as in time, and kinematics, a theory of objects
in motion through space, but though also outlined in a descriptive piece [11], a formal
development did not attain publication. It is clear that the extensions were conceived in a
Leśniewskian spirit.
502 P. Simons
Leśniewski based protothetic, his extended propositional logic, on material equiva-

lence and universal quantification, with definitions. In a part of his London dissertation,
later extracted and published as a paper in 1958 [11], Lejewski showed that material
implication, together with universal quantification and definitions, would also do the job.
It is a measure of the remarkable consistency of Lejewski’s work that over 30 years later,
in 1989, he published another paper showing how to fully formalize the 1958 system
using Leśniewskian terminological explanations [12].
3.3 Philosophical Reflections on Leśniewskian Logic
Lejewski not only expounded and extended Leśniewski’s views, but defended them
against all criticisms. This is apparent in the 1954 article on quantification. One frequently
made criticism of Leśniewski’s extreme extensionalism was that an extensional logic
such as his cannot adequately handle belief sentences such as ‘John believes that snow is
white’, and other intensional contexts. Lejewski’s account of belief contexts [13] is close
to that of Davidson’s paratactic view. Lejewski also reflected on the relationship between
natural language and the idealized languages of Leśniewski and other logicians: the
clearest statement of his view that a mutual give and take is required between natural and
artificial languages can be found in the 1979 article “Idealization of Natural Languages
for the Purpose of Logic” [14]. Though it does not purport to discuss Leśniewski
directly, Lejewski’s 1976 article “Ontology and Logic” [15] expounds a conception of
the relationship between those two disciplines which is purely Leśniewskian, denying
that either logic as a science or quantification as a device carries any ontological import
whatever.
3.4 Miscellaneous and Historical Pieces
Among Lejewski’s other pieces is another on propositional calculus derived from his
London dissertation [16], examining the groups of truth-functions that can be taken
as severally independent and jointly sufficient for functional completeness. Lejewski
honoured (and corrected) his London Doktorvater in the large commemorative volume
on Sir Karl Popper with a paper, “Popper’s Theory of Formal or Deductive Inference”
[17], which sympathetically reconstructs and revises Popper’s faulty attempts to reform
deductive logic.
In history of logic, apart from the magnificent Britannica article, Lejewski wrote two
papers on Theophrastus’ concept of prosleptic syllogisms [18, 19], a short Italian memoir
on Leśniewski and his systems [20], and an Arabic paper on Łukasiewicz. When the latter
died in 1956, Lejewski, at the request of the widow Regina Łukasiewicza, performed the
service of seeing the second edition (1957) of the classic monograph Aristotle’s Syllogistic
through the press, after Łukasiewicz had became too ill to complete the editing and proof-
reading himself. It was also through Lejewski that the bulk of Łukasiewicz’s post-war
letters and manuscripts came to be deposited at the John Rylands Library in Manchester.
Czesław Lejewski: Propagator of Lvov-Warsaw Ideas Abroad 503
Lejewski’s most straightforwardly philosophical piece is his 1976 article “Outline

of an Ontology” [21]. All that exists according to Lejewski are bulky and temporally
extended material bodies, and nothing else. His was thus a four-dimensionalist materialist,
a nominalist, and an atheist. The article denies that there are mereological atoms,
objects without proper parts, where Kotarbiński, whose views Lejewski otherwise closely
follows, had been agnostic about whether there are atoms or not.
Lejewski was also an assiduous reviewer of books and articles for various journals,
concentrating naturally but not exclusively on works connected with Polish philosophy
and logic.
4 Influence, Teaching, Personality
Because of the narrow scope of his interests, Lejewski’s influence was confined to
a small circle of logicians, mostly those interested in Leśniewski, such as his fellow
Leśniewski student Bolesław Sobociński, the New Zealand logician Arthur Prior, the
British philosopher Peter Geach, the historian of medieval logic Desmond Paul Henry,
and in mereology, the present author. In retirement Lejewski was pleased to act as mentor
to Manchester doctoral student Audoënus Le Blanc, whose work on mereology and
protothetic he considered an advance on his own. There is however very little secondary
literature on Lejewski’s work, and he did not aspire to a wide following, accepting with
equanimity the esoteric nature of his interests, but he inspired respect, most strikingly that
of Prior, among those who shared his interests and his conviction of the importance of
formal precision in philosophy.
As a teacher of undergraduates, Lejewski kept close to classical texts, from Aristotle to
Russell, though in advanced seminars he would venture into more technical work, proving
theorems on the board at a steady pace. His rather dry style and ontological asceticism
were not popular with students accustomed to the headier delights of existentialism or
Marxism, and his polite, somewhat old-fashioned reserve was considered aloof, but away
from the lecture room he could be outgoing and amusing. Friends in the local community
in Manchester knew him as ‘Czek’. He had no children, but he and both his wives were
fond of Shetland collie dogs, and were welcoming to visitors at their home in Cheadle
Hulme.
Works by Czesław Lejewski Mentioned in the Paper
1. Logic and existence. Br. J. Philos. Sci. 5, 1–16 (1954)

2. On Leśniewski’s ontology. Ratio 1, 150–176 (1958)
3. History of Logic. Encyclopaedia Britannica, vol. 23, 15th edn., pp. 234–250. Chicago (1974)
4. A contribution to the study of extended mereologies. Notre Dame J. Formal Log. 14, 55–67 (1973)
5. Consistency of Leśniewski’s mereology. J. Symb. Log. 34, 321–328 (1969)
6. Studies in the axiomatic foundations of boolean algebra. Notre Dame J. Formal Log. 1, 23–47, 91–106
(1960); 2, 79–93 (1961)
7. Aristotle’s syllogistic and its extensions. Synthese 15, 125–154 (1963)
504 P. Simons
8. A theory of non-reflexive identity and its ontological ramifications. In: Weingartner, P. (ed.)
Grundfragen der Wissenschaften und ihre Wurzeln in der Metaphysik, pp. 65–102. Pustet, Salzburg
(1967)
9. A system of logic for bicategorial ontology. J. Philos. Log. 3, 265–283 (1974)
10. Ontology: what next? In: Leinfellner, W., et al. (eds.) Language and Ontology. Proceedings of the 6th
International Wittgenstein Symposium, pp. 173–185. Hölder-Pichler-Tempsky, Vienna (1982)
11. On implicational definitions. Stud. Logica 8, 189–211 (1958)
12. Formalization of functionally complete propositional calculus with the functor of implication as the
only primitive term. Stud. Logica 48, 479–494 (1989)
13. Propositional attitudes and extensionality. In: Morscher, E., Neumaier, O., Zecha, G. (eds.) Philoso-
phie als Wissenschaft, pp. 211–228. Comes, Piding (1981)
14. Idealization of ordinary language for the purposes of logic. In: Alleton, D.J., Carney, E., Holdcroft,
D. (eds.) Function and Context in Linguistic Analysis, pp. 94–110. Cambridge University Press,
Cambridge (1979)
15. Ontology and logic. In: Körner, S. (ed.) Philosophy of Logic, pp. 1–28. Blackwell, Oxford (1976)
16. A propositional calculus in which three mutually undefinable functors are used as primitive terms.
Stud. Logica 22, 17–50 (1968)
17. Popper’s theory of formal or deductive inference. In: Schilpp, P.A. (ed.) The Philosophy of Karl
Popper, pp. 632–670. Open Court, La Salle (1974)
18. On prosleptic syllogisms. Notre Dame J. Formal Log. 2, 158–176 (1961)
19. On prosleptic premisses. Notre Dame J. Formal Log. 17, 1–18 (1976)
20. Ricordando Stanislaw Lesniewski [Remembering Stanisław Leśniewski]. Quaderni del Centro Studi
per la Filosofia Mitteleuropea 1, 1–46 (1989)
21. Outline of an ontology. Bull. John Rylands Univ. Library Manchester 59, 127–147 (1976)
22. Logic, ontology and metaphysics. In: Shanker, S.G. (ed.) Philosophy in Britain Today, pp. 171–197.
Croom Helm, London (1986) (Polish translation: Logika, ontologia, metafizyka. Filozofia Nauki 1,
15–35 (1993))
P. Simons ()
Department of Philosophy, Trinity College Dublin, Dublin, Ireland
Adolf Lindenbaum, Metric Spaces
and Decompositions
Robert Purdy and Jan Zygmunt
Abstract This paper revisits the life of Adolf Lindenbaum in light of new research
findings, then looks at two areas among many—metric spaces, and decompositions of
point sets—where his work has been underappreciated.
Keywords Adolf Lindenbaum · Biography · Bibliography · Metric spaces ·

Monomorphy · Decomposition · Warsaw school of mathematics · Wacław Sierpiński
AMS Subject Classification (2000) Primary 01A60, Secondary 01A70, 01A72, 51-03,
54-03, 54A05
1 Introduction
“I think your reasoning is really interesting for its simplicity and effective character.
I would just like to make one remark (which I’ve already mentioned to Erdős). As I
recall, Adolf Lindenbaum had a more general result—a proof of the existence of
sets not equivalent by countable decomposition in relation to an arbitrary family of
bijective transformations—not necessarily a family of isometric transformations (and
perhaps even a more general result for arbitrary cardinal numbers). I do not remember the
proof at all, and also I do not remember whether the family of bijective transformations
was subjected to some additional assumptions. I am under the impression (but can be
completely mistaken) that Lindenbaum announced his result without proof either in an
article in Fundamenta Mathematicae or in the reports of talks in the Annales de la
Société Polonaise de Mathématique. In any event it would be worthwhile to reconstruct
and announce the result. Overall it seems to me that there is an obligation to mathematics
and to the memory of Lindenbaum to encourage people to become acquainted with what
Lindenbaum left behind in print and to publish proofs of results [that he] supplied without
proof.”
Dedicated to Mariusz Pandura, in gratitude for his tireless research efforts.

506 R. Purdy and J. Zygmunt
The paragraph above is from a letter Alfred Tarski wrote to Wacław Sierpiński on
October 30th, 1946,1 discussing a pre-publication copy of Sierpiński [1947], a paper
which strengthened a result from Erdős [1943]: using the axiom of choice, Erdős had
proved (page 644) that for every cardinal there is a family of sets of real numbers,
no two of which can be decomposed into disjoint mutually congruent subsets. Citing
something he thought he recalled Tarski telling him, Erdős had credited Lindenbaum
for announcing, without proof, a less general version of that theorem. Sierpiński [1947]
obtained a more general version of Erdős’s 1943 result, without using the axiom of choice.
However, Erdős must have been misremembering what Tarski had told him. Tarski in fact
believed that Lindenbaum had achieved an even more general result than Sierpiński’s,
much earlier, and without the axiom of choice.
So do we (see the concluding paragraphs of our §4 below), though we cannot find any
published reference to it apart from Tarski’s 1946 letter to Sierpiński. Moreover, we agree
wholeheartedly with Tarski’s judgment that “there is an obligation to mathematics and to
the memory of Lindenbaum to encourage people to become acquainted with what Linden-
baum left behind in print and to publish proofs of results [that he] supplied without proof.”
It is the object of the present article to provide just such encouragement.
2 A Short Life
Lindenbaum, Adolf (1904–1941); Polish-Jewish mathematician and logician; docent of

the University of Warsaw; member of the Warsaw school of mathematics and the Warsaw
school of mathematical logic; early supporter of Fundamenta Mathematicae; co-founder
of the Polish Logical Society; Alfred Tarski’s closest collaborator of the inter-war period;
logical positivist and member of the Vienna Circle; member of the International Unity
of Science movement; anti-war campaigner; Polish Communist Party activist; Holocaust
victim (Figs. 1 and 2).
Fig. 1 Adolf Lindenbaum,

September, 1922
1 More precisely, it is from our translation of Tarski’s letter. The original letter was in Polish. McFarland–
McFarland–Smith [2014] offer their own translation on pp. 377–379, very close to ours, but with two
material differences: “efficiency” instead of “effective character”, and “two” instead of “ ”. We make
some changes to parts of their footnote 82 (ibid., page 378), while adopting other parts of it verbatim.
Adolf Lindenbaum, Metric Spaces and Decompositions 507
Fig. 2 Adolf Lindenbaum,

November, 1927
Fig. 3 Stefanja Lindenbaum.

Photograph from her
university student “indeks”,
October, 1926
Adolf Lindenbaum was born 12 June 1904 in Warsaw, the son of Mowsza Henoch aka
Maurycy Henryk (1878–1932) and Emilja née Krykus (1875–1939 or later). He had a
younger sister Stefanja (Fig. 3), born 22 March 1908.2
Mowsza was a businessman. On Adolf’s birth certificate he described himself as a
“ÒÓËÍÃÜËÍÝ”—an accordion word that can mean shop clerk, sales assistant, steward,
purser, branch manager, manager, director, superintendent, majordomo, or overseer. We
surmise that Mowsza’s father owned several businesses and put Mowsza in charge of one
or more of them. Soon afterward Mowsza switched to describing himself as a “kupiec”—
which simply means businessman. On a 1924 document he is named as one of the officers
of the Jewish Businessmen’s Mutual Assistance Society.
Adolf’s sister Stefanja entered pensja dla dziewczat˛ Pauliny Hewelkówny3 in 1917
and matriculated in 1926. She was accepted into the Faculty of Law at the University of
Warsaw on 2 September 1926. She attended all three trimesters of the 1926/27 academic
2 Some of this article draws on Marczewski–Mostowski [1971] and in several places is a straightforward
translation of that dictionary entry—a debt which the present authors are keen to acknowledge up front.
As well, some of this article overlaps with the paper Zygmunt–Purdy [2014], where readers will find a
more detailed treatment of Lindenbaum’s university student years, professional life and participation in
congresses. That said, much of the present material is new: some of it even overturns parts of Zygmunt–
Purdy [2014].
3 A private school for girls, at that time on the corner of Marszałkowska and Sienkiewicza (ul.
Marszałkowska 122). In 1919 the school was nationalized and renamed Państwowe Gimnazjum Żeńskie
im. Klementyny z Tańskich Hoffmanowej.
year, but did not sit any exams, and she formally withdrew from the university on 31
August 1927.
There is some slight evidence hinting at two more Lindenbaums in the household—an
S. Lindenbaum aka Z. Lindenbaum (possibly a younger sibling of Mowsza Henoch’s),
born 11 December 1886; and an M. Lindenbaum, born sometime in 1912—both of whom
appear to have been registered at the family’s home address in the 1930s and to have
emigrated to England either during or shortly after the Second World War.4
The family’s financial circumstances were boom and bust. Mowsza was in the movie
distribution and movie-theater franchising, leasing and financing businesses. From 1926
he was co-owner and general manager of Spółka Kinematograficzna “Kolos” (in Warsaw)
and “Kolos Małopolski” (in Kraków). In the later 1920s, Stefanja was brought into several
of the businesses as a co-owner and board member, and Mowsza stepped down from some
of their boards of directors in favour of his daughter. He ventured into movie production
in 1931 with a production company called Towarzystwo Kinematograficzne “Tempofilm”,
with which he had at least one box-office success that we know of.
Then he ran into financial difficulties. “Kolos Małopolski” went into bankruptcy,
destroying some 50,000 złotych of Mowsza’s net worth.5 He took out some very large
loans, and was unable to pay them back. He was last seen alive on 27 December 1932.
His body was recovered from the Vistula River next spring when the ice melted.
“Tempofilm” was legally dissolved on 26 January 1934. Some of the other businesses
survived. Records show that Stefanja remained a shareholder throughout most of the
1930s. But shareholders outside the Lindenbaum family were also recorded, and it is
not known if Stefanja held a controlling interest, or if the businesses were profitable after
Mowsza’s death, or if any of his erstwhile creditors had claims on the earnings.
There is no record of the elementary school Adolf Lindenbaum attended. In his first
year of secondary school, 1914–1915, he attended gimnazjum Rocha Kowalskiego, and
then from 1915 to 1922 gimnazjum Michała Kreczmara. In 1922 he entered the University
of Warsaw, and on 22 June 1928 was awarded a Ph.D. for a thesis titled “O własnościach
metrycznych mnogości punktowych” <On the metric properties of point sets>, written
under the supervision of Wacław Sierpiński.
Lindenbaum’s Ph.D. diploma reads, “primum in mathematica, deinde in physica
et in philosophia,” but in his bio-bibliography for Erkenntnis he added a telling
qualification: “Hauptfach–Mathematik; Nebenfächer–Philosophie, Experimentalphysik.”
In other words, not simply physics, but experimental physics—involving measurement.
He took courses in descriptive geometry and projective geometry, dealing with
transforms and invariants, and in mathematical astronomy and astrometry, Maxwell’s
4 At the present time this cannot be substantiated, as the Polish National Archives have “masked” the
relevant records, and the U.K. National Archives have “closed” them for 100 years, on the grounds
that they “contain sensitive personal information which would substantially distress or endanger a living
person or his or her descendants”. The “slight evidence” consists of scanned pages from the registry of
residents of the building at ulica Złota 45 inadvertently revealing some details that are less-than-perfectly
masked, and U.K. naturalization records HO 405/33558 and HO 405/33790.
5 Although purchasing power comparisons across 80 years are dicey, he is thought to have lost between
350,000 and half a million in 2009/2010 PLN zł.

equations, Planck radiation and Lichtquanta.6 He knew of Hendrik Lorentz’s group-

theoretic treatment of frame transforms and their invariants, and the role they played
in the geometrization of space-time. He wrote his thesis in the wake of the Banach-
Tarski paradox, which had possibly stirred some philosophical misgivings in him about
measures in physics—Hausdorff, Dirac, Liouville-Hamiltonian, Lorentz. With a nod to
the Vienna Circle and verificationism he took to using the term metrologia (metrology)
in the titles of some of his courses. Together with Edward Szpilrajn (Marczewski) he
co-authored an encyclopedia entry on measurement in geometry, aimed at a general
audience; at the bottom of the entry the reader is urged to consult a Polish translation of
Hermann von Helmholtz’s “Zählen und Messen”, which emphasizes the empirical nature
of measurement.7 Today the term “metrology” can be encountered in the literature on the
Hausdorff dimension of space-time.
Lindenbaum was a keen participant in student academic organizations and societies.
Through his papers, reviews, comments and personal contacts he exerted a strong
influence on younger mathematicians. Together with Alfred Tarski he became an active
contributor to two schools of scientific inquiry: the Warsaw school of mathematics, under
the intellectual influence of Wacław Sierpiński and Stefan Mazurkiewicz, and the Warsaw
school of mathematical logic, principally under the leadership of Jan Łukasiewicz and
Stanisław Leśniewski. He also kept in close contact with the Lwów school of mathematics
through Stefan Banach, Hugo Steinhaus and Stanisław Ulam, and two of his publications
show him encouraging and rising to the defense of a doctoral student of Steinhaus called
Sala Weinlös.
He published mainly in Fundamenta Mathematicae. His first major paper, “Contribu-
tions à l’étude de l’espace métrique. I.”, appeared in FM, volume 8 (1926), pp. 209–222.
Written while he was in his fourth year and loaded down with course work, it was a
remarkable accomplishment. Later he used it as part of his doctoral dissertation.
His second major paper, “Communication sur les recherches de la théorie des ensem-
bles,” Sprawozdania z posiedzeń Towarzystwa Naukowego Warszawskiego, Wydział III,
volume 19 (1926), pp. 299–330, was co-authored with Alfred Tarski. It was a monumental
work, setting out 30 pages of new results in general set theory and the arithmetic of
transfinite ordinals all presented without proofs. Some were results obtained jointly by
Lindenbaum and Mojżesz Dawid Kirszbraun, a classmate of his at gimnazjum Kreczmara.
Most came from work Lindenbaum and Tarski had undertaken prior to 1926 on the theory
of one-to-one transforms.
Despite the value and sheer multitude of his contributions to set theory, cardinal and
ordinal arithmetic, the axiom of choice, the continuum hypothesis, theory of functions,
measure theory, point set topology, geometry and real analysis, Lindenbaum’s name
continues to be associated principally with his work in mathematical logic, a field which
in the 1920s was not yet widely developed.
6 The courses “Teorja Promieniowania” and “Promieniowanie i Kwanty” were taught by Czesław
Białobrzeski. Mathematical astronomy and astrometry were taught by Michał Kamieński.

7 See [35], page 595.
His most important result in logic was his conjecture that any propositional calculus
can be characterized by a denumerable (i.e., finite or at most countably infinite) matrix.
He never published this conjecture. Its first appearance in print was by Łukasiewicz and
Tarski in 1930. Jerzy Łoś first published a full proof of it in 1949. Lindenbaum conceived
a method of constructing matrices by using the expressions of a propositional calculus
(or more strictly speaking, equivalence classes of expressions) as elements of the matrix.
For logicians in the 1920s, this idea was a revelation. Lindenbaum’s method spawned
waves of research, and it became generally accepted practice to refer to such algebras as
Lindenbaum algebras.
His second most important result, widely known among logicians as “Lindenbaum’s
Lemma,” was his theorem, framed in the terminology and concepts of Tarski’s <S,Cn>
methodology of deductive systems, that every Cn-consistent set of sentences in a language
S can be extended to form a Cn-consistent and Cn-complete deductive system in S. Or,
more loosely put, that every consistent theory, formulated in a suitable language and
assuming a suitable underlying logic, has a complete (maximal) and consistent extension.
Again, as with much of Lindenbaum’s legacy, he never stated it in print. Its statement and
proof were first published by Tarski (see, respectively, Tarski [1928] and Tarski [1930]),
who scrupulously attributed both the idea and the proof to Lindenbaum. The “Lemma”
(i.e., theorem) quickly became an essential tool in every logician’s toolkit. Some writers
have even ventured that Lindenbaum maximalization, with the notions of completeness
and consistency defined as in Tarski [1928], is the only essential thing that all logics have
in common.
Lindenbaum was the co-author (with Tarski) of a 1936 paper proving that all the
logical notions of Russell and Whitehead’s Principia Mathematica are invariant under
one-to-one transformations (automorphisms) of the domain of discourse of the model
onto itself: “Über die Beschränktheit der Ausdrucksmittel deduktiver Theorien”—a paper
which anticipated and did most of the heavy lifting for Tarski’s much later work on
“What are Logical Notions?” The results they obtained in this paper had many and various
applications, among other things to the foundations of geometry, to discussions of which
mathematical concepts count as purely logical and which as specifically mathematical,
to the study of the interdependence of primitive concepts in axiomatic systems, and
ultimately to the study of the independence of the axiom of choice. Lindenbaum and
Andrzej Mostowski documented their findings on this last problem in 1938 in the joint
paper, “Über die Unabhängigkeit des Auswahlaxioms und einiger seiner Folgerungen.”
Lindenbaum was an adept crossover artist, equally at home pursuing the program of
the axiomatists (David Hilbert, Bertrand Russell, the Italians Giuseppe Peano, Alessandro
Padoa and Mario Pieri) to reduce mathematics to language, truth and logic, and the
program of the algebraists (quintessentially Tarski) to reduce language, truth and logic
to mathematics. The same whiff of circularity, or more charitably coherentism, can be
found today in the interplay between model theory and proof theory. In his published
works and public lectures Lindenbaum concentrated on large themes, fundamental issues,
general concepts and synoptic solutions. He sought, throughout all, to apply whatever
means necessary to achieve the clearest possible understanding of the underlying reality
of things.
Lindenbaum was an adherent of Logical Empiricism. In March, 1930, he spent
time in Vienna where he met Rudolf Carnap, Herbert Feigl, Carl Hempel, Abraham
Fraenkel and Samuel and Lilian Broadwin. Later in the 1930s he participated in and
contributed to the International Unity of Science movement. He played an active role
in the movement’s founding congresses in Prague (Aug 31st–Sept 1st, 1934) and Paris
(Sept 15th–23rd, 1935), and he corresponded with Otto Neurath and Jørgen Jørgensen
on detailed arrangements for its next two congresses, in Copenhagen (June 21st–26th,
1936), and again in Paris (July 29th–31st, 1937). Neurath twice solicited Lindenbaum’s
bio-bibliography, in 1930 and again in 1934, for inclusion in his surveys (published in
Erkenntnis) of who’s who in the movement. In his article “After 6 years”, Synthese,
vol. 5, no. 1/2 (May–June, 1946), pp. 77–82 [and date-lined Oxford, December 19th,
1945, 3 days before his death], Neurath states outright that Adolf Lindenbaum had been
a member of the original Vienna Circle.
In a letter dated July 1st, 1935, Neurath invited Lindenbaum to speak on the subject
of formal simplicity (die formale Einfachheit) at the 1935 Paris congress. Lindenbaum
obliged with a lecture of the same title, which he gave in German on the morning of
September 18th in Room 1 (Vormittag, 18 September, Saal I). Two days later, on the
afternoon of September 20th, a debate took place in the same room on the question
of standardizing logical symbolism (Aussprache über Vereinheitlichung der logischen
Symbolik). The debate concluded by agreeing to establish a working committee charged
with advising on and promoting the international standardization of logical symbolism.
Lindenbaum was appointed to this committee.
Records show that Lindenbaum expected to take part in the 1937 Paris congress (July
29th–31st, 1937) in his capacity as a member of this committee, but at the last moment it
emerged that he was unable to attend: he was denied a travel document to leave Poland. In
place of attending in person, he sent a letter which was read aloud to the congress on the
morning of July 30th, expressing his and other Polish logicians’ concerns with the interim
results of the committee (“meine Meinung aussprechen, wobei ich im Voraus bemerken
möchte, dass die polnischen Logiker verschiedentlich anderen Standpunkt einnehmen.”).
Lindenbaum is known to have been an asystent in Łukasiewicz’s Philosophical
Seminar in the faculty of mathematics and natural sciences from (at least) the fall of 1931.
Samuel Eilenberg recalled him being in charge of the library at that time.8 In 1934 he
8 See Eilenberg [1993], page 1. Eilenberg writes:

In the academic year 1930–31 I was a first-year student at the University of Warsaw, while Karol
Borsuk was an assistant conducting exercises in real analysis. I was a member of a class which
was huge but he soon started to notice me and we got involved in several conversations. In the
spring of 1931 he received his doctorate and I attended the ceremony. At the same time I attended
a course on set theory given by Docent Bronisław Knaster. There were two other students in the
course, however, I was the only one who did all the homework. I struck up a friendship with
Knaster which lasted as long as he did. Set theory naturally led to Topology, which in Warsaw
meant strictly Set-theoretical Topology.
I remember a curious incident. In the fall of 1931 I was browsing through the Mathematics Library
and I came across a book entitled Topology by Solomon Lefschetz. I looked at the bibliography
to see to what extent the “Polish School” was quoted. I found only one reference. It was a paper
of Knaster, Kuratowski and Mazurkiewicz in volume 15 of Fundamenta Matematicae containing
a combinatorial proof of the Brouwer Fixed Point Theorem. I was very surprised to find no other
references and I conveyed my concern to Dr. Adolf Lindenbaum (an excellent logician) who was
then the assistant in charge of the library. He told me that it was a terminological misunderstanding,
that the book was not about Topology but about some sort of algebra.
successfully defended a habilitation thesis. Its title is lost, but there are strong indications
it might have been the paper, “Z teorii uporzadkowania
˛ wielokrotnego” <Sur la théorie de
l’ordre multiple>, Wiadomości Matematyczne, vol. 37 (1934), pp. 1–35, on an extension
of Cantor’s notion of multiply-ordered sets. On February 1st, 1935, he started lecturing
as a docent of the University of Warsaw, and from the commencement of the 1935–1936
academic year he took up the position of adiunkt (assistant professor) in the Philosophical
Seminar.
Lindenbaum’s political sympathies were left-leaning, anti-fascist and anti-war, and
some of his political activities were illegal for the time. He attended the World Congress
Against War held in Amsterdam on 27th–29th August 1932. He belonged to the Polish
Communist Party from at least the mid-1930s up until it was disbanded in 1938 by Stalin,
and campaigned for it in intelligentsia circles. In 1936, as one of the “editors and co-
workers” of Głos Współczesny, he signed a petition to Professor Halvdan Koht of the
Nobel Committee in Oslo, urging that Karl von Ossietzky, a German political journalist
imprisoned by the Nazis, be awarded the Nobel Peace Prize.9 Together with many writers
and social activists of the day Lindenbaum added his name to an open letter “to the
workers of Lwów” expressing solidarity with “the proletariat’s protest against the bloody
massacre [of April, 1936] of workers fighting for jobs, bread and freedom”.10
Lindenbaum took a keen interest in pedagogy, and in the second half of the 1930s
he delivered various lecture series to teachers’ groups and organizations. He was also
said to have been interested in art, literature, hiking and mountain climbing. One of his
lecture series to the Polish Teachers Union included a lecture on the creative and “artistic”
elements in mathematics. It seems he enjoyed teaching beyond the ranks of his own pro-
fession, and especially teaching other teachers. It appears, too, that he was an avid solver
of newspaper chess problems: his name pops up time and again in the mid-1920s in ABC
Nowiny Codzienne and Nasz Przeglad ˛ for having solved the previous issue’s challenge
puzzle. He even published a mathematical paper, “Sur le «problème fondamental» du jeu
d’échecs,” Rocznik Polskiego Towarzystwa Matematycznego (=Annales de la Société
Polonaise de Mathématique), vol. 13 (année 1934, publ. 1935), pp. 124–125. At the same
time it was rumored he liked “having a good time”, being among people, frequenting
cafés and partying. Interestingly though, in group photos he was never front row center.
He seemed to prefer the third or fourth row, or somewhere over by a wall (Figs. 4 and 5).
Around the end of October or the beginning of November 1935 Lindenbaum married
Janina Hosiasson (Fig. 6), an established philosopher of logic, fellow member of the
Lwów-Warsaw school of mathematical logic, fellow graduate of the University of
Warsaw, and four-and-a-half years older than him. She also was named by Neurath as
9 The petition was issued in the name of the editorial staff, co-workers and “friends” of Głos Współczesny
<Contemporary Voice>, a left-leaning monthly newspaper with national circulation, and was splashed
across the front page of the March 1936 issue. The signatures of editorial staff and co-workers were
grouped separately from those of “friends”. Lindenbaum’s signature was included among editorial staff
and co-workers.
10 Published in Lewar, 15th May, 1936, no. 4, page 10. Lewar was a biweekly literary magazine sponsored
and influenced by the Polish Communist Party from 1933 through 1936. Its name was a play on words,
combining “leverage” and “leftist”.
Fig. 4 First congress of mathematicians from Slavic countries, Warsaw, 23 September 1929. Adolf
Lindenbaum, arms folded, wearing a blazer with a boutonnière in the lapel, is standing by the door
having been a member of the original Vienna Circle. During their engagement they stayed
for 2 or 3 months with Adolf’s mother and sister in the Lindenbaums’ family home at
Złota 45/4 before moving into their own apartment at Krasińskiego 16/34 on the 31st of
October, 1935.
It was, according to Antoni Marianowicz (Kazimierz Jerzy Berman), a marriage
of convenience for both of them.11 Janina was in love with Antoni Ludwik Pański
(1895–1942), philosopher, statistician, social activist and eldest son of the neurologist
Aleksander Pański (Fig. 7). Janina and Antoni had been living together, on again off
again, since Janina’s late teens. During this time Antoni had a string of affairs with other
women, and was briefly married (to a singer, Elza Aftergut), but his love for Janina was
apparently “the real thing”. Awkwardly, they were second cousins.12 Marianowicz writes
that their respective “aunts and uncles” pressured them to choose other marriage partners.
11 SeeMarianowicz [1995], pp. 230–231, or Marianowicz [1995] (1), pp. 192–193.

12 Janina’s mother Zofja Hosiasson, née Feigenblatt, and Antoni’s mother Róża Pańska, née Seidemann,
were first cousins—(Antoni’s maternal grandfather Adolf Seidemann and Janina’s maternal grandmother
Leona Feigenblatt, née Seidemann, were brother and sister).
Fig. 5 Meeting of the mathematics, physics and astronomy circles in Warsaw, 5 May 1932. Łukasiewicz,
Leśniewski and Tarski are seated in the front row. Adolf Lindenbaum is standing farther right, one or two
rows farther back, behind and looking out from between two gloved men in wool coats. It is our guess that
in the very back row, plumb above the toe of Leśniewski’s airborne boot, stands a 27-year-old Mojżesz
Presburger
Marianowicz does not spell out why he thinks it was a marriage of convenience
for Adolf. It is not clear if Adolf regarded the marriage the same way Janina did. To
all outward appearances they cohabited just fine, and seemed to care for one another.
Certainly they held each other’s professional abilities in high regard. They were fired by
the same progressive social ideals and political convictions. But was Janina the apple of
Adolf’s eye? . . . the “little man” ( )13 of his eye? We are indebted to Arie Hinkis
for his suggestion that Adolf was delicate, feminine, the exact opposite of Tarski; that he
exhibited none of Tarski’s machismo, competitiveness or ego; and that only Tarski could
have “extracted” from the young Adolf the prodigious achievements of [26] and [26a].
Perhaps Lindenbaum was under social pressures of a different kind.
13 The biblical Hebrew origin of today’s expression (see: Deuteronomy 32:10, Psalms 17:8, Proverbs 7:2,
Lamentations 2:18).
Fig. 6 Janina Hosiasson. Photograph from her university student “indeks”, October, 1919
Fig. 7 The three Pański brothers, from left to right: Jerzy, Antoni and Wacław (Solski), circa 1920
On September 6th, 1939, Adolf and Janina abandoned their apartment and all their
belongings and fled Warsaw on foot,14 heading (Janina’s letters to Otto Neurath and G.E.
Moore suggest) either due east in the direction of Siedlce, or south-east in the direction of
D˛eblin. Janina writes to Neurath and Moore that progress on foot was slow and that the
road was repeatedly strafed by Luftwaffe planes. A friend with a motorcycle encountered
them on the road and gave Janina a ride on the saddle behind him, leaving Adolf to
continue on foot. The motorcyclist managed to get Janina as far east as Rivne15 where
he left her before heading back alone in the direction of Warsaw. From Rivne, Janina
made her way, catch as catch can, partly by train and partly by road, to Vilnius. There
she eventually learned through friends and acquaintances that Adolf was in Białystok.16
14 German forces invaded Poland on September 1st and within days German artillery shells were raining
down on Warsaw. Janina and Adolf fled their home under fire.
15 Polish: Równe; Ukrainian: PiÅÐÈ; Russian: PÑÅÐÑ; Hebrew: ; Yiddish: —a name that
reverberates in Aliyah consciousness. According to Polish Wikipedia, in 1939 Rivne had a population
of 41,500 persons, slightly more than half of whom (21,000) were Jews. See: https://pl.wikipedia.org/
wiki/R%C3%B3wne.
16 On September 17th Soviet forces entered Poland and on September 22nd Białystok came under Soviet
occupation. We do not know if Lindenbaum was already in Białystok by then, or if he arrived after it was
Fig. 8 Members of the Department of Mathematics and Physics, Pedagogical Institute, Białystok, spring
of 1941. Left to right: Stanisław Romanowski, Samuel Steckel, Adolf Lindenbaum, Salomon (Szlama)
Lubelski, Henryk Ferencowicz, Edward Litwinowicz
Adolf and Janina tried writing to each other, without much success. Apparently, most of
their letters to each other were not allowed to get through—confiscated by one postal
authority or another. Janina visited Adolf in Białystok for a day, then she returned to
Vilnius without him. Her letters to Neurath and Moore suggested that she and Adolf had
“agreed to disagree” about where best to try to survive.17
Adolf found work in the Pedagogical Institute in Białystok as a mathematics teacher
and the head of its mathematics department (Figs. 8 and 9).18 Then on June 22nd, 1941,
Germany invaded the Soviet Union, and on the same day, the Vilnius Uprising began.19
Within days, German forces were in Białystok, and not much later in Vilnius. Sometime
in Soviet hands. It would have been a difference of only a few days. In either case it would have suited
him, as he was attracted by Soviet communism—or at least, by his imagined picture of it.
17 On September 19th, 1939, Soviet forces wrested the city of Wilno from Poland and on October 28th
re-attached it to its ancestral home of Lithuania, upon which they bestowed (notional) “independent
statehood”, a dubious arrangement which lasted only until August 3rd, 1940.
18 The Pedagogical Institute was a Soviet institution.
19 In Zygmunt–Purdy [2014], §1, p. 299, we wrote that on June 22nd, 1941, Germany “declared war”
on the Soviet Union. Germany of course did nothing of the sort. It simply invaded, without bothering to
observe such niceties as telling anyone what it was doing. We thank Piotr Wojtylak for pointing this out.
Fig. 9 Street view of the

Pedagogical Institute in
Białystok, 1940
before the beginning of July, 1941, Adolf came to Vilnius and stayed, possibly for about
6 weeks, in a small satellite community on the eastern outskirts of the city called Pavilnys
(Polish: Kolonia Wileńska).20 By coincidence Pavilnys was where Anna Borkowska,
aka Mother Bertranda, famously hid members of HaShomer HaTza’ir in her Dominican
convent. However, there is no indication that Adolf had any knowledge of this.
Why he came when he did, and indeed at all, and why he chose to stay in Pavilnys,
rather than with his wife in her apartment downtown,21 remains unclear. Perhaps he was
finally persuaded of the wisdom of trying to emigrate to the West, and hoped that Janina’s
20 We have this on the authority of Professor Bogusław Wolniewicz, who cites testimony of Professor
Maria Renata Mayenowa (born Rachela Gurewicz), from a conversation he held with her on 26 April
1986. See Wolniewicz [2015].
21 Remarks attributed to Oskar Lange suggest that Janina and Antoni Pański were living in the same
apartment in Vilnius at that time, which might explain why Adolf chose to live elsewhere.
contacts in Vilnius might help him do so, but was wary of putting her in danger by openly
associating with her. Sara Bender writes that under the Soviets there was a systematic
campaign of destruction and arrests in Białystok from May to June, 1941, cut short only
by the German invasion, and that a “fourth wave of arrests began on the night of June 20th,
1941, when members of the NKVD went from house to house with their lists, sending
entire families, most of them Jewish, in cattle and freight trucks to the Soviet hinterland.”
Then on June 22nd, 1941, “the bombing of Białystok sowed panic in the city. As the Red
Army began to flee, [ . . . ] anyone who could, fled with the Russians”.22
Possibly he summoned his sister Stefanja to join him in Vilnius from wherever she
had been hiding. Or possibly she was already staying with him in Białystok and simply
came with him. We don’t really know where she had been staying before this, or how
or why she came to Vilnius when she did. All we know is that they both showed up
together. Sometime before the middle of August, 1941, Adolf and Stefanja were arrested
and shot. The timing of their arrests and murders—Adolf and Stefanja together—suggests
that Stefanja was staying with Adolf in Pavilnys.23
There is some evidence to suggest that Janina finally did marry Antoni Pański . . . either
immediately upon learning of Adolf’s death, or sometime earlier. In September, 1941,
both Antoni and Janina were arrested—he first, she a week later. She had two passports
in her possession at the time of her arrest, one in the name of Janina Lindenbaumowa,
the other in the name of Janina Pańska. Possibly Adolf had given her a divorce. Or she
had committed bigamy. Or one or both of the passports, or supporting documents used to
obtain them, had been forged.
Jerzy Dadaczyński writes that Janina applied for an American visa.24 We know from
American sources25,26 that she repeatedly tried, to no avail, to be sponsored into the U.S.
by the Rockefeller Foundation’s New School for Social Research as a “refugee scholar”.
Rudolf Carnap, Oskar Lange, William Gruen, Ernest Nagel, Albert Hofstadter, Alfred
22 See Bender [1997] (1), pp. 87–90.

23 According to the Dédicace, Fundamenta Mathematicae, vol. 33 (1945), p.V, Adolf Lindenbaum was
shot in Naujoji Vilnia (Polish: Nowa Wilejka). This was where Soviet forces withdrew to on October 28th,
1939, after handing Vilnius over to a notionally independent Lithuania. Naujoji Vilnia is one train stop—
4 km—east of Pavilnys/Kolonia Wileńska, and larger. However it must be underlined that by August,
1941, German forces were in control of both places. Stefanja and Adolf were not shot by the Soviets.
They were shot either by Germans or by Lithuanian collaborators.
24 Dadaczyński [2003].
25 SUNY, University at Albany, Science Library 352, M.E. Grenander Department of Special Collections
and Archives, German and Jewish Intellectual Émigré Collection (GER-017), Series 4: individual files
from Else Staudinger, Director of the American Council for Émigrés in the Professions (ACEP), Box 3,
folder 147.
26 The Rockefeller Archive Center, the Rockefeller Foundation (RF) Archives collection, Record Group
2, RG2 1940, Series 200, Box 192, folders 1368 & 1369; and RG2 1941, Series 200, Box 212, folder
1487.
Fig. 10 Journal entry recording the death of prisoner #4658, Antoni Pański, on 9 January 1942 in
the prison hospital. Reproduced from Monika Tomkiewicz’s book, Zbrodnia w Ponarach 1941–1944.
Monografie Komisji Ścigania Zbrodni Przeciwko Narodowi Polskiemu, vol. 43. Instytut Pami˛eci
Narodowej: Warszawa 2008. In her book this is Fig. 24 of 51 figures on 17 unnumbered pages at the
end of the book. Reproduced by kind permission of the author and Instytut Pami˛eci Narodowej
Tarski, Henry S. Leonard, Herbert Feigl, Mason W. Gross, G.E. Moore, C.A. Baylis, Carl
Hempel, J.C.C. McKinsey, Sidney Hook, Willard V.O. Quine, Victor F. Lenzen and C.J.
DuCasse all wrote letters to the Rockefeller Foundation supporting her applications.27 In
the end, the New School decided it was willing to accept her application to enter the U.S.,
but could not itself provide financial support for her, or recommend that the Rockefeller
Foundation provide financial support. Dadaczyński also writes that “probably” her efforts
to help her husband attracted the attention of the Lithuanian authorities that led to her
arrest. What those efforts were, we do not know.
Antoni Pański died in Lukiškės prison28 in Vilnius on January 9th, 1942 while under
interrogation, probably tortured to death. His death certificate, issued by the prison
hospital staff and handwritten in Lithuanian,29 identifies the cause of death as “Širdies
raumens degeneracija” <heart muscle degeneration>, which probably just meant his heart
stopped beating (Fig. 10).
Janina’s friends engineered a prison break for her, but she fluffed it. In April, 1942,
after 7 months of imprisonment, she was taken to Paneriai (Polish: Ponary) and shot.
27 Lord Russell, three of whose books Janina had translated, and himself an “economic immigrant” in the
U.S. at that time, on a work visa, declined to support her, claiming that he did not recall her.
28 Lithuanian: Lukiškiu˛ tardymo izoliatorius kalėjimas. Polish: wi˛ ezienie na Łukiszkach, or simply
Łukiszki. It was on the same street—Gedimino—as Janina’s apartment.
29 The prison and its hospital were under German direction and control but staffed by Lithuanians.
3 Metric Spaces30
Metric spaces interested Lindenbaum from early on, as can be seen from:
• his first published paper, [26], “Contributions à l’étude de l’espace métrique, I”,31 later
incorporated into his doctoral dissertation
• §5: “Théorie des ensembles des points” of the paper [26a], “Communication sur les
recherches de la théorie des ensembles”, co-authored with Tarski. §5 sets out results
pertaining to the decomposability of sets and their congruence in metric spaces, some
of which results were obtained by Lindenbaum alone, some jointly with Tarski, and
some jointly with M.D. Kirszbraun
• his doctoral dissertation, “O własnościach metrycznych mnogości punktowych” <On
metric properties of point sets>, submitted in 1927 and defended in 1928 (though never
published in its entirety)
• the short note [29a a], lifted from his doctoral dissertation, summarizing his talk at the
First Polish Mathematical Congress in Lwów in 1927
• the short note [31a a], “La projection comme transformation continue la plus générale”
• the short note [31ab], “Sur les figures convexes”
• the lecture “Badania nad własnościami metrycznymi mnogości punktowych”, given
at the Second Polish Mathematical Congress in Vilnius in 1931 (listed as [311a] in
Zygmunt–Purdy [2014], p. 306)
• the paper [33a], “Sur les ensembles localement dénombrables dans l’espace métrique”
It seems certain Lindenbaum’s interest in metric spaces was sparked most of all by
Tarski32 and also considerably by Banach and Kuratowski.33 Lindenbaum had a ringside
30 Metric spaces and decompositions are only two of many areas where Lindenbaum’s work is
underappreciated. The constraints of the present publication limit our treatment to these two areas. Future
articles will consider his contributions to sentential logics, metalogic, the general theory of sets, and the
independence of the axiom of choice.
31 The Roman “I” in the title implies that a sequel was planned. In footnote 1, p. 214 (see also footnote
1, p. 218) Lindenbaum indicated what he intended the sequel to be about: it was going to be a study
of the notion of the equivalence of point sets by decomposition, in the sense of Banach and Tarski (see
also [26a], p. 327). In the event, no sequel was ever published. Nor was his doctoral dissertation ever
published in its entirety, although both he and Sierpiński expected it to be: they each referred to it as
“à paraître” (see [33a], p. 106, note 18; and Sierpiński [1936], p. 32). As there is no surviving copy in
Warsaw University’s archives, or anywhere else we know of, we are unsure precisely how his dissertation
related to his publications. Many authors, including Lindenbaum himself, credited his dissertation for
results that were never set out in any of his (other) publications (see, e.g., Aronszajn [1932], p. 99, note
12 ; Kirszbraun [1934], p. 78, note 4, and p.102; Sierpiński [1936], p. 32; Lindenbaum [33a], p. 106,
note 18).
32 Lindenbaum and Tarski’s collaboration on decomposability and congruence of point sets in Euclidean
and general metric spaces began as early as 1923, when Lindenbaum was a freshman/sophomore (see
[26a], p. 327).
33 In [26], p. 210, one reads, “Je termine cette préface par remercier MM. Kuratowski et Tarski, qui ont
bien voulu prendre intérêt à ces recherches: j’en ai profité beaucoup.” Recognition of Kuratowski’s impact
on [26] is also expressed on p. 216 in a parenthetical suffix to the statement of Théorème 7 (“C’est M.
Kuratowski qui a su généraliser de cette manière intéressante une idée de ma démonstration primitive du
th. 8”) and on p. 222, note 1 (“Le problème auquel le théorème (II) donne réponse m’a été posé par M.
Kuratowski.”)
seat at the gestation and birth of Tarski’s “O równoważności wielokatów” ˛ <On the
equivalence of polygons> and Banach and Tarski’s famous so-called paradox, “Sur la
décomposition des ensembles de points en parties respectivement congruentes”, the latter
of which explicitly mentioned a result obtained by Lindenbaum and thus constituted, in a
“proxy” sort of way, Lindenbaum’s first published result. (We cite this reference below.)
On the other hand Wacław Sierpiński, Lindenbaum’s PhD thesis supervisor, probably
did not exert a formative influence on Lindenbaum’s interest in metric spaces . . . at least
not at first. Only when as editor of Fundamenta Mathematicae he recognized good work
in [26] and agreed to guide Lindenbaum’s doctoral dissertation did Sierpiński begin to
exert an influence on the direction of his pupil’s inquiries. And in due course vice-versa:
by [33a] and Sierpiński [1933], their mutual influence on each other started to show.
In the mid-1920s when Lindenbaum began investigating metric spaces the theory was
still in its formative stages of development. Maurice Fréchet introduced the concept,
using different terminology, in his doctoral dissertation in 1906. Felix Hausdorff, in his
Grundzüge der Mengenlehre,34 was the first to use the expression metrischer Raum in
place of Fréchet’s terminology to denote such entities.35 Lindenbaum’s [26] can thus be
seen as one of its earlier systematic treatments.
We present a selection of Lindenbaum’s results from [26] and [33a]. To do this, we
introduce some necessary terminology.
A space M is a collection of undefined entities called points, and a metric space in the
sense of Fréchet and Hausdorff is a pair <M, ρ> consisting of a space M and a real-valued
non-negative function ρ on the cross product M × M satisfying the following conditions:
(M1) ρ(x, y) = 0 if and only if x = y (law of coincidence)
(M2) ρ(x, y) = ρ(y, x) (law of symmetry)
(M3) ρ(x, y) + ρ(y, z) ≥ ρ(x, z) (the triangle law)
The function ρ is called a distance function, and the number ρ(x, y) is called the
distance between points x and y. The triangle law (M3) is so called because of its formal
expression of Euclid’s Proposition I. 20: “Any two sides of a triangle are together greater
than the third side.” The three laws as formulated above are consistent and independent.
Lindenbaum observed (see [26], p. 211) that both the law of symmetry (M2) and
the stipulation that ρ be non-negative can be derived, and thereby dispensed with, by
formulating the triangle law d’une façon plus avantageuse. Thus he showed that two
independent axioms sufficed36,37 :
(ML1) ρ(x, y) = 0 if and only if x = y (law of coincidence)
(ML3) ρ(x, y) + ρ(x, z) ≥ ρ(y, z) (modified triangle law).
34 See Hausdorff [1914].

35 The terms themselves—l’espace métrique and metrischer Raum—were not new. L’espace métrique
even predated Fréchet’s dissertation. See Couturat [1905], page 204; and Couturat [1905] (1), page 216.
However, these earlier usages were not related to the theory of metric spaces discussed here.
36 In the 1920s Stanisław Leśniewski and Jan Łukasiewicz, among others, advocated minimizing the
number of axioms in formalized deductive systems. For more on this issue see, e.g., Sobociński [1955].
37 He noted two other avantageuses modifications which could as well do the trick—“ρ(x, z) + ρ(y, z) ≥
ρ(x, y)” and “ρ(z, x) + ρ(y, z) ≥ ρ(x, y)”—the second of which he credited to Piotr Szymański.
Garrett Birkhoff, citing Lindenbaum’s [26], gave yet another modification (see Birkhoff [1944], p. 466):
“ρ(x, y) + ρ(y, z) ≥ ρ(z, x)”. Birkhoff thought this “circularity postulate,” as he called it, had “a clear
intuitive content: if one journeys from p to q and then from q to r, the minimum energy required to get
back to p is not more than that already expended.”
More precisely, he showed that: If ρ is a real-valued function defined on M × M and

satisfying (ML1) and (ML3), then the laws (M2) and (M3) hold, and ρ takes only non-
negative values, which is to say, <M, ρ> is a metric space in the usual sense.
A metric space <N, σ> is said to be a subspace of a metric space <M, ρ> iff
N ⊆ M, and
σ is a restriction of ρ to N × N.
It is obvious . . .
that every subset of M determines a unique restriction of ρ,
that any such restriction satisfies (ML1) and (ML3), and hence
that every subset of M determines a unique subspace of <M, ρ>.
Consequently it is common practice to speak of a subset N ⊆ M as being a subspace
of M (and M a superspace of N) without explicitly presenting them as ordered pairs
<M, ρ> and <N, σ>, or mentioning their distance functions. Moreover, when their distance
functions are explicitly mentioned, they are often presented using the same symbol
to designate both the distance function on the superspace M and its restriction to the
subspace N, provided this leads to no confusion.38
The relation of being a subspace is transitive. Hence, in regarding a subset N as a
subspace of M, we need not consider whether the metric on N is inherited directly from
M, or indirectly, from the metric on some intermediate subspace Q, where N ⊆ Q ⊆ M.
Thus, if N is a subset of M, we may refer to N as either a subset or a subspace of M.
The choice is one of emphasis only. If we refer to the subspace N, we are focusing our
attention primarily on N itself, qua metric space, whereas if we refer to the subset N, we
are considering N’s set-theoretic properties in relation to M.
[26] began by defining basic notions of point-set topology and metric spaces. It
cited Hausdorff’s 1914 classic Mengenlehre, but in its choice of definitions, which
Lindenbaum conceded “diffèrent souvent de celles qu’on trouve ailleurs” <often differ
from those found elsewhere>, it was mainly motivated by a desire to avoid using the axiom
of choice (AC).39 This is particularly evident in its definitions of a closure operation and
a compact set.
Namely, if A is a subset of M, then the closure A of A is the set A ∪ A , where A is
the derived set of A in the space <M, ρ>, i.e., the set of all accumulation points of A in
<M, ρ>.40 Hence the closure operation is not defined in the spirit of Fréchet as the set of
all limit points of A (or points “adherent to” A).41
38 A similar shorthand is frequently adopted in speaking of metric spaces of differing dimensions, where
the space of lower dimension can be considered as embedded in the higher-dimensional space: in this
case, too, their distance functions are often presented using the same symbol to designate both.
39 This attitude toward the axiom of choice is stated explicitly in [26] on page 212, footnotes 1 and 3.
40 Let a ∈ M and A ⊆ M. Recall that a is an accumulation point of A in the metric space <M, ρ> iff every
open sphere with centre a contains at least one point of A which is distinct from a (and consequently an
infinite number of points of A). It is easy to see that a’s being (or not being) an accumulation point of A
does not depend on the whole space M, but only on the subspace A∪{a}.
41 As a student of Sierpiński, Lindenbaum must certainly have known that the theorem “If A is closed
(=A), then A contains all its limit points” is provable without using the axiom of choice, whereas the
proof of the converse implication needs AC. See Sierpiński [1918].
Lindenbaum defined compactness by means of the Cantor condition: a metric space

<M, ρ> is said to be compact iff, for every finite decreasing sequence {Fk }of non-empty
closed subsets of M, . . . Fk+1 ⊆ Fk ⊆ . . . ⊆ F1 ⊆ M, the intersection {Fk : k < ∞}
is non-empty.42 And he commented that an alternative definition might read as follows: a
metric space <M, ρ> is compact iff either M is finite or, for every infinite subset X of M,
the derived set X is non-empty (i.e., every divergent subset of X is finite).43
An isometric transformation or isometry between metric spaces <M, ρ> and <N, σ>
is a surjective mapping f between the points of M and the points of N which preserves
distance. Thus:
(∗) σ (f (x), f (y)) = ρ (x, y) for every pair x, y of points in M.
It is easy to see that any f satisfying the above conditions must be one-to-one. So, for
any function f :M → N, if f satisfies (*) then it is an isometry between <M, ρ> and the
subspace <f (M), σ> of the space <N, σ> . . . called the image space of M in N under f.
One also says that such an f is an isometry of M into N.
Metric spaces <M, ρ> and <N, σ> are said to be isomorphic, or congruent, or super-
posable (in symbols <M, ρ> ∼ = <N, σ>; or simply M ∼ = N), iff there exists an isometry
between them.
Notice that any two subspaces A and B of the real line with the standard or “natural”
distance function σ(x, y) = |x – y| are superposable only by means of a translation or
rotation. That is, if <A, ρ> ∼
= <B, ρ> then the isometric transformation f establishing this
congruence is either f (x) = x + c, or f (x) = −x + c, where c is a constant.44
An interesting theorem due to Sierpiński states: Any linear set A contains no more than
one point p such that A–{p} ∼ = A. A corollary states: In any non-empty linear set A there
exists a point p such that A is not congruent to A – {p}. See Sierpiński [1954], page 7.
The congruence relation ∼ = is of course nothing more than the familiar concept of
isomorphism as applied to the class of metric spaces. It is an equivalence relation on
this class. But the class of metric spaces admits of another, more inclusive45 equivalence
relation, namely homeomorphism, which turns out to be much more important than simple
isometry.
Aware of this, Lindenbaum devoted the second half of §3 of [26] to “the topological
properties of congruence and the problem of extending a given congruence” (see p.
214), by which he apparently meant, laying some groundwork for relating congruence
42 To be more precise, Lindenbaum defined what it means for an arbitrary subset A ⊆ M to be compact in
a metric space <M, ρ>. Then if A is closed and compact in <M, ρ>, then the subspace <A, ρ> is compact.
In general the assumption of closedness cannot be omitted.
43 One can prove without AC that if a metric space is compact by the first definition, using the Cantor
condition, then it is also compact by the second definition. The proof of the converse implication requires
AC. In general topology, a topological Hausdorff space (a T2 space) satisfying the Cantor condition, or
equivalently the dual Borel condition for open sets, is called countably compact. In the class of metric
spaces, compactness and countable compactness are equivalent.
44 . . . though there are plenty of non-standard distance functions for which this is not true (readers of a
certain age may recall slide rules). Subspaces of the real line with the standard distance function are called
linear sets (not to be confused with linearly ordered sets).
45 An equivalence relation is said to be more (/less) inclusive iff the corresponding partition has coarser
(/finer) granularity.
to topology. Then in §4 he turned to the concept of monomorphism. We employ our own

numbering for the results of these last nine pages of [26], departing from Lindenbaum’s
original, slightly shambolic scheme:
Theorem 1 Every isometry is a homeomorphism.
Theorem 2 If an isometry f maps a compact set A onto a compact set B, then there exists
a unique isometry f* of A onto B which is identical with f on A.
Hausdorff [1914] defined a totally bounded metric space as: <M, ρ> is called totally
bounded iff, for every ε > 0, there is a finite subset A ⊆ M such that for every x∈M there
is an a∈A with ρ(x, a) < ε. This is equivalent to the condition that, for every ε > 0, M is
the union of a finite number of open spheres (or balls) of radii < ε. Any compact metric
space is totally bounded but not all totally bounded metric spaces are compact.
Theorem 3 If <M, ρ> is a totally bounded metric space, and f :M → M is an isometry of
M into itself, then the image f (M) is dense in M, i.e. f (M) = M.46
To prove Theorem 3, Lindenbaum used the Dedekind chain method, which was used
extensively in the 1920s by Sierpiński, Kuratowski and Banach.
In §4 of [26], a set is said to be monomorphic iff it is not congruent with any of its
proper subsets. Lindenbaum defined it thus: “The set A is monomorphic, if the relations
B∼ = A and B ⊆ A occur only when B = A” (Def. 3, p. 217), and he added that a set is
monomorphic if it is a minimal (“irréductible”) element in the class of all sets on which
it is superposable.
It follows that monomorphy (or non-monomorphy), although being a property of a
set qua metric space, does not depend on the distance function used to construct a metric
space out of the set; it inheres in the nature of the set itself, not in the nature of the distance
function. Hence the definition should more correctly be worded thus: a metric space is
monomorphic iff it is not congruent with any of its proper subspaces. Alternatively: iff all
distance-preserving transformations of the space into itself are surjective.
Lindenbaum remarked that the following hold for Euclidean spaces:
(a) There exist non-monomorphic linear sets;
(b) Every bounded linear set is monomorphic;
(c) There exists a bounded plane set which is not monomorphic.
He then derived some sufficient conditions for a metric space to be monomorphic.47
As mentioned earlier, we use a different numbering scheme from his:
Theorem 4 If <M, ρ> is a compact space, then it is monomorphic. Alternatively: Any
closed and compact subset of a metric space is monomorphic. (Théorème 8 in his original
numbering scheme)48
46 This readable formulation of Lindenbaum’s Théorème 7 (p.216) is due to Ryszard Engelking [1989],
p. 278.
47 In his words, “Le théorème 8, les corollaires 15 et 14, nous fourniront des conditions suffisantes, de
plus en plus générales, pour qu’un ensemble compact soit monomorphe.”

48 Theorem 4 was generalized by Tarski as theorem 17(T) in [26a], p. 329.
Theorem 5a Every bounded subset of a compact metric space which is both Fσ and Gδ
is monomorphic.49
Theorem 5b Every bounded set in an n-dimensional Euclidean50 space which is both
Fσ and Gδ is monomorphic.51
Several theorems in §5 of [26a] are closely related to [26]: We mention three of them
here: theorem 4, due to Kirszbraun and Lindenbaum, and theorems 5(L) and 14(L), due
to Lindenbaum alone.52 The first two each give sufficient conditions for an expanding53
mapping to be an isometry on a Euclidean space:
4: Let B ⊂ R be a bounded linear space, and let A ⊆ Rn for 1 ≤ n < ∞ be a subspace
of an n-dimensional Euclidean space. If δ(A) ≥ δ(B), that is, if the diameter of A is not
less than the diameter of B,54 and if f :A → B is an expanding, surjective mapping of
A onto B, then f is an isometry between A and B, i.e., A ∼
= B.
5(L): Let A ⊂ Rn for 1 ≤ n < ∞ be a bounded subspace of an n-dimensional Euclidean
space. Then any expanding mapping f :A → A of the subspace into itself is an isometry
between A and f (A), i.e., A ∼
= f (A).
The third one, theorem 14(L), gives a necessary and sufficient condition for a set to be
non-monomorphic:
14(L): A set is not monomorphic iff it has a denumerable non-monomorphic subset.
The first two of these theorems, particularly 5(L), seem to have gone unnoticed, or
been quickly forgotten, because Hans Freudenthal and Witold Hurewicz published a note
in Fundamenta Mathematicae in 1936 proving a theorem very closely related to 5(L)
using exactly Lindenbaum’s methods from [26].55
Indeed it can be argued that [26a] went generally unremarked before the Second World
War. Stanisław Ruziewicz’s review of it in JFM was perfunctory nearly to the point of
dereliction of the reviewer’s duty: scarcely 3½ or 4 lines of text, suggesting Ruziewicz
himself had only skimmed the work, and offering no reasons why anyone else should
49 In the Introduction to [26] Lindenbaum wrote, “Au §4 j’examine la propriété singulière d’un ensemble
de points d’être superposable avec son vrai sous-ensemble. On peut indiquer des ensembles plans bornés
jouissant de cette propriété paradoxale, bien qu’ils ne puissent être Fσ et Gδ à la fois, ni linéaires; donc,
à plus forte raison, ils ne sauraient être fermés, ni ouverts, cependant il y en a qui sont Fσ ou Gδ . Voilà le
sujet principal, mais, à ce propos, j’étudie encore de plus près la notion (bien élémentaire) de congruence
(§3).” And in a footnote he explained, “Un ensemble est Fσ , s’il est une somme dénombrable d’ensembles
fermés; s’il est complémentaire d’un Fσ (c.-à-d.: produit dénombrable d’ensembles ouverts)—il est Gδ .”
50 By “Euclidean” is meant, that the distance function is the “natural” or “standard” distance function on
√
Rn —i.e., the square root of the sum of the squares: (x2 − x1 )2 + (y2 − y1 )2 + . . . .
51 Theorem 5b for n = 2 was generalized by Tarski as theorem 18(T) in [26a], p. 330.
52 See [26a], pp. 327–329.
53 If <M, ρ> and <N, σ> are two metric spaces, then a mapping f :M → N fulfilling σ( f(x), f (y)) ≥ ρ(x, y)
for all x, y ∈M is called expanding.

54 For a given (fixed) distance function ρ, the diameter δ(X) of a set X is the farthest distance between any
two points in X—i.e., δ(X) = sup {ρ(x, y): x, y ∈ X}. Since a distance function is by definition real valued,
it is always possible to compare diameters of spaces of different dimension.
55 See Freudenthal and Hurewicz [1936].
want to do even that much. Other than Ruziewicz’s review, neither Zbl, JFM nor JSL
contains any reference to [26a] that would suggest it was studied, cited or worked on
by others during the interwar period. This was possibly owing to where it appeared:
in Sprawozdania z posiedzeń Towarzystwa Naukowego Warszawskiego <Minutes of
the meetings of the Warsaw Society of Arts and Sciences> as opposed to a weighty
mathematics journal. It was only during and (mostly) after the Second World War that,
thanks to Sierpiński’s diligently filling in almost all of its missing proofs, the work started
receiving serious attention and garnering citations. Then in 1958 Sierpiński’s monograph
Cardinal and Ordinal Numbers capped these efforts and placed the value of [26a] beyond
question.
It is instructive to view Lindenbaum’s paper [33a] “Sur les ensembles localement
dénombrables dans l’espace métrique” in the context of Sierpiński [1933] “Sur les
espaces métriques localement séparables”. Both papers are about local properties—then
as now a subject of lively interest in topology. They exchanged ideas, shared preliminary
drafts of their manuscripts,56 and published their results in consecutive papers in the same
issue of Fundamenta Mathematicae.
Sierpiński’s paper characterized locally denumerable57 sets in a separable metric
space. Alluding to this in [33a] at the bottom of page 101 Lindenbaum wrote, “Or,
M. Sierpiński a posé la question quels sont des ensembles localement dénombrables
quand l’espace métrique n’est pas séparable.” In fact Sierpiński’s paper did not explicitly
pose this question, only dangled it. Lindenbaum may have meant “m’a posé”, i.e., in
conversation, or in marginalia on a shared manuscript. In any event [33a] took up the
question and developed new set-theoretical tools to answer it. “Voici une réponse:” he
wrote.
Maurice Fréchet introduced the notion of a separable space in his doctoral dissertation
and the importance of the concept was quickly recognized. The definition of a separable
metric space can be expressed in various (equivalent) ways. For example, a metric space
is said to be separable . . .
iff it has a denumerable open base; or . . .
iff any open covering of the space admits a denumerable subcovering.
A topological space X is called separable iff it contains a denumerable subset D which
is dense in X, that is to say, for which D = X.
A metric space <M, ρ> is said to be locally separable at a point x ∈ M iff there exists
an open sphere centred on x which, qua subspace, is a separable space. Then a space
<M, ρ> is called locally separable (simpliciter) iff it is locally separable at every point
x ∈ M.
The central result of Sierpiński’s paper was its proof of the following “Théorème: Pour
qu’un espace métrique soit localement séparable, il faut et il suffit qu’il soit une somme
56 See[33a], p. 102, footnote 10, and p. 104, footnote 13; and Sierpiński [1933], p. 107, footnote 2.
57 We use the word “denumerable” in the sense of “at most denumerable”, i.e., either finite or at most
countably infinite. We understand, for example, that a singleton is denumerable. So is the empty set.
disjointe d’ensembles ouverts séparables.” <A metric space is locally separable iff it is a
disjoint sum of open separable sets>.58
Lindenbaum framed his “réponse” in brand-new set-theoretical concepts. These are
worth spelling out, as it is not generally realized how ground-breaking they were for 1933.
Let be a given class of sets, and let <M, ρ> be a given metric space. We will say that a
subset Z ⊆ M is locally (or has the property locally) at a point z ∈ M, iff there exists
a real number r > 0, and a set Y ∈ , such that . . .
Y ∩ S (z, r) = Z ∩ S (z, r)
. . . where S(z, r) is an open sphere of M centred on z with radius r.

Note that for any r satisfying the above condition there is a smaller one that does so
too, and thus an infinitely descending sequence of them. This follows from a rudimentary
property of open spheres, namely: that for every p ∈ S(z, r) there is an r such that 0 < r < r
and S(p, r ) ⊆ S(z, r).
Then we can say simply that Z is locally (or has the property locally) iff Z is locally
at z for all z ∈ Z. The class of all sets which are locally will be denoted by .59
We say that the set Z is locally in the restricted sense iff Z is locally at z for every
z ∈ M. Note that, since Z ⊆ M, clearly if Z is locally in the restricted sense then it is
locally simpliciter. The class of all sets which are locally in the restricted sense will
be denoted by .
The operators and share basic properties with a closure operator; they are . . .
(1) extensive:
(2) isotone: implies and
(3) idempotent: and
As an exercise in cardinal arithmetic, one can estimate the cardinalities of and
. Assume that the space <M, ρ> is infinite. Let be the cardinality of a dense set in
M, and be the cardinality of the class . Then the class is of cardinality . If
, then the class is also of cardinality .
For a given cardinal number , let be the class of all sets of cardinality which are
contained in M. Thus, for example, consists of the empty set and all singleton subsets
of M, while is the class of all finite subsets of M.
A set Z is said to be isolated in a space <M, ρ> if Z and its derived set Z are disjoint,
i.e., if no accumulation point of Z is in Z. It is said to be divergent in the space <M, ρ> if
it has no accumulation points in M, i.e., if its derived set Z = ∅. Every divergent set is
isolated; however the converse is not necessarily true.60 It is easy to see that the elements
58 Sierpiński admitted (p. 107, footnote 2) that Lindenbaum had pointed out to him that this theorem was
“implicitement contenu dans un théorème de M. Alexandroff (Math. Ann. 92, p. 299, Fundamentalsatz
2)”, but that Alexandrov’s proof was “plus compliquée que la nôtre”.
59 Always bearing in mind, of course, that this definition of is relative to the given metric space
<M, ρ>.
60 We allow ourselves to go off on a small sidetrack here. For readers who may be wondering if isolated
and scattered are the same notion: no, they are not. Every isolated set is scattered, but in general, not all
scattered sets are isolated.
of are isolated sets, while consists of divergent sets. More interestingly,

and .
Since is the class of all denumerable subsets of M, then is the class of
all locally denumerable subsets of M. If M is a separable space, then by definition
.
Lindenbaum’s main theorems on locally denumerable sets were as follows61:
Theorem 6 For a set Z to be locally denumerable, i.e., to belong to , it is necessary
and sufficient that there exists a sequence of positive real numbers {dn } (“distances”) and
a sequence of divergent sets {Zn } such that . . .

6.1 Z = { Zn : n < ∞}; and . . .
6.2 if n and m are positive integers, and z an arbitrary element of Zn , then for all points
y of Zm , with the possible exception of (at most) one point, ρ(x, y) ≥ dm .
Then by using Sierpiński’s result Lindenbaum reformulated the above Theorem 6 in

terms of open sets as:
Theorem 7 For a set Z to be locally denumerable it is necessary and sufficient that there
exists a class of open sets such that . . .
7.1 Z is contained in the union of the sets of ;
7.2 for every , the set Z ∩ G is denumerable; and . . .
7.3 if G1 and G2 are distinct elements of , then Z ∩ G1 ∩ G2 is empty.
Lindenbaum remarked ([33a], p.106, footnote 18) that the above Theorems 6 and 7
(plus several others) remained valid even in a class of spaces—broader than the class of
metric spaces strictly understood—where there may be no distance at all between distinct
points of M, i.e., where ρ need satisfy only the conditions . . .
(Mt 1.1) if x = y then ρ (x, y) = 0 (“half” the law of coincidence) ; and . . .

(Mt 2) ρ (x, y) ≤ ρ (z, x) + ρ (y, z) Szymański’s modified triangle law .
For a proof of this, he referred the reader to §22 of his doctoral thesis.62
This means that Lindenbaum was one of the first, along with E.W. Chittenden (1917)
and W.A. Wilson (1931), to broaden or generalize the notion of metric space, and to use
this generalized concept as a tool for solving topological problems.63 Various kinds of
generalized metric spaces were considered in later years, e.g., by Karl Menger (1935),
Garrett Birkhoff (1936), and Hugo Ribeiro (1943), but none of these authors recognized
or remarked on the fact that Lindenbaum had been there before them.
61 Again,using our own numbering scheme, not Lindenbaum’s original numbering.

62 “ . . . la
condition Mt 1.2 n’étant point essentielle [Cf. ma Thèse (Varsovie, 1927; à paraître), §22].”
63 Ryszard Engelking, in his treatise General Topology, makes extensive use of pseudometric spaces “as
a convenient tool” for investigating a wide variety of topological spaces.

4 Decomposition of Point Sets, and Their Equivalence

by Decomposition
Decomposition of point sets and their equivalence by decomposition, i.e., the congruence
of their respective parts, was a subject of lively interest among the University of
Warsaw’s mathematicians in the early-to-mid-1920s, when Adolf Lindenbaum entered
the university and began studying under them and working with them. Results in 1924
alone included Kuratowski’s “Une propriété des correspondances biunivoques” <A
property of bijections>, Banach’s “Un théorème sur les transformations biunivoques” <A
theorem on bijections>, Tarski’s “O równoważności wielokatów” ˛ <On the equivalence
of polygons>, and Banach and Tarski’s famous paradox “Sur la décomposition des
ensembles de points en parties respectivement congruentes” <On the decomposition of
point sets into respectively congruent parts>. The first two established some eyebrow-
raising facts about one-to-one mappings in purely abstract set-theoretical contexts. The
second two were directly about equivalence by decomposition—of polygons in plane
geometry, and of point sets in a Euclidean space of finite dimension. It was against
this backdrop that Tarski and Lindenbaum set out their results on the theory of cardinal
numbers and decompositions of abstract sets in [26a], “Communication sur les recherches
de la théorie des ensembles,” §2: “Propriétés des transformations univoques.”
To say that a set A is decomposed into a family F of sets means that F is a partition on
the set A, i.e., F is a family of non-empty disjoint subsets of A whose union ∪{X : X ∈ F}
is all of A. If is the cardinality of F, then A is said to be -decomposed, or decomposed
into parts. In the same way, A is said to be finitely decomposed, ℵ0 -decomposed, ℵ1 -
decomposed, 2ℵ0 -decomposed, etc.
Stefan Mazurkiewicz and Wacław Sierpiński had whetted Warsaw’s appetite 10 years
earlier with their 1914 paper “Sur un ensemble superposable avec chacune de ses deux
parties” <On a set congruent with each of its two parts> wherein they proved that
there were nonempty sets A, A1 , A2 ⊂ R2 such that A = A1 ∪ A2 , A1 ∩ A2 = ∅,
A∼ = A1 , and A ∼ = A2 . What made their result so striking64 was that the transforms they
employed to superpose A1 and A2 on A were rigid, i.e., they preserved all the “standard”
distance relationships on R2 . One of their transforms was a simple rotation through an
angle of 1 radian; the other was a straight-line translation over a distance of +1. And
they had defined all three sets “effectively”, which is to say, without resorting to the
axiom of choice, transfinite induction, or the well-ordering theorem; all three sets were
denumerable.
64 On the most obvious level, of course, their result was another example of the paradox of infinity, namely,
that the part could equal the whole—a family of paradoxes, actually, with an august lineage, from Zeno of
Elea in the fifth century BC, through Galileo’s Two New Sciences of 1638, to Bolzano’s Paradoxes of the
Infinite, published posthumously in 1851. But after Cantor’s work on cardinality, and certainly by 1914,
mathematicians had gotten used to such paradoxes. The novelty of Mazurkiewicz and Sierpiński’s result
lay not in showing that the part could equal the whole (which by then was old news), but in the partition
and the transformations they devised, which were truly novel, prefiguring and in a sense anticipating the
isometry group E(n) of Euclidean motions and the notion of E(n)-equidecomposability, and ultimately
the definition of a paradoxical set. See also Hausdorff’s 1914 paradoxical decomposition of the sphere.
Just as their result appeared, however, the First World War broke out, followed by the
Soviet westward offensive of 1918–1919 and the Polish–Soviet War of 1919–1921, all of
which combined to put a damper on Warsaw University’s research activities.
In 1921 Stanisław Ruziewicz, working in Lwów, picked up where Mazurkiewicz and
Sierpiński had left off, obtaining a related (though not fully analogous) result for a non-
denumerable set in R2 : “Sur un ensemble non dénombrable de points, superposable avec
les moitiés de sa partie aliquote” <On a non-denumerable point set, congruent with halves
of its proper subset>.
Using the axiom of choice Ruziewicz defined non-denumerable sets A, B, C, D ⊂ R2
such that A = B ∪ C ∪ D, C ∩ D = ∅, and A ∼ = C and A ∼ = D. The family F = {B, C, D}
was not a strict partition of A, only a cover of A, as B was not necessarily disjoint from
C ∪ D. In fact the set B played a rather similar role to the number 0 in Mazurkiewicz and
Sierpiński’s 1914 proof.65
This result of Ruziewicz went some way toward answering (but stopped short of
fully answering) a question that Hugo Steinhaus had earlier posed: Does there exist an
uncountable planar set which admits a 2-decomposition each of whose parts is congruent
with the whole?
Lindenbaum went the full distance and answered Steinhaus’s question in the affirma-
tive in the following theorems:
1. If A is a linear set congruent with each of two subsets B ⊆ A and C ⊆ A, then it is
congruent with a subset D ⊆ (B ∩ C) ⊆ A.
(See: [26a], p. 327, theorem 1(L).)
2. If A is a bounded planar set congruent with each of two subsets B ⊆ A and C ⊆ A, then
it is congruent with a subset D ⊆ (B ∩ C) ⊆ A.
(Ibid., theorem 2(L).)
Corollary of 1 + 2: No linear set, and no bounded planar set, can be decomposed into
two parts each of which is congruent with the whole set.
(See: [26], p. 218, footnote 1.)
Recall of course that Ruziewicz and Sierpiński [1914], and Hausdorff [1914], had
shown there do exist unbounded planar sets, and sets on the surface of a 3-dimensional
sphere, which can be so decomposed.
3. For every cardinal number , there exists an unbounded planar set which can be
decomposed into parts each of which is congruent with the whole. A similar set can
be constructed on the surface of a sphere.
(See: [26a], loc. cit., theorem 3*(L).)
65 Ruziewicz actually obtained this result some 2 years earlier, in the summer of 1919, and ran it past
Sierpiński for publication in the first issue of Fundamenta Mathematicae, i.e., the 1920 issue (Sierpiński
was its founding editor). Sierpiński suggested a simplification, which Ruziewicz incorporated. The
existence of the set B could be proved directly from Zermelo’s axioms (including the axiom of choice),
without any need of Zermelo’s theorem on well-ordering, or transfinite numbers. Sierpiński also urged
Ruziewicz to consult Hugo Steinhaus, who allegedly had an example of a non-denumerable planar set
congruent to its halves. As it turned out, Steinhaus did not have such an example. See the letters from
Sierpiński to Ruziewicz dated: 01 April 1919, 17 August 1919, 20 August 1919, and 19 April 1920, in
Wi˛esław [2004], pp. 141–143.
The above results nicely illustrated how Sierpiński’s own later research was shaped
by Lindenbaum’s. The asterisk* by 3*(L) meant that Lindenbaum had used the axiom
of choice in his (unpublished) proof. Twenty-one years later Sierpiński [1947b] proved
Lindenbaum’s third theorem without using the axiom of choice or any of its equivalents,66
and in so doing answered Steinhaus’s question in an effective manner: There is a
constructive proof that, yes, there does exist an uncountable planar set which admits a 2-
decomposition each of whose parts is congruent with the whole. Then Sierpiński went on
to supply the missing proofs for 1, 2, and the Corollary of 1+2 above, which Lindenbaum
had said he would deliver “à plus tard” but never got around to doing.
Many of the results in [26a] exploited and expanded on Stefan Banach’s [1924] “Un
théorème sur les transformations biunivoques” <A theorem on bijections>, applying it to
the theory of functions, cardinal arithmetic and decompositions of point sets.67 Banach’s
central theorem could be called “decompositions of pure sets”, or the DPS theorem. It
stated that:
(DPS) For arbitrary sets A and B, if f is a one-to-one transformation of A onto a subset
of B, and g is a one-to-one transformation of a subset of A onto all of B, then there
exist decompositions of A and B: A = A1 ∪A2 , B = B1 ∪B2 , A1 ∩ A2 = ∅ = B1 ∩B2 ,
such that . . . f (A1 ) = B1 and g(A2) = B2 .
Banach then set out a pair of useful properties which a 2-place relation between sets
might possess. He called them property (α), and property (β):
(α) Whenever A R B, there exists a bijection f :A → B such that for every X ⊆ A,
X R f (X).
(β) If A1 ∩ A2 = ∅ = B1 ∩B2 , and A1 R B1 , and A2 R B2 , then (A1 ∪ A2 ) R (B1 ∪ B2 ).
And with these tools to hand, he proved the following two further theorems:
DPS2 For a relation R with property (α), if A stands in relation R to some subset of B,
and some subset of A stands in relation R to B, then there exist decompositions of A
and B: A = A1 ∪ A2 , B = B1 ∪ B2 , A1 ∩ A2 = ∅, B1 ∩ B2 = ∅, such that A1 R B1 and
A2 R B2 .
DPS3 For a relation R with properties (α) and (β), if A stands in relation R to some
subset of B, and some subset of A stands in relation R to B, then A R B.
Some fundamental relations in set theory and topology turn out to possess properties
(α) and (β), e.g., equipollence of pure sets, similarity of ordered sets, homeomorphism
of topological spaces, and congruence of point sets. Properties (α) and (β) also crop
up (indeed figure prominently) in work on “les types de dimensions”, to use Fréchet’s
expression.68
66 Sierpiński
wrote: “La démonstration de A. Lindenbaum n’a pas été publiée et elle m’est inconnue.”
(p. 9).
67 See especially [26a], §2, “Propriétés des transformations univoques” <Properties of one-to-one
functions>, pp. 316–318.

68 See Arboleda [1981].
In [26a] Lindenbaum and Tarski used the terms “relation transformante” for a relation
possessing property (α), and “relation additive” for a relation possessing property (β).
Two of Lindenbaum’s results from [26a] were69 . . .
3(L) If A⊆B⊆C, A1 ⊆C, and function f :A → A1 is surjective, then there exist four sets
B1 , D, D1 , and E such that . . .
(a) A1 ⊆ B1 ⊆ C,
(b) B = D ∪ E and B1 = D1 ∪ E,
(c) D ∩ E = ∅ and D1 ∩ E = ∅, and
(d) f (D) = D1 .
4(L) If A⊆B⊆C, A⊆C1 , and function g:C1 → C is surjective, then there exist four sets
B1 , D, E, and E1 such that . . .
(a) A ⊆ B1 ⊆ C1 ,
(b) B = D ∪ E and B1 = D ∪ E1 ,
(c) D ∩ E = ∅ and D ∩ E1 = ∅, and
(d) g(E1 ) = E.
The first of these, 3(L), entailed Banach’s DPS theorem, even though it did not require
that the function f was one-to-one.
In [26a], §1, “Théorie des nombres cardinaux,” Lindenbaum used the above results
to help him prove 14(L) and 14(bis)70 relating to the Cantor–Bernstein theorem, which
provided Tarski with the machinery he was looking for to derive 15(T).71 The latter result
has since come to be known as Tarski’s Mean-value Theorem, and it has its counterpart
in theorem 5(T) of §2, p. 318, concerning decompositions.
Let n be a natural number. Point sets A and B in a metric space <M, ρ> are said to be
equivalent by n-decomposition—written A ≡n B—iff there exist two families of subsets,
FA = {A1 , A2 , . . . , An }, and FB = {B1 , B2 , . . . , Bn }, such that . . .
FA is an n-decomposition of A,
FB is an n-decomposition of B, and
Ak is congruent to Bk , i.e., Ak ∼
=Bk , for all k: 1≤k≤n;
. . . and they are said to be equivalent by finite decomposition—written A ≡f B—iff there
exists a natural number n for which A ≡n B. This can be extended in a natural way to
equivalence by -decomposition—written —where is an arbitrary transfinite
cardinal number.
The above definitions—of equivalence by n-decomposition, by finite decomposition,
and by -decomposition—are from Banach and Tarski [1924], who established funda-
mental properties of these relations. Firstly, and most obviously, that for n = 1, A ≡n B
is simply an isometry ∼ = in the space <M, ρ>. Secondly, that for all m ≥ n, A ≡n B
implies A ≡m B. Thirdly, that for fixed n, the relation ≡n is not transitive (simple
counterexamples suffice to show this). Fourthly, and more interestingly, that equivalence
69 See [26a], page 318.

70 See pp. 302–303.
71 Ibid, p. 303.
by finite decomposition—where you are free to choose a different n for each pair of sets—
is transitive. Their proof of this used what they called a “double network” method. And
since reflexivity and symmetry obviously hold, the relation ≡f is an equivalence relation.
Moreover, they showed that ≡f has Banach properties (α) and (β).
Lindenbaum and Tarski restated most of the above definitions and properties in [26a],
along with several new findings (some joint, some by Tarski or Lindenbaum alone).72
They demonstrated that:
(i) The relation of equivalence by -decomposition for is completely additive.
That is to say, if X1 , X2 , . . . , Xn , . . . and Y1 , Y2 , . . . , Yn , . . . are two sequences of
mutually disjoint sets, and for all naturals k, then ( ).
(See [26a], p. 328, theorem 6.)
(ii) If A ⊆ B ⊆ C and A ≡n C, then A ≡n+1 B ≡n+1 C.
(Ibid., theorem 9.)
Lindenbaum supplemented his result from [26], that every bounded linear set was
monomorphic, with:
13(L). There exists a bounded linear set A which has a proper subset B such that A ≡2 B.
(Ibid., page 329.)
He also considered combinatorial properties of congruence and decomposability, as in
a result mentioned in [26], p. 218, footnote 2:
• Let n be a natural number, and let A and B be subspaces of a metric space. If A ∼
=B
and A ∩ B contains fewer than n(n+1)
2 elements, then (A – B) ≡ n (B – A).
This was proved for the first time only by Sierpiński [1954], pp. 110–113, who also
showed, by means of a suitable counter-example on the straight line, that the number
n(n+1)
2 cannot be any greater.
In his 1943 paper “Some remarks on set theory”73 Paul Erdős recounted the following
interesting story:
Professor Tarski communicated to me the following result of Lindenbaum: There exist linear
sets no two of which are countable equivalent [by decomposition]. This result was never published,
and Tarski does not remember the details of the proof. I have succeeded in proving that if is any
cardinal number , then there exist linear sets no two of which are -equivalent. I do not
know whether my proof differs from that of Lindenbaum, but I have thought it might be worth
publishing, since the result has some interesting applications.74
Erdős’s proof used the axiom of choice in an essential way. Enter Sierpiński, who
4 years later was able to prove a generalization—a strengthening—of the same theorem75
without the axiom of choice, using the same von Neumann function that he and Ruziewicz
had used in the 1930s . . . work which Lindenbaum had surely been aware of at the
time. We have to conclude that, quite possibly, Sierpiński’s proof merely recapitulated
Lindenbaum’s original.
72 See [26a], §5, “Théorie des ensembles de points,” p. 328.

73 Erdős [1943].
74 Ibid., page 644.
75 See: Sierpiński [1947].
From 1931 on, Lindenbaum contributed importantly to investigating Fréchet’s “les

types de dimensions” of topological spaces.76
5 Decompositions and Equivalence of Polygons in Elementary

Geometry
Lindenbaum published only one short note in this area:

[37a a] “Sur l’équivalence de deux figures par décomposition en nombre fini de par-
ties respectivement congruentes.” Rocznik Polskiego Towarzystwa Matematycznego
(=Annales de la Société Polonaise de Mathématique), vol. 16 (année 1937, publ.
1938), p. 197.
This is a (six-line) summary of a lecture given by Lindenbaum on 30 September
1937 to the Third Polish Mathematical Congress in Warsaw. His talk began by setting
the problem in its historical perspective, then he outlined recent trends, which he
called “quantitatives”, then he worked through the proof of a theorem he and Zenon
Waraszkiewicz had obtained in 1932, and in conclusion he presented some other related
theorems without proofs. We state the Lindenbaum–Waraszkiewicz theorem below. Other
details as to what he said in the talk we can only surmise.
Fortunately we know quite a bit about Lindenbaum’s work in the theory of equiv-
alence of polygons from two of Tarski’s papers: [1931], “O stopniu równoważności
wielokatów”
˛ <On the degree of equivalence of polygons>, and [1931/32], “Uwagi o stop-
niu równoważności wielokatów”
˛ <Remarks on the degree of equivalence of polygons>.
In elementary geometry, two polygons are said to be equivalent if it is possible to
dissect them into the same finite number of respectively congruent polygons having no
common interior points . . . though the parts will invariably share common boundary
points: sides, or segments of sides, or vertices, etc.
This notion has a similar logical structure to the purely set-theoretical notion of
equivalence by finite decomposability, A ≡f B, discussed above. One important difference
between them is how they treat boundary points. Another is how they define congruence.
The notion of congruent planar figures in elementary geometry is much narrower than
the general idea of congruent metric spaces, and even differs in essential ways from the
definition of superposable subsets of R2 . Some of the other differences, however, are
more superficial: in elementary geometry it is common to use the symbol ‘≡’ without
a subscript, finiteness being taken for granted, and to use verbs like ‘divide’, ‘dissect’,
or ‘cut’ (as with scissors) instead of the septic ‘decompose’, which conjures images of
bacterial decay.
To appreciate how big a difference common boundary points make, consider √a square
√S
with sides of length 1, and an isosceles right triangle T with sides of length 2, 2 and 2.
76 See [31a ], [34a a], [36a ]; see also Sierpiński’s paper “Sur un problème concernant les types de
dimensions” in Fundamenta Mathematicae, vol. 19 (1932), pp. 65–71, wherein Sierpiński on pages
67–68 quotes verbatim Lindenbaum’s two-page proof of a generalization—a strengthening—of one of
Sierpiński’s own results.
Both S and T can be √cut exactly in half, into mirror-image pairs of isosceles right triangles
with sides of length 2, 1, and 1. But neither of these divisions is a decomposition, in the
set-theoretical sense of a partition, because the resulting halves are not disjoint: in both
cases they share an edge. And it is not obvious at first sight how to decompose S and T
into respectively disjoint parts such that S ≡f T. (Sierpiński [1954], p. 43, Theorem 15
offers one.)
The following two theorems about equivalence of polygons are provable in the usual
axiomatic systems of elementary geometry:
• If polygon V is a part of polygon W, then these polygons are not equivalent (known as
De Zolt’s axiom); and . . .
• Polygons V and W are equivalent if and only if they have equal areas (known as the
Wallace–Bolyai–Gerwien theorem).
Tarski was the first to ask if these (or analogous) theorems remain true when
equivalence is understood as set-theoretical equivalence ≡f by finite decomposition into
parts having no common points. Tarski [1924], “O równoważności wielokatów” ˛ <On
the equivalence of polygons>, answered this question in the affirmative using Banach’s
measure theorem for bounded planar sets (see Banach [1923]).
It is clear that no dissection of equivalent polygons into congruent parts can be unique:
two equivalent polygons can always be divided into congruent parts in various ways, with
respect to both the form and the number of those parts. Hence the question arises: what
is the smallest number of respectively congruent parts that two equivalent polygons can
be divided into? For equivalent polygons, W and V, this smallest number is called their
degree of equivalence, and is denoted by σ(W, V).77 (See Tarski [1931], pp. 37–38.)
Tarski attributed this definition to Lindenbaum in a footnote: “O ile nam wiadomo,
poj˛ecie to wprowadził Dr. Adolf Lindenbaum (Warszawa), który wraz z autorem artykułu
ustalił pewne własności tego poj˛ecia” <As far as we know, this notion was introduced by
Dr. Adolf Lindenbaum (Warsaw) who together with the author of this article established
some properties of the concept>. (See Tarski [1931], p. 38.) Oddly enough, none of the
properties or results that he reports in that paper are attributed to Lindenbaum.
Tarski defined a function τ as follows: “Let Q be a square with edge a, and let P be a
rectangle with edges x·a and a/x , where x is any positive real number. Polygons P and Q
are obviously equivalent, and it is easy to see that their degree of equivalence is a function
of x; we shall denote this function by the symbol τ(x). Thus τ(x) = σ(Q, P),” . . . and then
he urged his colleagues to join him in investigating its properties.
He himself offered some general results, including upper bounds on τ(x) for certain
values of x. For instance: τ(11 /3 ) ≤ 3; τ(21 /4 ) ≤ 4; τ(n) ≤ n for all natural numbers n.
This last inequality is easy to prove. A square with edge a can be dissected into n mutually
congruent rectangular strips with edges a and a/n, from which a rectangle with edges n·a
and a/n can be assembled by arranging all the strips end to end.
Tarski conjectured that: (i) τ(n) = n for every natural number n; and asked whether:
(ii) τ(x) ≥ 3 for every positive x different from ½, 1 and 2?
77 The symbol “σ” was chosen because of the first letter of the Polish word “stopień”, which means
“degree”.
1 I
1 I
2 2 II II
Fig. 11 Sierpiński’s sketch, showing how Lindenbaum’s examples work
In addition to Tarski himself, Henryk Moese, Adolf Lindenbaum, Bronisław Knaster

and Zenon Waraszkiewicz investigated the function τ. Moese published a detailed proof
of (i) in Moese [1932]. The others’ results were published en masse in Tarski [1931/32],
where it was noted that Zenon Waraszkiewicz was the first to come up with a proof of (i)
but did not publish his proof at the time.
Lindenbaum noticed that, “all the results so far obtained for the function τ(x) still hold
when, in the definition of this function, the square is replaced by any rectangle”, that is,
for a function τr (x) = σ(Q, P), where Q is a rectangle with edges a and b, and P is a
rectangle with edges x·a and b/x . (See Tarski [1931/32], p. 313) And he also obtained the
following:
• If x = n + 1 /p where n and p are naturals ≥ 1, then τ(x) = n + 1, i.e., τ(x) rounds x up.
• If 1 < x ≤ 2 then τ(x) = 2 if and only if x = 1 + 1 /p where p is a natural ≥ 1.
The second of these he obtained jointly with Zenon Waraszkiewicz, and its proof
was—in Tarski’s words—“somewhat complicated” and required “some subtle methods
of reasoning.” (See Tarski [1931/32], p. 312) It was this theorem and its proof that
Lindenbaum delivered to the Third Polish Mathematical Congress in Warsaw on 30
September 1937. In [37a a] it is formulated: For non-congruent rectangles a1 × a2 and
b1 × b2 each to be decomposed into 2 respectively congruent parts, it is necessary and
sufficient that either a1/ a2 or b1 /b2 equal k + 1/k for some whole positive k.
At the end of Tarski [1931/32] there is a set of “exercises” for the ambitious reader.
Presumably, all of the problems were known to their respective contributors to be solvable.
Lindenbaum offered the following challenge78:
• Prove that, if W is a convex figure situated in the plane, then s(W) ≥ δ(W )·ω(W
2
)
, where
s(W), δ(W) and ω(W) are respectively the area, diameter and width of W. Moreover, in
the above formula we will have equality if and only if W is a triangle.
Sierpiński, in a letter to Stanisław Ruziewicz dated Warszawa, 14/II/1932 r.,79 wrote:
“Lindenbaum znalazł ciekawy przykład równoważności przez rozkład kwadratu i pros-
tokata,”
˛ <Lindenbaum found an interesting example of equivalence by decomposition of
a square and a rectangle>, and he drew a sketch in his letter to show Ruziewicz how it
worked (Fig. 11).
It turned out, the decompositions Sierpiński sketched were crucial to the proof that
Lindenbaum and Waraszkiewicz devised.
78 Ibid., page 314.

79 See Wi˛esław [2004], page 158.
Note. A deeper and broader presentation of the papers Tarski [1924], Banach–Tarski
[1924], Tarski [1932], Tarski [1931/32] and Moese [1932], together with excellent English
translations of them, can be found in the book McFarland–McFarland–Smith [2014]. For
a survey of related results in the 70 years after Hausdorff’s, Banach’s and Tarski’s works,
see the fascinating monograph Wagon [1993].
6 Bibliography
6.1 Works by Adolf Lindenbaum
In Sect. 6.1 we select the main works of Adolf Lindenbaum. We follow the citation
style of the Lindenbaum bibliography in Zygmunt–Purdy [2014]. A work is cited by a
two-digit year in square brackets. Where several works appear in the same year, alphabetic
suffixes designate their order of appearance. Citations of abstracts and short notes are
distinguished by a superscripta ; reviews by a superscriptr.
The Lindenbaum bibliography in Zygmunt–Purdy [2014] identified him as the author
of seven reviews. That turned out to be a lamentable understatement. We are now able to
attribute 22 reviews to him, all of them published in Zentralblatt für Mathematik.
We also note that the Lindenbaum bibliography in Zygmunt–Purdy [2014] included
a fourth sub-part, canvassing his public lectures, with citations distinguished by a
superscriptl. We omit that sub-part here, as we are currently reviewing its contents. We
have learned that Lindenbaum delivered two lecture series, both titled „O matematyce”
<On Mathematics>, to the Instytut Wyższej Kultury Umysłowej Zwiazku ˛ Nauczycielstwa
Polskiego <Institute of Advanced Learning of the Polish Teachers Union>: the first a 15-
hour series over the month of July, 1938,80 and the second a 20-hour series sometime
between October 15th, 1938 and May 31st, 1939.81 These presently await detailed
documenting.
6.1.1 Papers
[26] Contributions à l’étude de l’espace métrique I. Fundamenta Mathematicae,

vol. 8 (1926), pp. 209–222. (The second part of [26] was announced on p. 214,
note 1, but was never published. See [26a], p. 327, note 1, for a correction.)
[JFM 52.0585.01 (E. Pannwitz)].
80 Cited in Ruch Pedagogiczny, rok 27, nr 6–10 (1938).

81 An advance notice was given in Głos Nauczycielski, nr 5, (25 September 1938), page 78. Some of the
titles of individual lectures in the second series are rather philosophical in tone: Co to sa˛ nauki scisłe
(matematyczno-przyrodnicze)? <What are exact sciences (mathematical-natural)?> Jakie nauki należa˛
do tej grupy? <What sciences belong to this group?> Trudności podziału nauk: przedmiot a metoda.
<Difficulties in the divisions of sciences: subject matter vs. method.> Metoda dedukcyjna. <The deductive
method.> Czy matematyka posiada przedmiot i jaki? <Does mathematics have a subject, and if so,
what?> Rachmistrz-technik a matematyk-badacz. <Number cruncher-technician versus mathematician-
researcher.> Element twórczy i “artystyczny” w matematyce. <The creative and “artistic” element in
mathematics.>
[26a] Communication sur les recherches de la théorie des ensembles (with A.Tarski).
Sprawozdania z posiedzeń Towarzystwa Naukowego Warszawskiego, Wydział
III Nauk Matematycznych i Przyrodniczych (=Comptes rendus des séances
de la Société des Sciences et des Lettres de Varsovie, Classe III), vol. 19
(1926), pp. 299–330. [JFM 57.1330.02 (S. Ruziewicz)].
(1) Reprinted in: Alfred Tarski, Collected Papers. Volume 1, 1921–1934, ed.
by S.R. Givant and R. N. McKenzie, Birkhäuser Verlag: Basel 1986, pp. 173–
204.
[30] Remarques sur une question de la méthode axiomatique. Fundamenta Mathe-
maticae, vol. 15 (1930), pp. 313–321. [JFM 56.0488.03 (A. Rosenthal)].
[30a] Sur les opérations d’addition et de multiplication dans les classes d’ensembles
(with A. Koźniewski). Fundamenta Mathematicae, vol. 15 (1930), pp. 342–
355. [JFM 56.0084.02 (R. Baer)].
[31] Sur les ensembles ordonnés. Comptes rendus hebdomadaires des séances
de l’Academie des Sciences, Paris, vol. 192 (1931), pp. 1511–1514. [JFM
57.0091.03 (A. Fraenkel); Zbl 0002.18405 (A. Kolmogoroff)].
[33] Sur les ensembles dans lesquels toutes les équations d’une famille donnée
ont un nombre de solutions fixé d’avance. Fundamenta Mathematicae, vol.
20 (1933), pp. 1–29. Errata, p. 287. [JFM 59.0095.02 (W. Hurewicz); Zbl
0006.34001 (B. Knaster)].
[33a] Sur les ensembles localement dénombrables dans l’espace métrique. Fun-
damenta Mathematicae, vol. 21 (1933), pp. 99–106; Errata p. 295. [JFM
59.0567.04 (G. Aumann); Zbl 0008.08806 (E. Čech)].
[33b] Sur les superpositions des fonctions représentables analytiquement. Comptes
rendus hebdomadaires des séances de l’Académie des Sciences, Paris, vol.
196 (1933), pp. 1455–1457. [JFM 59.0267.03 (A. Rosenthal); Zbl 0007.06005
(S. Saks)]. (For errata see [34a], p. 13, note 1.)
[34] Z teorii uporzadkowania
˛ wielokrotnego <Sur la théorie de l’ordre multiple>
(Polish with French summary). Wiadomości Matematyczne, vol. 37 (1934),
pp. 1–35. [JFM 60.0867.01 (S. Ruziewicz); Zbl 0009.30304 (Th. Motzkin)].
[34a] Sur les superpositions des fonctions représentables analytiquement. Funda-
menta Mathematicae, vol. 23 (1934), pp. 15–37; errata, op. cit., p. 304.
[JFM 60.0195.02 & JFM 60.0195.03 (A. Rosenthal)]; [Zbl 0009.30502 & Zbl
0010.01403 (S. Saks)].
[35] Miara w geometrji (with E. Szpilrajn) <Measure in geometry. Polish>. Świat
i życie. Zarys encyklopedyczny współczesnej wiedzy i kultury, vol. 3, Lwów–
Warszawa 1935, pp. 586–595.
[36] Sur la simplicité formelle des notions. Actes du Congrès International de
Philosophie Scientifique, Sorbonne, Paris 1935, VII Logique, Actualités
scientifiques et industrielles, vol. 394 (1936), Hermann & Cie : Paris 1936,
pp. 29–38. [JFM 62.1050.01 (W. Ackermann); JSL 2 (1937), pp. 55–56 (S. C.
Kleene)].
[36a] Über die Beschränktheit der Ausdrucksmittel deduktiver Theorien (with A.
Tarski). Ergebnisse eines mathematischen Kolloquiums, vol. 7 (1936), pp.
15–22. [JFM 62.0039.02 (F. Bachmann); Zbl 0014.38602 (A. Schmidt); JSL
1, p. 115–116 (B. Rosser)].
(1) On the limitations of the means of expression of deductive theories. In:

A. Tarski, Logic, Semantics, Metamathematics. Papers from 1923–1938.
Translated by J.H. Woodger, Clarendon Press: Oxford 1956, pp. 384–392.
(Revised English translation of Lindenbaum [36a].)
(2) Sur les limitations des moyens d’expression des théories déductives.
In: A. Tarski, Logique, sémantique, métamathématique. 1923–1944, vol. 2,
Philosophie pour l’Âge de la Science, Librairie Armand Collin: Paris 1974,
pp. 109–120. (Revised French translation of Lindenbaum [36a].)
(3) Reprinted in Alfred Tarski, Collected Papers. Volume 2, 1935–1944, ed. by
S.R. Givant and R. N. McKenzie, Birkhäuser Verlag: Basel 1986, pp. 205–212.
(4) O ograniczeniach środków wyrazu teorii dedukcyjnych. In: A. Tarski,
Pisma logiczno-filozoficzne. Volume 2, translated and annotated by Jan Zyg-
munt, Wydawnictwo Naukowe PWN: Warsaw 2001, pp. 147–157. (Polish
translation of [36a](1).)
[38] Über die Unabhängigkeit des Auswahlaxioms und einiger seiner Folgerungen
(with A. Mostowski). Sprawozdania z posiedzeń Towarzystwa Naukowego
Warszawskiego, Wydział III Nauk Matematyczno-Fizycznych (=Comptes
rendus des séances de la Société des Sciences et des Lettres de Varsovie,
Classe III), vol. 31 (1938), pp. 27–32. [JFM 64.0932.01 (Th. Skolem); Zbl
0019.29502, p. 295 (A. Schmidt); JSL 4, pp. 30–31 (A. A. Fraenkel)].
(1) On the independence of the axiom of choice and some of its consequences.
In: A. Mostowski, Foundational Studies. Selected Works, vol. 2, edited by
K. Kuratowski, W. Marek, L. Pacholski, H. Rasiowa, C. Ryll-Nardzewski,
and P. Zbierski. Studies in Logic and the Foundations of Mathematics,
vol. 93, North-Holland Publishing Company: Amsterdam 1979; and PWN—
Polish Scientific Publishers: Warszawa 1979, pp. 70–74. (English translation
of Lindenbaum [38] by M.J. Maczyński.)
˛ [Zbl 0425.01021 (E. Mendelson)].
6.1.2 Abstracts and Short Notes
[26a ] Sur l’arithmétique des types ordinaux. Rocznik Polskiego Towarzystwa Matem-
atycznego (=Annales de la Société Polonaise de Mathématique), vol. 5 (année
1926, publ. 1927), pp. 103–104. (Summary of §3 of [26a], presented at the
meeting of the Polish Mathematical Society, Warsaw Section, on 23 April 1926.)
[26a a] Sur l’indépendence des notions primitives dans les systèmes mathématiques
(with A. Tarski). Rocznik Polskiego Towarzystwa Matematycznego (=Annales
de la Société Polonaise de Mathématique), vol. 5 (année 1926, publ. 1927), pp.
111–113. (Presented at the meeting of the Polish Mathematical Society, Warsaw
Section, on 17 December 1926.)
(1) Reprinted in Alfred Tarski, Collected Papers. Volume 4, 1958–1979, ed. by
S.R. Givant and R.N. McKenzie, Birkhäuser Verlag, Basel 1986, pp. 538–540.
[27a ] Sur quelques propriétés des fonctions de variable réelle. Rocznik Polskiego
Towarzystwa Matematycznego (=Annales de la Société Polonaise de Math-
ématique), vol. 6 (année 1927, publ. 1928), pp. 129–130. [JFM 54.0298.01]
(Presented at the meeting of the Polish Mathematical Society, Warsaw Section,
on 24 October 1926.)
[29a ] Méthodes mathématiques dans les recherches sur le système de la théorie de

déduction. Ksi˛ega Pamiatkowa
˛ Pierwszego Polskiego Zjazdu Matematycznego
we Lwowie w 1927 roku (supplement to Annales de la Société Polonaise de
Mathématique), Kraków 1929, p. 36.
[29a a] O pewnych własnościach metrycznych mnogości punktowych.—Sur certaines
propriétés métriques des ensembles de points. Ksi˛ega Pamiatkowa
˛ Pierwszego
Polskiego Zjazdu Matematycznego, Lwów, 7–10. IX. 1927 (supplement to
Annales de la Société Polonaise de Mathématique), Kraków 1929, p. 96.
[31a ] Sur un ensemble linéaire extrêmement non homogène par rapport aux trans-
formations continues et sur le nombre des invariants de ces transformations.
Rocznik Polskiego Towarzystwa Matematycznego (=Annales de la Société
Polonaise de Mathématique), vol. 10 (année 1931, publ. 1932), pp. 113–114.
[JFM 58.0648.12] (Presented at the meeting of the Polish Mathematical Society,
Warsaw Section, on 16 January 1931.)
[31a a] La projection comme transformation continue la plus générale. Rocznik Pol-
skiego Towarzystwa Matematycznego (=Annales de la Société Polonaise
de Mathématique), vol. 10 (année 1931, publ. 1932), pp. 116–117. [JFM
58.0648.13] (Presented at the meeting of the Polish Mathematical Society,
Warsaw Section, on 8 May 1931.)
[31a b] Sur les figures convexes. Rocznik Polskiego Towarzystwa Matematycznego
(=Annales de la Société Polonaise de Mathématique), vol. 10 (année 1931,
publ. 1932), pp. 117–118. [JFM 58. 0805.05] (Presented at the meeting of the
Polish Mathematical Society, Warsaw Section, on 8 May 1931.)
[31a c] Sur les constructions non-effectives dans l’arithmétique élémentaire. Rocznik
Polskiego Towarzystwa Matematycznego (=Annales de la Société Polonaise
de Mathématique), vol. 10 (année 1931, publ. 1932), pp. 118–119. [JFM 58.
1001.14] (Presented at the meeting of the Polish Mathematical Society, Warsaw
Section, on 8 May 1931.)
[34a ] Sur le “problème fondamental” du jeu d’échecs. Rocznik Polskiego
Towarzystwa Matematycznego (=Annales de la Société Polonaise de
Mathématique), vol. 13 (année 1934, publ. 1935), pp. 124–125. (Presented
at the meeting of the Polish Mathematical Society, Warsaw Section, on 24
February 1933.)
[34a a] Sur le nombre des invariants des familles de transformations arbitraires. Rocznik
Polskiego Towarzystwa Matematycznego (=Annales de la Société Polonaise
de Mathématique), vol. 13 (année 1934, publ. 1935), p. 131. (Presented at the
meeting of the Polish Mathematical Society, Warsaw Section, on 19 January
1934.)
[34a b] Remarques sur le groupe des permutations de l’ensemble des nombres entiers.
Polonaise de Mathématique), vol. 13 (année 1934, publ. 1935), p. 131. (Pre-
sented at the meeting of the Polish Mathematical Society, Warsaw Section, on
19 January 1934.)
[34a c] Sur les relations contenues dans les relations ordinales. Rocznik Polskiego
Towarzystwa Matematycznego (=Annales de la Société Polonaise de Mathé-
matique), vol. 13 (année 1934, publ. 1935), p. 132. (Presented at the meeting of
the Polish Mathematical Society, Warsaw Section, on 19 January 1934.)
[36a ] Sur le nombre des invariants des familles de transformations arbitraires, II.
Polonaise de Mathématique), vol. 15 (année 1936, publ. 1937), p. 185. (Pre-
sented at the meeting of the Polish Mathematical Society, Warsaw Section, on
31 January 1936.)
[37a ] Numérotage des types logiques. Rocznik Polskiego Towarzystwa Matematy-
cznego (=Annales de la Société Polonaise de Mathématique), vol. 16 (année
1937, publ. 1938), p. 191. (Summary of a lecture given by Lindenbaum on 30
September 1937 to the Third Polish Mathematical Congress in Warsaw.)
[37a a] Sur l’équivalence de deux figures par décomposition en nombre fini de parties
respectivement congruentes. Rocznik Polskiego Towarzystwa Matematycznego
(=Annales de la Société Polonaise de Mathématique), vol. 16 (année 1937,
publ. 1938), p. 197. (Summary of a lecture given by Lindenbaum on 30
September 1937 to the Third Polish Mathematical Congress in Warsaw.)
[38a ] Sur les bases des familles de fonctions. Rocznik Polskiego Towarzystwa Matem-
atycznego (=Annales de la Société Polonaise de Mathématique), vol. 17 (année
1938, publ. 1939), pp. 124–126.
6.1.3 Reviews
[36r ] Tsao-Chen Tang, The theorem “p ≺ q. = .pq = p” and Huntington’s relation

between Lewis’s strict implication and Boolean algebra. Bull. Am. Math. Soc.
42 (1936), 743–746. Zentralblatt für Mathematik, vol. 15, no 4 (1936), p. 146.
[Zbl 0015.14601].
[36r a] Tsao-Chen Tang, A paradox of Lewis’s strict implication. Bull. Am. Math. Soc.
42 (1936), 707–709. Zentralblatt für Mathematik, vol. 15, no. 4 (1936), p. 146.
[Zbl 0015.14602].
[36r b] F. B. Fitch, A system of formal logic without an analogue to the Curry W
operator. Jour. Symb. Log. 1 (1936), 92–100. Zentralblatt für Mathematik, vol.
15, no. 6 (1936), p. 241. [Zbl 0015.24101].
[36r c] L. Kalmár, Zurückführung des Entscheidungsproblem auf des Fall von Formeln
mit einer einzigen, binären, Funktionsvariablen. Compositio Math. 4 (1936),
137–144. Zentralblatt für Mathematik, vol. 15, no. 8 (1936), pp. 338–339. [Zbl
0015.33804].
[37r ] W. Hetper, Bases de la sémantique. (Polish) Wiadom. Mat. 43 (1937), 57–86.
Zentralblatt für Mathematik, vol. 16, no. 5 (1937), p. 193. [Zbl 0016.19302].
[37r a] Z. Zawirski, Über die Anwendung der mehrwertigen Logik in der empirischen
Wissenschaft. Erkenntnis 6 (1937), 430–435. Zentralblatt für Mathematik, vol.
16, no. 5 (1937), p. 195. [Zbl 0016.19502].
[37r b] Jiro Hirano, Zum Zerlegungssatz im erweiterten einstelligen Prädikatenkalkül.
Proc. Phys.-Math. Soc. Japan, III, Ser. 19 (1937), 395–412. Zentralblatt für
Mathematik, vol. 16, no. 8 (1937), p. 337. [Zbl 0016.33702].
[38r ] V. F. von Seckendorff, Beweis des Induktionsschlusses der natürlichen Zahlen
aus der Dedekindschen Definition endlicher Mengen. Sitzungsber. Berlin. Math.
Ges. 36 (1937), 16–24. Zentralblatt für Mathematik, vol. 17, no. 4 (1938), p.
156. [Zbl 0017.15602].
[38r a] Carl G. Hempel, A purely topological form of non-Aristotelian logic. Jour.

Symb. Log. 2 (1937), 97–112. Zentralblatt für Mathematik, vol. 17, no. 6
(1938), p. 241. [Zbl 0017.24102].
[38r b] J.-L. Destouches, Groupe d’équivalence d’une théorie déductive. C. R. Acad.
Sci., Paris 205 (1937), 725–727. Zentralblatt für Mathematik, vol. 17, no. 6
(1938), p. 243. [Zbl 0017.24301].
[38r c] L. Kalmár, Zur Reduktion des Entscheidungsproblems. Norsk Mat. Tidsskr. 19
(1937), 121–130. Zentralblatt für Mathematik, vol. 17, no. 8 (1938), p. 337.
[Zbl 0017.33702].
[38r d] F. B. Fitch, Modal functions in two-valued logic. J. Symb. Log. 2 (1937), 125–
128. Zentralblatt für Mathematik, vol. 17, no. 8 (1938), pp. 337–338. [Zbl
0017.33703].
[38r e] L. P. Gokieli, Über den Funktionsbegriff. Trav. Inst. Math. Tbilissi 2 (1937), 1–
35. Zentralblatt für Mathematik, vol. 18, no. 1 (1938), p. 1. [Zbl 0018.00102].
[38r f] W. Wilkosz, Sur la notion de l’équivalence des systèmes déductifs. Annales de
la Soc. Polon. Math. 15 (1937), 161–164. Zentralblatt für Mathematik, vol. 18,
no. 1 (1938), pp. 2–3. [Zbl 0018.00204].
[38r g] Jiro Hirano, Einige Bemerkungen zum v. Neumannschen Axiomensystem der
Mengenlehre. Proc. Phys.-Math. Soc. Japan, III. Ser. 19 (1937), 1027–1045.
Zentralblatt für Mathematik, vol. 18, no. 1 (1938), p. 3. [Zbl 0018.00301].
[38r h] L. Couffignal, Sur un problème d’analyse mécanique abstraite: la théorie de la
réduction résulte de fonctions mécaniques. C. R. Acad. Sci., Paris 206 (1938),
1336–1338. Zentralblatt für Mathematik, vol. 18, no. 8 (1938), pp. 338–339.
[Zbl 0018.33805].
[38r i] W. Hetper, Le rôle des schémas indépendants dans le système de la sémantique
élémentaire. Arch. Towarz. Nauk. Lwów 9 (1938), 253–263. Zentralblatt für
[38r j] W. Hetper, Relations ancestrales dans le système de la sémantique. Arch.
Towarz. Nauk. Lwów 9 (1938), 265–280. Zentralblatt für Mathematik, vol. 19,
no. 4 (1938), p. 145. [Zbl 0019.14502].
[38r k] L. Chwistek, Remarques critiques concernant la notion de la variable dans le
système de la sémantique rationnelle. Arch. Towarz. Nauk. Lwów 9 (1939),
283–333. Zentralblatt für Matheatik, vol. 19, no. 4 (1938), p. 145. [Zbl
0019.14503].
[38r l] L. Couffignal, Solution générale, par des moyens mécaniques, des problèmes
fondamentaux de la logique déductive. C. R. Acad. Sci., Paris 206 (1938),
1529–1531. Zentralblatt für Mathematik, vol. 19, no. 4 (1938), p. 145. [Zbl
0019.14504].
[38r m] L. Couffignal, Les opérations des mathématiques pures sont toutes des fonctions
mécaniques. C. R. Acad. Sci., Paris 207 (1938), 20–22. Zentralblatt für
[39r ] J. J. Burckhardt, Zur Neubegründung der Mengenlehre. Jahresber. Dtsch. Math.-
Ver. 48 (1938), 146–165. Zentralblatt für Mathematik, vol. 19, no. 5 (1939), p.
201. [Zbl 0019.20101].
6.2 Works by Other Authors Referred to in This Paper
L.C. Arboleda
1981 Les Recherches de M. Fréchet, P. Alexandrov, W. Sierpiński et K. Kuratowski
sur la théorie des types de dimensions et les débuts de la topologie générale.
Archive for History of Exact Sciences, vol. 24, no. 4 (1981), pp. 339–388.
N. Aronszajn
1932 Sur les invariants des transformations continues d’ensembles. Fundamenta
Mathematicae, vol. 19 (1932), pp. 92–142.
S. Banach
1923 Sur le problème de la mesure. Fundamenta Mathematicae, vol. 4 (1923), pp.
7–33.
1924 Un théorème sur les transformations biunivoques. Fundamenta Mathemati-

cae, vol. 6 (1924), pp. 236–239.
S. Banach and A. Tarski

1924 Sur la décomposition des ensembles de points en parties respectivement
congruentes. Fundamenta Mathematicae, vol. 6 (1924), pp. 244–277.
(1) On decomposition of point sets into respectively congruent parts. In:
McFarland–McFarland–Smith [2014], pp.95–123. (English translation by
James T. Smith.)
S. Bender
1997 Mul mavet orev: Yehudei Bialystock be-milhemet ha-olam hasheniyah,
1939–1943, Am Oved Publishers Ltd: Tel Aviv 1997.
(1) The Jews of Białystok during World War II and the Holocaust, Brandeis
University Press: Waltham, Massachusetts 2008. (English translation by Yaffa
Murciano.)
G. Birkhoff
1944 Metric foundations of geometry, I. Transactions of the American Mathemat-
ical Society, vol. 55, no. 3 (May, 1944), pp. 465–492.
L. Couturat
1905 Les principes des mathématiques. Avec un appendice sur la philosophie des
mathématiques de Kant, F. Alcan: Paris 1905.
(1) Die philosophischen Prinzipien der Mathematik, Werner Klinkhardt:
Wien–Leipzig 1908. (German translation by Carl Siegel.)
J. Dadaczyński
2003 Hosiasson-Lindenbaumowa, Janina. Powszechna Encyklopedia Filozofii, vol.
4, Polskie Towarzystwo Tomasza z Akwinu: Lublin 2003, pp. 603–604.
S. Eilenberg
1993 Karol Borsuk—personal reminiscences, Topological Methods in Nonlinear
Analysis, vol. 1, no. 1 (1993), pp. 1–2.
R. Engelking
1989 General Topology. Revised and completed edition. Heldermann: Berlin 1989.
P. Erdős
1943 Some remarks on set theory. Annals of Mathematics, Series 2, vol. 44 (1943),
pp. 643–646.
H. Freudenthal and W. Hurewicz
1936 Dehnungen, Verkürzungen, Isometrien. Fundamenta Mathematicae, vol. 26
(1936), pp. 120–122.
F. Hausdorff
1914 Grundzüge der Mengenlehre. Veit & Comp: Leipzig 1914.
M.D. Kirszbraun
1934 Über die zusammenziehende und Lipschitzsche Transformationen. Funda-
menta Mathematicae, vol. 22 (1934), pp. 77–108.
J. Łoś
1949 O matrycach logicznych <On logical matrices. Polish>. Prace
Wrocławskiego Towarzystwa Naukowego, Seria B, vol. 42, Wrocław 1949.
A. McFarland, J. McFarland, James T. Smith (eds.)
2014 Alfred Tarski: Early Work in Poland—Geometry and Teaching. With a
Bibliographic Supplement. Birkhäuser: New York 2014.
E. Marczewski and A. Mostowski
1971 Lindenbaum Adolf (1904–1941). Polski Słownik Biograficzny <Polish Bio-
graphical Dictionary>, vol.17, Kraków 1971, pp. 364b–365b.
A. Marianowicz
1995 Życie surowo wzbronione. Czytelnik: Warszawa 1995.
(1) Life Strictly Forbidden. Valentine Mitchell: London 2004. (English trans-
lation by Alicja Nitecki.)
S. Mazurkiewicz and W. Sierpiński
1914 Sur un ensemble superposable avec chacune de ses deux parties. Comptes
Rendus hebdomadaires des séances de l’Académie des Sciences, Paris,
vol. 158 (1914), pp. 618–619. (Stefan Mazurkiewicz uses the name Etienne
Mazurkiewicz in French.)
H. Moese
1932 Przyczynek do problemu A. Tarskiego: “O stopniu równoważności
wielokatów”.
˛ Parametr, vol. 2 (1931–1932), pp. 305–309.
(1) A Contribution to the problem of A. Tarski: “On the degree of equivalence
of polygons”, in: McFarland–McFarland–Smith [2014], pp. 145–151. (English
translation by A. McFarland, J. McFarland and James T. Smith.)
S. Ruziewicz
1921 Sur un ensemble non dénombrable de points, superposable avec les moitiés de
sa partie aliquote. Fundamenta Mathematicae, vol. 2 (1921), pp. 4–7.
S. Ruziewicz and W. Sierpiński
1932 Sur un ensemble parfait qui a avec toute sa translation au plus un point
commun. Fundamenta Mathematicae, vol. 19 (1932), pp. 17–21.
W. Sierpiński
1918 L’axiome de M. Zermelo et son rôle dans la théorie des ensembles et l’analyse.
Bulletin International de l’Académie des Sciences de Cracovie. Classe des
Sciences Mathématiques et Naturelles. Serie A, Sciences Mathématiques,
Comptes-Rendus des Séances de l’année 1918, pp. 97–152.
1933 Sur les espaces métriques localement séparables. Fundamenta Mathemati-

cae, vol. 21 (1933), pp. 107–113.
1936 Sur une propriété du segment. Prace matematyczno-fizyczne, vol. 43 (1936),

pp. 25–32.
1947 Sur un théorème de A. Lindenbaum. Annals of Mathematics, Series 2, vol. 8

(1947), pp. 641–642.
1947a Sur la congruence des ensembles de points et ses généralisations. Commen-
tarii Mathematici Helvetici, vol. 19 (1947), pp. 215–226.
1947b Sur un ensemble plan qui se décompose en 2ℵ0 ensembles disjoints superpos-
ables avec lui. Fundamenta Mathematicae, vol. 34 (1947), pp. 9–13.
1954 On the Congruence of Sets and their Equivalence by Finite Decomposition.

Lucknow University Studies, no. XX. The Lucknow University: Lucknow
1954.
1958 Cardinal and Ordinal Numbers. Monografie Matematyczne, vol. 34, PWN–
Polish Scientific Publishers, Warszawa 1958.
(1) Second edition revised, 1965.
B. Sobociński
1956 On well constructed axiom systems. Rocznik VI Polskiego Towarzystwa
Naukowego na Obczyźnie <Yearbook VI of the Polish Society of Arts and
Sciences Abroad>, London 1955–56, pp. 54–65.
A. Tarski
1924 O równoważności wielokatów
˛ <On the equivalence of polygons. Polish with
French summary>. Przeglad ˛ matematyczno-fizyczny, vol. 2 (1924), pp. 47–60.
(1) On the equivalence of polygons, in: McFarland–McFarland–Smith [2014],
pp. 79–91. (English translation by A. McFarland, J. McFarland and James T.
Smith.)
1928 Remarques sur les notions fondamentales de la méthodologie des mathéma-
tiques. Rocznik Polskiego Towarzystwa Matematycznego (=Annales de la
Société Polonaise de Mathématique), vol. 7 (année 1928, publ. 1929), pp.
270–272.
(1) Remarks on fundamental concepts of the methodology of mathematics.
In J.-Y. Béziau (ed.), Universal Logic: An Anthology, Birkhauser 2012, pp.
67–68. (English translation by R. Purdy and J. Zygmunt.)
1930 Über Äquivalenz der Mengen in Bezug auf eine beliebige Klasse von Ab-
bildungen. Atti del Congersso internationale di Matematici, Bologna, 3–10
settembre 1928, vol. 6, Nicola Zanichelli: Bologna 1930, pp.243–252.
1930a Über einige fundamentalen Begriffe der Metamathematik. Sprawozdania
z posiedzeń Towarzystwa Naukowego Warszawskiego, Wydział III Nauk
Matematycznych i Przyrodniczych (=Comptes rendus des séances de la
Societé des Sciences et des Lettres de Varsovie, Classe III), vol. 23 (1930),
pp. 22–29.
(1) On some fundamental concepts of metamathematics. In: Tarski [1956], pp.
30–37. (Revised English translation of Tarski [1930].)
1930b Fundamentale Begriffe der Methodologie der deduktiven Wissenschaften. I.
Monatshefte für Mathematik und Physik, vol. 37 (1930), pp. 361–404.
(1) Fundamental concepts of the methodology of the deductive sciences. In:
Tarski [1956], pp. 60–109. (Revised English translation of Tarski [1930a].)
1931 O stopniu równoważności wielokatów˛ <On the degree of equivalence of
polygons. Polish>. Młody Matematyk, vol. 1 (supplement to Parametr, vol.
2) (1931), pp. 37–44. [Exercise 6 on p. 44 is due to A. L. = most probably
Adolf Lindenbaum.]
(1) Reprinted, and English translation by I. Wirszup, in: Alfred Tarski, Col-
lected Papers. Volume 1, 1921–1934, ed. by S.R. Givant and R.N. McKenzie,
Birkhäuser Verlag: Basel 1986, pp. 563–570, and pp. 573–580.
(2) On the degree of equivalence of polygons. In: McFarland–McFarland–
Smith [2014], pp. 135–144. (English translation by A. McFarland, J. McFar-
land and James T. Smith.)
1931/32 Uwagi o stopniu równoważności wielokatów ˛ <Remarks on the degree of
equivalence of polygons. Polish>. Parametr, vol. 2 (1931–32), pp. 310–314.
(1) Reprinted, and English translation titled “Further remarks about the degree
of equivalence of polygons” by I. Wirszup, in: Alfred Tarski, Collected
Papers. Volume 1, 1921–1934, ed. by S.R. Givant and R.N. McKenzie,
Birkhäuser Verlag: Basel 1986, pp. 597–601, and pp. 605–611.
(2) Remarks on the degree of equivalence of polygons. In: McFarland–

McFarland–Smith [2014], pp. 152–158. (English translation by A. McFarland,
J. McFarland and James T. Smith.)
1936 Grundzüge des Systemenkalküls. Zweiter Teil. Fundamenta Mathematicae,

vol. 26 (1936), pp. 283–301.
(1) Foundations of the calculus of systems. In: Tarski [1956], pp. 342–383.
(Revised English translation of Tarski [1936].)
1956 Logic, Semantics, Metamathematics. Papers from 1923 to 1938. (Translated

by J.H. Woodger.) Clarendon Press: Oxford 1956.
(1) Second, revised edition, with editor’s introduction and an analytic index,
J. Corcoran (ed.), Hackett Publishing Company: Indianapolis, Indiana 1983.
M. Tomkiewicz
2008 Zbrodnia w Ponarach 1941–1944. Monografie Komisji Ścigania Zbrodni
Przeciwko Narodowi Polskiemu, vol. 43. Instytut Pami˛eci Narodowej:
Warszawa 2008.
S. Wagon
1993 The Banach–Tarski Paradox. Paperback edition. Cambridge University
Press: Cambridge 1993.
W. Wi˛esław
2004 Listy Wacława Sierpińskiego do Stanisława Ruziewicza <Letters from
Wacław Sierpiński to Stanisław Ruziewicz. Polish>. Wiadomości Matem-
atyczne, vol. 40 (2004), pp. 139–167.
B. Wolniewicz
2015 Nota do biogramu Lindenbauma <A note on the life of Lindenbaum. Polish>.
Edukacja Filozoficzna, vol. 60 (2015), pp. 163–164.
J. Zygmunt and R. Purdy
2014 Adolf Lindenbaum: Notes on his life, with bibliography and selected refer-
ences. Logica Universalis, vol. 8, no. 3–4 (2014), pp. 285–320.
7 Acknowledgments
7.1 Institutions, Digital Sources and Individuals
The authors are immensely grateful for library and archival services and invaluable advice
and expertise provided by the following institutions, digital sources and individuals.
7.1.1 Universities
Archiwum Uniwersytetu Warszawskiego

Muzeum Uniwersytetu Warszawskiego; Jan Łukasiewicz fonds
Cambridge University Library
Bibliothek der Universität Konstanz
McMaster University Archives
University of Oxford, Bodleian Library
SUNY at Albany, M.E. Grenander Dept. of Special Collections & Archives
UC Berkeley, Bancroft Library
University of Pittsburgh, Hillman Library
Uniwersytet Wrocławski, Biblioteka Wydziału Nauk Społecznych
7.1.2 Other Archives
Archiwum Państwowe w Białymstoku

Archiwum Państwowe w Warszawie
Archiwum Akt Nowych
Archiwum Polskiej Akademii Nauk
Biblioteka Narodowa
Instytut Pami˛eci Narodowej
New York Public Library
The Rockefeller Archive Center
U.K. National Archives
Noord-Hollands Archief, Haarlem; Wiener Kreis Archief collection
§ÊâÓÉÃy̆ÐÞ ÃÓØiy̆ ¦ÓÑÇÊÈÐÔÍÃÌ ÅÑÄÎÃÔÙi (State Archives of Grodno Oblast)
7.1.3 Digital Sources
e-biblioteka Uniwersytetu Warszawskiego (e-bUW)

Jewish Records Indexing—Poland
Kujawsko-Pomorska Digital Library
Mazowiecka Biblioteka Cyfrowa
Narodowe Archiwum Cyfrowe
The Polish Digital Mathematics Library
Wielkopolska Biblioteka Cyfrowa
Yad Vashem Database
7.1.4 Individuals
Apart from persons mentioned in the text, particular thanks for their help in preparing this
article are due to:
Bethany J. Antos—The Rockefeller Archive Center
Brigitta Arden—Hillman Library, University of Pittsburgh
Godelieve Bolten—Noord-Hollands Archief, Haarlem (Wiener Kreis Archief )

Danuta Bondaryk—Archiwum Państwowe w Warszawie
Frank Bowles—Cambridge University Library
Jacqueline Cox—University of Cambridge, University Archives
Jerzy Dadaczyński—Uniwersytet Papieski Jana Pawła II w Krakowie
Anna Dziedzic—Archiwum Uniwersytetu Warszawskiego
Renata Grzegórska—Archiwum Uniwersytetu Warszawskiego
Colin Harris—University of Oxford, Bodleian Library
Arie Hinkis—Cohn Institute, Tel Aviv University
Danuta Kamińska—Archiwum Państwowe w Warszawie
Brian Keough—SUNY at Albany, M.E. Grenander Dept of Special Collections &
Archives
Marek Kietliński—Archiwum Państwowe w Białymstoku
Jean Lambert—CTC & Hughes Hall, University of Cambridge
Magdalena Masłowska—Archiwum Państwowe w Warszawie
Mariusz Pandura—Uniwersytet Wrocławski, Biblioteka Wydziału Nauk Społecznych
Brigitte Parakenings—Bibliothek der Universität Konstanz
Janusz Rudziński—Muzeum Uniwersytetu Warszawskiego
Wojciech Śleszyński—Uniwersytet w Białymstoku
Monika Tomkiewicz—Instytut Pami˛eci Narodowej. Oddział w Gdańsku
Piotr Wojtylak—Uniwersytet Opolski
Stephen Wordsworth—Council for At-Risk Academics, London South Bank University
Witold Żarnowski—Muzeum Niepodległości w Warszawie
7.2 Photograph Credits
7.2.1 Photographs of Adolf Lindenbaum (1922 and 1927), Janina Hosiasson (1919),
and Stefanja Lindenbaum (1926) by kind permission of the University of Warsaw
Archives.
7.2.2 Photograph of the three Pański brothers (circa 1920) by kind permission of
Spółdzielnia Wydawnicza ,Czytelnik” (Czytelnik Publishing House), Warsaw.
7.2.3 Photograph of the first congress of mathematicians from Slavic countries (1929) is
reproduced from Comptes-Rendus du I Congrès des Mathématiciens des Pays
Slaves, Warszawa 1929, edited by Franciszek Leja, Ksiażnica
˛ Atlas T.N.S.W.,
Warszawa 1930.
7.2.4 Photograph of the meeting of the mathematics, physics and astronomy circles
(1932) by kind permission of the Museum of the University of Warsaw, Jan
Łukasiewicz Fonds.
7.2.5 Photograph of the mathematics and physics faculty members of the Pedagogical
Institute of Białystok (1941) by kind permission of the National Archives in
Białystok.
7.2.6 Photograph of the façade of the Pedagogical Institute of Białystok (1940) by kind
permission of the Byelorussian Telegraph Agency (¶P „¤¨®µ£”).
© Copyright ownership in each instance resides with the above-named provider(s) of the
photograph(s).
R. Purdy
Tranby Avenue, Toronto, Ontario M5R 1N5, Canada
J. Zygmunt ()
University of Wrocław, Wrocław, Polska
Andrzej Mostowski: A Biographical Note
Marcin Mostowski
Abstract This is a short biography of the distinguished representative of Warsaw School

of Logic and his activity in post-war Poland.
Keywords Andrzej Mostowski · Axiom of choice · Skolem arithmetic · Generalized

quantifiers · Foundations of mathematics
Mathematics Subject Classification (2000) Primary 01A70, 01A60
1 Life
Andrzej Stanisław Mostowski was born in Lwów,1 on the first November 1913. He died
on the 22nd of August 1975 in Vancouver, British Columbia, Canada. His parents were
Zofia Kramstyk and Stanisław Mostowski. In 1920 his family moved to Warsaw.
In 1938 Mostowski defended his PhD thesis O niezależności definicji skończoności w
systemie logiki (On the Independence of Finiteness Definitions in a System of Logic) at
Warsaw University under supervision of Kazimierz Kuratowski and Alfred Tarski.
Just after the war he obtained habilitation degree at Jagiellonian University in Kraków
for the thesis Axiom of choice for finite sets. After graduation he got an university position
in Łódź.2 He moved to Warsaw in 1947, and then he created one of the most influential
schools of mathematical logic in the world at the Warsaw University.
2 Main Works
Selected most important research papers:
The author “Marcin Mostowski” is deceased.

1 In this time it was Austro–Hungarian part of Poland, then Poland, and nowadays Ukraine.
2 In this time Warsaw was completely destroyed and many institutions were temporarily moved to Łódź.

552 M. Mostowski
• Über die Unabhängigkeit des Wohlordnungssatzes vom Ordnungsprinzip, [2];

• On direct products of theories, [5];
• On a generalization of quantifiers, [7].
The most important research and survey books:
• Sentences Undecidable in Formalized Arithmetic: An Exposition of the Theory of Kurt
Godel, [6],
• Thirty years of foundational studies. Lectures on the development of mathematical
logic and the study of the foundations of mathematics in 1930–1964, [8]
• Constructible sets with applications, [9].
All of the mentioned works and many others are reprinted in [10] and [11]. He
also published two important monographs Logika Matematyczna [3] and jointly with
Kazimierz Kuratowski Teoria Mnogości [1]. They were very influential and useful as
university handbooks.
3 Main Scientific Achievements
The best known theorem proved by Andrzej Mostowski is the following:

Theorem 1 (Mostowski Collapse Lemma, [4]) Let us assume that a binary relation
R ⊆ A2 is well founded (∀B ⊆ A(B = ∅ ⇒ ∃x ∈ B(x is R-minimal in B))) and
extensional (∀x, y ∈ A(x = y ≡ ∀z(zRx ≡ zRy))), then there is unique transitive
set Z (∀x ∈ Z x ⊆ Z) and a bijection π : A −→ Z such that
∀x, y ∈ A(xRy ≡ π(x) ∈ π(y)).
Other broadly known ideas of Andrzej Mostowski are:

• Kleene–Mostowski hierarchy, classification of arithmetical sets via quantifier prefixes
required for defining them;
• the notion of generalized quantifier, formulated by him in [7].
However, some of less known results of Andrzej Mostowski are also very important and
influential. Let us mention two of them.
One of a few hard important3 results obtained before the second world war was the
following:
Theorem 2 ([2]) In Zermelo–Fraenkel set theory with atoms (ZFA) Zermelo theorem,
saying that every set can be well ordered, is independent of Ordering Principle, saying
that every set can be linearly ordered.
The proof of this theorem was based on the method proposed by Abraham Fraenkel.
Later on Andrzej Mostowski elaborated the method (see [9]). The basic construction is
nowadays called Fraenkel–Mostowski permutation models.
3 Let us observe that Andrzej Mostowski was 26 years old, when the result was published.
Andrzej Mostowski: A Biographical Note 553
Another important work was the method of obtaining decidability results for theories
of the product of two models having decidable theories, see [5]. As a corollary it gives
elimination of quantifiers for arithmetic of multiplication of natural numbers. This is
the first published proof of the theorem that first order arithmetic of natural numbers
is decidable. The theorem is commonly attributed to Thoralf Skolem.
4 Influence of His Works
In mathematical logic ideas of Andrzej Mostowski influenced mainly foundations of set

theory. He was also one of the important builders of model theory. Some of his ideas
are also present in the recursion theory. However, he was thinking rather in terms of
definability than computations.
In logic, in larger sense, Mostowski’s most influential idea was the concept of
generalized quantifiers. It is studied now not only by mathematicians, but also by
philosophers, linguists, and computer scientists.
5 Students
He promoted the following PhD students: Zofia Adamowicz, Krzysztof Apt, Maciej
Bryński, Andrzej Ehrenfeucht, Andrzej Grzegorczyk, Wojciech Guzicki, Andrzej Jan-
iczak, Stanisław Krajewski, Michał Krynicki, Moshe Machover, Mihály Makkai, Wiktor
Marek, Janusz Onyszkiewicz, Helena Rasiowa, Roman Sikorski, Kazimierz Wiśniewski,
Andrzej Zarach, Pawel Zbierski.
Probably, the most influential in logic were Andrzej Ehrenfeucht, Andrzej Grzegor-
czyk, and Helena Rasiowa.
6 Last Words
Andrzej Mostowski founded one of the greatest schools of foundations of mathematics.

How great and important he was becomes clear if on considers what happened later with
his group at Warsaw University. It completely disappeared.
Acknowledgement This work was supported by the Polish National Science Centre [2013/11/B/HS1/
04168].
554 M. Mostowski
References
1. Kuratowski, K., Mostowski, A.: Teoria mnogości: wraz ze wst˛epem do opisowej teorii mnogości.
Monografie Matematyczne, 27. Państwowe Wydawnictwo Naukowe (PWN), Warszawa (1952)
2. Mostowski, A.: Über die Unabhängigkeit des Wohlordnungssatzes vom Ordnungsprinzip. Fundam.
Math. 32(1), 12–36 (1939)
3. Mostowski, A.: Logika Matematyczna. Monografie Matematyczne, 18. Państwowe Wydawnictwo
Naukowe (PWN), Warszawa (1948)
4. Mostowski, A.: An undecidable arithmetical statement. Fundam. Math. 36(1), 143–164 (1949)
5. Mostowski, A.: On direct products of theories. J. Symb. Log. 17(1), 1–31 (1952)
6. Mostowski, A.: Sentences Undecidable in Formalized Arithmetic: An Exposition of the Theory of
Kurt Godel. North-Holland, Amsterdam (1952)
7. Mostowski, A.: On a generalization of quantifiers. Fundam. Math. 44(1), 12–36 (1957)
8. Mostowski, A.: Thirty years of foundational studies. Lectures on the development of mathematical
logic and the study of the foundations of mathematics in 1930–1964. Acta Philosophica Fennica,
Fasc. XVII. Barnes and Noble, Inc., New York (1966)
9. Mostowski, A.: Constructible sets with applications. Studies in Logic and the Foundations of Math-
ematics. North-Holland Publishing Co., Amsterdam; PWN-Polish Scientific Publishers, Warszawa
(1969)
10. Mostowski, A.: Foundational Studies, Vol. I. PWN, North-Holland (1979)
11. Mostowski, A.: Foundational Studies, Vol. II. PWN, North-Holland (1979)
Foundations and Philosophy of Mathematics
in Warsaw, the School of Andrzej
Mostowski and Philosophy
Marcin Mostowski
Abstract The paper is a mathematical and philosophical essay devoted to mathematical

logic school created and guided by Andrzej Mostowski. Firstly, we discuss some of the
main—still actual—achievements of Andrzej Mostowski, then we discuss weak sides of
his scientific project. They are computational and philosophical.
Scientific challenges of our times in logic are mainly computational. The Warsaw
school of mathematical logic did not support this direction.
Partially because of the political situation in Poland public philosophical discussions
were strongly influenced by hard politics, including personal politics in academic
institutions. Therefore, many people think that isolation of the foundations of mathematics
and philosophy was forced by the communist ideology. This impression is false. The
abyss between philosophy and the foundations was basically independent of the political
situation in Poland.
What can we learn from this experience?
Keywords Andrzej Mostowski · Axiom of Choice · Skolem arithmetic · Generalized

quantifiers · Foundations of mathematics · Philosophy of mathematics
Mathematics Subject Classification (2000) Primary 00A30, 03A05, 00A35; Secondary

03E25, 03C80, 03F30
1 Introduction
This work can be divided into two parts. The first one is devoted to three selected topics of
Andrzej Mostowski’s work in logic. Undoubtedly, I could not cover all his important and
influential works. The selection is partially personal and partially based on my opinion
what was characteristic to his scientific interests.
The author “Marcin Mostowski” is deceased.

556 M. Mostowski
The second part is a philosophical and historical essay devoted to relations of his school
to philosophy and a new way of computational thinking. In comparison to the first part it
is much more personal and informal.
2 Some Mathematical Works in the Foundations
We discuss here some selected ideas of Andrzej Mostowski.1
2.1 Permutation Models
One of the most fascinating idea in foundation of mathematics of twentieth century was
undoubtedly the Axiom of Choice (AC).2 AC says that for every family F of nonempty
sets there is a choice function f : F −→ F such that ∀A ∈ F f (A) ∈ A. Zermelo
invented it trying to prove his theorem, TZ: Every set can be well ordered. He proved TZ
assuming AC. However, having TZ it is easy to prove AC. For any family F of nonempty
sets, if we take any well ordering R on F then we can define a choice function taking
f (A) = the R-smallest element of A, for all A ∈ F .
The first question related to AC was whether it can be proved in Zermelo–Fraenkel
set theory (ZF). A partial answer was known relatively early. However it was done only
for the theory ZF extended by allowing elements not being sets, so called atoms, it is
called Zermelo–Fraenkel
set theory with atoms (ZFA).3 Models of ZFA are of the form
V (A) = α Vα (A), where V0 (A) = A and for all ordinal numbers
α we take Vα+1 (A) =
P (Vα (A)) ∪ Vα (A), for a limit ordinal λ we take Vλ (A) = α<λ Vα (A). We assume that
the set A is infinite.
Let us consider any permutation σ of the set A, that is a bijection σ : A −→ A. It
can be extended on all sets from the model by taking B σ = {x σ : x ∈ B}. We say that
B is stable for σ if B σ = B. For any set Z ⊆ A we say that a permutation σ fixes Z if
σ (a) = a, for every a ∈ Z.
Now we define S(A) a subuniverse of V (A) taking all sets B from V (A) such that
there is a finite Z ⊆ A such that for every permutation σ of the set A which fixes Z, B
is stable for σ . Of course all pure sets, having no atoms in their transitive closures, are
stable for any permutation of A, then all of them belong to S(A). Additionally all finite
and co-finite subsets of A belong to S(A). On the other hand no infinite and co-infinite
B ⊆ A belong to S(A). Because for any finite Z ⊆ A we can find a, b ∈ A − Z such
that a ∈ B and b ∈ B, then we take a transposition σ exchanging a and b and not moving
anything else. σ fixes Z and B σ = B.
The next step is proving that S(A) is a model for ZFA provided V (A) is a model for
ZFA. The class S(A) would not contain any well ordering of A. Otherwise, having a well
1 Some of the other topics are discussed in part 4 of this book.

2A very good monography of the topic can be found in [6].
3 A good introduction to ZFA can be found in [5].
Foundations and Philosophy of Mathematics 557
ordering R on A we can split A into two disjoint sets B containing even successors in the
sense R and the remaining part containing odd successors.
In this way we gave a sketchy proof of the following:
Theorem 1 If ZFA is consistent then neither AC nor TZ are provable in ZFA.
By refinement of the method of Fraenkel, Andrzej Mostowski proved the following:
Theorem 2 ([9]) In Zermelo–Fraenkel set theory with atoms (ZFA) Zermelo theorem,
saying that every set can be well ordered, is independent of Ordering Principle, saying
that every set can be linearly ordered.
Later on he elaborated the method. His book [11] presents the results. Nowadays the basic
construction is called Fraenkel–Mostowski permutation models.
These prewar ideas strongly influenced later research in the school of Andrzej
Mostowski for many years.
2.2 Skolem Arithmetic and Direct Products of Theories
In 1952 Andrzej Mostowski published the paper “On direct products of theories” reprinted
in [12]. It contains the first published proof of the theorem commonly attributed to Thoralf
Skolem, saying that first order arithmetic of multiplication is complete and decidable.
Neither the idea nor the details of the proof are well known. In 1929 Mojżesz
Presburger published his paper presenting his proof of decidability of first order arithmetic
of addition [14]. The proof can be found in almost every handbook of mathematical logic.4
It is one of the paradigmatic proofs by elimination of quantifiers.
The structure (P <ω (N), ∪, −, ∅) of finite sets of natural numbers with union and
difference has much more simpler first order theory TI = T h(P <ω (N), ∪, −, ∅). As
a matter of fact this theory is simply the theory of nonprincipal maximal ideals in atomic
infinite boolean algebras.5
Well known axioms of first order Peano arithmetic P A, with 0, S, +, × as primitive
notions, are the following:
(PA1)∀x (x = 0 ≡ ¬∃y x = S(y),
(PA2)∀x, y (S(x) = S(y) ⇒ x = y),
(PA3)∀x x + 0 = x,
(PA4)∀x, y x + S(y) = S(x + y),
(PA5)∀x x × 0 = 0,
(PA6)∀x, y x × S(y) = (x × y) + x
and the induction axiom scheme, for each arithmetical formula ϕ(x1 , . . . , xn , y):
(PA7ϕ )∀x1 , . . . , xn (ϕ(x1 , . . . , xn , 0)∧ ∀y (ϕ(x1 , . . . , xn , y) ⇒ ϕ(x1 , . . . , xn , S(y)))
⇒ ∀y ϕ(x1, . . . , xn , y)).
4 Professor Andrzej Grzegorczyk at least twice told me that Alfred Tarski decided that the result by
Presbuger was too weak for PhD. Undoubtedly intending this terrible mistake as a lesson for future
supervisors.
5 A proper boolean algebra can be obtained by adding all the complements of the elements of an ideal.
558 M. Mostowski
The arithmetic of addition, called also Presburger arithmetic, TP can be axiomatized by

(PA1)–(PA4) and (PA7ϕ ) restricted to formulae with 0, S, + as only primitive notions.
The arithmetic of multiplication, called also Skolem arithmetic TS , cannot be so easily
extracted from axioms of P A. However it can be defined as the set of all first order
consequences of P A which contain only multiplication and all quantifiers are of the form
∀x = 0 and ∃x = 0.6 On the other hand, we define TM as the first order theory of the
structure (N, ×). Of course TS ⊆ TM .
Theories TI and TP allow elimination of quantifiers. In the paper “On direct products of
theories” Andrzej Mostowski proves that this two methods give elimination of quantifiers
for TM . His theorem is essentially more general. It gives elimination of quantifiers for TS
as a corollary. It was generalized in the work [4].
However we are interested mainly in properties of TS . The method was elaborated for
this case by Patrick Cegielski [1]. Cegielski gives also an axiomatic characterization of
Skolem arithmetic, see also [15]. However the argument is still complicated.
The idea of the proof is based on so called Prime Factorization Theorem which says
that for each integer a > 0 there are uniquely determined a set of prime divisors of a:
Supp(a) = {q1 < q2 < . . . < qn } and a sequence of positive integers a1 , a2 , . . . , an such
that
a = q1a1 q2a2 . . . qnan .
The result of multiplication of a by
b = r1b1 r2b2 . . . rm
bm
is
c
c = s1c1 s2c2 . . . sk k ,
where Supp(b) = {r1 < r2 < . . . < rm }, Supp(c) = {s1 < s2 < . . . < sk }, Supp(c) =
Supp(a) ∪ Supp(b) and for i = 1, 2, . . . , k the exponent ci is either the sum of exponents
ai + bi , if si = qi = ri , or ci is one of ai or bi , if si divides only one of a, b.
Therefore each model M = (U, ×M ) of TM can be split into the model (I M , ∩, −, ∅)
for TI and the family of models (Mp )p∈P rimes M for TP , where I M = {Supp(a) ⊆
P rimes M : a ∈ U } and Mp = {Component (p, a) : a ∈ U }, Component (p, a) is
the greatest power of the prime p dividing a.
Reception of the proof of Andrzej Mostowski and later refinements is very poor
and in what follows we will discuss easier and less general argument for completeness
and decidability of TS . Because it is an axiomatic theory then it suffices to prove its
completeness.
6 Zero element in T is easy to define and inessential from the point of view of characterizing models of
S
this theory.
Much easier proof was given by Nadel [13], who proved the completeness of Skolem
arithmetic by using Ehrefeucht–Fraïssé games, [3].7
Theorem 3 Any two models of TS are elementary equivalent.
The proof given by Nadel combines winning strategies of ∃-player for two models of TI
and two models of TP for obtaining the same for two models of TS . As a byproduct it
gives that TM ⊆ TS .
This can be done simply by proving in P A proper statements justifying possibility of
needed representation. We give two examples of them.
Lemma 1 (Prime Factorization Theorem: A Multiplicative Version) In P A the
following statement is provable:
∀x∀y (x = y ≡ ∀p ∈ P rimes Component (p, x) = Component (p, y)).
Because the statement uses only multiplicative language then it is also provable in TS .
Lemma 2 (Selection: A Multiplicative Version) For each arithmetical first order
formula ϕ(x1 , . . . , xn , y, z) The following statement is provable in P A:
∀x1 , . . . , xn ∀b (∀p ∈ Supp(b)∃a ((P ow(p, a) ∧ ϕ(x1 , . . . , xn , p, a)) ⇒
∃b (Supp(b) = Supp(b ) ∧ ∀p ∈ Supp(b ) ϕ(x1 , . . . , xn , p, Component (p, b )))).
If additionally ϕ is in multiplicative language of TS then the statement is also provable

in TS .
Proof Let us fix a formula ϕ(x1, . . . , xn , y, z). Then we prove the statement in P A by
induction on the standard enumeration of primes p0 , p1 , p2 , . . .. Firstly we take any
x1 , . . . , xn , b.
We take as b0 any element such that P ow(p0 , b0 ) ∧ ϕ(x1 , . . . , xn , p0 , b0 ),8 or b0 = 1
if ¬ p0 | b. Let us assume that ∀p ∈ Supp(b)∃a ((P ow(p, a) ∧ ϕ(x1 , . . . , xn , p, a)).
Let us assume that bn is defined in such a way that ∀i ≤ n∀pi ∈ Supp(b ) ϕ(x1 , . . . ,
xn , p, Component (pi , bn )).
If ¬ pn+1 | b then we take bn+1 = bn , otherwise—by the assumption—we have a
such that P ow(pn+1 , a) ∧ ϕ(x1 , . . . , xn , pn+1 , a). In this case we take bn+1 = bn × a.
We stop at stage n when pn is the greatest prime in Supp(b), then we take b = bn . ,
*
7 The method is currently a standard one and can be found in many textbooks of mathematical logic. A
good presentation of it can be found e.g. in [2].
8 P ow(q, d) means that d is a power of a prime q.
560 M. Mostowski
2.3 Generalized Quantifiers
Another very influential work by Andrzej Mostowski was presented in his paper “On
a generalization of quantifiers”[10]. The paper does not contain any hard results, but
it presents a new very influential idea of generalized quantifiers. Traditionally, in
mathematics we have used two quantifiers: universal ∀ and existential ∃. In the Type
Theory they have many interpretations depending on the types of variables they can
bound. However, restricting our attention to first order—elementary language, they have
unique interpretation.
2.3.1 Basic Idea
Let us consider a model M with the universe U . Interpretations of quantifiers ∀ and ∃

depend only on U . So we take
∀U = {A ⊆ U : U − A = ∅},
and
∃U = {A ⊆ U : A = ∅}.
Now for every formula ϕ(x) with one free variable x we have:
M | Qx ϕ(x) if and only if {a ∈ U : M | ϕ(a)} ∈ QU ,
for any of quantifiers Q = ∀, ∃.

Are there more such quantifiers? Restricting our attention only to logical—topic
independent—notions, there are much more than two. Logicality condition for Q says
that for any bijection f : U −→ W and for any A ⊆ U :
A ∈ QU if and only if f (A) ∈ QW ,
where f (A) = {f (a) : a ∈ A}. Quantifiers satisfying this condition are determined by
classes of pairs of cardinal numbers, in the following sense KQ = {(card(A), card(U −
A)) : A ∈ QU }. For instance K∃ = {(κ1 , κ2 ) : κ1 > 0}. On the other hand QU can be
obtained back from KQ as follows:
QU = {A ⊆ U : (card(A), card(U − A)) ∈ KQ }.
The basic notion of a generalized quantifier is a quantifier in the above sense satisfying
the logicality condition. They are called also Mostowski quantifiers
Andrzej Mostowski observed that logics with such quantifiers are hard to axiomatize.
For instance the quantifier there are only finitely many—∃<ℵ0 , defined as
∃<ℵ0 U = {A ⊆ U : card(A) < ℵ0 },

allows finite axiomatization of the standard model of natural numbers, what means that,
by the Tarski undefinability of truth theorem, the set of theorems cannot be arithmetical,
therefore it is not axiomatizable.
The axiomatization can be given by (PA1)–(PA6) and instead of scheme (PA7ϕ ) we
take ∀x ∃<ℵ0 y y < x.9
Therefore, the result obtained by H. Jerome Keisler [7] giving an example of
axiomatizable logic was surprising. He gave a complete axiomatization for the logic with
the quantifier ∃>ℵ0 , defined as
∃>ℵ0 U = {A ⊆ U : card(A) > ℵ0 }.
Let us consider the logic F O(Q) first order logic with an additional quantifier Q
interpreted as arbitrary Mostowski quantifier. It means that formulae are interpreted in
models of the form (M, QU ), where M is a usual model and U is its universe.
Theorem 4 (Per Lindström10 ) The set of all universally valid arithmetical formulae in
F O(Q) is not arithmetical and therefore not axiomatizable.
Proof Let ϕ be a conjunction of (PA1)–(PA6) and the statement
ψ = ∀x(Qy y < x ≡ ¬Qy y < S(x)).
Let us observe that ψ says that Q gives different truth values on initial segments
determined by a and by S(a), for all a, what is possible only when these segments are
finite. Therefore for each arithmetical sentence ξ the conjunction (ϕ ∧ ξ ) is consistent in
F O(Q) if and only if ξ is true in the standard model of natural numbers.11 ,
*
2.3.2 Lindström’s Generalization
Currently the term generalized quantifiers is used in the sense given by Per Lindström in
[8]. For every finite sequence t = (t1 , . . . , tk ) of positive integers, we define a generalized
quantifier Q of type t as follows:
• On a syntactic level, for each formulae ϕ1 , . . . , ϕk , we have a new formula
Qx(ϕ1 (x1 ), . . . , ϕk (xk )), where x = x1 , . . . , xt , t = max(t1 , . . . , tk ) and xi =
x1 , . . . , xti , for i = 1, . . . , k;
• On a semantic level, for each nonempty U the set QU contains k-tuples (R1 , . . . , Rk ),
where Ri ⊆ U ti , for i = 1, . . . , k. Additionally we assume that the logicalness
condition is satisfied, in the following sense: for each bijection f : U −→ W and
9 The ordering is defined by x ≤ y ≡ ∃z x + z = y and x < y means x ≤ y ∧ x = y.

10 Personal communication at the joint meeting LMPhS and LC in Uppsala, 1991.
11 The argument given here is mine, but Per Lindström gave an argument in a similar style in a
conversation with Michał Krynicki and me (see footnote 10).

562 M. Mostowski
for each k-tuple (R1 , . . . , Rk ):
(R1 , . . . , Rk ) ∈ QU if and only if (f (R1 ), . . . , f (Rk )) ∈ QW .
For a model M with the universe U we define M | Qx(ϕ1 (x1 ), . . . , ϕk (xk )) if and only if
(ϕ1M,x1 , . . . , ϕkM,xk ) ∈ QU , where ϕiM,xi = {(a1, . . . , ati ) ∈ U ti : M | ϕi (a1 , . . . , ati )},
for i = 1, . . . , k.
Generalized quantifiers defined by Andrzej Mostowski are exactly Lindström’s quan-
tifiers of type (1).
3 Old School and Old Ideas
Here I am starting the esseistic part of the paper, changing to first person perspective (also
grammatically) from, usual in science, plural of majesty style.
The great project of foundations of mathematics started as both philosophical and
mathematical project. Some of the important researchers, as Bernard Bolzano and
Bertrand Russell, were mainly philosophically motivated. Others, like Gottlob Frege and
David Hilbert, were building on both philosophical and mathematical traditions, which
were for them not clearly separated. It is particularly striking in the case of Hilbert. He
has got his high position and influence in mathematics by purely mathematical works.
However his research program of grounding foundations of mathematics by reduction
to finitistic mathematics was formulated and justified in philosophical spirit. There were
also other influential logicians motivated from both sides, let me mention: Rudolf Carnap,
Willard Quine, Hilary Putnam, and Jon Barwise.
In Poland the most influential logicians of the first half of twentieth century, Jan
Łukasiewicz and Alfred Tarski, started with philosophical problems. Jan Łukasiewicz
started with the problem of justifying the basic logical laws and the problem of
determinism. Alfred Tarski started with the problem of defining truth.
These were old ideas. New ideas disappeared in philosophy. Some people in Poland
could think that the main reason was communistic dictature in years 1948–1989.
Andrzej Mostowski openly claimed that any philosophical discussions should be removed
from foundations of mathematics. The attitude can be justified by a political situation.
Philosophical faculties in Poland were dominated by the communist party expositors.12
All this unpleasant things did not touch mathematics. Stalin, who indirectly governed
in Poland in 1945–1953, thought that mathematics and physics should be independent,
everything else have to be penetrated by the communist party.
12 I remember,I was a teenager, a comment of my father Andrzej Włodzimierz Mostowski about the book
on philosophy and non-classical logics. He said: This is neither on philosophy nor on logic, this is about
who should be relegated and who can keep his position.
It is worth to mention that this attitude was not accepted by Mostowski‘s oldest
students, Andrzej Grzegorczyk and Helena Rasiowa. Grzegorczyk went to study philos-
ophy in 1970s, and Rasiowa strongly supported joining philosophical and mathematical
interests.13 However for all later students the attitude was obvious and acceptable.
In mathematics, philosophy was replaced by bad philosophy and in philosophy
philosophy was replaced by even worse philosophy.
What is important the same thing happened in many countries without communistic
dictatorship. I think that the main reason is the idea of autonomy of faculties. Why should
we confront relevance and importance of our results with people thinking in other way?
Of course philosophy is loosing in this confrontation. However, mathematics is loosing as
well.
4 The Old School and New Ideas: Computations
The other weakness of the school of Andrzej Mostowski was not absorbing the new idea
of computability. Andrzej Mostowski thought of computability not in terms of algorithms,
but in a sense of arithmetically definable sets. His only student really thinking in terms of
computability was Andrzej Grzegorczyk. Reception of his ideas was very poor because
he was far from the center. The center was Andrzej Mostowski. The department of
Mathematical Logic guided by Helena Rasiowa reoriented to computer science in the
late 1970s.14
In these times the old department of foundation of mathematics practically disap-
peared. The department of Helena Rasiowa was quickly developing and in majority
passed to the institute of computer science preserving the only strong group in logic at
mathematical faculty. It is symbolic that the main results establishing position of this
group in the new faculty were results related to Büchi–Rabin automata—the topic which
was introduced in Poland by Andrzej Włodzimerz Mostowski.15
5 What for Philosophy?
There is a common opinion in mathematics that a good achievement is just a difficult

proof. Joining it with a view that axiomatically defined ZFC is a good basis for
mathematics means that mathematics is a game for finding difficult proofs from ZF. In
other words all mathematical questions are of the form: is ϕ provable from ZFC. More
13 Iwas her student in 1982–1983.

14 Andrzej Salwicki told that they did not know works by Andrzej Grzegorczyk. Only later on they
recognized his works as relevant and important.
15 He is my father, and younger cousin of Andrzej Mostowski. Similarly as sons of Andrzej Mostowski:
Tadeusz and Jan, he lived in a shadow of Andrzej Mostowski. I was the the first person in the family who
took the topics of Andrzej Mostowski.
564 M. Mostowski
difficult proof, better result. Of course we know that ZFC is not complete. Therefore, we
allow questions of the form: is ϕ independent of ZFC.
I have asked a few mathematicians whether they would accept their activity as playing
such a game. Nobody answered yes. Mathematicians lost their philosophical sensitivity.
This is the only explanation of these contradictory views. They need serious philosophical
thinking.
On the other hand, philosophy, without real interaction with current science starts to
be infertile. It is going to problems from its history. So working on history of its history,
and so on.
6 Who Is Your Master?
In the late 1970s, when I was a young student, Krzysztof Maurin16 asked me “who is your
master?”. I was surprised, I did not know what to answer. After a while I answered “Jan
Łukasiewicz”. Łukasiewicz has died in 1956. After a few years I would answer “Alfred
Tarski”, but he died either. Never in my life I met anyone of them. After some time I
realized that I could not honestly answer such a question. I would say that I learned from
many people, frequently older ones, but I learned also a lot from younger people.
Later I was sometimes asked by my colleagues to give them a problem to work on.
When I was young I would answered “if you do not know it then you are not ready to
work in science”. With time I have changed my mind. By having a good problem and a
good support you get a few years in advance in your scientific carrier. This was probably
what I lost in my life.
Summarizing, we need masters, more masters than one.
Acknowledgement This work was supported by the Polish National Science Centre [2013/11/B/HS1/
04168].
References
1. Cegielski, P.: Théorie élémentaire de la multiplication des entiers naturels. In: Model Theory and
Arithmetic, pp. 44–89. Springer, New York (1981)
2. Ebbinghaus, H.D., Flum, J., Thomas, W.: Mathematical Logic. Springer, Berlin (1994)
3. Ehrenfeucht, A.: An application of games to the completeness problem for formalized theories.
Fundam. Math. 49(2), 129–141 (1960)
4. Feferman, S., Vaught, R.L.: The first order properties of algebraic systems. Fundam. Math. 47(1),
57–103 (1959)
5. Fraenkel, A., Bar-Hillel, Y., Levy, A.: Foundations of Set Theory, 2nd rev. edn. Studies in Logic and
the Foundations of Mathematics. Elsevier, Amsterdam (1973)
6. Jech, T.J.: The Axiom of Choice. Studies in Logic and the Foundations of Mathematics. North-
Holland/Elsevier, New York (1973)
16 He was in this time a very eminent professor working in mathematical analysis and mathematical
physics. He was also deeply interested in the philosophy of mathematics.

7. Keisler, H.J.: Logic with the quantifier “there exist uncountably many”. Ann. Math. Log. 1(1), 1–93
(1970)
8. Lindström, P.: First order predicate logic with generalized quantifiers. Theoria 32(3), 186–195 (1966)
9. Mostowski, A.: Über die Unabhängigkeit des Wohlordnungssatzes vom Ordnungsprinzip. Fundam.
Math. 32(1), 12–36 (1939)
10. Mostowski, A.: On a generalization of quantifiers. Fundam. Math. 44(1), 12–36 (1957)
11. Mostowski, A.: Constructible sets with applications. Studies in Logic and the Foundations of Math-
ematics. North-Holland Publishing Co., Amsterdam/PWN-Polish Scientific Publishers, Warszawa
(1969)
12. Mostowski, A.: Foundational Studies, Vols. I and II. PWN, North-Holland (1979)
13. Nadel, M.E.: The completeness of Peano multiplication. Isr. J. Math. 39(3), 225–233 (1981)
14. Presburger, M.: Über die Vollständigkeit eines gewissen Systems der Arithmetik ganzer Zahlen,
in welchem die Addition als einzige Operation hervortritt. In: Comptes Rendus du I congres de
Mathématiciens des Pays Slaves, pp. 92–101. Warszawa (1929)
15. Smorynski, C.: Logical Number Theory I; An Introduction. Springer, Berlin (1991)
Jerzy Słupecki (1904–1987)
Jan Woleński
Abstract This paper presents life and work in logic of Jerzy Słupecki, a distinguished
Polish logician.
Keywords Propositional calculus · Many-valued logic · Aristotelian syllogistic
Jerzy Słupecki (JS hereafter), a Polish mathematician and logician, was born in Harbin
(Manjuria, at that time Russia, now China) on August 29, 1904. Stanisław Słupecki and
Szcz˛esna Słupecki, née Jaszewska were his parents. The father was a military engineer in
Russian army (in the rank of captain) and worked for a company constructing railways
in the Far East. Most specifically, Stanisław Słupecki was employed in building the East
China Railway. His salary was high and provided an income sufficient the family to live
for living the family in very good material conditions. In particular, the Słupeckis had
resources to organize a private education of JS and his older brother at home. In 1915,
JS entered the secondary school in Tiumen (a city in the river-basin Obu in Western
Siberia, fairly far away from Harbin). Due to stormy times after the October Revolution
in Russia in 1917, the parents decided to bring back JS back to home in 1918. Since the
living conditions considerably worsened in the entire Russia after the revolution, JS had
to work in order to increase the income of the family. Moreover, he and his brother got
tuberculosis, a very dangerous illness at that time.
The above circumstances strongly motivated the Słupeckis to go back to Poland.
Stanisław Słupecki succeeded as the first member of the family; his wife and sons
returned in 1921. JS continued his education in Warszawa. His 4 years in the secondary
school in Tiumen were partially recognized in Poland in this sense that he had to pass
examinations to be allowed to continue his education in Polish secondary schools. After
successful examinations, JS entered the Tadeusz Rejtan State High School in Warszawa
and graduated in 1926. In the same year, he began his academic studies (architecture) at
the Warsaw Technical University (Politechnika Warszawska in Polish). He had to interrupt
This paper is based on a joint contribution of Jan Zygmunt and the present author (see [26]).

568 J. Woleński
studies because his health deteriorated due to relapse come back of tuberculosis. JS spend
a year in a sanatorium in Otwock, near Warszawa.
After returning from Otwock to Warszawa, JS changed his interests and decided to
study mathematics at the Warsaw University; in fact, mathematics was his beloved subject
in the years spent in the Far East. Since his father died, JS had to interrupt studies once
again and was forced to work in order to help his mother. He was appointed in the city
office in Warszawa and in the Historical Museum. Although he tried tries to continue his
mathematical studies, he could not do that regularly. This situation changed in 1932 and
JS began to study in a normal way. Andrzej Mostowski, Stanisław Hartman and Julian
Perkal, all students of mathematics, and Antoni Raabe, the studenta student of physics,
were close friends of JS at that time (and later as well). They frequently met in Lourse’s
Cafe located at the heart of the capital of Poland on the famous Krakowskie Przedmieście
Street. Due to his complex life story JS was considerably older than his fellow students;
the difference in the age amounted to about 10 years. And he was an authority to his
academic colleagues in many issues. The study years of JS were a difficult political time
in Poland, particularly in Warszawa. Several rightist parties and organizations proclaimed
and executed radically anti-Semitic policy, also at universities. In particular, they forced
Jewish students to occupy special parts in the lecture rooms (so-called bench-ghetto) and
claimed that Jews should be eliminated from the leadership in students’ scientific clubs,
including that acting at the activity in the Mathematical-Physical Faculty. JS decisively
opposed these tendencies and defended his Jewish fellow students.
JS as a student became particularly interested in mathematical logic. The University of
Warsaw was one of the most important centers in this field at that time. Jan Łukasiewicz
and Stanisław Leśniewski acted as professors of logic and the foundations of mathematics
(official denominations of their positions were different, but their scientific activities
belonged to logic sensu largo), and Alfred Tarski, Adolf Lindenbaum, Mordechaj
Wajsberg, Moses Presburger, Bolesław Sobociński and Stanisław Jaśkowski belonged to
the most important students of both professors. These people formed the Warsaw School
of Logic with Łukasiewicz, Leśniewski and Tarski as leaders. Łukasiewicz became the
main mentor of JS and the supervisor of his master thesis concerning three-valued logic.
JS graduated in 1935 and his diploma work won the Prize of the Council of the Faculty.
Since the University had a limited number of positions for young scholars, JS could not
pursue his academic career immediately after graduation. He was appointed as a teacher in
Maria Curie-Skłodowska High School in Warszawa; he taught introduction to philosophy
and mathematics. Yet he continued his scientific research and was considered as a member
of the Warsaw School of Logic. In 1938 he obtained PhD on the base of dissertation on the
axiomatization of many-valued system of logic (see [5]). Łukasiewicz was the supervisor
and Wacław Sierpiński, a famous mathematician acted as one of referees. JS’s PhD thesis
contained new and original logical results and reaffirmed his reputation in the Warsaw
Logical Group. At that time, Lourse’s Café was not only a place for students-discussions,
but also the locum of real scientific debates with participation of Łukasiewicz, Leśniewski
(he also strongly influenced logical interests of JS) and Tarski. The relationships between
JS and Łukasiewicz were fairly close. JS and his wife Stanisława, nee Izydorzak (they
married in 1937 and had the daughter) spent the New Year’s Eve 1938/1939 (the last one
before the World War II) in Łukasiewicz’s home in Warszawa.
Jerzy Słupecki (1904–1987) 569
JS lived in Warszawa almost the entire years 1939–1944. He participated in the

clandestine education system and lectured in a private grammar school as well as at the
underground Warsaw University. He taught together with Łukasiewicz, Mostowski and
Sobociński; also Henryk Hiż joined this group. Moreover, JS was an active member of
Polish military resistance movement (AK—Armia Krajowa, Home Army). In particular,
he participated in the operation “Żegota”, that is, helping Jews. In fact, thousands of Jews,
mostly children, were saved by “Żegota”. When the Warsaw Uprising began on August
2, 1944, JS took part in it as a soldier in Praga, the right-bank, relatively to the direction
of the river Vistula, part of Warszawa. After the uprising, he was taken by Germans and
transported to Germany for forced labor (he was released due to his bad health). JS was
very proud of his war activities. In private talks (the present author had an opportunity
to listen to his recollections) about his past, he stressed that he considered his teaching
military service in the years 1939–1944 as something of the utmost importance.
After liberation of Warszawa in January 1945, JS worked for the Telecommunication
School in Warszawa and the secondary school (Maria Curie-Skłodowska in which he was
teaching before 1939). In 1945, Maria Curie-Skłodowska University was established and
JS was appointed as Associate Professor of and the Head of the Department of Logic
and the Foundations of Mathematics. This meant the real beginning of his academic
career. In 1947, he obtained his habilitation on the base of the dissertation on Aristotle’s
syllogistic (see [7]); the procedure was completed at Jagiellonian University in Kraków.
After obtaining the habilitation JS moved to Wrocław and became Assistant Professor
of Logic and Methodology of Science at Wrocław University and Wrocław Polytechnic
(both schools acted together; after their separation, JS remained at the former). In the years
1953–1955, he served as the Dean of the Faculty of Mathematics, Physics and Chemistry.
JS was nominated as the full professor in 1962; he was the head of the department until
his retirement in 1974.
JS worked not only for Wrocław University, but also for Higher Pedagogical School
(a counterpart of a college in American education system; HPS hereafter) in the same
city. This school was moved to Opole in 1954; JS was its rector in 1962–1966 and the
professor until 1974. Another important position was held by JS in the Polish Academy of
Sciences. In particular, he was associated with the Institute of Philosophy and Sociology
of this academy since 1957. The Department of Applications of Logic was established
as part of the Institute of Philosophy and Sociology and JS was its head in the years
1970–1974. He also served as the Editor-in-Chief of Studia Logica in 1964–1969. He
helped in the editorial work in “Matematyka (Mathematics) and “Zeszyty Naukowe WSP
w Opolu” (The Scientific Journal of the School of HPS in Opole). He was a member of
Polish Mathematical Society, Polish Philosophical Society, Wrocław Scientific Society
and Opole Society of the Friend of Science.
In spite of the fact that JS’s position at Wrocław University was more prestigious, he
considered Opole as the place of his main academic activity. It was caused by the situation
of logic at Wrocław University, particularly among mathematicians. The first generation
of mathematicians working at Wrocław University came from Lvov and inherited the
style of doing mathematics represented by Lvov Mathematical School, in particular,
by Stefan Banach and Hugo Steinhaus (the latter became professor in Wrocław). This
group recognized mathematical logic as an important field, but secondary as compared
with such branches of mathematics, like analysis (functional analysis, in particular),
570 J. Woleński
algebra, geometry or applied mathematics. On the other hand JS grew in Warszawa, in the
environment in which mathematical logic and the foundations of mathematics belonged
to the very heart of mathematics. This view was held by Łukasiewicz, Leśniewski and
Tarski as well as accepted by Zygmunt Janiszewski and Wacław Sierpiński, the leaders of
Warsaw Mathematical School. The Warsaw School’s logical ideology regarded logic as an
autonomous branch of mathematics and having its own scientific problems, unnecessarily
connected with other mathematical issues. Moreover, for the matter concerning the
meaning of the phrase “the foundations of mathematics”, logicians and mathematicians
working in Warszawa saw it as referring just to the foundations of mathematics as
something fundamental or basic for the theoretical and practical aspects of mathematics
as well as the practice of doing it in a proper manner, whereas the Lvov tradition saw the
foundations of mathematics as one in the entire variety of mathematical disciplines. This
view about mathematical logic and its place in mathematics was also continued by the
generations of mathematicians working in Wrocław after 1945.
Consequently and somehow paradoxically, JS, influenced by his teachers, considered
himself rather as a logician than a mathematician. Although he was respected in Wrocław
for his achievements and personality (he was a very kind person, always ready to
help other people), he had to feel alienated to some degree from the main stream of
mathematical research conducted at Wrocław University. He did not have many students,
at least not many became logicians in his sense of belonging to this profession. In Lublin,
he trained two later distinguished scholars, namely Czesław Ryll-Nardzewski and Jerzy
Łoś (both were his assistants), but both chose their own scientific way very soon and
went in the direction of the foundations, not very much related to logic. Słupecki used to
say that his two pupils did their own scientific “housekeeping” and although it sounded
somehow sad in his mouth, he understood this situation. In Wrocław, he supervised
PhD dissertations of Edward Bałuka, Ludwik Borkowski, Edmund Glibowski, Bogusław
Iwanuś, Tadeusz Kubiński, Jerzy Nowak, Witold A. Pogorzelski or Juliusz Reichbach.
Although most of them worked in logic and achieved fairly important results, no Wrocław
School of Logic arose. Anyway, several circumstances decided that JS observed a very
deep contrast between the status of logic in Warszawa (also after 1945) as compared with
Wrocław.
Certainly, JS’s deep intention was to repeat the Warsaw experience at least to some
degree. Since HPS was a entirely new academic institution without any limit bounding the
style of teaching and doing science, JS could start from scratch and try to realize his dream
in cultivating logic. In the years 1954–1964, HPS had the Department of Mathematics
without specified particular mathematical fields. JS succeeded in establishing the Depart-
ment of the Foundations of Mathematics and began a very intensive teaching of logic. He
trained, among others, Grzegorz Bryll, Katarzyna Hałkowska, Marian Maduch, Krystyna
Piróg-Rzepecka Tadeusz Prucnal and Urszula Wybraniec-Skardowska. These logicians
constituted the Slupecki School of Logic. Thus his dream was fulfilled. Moreover, JS’s
seminar conducted in Opole became very famous over the entire country and collected
many logicians from other cities: Gliwice, Katowice, Kraków and Wrocław. In fact, JS
strongly influenced investigations of many logicians working outside Opole, including
(I omit earlier mentioned names) Paweł Bielak, Janusz Czelakowski, Jacek Kabziński,
Grzegorz Malinowski, Stanisław J. Surma, Marek Tokarz, Ryszard Wójcicki, Andrzej
Wroński and Jan Zygmunt. Every issue of the scientific journal of HPS had a considerable
section of mathematical logic in which were published results obtained by JS, his students
and participants of his seminars. JS co-authored many papers with his students. He also
published textbook in mathematical logic and the foundations of mathematics.
JS died on January 15, 1987 in Wrocław. Although the day of his funeral ceremony was
very cold, many people, friends, colleagues, an students, participated in it. It documented
that JS was well remembered for his unusual personality. His death was symbolic for the
history of Warsaw School of Logic. JS was the last logician who graduated in Warszawa
before 1939 that is, in golden period (1918–1939) of the development of logic in Poland;
Hiż and Lejewski, who studied with Łukasiewicz and Leśniewski, obtained their PhD
degrees after the end of World War II.
The last section of this paper is devoted to a brief presentation of JS’s logical research
(I proceed by keeping the chronological order). JS was a very faithful inheritor of the
tradition of Warsaw School of Logic. Although he considered Leśniewski as the most
brilliant mind in this circle, Łukasiewicz was his beloved teacher (this statement is based
on my private conversation with JS). JS continued Łukasiewicz’s style of doing logic
as well as investigated similar problems as his main teacher; these questions mostly
concerned propositional calculi. As I already mentioned, first results of JS concerned
many-valued logic, more specifically the three-valued system. He solved the problem of
functional-completeness (or functional fullness) of three-valued logic (see [3, 4, 6]). The
question concerns the choice of sentential functors (connectives) sufficient for defining
all operations of a given calculus. For instance, negation and conjunction (disjunction,
implication) allow to define all remained sentential functions in two-valued logic; the
same role is played by two Sheffer connectives. On the other hand, negation and
equivalence are not enough. It was known that full bases for two-valued logic are not
sufficient for many-valued logic. JS introduced the T-functor such that, in the case of
three-valued logic, the value Tp = 12 for any value of p. Employing the T-operation. JS
showed that the system Ł3 (three-valued logic of Łukasiewicz) can be axiomatized by
supplementing already known axioms by the formulas CTpNTp and CNTpTp. He also
generalized his result on the fullness of many-valued logic by formulating the criterion
(called the Słupecki criterion) universally applicable to the problem in question. Roughly
speaking, this criterion points out that the functional fullness depends on the definability
of all unary connectives and having a truth-table by at least one binary connective.
Aristotle’s syllogistic was one of favorite topics of Łukasiewicz. JS joined this
path of investigations even before 1939. One of open problems concerned decidability.
Łukasiewicz himself obtained a partial result and observed that the final solution depends
on introducing rejection rules next to assertion rules (likes modus ponens) usually used as
inference rules in logical system. JS showed that a special form of such a (rejection)
rule added to syllogistic results in decidability of this system (see [7]). According to
Łukasiewicz himself, this result of JS should be recognized as the most important
formal discovery concerning syllogistics after its construction by the Stagirite. The
rejection function was intensively investigated by JS and his students from a more general
point of view, that is, in the framework of general theory of consequence operation
(see [17, 18]). In particular, Słupecki, Bryll and Wybraniec-Skardowska introduced the
operation Cn−1 —the rejection consequence operation. We can say that a sentence A is
rejected on the base of a set of X of sentences, if the falsity of X implies the falsity of
A. Yet this definition is based on the classical consequence operation Cn introduced by
572 J. Woleński
Tarski in 1930s. The question is how to define rejection directly, that is, not appealing
to being false. Cn−1 realizes this task. Roughly speaking, a sentence A is rejected in the
sense of Cn−1 on the basis of the set of sentences X if and only if some sentence B
from X is deducible from A. The definition is conceived in such a way that the operation
Cn−1 is a topological closure operator, because Cn−1 ∅ = ∅ (in the case Cn as defined by
Tarski, the set of logical consequences of the empty set is not empty). JS together with
Pogorzelski investigated deductive systems based on non-classical logics. These studies
generalized Tarski’s works and showed how to defined consequence operations associated
with logics other than classical two-valued system (see [12]). The same authors presented
a new syntactic proof of the Gödel’s completeness theorem (see [13]).
JS essentially contributed to popularization of two Leśniewski’s logical systems,
namely protothetics and ontology (see [8, 9]). His papers on these topics were based
on lecture notes written by himself before 1939. On the other hand, he also generalized
Leśniewski’s mereology (see [11]). Other work of JS concerned a satisfactory intuitive
interpretation of many-valued logic and modal logic based on many-valueness (see
[14, 16]). Finally let me mention JS’s writings on Łukasiewicz (see [10, 19]) and the
entire Warsaw School of Logic (see [15, 18]). As a person who witnessed the glorious
past of Polish logic, he formulated many very interesting remarks enlightening many
issues in the history of contemporary logic. There is a problem with evaluating the logical
work of JS. He is sometimes treated as a great loser of logic, because he worked inside
an old-fashioned paradigm of logical investigations, dominant before 1939, but changed
after World War II in the direction of the foundations of mathematics. This paradigm
concentrated on logical calculi and was deeply associated with using the minimum
mathematical methods. For example, logicians working in this style were not particularly
interested in algebraization of logic. On the other hand, JS essentially developed the old
paradigm and solved several fundamental problems generated by it.
References and Bibliography
A. Selected Works by Jerzy Słupecki
Books:
1. O dowodzie matematycznym [On Mathematical Proof]. PZWS, Warszawa (1962). With W.A.
Pogorzelski
2. Elementy logiki matematycznej i teorii mnogości. PWN, Warszawa (1963). With L. Borkowski (Eng.
Tr.: Elements of Mathematical Logic and Set Theory. Pergamon Press/PWN, Oxford (1967))
Papers:
3. Der volle dreiwertige Aussagenkalkül. Comptes Rendus de Séances de la Société de Sciences et de
Lettres de Varsovie, Cl. III, 29, 9–11 (1936) (Eng. tr. in: McCall, S. (ed.) Polish Logic 1920–1939,
pp. 335–337. Oxford University Press, Oxford (1967))
4. Kryterium pełnosci wielowartościowych systemów rachunku zdań [A Criterion of Fullness of Many-
Valued Systems of Propositional Logic]. Comptes Rendus de Séances de la Société de Sciences et de
Lettres de Varsovie, Cl. III, 32, 102–109 (1939) (Eng. tr.: Stud. Logica 30, 153–157 (1972))
5. Dowód akjsomatyzowalności pełnych systemów wielowartościowych rachunków zdań [Proof of
Axiomatiozability of Full Many-Valued Systems of Calculus of Propositions]. Comptes Rendus de
Séances de la Société de Sciences et de Lettres de Varsovie, Cl. III (1939) (Eng. tr.: Stud. Logica 29,
155–168 (1971))
6. Pełny trójwartościowy rachunek zdań [Full three-valued sentential calculus]. Ann. Univ. Mariae
Curie-Skłodowska Sect. F, 193–208 (1946)
7. Z badań nad sylogistyka˛ Arystotelesa. Travaux Comptes Rendus de Séances de la Société de Sciences
et de Lettres de Wrocław, Wrocław (1948) (Eng. tr.: Stud. Philos. 4, 275–300 (1949–1950))
8. St. Leśniewski’s Protothetics. Stud. Logica I, 44–112 (1953)
9. St. Leśniewski’s calculus of names. Stud. Logica III, 7–76 (1955) (Repr. in: Srzednicki, J. et allies
(eds.) Leśniewski’s Systems. Ontology and Mereology, pp. 59–122. Nijhoff, The Hague (1984))
10. The logical works of J. Łukasiewicz. Stud. Logica 8, 7–56 (1958). With L. Borkowski
11. Toward a generalized mereology of Leśniewski. Stud. Logica 8, 131–163 (1958)
12. Podstawowe własności systemów dedukcyjnych opartych na nieklasycznych logikach [Basic proper-
ties of deductive systems based on non-classical logics]. Part I, Stud. Logica 9, 163–176 (1960); Part
II, Stud. Logica 10, 77–98 (1960). With W.A. Pogorzelski
13. A variant of the proof of completeness of the first-order functional calculus. Stud. Logica 12, 125–134
(1961). With W.A. Pogorzelski
14. Próba intuicyjnej interpretacji logiki trójwartościowej Łukasiewicza [An attempt of an intuitive
interpretation of Łukasiewicz’s three-valued logic]. In: Rozprawy logiczne. Ksi˛ega Pamiatkowa˛ ku
czci profesora Kazimierza Ajdukiewicza, pp. 185–191. PWN, Warszawa (1964)
15. Logic in Poland. In: Klibansky, R. (ed.) Contemporary Philosophy. A Survey I: Logic and Foundations
of Mathematics, pp. 190–201. La Nuova Italia, Firenze (1968)
16. A generalization of modal logic. Stud. Logica 21, 7–17 (1971)
17. A theory of rejected propositions. Part I, Stud. Logica 29, 75–123 (1971); Part II, Stud. Logica 39,
97–145 (1972). With G. Bryll and U. Wybraniec-Skardowska
18. Warszawska szkoła logiczna [The Warsaw School of Logic]. Wiadomości Matematyczne 15, 65–72
(1972)
19. Jan Łukasiewicz. Wiadomości Matematyczne 13, 73–78 (1972)
B. Selected Papers on Jerzy Słupecki
20. Bryll, G., Iwanuś, B., Piróg-Rzepecka, K.: Działalność naukowa prof. Jerzego Słupeckiego (w 76
rocznic˛e urodzin) [Scientific Activity of J. Słupecki (on the Occasion of His 76th Birth Anniversary)].
Zeszyty Naukowe WSP w Opolu, Matematyka 4, 7–34 (1983)
21. Bryll, G., Hałkowska, K., Piróg-Rzepecka, K.: Działalność naukowa prof. Jerzego Słupeckiego w
środowisku opolskim [Scientific Activity of Prof. Jerzy Słupecki in Opole], pp. 27–37. Zeszyty
Naukowe WSP w Opolu, Matematyka 28 (1992)
22. Duda, R.: Słupecki Jerzy (1904–1987) – logik. In: Duda, R., Matematycy XIX and XX w. zwiazani ˛
z Polska˛ [Mathematicians of the 19th and 20th Century Associated with Poland], pp. 437–439.
Wydawnictwo Uniwersytetu Wrocławskiego, Wrocław (2012)
23. Hartman, S.: Wspomnienia o Jerzym Słupeckim [Recollections on Jerzy Słupecki]. Zeszyty Naukowe
WSP w Opolu, Matematyka 28, 21–25 (1992)
24. Iwanuś, B., Piróg-Rzepecka, K.: Działalność naukowa prof. Jerzego Słupeckiego [Scientific Activity
of Prof. Jerzy Słupecki]. Ruch Filozoficzny 45(3), 227–239 (1988)
25. Woleński, J.: Jerzy Słupecki (1904–1987). Wiadomości Matematyczne 28(2), 183–194 (1990)
26. Woleński, J., Zygmunt, J.: Jerzy Słupecki (1904–1987): life and work. Stud. Logica XLVIII, 401–411
(1989)
27. Zygmunt, J.: A bibliography of the published works of Jerzy Słupecki. Stud. Logica XLVIII, 413–421
(1989)
J. Woleński ()
University of Information, Technology and Management, Rzeszów, Poland
Rejection in Łukasiewicz’s and Słupecki’s
Sense
Abstract The idea of rejection originated by Aristotle. The notion of rejection was
introduced into formal logic by Łukasiewicz. He applied it to complete syntactic char-
acterization of deductive systems using an axiomatic method of rejection of propositions.
The paper gives not only genesis, but also development and generalization of the notion
of rejection. It also emphasizes the methodological approach to biaspectual axiomatic
method of characterization of deductive systems as acceptance (asserted) systems and
rejection (refutation) systems, introduced by Łukasiewicz and developed by his student
Słupecki, the pioneers of the method, which becomes relevant in modern approaches to
logic.
Keywords Aristotle’s syllogistic · Genesis of rejection notion · Rejection in

Łukasiewicz’s sense · Słupecki’s solution of decidability problem of Aristotle’s
syllogistic · Deductive systems · Axiomatic method of rejection · Refutation systems ·
Słupecki’s rejection function · Generalization of rejection notion
Mathematics Subject Classification (2000) Primary 01A60, 03B22; Secondary 03B30
1 Introduction
The European logic as a science arose in fourth century BC in ancient Greece. Aristotle
is thought to be the creator of logic. First of all, he is recognized as the creator of
the first formal—logic system, deductive system, the so called Aristotle’s syllogistic,
which together with the theory of immediate reasoning (square of opposition, conversion,
obversion, contraposition, inversion) is treated as traditional logic in a narrower meaning.
It is the logic of names.
The idea of rejection of some sentences on the basis of others was originated by
Aristotle, who in his systematic investigations regarding syllogistic forms not only proves
the proper (true) forms but also rejects the false (invalid, erroneous) ones. Aristotle to
reject some false syllogistic forms very often used examples, but he also used another way
of rejecting false forms by reducing them to other ones, already rejected (see Łukasiewicz
[22, 23]).

The method of rejecting sentences always functioned in empirical sciences in connec-

tion with the procedure of refutation of hypotheses, and we can assume that it was also
known to Stoics: because within the five ‘unprovable’, hypothetical syllogistic rules, their
logic of sentences, formulated probably by Chrisippus, the rule of deduction called in
Latin modus tollendo tollens, or modus tollens (If p, then q; not-q, so not-p) appears as
the second one.
Although the method of rejection of sentences has always existed in empirical science,
in relation to the procedure of refutation of hypothesis, the Aristotle’s idea of rejection of
some sentences on the basis of the others has never been properly understood by logicians
or mathematicians, especially the ones convinced that rejection of closed sentences of the
language of deductive system can be always replaced by introducing into such system the
negation of such sentences.
The proper understanding of Aristotle’s ideas and implementation of the concept of
rejection into the formal researches on logical deductive systems (including on Aristotle’s
syllogistics) we owe to Jan Łukasiewicz—the co-founder of world famous Warsaw
Logical School, which functioned in the interwar period (1918–1939). Łukasiewicz and
his pupil—Jerzy Słupecki are classified as pioneers of meta-logical studies on the concept
of rejection and related to it the notion of saturation (in Słupecki’s terminology—Ł-
decidability) of deductive systems.
The notion of rejection was introduced into formal logic by Łukasiewicz in his work
“Logika dwuwartościowa” (“Two-valued logic”) [20], in which, apart from the term
“assertion” (introduced by Frege), Łukasiewicz introduces also the term “rejection”.
In adding “rejection” to “assertion” he, as he states himself, followed Brentano, but
he did not mention any more about it. The notion of rejected proposition later played
an important role in his research on Aristotle’s syllogistic [22, 23], as well as in his
metalogical studies of some propositional calculi [24, 25].
In that research Łukasiewicz makes use of the idea of rejection originated by Aristotle.
Łukasiewicz applied it to complete a syntactical characterization of deductive systems
using an axiomatic method of rejection, introduced by him into the formal logic in his
paper on Aristotle’s syllogistic [22], and then, after the war, in a monograph [23], which
followed many years of research on Aristotelian logic, and which included the results
presented in the paper prepared before the war.
As was pointed out by Łukasiewicz (see [23, p.67]),
• Aristotle, in his systematic investigations of syllogistic forms, not only proves the true
ones but also shows that all the others are false, and must be rejected.
Further on (p.74), Łukasiewicz observes:
• Aristotle rejects invalid forms by exemplification through concrete terms. This proce-
dure is logically correct, but it introduces into the systems terms and propositions not
germane to it. There are, however, cases where he applies a more logical procedure,
reducing one invalid form to another already rejected. On the basis of this remark, a
rule of rejection could be stated corresponding to the rule of detachment by assertion;
this can be regarded as the commencement of a new field of logical inquiries and of
new problems that have to be solved.
Rejection in Łukasiewicz’s and Słupecki’s Sense 577
• Modern formal logic, as far as I know—writes Łukasiewicz (see [23, p.71])—does not
use ‘rejection’ as an operation opposed to Frege’s ‘assertion’. The rules of rejection
are not yet known.
As a rule of rejection corresponding to the rule of detachment by assertion, Łukasiewicz
adopts [22, 23] the following rule, which was anticipated by Aristotle:
• the rule of rejection by detachment:
if the implication “If α, then β” is asserted, but its consequent β is rejected, then its antecedent α
must be rejected, too.
As a rule of rejection corresponding to the rule of substitution for assertion

Łukasiewicz adopts [22, 23] the following rule, which was unknown to Aristotle:
• the rule of rejection by substitution:
if β is a substitution instance of α, and β is rejected, then α must be rejected, too.
Both rules enable us to reject some syllogistic forms, provided that some other forms
have already been rejected.
As we mentioned above, Aristotle used the procedure of rejection of some forms by
means of concrete terms, but such a procedure, though correct, introduces into logic terms
and propositions that are not germane to it.
To avoid this difficulty, Łukasiewicz rejects some forms axiomatically, which leads him
to biaspectual axiomatic characterization of deductive systems analyzed by him [22–25].
The idea of rejection was first used by Łukasiewicz for two-level syntactic description
of Aristotle’s axiomatic system of syllogistics: as a system with respect to acceptance
(the first level) and as a system with respect to rejection (the second level). The sentence
rejected from the system he understood as a false sentence, or a sentence which due to
some reasons we cannot classify as a thesis of this system. The idea of rejection was used
by Łukasiewicz while studying the decidability (saturation) of the system: each sentence,
which is not the thesis of a decidable (saturated) system, is rejected.
Reconstruction of concepts of rejection and decidability (saturation) used by
Łukasiewicz, was done by his pupil—Jerzy Słupecki [39, 40] (see also Słupecki et
al. [45]). Słupecki modified, developed and later generalized the concept of rejected
sentence, he also made its certain formalization. He also inspired systematic, formal
studies on this and related concepts, and also initiated research on the decidability of
many deductive systems.
In this paper1 I am starting from Słupecki’s and mine reconstruction of concepts
rejection and decidability—the notions introduced and used by Łukasiewicz (Sect. 2).
Later on (Sect. 3), I am discussing the problem of decidability of Aristotle’s syllogistic,
set by Łukasiewicz, and I outline its solution given by Słupecki. To follow with, I am
describing Słupecki’s modification and generalization of the concept of rejection and
also presenting the importance of Słupecki’s research on syllogistics for contemporary
metalogic (Sect. 4). Later on, I am showing the relation of idea of decidability in
Łukasiewicz’s sense with the common concept of decidability given by Słupecki, and
I am discussing the decidability of more important logical systems (Sect. 5). Finally, I
1 The paper is elaborated on the basis of my works [63–65] and Słupecki et al. [45].
am presenting Słupecki’s generalization of the concept of rejection to the function of

rejection, and its formalization in the theory of rejected sentences (Sect. 6). The final notes
relate to presentation of different than Łukasiewicz’s way of two-level formalization of
deductive systems (Sect. 7).
2 Reconstruction of Concepts of Rejection and Decidability: The

Notions Introduced and Used by Łukasiewicz
The notion of decidability of the deductive system that was used by Łukasiewicz in his
research on Aristotle’s syllogistic [22, 23] and systems of propositional calculi [24, 25] is
based on the notion of a rejected sentence introduced by him.
The main idea of a syntactic biaspectual characterization of deductive systems in
Łukasiewicz’s sense is compatible with providing both:
• the axioms and inference rules for the given deductive system, which intuitively lead
from some true formulas to true ones of this system
and
• the rejected axioms (treated as false formulas of this system) and rejection (refutation)
rules of this system, which intuitively lead from some false formulas to false ones of
this system.2
Łukasiewicz used the terms ‘decidable system’ and ‘consistent system’ in the meaning
different from the one accepted in logic. Łukasiewicz does not give clear definitions of
these terms, but the context points out that he used them in the following meaning:
• The system is decidable if every its expression which is not its thesis is rejected on the
ground of finite number of axiomatically rejected expressions;
• The system is consistent if none of its thesis is rejected.
Łukasiewcz did not use the term ‘decidable system’ consequently. He also employed
interchangeably the terms ‘saturated system’ or ‘categorical system’.
In nomenclature introduced by Słupecki, decidability of a deductive system in
Łukasewicz’s sense was called Ł-decidability and its consistency was called Ł-
consistency.
The meaning of the term ‘decidable system’ compatible with the understanding of the
notion of a decidable system by Łukasiewicz gives the following definition:
• The deductive system determined by means of the ordered triple:
F, A, R!
2 Sucha syntactic formalization of some propositional calculi was also, probably independently, intro-
duced by Rudolf Carnap [6, 7]; see Citkin [10].
where F is the set of all well-formed formulas of this system, A is the set of its axioms
and R is the set of its primitive inference rules, is Ł-decidable if and only if there exist
finite sets: the set A−1 of rejected axioms (included in F ) and the set R −1 of primitive
rejected rules, such that the following two conditions are satisfied:
I. T ∩ T −1 = ∅ and II. T ∪ T −1 = F,
where T is the set of all theses of the system, T −1 is the set of all rejected formulas (the
smallest set including the set A−1 and closed with respect to every relation determined by
rejected rules of the set R −1 ).
The conditions I and II we call, respectively, Ł-consistency of the deductive system
and Ł-completeness of the system.
Characterizing the deductive system by means of tuples
F, A, R; A−1 , R −1 ! or F, T ; T −1 !,
the definition given above can be defined as follows:

The deductive system is Ł-decidable if and only if the following conditions are
satisfied:
I. the set of all its theses (asserted expressions) determining the system is disjoined with
the set of all its rejected formulas,
II. every propositional formula is either asserted or rejected.
3 The Problem of Decidability of Aristotle’s Syllogistics, Set by

Łukasiewicz, and Its Solution Given by Słupecki
3.1 Łukasiewicz’s Biaspectual Formalization of Aristotelian

Logic AS
On the first level AS is characterized as follows (cf. Łukasiewicz [21]):

The Vocabulary of AS:
• constant symbols of classical logic CL, i.e., the connectives of CL,
• primitive terms of AS: constants a and i which are sentence forming functors of two-
term arguments: ‘all . . . are . . . ’ and ‘some . . . are . . . ’,
• nominal variables: S, P , M, N, . . . .
Well-Formed Expressions of AS:
• atomic affirmative expressions: formulas of the form of S a P and S i P , which are
read: ‘all S are P ’, ‘some S are P ’, respectively, compound expressions: formulas that
are created from the atomic ones using the connectives of CL,
• the set F of all well-formed expressions—the smallest set of formulas including the
atomic expressions and closed under the connectives of CL.
Defined Terms of AS:

• the remaining constants of Aristotelian logic: e and o, i.e. the functors: ‘no . . . are . . . ’
and ‘some . . . are not . . . ’ which are defined as follows:
df
D1. S e P = ∼S i P ,
df
D2. S o P = ∼S a P .
The negative expressions S e P and S o P are read, respectively: ‘no S are P ’ and ‘some
S are not P ’. Apart from the atomic expressions, we include them into the so called simple
expressions.
Axioms of AS:
A+ 1. S a S,
A+ 2. S i S,
A+ 3. M a P ∧ S a M → S a P (Barbara),
A+ 4. M a P ∧ M i S → S i P (Datisi).
Axioms A+ 1 and A+ 2 are the two laws of identity; Aristotle did not accept them.
Primitive Inference Rules for AS:
rŁˆ: the rule of definitional replacement (according to D1, S e P may be everywhere
replaced by ∼S i P , and, according to D2, S o P may be everywhere replaced by
∼S a P );
r + 1: the rule of detachment (modus ponens) (if ‘α → β’ and α are asserted expressions
of a system, then β is an asserted expression);
r + 2: the rule of substitution (if α is an asserted expression of the system, then any
expression produced from α by a valid substitution is also an asserted expression;
the valid substitution is to put, for term-variables, other term-variables).
The schemes of the rules r + 1 and r + 2 are as follows:
r + 1: #α→β r + 2:
#α #α
#β # e(α)
The symbol ‘#’ is a sign of assertion introduced by Frege, whereas, the expression ‘e(α)’
denotes a substitution instance of α.
Characterization of the system AS on the second level consists in supplementing it
with rejected axioms and rejection rules.
Łukasiewicz formulates the following rejected axioms and rejection rules:
Rejected Axioms of AS:
A− 1. P a M ∧ S a M → S i P ,
A− 2. P e M ∧ S e M → S i P .
Primitive Rejection Rules for AS:
r − 1: the rule of rejection by detachment (reverse modus ponens),
r − 2: the rule of rejection by substitution.
The schemes of the rules are the following:
r − 1: #α→β r − 2:
-β - e(α)
-α -α
The symbol ‘-’ is the sign of rejection.

The characterized system AS is determined by the following ordered 5-tuple, which
may be called the basis of AS:
F, A+ , R + ; A− , R − !, (B)
where F is the set of all well-formed formulas of this system; A+ —the set of its axioms;
R + —the set of its primitive inference rules; A− —the set of its rejected axioms, and R − —
the set of its primitive rejection rules. The tuples
F, A+ , R + ! and F, A− , R − !
determine, respectively, the set T + of all theses of this system and the set T − of all its
rejected formulas.
The first tuple may be called the assertion system for AS, whereas, the second one may
be called the refutation system for AS.
• T + is the set of all well-formed formulas derivable from the set of theses of
metalogically formulated CL and axioms of A+ by means of inference rules of R + ,
while
• T − is the set of all well-formed formulas derivable from the rejected axioms A− by
means of theses of T + and rejection (refutation) rules of R − . So
T + = Cn+ (CL ∪ A+ , R + ),
and T + is the smallest set including CL ∪ A+ and closed under the inference rules of R + .
T − = Cn− (T + ∪ A− , R − ), D(Ł)
and T − is the smallest set including A− and closed under the rejection rules of R − . The
set T − is the set of all rejected expressions of the system in Łukasiewicz’s sense.
To the set T + there also belong all 24 valid syllogistic forms, the laws of logical square
and the laws of conversion, and, to the set T − of all rejected formulas, there belong all the
remaining 232 invalid forms. However, it turned out that there exists such well-formed
expression of AS, which is neither a thesis of this system nor a rejected expression of the
set T − . Such, for example, is the formula:
S i P → (∼S a P ∧ P a S). (Fl)

In order to remove this difficulty, we could reject the expression (Fl) axiomatically.
However, a question arises whether there exists some other formula of the same kind as
(Fl), or, may be, an infinite number of such formulas, which can be called undecidable on
the strength of our basis (B). Therefore, we may only claim that the following condition
holds:
T + ∪ T − ⊂ F.
3.2 The Problem of Ł-Decidability of AS
The system AS whose basis is (B) analyzed by us, is not saturated or decidable in the
sense that it is both 1◦ Ł-consistent and 2◦ Ł-complete, i.e.
1◦ T+ ∩T− = ∅ and 2◦ T+ ∪ T− = F.
As we know, a system satisfying both conditions was called by Słupecki an Ł-decidable

system (see Sect. 2).
The problem concerning the finite Ł-decidability of Aristotelian syllogistic was raised
by Łukasiewicz in December 1937, during his seminar on Mathematical Logic at the
University of Warsaw.
Łukasiewicz presented the problem in the form of the following questions:
Q1. Are the axioms of A+ for AS together with the inference rules of R + for AS
sufficient to prove all true expressions of the AS?
Q2. Are the rules of rejection of R − = {r − 1, r − 2} for AS sufficient to reject all false
expressions (every formula of F that is not a thesis of T + ), provided that a finite
number of them are rejected axiomatically?
3.3 Słupecki’s Solution of the Problem of Ł-Decidability of AS
Jerzy Słupecki, who participated in Łukasiewicz’s seminar, solved in 1938 the problem
providing a basis for which the system of Aristotelian syllogistic is Ł-decidable (see
Łukasiewicz [22, 23]). His answer to the question Ql was positive; to the second one,
negative.
Słupecki was able to prove that it is not possible to reject all the false expressions
of AS by means of the rules r − 1 and r − 2, provided a finite number of them is rejected
axiomatically.
This way Słupecki gave a negative answer to the question Q2:
(i) The system AS with the basis (B) is not Ł-complete with any finite set of rejected
axioms of A− .
Słupecki extended the system AS, adding to it a new rejection rule, called by Łukasiewicz
Słupecki’s rule of rejection. It is denoted by r − S and has the following scheme:
r − S: -α→γ
-β→γ
-α∧β →γ
where α and β denote negative expressions in the form: S e P or S o P and γ denotes

a simple expression or an implication the consequent of which is a simple expression and
the antecedent, a conjunction of such expressions.
The Słupecki’s rule says: If the expression γ does not follow from any of two negative
expressions then it does not follow from their conjunction.
Słupecki’s rule is closely related to the principle of traditional logic (ex mere negative
nihil sequitur).
As was noticed by Łukasiewicz, having added Słupecki’s rule, it is enough to adopt
merely one rejected axiom, namely, A− 1.
Słupecki demonstrated that:
(ii) System of Aristotle’s syllogistic determined by the following base:
F, A+ , R + ; {A− 1}, R − ∪ {r − S}! (BS)
with the refutation system:

(iii) F, {A− 1}, R − ∪ {r − S}!,
is Ł-decidability system, i.e.
1◦ T + ∩ T −S = ∅ and 2◦ T + ∪ T −S = F , where
T −S = Cn− (T + ∪ {A− 1}, R − ∪ {r − S}), D(Ł)
is the set of all rejected propositions, i.e. the set of propositions derivable from the axiom
A− 1 by means of the thesis of AS and Łukasiewicz’s rejection rules and Słupecki’s
rejection rule.
And the problem of Ł-decidability of AS has been solved: any well-formed formula of
AS is either a thesis or is a rejected formula of AS.
It is clear that Słupecki extended the notion of the rejected proposition used by
Łukasiewicz, because:
T − ⊂ T −S . D(Ł) ⊂ D(Ł)
The results obtained by Słupecki were summarized by Łukasiewicz in his work [22]
containing also the text of his paper on Aristotle’s syllogistic.
The results of research of both Łukasiewicz and Słupecki were later, after the
war, presented in detail in Łukasiewicz’s monograph [23]. In both works Łukasiewicz
expressed his high opinion of Słupecki’s findings, which, in the words of Łukasiewicz
[22] were
organically united with researches of the author [. . . ] the author regards as the most significant
discovery made in the field of syllogistic since Aristotle.
3.4 Słupecki’s Definition of a Rejected Proposition
Słupecki failed to publish his findings before the war. After the war ends, Słupecki
published them in [38] and in a monograph [39]. In the monograph [39], in his proof
of Ł-completeness (condition (ii), 2◦ ), he also used a definition of the rejected proposition
different than Łukasiewicz D(Ł), and additionally, he modified its extension D(Ł),
adopted earlier by himself. Instead of Łukasiewicz’s definition D(Ł), Słupecki adopts the
following equivalent definition:
D(Sł). A rejected proposition on the ground of the basis
F, A+ , R + ; A− , ∅!, (B \ R − )
is such an expression for which there exists a rejected axiom of the set A− which is
derivable from it and theses of the set T + by means of inference rules of R + .
Denoting a set of all rejected propositions in the sense of the definition D(Sł) by
Cn−1 (T + ∪ A− , R + ), we obtain the following symbolic notation of it:
α ∈ Cn−1 (T + ∪ A− , R + ) ⇔ ∃β ∈ A− (β ∈ Cn+ (T + ∪ {α}, R + )), D(Sł)
where Cn+ is a consequence operation with respect to the set T + of all theses of the
system and its set of rules R + .
The definition of a rejected proposition D(Sł) is closer to Aristotle’s idea of refutation
of syllogisms by means of reducing them to syllogisms rejected earlier.
We note (see D(Ł)) that:
T − = Cn− (T + ∪ A− , R − ) = Cn−1 (T + ∪ A− , R + ). D(Ł) ≈ D(Sł)
Thus, the notions of rejected propositions, both the one used by Łukasiewicz and that
introduced by Słupecki in the form of the definition D(Sł), are equivalent.
3.5 Słupecki’s Definition of an Extended Notion of Rejected

Proposition
We will reconstruct Słupecki’s definition of an extended notion of the rejected proposition,

equivalent to the definition D(Ł).
D(Sł). A rejected proposition on the ground of the basis
F, A+ , R + ; {A− 1}, {r − S}! (BSł)
is either a rejected axiom A− 1 or a proposition rejected with respect to those rejected

earlier in the sense of D(Sł; A− /X), or an expression rejected on the basis of those
rejected earlier, by means of using Słupecki’s rule.
Let Cn (T + ∪ {A− 1}, R + ; {r − S}) be the set of all rejected propositions in the sense of
D(Sł). We may note that
T −S = Cn− (T + ∪ {A− 1}, R − ∪ {r − S}) = Cn (T + ∪ {A− 1}, R + ; {r − S}).

D(Ł) ≈ D(Sł)
3.6 Three Different Ways of Understanding the Notion of Rejected

Proposition
It is easy to see that the given definitions of rejected propositions provide three different
ways of understanding this notion:
D(Sł) ⊂ D(Sł). Cn−1 (T + ∪ A− , R + ) ⊂ Cn (T + ∪ {A− 1}, R + ; {r − S})

/ /
T− T −S
/ /
− − − −
D(Ł) ⊂ D(Ł). Cn (T ∪ A , R ) ⊂ Cn (T ∪ {A 1}, R − ∪ {r − S})
+ + −
The first of them refers to Łukasiewicz’s understanding of the rejected proposition (see
D(Ł)) and to its strengthening given by Słupecki (see D(Ł)); the second and the third
ones, to Słupecki’s understanding of the rejected proposition (see D(Sł) and D(Sł). At the
same time, the second one refers directly to Aristotle’s method of rejection of syllogisms
by reducing them to previously rejected syllogisms and makes it possible to simplify the
procedure of rejection without supplementing the system with the rules of rejection, and
the third one (see D(Sł)) is a combination of both former methods.
4 Notions of Rejection in a Deductive System and Notions

of Ł-Decidability
4.1 Three Different Notions of Rejection
We are adapting the previous definitions of rejection for AS for any deductive system.
Let S be any deductive system with biaspectual formalization and with the basis
FS , A+ + − −
S , RS ; AS , RS ∪ RS ! (BS )
determined, respectively, by computable sets: the set FS of all well-formed formulas, the
set A+ + −
S of axioms (asserted axioms), the set RS of inference rules, the set AS of rejected
axioms and by the set RS− ∪ RS of rejection rules, with an assumption that the sets RS+
and RS− are sets of mutual dual rules, while the set RS is a set of non-dual rejection rules.
Let TS+ be a set of all theses of S. Then, according to Łukasiewicz’s conception, on the
analogy to D(Ł),
the set 1 TS− of all rejected propositions of S, with respect to the set TS+ and the basis
(BS ) (the refutation system FS , A− −
S , RS ∪ RS !) is defined as follows:
1
TS− = Cn− + − −
S (TS ∪ AS , RS ∪ RS ) DS (Ł)
And 1 TS− is a set of all formulas with rejection proofs in Łukasiewicz’s sense, i.e. it is a
set of all formulas derivable from rejected axioms of A− +
S by means of theses of TS and
rejection rules of RS− ∪ RS ; i.e. 1 TS− is the smallest set including A−
S and closed under
the rejection rules of RS− ∪ RS .
If RS = ∅, the basis (BS ) of the system S can be replaced by the basis
FS , A+ + −
S , RS ; AS , ∅! (BS \ RS− )
and the set 2 TS− of all rejected propositions of S with respect to the set TS+ and the basis
(BS \ RS− ), on the analogy to D(Sł), is defined as follows:
2
TS− = Cn−1 + − +
S (TS ∪ AS , RS ) DS (Sł)
and 2 TS− is a set of all propositions with rejection proofs in Słupecki’s sense, i.e. it is a set
of all such formulas from which, and from theses of TS+ , and by means of inference rules
of RS+ a rejected axiom of A− S is derivable.
If RS = ∅, the basis (BS ) can be replaced by the basis
FS , A+ + −
S , RS ; AS , RS ! (BS )
and the set 3 TS− of all rejected propositions of S with respect to the set TS+ and the basis
(BS ), on the analogy to D(Sł), can be defined as follows:
3
TS− = Cn S (TS+ ∪ A− +
S , RS ; RS ) DS (Sł)
and the set 3 TS− is a set of rejected propositions with rejected proofs in Słupecki-
Łukasiewicz’s sense, i.e. it is a set of all such propositions every one of which is either:
1◦ a rejected axiom of A− ◦
S , or 2 a proposition rejected with respect to those rejected
◦
earlier in Słupecki’s sense, or 3 a proposition rejected on the basis of those rejected
earlier by means of rejection rules of RS .
Let us note that the given above definitions DS (Ł), DS (Sł) and DS (Sł) are in some
extent a simplification and require the precise definitions of the above-mentioned rejection
proofs on the basis of any set X ⊆ FS , with respect to the bases (BS ), (BS \ RS− ) and (BS )
(see Słupecki [43], Słupecki and Bryll [44], Wybraniec-Skardowska [63, 64]).
Let us observe that among the set of rejected propositions defined above, the following
relationships hold:
a. If RS = ∅ then 2 TS− = 1 TS− b. 2 TS− ⊆ 1 TS− = 3 TS− .
4.2 Three Different Notions of Ł-Decidability; Decidability
The three different definitions of the sets of rejected propositions of the system S entail
three different definitions of Ł-decidability of this system.
Definition The system S is Ł-decidable if and only if, for some i = 1, 2, 3, S is i Ł-
decidable, i.e. it satisfies the two following conditions:
1◦ TS+ ∩ i TS− = ∅, 2◦ TS+ ∪ i TS− = FS .
The condition 1◦ is called i Ł-consistence condition and the condition 2◦ is called i Ł-

completeness condition of the system S.
A question arises: What the relationship between Ł-decidability and decidability in the
usual sense is? The answer was given by J. Słupecki.
Słupecki’s Theorem [43]: If the system S is Ł-decidable and any rejection rule, except
for the rule of rejection by detachment, is computable, then the system S is decidable in
the usual meaning.3
It is easy to show that
Corollary The system AS of Aristotle’s syllogistic with the bases (BSŁ) is decidable.
In the next section we will present Ł-decidable systems that are also decidable.
5 More Important Findings Concerning Ł-Decidability

of Deductive Systems
The axiomatic rejection method introduced by Lukasiewicz to complete, biaspectual char-

acterization of deductive systems (which was effectively used and developed by Słupecki)
provided a broad response in literature after the Second World War. Łukasiewicz, already
living in Dublin, uses such method in his research on intuitionistic logic [24], as well
3 Słupecki’s theorem is an immediate consequence of the following theorem of the theory recursion, which
we quote from Grzegorczyk’s book [14, p. 355 in the Eng. ed.]: If the union of two recursively enumerable
disjoint sets T and S is computable set, then the sets T and S are also computable.
as in a four-valued modal system of propositional calculus [25], built by himself. At the

same time, in Poland, further studies, inspired by Słupecki on Ł-decidability of deductive
systems and the very notion of rejected proposition were taken up. In studies on Ł-
decidability and providing complete refutation systems for logical systems (assertion
systems) one of three methods, modeled on those described in the previous section, is
usually used. In the next sections we will present a few, more important results, connected
with this research.4
5.1 Calculi of Names
J. Słupecki’s research on Aristotle’s syllogistic was continued mainly by B. Iwanuś. Using

the Słupecki-Łukasiewicz’s methods, Iwanuś managed to prove Ł-decidability of a few
systems of calculi of names.
• Iwanuś [17] gave a proof of Ł-decidability of the whole traditional calculus of names,
i.e. the system of Aristotle’s syllogistic enriched by nominal negation.
• Another interesting, though much later obtained result of Iwanuś’s research [18] is a
proof of Ł-decidability of the system of Aristotle’s syllogistic built by Słupecki [38].
In this system, the two initial Łukasiewicz’s axioms (laws of identity) of the system
AS (those that are absent in Aristotle’s logic) are replaced with the following axioms:
S a P ⇒ S i P, S i P ⇒ P i S.
In Słupecki’s system, unlike in the system AS, it is permissible for variables to

represent empty names. A complete refutation system for the system given by Iwanuś
[18] is based on three rejected axioms and one rejected rule that is germane to the
traditional calculus of names.
• Iwanuś also, in [16], gave a proof of Ł-decidability of a certain version of elementary
ontology, distinguished from the system of Leśniewski’s Ontology (both terms were
invented by Słupecki in [41], who presented a system of calculus of names, based
on Leśniewski’s findings). In elementary ontology it is possible to interpret both the
asserted system AS and other asserted syllogistic systems richer than AS, with nominal
negation. Iwanuś gave a complete refutation system for the version of the system of
elementary ontology; Iwanuś’s refutation system consist of two independent rejected
axioms and one non-Łukasiewicz’s specific rejection rule.
4 G. Bryll (as well as his book [2]) and T. Skura have been very helpful in verifying certain significant
facts.
5.2 Propositional Logic
As we have already mentioned, Łukasiewicz used to apply the axiomatic method of

rejection also to some systems of propositional calculus.
5.2.1 Classical Logic
Łukasiewicz mentioned in [25], that the classical propositional calculus with the inference
rules r + 1 and r + 2 is Ł-decidable. A complete refutation system for it determines one
rejected axiom, namely the sentential variable p, and two Łukasiewicz’s rules: r − 1 and
r − 2.
Let us note that the method of rejection of false formulas (i.e. non-theses) used by
Łukasiewicz can be replaced by Słupecki’s method, omitting Łukasiewicz’s rules, i.e.
applying the following principle: If, from a formula of propositional calculus and the set
of theses, it is possible to deduce, according to the rules r + 1 and r + 2, the rejected axiom
p, then the formula is rejected (see D(Sł)).
For the classical first-order calculus rejected axioms and rejected rules were formulated
by T. Skura in [32]. Skura made use of the ‘tableaux’ method.
The rejection procedure accompanying the syntactic characterization of systems of
propositional logic became a standard among logicians.
We will limit our presentation of results in the scope of the rejection procedure in the
systems to a brief mention of only intuitionistic logic and extensions, modal logic and
Łukasiewicz’s logics.
5.2.2 Intuitionistic Logic and Extensions
In his research on intuitionistic propositional calculus, Łukasiewicz [24] advanced the

hypothesis that it is Ł-decidable. Moreover, he supposed that a sole rejected axiom of it is
a propositional variable, and that the rejection rules are: r − 1, r − 2 , and one special rule
(Gödel rule) which states that the disjunction α ∨ β is rejected whenever so are α and β.
Thanks to Kreisel-Putnam [19], we know that the rules proposed by Łukasiewicz do not
suffice to reject all non-theses of the intuitionistic calculus. It is also known that there is
no finite set of rejected axioms that, together with Łukasiewicz’s rules, gives a complete
refutation of intuitionistic system (cf. Maduch [28]).
Ł-decidability of the intuitionistic propositional logic was achieved by Scott [29] using
a countable number of non-structural rejection rules. However, Scott’s results, which were
presented in Summaries of Talks at Cornell University, were inaccessible to Słupecki’s
circle in behind the Iron Curtain (then) Poland.
Independently of Scott’s results, the proof of Ł-decidability of the intuitionistic
propositional calculus was provided by Dutkiewicz [11]. In his approach, a complete
refutation intuitionistic system is compounded of one axiom and three rejection rules:
Łukasiewicz’s rules, and a new, original rejection rule, which is, in fact, an infinite
countable class of rejection rules of a common scheme. In his proof, Dutkiewicz uses the
method of rejection modeled on the one applied by Łukasiewicz, as well as the method of
Beth’s semantic tableaux.
Another proof of Ł-decidability for the intuitionistic calculus was given by Skura
[30, 36], who, while defining a complete intuitionistic refutation system, added to
Łukasiewicz’s rules a new rule, or rather, a class of structural rules of rejection, the
number of which is infinite. Skura in [31], provided a complete refutation system for
certain intermediate logics.
5.2.3 Modal Logic
The first research into a complete syntactic characterization of a modal propositional

calculus was undertaken by Łukasiewicz [25]: he extends the four-valued modal system,
built by himself, by two rejected axioms and his rejection rules, obtaining Ł-decidability
of this system.
Afterwards, Słupecki initiated research on Ł-decidability of Lewis system S5. In his
and Bryll’s paper [44], the proof of Ł-decidability was achieved with an assumption of
one rejected axiom (the prepositional variable p) and, apart from Łukasiewicz’s rules, a
class of rejection rules of the common scheme.
Skura [31, 32, 34] gave a simpler proof of Ł-decidability of Lewis system S5, assuming
that the language of this system was supplemented with a symbol ‘⊥’, i.e. the constant
of falsity. Skura adopts this constant as the rejected axiom and extends the systems of
Łukasiewicz’s rules by: 1) a rule stating that: if formula p is rejected, the formula p is
rejected, too; and 2) a class of structural rejection rules of the same scheme.
Skura in [31, 35, 36], using the algebraic method, also provided a complete refutation
system for the logic S4 and for some of its extension (Grzegorczyk’s logic).
A little earlier, Goranko [12, 13] formulated a complete refutation system for some
normal modal propositional logics (including S4 and Grzegorczyk’s logic) that are
characterized by a class of finite trees. His refutation systems for these logics are based
on the same rejected axiom (the constant ‘⊥’), Łukasiewicz’s rules, and a class of non-
structural rejected rules of the same scheme.
The method of constructing refutation systems corresponding to classes of finite
models was used by Skura [31] for intermediate logics and, by Skura [33] and Goranko
[13], for certain normal modal logics. Skura in [33], showed that the refutation systems
can be useful in such cases when a given system of logic cannot be characterized by any
class of finite models: there is a decidable modal logic without a finite model property
that has a simple refutation system.
Tomasz Skura is regarded as an expert on the methods in refutation systems; his book
[37] is devoted to refutation methods in modal propositional logics.
5.2.4 Łukasiewicz’s Many-Valued Logics
Researches into Ł-decidability of Łukasiewicz’s sentential calculus were conducted in the

Opole circle of logical research, which, for many years, was led by Jerzy Słupecki. Bryll
and Maduch [3], formulated a uniform method of rejection of formulas in an n + 1-valued
implicational, implicative-negative and definitionally complete Łukasiewicz’s calculus. In
these systems the same formula may be adopted as the sole rejected axiom5
C(Cp)n q(Cp)n−1 q
(for n = 1 we get, in particular, the rejected axiom: CCpqq in the classical implicational
calculus). The only rejection rules are, here, Łukasiewicz’s rules.
A complete refutation system for ℵ0 -valued Łukasiewicz’s calculus was built by Skura
[32] by extension of Łukasiewicz’s rejection rules. Research into Ł-decidability of this
system has been also conducted by Bryll [2]. His research was continued by R. Sochacki.
Sochacki [50] gave complete refutation systems for all invariant Łukasiewicz’s many-
valued logics (in which the rule of rejection by substitution was eliminated; see also
Sochacki [48], Bryll and Sochacki [4, 5].
Sochacki also built refutation systems for selected many-valued logics: the k-valued
logic of Sobociński and some systems of nonsense logic [49, 51].
5.3 The Generalized Method of Natural Deduction
The method of rejection introduced to metalogical investigations by Łukasiewicz in many

cases can be replaced with the generalized method of natural deduction. The latter is
applicable to all propositional calculi which have finite adequate matrices, as well as to
the intuitionistic propositional calculus and the first-order predicate calculus, that is, to
almost all logics discussed in this section.
The basis for such systems consists then of only assertion rules and rejection rules; the
sets asserted and the rejected axioms are empty sets. The method used in the proofs is
similar to Słupecki-Łukasiewicz’s method of rejection, though they are apagogic proofs
(by reductio ad absurdum).
This method refers to the ‘tableaux’ method. It is presented by Bryll [2], who in his
studies refers to results obtained by Hintikka [15], Smullyan [47], Suchoń [54], Surma
[55, 56] and Carnielli [8, 9].
6 Rejection Operation
As was noticed by Słupecki, the notion of rejected proposition is so general that it is most
convenient to base studies concerning this notion on Tarski’s theory of deductive systems,
i.e. the axiomatic theory of consequence built by Alfred Tarski [57]. Let us recall that the
only primitive notions of this theory are:
5 It is saved in Łukasiewicz’s, so called Polish notation.

• the set F of all propositions (well-formed formulas) of an arbitrary but fixed language
of a given system and the consequence operation:
Cn+ : 2F → 2F ;
• the symbol ‘Cn+ (X)’ denotes the set of all consequences of the set X ⊆ F .
Properties of deductive systems characterized by the bases with refutation systems can
be established by means of Tarski’s theory of consequence enriched by the following
definition of rejection operation:
Cn−1 : 2F → 2F ,
determined by the consequence operation Cn+ (see D(Sł)): For any X ⊆ F, α ∈ F
α ∈ Cn−1 (X) ⇔ ∃β ∈ X(β ∈ Cn+ ({α})) MD(Sł)
According to the definition MD(Sł): a proposition α is a rejected proposition on the

ground of the set X if and only if at least one of the propositions of X is derivable from
(is a consequence of) α; the symbol ‘Cn−1 (X)’ denotes the set of all propositions rejected
on the ground of propositions of X.
The following theorem helps to understand the intuitive sense of the definition MD(Sł):
∀X ⊆ F (X ⊆ Y ⇒ Cn+ (X) ⊆ Y ) ⇒ ∀X ⊆ F (X ⊆ Y ⇒ Cn−1 (X) ⊆ Y ).
If Y is the set of all true propositions of F (then the set Y is the set of all false propositions
of F ), then, according to the above theorem, we can state that:
If consequence operation always leads from true propositions to true propositions, then a rejected
operation always leads from false propositions to false propositions.
So, the defined rejection operation (function) Cn−1 is a generalization of the notion of
rejection introduced by Łukasiewicz.
The definition MD(Sł) of the function Cn−1 was formulated by Słupecki [42]. Słupecki
proved that the function satisfies all axioms of Tarski’s general theory of deductive
systems [57] for the consequence Cn+ and that it is additive. Thus, the rejected operation
is another consequence operation and is called a rejection consequence. It is clear that
Every deductive system with a bi-level formalization (with the assertion system and the refutation
system) can be characterized by the basis:
F, Cn+ , Cn−1 !
and that the extension of Tarski’s theory of the definition MD(Sł) describes every such system.
This theory has been developed in the form of the theory of rejected propositions by
Wybraniec-Skardowska [62], and later also by Bryll [1]. Their researches have been a
continuation of investigations initiated by J. Słupecki and have been conducted under his
supervision to be later presented in co-authored papers (see Słupecki et al. [45, 46]).
The theory of rejected propositions contains many significant theorems that are not
counterparts of any theorem of Tarski’s theory [58]. The following are examples of such
theorems about the rejection operation Cn−1 :
• Cn−1 (∅)
= ∅ – it is normal,
• Cn−1 ( {X ⊆G : G ⊆ F }) = {Cn−1 (X) : X ⊆ G ⊆ F } – it is complete additive,
• −1 −1
Cn (X) = {Cn ({α}) : α ∈ X} – it is unit operation,
• α ∈ Cn−1 (X) ⇔ ∃β ∈ X(Cn−1 ({α} ⊆ Cn−1 ({β})) – it is a unit consequence.
In the theory the following formulation of the rule of rejection by detachment is valid:
• c αβ ∈ Cn+ (X) ∧ β ∈ Cn−1 (Y ∪ X) ⇒ α ∈ Cn−1 (Y ∪ X),
while in the Tarski’s theory the formulation of rule of detachment has the form:
• c αβ ∈ Cn+ (Y ) ∧ α ∈ Cn+ (Y ∪ X) ⇒ β ∈ Cn+ (Y ∪ X).
The set Y ⊆ F can be understood as a set of axioms A+ or the set of theses T + of a given
deductive system.
7 Rejection Operation as a Primitive Notion
There is a possibility of the axiomatization of a theory of rejected propositions in a dual

and equivalent way, i.e. assuming that a primitive notion is the rejection consequence
Cn−1 , while Cn+ is defined operation.
Such dual theory of rejected propositions was formulated by Wybraniec-Skardowska
[62] (see also [66, 67]). It can be understood as the theory describing the deductive
systems with the basis
F, Cn−1 , Cn+ !.
A theory of rejected propositions was developed also to formalize some problems of

methodology of empirical sciences, mainly by Bryll [1] (see also Słupecki et al. [46]).
The theories of rejected propositions have a natural interpretation, which was given by
Staszek [53]: the set Cn−1 (X) can be understood as a set of all rejection proofs on the
ground of a proposition of the set X, in the sense relating to the nature of rejection proofs
used by Słupecki in his researches on Aristotle’s syllogistic.
In the theory of rejected propositions can be defined the notion of Ł-decidability.
7.1 Rejection Operation as a Finitistic Consequence: Dual

Consequences
Rejection function Cn−1 can be generalized into a dual, finitistic consequence in the
usual meaning. The notion of the dual consequence dCn+ relating to the consequence
Cn+ was introduced by Wójcicki [61] by definition:

α ∈ dCn+ (X) ⇔ ∃Y ⊆ X ∧ card(Y ) < ℵ0 ( {Cn+ ({β}) : β ∈ Y } ⊆ Cn+ ({α})).
The dual consequence dCn+ is stronger than the rejected consequence Cn−1 (i.e. Cn−1 ≤
dCn+ ), though the former is linked with the latter by a number of interesting relationships
(see Spasowski [52]).
Facts: a. dCn+1 = Cn−1 and b. dCn−1 = Cn+1 ,
+1
where the unit consequence Cn is defined as follows:
α ∈ Cn+1 (X) ⇔ ∃β ∈ X(α ∈ Cn+ ({β}).
Certain studies of generalization of the notion of rejected expression in the form of a

function of a consequence of rejection or a dual consequence are discussed by Wybraniec-
Skardowska and Waldmajer [68].
The dual consequences Cn+ and dCn+ as well as Cn+1 and Cn−1 can be used to
study both true (asserted), and respectively, false (rejected) contents of a given theory.
The fact was noticed by Woleński [60] in his studies relating to Popper’s conception of a
comparison of scientific theories by their contents.
References
1. Bryll, G.: Kilka uzupełnień teorii zdań odrzuconych [Some supplements to the theory of rejected
sentences]. In: Wybraniec-Skardowska, U., Bryll, G.: Z badań nad teoria˛ zdań odrzuconych [Investi-
gations on the theory of rejected propositions]. Zeszyty Naukowe Wyższej Szkoły Pedagogicznej w
Opolu, Matematyka, seria B: Studia i Monografie, no 22, Opole (1969), pp. 133–154
2. Bryll, G.: Metody odrzucania wyrażeń [The methods of the rejection of expressions]. Akademicka
Oficyna Wydawnicza PLJ, Warszawa (1996)
3. Bryll, G., Maduch, M.: Aksjomaty odrzucone dla wielowartościowych logik Łukasiewicza [Rejected
axioms for Łukasiewicz’s many-valued logics]. Zeszyty Naukowe Wyższej Szkoły Pedagogicznej w
Opolu, Matematyka 6, 3–17 (1968)
4. Bryll, G., Sochacki, R.: Aksjomatyczne odrzucanie w inwariantnych rachunkach zdaniowych
Łukasiewicza [Axiomatic rejection in invariant Łukasiewicz’s propositional calculi]. Zeszyty
Naukowe Uniwersytetu Opolskiego, Matematyka 29, 29–37 (1995)
5. Bryll, G., Sochacki, R.: Elimination of the rejection rule by substituting in the three-valued
Łukasiewicz calculus. Acta Universitatis Wratislaviensis, Logika 18, 129–131 (1998)
6. Carnap, R.: Introduction to Semantics. Harvard University Press, Cambridge (1942)
7. Carnap, R.: Formalization of Logic. Harvard University Press, Cambridge (1943)
8. Carnielli, W.A.: Systematization of finite many-valued logics. J. Symb. Log. 52(2), 473–493 (1987)
9. Carnielli, W.A.: On sequents and Tableaux for many-valued logics. J. Non-Classical Logics 8(1),
59–76 (1991)
10. Citkin, A.: A meta-logic of inference rules. Syntax. Log. Logical Philos. 24, 313–337 (2015)
11. Dutkiewicz, R.: The method of axiomatic rejection for the intuitionistic propositional logic. Stud.
Logica 48(4), 449–459 (1989)
12. Goranko, V.: Proving unprovability in some normal modal logics. Bull. Sect. Log. Polish Acad. Sci.
20, 23–29 (1991)
13. Goranko, V.: Refutation systems in modal logic. Stud. Logica 53, 299–324 (1994)
14. Grzegorczyk, A.: Zarys logiki matematycznej. PWN, Warszawa (1961) (English trans.: Outline of
mathematical logic. R.Reidel Publishing Company, Dordrecht (1974))
15. Hintikka, J.: Form and content in quantification theory. Acta Philosophica Fennica 8, 7–55 (1955)
16. Iwanuś, B.: On Leśniewski’s elementary ontology. Stud. Logica 31, 73–147 (1972)
17. Iwanuś, B.: Proof of decidability of the traditional calculus of names. Stud. Logica 32, 73–90 (1973)
18. Iwanuś, B.: Z badań nad sylogistyka˛ Arystotelesa [Investigations on Aristotle’s syllogistic]. Zeszyty
Naukowe Wyższej Szkoły Pedagogicznej w Opolu, Matematyka 28, 41–56 (1992)
19. Kreisel, G., Putnam, H.: Unableitbarkeits-beweismethode für den intuitionistischen Aussagenkalkül.
Archiv für Mathematische Logik und Grundlagenforschung 3, 205–220 (1957)
20. Łukasiewicz, J.: Logika dwuwartościowa [Two-valued logic)] Przeglad ˛ Filozoficzny 23, 189–205
(1921) (English trans. by Wojtasiewicz, O.A. in: Łukasiewicz, J.: Selected Works. North-Holland,
Amsterdam (1970). Borkowski, L. (ed.), pp. 89–109)
21. Łukasiewicz, J.: Elementy logiki matematycznej [an authorized collection of Łukasiewicz’s lectures].
ed. Presburger, M. (1929). (Second edition, PWN, Warszawa (1958); English trans.: Elements of
Mathematical Logic. Pergamon Press, Oxford; The Macmillan Company, New York (1963))
22. Łukasiewicz, J.: O sylogistyce Arystotelesa [On Aristotle’s Syllogistic]. Sprawozdania z Czynności
i Posiedzeń Polskiej Akademii Umiej˛etności, no 44 (1939) (Reprinted in: Łukasiewicz, J.: Z
zagadnień logiki i filozofii. Pisma wybrane [Problems in logic and philosophy. Selected papers].
PWN, Warszawa (1961). Słupecki, J. (ed., selection, introduction and footnotes), pp. 220–227)
23. Łukasiewicz, J.: Aristotle’s Syllogistic from the Standpoint of Modern Formal Logic. Clarendon
Press, Oxford (1951) (Second edition in 1955)
24. Łukasiewicz, J.: On the intuitionistic theory of deduction. Indag. Math. Ser. A 3, 202–212 (1952)
(Reprinted in: Łukasiewicz, J.: Selected Works. North-Holland, Amsterdam (1970). Borkowski, L.
(ed.), pp. 325–340)
25. Łukasiewicz, J.: A system of modal logic. J. Comput. Syst. 1(3), 111–149 (1953) (Reprinted in:
Łukasiewicz, J.: Selected Works. North-Holland, Amsterdam (1970). Borkowski, L. (ed.), pp. 352–
390)
26. Łukasiewicz, J.: Z zagadnień logiki i filozofii. Pisma wybrane [Problems in logic and philosophy.
Selected papers]. PWN, Warszawa (1961). Słupecki, J. (ed., selection, introduction and footnotes)
27. Łukasiewicz, J.: Selected Works. North-Holland, Amsterdam (1970). Borkowski, L. (ed.)
28. Maduch, M.: O Łukasiewiczowskich regułach odrzucania [On Łukasiewicz’s rules of rejection].
Zeszyty Naukowe Wyższej Szkoły Pedagogicznej w Opolu, Matematyka 13, 115–122 (1973)
29. Scott, D.: Completeness proofs for intuitionistic sentential calculus. In: Summaries of talks presented
at the Summer Institute of Symbolic Logic, Cornell University (1957) (Second edition by: Commu-
nications Research Div., pp. 231–242, Princeton (1960))
30. Skura, T.: A complete syntactic characterization of the intuitionistic logic. Rep. Math. Log. 23, 75–80
(1989)
31. Skura, T.: Refutation calculi for certain intermediate propositional logics. Notre Dame J. Formal Log.
33, 552–560 (1992)
32. Skura, T.: Some results concerning refutation procedures. Acta Universitatis Wratislaviensis, Logika
15, 83–95 (1993)
33. Skura, T.: Syntactic refutation against finite models in modal logic. Notre Dame J. Formal Log. 35,
573–582 (1994)
34. Skura, T.: A Łukasiewicz-style refutation system for the modal logic S5. J. Philos. Log. 24, 573–582
(1995)
35. Skura, T.: Refutation and proofs in S4. In: Wansing, H. (ed.) Proof Theory of Modal Logic, pp. 45–51.
Kluwer Academic, Dordrecht (1996)
36. Skura, T.: Aspects of refutation procedures in the intuitionistic logic and related modal systems. Acta
Universitatis Wratislawiensis, Logika 20, Wrocław (1999)
37. Skura, T.: Refutation Methods in Modal Propositional logic. Semper, Warszawa (2013)
38. Słupecki, J.: Uwagi o sylogistyce Arystotelesa [Remarks about Aristotle’s syllogistic]. Annales
Universitatis M. Curie-Skłodowska I(3), 187–191 (1946)
39. Słupecki, J.: Z badań nad sylogistyka˛ Arystotelesa. Prace Wrocławskiego Towarzystwa Naukowego
(B), 6, Wrocław (1948) (English trans. in: Słupecki, J.: On Aristotelian syllogistic. Stud. Philos. 4,
275–300 (1951))
40. Słupecki, J.: On Aristotelian syllogistic. Stud. Philos. 4, 275–300 (1951)
41. Słupecki, J.: St. Leśniewski’s calculus of names. Stud. Logica 3, 7–76 (1955)
42. Słupecki, J.: Funkcja Łukasiewicza [The Łukasiewicz function]. Zeszyty Naukowe Uniwersytetu
Wrocławskiego, seria B, 3, 30–40 (1959)
43. Słupecki, J.: Ł-decidability and decidability. Bull. Sect. Log. Polish Acad. Sci. 1(3), 38–43 (1972)
44. Słupecki, J., Bryll, G.: Proof of Ł-decidability of Lewis system S5. Stud. Logica 12, 99–107 (1973)
45. Słupecki, J., Bryll, G., Wybraniec-Skardowska, U.: Theory of rejected propositions, part I. Stud.
Logica 29, 75–123 (1971)
46. Słupecki, J., Bryll, G., Wybraniec-Skardowska, U.: Theory of rejected propositions, part II. Stud.
Logica 30, 97–145 (1972)
47. Smullyan, R.M.: First Order Logic. Springer, Berlin (1968)
48. Sochacki, R.: Axiomatic rejection in the implicational-negational invariant sentential calculi of
Łukasiewicz. Bull. Sect. Log. Polish Acad. Sci. 36, 1–6 (2007)
49. Sochacki, R.: Rejected axioms for the “nonsense-logic” W and the k-valued logic of Sobociński. Log.
Logical Philos. 17, 321–327 (2008)
50. Sochacki, R.: Metody refutacyjne w badaniach nad systemami logicznymi [Refutation methods
in research on logical systems]. Studia i Monografie, nr 445, 1–122, Wydawnictwo Uniwersytetu
Opolskiego, Opole (2010)
51. Sochacki, R.: Refutation systems for a system of nonsense-logic. Log. Logical Philos. 20, 233–239
(2011)
52. Spasowski, M.: Some connections between Cn, Cn−1 , dCn. Bull. Sect. Log. Polish Acad. Sci. 2(1),
53–56 (1973)
53. Staszek, W.: A certain interpretation of theory of rejected propositions. Stud. Logica 30, 147–150
(1973)
54. Suchoń, W.: Smullyan’s method of constructing Łukasiewicz’s m-valued implicational-negational
sentential calculus. In: Wójcicki, R., Malinowski, G. (eds.) Selected Papers on Łukasiewicz’s
Sentential Calculi, pp. 119–124. Ossolineum, Wrocław-Warszawa (1977)
55. Surma, S.J.: A method of constructing Łukasiewicz’s algebras and its applications to a Gentzen-style
characterization of finite logics. Rep. Math. Log. 2, 49–54 (1974)
56. Surma, S.J.: An algorithm of axiomatizing every finite logic. Rep. Math. Log. 3, 57–62 (1974)
57. Tarski, A.: Fundamentale Begriffe der Methodologie der deductiven Wissenschaften I. Monatshefte
für Mathematik und Phisik 37, 361–404, Leipzig (1930) (English trans.: Fundamental concepts of
methodology of the deductive sciences. In: Tarski, A.: Logic, Semanics, Metamathematics. Papers
from 1929 to 1938. Second edition editing and introduced by John Corcoran, Hackett Publishing
Company, Indianapolis, Indiana (1983) (First edition published in 1956 by OUP; trans. by Woodger
JH), pp. 60–109)
58. Tarski, A.: Über einige fundamentale Begriffe der Metamathematik. Comptes Rendus des séances
de la Société des Sciences et des Lettres de Varsovie 23, 22–29 (1930) (English trans.: On some
fundamental concepts of metamathematics. In: Tarski, A.: Logic, Semanics, Metamathematics. Papers
from 1929 to 1938. Second edition editing and introduced by John Corcoran, Hackett Publishing
Company, Indianapolis, Indiana (1983) (First edition published in 1956 by OUP; trans. by Woodger
JH), pp. 30–37)
59. Tarski, A.: Logic, Semanics, Metamathematics. Papers from 1929 to 1938. Second edition editing
and introduced by John Corcoran, Hackett Publishing Company, Indianapolis, Indiana (1983) (First
edition published in 1956 by OUP; trans. by Woodger JH)
60. Woleński, J.: On comparison of theories by their contents. Stud. Logica 48, 617–622 (1989)
61. Wójcicki, R.: Dual counterparts of consequence operation. Bull. Sect. Log. Polish Acad. Sci. 2(1),
54–57 (1973)
62. Wybraniec-Skardowska, U.: Teoria zdań odrzuconych [The theory of rejected propositions]. In:
Wybraniec-Skardowska, U., Bryll, G.: Z badań nad teoria˛ zdań odrzuconych [Investigations on
the theory of rejected propositions]. Zeszyty Naukowe Wyższej Szkoły Pedagogicznej w Opolu,
Matematyka, seria B: Studia i Monografie, no 22, Opole (1969), pp. 5–131
63. Wybraniec-Skardowska, U.: Badania Jerzego Słupeckiego nad sylogistyka˛ Arystotelesa i ich rezonans
we współczesnej logice [Jerzy Słupecki’s investigations on Aristotle’s syllogistic ad their response in
the contemporary logic]. Zeszyty Naukowe Wyższej Szkoły Inżynierskiej w Opolu, seria Matematyka
4, 35–61 (1983)
64. Wybraniec-Skardowska, U.: On the notion and function of the rejection of propositions. Acta
Universitatis Wratislaviensis, No 2754, Logica 23, 179–202 (2005)
65. Wybraniec-Skardowska, U.: Rejection in traditional and modern logics. In: Proceedings of the 10th
Panhellenic Logic Symposium, Samos, Greece, 11–15 June 2015, pp. 59–62 (2015)
66. Wybraniec-Skardowska, U.: On the mutual definability of the notions of entailment, rejection, and
inconsistency. Axioms 5, 15 (2016). https://doi.org/10.3390/axioms5020015
67. Wybraniec-Skardowska, U., Bryll, G.: Z badań nad teoria˛ zdań odrzuconych [Investigations on
the theory of rejected propositions]. Zeszyty Naukowe Wyższej Szkoły Pedagogicznej w Opolu,
Matematyka, seria B: Studia i Monografie, no 22, Opole (1969)
68. Wybraniec-Skardowska, U., Waldmajer, J.: On pairs of dual consequence operations. Log. Univers.
5, 177–203 (2011)
Bolesław Sobociński: The Ace of the Second
Generation of the LWS
Abstract The here presented biography of Bolesław Sobociński (1906–1980) describes

the history of his life, his patriotic activity and scientific results. Sobociński, as a pupil
of Jan Łukasiewicz and Stanisław Leśniewski, was one of the main representatives of
the second generation of the Lvov-Warsaw School, a member of the Warsaw School
of Logic. He is known as the most influential popularizer of Leśniewski’s systems of
prothotetics, ontology and mereology, on the international stage, as well as the author
of many important results in symbolic logic: many-valued logics, modal logics, as well
in set theory and theory of algebras. The international community of logicians knows
Sobociński as the creator and, for many years, editor of the prestigious scientific periodical
Notre Dame Journal of Formal Logic. The material is divided in the following parts:
(1) Life, (2) National service, (3) Academic career, (4) Interests and achievements,
(5) Sobociński presonally, (6) Selected publications by Bolesław Sobociński.
Keywords Bolesław Sobociński · Warsaw School of Logic · Leśniewski’s systems ·

Symbolic logic
In search of source materials for Bolesław Sobociński’s biography, I came across the
work entitled “Profesor Sobociński i kolega Bum” [“Professor Sobociński and colleague
Bum”]1 by a Polish mathematician Krzysztof Tatarkiewicz (1923–2011) [18]. This text is
exceptional, mainly because the author was able to capture the personality of Sobociński
through a direct contact with him, through anecdotes heard from the members and
supporters of the Lvov-Warsaw School, through the recollection and comments of his
own father—Władysław Tatarkiewicz (1886–1980). It is especially worth noting how
Tatarkiewicz made an attempt to refer to a particularly difficult chapter in Sobociński’s
life: his national service in the Second World War and in the early post-war period. The
great historical value of Tatarkiewicz’s paper stems from historical data and caution in
passing judgment too quickly.
1 English translations of Polish titles are put in square brackets.

Sobociński’s biography was written and published by K. Tatarkiewicz in Polish. The

aim of this paper is to present one of the leading personalities of the second-generation of
the Lvov-Warsaw School, and the Warsaw School of Logic more specifically, to a broader
public, partly drawing on the text of this biography. We repeat a great deal of information
after K. Tatarkiewicz, supplementing it, whenever possible, with data from other texts
about Sobociński, and verifying the entire piece with our own research conducted in
the Warsaw archives. We also add a commentary on the academic achievements of
Sobociński.
1 Life
Bolesław Sobociński was born in St. Petersburg (Russian Empire) on 28.06.1906. He was
the only child of Waleria, née Jasiewicz, and Antoni Sobociński. Waleria Sobocińska was
the daughter of Katarzyna, née Misiewicz, and Karol Jasiewicz. She was born in Pułtusk
(now Poland) on 31.05.1873 and died in Warsaw on 2.02.1942. Antoni Sobociński was
the son of Bolesława, née Sokołowska, and Antoni Sobociński; his sister Wanda (married
name Chawłowska) was born in Moscow around 1867 and died in Warsaw on 24.06.1934
[5]. It is known that Sobociński’s parents married in 1905.
In 1916, Sobociński started attending the Catholic Gymnasium at the Saint Catherine
of Alexandria Church in St. Petersburg. After 2 years, his parents decided to change his
schooling to home education. In his 1926 biography from [1], Sobociński explains this
change, stating “[. . . ] in 1918, I was taken away by my parents due to the gymnasium’s
nationalisation and the introduction of the Bolshevik curriculum. Because of the condi-
tions created by the Russian Revolution, I had to abandon regular education.”*2 (A deep
dislike for communism in general, and not just Russian, was a distinctive characteristic of
Sobociński’s outlook in his adult life.)
Following the Riga treaty, the Sobocińskis were awarded Polish citizenship, and in
October 1922 they came to Warsaw as repatriates. Here, Bolesław obtained his extramural
baccalaureate (he started attending courses at the humanistic gymnasium in February
1923, and passed his maturity examination on 23.02.1926 [1]). In 1926, he started his
studies in philosophy at the Warsaw University, where he obtained a Master of Arts degree
(30.06.1930), a doctoral degree (30.06.1936), and successfully completed the habilitation
(14.06.1939).
At the beginning of the war (around 6.09.1939), Sobociński fled from Warsaw and
headed to the estate of W. Tatarkiewicz’s mother-in-law, Ksawera Potworowska, located
near Lublin (Radoryż). There he met Krzysztof Tatarkiewicz, who was sixteen at the
time. Sobociński planned to reach the estate of the Skirmunt family (at that time in
Poland, now Belarus), but he was stopped (by a “gang of peasants”—as he described the
2 Translations of Polish fragments done by Daniela Szczygieł are marked with: *.

Bolesław Sobociński: The Ace of the Second Generation of the LWS 601
attackers) and taken to a town nearby (Motol).3 Sobociński kept his acquaintance with the
Skirmunts secret (the whole family was murdered in 1939 by the Bolsheviks); in this way,
he saved his life and was able to leave Motol. He travelled to Vilnius first (at that time
under Lithuanian). There, he gave a lecture entitled “O prototetyce prof. Leśniewskiego”
[“On Professor Leśniewski’s Prothotetics”] at a meeting of the Vilnius Philosophical
Society (18.10.1939). He went to the estate belonging to his maternal relative, Karol
Parczewski (1875–1957), where he stayed from December 1939 to mid-1941 (Stończe,
Lithuania, at that time not occupied by the Soviets). In 1941, he returned to Warsaw.
In February 1942, Sobociński’s mother died. Until the outbreak of the Warsaw Uprising
against Germans in 1944, he attended seminars at the underground Warsaw University,
where he gave lectures in logic and edited the manuscripts by his master, Stanisław
Leśniewski (deceased just before the war). He actively participated in the independence
conspiracy within the structures of the underground National Armed Forces. Sobociński
survived the Warsaw Uprising and he was a witness to the last hours of the life of his
friend and companion in the underground fight, Jan Salamucha (1903–1944), who was
killed in the uprising by SS-Sturmbrigade RONA4 soldiers. After the fall of the uprising,
K. Tatarkiewicz met Sobociński at a temporary camp near Warsaw (around 12 or 13
August 1944). It is also known that in 1944 Sobociński helped to organise the journey
of the Łukasiewicz family to Switzerland. Together with Leśniewski, Jan Łukasiewicz
(1878–1956) played a crucial role in his academic career. There are few known facts
about Sobociński’s life during the following 2 years, until September 1946. This lack of
data is connected with his pro-independence activity. Sobociński held prominent functions
within the underground National Armed Forces, and for the rest of his later life he kept
absolute discretion regarding any issues related to this activity. Before the war (around
1923) he met his future wife Ewa, née Wrześniewska (1909–2005). Born in Warsaw, she
graduated in Philosophy at the Warsaw University (she was a student of W. Tatarkiewicz).
Later she was activist of the external women’s Wiara i Wola [Faith and Will] structure
cooperating with the Polish Organisation.5 (Wrześniewska held a managerial function
there [14, p. 251].)
After the war, in 1945, Sobociński was offered a position at the newly-established
University in Łódź, which he declined. Two or three days before starting this job, he
withdrew from public life. At the same time, he was wanted by the communistic Office of
Public Security.
In September 1946, he secretly left Poland, together with Ewa Wrześniewska.6 They
travelled to Regensburg via Katowice and they eventually married in Germany.
Between 1946 and 1949, the Sobocińskis stayed in Brussels. During this time,
Sobociński worked at the Polish Scientific Institute in Belgium. Between March 13 and
3 We have some information about Sobociński’s adventures at the beginning of the war, ended with his
stay at the Stończe estate, through his direct testimony, which he gave in his letters to Father Bocheński,
parts of which were published in [13, pp. 123–133] (we know their full versions from [6]). These events
were also described by Tatarkiewicz in [18].
4 RONA—Russkaya Osvoboditelnaya Narodnaya Armiya.
5 The Polish Organisation was formed already in 1934 as a secret right-wing organisation and a
continuation of the National Radical Camp, later (partly) included in the National Armed Forces.
6 On Łukasiewicz’s request, their travel was facilitated by Zbigniew Jordan (1911–1977).
30, 1948 they went to Freiburg (Switzerland) to see Father Józef M. Bocheński (1902–
1995). Together with his wife, Sobociński visited the Łukasiewicz family in Dublin (July
1947). He also travelled to Paris.
In December 1949, the Sobocińskis moved to the States, where they remained for the
rest of their lives. Thanks to the efforts of Father Bocheński, Sobociński was employed
at the College of St. Thomas in St. Paul, USA (Minnesota). (He also availed himself the
help of Marian Heitzman (1899–1964), already employed there.)
During the first period of their stay in the States, they were not entirely successful.
Sobociński lost his job after the first semester. Due to financial constraints, Ewa
Sobocińska took up a job as a typist. From c. 1951 to c. 1955, Sobociński got a permanent
position at the Institute of Applied Logic in St. Paul. Ewa Sobocińska also became
employed there as an IT engineer.
In 1955, the Sobocińskis received American citizenship, and a year later they moved
to South Bend, USA (Indiana), where Sobociński became employed at the Faculty of
Philosophy at the University of Notre Dame. Mrs Sobocińska was able to get a job at
the same University, in the technical department. Due to financial stability they bought a
house in South Bend [18, p. 141]. In 1972, Sobociński retired.
Bolesław Sobociński died on 21.11.1980, after a long illness. His wife Ewa died on
15.01.2005. Both of them were buried at the Cedar Grove university cemetery (St. Joseph
County Indiana, USA).
The Sobocińskis had no children.
2 National Service
Sobociński was actively involved in the social and political life of pre-war Poland. In
1934, he became involved with the National Radical Camp (a fraction ABC), and the
Polish Organisation.7
A reconstruction of Sobociński’s patriotic activity and his ties with the Polish
underground after the start of the war is clearly difficult: the information available is
incomplete, and the fragments available were preserved despite the planned distortion
of the image of right-wing pro-independence groups by the communist authorities. In
1949, Sobociński wrote to Bocheński: “It is obvious that my person there [in Warsaw] is
completely unpalatable (no less than 7 trials ended with a death sentence—all executed,
in which I played the role of the lead but absent actor and leader). [. . . ] Matters are also
complicated by some other concerns (the wish not to endanger those who survived) so
that even anti-communist elements know nothing about me”* [13, p. 128].8
We will only outline this chapter of Sobociński’s life, referring to information from [14,
15, 18]. As far as it was possible, we compared these datas with the documents available
in the Polish Institute of National Remembrance [4].
7 Cf.footnote 5.
8 Former employee of the Office of Public Security—Lt.-Col. Różański, called Sobociński “a famous spy”
in his report [7].
We know that after his return to Warsaw from the Stończe estate, in spring of
1941, Sobociński became involved in the activities of the Polish Organisation. He
operated under the organisational nickname “Rawicz” (or “prof. Rawicz”).9 He joined the
Christian-National Lizard Union (National Order) and led the 7th group responsible for
secret service [14, p. 247]. In the spring of 1942, he became a member, and subsequently
the chairman of the Executive Committee of the Polish Organisation, to which also
Salamucha belonged. He performed important functions in the National Armed Forces:
he was the head of the IVa Office of the Central Intelligence Service of the National
Armed Forces Command. In 1943, he was a full member—together with Salamucha—of
the highest level (A) of the Political Committee (the Polish Organisation leadership). In
July 1944, he participated in the first secret meeting of A level members (the meeting
was organised in his flat) during the occupation. At the meeting, an evacuation plan for
the Polish Organisation leaders was also discussed. Sobociński and Salamucha decided
to stay in Warsaw [14, pp. 329–331].10 Members of the Political Committee met again in
November 1943 in Kraków, to discuss the possibility to “establish tactical contact with the
Germans”* [14, p. 347] and broadcast an anti-communist Polish radio station. The leaders
of the Political Committee opposed this plan and resigned. We do not know Sobociński’s
position on this matter. However, we do know that in 1945, he was elected chairman of
the Political Committee, and held this position until his departure in the autumn of 1946.
(From January 1945 he was the political commissioner for the “South” and “West” areas
of the National Armed Forces.) Thanks to his foresight, Sobociński avoided arrest by
the Soviet authorities, and he managed to avoid the so-called Trial of the Sixteen. On
27.03.1946, delegates of the Polish government and the Home Army were deceitfully
invited to take part in another round of talks about a “compromise” on the composition
of “the temporary Government of National Unity” established under the Yalta agreement
(1945), on which basis Poland fell under the Soviet control. The meeting (in Pruszków
near Warsaw) was attended by the Polish Deputy Prime Minister J. S. Jankowski, last
Commander-in-Chief of the Home Army, General L. Okulicki, and 14 other leaders of
Underground Poland. Sobociński was also to attend that meeting, but having explored
the site, he decided not to go and thus was not taken to Moscow for a staged Stalinist
trial. Until September 1946, Sobociński went into hiding (he was wanted by the Office
of Public Security) and in the end—after consulting with the leadership of the Internal
Organisation—he made a decision to leave the country. We do not know the extent of
Sobociński’s political activity after his departure from Poland. His correspondence with
Bocheński leads us to conclude that at least at the beginning of his stay in Brussels
Sobociński was an energetic political activist. He wrote: “[. . . ] I am of the opinion that
the fuss from 1939 has not ended yet, we live in times of respite, short or long, and for this
9 Among his relatives and friends he was called “Bum”; this nickname (used also in [4]) was supposedly
given to Sobociński to commemorate a comment he made during one of the bombardments of Warsaw
(according to Tatarkiewicz, Sobociński allegedly said “in a calm voice and with a characteristic
intonation: What a great boom”* and he made his friends laugh [18, p. 136]. Sobociński stuttered and
tried to “help” himself by using the word “bum”). Muszyński mentions some other nicknames used by
Sobociński: “Professor”, “Uncle” [14, p. 247].
10 Tatarkiewicz denies the claim that during the uprising, Sobociński allegedly helped crack ciphers at the
Home Army Command [18, p. 130].

reason, my presence in Europe will be necessary.”* (letter of 16.12.1946 [6]).11 However,

we find no historical records describing his further political involvement.
3 Academic Career
Sobociński was awarded the degree of Doctor of Philosophy 30.06.1936. His doctoral
thesis entitled “Aksjomatyzacja pewnych wielowartościowych systemów teorii dedukcji”
[“Axiomatisation of certain many-valued systems of the theory of deduction”] was
supervised by Łukasiewicz.
As already mentioned, in 1930 Sobociński was awarded an MA in Philosophy at the
Faculty of Humanistics at the Warsaw University. His dissertation, entitled “Z badań
nad teoria˛ dedukcji” [“Investigations on the theory of deduction”*], was also supervised
by Łukasiewicz. During his studies in philosophy, he was taught and examined, among
others, by W. Tatarkiewicz (history of modern philosophy), T. Kotarbiński (logic),
W. Witwicki (psychology), K. Ajdukiewicz (theory of cognition, Frege’s logic), J.
Łukasiewicz (Kant’s philosophy, contemporary philosophy), W. Wasik ˛ (Polish philos-
ophy), W. Sierpiński (set theory), S. Leśniewski (introduction to mathematics, group
theory) [3].
After completing his master’s studies, Sobociński maintained and developed his
interest in logic and became actively involved in publishing.
From 1931 (or 1932) and until the outbreak of the war, he worked as the secretary of
Przeglad˛ Filozoficzny (at that time edited by W. Tatarkiewicz; in 1939 Sobociński took up
the position of editor while Tatarkiewicz became editor-in-chief).
In 1936, he became a member of the Interim Board of the Polish Association for
Logic (together with A. Lindenbaum, A. Mostowski, and A. Tarski), and 2 years later, he
started his work as an editorial secretary for the journal Collectanea Logica established on
Łukasiewicz’s initiative and published by the Polish Association for Logic. In the same
year, he also became the secretary for Organon (founded in 1936 by S. Michalski as
an international review and published by the Mianowski Institute for the Promotion of
Science and Letters).12
In 1934, he attended a congress for the students of the new Faculty of Mathematics
and Natural Sciences at the Warsaw University (where he later defended his doctorate).
Immediately after his doctoral defence, Sobociński began working there as the first
titular assistant (1934 (36?)–1938), and later as a senior assistant (1938–1939) for the
11 Maybe he knew about and took into account the possible implementation of the Operation Unthinkable
by Churchill?
12 Unfortunately, the proofs of almost all texts that were due to be published in these journals were
burnt during the Warsaw Uprising (in the flats of Sobociński and W. Tatarkiewcz). The first issue of
Collectanea Logica, printed in August 1938, did not come out because of the outbreak of the war, and
was burnt together with the second volume prepared by Sobociński, in his apartment. In additions to the
journals, Leśniewski’s original manuscripts, as well as their editions done by Sobociński, were also burnt.
According to Tatarkiewicz [18, p. 126], at that time a few dozen academic papers were lost (including 13
texts for Collectanea, 4 of which were authored by Sobociński).
Philosophy of Mathematics Seminars led by Leśniewski. Thus, he formally joined the

Warsaw School of Logic.
During that period, Sobociński cooperated closely with Leśniewski, and maintained
his interest in protothetics, ontology, and mereology for many years.
Leśniewski died on 13.05.1939, only a few weeks before Sobociński’s habilitation.
On 14.06.1939, Sobociński passed habilitation colloquium at the Faculty of Mathe-
matics and Natural Sciences of the Warsaw University presenting dissertation entitled “Z
badań nad prototetyka” ˛ [“Research on Protothetics”]. The dissertation was reviewed by
Łukasiewicz and S. Mazurkiewicz (the committee also comprised Białobrzeski [3]). The
Polish Ministry of Religion and Public Education was unable to approve the habilitation
before the outbreak of the war.
Understandably, the war also ruined the plans for Sobociński’s academic career;
however, it did not diminish his interest in academic matters.
During his stay at the Stończe estate (1939–1941), Sobociński conducted research in
logic, and went back to academic research and teaching activities immediately after his
return to Warsaw. As already mentioned, he participated in secret seminars; he also took
over Łukasiewicz’s lectures in the underground Warsaw University. At the same time,
he took part in a seminar on the history of logic, conducted by Łukasiewicz and A.
Krokiewicz. Together with Łukasiewicz, he was an academic supervisor for H. Rasiowa
(her first master’s thesis, written under the direction of Łukasiewicz and Sobociński, was
burnt during the Warsaw Uprising). In the autumn of 1943, after Łukasiewicz’s departure,
Sobociński delivered lectures in logic for mathematicians (among his students were K.
Szaniawski and K. Tatarkiewicz). Until the outbreak of the uprising, he worked on editing
Leśniewski’s notes given to him in 1939 by Leśniewski’s widow. As mentioned earlier,
the results of his work (approx. 1000 pages of text) and the original manuscripts by
Leśniewski, were also burnt during the Uprising.
After the end of the war, it was possible to attempt to complete the habilitation
procedure. On 14.09.1945, the Board of the Faculty of Mathematics and Natural
Sciences of the Warsaw University applied to the Ministry of Education for approval
of Sobociński’s habilitation, which had taken place in 1938/39. The Ministry granted the
application in May 1947, but Sobociński did not receive its official response—he had
already gone abroad [3].
Also in 1945, Kotarbiński made efforts to secure for Sobociński the position of
associate professor and head of the Department of Deductive Sciences at the Faculty
of Mathematics and Natural Sciences of the Łódź University (at that time, Kotarbiński
was Rector of the Łódź University [3]). On 15.09.1945, Sobociński accepted the post,
but had to miss his first classes.13 As we already know, due to the threat of arrest, he
decided to leave Poland illegally. As late as 15.11.1946, Kotarbiński, as Rector of the Łódź
University, informed the Ministry of Education that “the appointment of Dr Bolesław
Sobociński associate professor of deductive sciences at the Faculty of Mathematics and
Natural Sciences of the Łódź University is no longer valid”* [3]. The end of his formal
13 According to K. Tatarkiewicz, who refers to information obtained from C. Lejewski, Sobociński was
also to be employed at the Maria Curie-Skłodowska University (Lublin), and allegedly failed to assume
this post in the same way. However, Tatarkiewicz could not find any documents in the archives of the
Maria Curie-Skłodowska University to prove this [18, p. 132].
ties with the Warsaw University and the Łódź University can be dated, respectively, to
28.02.1946 (when of the Rector of the Warsaw University decided to cease salary payment
[3]) and 5.12.1947 (when the Ministry of Education was notified by the Rector of the Łódź
University that Sobociński “failed to report to assume duties and his place of stay is [. . . ]
unknown”* [3]).
Sobociński’s departure from Warsaw to Brussels in 1946 is associated with more
professional troubles. As already mentioned, Sobociński was employed by the Polish
Scientific Institute in Belgium (headed by J.F. Drewnowski). He was given a research
position after Łukasiewicz’s retirement and was supposed to continue teaching classes
in. The stay in Brussels was difficult for Sobociński: he struggled financially, feared
another war turmoil,14 was troubled by the lack of scientific contact with logicians and
philosophers he had known previously.
The departure for the States in 1949 was to open up a new and better life, as well as
more professional opportunities. With recommendations from Bocheński and Heitzman,
Sobociński became employed at the College of St. Thomas, as previously mentioned, as
a philosophy teacher. He taught Thomism, but his classes were not popular (according to
Bocheński, Sobociński’s diligence reflected in his rigorous reference to original source
texts was not entirely appropriate, considering the rather basic level of the school [9,
p. 120]). In the same year, Sobociński’s lectures were suspended. Between 1951 and
1955, Sobociński acted as the Director of Research at the Institute of Applied Logic at
St. Paul’s, where he dealt with logical issues for the purposes of the computer industry.
Thanks to Bocheński’s help once again, in 1956 he was invited to an interview by the
authorities of the Notre Dame University. According to Bocheński, because of a somewhat
unfortunate course of the interview, Sobociński was offered the lowest possible position of
Research Associate.15 However, in 1961 (1958?), he was given the position of a Professor
of Philosophy, and held this post until his retirement (1972). He gave lectures even in the
autumn of 1979 (this date is provided by Rickey [8, p. 11]).
At Notre Dame, Sobociński was successful again with his research (which we will
describe in the next section) and editorial efforts; he also proved to be an excellent teacher.
In the early years of his work, he may not have been as popular with his students as
Łukasiewicz, but at Notre Dame he was able to restore, to a certain degree, the atmosphere
of the Warsaw School of Logic and encourage a big group of young logicians to become
involved in academic work. Over the entire period of his career, he taught such courses as:
symbolic logic, modal logic, combinatory logic, algebraic logic, metalogic, Leśniewski’s
systems, basic mathematics, logical basis of set theory, theory of cardinal and ordinal
numbers (a detailed list of all subjects is provided in [8, pp. 9–10]). He successfully
supervised fifteen Doctors of Philosophy (R.E. Clay, J.T. Canty, and V. F. Rickey among
others); he also worked with other students on the results published by them between
14 In his correspondence with Bocheński he mentioned anxiety in each letter, and repeatedly enquired
about the possibility of leaving Belgium. In 1948, he asked Bocheński several times to help him to obtain
a Spanish visa (due to concerns about the Berlin Blockade by the Soviets?). Eventually, he declared he
wanted to leave Europe altogether [6].
15 Bocheński recalls that during the meeting he had a problem with communication due to poor knowledge
of English.
1972 and 1979. (A list of Sobociński’s students and doctoral students is also provided in
[8, pp. 7–9].)
In the States, he resumed his publishing activities. Between 1953 and 1954, he was the
editor of The Journal of Computing Systems. In 1960, he founded one of the prestigious
logic journals, the Notre Dame Journal of Formal Logic (NDJFL), and edited it until 1978.
He published almost 70 papers in the NDJFL (the last one in 1979).
In his recollection of Sobociński, C. Lejewski states: “The academic achievements of
Sobociński and his contributions in the areas of teaching and editing are clearly more than
average”* [10, p. 401]. Let us now take a brief look at the main directions and results of
his research.
4 Interests and Achievements
As Lejewski accurately remarks, Sobociński’s intellectual inclinations combined the

minds of his two great teachers: Jan Łukasiewicz and Stanisław Leśniewski [11]. His
philosophical background and the “mathematical” way of thinking, were inherited by
Sobociński from Łukasiewicz; likewise, a great share of his intellectual involvement
was dedicated to the work of Leśniewski—a prominent formalist, who officially denied
(almost) all existing philosophical perspectives, but created deeply philosophical systems:
protothetics, ontology and mereology. Sobociński researched and improved these theories
in the “mathematical” style of Łukasiewicz.
His early philosophical interests are known because of his collaboration with the
Kraków Circle—the Catholic “branch” of the Lvov-Warsaw School, established in 1936
at the initiative of Bocheński and Salamucha (with support from Łukasiewicz). Bocheński
considered Salamucha, Drewnowski, Sobociński and himself as key members of the
association. The most prominent exponent of the Circle—whose aim was to improve
methods of practising “Catholic” philosophy and theology—was Salamucha. Sobociński
did not publish any papers concerning the topics researched by the school, and had an
“advisory” role on the subject of logic. However, as evidenced by his correspondence
with Father Bocheński, Sobociński also had his own, original philosophical reflections
going beyond mere formalism, linked to the topics addressed by the Circle. In his letters,
he discussed the issue of the existence of universals, and proposed their formalisation;
he considered the possibility to formalise the concept of the Universe on the basis
of mereology, and outlined the applications of mereological concepts to theological
issues; he also wrote remarks on analogy [6]. Recently, there have been attempts at
analysing and reconstructing some of these philosophical ideas of Sobociński: [12, 16],
and also [17]. The main ideas of the Kraków Circle were still at play in a few activities
undertaken jointly by Bocheński and Sobociński after the war, during their stay at Notre
Dame. Here, together with theologian and philosopher Ivo Thomas (1912–1976),16 they
actively promoted the logical style of doing philosophy, and in 1956 their efforts led to a
16 Ivo
Thomas was invited to Notre Dame thanks to Bocheński’s efforts; he collaborated in the creation of
NDJFL, and attended Sobociński’s seminars. Bocheński recollects that their team was called “Sobo, Ivo
and Bobo” [9, p. 222].
meeting of some hundred logicians and mathematicians to discuss the status and nature of
universals (among others, the discussion was attended by N. Goodman and A. Church).17
If we measure the creative contribution to a particular area of scientific research by
the quantity and quality of publications, then we can safely say that Sobociński achieved
remarkable results in logic. He published often, and his works were of recognised and
substantial value. His achievements include almost 90 academic texts.
Lejewski classifies Sobociński’s outputs into the following research areas: (1) symbolic
logic: sentential systems and Leśniewski’s systems, (2) border between logic and
mathematics: set theory, axiomatics of algebraic systems, (3) history of logic [11]. This
list can be supplemented by adding Sobociński’s interest in history of contemporary
Polish philosophy and philosophical didactics.
Before the war, Sobociński wrote several papers of a rather local nature. He published
a guide to philosophical studies at the Warsaw University, a report on Polish publishing
in the area of philosophy from 1918 to 1936, and a description of the current trends in
Polish philosophy, after the 3rd Polish Philosophical Meeting in Kraków in 1936.
Sobociński’s contribution to the history of logic includes two brief biographies of
Łukasiewicz, a brief biography of Salamucha, and a text in Italian on the history of the
Polish school of logic (from 1957).
The main core of Sobociński’s achievements are in the area of symbolic logic and
concern several issues in logic and mathematics.
Let us start with certain issues in sentential logics—the subject matter of Sobociński’s
master’s, doctoral and post-doctoral dissertations. In his master’s dissertation, Sobociński
positively solved two proof-theoretic problems concerning classical sentential logic, first
put by Łukasiewicz and Tarski (the original text can be found in [1]). He proved that: (i)
there exists one sole organic axiom for sentential logic, made only of original symbols
of implication and negation. (The axiom is organic, when none of its sub-formulae is a
thesis of a given logic). Sobociński was the first to formulate such an axiom, which—
using the Polish notation (without brackets)—comprised 139 symbols. Subsequently, he
formulated further, shorter formulae of that type (with 47 and 27 symbols). Addressing
Tarski’s problem, he demonstrated that (ii) for each natural number n, classical sentential
logic has a basis comprising exactly n elements. (A set of formulae Y is a basis for a
set of formulae X, when Y is an independent set of formulae—i.e. where for any of
its sub-sets Z, which has exactly the same consequences as Y , Y = Z—and when
Y and X have the same consequences). It can be said that according to (ii), for any
number n there is a complete set of n logically independent axioms for classical sentential
logic. Research on the deductive minima of classical sentential logic was conducted by
Sobociński mainly before the war (among other things, he found an axiomatic system
for the implication-conjunction fragment of classical sentential logic). In his doctoral
thesis, he examined selected n-valued logics with n − 1 distinguished values, for which he
developed axiomatic systems. His post-doctoral thesis concerned Leśniewski’s sentential
system—protothetics, which he continued to study also after the war. He analysed its
systems in the 1960s, identifying different possible sets of axioms (he found the shortest,
sole axiom of protothetics with the only original symbol of equivalence). His results
17 Sobociński did not participate in this meeting and we do not know to what extent he was involved in its
organisation.
are considered to be crucial technical improvements of this theory. He also improved,

axiomatised, and extended ontology and mereology. Thanks to him, Leśniewski’s theories
became known abroad. He also redefined other concepts by his master: he recreated
Leśniewski’s proposal for the solution of Russell’s antinomy and promoted Leśniewski’s
ideas concerning the methodological correctness of axiomatic systems of deductive
theories. Similarly to Leśniewski, he preferred classical logic to many-valued logics. He
inspired many students with Leśniewski’s ideas.
After the war, Sobociński became interested in modal logics, and achieved significant
results in this field as well. He wrote almost 20 papers about various issues in sentential
modalities. One of his first results was a proof that of modal systems T of R. Feys and
M of G.H. von Wright are deductively equivalent. He also showed the same equivalence
between systems M and M of von Wright with S4 and S5 of C.I. Lewis. He analysed
several axiom systems of Lewis and Feys. He established that system T has infinitely
many non-equivalent modalities. He introduced a division of modal logics into regular
(i.e. where the elimination of the modal symbols from any thesis yields a classical
tautology), and irregular. He also proved that each of Lewis’s modal systems containing T
must be regular, and that there exist irregular extensions of system T o of Feys-von Wright.
He defined a certain family of modal irregular logics which are sublogics of system S5.
With respect to the last group of issues mentioned by Lejewski, we may notice that
Sobociński was particularly interested in deductive dependencies between some key
theorems of set theory. He researched logical relationships between selected formulae
concerning cardinal numbers and the axiom of choice, generalised continuum hypothesis,
and Cantor’s hypothesis on alephs. He became interested in the theory of lattices in the
early 1960s, and he maintained this interest—as in the case of the theory of algebras—
until the end of his activity. He wrote papers on diverse axiom systems for different types
of lattices and algebras as late as 1978 and 1979.
5 Sobociński Personality
We do not know whether Sobociński was a “mysterious” person, but he surely made
surprisingly contrasting impressions under different circumstances and on different
people. In a letter to Kazimierz Twardowski, Leśniewski described him as a “Very decent
and well-educated person”* [20]. In same statements by Bocheński, K. Tatarkiewicz, and
Lejewski, the reader will easily find the picture of a fragile man and slightly confused
scholar (“eccentric”) as well as a courageous and at the same time cautious WWII
combatant (as evidenced by the events in Pruszków).
Sobociński was tall and slim (thin), had a “soft” eastern accent (he also stammered and
had a lisp?) and had troubles with the English pronunciation (at least in the first period
of his stay in America). To Bocheński, he seemed somewhat very “candid” (Bocheński
described him as “naive”). Before the war he wore old-fashioned clothes, thus attracting
attention (even during the war, he did not change his style and used to wear bow tie, a
bowler hat, and an umbrella. . . ).18 He never had a driving licence. In his academic work,
he combined research creativity, exceptional memory, determination to popularise the
works of his master, efficiency in independent and original logical research, and care about
teaching. In his political activity he was an efficient and discrete organiser of war secret
services, as well as an industrious conspirator who bravely fought for independence.
Perhaps the multiple facets of Sobociński’s personality were the result of the times in
which he lived. It is true, however, that by combining them with natural talents, he gave
witness to two cardinal virtues: bravery and wisdom. Thanks to them, he has a lasting
place in the contemporary history of Poland, and has made a significant contribution to
contemporary logic.
6 Selected Publications by Bolesław Sobociński
A complete list of publications, excluding reviews written by Sobociński over many years
of his activity, is to be found in [19]. (Before 1939, Sobociński published 13 reviews,
mainly in Przeglad ˛ Filozoficzny, but also in Organon and Nauka Polska; between 1938
and 1956, 11 reviews were published in the Journal of Symbolic Logic [8, pp. 5–7].)
The list compiled by K. Tatarkiewicz was provided to him by Professor J. Woleński and
contains a collection of articles published outside of NDJFL. The list was completed
by Rickey [8]. Tatarkiewicz also enumerates all of Sobociński’s papers published in the
NDJFL, and several other articles published after the list prepared by Rickey.
We will not include Sobociński’s publications in NDJFL. A full list is already available
on the Leibniz—Zentrum für Informatik, Schloss Dagstuhl, Universität Trier: http://dblp.
uni-trier.de/pers/hd/s/Sobocinski:Boleslaw.
Since 1957, Sobociński published only in English. Almost all earlier, significant works
were translated or reconstructed in English. In line with [19], we present the articles in a
chronological order, and English versions are followed by original Polish titles.
(1932) Z badań nad teoria˛ dedukcji. [An Investigation on the Theory of Deduction]
Przeglad
˛ Filozoficzny 35, 171–193 (master’s degree thesis)
(1933) a. Informator o studjach i egzaminach z nauk filozoficznych na Uniwersytecie
Warszawskim. [A Guidebook About Studies and Examinations from Philo-
sophical Sciences at the Warsaw University] Warszawa
b. Polish original of: (1967) Successive Simplifications of the Axiom-System
of Leśniewski’s Ontology. In: McCall F. (ed.) Polish Logic 1920–1939,
pp. 188–200. Oxford (O kolejnych uproszczeniach aksjomatyki Ontologji
prof. St. Leśniewskiego. In: Ksi˛ega Pamiatkowa
˛ ku uczczeniu 15-lecia
pracy nauczycielskiej w Uniwersytecie Warszawskim prof. Tadeusza
Kotarbińskiego. Warszawa)
18 According to K. Tatarkiewicz: “[. . . ] Sobociński’s outfit drew attention. [. . . ] in the interwar period, he
would always wear a black suit with a waistcoat, a white shirt with a stiff collar, a black bow tie, and a
bowler hat. Even before 1939 such attire would be worn only for formal occasions, and was almost never
seen in the streets. But B. Sobociński started his journey around the country in this outfit—looking rather
sensational. . . In addition to a suitcase, he always held a black umbrella.”* [18, pp. 127–128].
(1935) Aksjomatyzacja implikacyjno-konjunkcyjnej teorii dedukcji. [Axiomatisation

of the Implication-Conjunctive Theory of Deduction] Przeglad ˛ Filozoficzny
38, 85–95
(1936) a. Polskie wydawnictwa filozoficzne w latach 1918–1936. [Polish Philosophical
Publications in the Years 1918-1936] Nowa Ksiażka˛ 3, 13–121
b. Tendencje rozwojowe współczesnej filozofii polskiej (Refleksje na marginesie
III Polskiego Zjazdu Filozoficznego. Kraków 24–27.IX.1936). [Development
Tendencies of Contemporary Polish Philosophy] Nowa Ksiażka ˛ 3, 433–437
c. Aksjomatyzacja pewnych wielowartościowych systemów teorji dedukcji. [Ax-
iomatisation of Certain Many-Valued Systems of the Theory of Deduction]
Roczniki Prac Naukowych Zrzeszenia Asystentów Uniwersytetu Józefa Pił-
sudskiego w Warszawie. T.1, 399–419 (doctoral thesis)
(1939) a. Habilitation in Polish reconstructed as: (1949) An Investigation on Proto-
thetics. Cahiers de l’Institut d’Études Polonaises en Belgique 5, V+44. In:
McCall F. (ed.) Polish Logic 1920–1939, pp. 201–206. Oxford (Z badań nad
prototetyka.
˛ Collectanea Logica 1, 171–177)
b. Polish text reconstructed as: (1953) Axiomatization of a Conjunctive-Negative
Calculus of Propositions. Journal of Computing Systems, 229–242 (Aksjo-
matyzacja konjunkcyjno-negacyjnej teorii dedukcji. Collectanea Logica 1,
179–193)
(1949) French text translated as: (1984) Leśniewski’s Analysis of Russell’s Paradox.
In: Srzednicki, J.T., Rickey, V.F. (eds.) Leśniewski’s Systems. Ontology and
Mereology, pp. 11–44. The Hague-Wrocław (L’analyse de l’antinomie russel-
lenne par Leśniewski I, II, III, IV. Methodos 1, 94–100, 200–228, 308–316; 2
237–257)
(1953) a. Axiomatization of Partial System of Three-Value Calculus of Propositions.
Journal of Computing Systems 1, 23–55
b. On a Universal Decision Element. Journal of Computing Systems 1, 71–80
c. Note on a Modal System of Feys-von Wright. Journal of Computing Systems
1, 171–178
(1954) Z badań nad aksjomatyka˛ prototetyki Stanisława Leśniewskiego. [An Investiga-
tion on the Axiomatisation of Prototetics by Stanisław Leśniewski] Yearbooks
of the Polish Society of Arts and Sciences Abroad, for years 1953/1954, 4,
18–20
(1955) a. Note on a Problem of Paul Bernays. Journal of Symbolic Logic 20, 109–114
b. Studies in Leśniewski’s Mereology. Yearbooks of the Polish Society of Arts and
Sciences Abroad, for years 1954/1955, 5, 34–65 (Reprinted in: Srzednicki, J.T.,
Rickey, V.F. (eds.) Leśniewski’s Systems. Ontology and Mereology, pp. 217–
227. The Hague-Wrocław)
(1956) a. On Well-constructed Axiom-Systems. Yearbooks of the Polish Society of Arts
and Sciences Abroad, for years 1955/1956, 6, 54–65
b. In Memoriam Jan Łukasiewicz (1878–1956). Philosophical Studies 6, 3–49
(1957) a. Jan Łukasiewicz. Yearbooks of the Polish Society of Arts and Sciences Abroad,
for years 1956/1957, 7
b. La Genesis de la Escuela Polaca de Logica. Revista Orient Europe 7, 25, 83–95
(1958) Jan Salamucha (1903–1944): A Biographical Note. The New Scholasticism 32,
327–333
(1995) Lata wojny. . . Listy do Józefa M. Bocheńskiego. [Years of War ... Letters to
Józef M. Bocheński] In: Markiewicz, B., Jadacki, J. (eds.) Próg istnienia.
Zdziesiatkowane
˛ pokolenie, pp. 123–128. Polskie Towarzystwo Filozoficzne,
Warsaw
Acknowledgements Notes by Professor Jacek Jadacki [7] (and [20]) were an important source of
information, and I wish to thank him for allowing me to use them. I wish to thank Professor V. Frederick
Rickey for bringing amendments to some of the bibliographic details, as well Marek J. Minakowski,
PhD, for helping me reconstruct the family connections of Bolesław Sobociński. The dates and historical
facts were verified, as far as possible, in the Warsaw archives. Data from the archives of the Notre Dame
University in South Bend are quoted from [18]. I am grateful to Mr Zbigniew Siemaszko for help in
arranging information on Sobociński’s political activity.
References
Archival Materials
1. Student files and a folder with the master’s degree thesis along with examination reports, Archives of
the Warsaw University, catalogue number RP21816, KEM 829
2. Employee records, Archives of the Warsaw University, catalogue number K996
3. Personal Files from the Ministry of Higher Education Institutions and Science, Archive of New Files,
catalogue number MSW, 553
4. Files from the Polish Institute of National Remembrance, catalogue numbers: IPN BU 0203/2411,
IPN Kr 074/216/1, IPN KR 074/216/2
5. Death records at the Archive of the Roman Catholic Diocese of Warszawa–Praga, entry on Wanda
Chawłowska, reference number 493/1934
6. Letters by B. Sobociński to J.M. Bocheński. Copies of manuscripts
7. Archival materials of Professor Jacek Jadacki, with information obtained from Henryk Hiż and Witold
Strawiński
8. Rickey, V.F.: Bibliography of Bolesław Sobociński, typescript from the collection of Professor J.
Woleński, donated by K. Tatarkiewicz to the Archives of the Warsaw University (appended to the
files [2]), p.12 (1981)
Others
9. Bocheński, J.: Wspomnienia [Memoirs], 2nd edn. Philed, Kraków (1994)

10. Lejewski, C.: Ś.P. Bolesław Sobociński. Znak, XXXVI, No (2–3), February–March, 351–352 Kraków
(1984)
11. Lejewski, C.: Powojenna działalność polskich logików na Zachodzie [Post-war activity of Polish
logicians in the West]. In: Strzałkowski, W. (ed.) Filozofia polska na obczyźnie [Polish philosophy
abroad], pp. 161–175. Polskie Towarzystwo Naukowe na Obczyźnie, Londyn (1987)
12. Łyczak, M., Porwolik, M., Świ˛etorzecka, K.: The Universe in Leśniewski’s Mereology: Some Com-
ments on Sobociński’s Reflections. Axioms 5(3), 23 (2016). https://doi.org/10.3390/axioms5030023
(open access http://www.mdpi.com/2075-1680/5/3/23)
13. Markiewicz, B., Jadacki, J. (eds.): Próg istnienia. Zdziesiatkowane
˛ pokolenie. [Threshold of existence.
Decimated generation] Polskie Towarzystwo Filozoficzne, Warszawa (1995) (includes: Sobociński,
B., Lata wojny. . . Listy do Józefa M. Bocheńskiego, 123–128)
14. Muszyński, W.J.: Duch młodych. Organizacja Polska i Obóz Narodowo-Radykalny w latach 1934–
1944. Od studenckiej rewolty do konspiracji niepodległościowej [The Spirit of the Young: The Polish
National Radical Camp in the Years 1934–1944. From Student Revolts to Conspiracy for Polish
Independence]. Instytut Pami˛eci Narodowej, Warsaw (2011)
15. Siemaszko, Z.S.: Narodowe Siły Zbrojne [National Armed Forces]. London (1982) (3rd edn., Warsaw
(1991))
16. Świ˛etorzecka, K., Porwolik, M.: O pewnym przesadzie
˛ dotyczacym
˛ Uniwersaliów. Uwagi do sformal-
izowanego przez Bolesława Sobocińskiego argumentu na rzecz tezy o nieistnieniu powszechników
podanego przez Stanisława Leśniewskiego [About a superstition concerning Universals. Comments
on Bolesław Sobociński’s formalised argument for the non-existence of Universals as presented by
Stanisław Leśniewski*]. In: Brożek, A., Chybińska, A., Grygianiec, M., Tkaczyk, M. (eds.) Myśli o
j˛ezyku, nauce i wartościach. Seria druga. Profesorowi Jackowi Juliuszowi Jadackiemu w 70. rocznic˛e
urodzin [Thoughts about language, science and values. Second series. To Professor Jacek Juliusz
Jadacki on the 70th anniversary of his birth]. Semper, Warszawa (2016), 252–263
17. Świ˛etorzecka, K., Porwolik, M.: Bolesław Sobociński on Universals. From Leśniewski’s Nominalism
to Meta-Conceptualism of Sobociński. In: Garrido, Á., Wybraniec-Skardowska, U. (eds.) The Lvov-
Warsaw School. Past and Present, Studies in Universal Logic. https://doi.org/10.1007/978-3-319-
65430-0_42
18. Tatarkiewicz, K.: Profesor Sobociński i kolega Bum [Professor Sobociński and colleague Bum].
Wiadomości Matematyczne Seria 2, XXXIV, 122–146 (1998)
19. Tatarkiewicz, K.: Logik i polityk (Bolesław Sobociński) [Logician and politician (Bolesław Sobo-
ciński)]. Materiały i Konferencje, Szczecin University, vol. 3, pp. 167–183 (1998)
20. Jadacki, J.: Stanisław Leśniewski: geniusz logiki [Stanisław Leśniewski: Genius of Logic]. Epigram,
Warsaw (2016)
K. Świ˛etorzecka ()
Bolesław Sobociński on Universals
Leśniewski’s Nominalism and Sobociński’s
Metaconceptualism
Kordula Świ˛etorzecka and Marek Porwolik
Abstract The present paper proposes a comparative analysis of two standpoints on

the existence and nature of universals hold by Stanisław Leśniewski and Bolesław
Sobociński. We consider first the nominalistic argumentation of Leśniewski formalized
by Sobocińki and described in the correspondence with J. M. Bocheński in 1956. Sobo-
ciński’s formalization revealed a fundamental pragmatic weakness of the reconstructed
argumentation which was also mentioned by Sobociński. He himself was aware of
the difficulties connected with an adequate interpretation of the crucial axiom, whose
acceptance Leśniewski imputed to supporters of all theories of universals. Finally, the
problem of the existence and nature of universals was elaborated by Sobociński also
in a separate typescript “Uwagi w sprawie powszechników” (Remarks on universals).
The view formulated by Sobociński comes from a combination of the methodology of
deductive systems and the conceptualist standpoint. From the philosophical perspec-
tive Sobociński’s idea is both interesting and original, but it remained unknown to
philosophers and logicians in general. For these reason we describe it and compare
it with Leśniewski’s approach. We use in this description epistemological notions of
R. Suszko. Our analysis enables to speak about universals in sense of Leśniewski, which
are described by some universal in sense of Sobociński.
Keywords Bolesław Sobociński · Stanisław Leśniewski · Universals · Nominalism ·

Conceptualism
Research was carried out within the project Logic, Concepts, and Communication (LogiCCom, IP-2014-
09-9378) financed by Croatian Science Foundation (HRZZ).

616 K. Świ˛etorzecka and M. Porwolik
The medieval problem of universals was undertaken by members of the Lvov-Warsaw

School.1 A number of discussions concerning the nature and existence of general
objects were conducted in the new, analytical style, sometimes utilizing formal methods.
The main opponent of the views supporting the existence of universals was Stanislaw
Leśniewski, one of the most committed nominalists of the school. What is interesting, his
argumentations influenced the discussion mainly thanks to the information provided by
his friend, Tadeusz Kotarbinski, who reported on Leśniewski’s claims in his writings.2
Regardless of these historical circumstances, it bears emphasizing that Leśniewski’s
theories constitute interesting material, which has already been formalized a few times.
One of these formalizations will be the subject of our attention in the present paper. The
author of the formalization to be discussed here, is Boleslaw Sobociński, a close co-
worker and promoter of Leśniewski’s achievements. Sobociński described his theory in
the correspondence with Father Bocheński in 1956, which was published in 2004 in [19].
In this frame he wanted to reconstruct as closely as possible Leśniewski’s argumentation
of 1927, concerning the non-existence of universals [9]. Indeed, it can be claimed that this
goal was achieved: Sobociński was a fine expert on all the systems created by Leśniewski,
including his ontology, which he chose as the formal frame for his reconstruction. He
was well acquainted with his master’s philosophical views. As an exceptionally capable
formalist, Sobociński helped Bocheński in some of his analyses, although he usually
preferred to remain only a logical consultant, who kept a number of the discussed
philosophical decisions in perspective.3 Thus, it can be assumed that his formalization
was actually meant to provide the addressee with the explanation of Leśniewski’s line
of thought. Independently of this direct goal, Sobociński’s formalization revealed a
fundamental pragmatic weakness of the reconstructed argumentation. Sobociński himself
was aware of the difficulties connected with the adequate interpretation of the crucial
axiom, whose acceptance Leśniewski imputed to supporters of all theories of universals.
In a letter to Bocheński he tried to sketch out the possibility to avoid the problem,
but he failed to develop his analysis. Finally, the problem of the existence and nature
of universals proved to be important enough for him to elaborate on his opinion in
the next letter to Bocheński, where he added an attachment titled “Uwagi w sprawie
powszechników” (Remarks on universals). The view formulated by Sobociński comes
from a combination of the methodology of deductive systems and the conceptualist
standpoint. From the philosophical perspective Sobociński’s idea is both interesting and
original, but it remained unknown to philosophers and logicians in general. As far as we
know, the reprint of Remarks in the original version was published only once in [19] and
1 The historical presentation of all the discussion is shown by M. Grygianiec in [5].

2 Such a role was played by the work “Sprawa istnienia przedmiotów idealnych” (The problem of the
existence of ideal objects) [6] and by fragments devoted to Leśniewski’s argumentations from “Elementy
teorii poznania, logiki formalnej i metodologii nauk” (Elements of cognitive theory, formal logic, and
methodology of science) [7], which elicited replies from R. Ingarden and K. Ajdukiewicz. A report on
this discussion can be found in [5, pp. 110–113].
3 Such intentions can be found in the text of Sobociński’s letters. R. Murawski names Sobociński “an ex-
pert on logical problems” who has collaborated with all members of the Cracow Circle (J. M. Bocheński,
J. Salamucha, F. Drewnowski). He participated in all of the meetings of the Circle, but he did not published
any works related to the philosophical topics considered by the Circle [11, pp. 360–361].
Bolesław Sobociński on Universals 617
it has not been subject to any analysis. For these reasons we have decided to present it in
comparison with Sobociński’s formalization of Leśniewski’s nominalist argumentation.
1 Leśniewski’s Nominalist Argumentation in Sobociński’s

Version
As it has already been noted, Sobociński formulated his formalization in 1956, but the
fragments of correspondence which contained it were published almost 50 years later. The
author of the publication, T. Waragai, received the manuscripts from Father Bocheński in
1982. However, in the literature we can find Sobociński’s formalization published earlier.
In 1997 it was reprinted by Woleński [20] and quoted from this source by Grygianiec
in 2000 [5]. Additionally, also Urbaniak [18] referred to Gryganiec’s presentation in his
own paraphrase of Leśniewski’s argumentation. Urbaniak used Sobociński’s main specific
axiom, but he did not follow the original structure of argumentation, and he also showed
the possibility of simplifying Leśniewski’s (and Sobociński’s) reasoning. In 2016 our
work was published [17], which we are going to refer to in the present paper.
Leśniewski expressed his claim that general objects do not exist several times, and he
modified his justification. The first argument was published in 1911, and its amended
version was published in 1913. In 1927 Leśniewski provided an argument based on
the idea of the general object, which was similar to the one he used in his previous
justifications, but this time his reasoning had a new, different structure. The source
analyzed by Sobociński was the text of 1927.
The argumentation of 1927 had a broader range than the previous two arguments
concerning the non-existence of general objects did (1911, 1913), and was supposed
to represent Leśniewski’s radical nominalist point of view. At that time, Leśniewski
believed that only particulars exist in opposition to any non-individual objects, which also
comprise “the so-called properties” and “the so-called relations” [9, p. 314]. He claimed
that whoever has “a tendency to argue that “if X is a general object in relation to objects
a, X is b, and Y is a, then Y must be b” ”, he must accept that “that sentence involves
the sentence “if there are at least two different a’s, there is no general object in relation to
objects a” ”. The whole reasoning was formulated as a sequence of numbered sentences
with a commentary on the inferential connections among them. Sobociński conducted a
formal reconstruction of this text on the basis of Leśniewski’s elementary ontology.
1.1 Original Text
Let us start with the presentation of manuscript reprinted in [19, pp. 211–214].
Sobociński uses two types of individual variables: X, Y, . . . ; a, b, . . . (symbol a is also
used as an individual constant); two-argument predicate ε (. . . is . . . ) in the sense assumed
in Leśniewski’s ontology; constant opα, which appears in such contexts as Xεopα(a) (X
is a general object in relation to objects a). We assume here standard notation of logical
connectives, quantifiers, and brackets.
He provides the following definitions:

Def=. ∀X, Y (X = Y ↔ XεY ∧ Y εX)
Def=. ∀X, Y (X = Y ↔ XεX ∧ Y εY ∧ ¬(X = Y ))
Defι. ∀X, Y (Xει(Y ) ↔ XεX ∧ X = Y )
(Y is identical with individual X)
Defδ. ∀X, Y (Xεδ(Y ) ↔ XεX ∧ X = Y )
(Y is different from individual X)
and writes two ontological theses used in the argumentation:

(T1) ∀X, Y (XεY → XεX)
(T2) ∀X¬(X = X)
Formalizations of the subsequent sentences from Leśniewski’s argumentation in

Sobociński’s version are following4:
The empirical assumption:
(Z1) ∃Z, Y (Zεa ∧ Y εa ∧ ¬(Z = Y )), a is an auxiliary constant
there are at least two different [objects] a [the antecedent of the implication proved by Leśniewski]
The “philosophical”5 assumption:

(A1) ∀X, Y, Z(Xεopα(a) ∧ XεZ ∧ Y εa → Y εZ)
1. if X is a general object in relation to objects a, X is Z and Y is a, then Y is Z
Now he formulates the following derivation:

(A2) ∀X, Z (Xεopα(a) ∧ X = Z ∧ Zεa → Z = Z)
2. if X is a general object in relation to objects a, X is different from Z, and Z is a, then Z is different
from Z
(A3) ∀X, Z, Y (Xεopα(a) ∧ X = Z ∧ Y εa → Y = Z)
3. if X is a general object in relation to objects a, X is identical with Z, and Y is a, then Y is identical
with Z
(A4) ∀X, Z (Xεopα(a) ∧ Zεa → X = Z)
4. X is a general object in relation to objects a, and Z is a, then X is identical with Z
(A5) ∀X, Z, Y (Xεopα(a) ∧ Zεa ∧ Y εa → Xεopα(a) ∧ X = Z ∧ Y εa)
5. if X is a general object in relation to objects a, Z is a, and Y is a, then (X is a general object in
relation to objects a, X is identical with Z, and Y is a)
(A6) ∀X, Z, Y (Xεopα(a) ∧ Zεa ∧ Y εa → Y = Z)
6. if X is a general object in relation to objects a, Z is a, and Y is a, then Y is identical with Z
(A7) ∀X¬(Xεopα(a))
7. there is no general object in relation to objects a [the consequent of the implication proved by
Leśniewski]
4 Under the subsequent symbolic expressions we quote Leśniewski’s original words from [9] (only
variables are changed).

5 Sobociński uses quotation marks here.
The proof is done in frame of natural deduction system. For auxiliary constant a it is
assumed that formulas A1 and Z1 are fulfilled, and as a result: A7 is also fulfilled.
Sobociński may have assumed rules that give the possibility to introduce or eliminate
auxiliary constants, which are equivalent to those used by Słupecki in [13, pp. 19–25].
Słupecki’s calculus fulfills the deduction theorem with restriction on the rule introducing
∀. Thus, we can say that Sobociński actually proved the following implication:
T LS. A1 ∧ Z1 → A7 (a is an individual variable)
or its general closure.
In fact, the subject of Leśniewski’s original argumentation is the implication (or its
general closure) formalized by T LS (or the general closure of T LS).6
Now, our task is to describe the theory which enables to express Sobociński’s
formalization.
1.2 The Formal Background of Sobociński’s Formalization
In the letter to Bocheński we read:

This formalization shows that Leśniewski’s argumentations from that period of time complied with
ontology rules, although the last one was constructed only in 1920.7 [19, p. 214]
We assume that the formal basis of the analyzed argumentation is Leśniewski’s

ontology, or actually, its small fragment, which, following A. Pietruszczak, we call little
ontology eO.
The vocabulary of eO comprises: individual variables: u, v, w, x, y, z, . . .; predicate
constant ε (is), logical connectives: ¬, ∧, ∨, →, ↔, ∀, ∃, and parentheses: ( , ). The terms
are individual variables. We construct the formulas in the standard way.
The axioms characterizing eO are:
(0) all theses of the classical propositional logic,
the formulas of schemas:
(1) ∀xA(x) → A(y/x)
(2) ∀x(A → B) → (A → ∀xB), where x ∈ F V (A)
(3) ∃xA ↔ ¬∀x¬A
and Leśniewski’s axiom for ε:
AO xεy ↔ ∃z(zεx) ∧ ∀z, u(zεx ∧ uεx → zεu) ∧ ∀z(zεx → zεy)
The primitive rules are MP: # A → B, A ⇒ # B; Gen: # A ⇒ # ∀xA.
In our presentation we take the definitions of =, = ι rewritten in the introduced
language and also
(⊂) x ⊂ y ↔ ∀z(zεx → zεy) (subsumption)
(∠) x∠y ↔ ∃zzεx ∧ ∀z(zεx → zεy) (strong subsumption)
6 The actual aim of the proof is discussed by K. Ajdukiewicz in his remarks on T. Kotarbiński’s report on
Leśniewski’s idea [1, p. 227]. If someone accepts A1 (as an axiom), then the argument justifies implication
Z1 → A7, and not the thesis of the non-existence of universals.
7 Sobociński might have confused the dates of the argumentations of 1913 and 1927.
We also use the following

Fact 1 In the eO theory and its extensions with the appropriate definitions from the
above-mentioned ones the following formulas are provable:
(i) xεx ∧ yεx → xεy
(ii) xεy ∧ yεz → xεz
(iii) xεy → xεx
(iv) xεx → (xεy ↔ x ⊂ y)
(v) x∠y ∧ yεz → xεz
The proofs for (i)–(iii) and (v) can be formulated in the same way as in [8, p. 33]: cf. T6,
T7, T14, for (v)—cf. T119 (p. 37); for (iv) cf. T13.2 in [13, p. 39].
The eO theory is weaker than Iwanuś’s elementary ontology and than a stronger
version of Leśniewski’s EO ontology, where it is assumed that the set of theses comprises
all substitutions of the equivalents of the definition schema:
() ∃z∀u(uεz ↔ uεx ∧ φ),
where x is any variable different from z and φ is any formula in which z is not free.
Schema allows us to introduce to EO definitions of name-forming functors and named
constants, constructed in the way Leśniewski wanted:
def-f ∀u(uεf (x1 . . . xn ) ↔ uεxi ∧ φf ), where xi {u, x1 , . . . , xn }
def-n ∀u(uεn ↔ uεu ∧ φn )
In def-f and def-n on both sides of ↔ there are the same free variables.
The relationships between eO, Iwanuś’s ontology, and EO are described, for instance,
in [12], and we use this source in our presentation.
We use eOgen to mark the conservative extension of eO, created through adding
one-argument function constant gen to the dictionary of eO, and through the respective
extension of the set of terms and the set of formulas. Symbol gen is a counterpart of opα
from Sobociński’s formalism.
Symbol T [F ] is used for any extension of theory T with formula F .
First, we consider the equivalent of “philosophical” assumption A1:
ALS. xεgen(y) ∧ xεu ∧ vεy → vεu
This implication follows directly from the substitution of def-f
∀u(uεgen(x) ↔ uεu ∧ ∀v(uεv ↔ ∀z(zεx → z εv))).8
So, eOgen [ALS] is consistent.
The crucial step in Leśniewski’s argumentation is showing that just with the use of
ontology theses and definitions the consequence of “philosophical” assumption A1 is
that “if X is a general object in relation to objects a, Z is a, and Y is a, then Y is identical
with Z”. That consequence is expressed in implication A6 and now it is written
NR. xεgen(y) ∧ zεy ∧ vεy → z = v.
ALS implies formula NR in eOgen with the definition of identity (=). In a similar way
to Sobociński, who uses T1, we use thesis (iii) from Fact 1, but later on we significantly
8 This formula is equivalent to the definition of universal assumed in Luschei’s formalization of

Leśniewski’s argument (cf. [10, p. 309]).
shorten Sobociński’s original version. We use Urbaniak’s idea, in which the key role is
played by thesis (i) [18, pp. 97–98]:
Fact 2 eOgen [=] # ALS → NR
Proof
1. xεgen(y) ∧ xεx ∧ vεy → vεx ALS
2. xεgen(y) ∧ xεx ∧ vεy → vεx ∧ xεx 1
3. xεgen(y) ∧ vεy → vεx ∧ xεx (iii), 2
4. xεgen(y) ∧ vεy → xεv (i), 3
5. xεgen(y) ∧ xεx ∧ zεy → zεx ALS
6. xεgen(y) ∧ vεy ∧ zεy → zεv (iii), (i), 4, 5
7. xεgen(y) ∧ vεy ∧ zεy → vεz as: 1–6
8. xεgen(y) ∧ vεy ∧ zεy → zεv ∧ vεz 6, 7
9. xεgen(y) ∧ vεy ∧ zεy → v = z 8, =
Next, we add the definition of the diversity of individuals (=) to eOgen , and following
Sobociński’s argumentation, we take into account the counterpart of implication Z1 →
A7:
RNP . zεy ∧ uεy ∧ z = u → ¬xεgen(y)
Using just the rules of classical logic we notice that eOgen [=, =] # NR → RNP and
therefore also the following:
Fact 3 eOgen [=, =] # ALS → RNP
By classical logic the implication ALS → RNP is an equivalent of T LS, proved by
Sobociński.
Sobociński comments on the exposition given by Leśniewski in the following way:
Leśniewski’s argumentation does not lead to the conclusion that universals as such do not
exist [. . . ]. It only states that the theory of universals in which assumption A1 is used [. . . ] is
contradictory. [19, p. 215]
We can say that the main thesis of the theory of universals can be formulated in the
following way:
T P . ∃x∃y(xεgen(y) ∧ ∃z∃u(zεy ∧ uεy ∧ z = u))
In the closure of our reconstruction we can notice that:
Fact 4 eOgen [=, =, ALS, T P ] is a contradictory theory.
(Directly from the Fact 3.)
1.3 Universals in Theories with ALS
Now, we want to consider the consequences of accepting ALS in elementary ontology,

and whether it can be assumed that any of these ideas corresponds to the attacked views
of the supporters of universals. As we will see, ALS leads to a sort of collapse: it comes
out that every universal representing any non-contradictory object is identical with it. In
order to show this effect, we add the definition of the contradictory object to eOgen as an
axiom:

( ) xε ↔ xεx ∧ ¬(xεx)
Now, we can notice that:

Observation 1a. eOgen [=, ⊂, ∠, ] # ¬(y ⊂ ) → (ALS → (xεgen(y) → x =
y))9
Proof
1. xεgen(y) ∧ xεx → ∀v(vεy → vεx) ALS
2. ¬(y ⊂ ) ∧ xεgen(y) ∧ xεx → ∀v(vεy → vεx) ∧ xεx ∧ ¬(y ⊂ ) 1
3. ¬(y ⊂ ) ∧ xεgen(y) ∧ xεx → ∀v(vεy → vεx) ∧ xεx ∧ ∃z(zεy∧¬zε ) ⊂, 2
4. ¬(y ⊂ ) ∧ xεgen(y) ∧ xεx → y∠x ∧ xεx ∠, 3
5. ¬(y ⊂ ) ∧ xεgen(y) ∧ xεx → yεx ∧ xεx (v), 4
6. ¬(y ⊂ ) → (xεgen(y) → x = y) (i), =, 5

Restriction ¬(y ⊂ ) is important because: eOgen [=, ⊂, ∠, ] # ALS → (xεgen(y)
→ x = y). It is sufficient to take into account any ε -structure < 2{a} , gen gen, ε >,
in which gen,gen ε are interpretations of symbols gen and ε respectively, where
ε = {< {a}, {a} > }, gen gen(∅) = gengen({a}) = {a}. In this case, for the valuation,
which gives {a}, ∅ to the variables x, y respectively, formula ALS is fulfilled because of
the falsity of its antecedent (< d, ∅ >∈ ε for every d ∈ 2{a} ), and it is not true that x = y
(< ∅, {a} >∈ ε ).
On the other hand, we can notice that our equivalent of the “philosophical” assumption
is derivable from the implication, where it is said that being a universal of any object y is
sufficient to be identical with it:
Observation 1b. eOgen [=] # (xεgen(y) → x = y) → ALS
Proof Let us assume that 1. xεgen(y) → x = y. From (ii) and (=) we get: 2. x = y →
(xεu → yεu), and from AO: 3. x = y → (xεu → ∀v(vεy → vεu)). From 1 and 3:
xεgen(y) ∧ xεu ∧ vεy → vεu.
Observations 1a and 1b allow to notice that:

Observation
2. Theories eOgen [=, ⊂, , ¬(y ⊂ ) → ALS] and eOgen [=, ⊂, ,
¬(y ⊂ ) → (xεgen(y) → x = y)] are deductively equivalent.
A supporter of the theory of universals, who would accept ALS in its limitation
to universals corresponding to nonempty objects, can equivalently accept a theory in
which being a universal of any nonempty object means being identical with it. The
problem is that for a supporter of universals, even for one who ascribes to universals
the same status of individual objects as to objects for which they are universals, that
latter claim is unacceptable and any dispute based on it (or equally on the basis of ALS
limited to nonempty objects) is simply futile. This line of defense against Leśniewski’s

9 Letus notice that although the thesis of EO is ¬∃x(xε ), also ∃x(x ⊂ ). The fact that x ⊂ is
only equivalent to the fact that: ¬∃z(zεx)—we can say that in this situation x is an empty object.
argumentation is more effective than trying to modify the definition of a universal, which
is criticized even by Sobociński himself:
I do not know if there is any chance of weakening any premises of assumption A1, which
would not lead to at least paradoxical conclusions. Adding the intuitive, as it appears, premise:
∀v(vεopα(a) → ¬(vεa)) (e.g. a universal of a cat is not a cat) results in a paradoxical thesis.
[19, p. 215]
In fact, the assumption considered by Sobociński even strengthens Leśniewski’s

argument. As it can be easily shown, the assumption
S. xεgen(y) → ¬xεy
immediately leads to

Observation 3. eOgen [=, ∠, , S] # ¬(y ⊂ ) → (ALS → ¬xεgen(y))
The direct proof from Observation 1a and S.
This time being a universal cannot be reduced to identity, but it is an empty attribute (!).
The solution that does not fall into the described extremes and that accepts the
existence of objects that are described in T P could consist of weakening ALS in such a
way that we would limit attributes that are vested to universal x: x can have any attributes
except being x itself.10 This idea seems to be effective. If we assume:
ALS ∗ . xεgen(y) → (¬uεx ∧ xεu ∧ vεy → vεu)
instead of ALS, then we do not get T LS, which leads to a contradiction with T P . In fact:
Observation 4. eOgen [=, =] # ALS ∗ → RNP
Proof Let us remember: ALS ∗ → RNP has the form:
(xεgen(y) → (¬uεx ∧ xεu ∧ vεy → vεu)) → (zεy ∧ v εy ∧ z = v → ¬xεgen(y)).
We can consider ε -structure < 2{a,b} , gen gen, ε > (where: a = b), where gen gen, ε are
interpretations of functors gen and ε respectively, such that: ε = {< {a}, {a} >,
< {b}, {b} >, < {a}, {a, b} >, < {b}, {a, b} >}, gen gen(∅) = gengen({a}) = gen
gen({b}) =
gen({a, b}) = {a, b}. For the variables: x, y, z, v we take the valuation that ascribes
∅, gen
to them respectively: {a}, {a, b}, {a}, {b}. Then: (1) for every u, v-variant of our valuation
ALS ∗ is true; (2) zεy ∧ v εy ∧ z = v is true because < {a}, {a, b} >, < {b}, {a, b} >∈ ε
and (3) ¬xεgen(y) is false because gen gen({a, b}) = {a, b}.
However, we can say that this improvement of the idea of universals by means of
weakening axiom ALS to ALS ∗ would not be satisfactory for numerous supporters
of universals, including the author and the addressee of the letter we are studying.
Bocheński, a moderate realist when it comes to universals, wrote about his point of view
in the following way “Nominalists claim that in objects there are no universals, and that
only general words exist; followers of Plato claim that universals exist in objects. Well,
following Aristotle, I reject both views. I deny the existence of universals in objects, but I
think that in those objects there is something real that allows us to create general notions”
[4, p. 224]. Nevertheless, in his views, general notions are supposed to be “the content of
properties” [2, p. 146], and those, in line with the Aristotelian-Thomistic thought, belong
to a different ontic category than individuals. Bocheński analyzed the idea of universals
twice in [2] and [3]. He expressed his formalizations in the language with two types of
10 We owe this conception to M. Łyczak.

variables, and in this way he respected Ajdukiewicz’s main objection to the nominalist
view of universals shown in Kotarbiński’s (and Leśniewski’s) version. As Ajdukiewicz
argues, the linking word “is” is used by Aristotle as a functor of a individual name and
a predicative name, or of two predicative names, and the latter use, which is necessary
to express the Aristotelian definition of a universal, is absent from Leśniewski’s ontology
[1, pp. 227–228]. Actually, elementary ontology is a theory of objects belonging to one
category, its language contains only one type of variables, and symbol gen (opα) creates
terms belonging to the same category as individual variables. In a language constructed in
this way, universals, if they exist, can only belong to the same category as individuals, for
which they are universals, so they can only be something similar to Plato’s ideas. Thus,
if ontology is supposed to be a frame for discussion in the case of universals, then just
because of its language the argument may concern some versions of Platonism [10, p. 23].
Sobociński’s idea differs from the analyzed formalization in another way, and this is now
the point of our interest.
2 Sobociński’s Metaconceptualism
In a letter to Bocheński, Sobociński writes the following words:

[. . . ] the problem of universals does not belong to the area of logic (and to the deductive
mathematical systems in general). This means that a system of logic (and mathematics), in whose
assumptions there are these or other concepts concerning universals is not a scientific system.
However, this does not mean that the problems of universals do not belong to the so-called science.
[19, p. 209]
Then, he adds the following commentary on Leśniewski’s reasoning:

I claim that the problem of universals can be positively solved on a completely different ground.
Namely: a universal is a deductive theory describing objects belonging to a particular class in an
adequate way (but not in a specific way). [19, p. 215]
Actually Sobociński developed this idea in his Remarks added to the correspondence
with Bocheński. In the present paper we describe this approach, complemented with a
few our original comments. In our sketch we use elements of the semantic description
of cognitive processes suggested by R. Suszko from [16]. We set together these two
conceptions motivated by their concurrence.
2.1 The Original Text of Remarks
Sobociński divided the exposition of his conception into three parts without titles. In
the first part, he presented his intuitions concerning the distinguished components of
the cognitive process, which lead to knowledge expressed in certain types of axiomatic
systems. He considered the core of each system of this type, that is a set of specified
axioms, to be a universal (although sometimes the term “universal” was used by him to
name the whole deductive system founded on such a set of axioms). This metatheoretical
approach was explained in the second part of the analyzed text. Here the author
presented also his own views on the existence of universals, which he described as
being conceptualistic and he proposed a specific way of understanding both Platonic
and Aristotelian realism. In the third part he provided some negative remarks: he gave
examples of theories which, thanks to sufficient formal means, allow us to recreate the
definitions of the original ideas determined in other systems in the axiomatic way. In this
situation, the recreated idea cannot be a candidate to be a universal in the sense suggested
by Sobociński.
Sobociński begins his consideration with the following view of the pre-theoretical
phase of the cognitive process:
The human mind only has the ability to perceive sensory data, as well as the ability to abstract and
reason, which is its characteristics. Thus, our view of reality is determined by the properties of our
senses and the way we reason. [19, p. 220]
The process of abstraction leads us to “distinguishing in our mind a distributive class

[. . . ] of objects that are of interest to us at the moment”11 The distinguished classes can
be homogeneous or heterogeneous. An example of the former type is a class of all people;
the latter can be exemplified by a class of people and natural numbers. When we provide
an axiomatic description for a homogeneous class, such a description can be considered
as a universal:
[. . . ] the definition created through the postulates (with the assumption that it is intrinsically non-
contradictory) of a class of some homogenous objects, and constructed by our mind from a specific
point of view, can be, it appears to me, treated as a “universal” of these objects for the sake of our
minds, so, as I think, it is an approach similar to conceptualism. [19, p. 223]
Thanks to the development of the epistemic subject’s knowledge, universals can

undergo changes:
The axiomatic system, which is a more or less appropriate definition through postulates, because
of its nature must be a system that deductively is highly complicated. Obviously, in most cases we
only create a very imprecise and imperfect system of this kind. Improving our knowledge about
objects of a given class means improving this system in us. [19, p. 223]
That improving is possible within the limits of non-contradiction: “Our only limitation
in this respect is the intrinsic contradiction in the system”, although the created systems
(and the universals that generate them) can be (extrinsically) contradictory with one
another, but at least they ought not to be subsystems of one another: “The distributive
classes described in such a way are systems standing next to one another, but included in
one another”. Sobociński connects explications from Platonic realism with the problem
of what else, apart from axioms, constitutes a component of our knowledge, resulting
from the formulated deductive system. The full knowledge of this type is possessed only
by God:
If we had a precise and thorough knowledge concerning a specific deductive system which
determines a class of some homogenous objects, then we had a precise description of those objects.
Such a knowledge may be possessed only by the infinite mind, that is God, and I think that in this
sense universals exist as ideal objects (Platonic realism). [19, p. 223]
11 In the further part of the text he will add that the effect of abstraction depends also on “our knowledge
and other external circumstances”.

Some universals can be treated as corresponding to the theory called by Sobociński

“Aristotelian realism”. Whether a given universal is a universal understood in accordance
with Aristotelian realism depends on the way individuals characterized by the universal
exist. Although in general creating universals is not connected with the (real) existence
of objects that they are supposed to refer to, certain universals cannot be universals in
the sense of Aristotelian realism because their realization contradicts the structure of the
universe planned by God:
[. . . ] it is easy for us to create in this way [that is, through determining non-contradictory
axioms for a given set] concepts of objects that not only do not exist in reality, but that cannot
even be realized, for instance, the concept of “Greek gods”. Thus, I claim that we can even
possess such a concept, but it is determined by such assumptions that are too weak to cause any
intrinsic contradiction in the system. However, if we modify such a system by adding assumptions
concerning the existence of God, then there is an intrinsic contradiction. In other words, in any
concept of “Greek gods” the term “god” is used with so weak a meaning that it does not cause
any contradiction in the system. Concepts of this kind cannot be realized (in my understanding of
Aristotelian realism) because they might cause a contradiction in the universe. Concepts that are
less roughly thought could be realized, as I think, if God wanted it and if they did not clash with
his plan of the world, in which we dwell. [19, p. 224]
A system, whose axioms constitute a universal “must possess its own deductive basis,
that is a system of logic”, which is “an integral part of a given system”. The choice of a
particular logic depends on “the will and the starting point” of the epistemic subject.
Comparing universals must be done on the basis of one logic, which may differ from
logics that function as the formal ground for these universals. Such an external logic—
a general logic—ought to be “extensional, saturated with functors because of sentential
functors,12 and it should be compatible with properties of colloquial language. These
assumptions mean that the logic must be the two-valued logic”. General logic ought not
to contain any existential assertions, and the assertions concerning the logic itself can only
have the relative character (as it happens in the case of the set theory).
In the last part of Remarks Sobociński provides examples, which show that for him
a universal is the way of describing certain concepts, and not concepts themselves. Real
numbers can be defined on the ground of the theory of natural numbers (with the help
of Cauchy sequences or Dedekind cuts), but this definition is not a universal. In this
situation real numbers are logical creations, which differ from the categorial point of
view from natural numbers, which are identified with the use of an object of the lowest
rank, representing a universal, a natural number. The construction of a universal of a real
number means the determination of the axiom of a deductive theory, in which axioms are
true in the set of real numbers:
I claim that the right idea of one or another arithmetic is based on its definition through postulates,
among which there must be assumptions of one or another logic. This means that the concepts
of natural numbers are given to us by logic and, for instance, Peano axioms. In a similar way
we receive concepts of rational numbers, real numbers, complex numbers, etc. as well as other
so-called mathematical objects. [19, p. 225]
12 For any sentential functor, an expression created by this functor and any sentence, should also have a
logical value.
Finally, Sobociński presents also his conceptualistic view on the existence of mathe-
matical objects, which corresponds with his theory of universals:
Thus, do numbers and other mathematical objects exist? Undoubtedly, they exist as concepts in our
minds. In people who are not advanced at mathematics these concepts are imperfect [because they
are described by relatively weak universals], but in people who are experienced in mathematics,
they are much better developed [because their universals have been “improved”]. Do they exist in
reality (in the sense of moderate realism)? They do not exist as real objects. However, in concepts
(universals) of real objects, which, in line with the approach presented here, are deductive systems
there is one or another mathematical theory. Thus, objects realized in accordance with a universal
understood in this way fulfill, among others, the rules of this theory. For that reason we may use
assertions belonging to this theory when we examine and describe these objects. [19, p. 225]
2.2 Sobociński’s Universals in the Frame of Suszko’s Epistemic

Approach
Sobociński expressed his conceptualist standpoint in the frame of the metatheory of de-
ductive systems and for that reason we call it “metaconceptualism”. Obviously, this notion
exceeds metalogic, because of the fact that some of its important fundamental assumptions
concern human activities that are considered still in epistemology. The description of the
pre-theoretical cognitive processes, whose results are deductive theories, was undertaken
also by Suszko in [16] and this description is now used. Suszko meant his epistemological
analysis to be an introduction to research into diachronic logic. That logic was supposed
to help in the description of the changeability of knowledge that involves improving the
conceptual apparatus associated with a chosen model or improving knowledge (with the
use of a given conceptual apparatus).
Suszko’s approach and Sobociński’s theory turn out to be very similar when it comes
to certain observations and solutions. For that reason we use distinctions suggested by
Suszko in our explanations of Sobociński’s conception.13 Our proposal, just like Suszko’s
epistemological analysis, is only a partially formalized sketch, which we formulate in
order to encourage our readers to develop further research, and which we do not consider
to be a ready interpretation of Sobociński’s views.
Suszko starts his reflection with combining certain a conception of cognitive acts and
the model-theoretic view on the interpretation of languages and theories, which are results
of those acts.14 He restricts his investigations to theories expressed in standard formal
languages, that is, such languages that can be described with Ajdukiewicz’s method of
semantic categories. In line with Suszko’s conception, for every language it is possible to
assign a family of models founded on any non-empty set of individuals, that differ among
13 In 1946, only a few days before Sobociński left Poland forever, the young Suszko visited him in Warsaw.
Many years later Suszko told Professor Mieczysław Omyła about that only meeting with Sobociński,
saying that they discussed logic. We do not know what problems they discussed during that meeting.
Although the similarities between the approaches described by us may be accidental, to us it seems to be
interesting also in the context of that meeting.
14 Suszko was involved in his research from the early 1950s. The text of 1966 [16] is a summary of this
research. Technical details assumed in [16] which are assumed also here, are elaborated in [15].
one another when it view of their characterization, which contains properties and relations
occurring between these individuals—the so-called aspects of individuals. Equivalently,
having in mind a specific model, we may link with it a family of symbolic languages
that can be used to describe that model. The language that is interpreted in our model in
accordance with the certain circumstances of the sense of that language constitutes the
conceptual apparatus for the selected model, and that model is called “the proper model”
in those circumstances.
Developing the conceptual apparatus for the proper model is the primary cognitive
act: “its course is obviously extremely complicated. [. . . ] conducting the development of
the conceptual apparatus for model M in a given group of people would not be possible
unless that group of people were involved in a sort of direct contact with model M as part
of their activity (sensory perception, practical activity).” [16, pp. 529–530]. Following
Sobociński we would say that the ability to perceive sensory data precedes the ability of
abstraction and this results in distinguishing the set of individuals, which we are going
to describe in the frame of a formed deductive system. The effect of abstraction, about
which Sobociński speaks, would be named by Suszko “the characterization of a given set
of considered individuals” which constitute a given model.
Let us consider Sobociński’s theory with reference to the systems expressed in
such languages that are described by Suszko. The parameter of circumstances, which,
according to Suszko, involves the social group that uses a particular language and the
totality of activities performed by the members of that group, will be replaced with the
relativization of the proper model to the defined epistemic subject x, who has a knowledge,
which determines state i.
Following Suszko, we accept that
Definition 1 A model considered by x in i is an ordered pair:
Mxi =< U, Cxi >
where U = ∅ is a set of individuals and Cxi = {c1 , . . . , cn } is a characterization of

individuals from U considered by x in i.
Every model Mxi is an effect of an abstraction undertaken by x.
Depending on a certain state of knowledge, x may investigate different characteriza-
tions.
All models founded on a given U , which may be considered by x form epistemic space
of x:
Definition 2 Epistemic space of x is Mx = {U } × {Cxi }i∈I .
(Perhaps, if x is a human being the family {Cxi }i∈I should be finite.)
Let us now consider some given element of Mx : Mx =< U, Cx > with Cx =
{c1 , . . . , cm } which is determined by the given stage of conceptual knowledge of x.
Mx determines a certain type of symbolic languages adequate to describe Mx which
we call L(Mx ). Each of these languages contains specific terms (non-logical constants)
which correspond to elements of Cx . Languages from L(Mx ) do not have more specific
terms as is needed to describe elements of Cx .
We choose certain language L ∈ L(Mx ) as the conceptual apparatus used by x to
describe model Mx . We call Mx a proper model of L. Now, epistemic subject x chooses
certain logic L expressed in language L. According to Sobociński, L may be non classical

logic. We use the notation L[A] to speak about extension of L by the set of formulas
A ⊆ L. The set of all sentences of L true in model Mx we name like Suszko V er(L, Mx ).
Extensions of L are theories based on L. The act of forming a theory based on L is
named by Suszko as “the secondary cognitive act” [16, p. 530]. At least some sets of
formulas which may be added to L are considered by Sobociński as universals.
We take the set of all L-consistent theories which are adequate to Mx :
Definition 3 T(Mx ) = {T : ∃A T = L[A] and T ⊆ V er(L, Mx )}
We use primitive predicate GEN(x, A) to be read: set (of formulas) A is an universal
for x.
Let us propose the following few axioms characterizing universals in Sobociński’s
sense:
(A1) GEN(A, x) ⇒ L[A] ∈ T(Mx ) (every universal is an axiomatization of some theory
adequate for the proper model)
(A2) ∃A (GEN(A, x)) (conceptualism)
(A3) GEN(A, x) ⇒ A = ∅ (universals are not empty)
(A4) GEN(A, x) ⇒ A ∩ L = ∅ (universals do not belong to logic)
In order to express Sobociński’s belief, which he calls “Platonic realism”, we take the
abbreviation Gx to be read x is God.
Following Sobociński we say that God knows all consequences of any his universal:
(A5) Gx ⇒ (GEN(A, x) ⇒ A = L[A] − L)
Concerning the question of improving of universals we would say only that:
(A6) GEN(A, x) and A ⊆ B and B ⊆ V er(L, Mx ) and B ∩ L = ∅ ⇒ GEN(B, x)
Let us note that the development of a given deductive system in order to improve a
universal is connected only with adding axioms to a new theory, but not with enriching
the conceptual apparatus. Changing the conceptual apparatus causes the need to redefine
a selected model that we aim at and selecting a respectively changed language.
Our axioms allow us to formulate at least a few dependencies which we find in
Sobociński’s text:
(t1) GEN(A, x) ⇒ A is L-consistent (A1)
(universals are non-contradictory with their logic)
(t2) not (GEN(L, x)) (A4)
(logic is not a universal)
(t3) GEN(A, x) and A ⊆ B ⇒ B = L (A4)
(improving of universals is done outside of logic)
(t4) GEN(A, x) and GEN(B, x) ⇒ GEN(A ∪ B, x) (A1, A6, Def3)
(the sum of universals is also a universal)
(t5) GEN(A, x) and GEN(B − A, x) ⇒ GEN(B, x) (A1, A6, Def3)
(if B-complement of universal A is a universal, then B is also an universal)
We hope that Sobociński would accept the above considerations as a kind of sketch of
a general theory of universals (obviously, requiring further complementation). Sobociński
might expect us to include the conditions of adequacy and independence of axioms

constituting each universal along with the condition of non-contradiction.15 Sobociński
lists these requirements, following Leśniewski, in [14] as the basis for all well-constructed
axiom systems. Axiom systems ought to fulfill also other conditions, which were classified
by Sobociński into the following groups: conditions concerning the primitive terms of
axiom theories, specific conditions concerning groups of axioms from any theories and
conditions that generate Leśniewski’s systems or theories based on them. All additional
requirements in his views result from these basic ones and their nature is rather aesthetic.
We believe that a contemporary analysis of them might show that their nature is rather
pragmatic, connected with the pragmatic criteria of choosing axioms for a particular
theory. This question remains open.
The conditions of well-constructed axiom systems grouped by Sobociński only have
the necessary character. If universals are supposed to be well-constructed axiom systems,
these conditions only have such a meaning for them. What then decides that a given axiom
system is a universal in Sobociński’s understanding?
Let us analyze two of the already described theories.
In line with Fact 4, set {=, =, ALS, T P } is non-consistent on the ground of eOgen and
as such it cannot be a universal for a given epistemic subject, whose cognitive apparatus
is the language of system eOgen , chosen by that subject to describe any model from his
epistemic space, which is suitable for being an interpretation of that language. Next, set
{ALS} is consistent on the ground of eOgen and it fulfills all the conditions of well-
constructed axiom systems for theory eOgen [ALS]. Is {ALS} that determines the notion
of a universal in Leśniewski’s understanding suitable to be a universal in Sobociński’s
understanding? Let us take into account epistemic subject S that uses system eOgen
enriched with equivalences =, ⊂, ∠, and ι, which he treats as theses of the logical
nature. In line with Observation 1a, on the basis of a selected formal ground, subject S
will be forced to accept that the notion of a universal in Leśniewski’s understanding is
essentially weaker than identityin the sense of ι, because in his theory ALS implies that
xεgen(y) → xει(y), if ¬(y ⊂ ). Next, let us assume that the right model for S is MS∗ —
ordered pair consisted of a non-empty set U, founded on some set of objects I nd: U ⊆
2I nd (elements of set U are individuals in our model) and characteristic CS∗ , containing at
relation ε , a semantic correlate of predicate ε, ∅—the semantic correlate of constant
least:
gen, ι : U → U, which are counterparts of constants
, and functions: gen gen and ι. In
line with what is determined by the axioms of system eOgen [=, ⊂, ∠, , ι] extended by
ALS we notice that the function of being a universal in Leśniewski’s understanding, that
is gen assigns the following: (a) to every singleton the same singleton, (b) to every set
bigger than a singleton it assigns ∅, and (c) to ∅ it assigns ∅ or any singleton. Only the
second version of case (c) distinguishes function gen from identity ι (!), which assigns
the same singleton to every singleton, and ∅ to every non-singleton. It’s hard to say if S
would accept {ALS} as a universal in Sobociński’s understanding in this situation. It can
15 The notion of the adequacy of axioms was distinguished by Sobociński from the notions of the adequacy
and completeness of theories. We assume the following simplification of his explanation: (1) a set of
formulas expressed in a language of a given theory is adequate when the set of all the theses of a given
theory can be deduced from that set of formulas (cf. [14, pp. 56–57]); (2) independence is understood as
such a property of a set of axioms that none of its elements can be deduced on the grounds of a specific
theory from other axioms.
be said that there are no formal obstacles to do that, but it is hard to accept that being
a universal in Leśniewski’s understanding, which corresponds to function gen from CS∗
constitutes an interesting (not only from the philosophical point of view) aspect of objects
from set U.
Acknowledgements We would like to express our gratitude to Professor M. Omyła for his advice
concerning R. Suszko’s epistemological analysis, and to M. Łyczak for valuable seminar discussions.
Fragments of Sobociński’s and Suszko’s texts were translated by Marcin Kuczok.
References
1. Ajdukiewicz, K.: W sprawie uniwersaliów. [On universals] Przeglad ˛ Filozoficzny 37, 219–234 (1934)
2. Bocheński, J.M.: Powszechniki jako treści cech w filozofii św. Tomasza z Akwinu. [Universals as the
content of properties in st. Thomas Aquinas’ philosophy] Przeglad ˛ Filozoficzny 41, 136–149 (1938)
3. Bocheński, J.M.: The Problem of Universals. In: Bocheński, J.M., Church, A., Goodman, N. (eds.)
The Problem of Universals, pp. 33–54. Notre Dame Press, Notre Dame (1956) (Polish translation:
Zagadnienie powszechników. In: Parys, J. (ed.) Logika i filozofia. Wybór pism, pp. 79–105. PWN,
Warszawa (1993). Trans. T. Baszniak)
4. Bocheński, J.M.: Wspomnienia. [Recollections] Philed, Kraków (1994)
5. Grygianiec, M.: Leśniewski przeciw powszechnikom. [Leśniewski against universals] Filozofia Nauki
3/4, 109–125 (2000)
6. Kotarbiński, T.: Sprawa istnienia przedmiotów idealnych. [The Problem of the Existence of Ideal
Objects] In: Ksi˛ega pamiatkowa
˛ ku uczczeniu 25-letniej działalności nauczycielskiej na katedrze filo-
zofii w Uniwersytecie Lwowskim Kazimierza Twardowskiego, pp. 149–170. Uniwersytet Lwowski,
Lwów (1921)
7. Kotarbiński, T.: Elementy teorii poznania, logiki formalnej i metodologii nauk. [Elements of Epis-
temology, Formal Logic, and Methodology of Sciences] Ossolineum, Wrocław-Warszawa-Kraków
(1961)
8. Leśniewski, S.: Definitions and theses of Leśniewski’s Ontology. In: Srzednicki, J.T.J., Stachniak,
Z. (eds.) S. Leśniewski’s Lecture Notes in Logic, pp. 29–58. Kluwer Academic Publishers, Dor-
drecht/Boston/London (1988)
9. Leśniewski, S.: O podstawach matematyki [On the foundation of mathematics]. Przeglad ˛ Filozoficzny
30, 164–206 (1927); In: Jadacki, J. (ed.) Leśniewski. Pisma zebrane [Leśniewski. Collected Works],
vols. 1 and 2, pp. 295–337. Semper, Warszawa (2015)
10. Luschei, E.C.: The Logical Systems of Leśniewski. North-Holland Publishing Company, Amsterdam
(1962)
11. Murawski, R.: Cracow circle and its philosophy of logic and mathematics. Axiomates 25, 359–376
(2015)
12. Pietruszczak, A.: O teoriach pierwszego rz˛edu zwiazanych
˛ z elementarnym fragmentem ontologii
Leśniewskiego. [On first order theories connected to Leśniewski’s fundamental fragment of ontology]
In: Perzanowski, J., Pietruszczak, A. (eds.) Logika and Filozofia Logiczna. [Logic and Logical
Philosophy] FLFL 1996–1998, pp. 127–168. Wydawnictwo UMK, Toruń (2000)
13. Słupecki, J.: S. Leśniewski’s Calculus of Names. Stud. Logica 3(1), 7–73 (1955)
14. Sobociński, B.: On well-constructed axiom-systems. Rocznik Polskiego Towarzystwa Naukowego na
Obczyźnie 6, 54–65 (1955/1956)
15. Suszko, R.: Syntactic structure and semantical reference I. and II. Stud. Logica VIII, 213–144 (1958),
IX, 63–93 (1960)
16. Suszko, R.: Logika formalna a niektóre zagadnienia teorii poznania. [Formal logic and some aspects
of epistemology] In: Pawłowski, T. (ed.) Logiczna teoria nauki. Wybór artykułów. [Logical Theory
of Science. A Selection of Articles] PWN, Warszawa, 505–576 (1966)
17. Świ˛etorzecka, K., Porwolik, M.: O pewnym przesadzie
˛ dotyczacym
˛ uniwersaliów. Uwagi do sformal-
izowanego przez Bolesława Sobocińskiego argumentu na rzecz tezy o nieistnieniu powszechników
podanego przez Stanisława Leśniewskiego [About a superstition concerning universals. Comments

on Bolesław Sobociński’s formalised argument for the non-existence of universals as presented by
Stanisław Leśniewski]. In: Brożek, A., Chybińska, A., Grygianiec, M., Tkaczyk, M. (eds.) Myśli o
j˛ezyku, nauce i wartościach. Seria druga. Profesorowi Jackowi Juliuszowi Jadackiemu w 70. rocznic˛e
urodzin [Thoughts about Language, Science and Values. Second series. To Professor Jacek Juliusz
Jadacki on the 70th Anniversary of His Birth]. Semper, Warszawa 252–263 (2016)
18. Urbaniak, R.: Leśniewski’s Systems of Logic and Foundations of Mathematics. Springer, Cham
(2014)
19. Waragai, T.: Letters and typescripts send to Bocheński from Sobociński: Historical Documents about
Leśniewski’s refutation of universals. Reports of the Keio Institute of Cultural and Linguistic Studies
36, 203–228 (2004)
20. Woleński, J.: Szkoła Lwowsko-Warszawska w polemikach. [The Lvov-Warsaw School in Polemics]
Scholar, Warszawa (1997)
K. Świ˛etorzecka • M. Porwolik ()

Many-Valued Logics in the Iberian Peninsula
Angel Garrido
To Prof. Dr. Solomon Marcus (1925–2016), who recently passed away.
Abstract The roots of the Lvov-Warsaw School (LWS, by acronym) can be traced back
to Aristotle himself. But in later times we better put them into thinking GW Leibniz and
who somehow inherited many of these ways of thinking, such as the philosopher and
mathematician Bernhard Bolzano. Since he would pass the key figure of Franz Brentano,
who had as one of his disciples to Kazimierz Twardowski, which starts with the brilliant
Polish school of mathematics and philosophy dealt with. Among them, one of the most
interesting thinkers must be Jan Lukasiewicz, the father of many-valued logic.
Jan Łukasiewicz (1878–1956) began teaching at the University of Lvov (now Lwiw;
former Lemberg, but also Leópolis), and then at Warsaw, but after World War II must to
continue in Dublin. Some questions may be very astonishing in the CV of Łukasiewicz.
For instance, that a firstly Polish Minister of Education in Paderewski cabinet, into the new
Polish Republic, and also Rector for two times at Warsaw University, was awarded with
a Doctorate ‘Honoris Causa’ in spring 1936, at University of Münster, into the maximum
of effervescency of Nazism in Germany. The explanation must be their good relation with
a very good friend, the former theologian, and then logician, Heinrich Schölz, who was
the first Chairman of Mathematical Logic in German universities.
Łukasiewicz firstly studied Law, and then Mathematics and Philosophy in Lvov (then
Lemberg). His doctoral supervisor was Kazimierz Twardowski, and in 1902 he obtain
his Ph. D. title with a very special mention: ‘sub auspiciis Imperatoris’ (i.e., under the
auspices of the Kaiser). Also he received a doctorate ring with diamonds from the Kaiser
of the Austro-Hungarian Empire, Franz Joseph I.
From 1902, Łukasiewicz was employed as a private teacher, and also as a desk in the
Universitary Library of Lvov. So it was until 1904 when he obtained a scholarship to study
abroad. He defends his ‘Habilitationschrift’ in 1906, entitled “Analysis and construction
of the concept of cause”. This permits to give university courses. His first lectures were
on the Algebra of Logic, according to the recent translation to Polish of this book of the
French logician Louis Couturat.
Between 1902 and 1906, Lukasiewicz continued his studies in the universities of Berlin
and Leuwen (Lovaina). In 1906, by his ‘Habilitationschrift’, he obtain the qualification as

634 A. Garrido
university professor at Lvov. And the, in 1911, he was appointed as associate professor in
his ‘alma mater’ (Lemberg).
Jan Łukasiewicz was also very active in historical research on logic, giving a new and
up-to-date interpretation of Aristotle’s syllogism and of the Stoics’ propositional calculus.
According to Scholz, the better pages on history of logic are due to him. And also, as
Arianna Betti says, “Jan Lukasiewicz is first and foremost associated with the rejection of
the Principle of Bivalence and the discovery of Many-Valued Logic.”
The discovery of MVL by Lukasiewicz was in 1918, a little earlier than Emil Leon
Post. According to Jan Wolenski, “although Post’s remarks were parenthetical and
extremely condensed, Lukasiewicz explained his intuitions and motivations carefully and
at length. He was guided by considerations about future contingents and the concept
of possibility”. So, he introduces, firstly, three-valued logic, then four-valued logic,
generalized to logics with an arbitrary finite number of veritative values, and finally, to
logics with a countably infinite-valued number of such values.
Very noteworthy is his treatment of the history of logic in the light of the new formal
logic (then called Logistics). Thus, not only he addressed the issue of future contingents
departing from Aristotle, but also put in value logic of the Stoics, at least so far taken. In
fact, Heinrich Scholz said, rightly, that Lukasiewicz had written the most lucid pages on
the history of logic.
Keywords Aristotle · G.W. Leibniz · Non-classical logics · Many-valued logics ·

Fuzzy logic · Fuzzy measures and integrals · Applied logic in the Iberian Peninsula
Mathematics Subject Classification (2000) Primary 01A20, 01A55, 01A60, 03A05,

03B50, 03B52, 03B60, 03B80, 03E72, 03E75, 03G20, 08E72
1 Introduction
As we know, logic is the study of the structure and principles of correct reasoning,
and more specifically, attempts to establish the principles that guarantee the validity of
deductive arguments [14, 15]. The central concept of validity is for logic, because when
we affirm the validity of an argument are saying that it is impossible that its conclusion is
false if its premises are true.
Propositions are descriptions of the world, that is, are affirmations or denials of events
in various possible worlds, of which the “real world” is just one of them [20, 21, 30, 31].
There is a long philosophical tradition of distinguishing between truth necessary (a priori
or “logical”) and facts “contingent” (a posteriori or “factual”).
Both have really led the two concepts of logical truth, without being opposed to each
other, are quite different: the conception of truth as coherence, and the conception of truth
as correspondence. According to the point of view of consistency, a proposition is true or
false depending on their relationship with respect to a given set of propositions, because
of the rules of that system. Under the terms of correspondence, a proposition is true or
false, if it agrees with reality, that is, the fact referred to [22, 24].
Many-Valued Logics in the Iberian Peninsula 635
To further enhance the complexity of the problem, not only analyze trueness or falsity
of propositions, but also of theories, ideas and models [10, 19]. And so, we allow new and
different conception of truth.
The basic idea underlying all these approaches is that of an intrinsic dichotomy
between true and false. This opposition implies the validity of two fundamental laws of
classical logic:
• Principle of excluded middle: Every proposition is true or false, and there is another
possibility.
• Principle of non-contradiction: No statement is true and false simultaneously.
Such fundamental ideas produce some series of paradoxes and dissatisfaction that is based
on the need to overcome this strict truth-bivalence of classical logic.
Searching for the origins could lead too far and eventually disperse, which, as we know
is not very convenient for a job pretending to be research. So we will refer to these first
signs that appear in the East (China, India. . . ), and then we may analyze the problem of
“future contingents”, treated by Aristotle in Peri hermeneias.
For a man may predict an event ten thousand years beforehand, and another may predict the
reverse; that which was truly predicted at the moment in the past will of necessity take place
in the fullness of time. (Aristotle, Peri Hermeneias, ch. 9)
About Future Contingent Propositions, we must remember that they are statements
about states of affairs in the future that are neither necessarily true nor necessarily false.
Suppose that a sea-battle will not be fought tomorrow. Then it was also true yesterday
(and the week before, and last year) that it will not be fought, since any true statement
about the case that will be was also true in the past. But all past truths are now necessary
truths; therefore it is now necessarily true that the battle will not be fought, and thus the
statement that it will be fought is necessarily false. Therefore it is not possible that the
battle will be fought. In general, if something will not be the case, it is not possible for it
to be the case [23, 29, 36].
As we know, although the starting point of Leibniz “calculus universalis” were
Stagirite’s theories, Leibniz ends to be dependent from the ideas of Aristotle, to
finally develop its own axiomatic system, a more general type, based on applying the
Combinatorial Instrument to syllogistic [18, 34].
That issue (Future Contingent’s problem, with variations) would be then crucial in
medieval times, as during the Scholasticism, with William of Ockham, and Duns Scotus,
or Richard of Lavenham, among others, looked at from different point of views, for its
relationships with Determinism and ‘Divine Foreknowledge’. Then, this issue is taken up
by Spanish Jesuit F. Luis de Molina (and the famous controversy ‘De Auxiliis’ maintained
with the Dominican Fray Domingo Báñez), or Francisco Suarez, and even the great
polymath G.W. Leibniz dedicated his time [15, 20].
The controversial “De Auxiliis” involves two key works: the Concordia, from the
Jesuit Father Luis de Molina (1535–1600), and the Apology, from Fray Domingo Báñez
(1528–1604), a Dominican School of Salamanca and San Esteban’s Convent. In essence,
it represented the possible antagonism between free will of humans and efficacy of Divine
Grace. In short: Is Omniscience and Omnipotence compatible with the man’s freedom?
The discussion took a particularly interesting way during the Middle Age. In this period
philosophy was interconnected with theology. And one of the most important theological
issues was precisely the problem of future contingents, in its direct relationship with
636 A. Garrido
Christian doctrine. According to this tradition, related with the Divine Foreknowledge.
It includes knowledge of future possibilities to be made by human beings. But this
assumption seems to lead to a simple argument. It leads from foreknowledge to the need
of future events: now known as God and I will take the decision tomorrow, it’s true
that my choice of morning is given. My choice then, it seems necessary but not free.
Therefore, there appears to be no basis for claim that we have freedom of choice among
alternatives. The conclusion, however, would violate the idea of human freedom and of
moral responsibility [16, 35, 37].
Even then there is a dark time for the logic, and reappearing in the nineteenth century,
philosophers and mathematicians such as George Cantor, Augustus De Morgan and
George Boole, Gottlob Frege, . . . . There was born the new set theory, now called “classic”,
but then also had terrible enemies, as the then almighty Leopold Kronecker, who from his
professorship in Berlin did everything possible to hinder the work of Cantor, and the rise
of those new ideas.
As Kluge said, Frege’s logico-mathematical and philosophical speculations were
not historically isolated phenomena that arose completely de novo, as it were like
fulgurations of genius out of a conceptually unclouded sky. They were more like
nodal points in a long series of speculative endeavors that began with people like
Raymon Lull in the Middle Ages, continued through René Descartes, Athanasius Kircher,
Jakob Böhme, and G.W. Leibniz, that drew on the thoughts of people like Giordano
Bruno and Blaise Pascal, J.D. Gergonne and Thomas Hobbes, Pierre Gregoire and
Bernhard Bolzano, and in turn constituted the basis of much contemporary thought—the
works of Bertrand Russell and Rudolf Carnap, Edmund Husserl and Ludwig Wittgen-
stein, Alonzo Church, Strawson, and Willard van Orman Quine immediately come to
mind.
Remember that, and according to SEP (Stanford Encyclopedia of Philosophy), Bernard
Bolzano (1781–1848) was an outstanding mathematician and one of the greatest logicians
or even (as some would have it) the greatest logician who lived in the long stretch of
time between G.W. Leibniz and Gottlob Frege. As far as logic is concerned, Bolzano
anticipated almost exactly 100 years before Alfred Tarski and Rudolf Carnap their
semantic definitions of logical truth and logical consequence; and in mathematics he is
not only known for his famous Paradoxes of the Infinite, but also for certain results that
have become and still are standard in textbooks of mathematics such as the Bolzano-
Weierstrass theorem. Bolzano also made important contributions to other fields of
knowledge in and outside of philosophy. Due to the versatility of his talents and the
various fields to which he made substantial contributions, Bolzano became one of the
last great polymaths in the history of ideas.
2 Many-Valued Logics and the Lvov-Warsaw School
Parallel to this, there arises a new kind of thought and way of seeing must be the
act of philosophizing: the Polish Lwow-Warsaw School (LWS, by acronym). This is
happening like tributaries of a great river and sub-tributaries, departing from Leibniz,
from masters to disciples [11, 18, 35, 38, 42, 43]. Start with the aforementioned Bernard
Bolzano, which influence-much about his intellectual heir, Franz Brentano. This, in turn,
greatly influence on all his subsequent students. Among these disciples of Franz Brentano
will be one that particularly interested us. This was the Polish philosopher Kazimierz
Twardowski, who shared many characteristics with his teacher: love for precision and
clarity of ideas, charisma among those who treated him, preference for the spoken to
the written word, etc. . . From his chair in the city of Lvov spread many of the ideas
of Franz Brentano, adding their own. Its members took the logical-philosophical and
mathematical studies in Poland to the forefront of global world research. It was during
the “interbellum”, or period between the two World Wars, i.e. ranging from 1918–1939.
Then, rouse the Diaspora, after the war and by the strong communist dictatorship. Jan
Łukasiewicz introduced the three-valued logic and then generalized to the infinite-valued
[22, 24]. He was the effective mentor of Alfred Tarski, whereas officially it was Stanisław
Leśniewski.
The biographers of Tarski, Anita and Solomon Feferman, state that “along with
his contemporary, Kurt Gödel, he changed the face of logic in the twentieth century,
especially through his work on the concept of truth and the theory of models.” Tarski
had gone to the US to participate in a conference when Nazi troops invaded his
native Poland and could not return to it. Over time, he created in California the
most powerful logical school of his time; in fact, you can consider continuing the
tradition inaugurated by the LWS, outside the continent in ruins (Europe). Its ‘Semantic
Theory of Truth’ is one of the greatest achievements of the human thinking of all
time.
Many notable names were among the members of this school of logic, but could cite
[35, 37, 38] to:
• Jan Łukasiewicz,
• Stanisław Leśniewski,
• Kazimierz Ajdukiewicz,
• Tadeusz Kotarbiński,
• Mordechai Wajsberg,
• Alfred Tarski,
• Jerzy Słupecki,
• Andrzej Mostowski.
Also must be cited:
• Jan Woleński (as vindicator of the LWS’ memory),
• Helena Rasiowa,
• Roman Sikorski,
• Zdzisław Pawlak,
• Andrzej Skowron,
• Roman Murawski,
• etc.
Among them, one of the most interesting must be Jan Łukasiewicz, the father of many-
valued logics (MVLs, by acronym). Jan Łukasiewicz began teaching at the University of
Lvov, and then at Warsaw, but after World War II he had to continue at the Royal Academy
of Dublin, and then at Manchester.
638 A. Garrido
At first, Jan Łukasiewicz introduced the three-valued logic and then generalized to the
infinite-valued. That possibility modulation can be expressed by a membership function,
with values which run over all real numbers of the unit real interval, [0, 1], instead of
being reduced to the dichotomic {0, 1} of classical logic:
• True vs. False,
• 0 vs 1,
• White vs Black,
• etc.,
so, allowing the treatment of uncertainty and vagueness, important not only from the
theoretical point of view, but from the applications.
The logical research of Łukasiewicz suffered a long slumber, until Zadeh, among
others. The one who would see the potential utility in 1965, firstly obtaining a generalized
version of the classical theory of sets, now denoted by FST, acronym of the so-called
“Fuzzy Set Theory”, and later, its application to logic, introducing the “Fuzzy Logic”.
We must not forget that Zadeh, an engineer, knew Jan Łukasiewicz’s research as
explained by his colleague, the brilliant American logician Stephen Cole Kleene [21].
According to this theory, we have a transfer function derived from the characteristic
function usually called the “membership function”, which runs from the universe of
discourse, U, until the unit closed interval of reals, which is [0, 1]. Not so in the sets
“classic” or “crisp sets”, where the range of the function is reduced to a set consisting of
only two elements, namely was the {0, 1}. Therefore, fuzzy set theory is a generalization
of classical set theory [14].
We may mention the uncertainty principle of quantum physics by Werner Heisenberg.
The theory of “vague sets” (today, so-called Fuzzy Sets) proceeds from the quantum
physicist and German philosopher Max Black (1937), also analyzes the problem of
modeling “vagueness”. He differs from Russell in that he proposes that traditional logic
can be used by representing vagueness at an appropriate level of detail and suggests
that Russell’s definition of vagueness confuses vagueness with generality. He discusses
vagueness of terms or symbols by using borderline cases where it is unclear whether the
term can be used to describe the case. When discussing scientific measurement he points
out “the indeterminacy which is characteristic in vagueness is present also in all scientific
measurement”.
An idea put forward by Black is the idea of a consistency profile or curve to enable
some analysis of the ambiguity of a word or symbol. To the fuzzy logic researcher of today
these curves bear a strong resemblance to the membership functions of (type-1)-fuzzy
sets. Also may be considered the subsequent contribution of the Polish Jan Łukasiewicz
(1878–1956).
So, they must have greatly influenced Lofti A. Zadeh (b. 1921) to publish his seminal
paper in the journal Information and Control, and 3 years later (since 1968), the so-called
“Fuzzy Algorithm” [39–41].
In 1923, the British logician Bertrand Russell wrote that all traditional logic habitually
assumes that precise symbols are being employed. It is, therefore, not applicable to this
terrestrial life but only to an imagined celestial existence.
And Lofti Asker Zadeh, says that according the complexity of a system increases, our
ability to make precise and yet significant statements about its behavior diminishes until
a threshold is reached beyond which precision and significance (or relevance) become
almost mutually exclusive characteristics.
For such reasons—during the last decades—some very powerful ‘Fuzzy Mathematics’
has been developed, basically in Japan, but also in Europe, where these ideas came to
fruition, creating a powerful technological “boom”, with new techniques based on “fuzzy”
concepts. This trend was particularly strong in Oriental countries, such as South Korea,
China or India.
And much later these ideas, even more applications came to Western countries, both
European and American, with brilliant studies both from a mathematical point of view
and its philosophical implications, as always connected therewith. Very active research
groups [15], where papers on Many-Valued Logic have been published, proceed currently
from good European universities, for instance of:
• Warsaw,
• Prague,
• Ostrava,
• Vienna,
• Lisbon,
• Opole,
• Barcelona,
• Madrid,
• Toulouse,
• Pamplona,
• Granada,
• etc.
Today, some emerging countries, such as Brazil (Newton Da Costa or Jean-Yves Béziau),
India or Turkey, are becoming related with the investigation of all these theories and
associated methods, paraconsistent logics and so one [1–9].
From the above it follows that you may need a radical rethink of our classical
concepts of truth and falsehood, replacing the concept of fuzziness (vagueness or
fuzziness) as a result of which the truth or falsity are only extreme cases. By fuzziness
we understand the fact that a proposition may be partially true and partially false
simultaneously. A person is not just tall or short, but partially may participate in both
features, so that only above and below certain heights it is necessary to call upper
or lower bound, while in the intermediate zone of both heights exist as a graduation
which is ceasing to be high. It seems intuitively clear that the concept of fuzziness is
rooted in most of our ways of thinking and speaking [16]. Another separate issue is
the valuation of that each individual granted such a fuzzy character (the glass half full
or half empty), which depend on subjective psychological issues and are difficult to
evaluate.
The fuzzy principle states that everything is a matter of degree. It will be its more
famous “leitmotiv”. All propositions acquire a truth value between one (true) and zero
(false), inclusive. The allocation of these extreme values will only be given in the case of
logical truths or falsehoods or strong inductions: “All men are mortal” can be an example
of strong induction, since there is no counterexample.
640 A. Garrido
The arguments for introducing the concept of fuzziness in logic have already been
exposed, but it will be necessary to examine in detail some key aspects:
(a) The historical background and methodological concept.
(b) The possibility of building an infinite-valued formal language, and if so, try to define
their properties and laws.
(c) The philosophical and practical consequences stemming from such introduction.
One of the most interesting cases in the history of AI is the country of Romania [25, 27,
28]. We have the greatest landmark in the person of the mathematician Grigore Constantin
Moisil (1906–1973), who introduced Computer Science in the country; after he had left a
very brilliant school of researchers from Romania devoted to mathematics and AI, many
of them scattered around the world by the ‘economic diaspora’, after the Communist
period. After World War II, Grigore C. Moisil started teaching Mathematical Logic and
Algebra at Iasi and Bucharest, as he understood that the new emerging field of computers
would have enormous repercussions for the social fabric of society. He continued working
about the ideas of Shannon on Circuits, and some Łukasiewicz’s fundamental advances
on Many-Valued Logics, where the Fuzzy Logic eventually derived from.
The Łukasiewicz-Moisil Algebras (LMA) was created by G. Moisil as an algebraic
counterpart for the many-valued logics of J. Łukasiewicz. They are an attempt to give
semantic consistency to n-valued logics. This theory has developed to a considerable
extent both as an algebraic theory of intrinsic interest and in view of its applications to
logic and switching theory.
The study of LMA was followed by G. Georgescu and A. Iorgulescu, from Bucharest;
also are very relevant C. Calude, G. Paun (membrane computing), and some others, in
different areas. Also worthy of mentioning is the figure of Solomon Marcus (1925–2016),
an inspired disciple of G.C. Moisil, because Marcus has made great contributions to many
fields of Mathematics, such as Logic, Analysis, or Computational Linguistics, of which
he is one of the founders and a principal contributor [25].
Antonio Monteiro (1907–1980), mathematician born in Portuguese Angola, showed
that for every monadic Boolean-algebra we can construct a 3-valued Łukasiewicz-algebra,
and that any 3-valued Łukasiewicz-algebra is isomorphic to a Łukasiewicz-algebra thus
derived from a monadic Boolean-algebra. Roberto Cignoli says about it that since it
was shown by Halmos that monadic Boolean-algebras are the algebraic counterparts
of classical first order monadic calculus, Monteiro considered that the representation
of 3-valued Łukasiewicz-algebras into monadic Boolean-algebras gives a proof of the
consistency of 3-valued Łukasiewicz-logicas relative to classical logic. He showed that,
from the algebraic point of view, the three-valued Łukasiewicz-logic stands in the same
relation to constructive logic with strong negation as classical logic does to intuitionistic
logic.
Of course there is an increasing production of publications on the area. But many of
the best papers on Many-Valued Logics currently come from good European Universities
and very active research groups. This is possible because very remarkable researchers on
MVLs (in particular on Mathematical Fuzzy Logic) have created a solid and consistent
basis for these theories. Such has been the case for Petr Hájek, from the Charles
University (Prague), P. Cintula, Jan Pavelka, Líbor Behounek, or Vilem Novak, from
Ostrava. They have powerful research groups, with publications which are among the
most internationally valued in this field [16, 26].
And they are not alone, as in France we have the important task of dissemination
and investigation of D. Dubois and H. Prade, Elie Sanchez (1944–2014), or B. Bouchon-
Meunier. In Germany, H.-J. Zimmerman, or S. Gottwald (1943–2015). Also we find US
researchers as Z. Wang, G. Klir or R.R. Yager; in Hungary (J. Fodor, . . . ), Canada (W.
Pedrycz, . . . ), UK, Pays Bas (E. Kerre, B. de Baets, G. Cooman, M. De Cock, . . . ), Italy
(G. Gerla, A. Di Nola, . . . ), Austria, Argentina (R. Cignoli), Brasil, Turkey, etc.
In Poland they follow the great tradition of the LWS of logic and mathematics, and
with contributions to research the uncertainty topic through the Rough Sets, by Zdzisław
Pawlak (1926–2006), and continued by Andrzej Skowron, among others [15, 38].
3 Reception of Many-valued Logics and Fuzzy Logic in the

Iberian Peninsula
As remarkable precursor of Automatics, we must mention the mathematician and Spanish

engineer Leonardo Torres Quevedo (1852–1936).
Also, in medieval times, Raymond Lully, and his famous book, Ars Magna.
One of the first Hispanic scholars giving notice of the new currents was Juan David
Garcia Bacca, who in 1936 published his Introduction to modern logic, a work very
praised by I. M. Bocheński and Heinrich Schölz.
Later, try so eminent teachers, among them Alfredo Deaño (editor of the Spanish
translation of Łukasiewicz’s selected papers), Miguel Sánchez-Mazas (interpreting in
the more deep sense the logico-mathematical works and ideas of Gottfried Whilhelm
Leibniz, as—for instance—the known “characteristica universalis”), or Manuel Sacristán
(prosecuted in ‘academia’ due to its Marxist point of views), all them very often clashing
against a very conservative and not so good innovative ideas [32, 33].
One very good initiative has been the creation in the old mining town of Mieres, and
by the Government of Asturias, named the ‘Research Center for Artificial Intelligence
and Soft Computing’, initially around someone as well-known as Enric Trillas, who can
be considered the father of the introduction of Fuzzy Logic in the Spanish University
curricula. This center has attracted many of the most famous international researchers,
such as well-known Japanese Professor Michio Sugeno. His topics of research are very
broad working, but revolve around fuzzy methods, as well as philosophical implications
these carry.
Although I have left it for last, a name should not be omitted from those that appear
only from time to time in Spain. I am referring to the Father Pablo Domínguez Prieto
(1966–2009), Spanish philosopher and theologian [11–13], who wrote the first major
book in Spain on the LWS, from his PhD thesis, at Madrid (1993). Such work is so-
called Indeterminación y Verdad. La polivalencia lógica en la Escuela de Lwow-Warsaw
(Indeterminacy and Truth), and was published in 1995, with a foreword by Arch. J.M.
życiński, and showing a very strong influence by Jan Woleński.
Pablo can be considered as one of the Spanish precursors in the study of MVLs, from
the philosophical and historical point of view.
Another interesting Spanish author who has been reporting these new streams of logic
is Prof. Julián Velarde, with Polyvalent Logic, or his book Formal Logic, a volume II
642 A. Garrido
belonging to its History of Logic, all of them around the University of Oviedo and its
service publications, or later, to the Editorial Pentalfa. Also of great interest may be his
work Gnoseology of Fuzzy Systems, which analyzes the deep philosophical connections
of these issues [34].
New research groups have been formed in recent times [17], as the Spanish institution
CSIC (Consejo Superior de Investigaciones Científicas), centered in Barcelona, led by
Lluis Godo and Francesc Esteva. Or the group that belongs to the UPNA (Public
University of Navarra), headed by Humberto Bustince. Or in the University of Granada
(lead by Miguel Delgado Calvo-Flores).
Observe that the impulse to the study of Mathematical Logic in Spanish universities
came, among others, through the aforementioned Prof. Miguel Sánchez-Mazas, and also
by Prof. Manuel Garrido.
In Portugal the origins of the study of AI are linked to the names of Luis Moniz Pereira,
Helder Coelho and Fernando Pereira, who in 1973 created the LNEC, within which the
following year formed a division of Computer Science [16].
In 1977 the programming language called DEC-10 Prolog is introduced, which Helder
Coelho contributes greatly to propagate in Brazil.
In 1984 is created the Portuguese Association for AI (being APPIA, by acronym),
which maintains its vitality with many publications, and also organizing congresses.
Currently, there exist in Portugal four basic areas of work, related respectively with:
• Learning;
• Knowledge Representation;
• Knowledge in general, and
• Applications.
The research on Logic and AI in Portugal maintains its vitality through good publications,
and also organizing very important conferences, in Lisbon, Porto, etc.
4 Final Note
In early January, about 700 scientists specialized in technological areas signed an open
letter, warning of the dangers of AI. There have been many reactions to this paper,
with catastrophic scenarios described in science fiction that warn of a revolution of
the machines and extinction of humanity. However, the current risks are different. For
instance, the ability of processing huge amounts of data by computers, which can be
released to those who are in control. Today, the main threat of artificial intelligence is a
misuse of the abilities of the equipment, which can extract and analyze data in bulk.
The director of the Institute of Artificial Intelligence (IIA) of Barcelona, Ramón López
de Mántaras, also adds autonomous weapons as a threat: “in order to have robots soldiers
is very worrying, because for a robot it is almost impossible to distinguish between an
innocent civilian and a fighter”.
“Today we can ensure that none of the current robots, nor any that will be in short,
medium and even long term fairly, would be out of control”.
“The problem is not in technology itself, but in humanity. Is more likely to be the
man with evil intentions, who may produce a very possible war between humans and
machines”, he qualifies.
The loss of workplaces is another hazard of AI.
According to López de Mántaras,
“So far the robots moved people from repetitive or dangerous task, but with advances
in artificial intelligence begin to endanger related services sector; for example, jobs”.
“Experts agree that education is the most important measure. We will look for other
jobs where creativity is essential, and therefore will require investment in education to
add art between engineering and mathematics”.
Meanwhile, “it must be given much more importance to lifelong learning so that people
can be recycled more easily and be able to change careers. At least so far, the technological
changes, which have destroyed workplaces, have also created other ones instead.”
Acknowledgements This work was supported by the MICINN’s Research Project FFI2016-77574-P,
and the Investigation Group of our Spanish University (UNED) that we belong to, into a section
about ‘Science and Technique’, that belongs to the Project entitled “Estudio sistemático de las lecturas
heideggerianas de Jacques Derrida. Confluencias y divergencias”. Being Principal Researcher (IP), Prof.
Dr. Cristina De Peretti.
References
1. Belnap, N.D.: A useful four-valued logic. In: Dunn, J.M., Epstein, G., Wolf, R.G. (eds.) Modern Uses
of Multiple-Valued Logic, pp. 8–37. Reidel, Dordrecht (1977)
2. Béziau, J.-Y.: Universal logic. In: Childers, T., Majer, O. (eds.) Logica’94: Proceedings of the 8th
International Symposium, pp. 73–93. Czech Academy of Sciences, Prague (1994)
3. Béziau, J.-Y.: La véritable portée du théoréme de Lindenbaum-Asser. Log. Anal. 42(167–168), 341–
359 (1999)
4. Béziau, J.-Y.: From paraconsistent logic to universal logic. Sorites 12, 5–32 (2001)
5. Béziau, J.-Y.: Sequents and bivaluations. Log. Anal. 44(176), 373–394 (2001)
6. Béziau, J.-Y.: Many-valued and Kripke semantics. In: J. van Benthem et al. (eds.) The Age of
Alternative Logics, pp. 89–91. Springer, Dordrecht (2006)
7. Béziau, J.-Y.: 13 questions about universal logic. Bull. Sect. Log. 35(2–3), 133–150 (2006)
8. Béziau, J.-Y.: From consequence operator to universal logic: a survey of general abstract logic. Logica
Universalis, pp. 3–17. Birkhäuser, Basel (2007)
9. Birkhoff, G.: Universal algebra. In: Proceedings of the First Canadian Mathematical Congress 1945,
pp. 310–326. University of Toronto Press, Toronto (1946)
10. Da Costa, N., Béziau, J-Y.: Logiques Classiques et Non-Classiques: essais sur le fondements de la
logique. Editora Masson et Cie, Paris (1997)
11. Domínguez Prieto, P.: Indeterminación y Verdad. La polivalencia lógica en la Escuela de Lvov-
Varsovia. Nossa y Jara Editores, Móstoles (1995)
12. Domínguez Prieto, P.: Teoría del Contorno Lógico. Nossa y Jara Editores, Móstoles (1999)
13. Domínguez Prieto, P.: Lógica Modal y Ontología. Nossa y Jara Editores, Móstoles (2001)
14. Dubois, D., Prade, H.: Fuzzy Sets and Systems: Theory and Applications. Academic, New York,
(1980)
15. Garrido, A.: Lógicas de nuestro tiempo. Madrid, Editorial Dykinson (2014)
16. Garrido, A.: Lógica Aplicada. Vaguedad e Incertidumbre. Madrid, Editorial Dykinson (2014)
17. Garrido, A.: The Challenges of Artificial Intelligence. Acta Universitatis Apulensis (AUA), Special
Issue ICTAMI 2015, pp. 155–165. Alba Iulia (2015)
18. Garrido, A.: Leibniz y las Nuevas Lógicas. Editorial Comares, Granada (2016)
644 A. Garrido
19. Goguen, J.A.: The logic of inexact concepts. Synthese 19(3/4), 325–373 (1969)
20. Hájek, P.: Metamathematics of Fuzzy Logic. Kluwer, Boston/Dordrecht (1998)
21. Kleene, S.C.: Introduction to Metamathematics. North-Holland, Amsterdam/Van Nostrand, New York
(1952)
22. Łukasiewicz, J.: Elements of Mathematical Logic. Państwowe Wydawnictwo Naukowe (PWN),
Warsaw (1963)
23. Łukasiewicz, J.: Selected Works, L. Borkowski (ed.). North Holland, Amsterdam (1970)
24. Łukasiewicz, J.: Écrits logiques et philosophiques. Librairie Philosophique Jacques Vrin, Paris (2013)
25. Marcus, M., Marcinkiewicz, M.: Building a large annotated corpus of English: The Penn Treebank.
Computat. Linguist. 19(2), 313–330 (1993)
26. Mira, J., et al.: Aspectos Básicos de Inteligencia Artificial. UNED (1995)
27. Moisil, G.: Încercări vechi şi noi în logica neclasică. Bucuresti (1965)
28. Moisil, G.: Elemente de logică matematică şi teoria mulţimilor. Bucuresti (1968)
29. Murawski, R.: Essays in the Philosophy and History of Logic and Mathematics. Poznań Studies in
the Philosophy of the Sciences and the Humanities. Rodopi (2010)
30. Rasiowa, H., Sikorski, R.: The Mathematics of Metamathematics. Polish Scientific, Warszawa (1963)
31. Rosser, J.B., Turquette, A.R.: Many-Valued Logics. North-Holland, Amsterdam (1952)
32. Sacristán, M.: Introducción a la Lógica y al Análisis Formal, Colección del Círculo de Lectores.
Barcelona (1990)
33. Sánchez-Mazas, M.: Obras Escogidas. Servicio Editorial Universidad del País Vasco (UPV). En dos
volúmenes. Edición de Javier de Lorenzo. Donostia/ San Sebastián (2002)
34. Velarde, J.: Leibniz y la lógica. Themata. Revista de Filosofía 29 (Treinta años de estudios
leibnizianos), 217–231 (2002)
Dordrecht (1989)
36. Woleński, J.: Historico-Philosophical Essays, vol. I. Copernicus Center, Krakow (2013)
37. Wolenski, J., Zygmunt, J.: Jerzy Slupecki (1904–1987): Life and Work. Studia Logica: Int. J. Symb.
Logic, 401–411 (1989)
38. Wybraniec-Skardowska, U.: Polish Logic. Some Lines from a Personal Perspective. Publications of
39. Yager, R.R.: Fuzzy Sets and Applications: Selected Papers by L.A. Zadeh. Wiley, New York (1987).
Yager et al. (eds.)
40. Zadeh, L.A.: Fuzzy sets. Inf. Control 8, 338–355 (1965)
41. Zadeh, L.A.: Fuzzy Logic and approximate reasoning. Synthese 30, 407–428 (1975)
42. Zimmermann, H.J.: Fuzzy Set Theory and Applications. Kluwer Academic Press, Boston (1991)
43. Zinoviev, A.A.: Philosophical Problems of Many-Valued Logic. Reidel Verlag, Dordrecht (1963)
A. Garrido ()
Department of Fundamental Mathematics, Faculty of Sciences of the National University of Distance
Education (UNED), Madrid, Spain
Ontology of Logic and Mathematics
in Lvov-Warsaw School
Roman Murawski
Abstract The aim of the paper is to consider ontological views connected with math-
ematics and logic of main representatives of Lvov-Warsaw School of Philosophy.
In particular views of the following scholars will be presented and discussed: Jan
Łukasiewicz, Stanisław Leśniewski, Alfred Tarski, Tadeusz Kotarbiński and Kazimierz
Ajdukiewicz. We shall consider also views of Andrzej Mostowski who belonged to the
second generation of the school as well as of Leon Chwistek who was not directly the
member of this group but whose conceptions are of interest.
Keywords Philosophy of logic · Philosophy of mathematics · Ontology · Platonism ·

Nominalism · Intuitionism
Mathematics Subject Classification (2000) Primary 00A30; Secondary 03A05, 01A60
The aim of the paper is to describe ontological views concerning logic and mathematics
of representatives of Lvov-Warsaw School of Philosophy.1 As will be shown there was
in fact no common conception. The element unifying members of the school was not
a particular philosophical doctrine but rather the method of practicing and developing
philosophy. I. Dambska
˛ charactered it in the following way:
There was no common doctrine, no uniform view of the world shared by Lvov philosophers.
What formed the foundation of the spiritual community of those scholars was not the content of
conceptions but rather the way, the method of doing philosophy as well as the common scientific
language. Thanks to that feature members of this group could be both spiritualists and materialists,
nominalists and realists, logicians and psychologists, philosophers of nature and theoreticians of
art.2 [8, p. 17]
1 On Lvov-Warsaw School in Philosophy see Woleński [38] and [39] as well as Murawski [35] and [36].
2 Nie łaczyła
˛ bowiem filozofów lwowskich jakaś wspólna doktryna, jakiś jednolity poglad ˛ na świat. To,
co stworzyło podstaw˛e wspólnoty duchowej tych ludzi, to była nie treść nauki, tylko sposób, metoda
filozofowania i wspólny j˛ezyk naukowy. Dlatego wyjść z tej szkoły mogli: spirytualiści i materialiści,
nominaliści i realiści, logistycy i psychologowie, filozofowie przyrody i teoretycy sztuki.

646 R. Murawski
1 Jan Łukasiewicz: (Neo)Platonism
Let us start by considering views of Jan Łukasiewicz (1879–1956). One should begin
by stressing his anti-psychological attitude. Psychologism claims that the objects inves-
tigated by logic and mathematics exist as psychological beings and are got to know
like other psychological facts. This approach was popular in the philosophy of logic
and mathematics at the end of the nineteenth century. It was criticised in particular by
Frege, Husserl and Meinong. In the paper “Logika a psychologia” [Logic and psychology]
(cf. [21]) Łukasiewicz formulated his arguments against psychologism. First he stated
that laws of psychology are empirical and consequently only probable whereas laws
of logic are certain. Laws of both those disciplines are also of different character: the
laws of logic concern relations between the truth and falsity of judgements whereas the
laws of psychology state the relations between psychological phenomena. Łukasiewicz
concludes:
Exposing the attitude of logic towards psychology can be to the advantage of both sciences.
Logic will be cleared from the weeds of psychologism and empiricism, which choke its right
development and the psychology of cognition will get rid of a priori traces which hid the light of
the sincere splendour of its truths. Since one should remember that logic is an a priori science,
like mathematics, whereas psychology, like any other natural science, is based and must be based
on experience.3
Łukasiewicz stressed explicitly the apriorism of logic. In the paper “O twórczości w

nauce” [On creativity in science] (cf. [25]) he wrote:
Logic is an a priori science. Its theorems are true by virtue of definitions and axioms flowing from
reason and not from experience. This science is a domain of pure mental creativity. [. . . ] Logical
and mathematical judgements are truths only in the world of ideal beings. We will never know
whether some real objects correspond with these beings.
A priori constructions of the mind, being part of every synthesis, imbue the whole science with an
ideal and creative element.4
Łukasiewicz admitted that logic and mathematics have a nominalistic robe (see for
example his paper “Logistyka a filozofia” [Logistics and philosophy], [23, p. 119]) but
simultaneously he saw some difficulties in the nominalistic approach. An individual can
create only a finite number of inscriptions. Hence a set of inscriptions is finite what would
mean that the set of theses of logic and mathematics would be finite as well but “on this
basis it would be as difficult to practise logistics, especially metalogistics, as to build
3 Wyświetlenie stosunku logiki do psychologii przynieść może korzyści obu tym naukom. Logika oczyści
si˛e z chwastów psychologistycznych i empirystycznych, które tłumia˛ jej prawidłowy rozwój, a psycholo-
gia poznania pozb˛edzie si˛e naleciałości apriorycznych, spod których szczery blask jej prawd nie mógł
jakoś dotad
˛ zajaśnieć. Należy bowiem pami˛etać, że logika jest nauka˛ aprioryczna,˛ tak jak matematyka,
a psychologia, tak jak każda nauka przyrodnicza, opiera si˛e i opierać si˛e musi na doświadczeniu [21, p.
491].
4 Logika jest nauka˛ aprioryczna.˛ Twierdzenia jej sa˛ prawdziwe na mocy określeń i pewników płynacych˛
z rozumu, nie z doświadczenia. Nauka ta jest dziedzina˛ czystej twórczości myślowej. [. . . ] Sady
˛ logiczne
i matematyczne sa˛ prawdami jedynie w świecie bytów idealnych. Czy bytom tym odpowiadaja˛ jakieś
przedmioty rzeczywiste, o tym zapewne nigdy si˛e nie dowiemy.
Aprioryczne konstrukcje umysłu, wchodzac ˛ w skład każdej syntezy, przepajaja˛ cała˛ nauk˛e pierwiastkiem
idealnym i twórczym [22, pp. 13–14].
Ontology of Logic and Mathematics in Lvov-Warsaw School 647
arithmetic on the assumption that the set of natural numbers is finite” [25, p. 224].5 It
would also lead to make logic dependent on certain empirical facts, i.e., on the existence
of inscriptions, which is difficult to accept.
According to Łukasiewicz the nominalism of logic and mathematics is virtual.
Moreover, logic was developed without solving the problem of its nominalism. In his
article “Logistyka a filozofia” [Logistics and philosophy] he wrote:
We have so far been little worried by these difficulties, and this is the strangest point. It was so
probably because, while we use nominalistic terminology, we are not true nominalists but incline
toward some unanalysed conceptualism or even idealism.6 [25, p. 224]
Łukasiewicz himself thought that the objects that logic investigated existed only
beyond the sphere of inscriptions. He did not develop some alternative to nominalism—
he just formulated his personal view. But his view resulted from his personal religious
convictions—influenced by these convictions Łukasiewicz opted for the Neoplatonic
interpretation of logic. In the paper “W obronie logistyki” [In defence of logic] he wrote:
In concluding these remarks I should like to outline an image which is connected with the most
profound intuitions which I always experience in the face of logistic. That image will perhaps
shed more light on the true background of that discipline, at least in my case, than all discursive
description could. Now, whenever I work even on the least significant logistic problem, for
instance, when I search for the shortest axiom of the implicational propositional calculus I always
have the impression that I am facing a powerful, most coherent and most resistant structure. I sense
that structure as if it were a concrete, tangible object, made of the hardest metal, a hundred times
stronger than steel and concrete. I cannot change anything in it; I do not create anything of my own
will, but by strenuous work I discover in it ever new details and arrive at unshakable and eternal
truth. Where is and what is that ideal structure? A believer would say that it is in God and is His
thought.7 [25, p. 249]
Łukasiewicz stressed that this was his personal view. He was of the opinion that logic is
neither called nor allowed to solve the eternal philosophical debate concerning universals.
5 na takiej podstawie byłoby równie trudno uprawiać logistyk˛ e, a zwłaszcza metalogistyk˛e, jak trudno
byłoby zbudować arytmetyk˛e na gruncie założenia, że zbiór liczb naturalnych jest skończony [23, p.
120].
6 Mało dotychczas przejmowaliśmy si˛ e tymi trudnościami i to jest w tym wszystkim najdziwniejsze.
Działo si˛e to chyba dlatego, że używajac
˛ terminologii nominalistycznej, nie jesteśmy naprawd˛e nomi-
nalistami, lecz hołdujemy jakiemuś nie zanalizowanemu konceptualizmowi czy nawet idealizmowi [23,
p. 120].
7 Chciałbym na zakończenie tych uwag nakreślić obraz zwiazany ˛ z najgł˛ebszymi intuicjami, jakie
odczuwam zawsze wobec logistyki. Obraz ten rzuci może wi˛ecej światła na istotne podłoże, z jakiego
przynajmniej u mnie wyrasta ta nauka niż wszelkie wywody dyskursywne. Otóż ilekroć zajmuj˛e si˛e
najdrobniejszym nawet zagadnieniem logistycznym, szukajac ˛ np. najkrótszego aksjomatu rachunku
implikacyjnego, tylekroć mam wrażenie, że znajduj˛e si˛e wobec jakiejś pot˛eżnej, niesłychanie zwartej i
niezmiernie odpornej konstrukcji. Konstrukcja ta działa na mnie jak jakiś konkretny dotykalny przedmiot,
zrobiony z najtwardszego materiału, stokroć mocniejszego od betonu i stali. Nic w niej zmienić nie
mog˛e, nic sam dowolnie nie tworz˛e, lecz w wyt˛eżonej pracy odkrywam w niej tylko coraz to nowe
szczegóły, zdobywajac ˛ prawdy niewzruszone i wieczne. Gdzie jest i czym jest ta idealna konstrukcja?
Filozof wierzacy
˛ powiedziałby, że jest w Bogu i jest myśla˛ Jego [24, p. 165].
648 R. Murawski
2 Stanisław Leśniewski: Nominalism
Nominalism mentioned above was the philosophical doctrine of another representative

of Lvov-Warsaw School in Philosophy and, together with Łukasiewicz, the founder of
Warsaw School of Logic, Stanisław Leśniewski (1886–1939). This doctrine had strong
influence even on the contents as well as on the form of his logical constructions. His
views Leśniewski called constructive nominalism.
Leśniewski treated language as a collection of concrete inscriptions and expressions of
a language as finite sequences of signs. Two inscriptions of the same shape were treated
by him as two separate, different inscriptions. In his opinion there only exist as many
expressions as they have been written. One cannot speak of some potential existence
of expressions. Consequently a given logical system contains only so many theorems
as they have been written until a given moment, i.e., every logical system consists of
only a finite number of theorems. Leśniewski did not allow the existence of any general
objects, in particular of common properties of individual objects. Another consequence
of Leśniewski’s nominalism was the fact that two equivalent systems, for example the
system of propositional calculus based on negation and implication and the system based
on negation and disjunction as primitive connectives usually treated as two variants of
the same logic should be treated now as two different systems. Leśniewski’s systems are
never something complete at a given moment.
Leśniewski connected the described view with the so-called intuitive formalism.
According to it a language of logic—uniquely and completely codified—says always
“something” and about “something”. In the work “Grundzüge eines neuen System der
Grundlagen der Mathematik” [17] he wrote:
Having no predilection for various ‘mathematical games’ that consist in writing out according
to one or another conventional rule various more or less picturesque formulae which need not
be meaningful, or even—as some of the ‘mathematical gamers’ might prefer—which should
necessarily be meaningless, I would not have taken the trouble to systematize and to often check
quite scrupulously the directives of my system, had I not imputed to its theses a certain specific
and completely determined sense, in virtue of which its axioms, definitions and final directives,
have for me an irresistible intuitive validity.8 [18, p. 487]
In the work “O podstawach matematyki” [On the foundations of mathematics] (cf.

[16]) one reads:
They encouraged the disappearance of the feeling for the distinction between the mathematical
sciences, conceived as deductive theories, which serve to capture various realities of the world
in the most exact laws possible, and such non-contradictory deductive systems, which indeed
ensure the possibility of obtaining, on their basis, an abundance of ever new theorems, but which
8 Da ich keine Vorliebe für verschiedene «Mathematikspiele» habe, welche darin bestehen, dass man
nach diesen oder jenen konventionellen Regeln verschiedene mehr oder minder malerische Formeln
aufschreibt, die nicht notwendig sinnvoll zu sein brauchen oder auch sogar, wie es einige der «Math-
ematikspiele» lieber haben möchten, notwendig sinnlos sein sollen,—hätte ich mir nicht die Möhe
der Systematisierung und der vielmaligen skrupulösen Kontrollierung der Direktiven meines Systems
gegeben, wenn ich nicht in die Thesen dieses Systems einen gewissen ganz bestimmten, eben diesen
und nicht einen anderen, Sinn legen wörde, bei dem för mich die Axiome des Systems und die in
den Direktiven zu diesem System kodifizierten Schluss- und Definitionsmethoden eine unwiderstehliche
intuitive Geltung haben [17, p. 78].
simultaneously distinguish themselves by the lack of any connection with reality of any intuitive,
scientific value.9 [18, pp. 177–178]
Leśniewski treated formal systems as a means to transmit certain information about

the world and as a way to express what is intuitively true. This may seem not to be fully
in accordance with his nominalism and radical formalism. However he did not consider
those views as contradictory. In fact in “Grundzüge eines neuen System der Grundlagen
der Mathematik” [17] he wrote:
I see no contradiction, therefore, in saying that I advocate a rather radical ‘formalism’ in the
construction of my system even though I am an obdurate ‘intuitionist’. Having endeavoured to
express some of my thoughts on various particular topics by representing them a s a series of
propositions meaningful in various deductive theories, and to derive one proposition from others
in a way that would harmonize with the way I finally considered intuitively binding . . . .10 [18, p.
487]
For Leśniewski logic was the description of most general features of being (the same—
under the influence of Leśniewski—was claimed by Kotarbiński). Hence it plays the role
of a general theory of objects. This view was in accordance with the fact that Warsaw
School of Logic rejected the so-called analytic interpretation of logic, i.e., the thesis that
logic and mathematics are the set of tautologies that do not say anything about the world.
Logic and mathematics were thought to refer to the formal aspects of reality. Add that
Leśniewski rejected also the conventionalism in the style of Poincaré.
Leśniewski took a firm stand in the dispute concerning universals—he rejected the
existence of any ideal and general objects. In the paper “Krytyka logicznej zasady
wyłaczonego
˛ środka” [Critique of the logical principle of excluded middle] (cf. [20])
he gave the proof of non-existence of such objects that became very popular in Poland.
In the proof the concept of a feature as well as the principle of excluded middle and
the principle of contradiction were used. It was quoted—with some modifications—by
Kotarbiński in the paper “Sprawa istnienia przedmiotów idealnych” [The problem of
existence of ideal objects] (cf. [11]) and repeated in his book Elementy teorji poznania,
logiki formalnej i metodologji nauk [Elements of the theory of cognition, formal logic
and methodology] (cf. [12]). It became one of the justifications of reism propagated by
him. Leśniewski return to his proof in the work “O podstawach matematyki” [On the
foundations of mathematics] (cf. [16, pp. 183–184]) where he gave a new version of it in
9 Sprzyjało to zanikowi poczucia różnicy mi˛edzy naukami matematycznymi pojmowanymi jako teorie
dedukcyjne, służace
˛ do uj˛ecia w prawa możliwie ścisłe różnorodnej rzeczywistości świata, a takimi
niesprzecznymi systemami dedukcyjnymi, które zabezpieczaja˛ wprawdzie możność otrzymania na ich
gruncie obfitości wciaż
˛ nowych twierdzeń, odznaczajacych
˛ si˛e jednak jednocześnie brakiem jakichkol-
wiek łacz
˛ acych
˛ je z rzeczywistościa˛ walorów intuicyjno-naukowych [16, p. 166].
10 Ich sähe keinen Widerspruch darin, wenn ich behaupten wollte, dass ich eben deshalb beim Aufbau
meines Systems einen ziemlich radikalen «Formalismus» treibe, weil ich ein versteckter «Intuitionist»
bin: indem ich mich beim Darstellen von verschiedenen deduktiven Theorien bemöhe, in einer Reihe
sinnvolle Sätze eine Reihe von Gedanken auszudröcken, welche ich öber dieses oder jedes Thema hege,
und die einen Sätze aus den anderen Sätzen auf eine Weise abzuleiten, die mit den Schlussweisen
harmonisieren wörden, welche ich «intuitiv» als för mich bindend betrachte . . . [17, p. 78].
650 R. Murawski
which the concept of “feature” does not appear. The proof was preceded by the following
explanations:
At the time I wrote that passage [Leśniewski says about the appropriate fragment of his [19]—
my remark - RM] I believed that there are in existence in this world so called features and so
called relations, as two special kinds of objects, and I felt no scruples about using the expressions
‘feature’ and ‘relations’. It is a long time since I believed in the cxistence of objects which are
features, or in the existence of objects which are relations and now nothing induces me to believe
in the existence of such objects [. . . ] and in situations of a more ‘delicate’ character I do not use
the expressions ‘feature’ and ‘relation’ without the application of various extensive precautions
and circumlocutions. I also have no inclination at present—considering the possibility of various
interpretational misunderstandings—to ascribe this or that opinion on the question of ‘general
objects’ to the authors mentioned in the passage mentioned above.11 [18, p. 198]
3 Alfred Tarski: Nominalism
A follower of nominalism was also Alfred Tarski (1901–1983). This pronominalistic

attitude was the source of the fact that in the interwar period he treated language as a
set of sentences understood in a strictly nominalistic way as physical objects. However
his sympathies towards nominalism were in fact stronger. Mostowski wrote about this in
the following way:
Tarski, in oral discussions, has often indicated his sympathies with nominalism. While he never
accepted the » reism« of Tadeusz Kotarbiński, he was certainly attracted to it in the early phase of
his work. However, the set-theoretical methods that form the basis of his logical and mathematical
studies compel him constantly to use the abstract and general notions that a nominalist seeks to
avoid. In the absence of more extensive publications by Tarski on philosophical subjects, this
conflict appears to have remained unresolved. [32, p. 81]
Tarski’s pronominalistic attitude is confirmed in various sources. Firstly, it was Tarski’s

remark (preserved on a tape cassette) made during the symposium organised by the
Association for Symbolic Logic and the American Philosophical Association, held in
Chicago on 29th–30th April 1965, and dedicated to philosophical implications of Gödel’s
incompleteness theorem. Tarski said:
I happen to be, you know, a much more extreme anti-Platonist. [. . . ] However, I represent this
very [c]rude, naive kind of anti-Platonism, one thing which I would describe as materialism, or
nominalism with some materialistic taint, and it is very difficult for a man to live his whole life
with this philosophical attitude, especially if he is a mathematician, especially if for some reasons
he has a hobby which is called set theory. [9, p. 52]
11 W czasie, gdy ust˛ ep ten [chodzi tu o stosowny fragment pracy [19]—uwaga moja, R.M.] pisałem,
wierzyłem, iż istnieja˛ na świecie tzw. cechy i tzw. stosunki jako dwa specjalne rodzaje przedmiotów,
i nie odczuwałem żadnych skrupułów przy posługiwaniu si˛e wyrazami “cecha” i “stosunek”. Obecnie
nie wierz˛e już od dawna w istnienie przedmiotów b˛edacych˛ cechami, ani też w istnienie przedmiotów
b˛edacych
˛ stosunkami, nic mnie też nie skłania do wierzenia w istnienie takich przedmiotów [. . . ],
wyrazami zaś “cecha” i “stosunek” staram si˛e w sytuacjach o cokolwiek “delikatniejszym” charakterze
nie posługiwać bez daleko idacych
˛ ostrożności i omówień. Nie mam dziś także skłonności—wobec
możliwości rozmaitych nieporozumień interpretacyjnych—do przypisywania tych lub innych pogladów ˛
w sprawie „przedmiotów ogólnych” tym lub innym z autorów, wymienionym w ust˛epie wyżej przytoc-
zonym [16, p. 183].
Fefermans’ book [9] contains more similar words concerning Tarski himself or other
people’s opinions about Tarski. These opinions were expressed on Tarski’s 70th birthday
celebrations and remembered by Chihara, Chateaubriand and the Fefermans:
I am a nominalist. This is a very deep conviction of mine. It is so deep, indeed, that even after
my third reincarnation, I will still be a nominalist. [. . . ] People have asked me, » How can you,
a nominalist, do work in set theory and logic, which are theories about things you do not believe
in?« [. . . ] I believe that there is a value even in fairy tales.
[I am] a tortured nominalist.
Elsewhere Tarski has said more specifically that he subscribed to reism or concretism (a kind of
physicalistic nominalism) of his teacher Tadeusz Kotarbiński. [9, p. 52]
Also Tarski’s letter to Woodger, dated 21st November 1948, testifies to the importance
he attached to nominalism:
The problem of constructing nominalistic logic and mathematics has intensively interested me
for many-many years. Mathematics—at least the so-called classical mathematics—is at present
an indispensable tool for scientific research in empirical sciences. The main problem for me is
whether this tool can be interpreted nominalistically or replaced by another nominalistic tool which
should be adequate for the same purposes. [27, p. 147]
On many occasions Tarski stressed his sympathies towards Kotarbiński’s reism and
physicalism. He also translated into English (together with David Rynin) Kotarbiński’s
work “Zasadnicze myśli pansomatyzmu” [The Fundamental Ideas of Pansomatism]
(cf. [13]). The translation was published in Mind, one of the most important English
periodicals dedicated to philosophy. It was included in Tarski’s Collected Works [37].12
More details about Tarski’s sympathies and inclinations towards nominalism can be
found in the recently discovered protocols of Carnap from the discussions conducted at
Harvard in the academic year 1940/1941. Besides Carnap the other participants were
Tarski and Quine as well as—occasionally—Russell.
In the protocol of 10th January 1941 Carnap wrote down the following remarks
concerning nominalism and finitism:
Tarski: I understand basically only languages which satisfy the following conditions:
1. Finite number of individuals;
2. Realistic (Kotarbinski): the individuals are physical things;
3. Non-platonic: there are only variables for individuals (things) not for universals (classes and so
on) [26, p. 342].13
Mancosu notices [26, p. 343] a mistake: instead of ‘realistic’ it should be ‘reistic,’

which is confirmed by the reference to Kotarbiński.
Carnap’s notes also contain the following exchange of views:
I [Carnap]: Should we construct the language of science with or without types?
He [Tarski]: Perhaps something else will emerge. One would hope and perhaps conjecture that
the whole general set theory, however beautiful it is, will in the future disappear. With the higher
types Platonism begins. The tendencies of Chwistek and others (‘Nominalism’) of speaking only
12 It also testifies to Kotarbiński’s strong influence on Tarski.

13 Tarski: Ich verstehe im Grunde nur eine Sprache die folgende Bedingungen erföllt: [1] Finite Anzahl der
Individuen; [2] Realistisch (Kotarbiński): Die Individuen sind physikalische Dinge; [3] Nicht-platonisch:
Es kommen nur Variable för Individuen (Dinge) vor, nicht för Universalien (Klassen usw.).
652 R. Murawski
of what can be named are healthy. The problem is only how to find a good implementation.14 [26,
p. 334]
Of special interest—in the context of the problem of passing from the systems of the
theory of classes—is also Carnap’s summary of his conversation with Tarski on 12th
February 1941:
The Warsaw logicians, especially Leśniewski and Kotarbiński saw a system like PM (but with
simple type theory) as the obvious system form. This restriction influenced strongly all the
disciples; including Tarski until the ‘Concept of Truth’ (where the finiteness of the levels
is implicitly assumed and neither transfinite types nor systems without types are taken into
consideration; they are discussed only in the Postscript added later). Then Tarski realized that
in set theory one uses with great success a different system form. So he eventually came to see this
type-free system form as more natural and simpler.15 [26, p. 335]
One should notice that Tarski’s research practice, in particular his investigations
concerning set theory or the theory of models, contradicted in fact his nominalism to a
certain extent and would rather suggest that he was a follower of Platonism (this explains
the question mark in the title of this section). This discrepancy can be explained by the
spirit and ideological canon of the Polish School. According to them, research should not
be limited by any a priori philosophical foundations and all correct methods should be
allowed and applied.
4 Tadeusz Kotarbiński: Reism
Presenting philosophical views of Tarski we have mentioned Tadeusz Kotarbiński (1886–

1981) and his doctrine of reism to which Tarski referred. Let us say now something more
about this.
Reism is by Kotarbiński both a semantical and an ontological doctrine, moreover both
levels are in a certain sense parallel. Kotarbiński admitted that developing reism he used
some logical ideas of Leśniewski explained by the latter in his system of the calculus
of names called ontology. In the Preface to Elementy teorii poznania, logiki formalnej i
metodologii nauk [Elements of epistemology, formal logic and methodology of science]
14 Ich: Sollen wir vielleicht die Sprache der Wissenschaften mit oder ohne Typen machen? Er: Vielleicht
wird sich etwas ganz Anderes entwickeln. Es wäre zu wönschen und vielleicht zu vermuten, dass die
ganze allgemeine Mengenlehre, so schön sie auch ist, in der Zukunft verschwinden wird. Mit den höheren
Stufen fängt der Platonismus an. Die Tendenzen von Chwistek und anderen (» Nominalismus«), nur öber
Bezeichenbaren zu sprechen, sind gesund. Problem nur, wie gute Durchföhrung zu finden.
15 Die Warschauer Logiker, besonders Leśniewski und Kotarbiński, sahen ein System wie PM (aber mit
einfacher Typentheorie) ganz selbstverständlich als die Systemform an. Diese Beschränkung wirkte stark
suggestiv auf alle Schöler; auf T. selbst noch bis zu » Wahrheitsbegriff« (wo weder transfinite Stufen
noch stufenloses System betrachtet wird, und Endlichkeit der Stufen stillschweigend vorausgesetzt wird,
erst im später hinzugefögten Anhang werden sie besprochen). Dann aber sah T., dass in der Mengenlehre
mit grossem Erfolg eine ganz andere Systemform verwendet wird. So kam er schliesslich dazu, diese
stufenlose Systemform als natörlicher und einfacher zu sehen.
(cf. [14]) Kotarbiński wrote:

Still, I have learnt most things from Prof. Dr Stanisław Leśniewski. I admit that in many places
of the book. And they are the most important and clearest points. Besides I admit that all my
thoughts are deeply saturated with the influences of that extraordinary mind whose precious gifts
I have used, thanks to good luck, almost every day for a number of years. I am undoubtedly a
disciple of my colleague Leśniewski whom here I thank cordially and respectfully for all that he
has ever taught me.16 [15, pp. 9–10]
Add that Leśniewski himself valued his collaboration with Kotarbiński. He admitted
that he owed him a lot (see for example [36, p. 93]).
The source of Kotarbiński’s reism were his doubts concerning the existence of
properties and other ideal objects. He expressed them for the first time in his paper
“Sprawa istnienia przedmiotów idealnych” [The problem of existence of ideal objects]
(cf. [11]). He criticized there conceptions assuming the existence of ideal objects. He
wrote that there were no foundations to assume the existence of such objects. He tried to
show that there were no imaginary (only conceivable) objects, no mathematical objects;
there were no types (universals), features, relations, intentional objects, thinking processes
and psychological contents.
Reism was explained by Kotarbinski in his book Elementy [Elements] (cf. [12] and
[14]) and in various papers. Reism in the ontological sense can be reduced to the following
two theses: (1) every object is a thing, (2) no object is a state, a relation, a feature.17
Kotarbiński assumes also that things are bodies, and thus extensive beings existing
in time and space. Therefore, we are dealing with somatism strengthened to become
pansomatism—there are only bodies. This distinguishes reism from other concretisms,
for example from the concretism of Leibniz who towards the end of his life assumed that
there were only concrete entities (note that this concretism was of spiritualistic nature
because those concrete entities were spiritual monads). Reism can be seen as a certain
interpretation of Leśniewski’s ontology (the latter was not a reist although he was a
nominalist).
Reism faces various difficulties when applied to logic and mathematics. Using the
language of reism one can speak about sets in a distributive sense that is fundamental
for set theory, on which in turn the whole building of mathematics is constructed, but
only providing that those statements refer to the elements of these sets. Hence it allows
us to develop the elementary algebra of sets but not to define, for instance the concept
of finite or infinite set. However, it is not sufficient for mathematics. Leśniewski was
aware of these difficulties and proposed to use the concept of a set in a collective sense
(mereological)—such an approach does not allow realising all that mathematicians expect
of set theory. It should be added that reism had numerous followers, the greatest one being
16 Najwi˛ecej wszelako nauczyłem si˛e od prof. dra Stanisława Leśniewskiego. W wielu miejscach ksiażki ˛
wyraźnie z tego zdaj˛e spraw˛e. Ale to sa˛ punkty najważniejsze i najwyraźniejsze. Poza tym, przyznaj˛e, cała
myśl moja przesycona jest do gł˛ebi wpływami tego niezwykłego umysłu, z którego bezcennych darów los
przychylny pozwolił mi przez szereg lat korzystać w obcowaniu niemal codziennym. Jestem niewatpliwie ˛
uczniem kolegi Leśniewskiego, któremu na tym miejscu serdecznie i z gł˛ebokim szacunkiem dzi˛ekuj˛e za
wszystko, czego mnie kiedykolwiek nauczył [14, pp. 9–10].
17 A clear reference to the four categories proposed by W. Wundt can be seen here.
654 R. Murawski
Alfred Tarski.18 Furthermore, reism, thanks to its logical tools, allows achieving more
than any other nominalism.
In the ontology of mathematics Kotarbiński proclaimed himself in favour of nominal-
ism. In Elementy he wrote:
In this variety of opinions, let us single out, and declare for, the position of nominalism. [. . . ]
no object is a number, and [. . . ] neither arithmetic, nor the theory of numbers, nor—a fortiori—
mathematics in general build statements which might strictly be called statements about numbers
in the same sense in which zoology makes statements about animals.19 [15, p. 317]
Mathematics speaks about all things—and hence, its universality.

Kotarbiński—firmly refuting the conception that mathematics investigates a certain
world of ideal objects dependent on time, space and cognitive mind—did not follow any
concrete conception. He stated that mathematics can be characterised in at least three
ways:
(1) as the body of systems in which theorems are justified only in a deductive way and
‘the theorems are formulated correctly as statements containing only the following
types of signs—variables, connectives, what are called ‘names of numbers”, ‘names
of sets”, ‘names of figures”, or terms defined by such signs, names of relations (such
as ‘greater than”, ‘equal to”, etc,) and finally punctuation marks and signs informing
about the role of the remaining signs’ [15, p. 322]20—mathematics thus understood
embraces the whole formal logic (in its propositions these ‘names’ do not occur) and
the so-called proper mathematics;
(2) as proper mathematics or mathematics in a narrower sense, which is characterised by
the fact that those ‘names’ occur in its thesis;
(3) as a science that is characterised like proper mathematics but adding the condition
that its propositions have the feature of apriority, i.e., its axioms are assigned the
feature of obviousness, and justifying its theorems we do not refer to empirical data.
Add that according to Kotarbiński nominalism is consistent with the thesis about
aprioristic character of mathematics a science.
18 It is worth quoting the words of Andrzej Mostowski uttered after returning from a conference dedicated
to the foundations of set theory: “Just imagine that there I sighed for reism. The presented conceptions
resulted from so breakneck speculations, so unattainable for intuition and so incomprehensible that reism
seemed to be an oasis where one can breathe fresh air” (cf. [10, p. 73]).
19 W tym nadmiarze rozmaitych stanowisk niechaj nam wolno b˛ edzie wyróżnić stanowisko nominalizmu
i przy nim si˛e opowiedzieć. [. . . ] żaden przedmiot nie jest liczba˛ i [. . . ] ani arytmetyka, ani tzw. “teoria
liczb”, ani tym bardziej matematyka w ogóle nie buduja˛ zdań, które by można nazwać ściśle zdaniami o
liczbach w tym sensie, w jakim np. zoologia mówi o zwierz˛etach [14, p. 373].
20 których twierdzenia wypowiada si˛ e poprawnie w zdaniach, zawierajacych ˛ tylko nast˛epujace
˛ rodzaje
znaków: symbole zmienne, spójniki, tzw. » nazwy liczb«, tzw. » nazwy zbiorów«, tzw. » nazwy figur«,
lub terminy przez takie znaki zdefiniowane, dalej terminy stosunkowe, jak » wi˛ekszy«, » równy« itp.,
wreszcie znaki przestankowe oraz znaki informujace ˛ o roli pozostałych znaków [14, p. 379].
5 Leon Chwistek: Nominalism
Talking about nominalism one should mention also Leon Chwistek (1884–1944). Though
he did not belong directly to Lvov-Warsaw School, he went always along his own paths
being a “separate” scholar, nevertheless his conceptions were important.
Chwistek declared himself as a nominalist. According to him the subject of deductive
sciences, hence also of mathematics, are expressions constructed in them according to
accepted rules of construction. Hence the subject of mathematics are not ideal objects
like points, lines, numbers or sets. Expressions being subjects of mathematics are physical
objects given us in experience. They can be transformed according to accepted rules. In
every system one accepts such rules as well as some expressions that play the role of
axioms and that form the base on which theorems are deduced. Transformation rules and
axioms are chosen in such a way that expressions could be interpreted as descriptions of
considered states of affairs. To be able to apply deductive theories to specific sciences
and generally to perceive concrete areas of reality, the elements of the latter should be
schematised.
In particular geometry is—according to Chwistek an experimental discipline. In
Chapter VIII of Granice nauki [Limits of Science] he wrote:
Geometry is an experimental science. It depends upon the measurement of segments, angles, and
areas. The Egyptians conceived it in this way and it has remained essentially the same up to this
very day. Today what is generally regarded as geometry, i.e. what is included in textbooks, is the
peculiar mixture of experimental geometry and the geometrical metaphysics which was inherited
from the Greeks as Euclid’s Elements.21 [5, p. 170]
The rise of the systems of non-Euclidean geometry of Bolyai, Gauss and Lobachevsky
in the nineteenth century—regarded by Chwistek as the most important achievement in
exact sciences—abolished in his opinion Kant’s idealism. These geometries showed that,
for example, the concept of a straight line is not of an objective character, but depends
on the accepted axioms. It may suggest that conventionalism is the proper philosophy
for geometry. Indeed, in his first works, e.g. the paper “Trzy odczyty odnoszace˛ si˛e do
poj˛ecia istnienia” [Three Talks Concerning the Concept of Existence] (see [3]), he states
that the existence of systems of non-Euclidean geometry, which are consistent, refutes
the thesis of the a priori character of geometry. It seems that he would tend to accept
conventionalism, although he does not state this explicitly. However in Granice nauki
[4] he explicitly and categorically rejected conventionalism claiming that geometry—
similarly as all other fundamental experimental sciences—should be based on the theory
of expressions. This is because conventionalism introduces hypothetical entities, as was
the case in John Stuart Mill’s works or later Poincaré’s, a promoter of this direction.
Chwistek wrote:
It seems that it is impossible to attain a general concept of geometry without using formulae.
It is therefore clear that the conception of geometry as the science of ideal spatial constructions
21 Geometria jest nauka˛ doświadczalna.˛ Polega ona na mierzeniu odcinków, katów

˛ i powierzchni. Tak
pojmowali ja˛ Egipcjanie i taka˛ pozostała w istocie swojej do dzisiaj. To, co uważa si˛e powszechnie
za geometri˛e za naszych czasów, tj. to, o czym pisze si˛e w podr˛ecznikach, jest osobliwa˛ mieszanina˛
geometrii doświadczalnej i metafizyki geometrycznej, która˛ pozostawili nam w spadku Grecy pod
postacia˛ elementów Euklidesa ([4, p. 190]; see also [6, p. 170]).
656 R. Murawski
must be nullified . . . . To speak of different four-dimensional space-times it is necessary to employ

five-dimensional spacetime. It is clear that all this has only as much meaning as do mathematical
formulae.22 [5, pp. 186–187]
In a similar way as geometry one should treat arithmetic, mathematical analysis

and other mathematical theories obtaining in this way consequently their nominalistic
interpretations.
6 Kazimierz Ajdukiewicz and Ontology of Mathematics
Let us turn back to Lvov-Warsaw School and consider ontological views of Kazimierz
Ajdukiewicz (1890–1963). He considered the ontology of mathematics and logic in his
Habilitationsschrift entitled Z metodologii nauk dedukcyjnych [From the Methodology of
Deductive Sciences] from 1921 (cf. [1]) in which he discussed the problem of existence,
in particular the problem what does it mean “to exist” in deductive sciences. He wrote
there:
An analysis of meaning of the word ‘exist’ as used in deductive theories does not amount to the
problem: what kind of existence is among the attributes of existing objects of deductive theories;
our own position permits us to doubt whether any kind of being at all is among the attributes of
these objects. Our problem then is not the question what kind of being is attributable to objects
under discussion, but the question what is the meaning of the word ‘exist’ as used in deductive
theories. It may be that it is being used quite erroneously and has nothing at all to do with
existence.23 [2, p. 34]
Ajdukiewicz argued in the considered work that the existence in deductive sciences
cannot be identified with consistency and that consistency is neither sufficient nor
necessary condition of existence. He claimed that the necessary conditions of existence
are: (I) being included into the domain of the given theory, and (II) consistency:
My contention is, namely, that for an object p defined by (p) to exist it is necessary that p be an
element of the domain of the given theory, in other words that (p) entailed A(p) [. . . ].
In order to exist an object must, therefore, satisfy another requirement—besides the above
condition of being an element of the domain of the theory—scil. its definition must not have
any consequences inconsistent with the consequences of A(p). [. . . ]
22 Okazuje si˛ e, że dotarcie do ogólnego poj˛ecia geometrii bez formuł jest niemożliwe. Jasne jest, że
idac
˛ ta˛ droga,˛ musimy dojść do unicestwienia geometrii jako nauki o idealnych utworach przestrzen-
nych. [. . . ] żeby mówić o różnych czterowymiarowych czasoprzestrzeniach, musimy si˛e odwołać do
czasoprzestrzeni pi˛eciowymiarowej. Jest jasne, że wszystko to ma tyle sensu, ile zawieraja˛ go formuły
matematyczne [4, pp. 186–187].
23 Analiza znaczenia wyrazu “istnieć” w naukach dedukcyjnych nie jest zatem równoznaczna z zagad-
nieniem: jaki rodzaj istnienia przysługuje istniejacym

˛ przedmiotom nauk dedukcyjnych; problemat nasz
pozwala nam w ogóle watpić ˛ o tym, czy jakikolwiek rodzaj bytu przedmiotom tym przysługuje. Kwestia˛
nasza˛ zatem nie jest pytanie, co za rodzaj bytu maja˛ przedmioty przez nas rozważane, ale co znaczy wyraz
“istnieć” w naukach dedukcyjnych. Być może, że jest on całkiem mylnie używany i nie ma z istnieniem
nic wspólnego [1, p. 46].
Objects which do not satisfy either the first or the second requirement, do not exist, are nonexisting.
From existing and nonexisting objects we ought to distinguish objects which are possible in the
given theory.24 [2, pp. 42–43]
Ajdukiewicz comes to the conclusion that if an object is to exist it must satisfy the
requirements (I) and (II) as well as ‘not restrict the domain of possible objects’25 [2, p.
44] in the given theory. And he concludes his considerations in the following way:
In the deductive sciences we do not speak of existence in absolute sense but only relatively to a
given system. For there exist Euclidean straight lines and non-Euclidean straight lines; however,
both cannot co-exist and their co-existence would be a consequence of their existence if this word
were taken in either case in the absolute sense. We may only speak of existence in a system as we
speak of inclusion in a domain. Nevertheless it is possible to construct a ‘universe’ consisting of the
domains of several compatible theories, thus forming a system whose axioms would be all axioms
of all compatible theories. We could then speak of absolute existence, not quite absolute, though,
since it would be possible by choosing various theories, to construct many such ‘universes,’ self-
compatible but mutually exclusive.26 [2, p. 45]
7 Andrzej Mostowski: Nominalism, Reism, Constructivism
Let us finish our considerations by presenting ontological views of Andrzej Mostowski

(1913–1975). He is usually treated as a representative of the second generation of Lvov-
Warsaw School. He was disciple of Tarski, some influence of Alfred Lindenbaum on him
can be seen.
Mostowski inherited from Tarski general philosophical views, in particular tendency
towards empiricism and apparent respect for nominalism. His sympathies were also, as
it seems, with Kotarbiński’s reism, i.e., the view that there exist only individual physical
things. However he avoided in his logical and mathematical works explicit philosophical
declarations. Nevertheless there are some exceptions from that tendency.
24 Twierdz˛ e mianowicie, że koniecznym warunkiem na to, by przedmiot określony przez (p) istniał,
jest iżby przedmiot p należał do zakresu danej teorii, czyli iżby z (p) wynikało A(p) [. . . ].
Musi tedy przedmiot na to, aby istniał, spełniać prócz pierwszego (wyżej wymienionego warunku
zawierania si˛e) warunek drugi, musi mianowicie jego określenie nie posiadać nast˛epstw sprzecznych z
nast˛epstwami A(p). [. . . ]
Przedmioty, które nie czynia˛ zadość pierwszemu albo drugiemu warunkowi, nie istnieja˛ i sa˛ nieist-
niejace.
˛ Prócz przedmiotów istniejacych ˛ i nieistniejacych
˛ należy jeszcze rozróżnić, naszym zdaniem,
przedmioty możliwe w danej teorii [1, pp. 59–60].
25 nie ograniczał [on] zakresu przedmiotów możliwych [1, p. 62].
26 O istnieniu bezwzgl˛ ednym w naukach dedukcyjnych nie mówimy wcale. Zawsze tylko o istnieniu
w pewnym systemie. Wszakże istnieja˛ i proste euklidesowe, i nieeuklidesowe, obie nie moga˛ jednak
współistnieć, a współistnienie ich byłoby konsekwencja˛ ich istnienia, gdyby ten wyraz wziać ˛ w
odniesieniu do obu w tym samym sensie bezwzgl˛ednym. Można wi˛ec mówić tylko o istnieniu w pewnym
systemie, podobnie jak o zawieraniu si˛e tylko w pewnym zakresie. Niemniej jednak można utworzyć
“uniwersum” z zakresów kilku zgodnych z soba˛ teorii, tworzac ˛ system, którego aksjomaty byłyby
wszystkimi aksjomatami wszystkich teorii zgodnych. Można by wtedy mówić o istnieniu bezwzgl˛ednym,
jakkolwiek niezupełnie bezwzgl˛ednym, bo można by, dobierajac ˛ rozmaite teorie, potworzyć wiele takich
“uniwersów” w sobie zgodnych, lecz mi˛edzy soba˛ wykluczajacych ˛ si˛e [1, p. 63].
658 R. Murawski
In the paper “A Classification of Logical Systems” [29] Mostowski assumes a certain

philosophical presumption on the analysed logical systems. Declaring at the beginning
of his work that “The subject itself as well as the method of its presentation will be of a
mathematical rather than philosophical character” [29, p. 245] he openly states:
Although our investigations will be purely formal we shall nevertheless accept a definitive
philosophical point of view with respect to logical systems. We shall not consider logical systems
as void schemata deprived of any interpretation. On the contrary we shall assume the objective
existence of a kind of » mathematical reality« (e.g. of the set of all integers or the set of all real
numbers). By objective existence we mean existence independently of all linguistic constructions.
[29, pp. 246–247]
The task of logical systems is—according to Mostowski—just to describe that

“mathematical reality”. Consequently, every sentence of logic is equipped with a certain
meaning—it says that mathematical reality is entitled to have this or that property.
The fact that there exist true sentences that are unprovable in a given system—what
is a consequence of Gödel’s incompleteness theorems—can be explained by the bigger
complexity of the properties of this “mathematical reality” than the complexity of the
properties that can be deduced from axioms by the accepted inference rules. Mostowski
concludes in a characteristic way:
We do not intend to defend the philosophical correctness or even the philosophical acceptability
of the point of view here described. It is evident that it is entirely opposite to the point of view of
nominalism and related trends. [29, pp. 247]
One can easily see here certain tension between aforementioned inclinations towards
nominalism and his concrete logical and mathematical investigations.
Philosophical questions, in particular ontological ones, appear in Mostowski’s papers
devoted to set theory. Considering Gödel’s and Cohen’s results on consistency and
independence of Axiom of Choice and Continuum Hypothesis Mostowski claimed that
they can be treated as “one of the most important arguments against mathematical
Platonism” [33, p. 176]. After Gödel’s and Cohen’s results it is possible to construct
consistent but mutually inconsistent set theories. If such theories are constructed “we
shall be forced to admit that in the match between Platonism and formalism the latter has
again scored one point” [33, p. 182]. Since metamathematical results on set theory do
not provide decisions concerning the way of existence of sets, and in general of objects
of mathematics, and consequently the controversy between formalism and Platonism,
Mostowski concludes in the paper “Sets” [34]:
Whatever the final outcome of the fight between these two opposing trends will be, it is obvious
that we should concentrate on the study of concepts which seem perfectly clear and perspicuous
to us. [34, p. 28]
At another place however he wrote that “the ultimate formulation of axioms of set
theory should be preceded by a discussion of the fundamental assumptions of this theory,
including the constructive standpoint” [30, p. 20].
Mostowski had a keen interest in constructivism. However it should be added that he
was more interested in its aims than in proposed solutions (see for example [31, p. 192]).
In the monograph Logika matematyczna [Mathematical Logic] [28] he wrote even:

I am inclined to think that a satisfactory solution of the foundations of mathematics will happen on
the way shown by constructivism or a similar direction. However, one cannot now write a textbook
of logic on this basis.27 [28, p. VI]
Mostowski saw advantages of constructivism in the fact that:

[. . . ] it wants to inquire into the nature of mathematical entities and to find a justification for
the general laws which govern them, whereas Platonism takes these laws as granted without any
further discussion. [31, p. 192]
According to Mostowski constructivistic conceptions are closer to nominalism than

to Platonism. This implies that constructivism does not accept general concepts of
mathematics as being given but attempts to construct them. “This leads to the result that
one can identify mathematical concepts with their definitions” [31, p. 178]. In arithmetic
constructivism allows us to give up assuming actual infinity or to use solutions requiring
only the nominalist approach. Whereas one of the advantages of nominalism is that many
important mathematical theories have been satisfactorily reconstructed on the nominalist
basis, and these reconstructions have turned out to be equivalent to the classical theories.
Mostowski was aware of some limitations of constructive methods in mathematics
and of the fact that they do not suffice (cf. for example [34, pp. 29–32]). Nevertheless
he investigated principles of constructivism. He was of the opinion that sometimes
constructivism is philosophically more satisfactory—this is the case of arithmetic or of
applied mathematics where it seems to reveal new promising perspectives.
It should be noticed that Mostowski considered constructivism in a way connected
with the classical point of view. Hence he was not connected with pure constructivism of
Brouwer, Heyting and other intuitionists. In fact he represented rather certain combination
of constructivism and set-theoretical program. This combination formed according to him
the base on which the foundations of mathematics should be developed.
8 Conclusion
The described panorama of ontological views concerning logic and mathematics of

representatives of Lvov-Warsaw School shows that one does not find here any dominating
position, just opposite, there is a full range of positions from Platonism through
constructivism till nominalism. What was the reason of that? First of all it should
be stressed that philosophical views were formulated in Lvov-Warsaw School on the
margin of proper logical or metamathematical investigations. One was convinced that
mathematical and logical investigations should not bounded by any a priori philosophical
assumptions. Mathematics and logic should be autonomous and neutral with respect to
philosophy. Hence opinions about philosophical aspects of logic or mathematics were
rather fragmentary and incomplete, they concerned first of all particular issues connected
with problems actually studied. Formulated remarks were often simply comments to
27 Jestemskłonny mniemać, że zadowalajace ˛ rozstrzygni˛ecie zagadnienia podstaw matematyki nastapi

˛ na
drodze wskazanej przez konstruktywizm lub kierunek do niego zbliżony. Na tej jednak podstawie nie
można by już teraz napisać podr˛ecznika logiki.
660 R. Murawski
concrete technical results from the foundations of mathematics or logic. Exceptions were
here Leśniewski and Chwistek whose logical investigations were a consequence and result
of some philosophical considerations.
Philosophical views were generally treated as private matters. They should not bound
the research activity in logic, mathematics or the foundations of mathematics and during
the very investigations of concrete mathematical or logical problems should be suspended.
Moreover it happed that declared philosophical views were in fact not compatible with
research practice. It can be easily seen for example by Tarski who declared himself
as nominalist but simultaneously in his research practice used without any restrictions
infinitistic methods far from what has been accepted by nominalists.
References
1. Ajdukiewicz, K.: Z metodologii nauk dedukcyjnych. Nakładem Polskiego Towarzystwa Filo-

zoficznego, Lwów (English translation see Ajdukiewicz, K.: From the Methodology of the Deductive
Sciences. Stud. Logica 19, 9–45 (1966))
2. Ajdukiewicz, K.: From the Methodology of the Deductive Sciences. Stud. Logica 19, 9–45 (1966)
3. Chwistek, L.: Trzy odczyty odnoszace ˛ si˛e do poj˛ecia istnienia. Przeglad˛ Filozoficzny 20, 122–151
(1917). (Reprinted in: Chwistek, L.: Pisma filozoficzne i logiczne, Tom 1. Państwowe Wydawnictwo
Naukowe, Warszawa (1961), pp. 3–29)
4. Chwistek, L.: Granice nauki. Zarys logiki i metodologii nauk ścisłych. Ksiażnica-Atlas,
˛ Lwów-
Warszawa (1935) (Reprinted in: Chwistek, L.: Pisma filozoficzne i logiczne, Tom 2. Państwowe
Wydawnictwo Naukowe, Warszawa (1963), pp. 1–232; English translation see Chwistek, L.: The
Limits of Science. Outline of Logic and of the Methodology of the Exact Sciences. Kegan Paul,
Trench Trubner and Co, New York/London (1948). Translated by H.Ch. Brodie and A.P. Coleman
(Reprinted by Routledge, London (2000)))
5. Chwistek, L.: The Limits of Science. Outline of Logic and of the Methodology of the Exact Sciences.
Kegan Paul, Trench Trubner and Co, New York/London (1948). Translated by H.Ch. Brodie and A.P.
Coleman (Reprinted by Routledge, London (2000))
6. Chwistek, L.: Pisma filozoficzne i logiczne, Tom 1. Państwowe Wydawnictwo Naukowe, Warszawa
(1961)
7. Chwistek, L.: Pisma filozoficzne i logiczne, Tom 2. Państwowe Wydawnictwo Naukowe, Warszawa
(1963)
8. Dambska
˛ I.: Czterdzieści lat filozofii we Lwowie. In: Pi˛ećdziesiat˛ lat filozofii w Polsce. Przeglad
˛
9. Feferman, A.B., Feferman, S.: Alfred Tarski. Life and Logic. Cambridge University Press, Cambridge
(2004)
10. Kotarbińska, J.: Głos w dyskusji. Studia Filozoficzne 5 (1984)
11. Kotarbiński, T.: Sprawa istnienia przedmiotów idealnych. Przeglad ˛ Filozoficzny 23, 149–170 (1920)
12. Kotarbiński, T.: Elementy teorji poznania, logiki formalnej i metodologji nauk. Wydawnictwo
Zakładu Naukowego im. Ossolińskich, Lwów (1929)
13. Kotarbiński, T.: Zasadnicze myśli pansomatyzmu. Przeglad ˛ Filozoficzny 38, 283–294 (1935)
14. Kotarbiński, T.: Elementy logiki, teorii poznania, logiki formalnej i metodologii nauk, 2nd edn. Os-
solineum, Wrocław-Warszawa-Kraków (1961) (3rd edition: Ossolineum, Warszawa (1986); English
translation see Kotarbiński, T.: Gnosiology: The Scientific Approach to the Theory of Knowledge.
Pergamon, Oxford/Warszawa, 1966)
15. Kotarbiński, T.: Gnosiology: The Scientific Approach to the Theory of Knowledge. Pergamon,
Oxford/Warszawa (1966)
16. Leśniewski, S.: O podstawach matematyki. Wst˛ep oraz §§1–3. Przeglad ˛ Filozoficzny 30, 164–206
(1927)
17. Leśniewski, S.: Grundzüge ge eines neuen System der Grundlagen der Mathematik. Fundamenta
Mathematicae 14, 1–81 (1929)
18. Leśniewski, S.: Collected Works. PWN (Polish Scientific Publishers) and Kluwer Academic Publish-
ers, Dordrecht-Boston-London (1992). Edited by S.J. Surma, J.T. Srzednicki, I. Barnett, V.F. Rickey
19. Leśniewski, S.: Czy prawda jest wieczna czy teź wieczna i odwieczna?. Nowe Tory 18, 493–528
(1913)
20. Leśniewski, S.: Krytyka logicznej zasady wyłaczonego
˛ środka. Przeglad
˛ Filozoficzny 16, 315–352
(1913) (Reprint (with certain omissions) in: Jadacki, J.J., Bigaj, T., Lissowska, A. (eds.) Co istnieje?,
vol. 1, pp. 143–167. PETIT, Warszawa (1996))
21. Łukasiewicz, J.: Logika a psychologia. Przeglad ˛ Filozoficzny 10, 489–491 (1907)
22. Łukasiewicz, J.: O twórczości w nauce. In: Ksi˛ega pamiatkowa
˛ ku uczczeniu 250-tej rocznicy założe-
nia Uniwersytetu Lwowskiego przez Króla Jana Kazimierza r. 1661, vol. 1, pp. 1–15. Uniwersytet
Lwowski, Lwów (1912) (Second revised and enlarged edition in: Poradnik dla samouków. Wskazówki
metodyczne dla studjujacych
˛ poszczególne nauki. New edition, Vol. 1, pp. 15–39. Wydawnictwo A.
Heflera i St. Michalskiego, Warszawa (1915)
23. Łukasiewicz, J.: Logistyka a filozofia. Przeglad˛ Filozoficzny 39, 115–131 (1936)
24. Łukasiewicz, J.: W obronie logistyki. In: Myśl katolicka wobec logiki współczesnej. Studia Gnesnen-
sia 15, 12–26, 159–165 (1937)
25. Łukasiewicz, J.: Selected Works. North-Holland Publishing Company, Amsterdam-London and PWN
(Polish Scientific Publishers), Warszawa (1970). Edited by L. Borkowski
26. Mancosu, P.: Harvard 1940–1941: Tarski, Carnap and Quine on a finitistic language of mathematics
for science. Hist. Philos. Log. 26, 327–357 (2005)
27. Mancosu, P.: Tarski’s Engagement with Philosophy. In: Lapointe, S., Woleński, J., Marion, M.,
Miskiewicz, W. (eds.) The Golden Age of Polish Philosophy. Kazimierz Twardowski’s Philosophical
Legacy, pp. 131–153. Springer, Dordrecht-Heidelberg-London-New York (2009)
28. Mostowski, A.: Logika matematyczna. Kurs uniwersytecki. Monografie Matematyczne, Warszawa-
Wrocław (1948)
29. Mostowski, A.: A Classification of Logical Systems. Stud. Philos. 4, 237–274 (1949–1950) (pub-
lished in 1951) (Reprinted in: Murawski, R.: Filozofia matematyki i logiki w Polsce mi˛edzywojennej.
Monografie Fundacji na rzecz Nauki Polskiej, Wydawnictwo Naukowe Uniwersytetu Mikołaja
Kopernika, Toruń (2011), vol. II, pp. 154–191)
30. Mostowski, A.: The Present State of Investigations of the Foundations of Mathematics. Rozprawy
Matematyczne 9, 1–48 (1955). In collaboration with: A. Grzegorczyk, S. Jaśkowski, J. Łoś, S. Mazur,
H. Rasiowa and R. Sikorski
31. Mostowski, A.: On Various Degrees of Constructivism. In: Heyting, A. (ed.) Contructivity in
Mathematics. Proceedings of the Colloquium held in Amsterdam, 1957, pp. 178–194. North-Holland
Publishing Company, Amsterdam (1959)
32. Mostowski, A.: Tarski Alfred. In: Edwards, P. (ed.) The Encyclopedia of Philosophy, vol. 8, pp. 77–
81. The Macmillan Company and the Free Press, New York/Collie-Macmillan, London (1967)
33. Mostowski, A.: Niesprzeczność i niezależność hipotezy kontinuum. Roczniki Polskiego Towarzystwa
Matematycznego, Seria II: Wiadomości Matematyczne 10, 175–182 (1968)
34. Mostowski, A.: Sets. In: Scientific Thought. Some Underlying Concepts, Methods and Procedures,
pp. 1–34. Mouton/UNESCO, The Hague (1972)
35. Murawski, R.: Filozofia matematyki i logiki w Polsce mi˛edzywojennej. Monografie Fundacji na rzecz
Nauki Polskiej, Wydawnictwo Naukowe Uniwersytetu Mikołaja Kopernika, Toruń (2011)
37. Tarski, A.: Collected Papers, vols. 1–4. Birkhäuser, Basel-Boston-Stuttgart (1986). Eds. S.R. Givant
and R.N. McKenzie
38. Woleński, J.: Filozoficzna szkoła lwowsko-warszawska. Państwowe Wydawnictwo Naukowe,
Warszawa (1985)
Dordrecht-Boston-London (1989)
R. Murawski ()
Adam Mickiewicz University in Poznań, Faculty of Mathematics and Computer Science, Poznań, Poland
Part III
The War and Post-War Period
A View of Revival of Mathematical
Logic in Warsaw, 1945–1975
Victor W. Marek
To the memories of my teachers: Andrzej Mostowski, Helena Rasiowa,

and Andrzej Grzegorczyk
Abstract This is a (biased) account of the revival of Foundations of Mathematical Logic,

and more generally Foundations of Mathematics after WWII. The perspective is limited
to years 1945–1975. This coincided with the activities of Professor Andrzej Mostowski,
the principal leader of the foundational research in Poland during the reported time.
Moreover, in this text, we focus on the development of Foundations in Warsaw, which
misses an important aspect of the foundational research in Poland of the time, specifically
in Wrocław. Even the description of the Warsaw efforts is incomplete, not covering
Foundations of Geometry. We focus on activities of three main researchers: Andrzej
Mostowski, Helena Rasiowa and Andrzej Grzegorczyk.
Keywords Logic in Warsaw · Andrzej Mostowski · Helena Rasiowa · Andrzej

Grzegorczyk
Mathematics Subject Classification (2000) Primary 01A60, 03-03
1 Introduction
World War II separates significant achievements of mathematical community of Poland

during the so-called “Second Polish Republic” (1918–1939) and very different “Polish
Peoples Republic” that was created after liberating Poland from German occupation.
The war resulted in significant territorial changes. In effect, Poland was moved some
200 miles to the west, loosing significant part of its Eastern part and gaining territories
to the West of the pre-war borders that had previously been German. Poland lost two
significant academic centers: Lwów, (now Lviv in Ukraine) and Wilno (now Vilnius in
Lithuania). Instead, Poland gained important centers of science: Wrocław (previously
German Breslau) as well as Gdańsk (previously a free city). Significant part of Prussia
became part of Polish territory.

666 V. W. Marek
Science in Poland, and in particular Mathematics (thus also Mathematical Logic)

suffered greatly as a result of German occupation and, more generally, World War II. The
losses were both due to extermination of intellectuals by Germans (especially of those
with Jewish roots) and emigration. Logicians such as Adolf Lindenbaum, Mordechaj
Wajsberg, Mojzesz Presburger, and others were killed by Germans. Several important
logicians, in particular Alfred Tarski, Jan Łukasiewicz, Bolesław Sobociński and others
left Poland and emigrated as the result of the war and continued their work in foreign
institutions.
There was a conscious effort to re-create Jan Kazimierz University of Lwów in
Wrocław, and to some extent, Lwów mathematics, with Hugo Steinhaus continuing
Lwów traditions. Likely the result of WWII, the death of Stefan Banach in 1945, in a
relatively early age of 53, deprived Poland of one of its great mathematicians open to
many foundational challenges.
2 The Beginnings
It is very hard to imagine the amount of destruction both of Warsaw and Wrocław as the
result of WWII. The western part of Warsaw (where the Warsaw University was, and is
now) was destroyed in 95%. This author, born in the second half of WWII, remembers
well the sea of ruins in parts of Warsaw. This was the result of a systematic destruction
of that part of Warsaw by Germans after the fall of Warsaw Uprising (August–October
1944.) Wrocław, converted by Nazis into a fortress, was similarly destroyed. Both places
were, essentially, human deserts.
During the WWII, there were several underground Universities in Warsaw, functioning
in spite of prohibition by the occupiers of both secondary and university education. At
a significant risk, the principal Warsaw mathematicians (Sierpiński, Kuratowski, and
others) were conducting lectures and doing research. All of them left Warsaw with the
rest of the population when Germans started their destruction of the city.
Among the other mathematicians active in the underground Warsaw University was
Andrzej Mostowski, a student of Lindenbaum, Tarski, Gödel and Kuratowski. Mostowski
conducted an active research during WWII (a story of his “black notebook” containing
the results of his research and the alternative he faced as he was leaving Warsaw in 1944
is told in several publications devoted to his life). Mostowski, along with the rest of
population of Warsaw had to leave the town. Eventually, he found his way to Kraków. In
1945, Mostowski received his habilitation degree at Jagiellonian University. Mostowski
returned to Warsaw in 1946, and was associated with the revived Warsaw University till
the end of his life in 1975.
The faculty of Warsaw University with interests in widely understood Logic and
Foundations, Kazimierz Kuratowski and Waclaw Sierpiński, while still active, had
principal interests in a variety of areas, with Kuratowski mainly interested in Topology,
and Sierpiński in Number Theory and Combinatorial Set Theory. Both of them had
residual interest in Foundations, but it was not the principal area of their research. In
this situation, the fundamental role in rebuilding foundational research in Poland fell
on Mostowski, who, eventually, educated a large group of students who helped him
A View of Revival of Mathematical Logic in Warsaw, 1945–1975 667
in returning Warsaw to one of the prime centers in foundational research. Already in

1946, both Helena Rasiowa and Wanda Szmielew started their research in logic (both
studied in the underground Warsaw University during German occupation). Eventu-
ally, both Professor Rasiowa and Professor Szmielew developed interests in areas not
directly related to Professor Mostowski research (with Rasiowa mainly interested in
Algebraic Foundations of Logic, and Szmielew working in Foundations of Geometry).
Both Rasiowa and Szmielew educated groups of logicians following the directions of
research outlined above. Additionally, Professor Rasiowa interests were later influenced
by Computer Science applications of Logic. The results of Professor Szmielew initially
developed during German occupation and then continued during her stay at the University
of California, Berkeley, 1949–1950, (decidability of the theory of Abelian groups) were
instantly recognized as a major result. Even earlier, 1948–1949, Professor Mostowski
spent a year as a visitor at the Institute of Advanced Studies, in Princeton, NY.
Mostowski’s relation with Gödel (since Mostowski’s extended visit in Vienna in 1936)
played an important role in Mostowski’s research in several areas of Foundations,
specifically in the metamathematics of set theory and related formalisms.
3 Andrzej Mostowski and His Students
The most amazing aspect of Mostowski’s activities during the first 15 years after WWII is
the breadth of his interests as reflected in his publications. He found significant results in
Recursion Theory, Set Theory, Proof Theory, and Model Theory. This needs to be related
to the fact that the notebooks that he wrote during WWII were lost (Germans allowed
inhabitants of Warsaw expelled from the city to take one kilogram of items, and he (as
stated to this author) chose a kilogram of bread). Reconstruction of the results and their
proofs took time, and not all was remembered. Still, results on independence of various
set-theoretic statements, results on existence of models of arithmetical theories, results
on the arithmetical hierarchy, on proof theory of intuitionistic logic and results on the
incompleteness of Peano arithmetic and its extensions (culminating in the monograph
with R. Robinson and A. Tarski) all followed in quick succession. The handbook on
Mathematical Logic (in Polish, 1948) was a powerful tool in educating generations
of Polish logicians.1 Almost every year in the 1950s Mostowski was publishing work
that was widely read, commented, and even more importantly further expanded by
mathematicians all over the world. It is enough to mention his work on the direct
product of theories (significantly expanded by Solomon Feferman and Robert Vaught),
generalized quantifiers (which pioneered the area of significant importance both in
mathematical logic and (much later) computer science, model theory of second order
arithmetic (together with Andrzej Grzegorczyk and Czesław Ryll-Nardzewski), and one
of most important tools of model theory—existence of models with indiscernibles (joint
work with Andrzej Ehrenfeucht). This created the opportunity for many individuals to
study with Professor Mostowski.2
1I believe that one of the articles in the present volume discusses this monograph.
2 The author of the present article was Mostowski’s Ph.D. student in 1964–1968.
668 V. W. Marek
It is safe to say that by the end of 1950s, Warsaw, and more generally, Poland,
returned to its place as one of most important centers of research in the Foundations
of Mathematics. Moreover, by that time, a cadre of researchers both within direct circle
of Mostowski, as well as circles of his former students, such as Andrzej Grzegorczyk,
Helena Rasiowa and Wanda Szmielew was active in Warsaw. Each of these individuals
(and, of course, Professor Mostowski himself) directed a seminar at Warsaw University.
For that reason, the access to the knowledge in the areas of Foundations was easy both for
students and faculty.
The unusually wide interests of Mostowski continued into 1960s. The breakthrough in
the area of independence proofs in Set Theory due to Paul J. Cohen using the technique
of forcing influenced Mostowski and his group of collaborators. Mostowski devoted to
this topic a monograph that presented Cohen arguments in a new perspective. Even more
importantly, in a series of lectures at Helsinki (subsequently published as a monograph),
Mostowski presented a unified perspective of Foundations of Mathematics of the middle
of 1960s. It established him as one of the leaders and visionaries of the area.
Continuing exploration of other areas of foundations, models of second order arith-
metic, impredicative theories of classes, abstract model theory, weak systems of arithmetic
and other current areas of research, Mostowski constantly expanded the perspective of
research in Warsaw. A number of scientific meetings in Warsaw, culminating with the
Foundations of Mathematics Semester at the Banach Center in 1973, firmly established
Warsaw as a place worth visiting. During 1960s and first half of 1970, there was a constant
stream of visiting logicians benefiting of Mostowski’s hospitality and the cutting edge of
foundational research in Warsaw.
Professor Mostowski’s death in 1975, at age 62, was a significant blow to foundational
work in Warsaw, and more generally, in Poland. Additionally, later on, important political
disturbances, including martial law, resulted in further losses in the foundational research
in Warsaw.
A most comprehensive and reasoned bibliography of Professor Andrzej Mostowski is
provided by Jan Zygmunt [12] and supersedes earlier efforts by students of Mostowski.
The bibliography lists over 110 papers and their translations including papers published
posthumously) as well as over 15 monographs and handbooks. Together with a large
number of reviews of papers, this bibliography presents the enormous body of highest-
class research and other contributions to Foundations.
At present, Professor Mostowski has 18 direct descendants in the Mathematical
Genealogy Database as well as 300 indirect descendants. Since some Mostowski’s
students and their descendants moved to Computer Science and Philosophy, the number
of indirect descendants is, clearly, bigger.
4 Helena Rasiowa and Her Group
Professor Helena Rasiowa started her Mathematics studies in 1938. The World War
II, with its consequences for Polish citizens (with dual occupation at first, and then
suffering under German occupation) limited opportunities for individuals interested in
higher education to “underground University”. Rasiowa was studying in that institution,
and wrote her M.Sc. thesis under the direction of Łukasiewicz and Sobociński. She
described to her younger collaborators (this author included) the horrible experience
of being buried under ruins of a house bombed during the Warsaw Uprising of 1944.
As the rest of the population of Warsaw she was expelled of the city, and survived
as a secondary school teacher. Andrzej Mostowski brought her back to the Warsaw
University where she started her Ph.D. studies under his guidance. Her M.Sc. thesis was
reconstructed (recall Mostowski’s experience mentioned above). Following Mostowski’s
work on algebraic methods of logic (the area 100 years old at the time, already), Rasiowa
developed algebraic techniques for modal and intuitionistic logics. Soon, jointly with
Roman Sikorski, she developed a comprehensive theory of algebraic methods in logic.
Several papers in this area resulted in habilitation at the Mathematical Institute of the
Academy of Sciences. The essence of that theory consisted in expanding the notion
of semantics to one where values form a Boolean Algebra (or Heyting algebra, in
intuitionistic case). A fundamental result of this area, often called Rasiowa-Sikorski
Lemma proves the existence of filters closed under denumerable family of meets. This
result provides an alternative means to prove completeness theorem for predicate calculus
and is Baire Theorem in disguise. It also explains why Cohen’s technique of forcing
works (although Rasiowa and Sikorski did not take an opportunity to provide this,
maybe the most spectacular, application of their techniques in their famous monograph
“Mathematics of Metamathematics”).
The work on the algebraic methods in logic continued throughout the entire further life
of Rasiowa. She created a large group of collaborators who worked on algebraic methods
in a variety of logics, including logics that stemmed of application of formal methods in
Computer Science.
Professor Rasiowa was very active in the science administration, both at Warsaw Uni-
versity (she served as a Dean of Mathematics and Physics, later Mathematics, Mechanics
and Computer Science for many years) and international scientific institutions. She helped
to establish Studia Logica as a leading journal on the borderline of Mathematics and
Philosophy, and Fundamenta Informaticae (together with Zdzisław Pawlak).
Beginning in 1960s Professor Rasiowa promoted foundational issues in Computer
Science; her contributions to the logic of programs (algorithmic logic) formed a widely-
read text published by the Institute of Computer Science of Polish Academy of Sciences.
Professor Rasiowa supervised 19 PhD dissertations. At present she has 12 direct
descendants in the Mathematical Genealogy Database as well as 93 indirect descendants.
Since some (in fact most) of her students and their descendants moved to Computer
Science, the number of indirect descendants is, clearly, bigger.
The paper by Bartol, Orłowska and Skowron includes the full list of Rasiowa’s work.
5 Andrzej Grzegorczyk
Of the four main leaders of mathematical logic, and more generally, foundations of
mathematics in post 1945 Poland, the youngest researcher was Professor Andrzej
Grzegorczyk (1922–2014).
670 V. W. Marek
Professor Grzegorczyk studied (like Wanda Szmielew and Helena Rasiowa) at under-
ground universities in Warsaw where he studied chemistry, physics and philosophy. He
completed his studies in Cracow, immediately after the war; his MA thesis generalized
Leśniewski’s system to higher types. His interests at the time were motivated by
philosophical questions—and involvement in widely understood philosophy continued
throughout his entire scientific career. After WWII, Grzegorczyk turned to mathematical
logic, with his Ph.D. (written under the direction of Andrzej Mostowski) devoted to formal
topology. Several important contributions to Foundations of Mathematics followed. Of
these, the hierarchy of primitive-recursive functions is commonly called Grzegorczyk
hierarchy. The paper on the hierarchy can be treated as one of the first research papers on
computational complexity of recursive functions. In 2003 Grzegorczyk proposed an new
proof for Gödel’s undecidability result. The proof omits arithmetization. Grzegorczyk’s
proof is much simpler than the original Gödel’s proof. Other important contributions
involved modal logic (with the logic commonly called Grz, Leśniewski’s mereology
and λ-calculus). Joint work with Mostowski and Ryll-Nardzewski on ω-models of
second-order arithmetic established fundamental properties of these structures, creating
foundations for significant body of work in Warsaw.
Professor Grzegorczyk, like many philosophically-oriented logicians was concerned
with the issues of syntax and definability of fundamental notions. This, to some extent,
resulted in tensions between Professor Grzegorczyk and other researchers in foundations,
esp. mathematicians. The point was that mathematicians proved correctly results over
the millennia, and these results, and more importantly, their proofs are still valid. This
results in a certain attitude among mathematicians (“this is philosophy, not mathematics”)
which alienates some philosophical thinkers. Eventually, Professor Grzegorczyk left
mathematics (at least formally) transferring to the Institute of Philosophy and Sociology
of the Academy of Sciences.
Of the researchers of foundations of older generation, Professor Grzegorczyk was
the most concerned with the questions of ethics. A rigid set of values and opinions,
often going “against the grain” distinguished Grzegorczyk among his peers. During the
“real socialism” he was willing to protest against injustices (this author benefited from
Grzegorczyk’s help and encouragement when subjected to repressions), but was willing
to speak against popular opinions when he felt that these bring more harm than societal
benefit. As a result, Grzegorczyk stood out among many, but “not with many”.
The complex nature of the research of Professor Grzegorczyk (firmly placed on the
intersection of Mathematical Logic and Philosophy) resulted in some controversies, esp.
from the mathematical side and critiques of his fundamental contribution to Foundations
of Mathematics, Outline of Mathematical Logic.
Mathematical Genealogy Projects lists one direct descendant of Professor Grzegorczyk
and 6 indirect descendants. Since the unique descendant listed in that project is a
Computer Scientist, the number of descendants is, likely, bigger. The influence of
Professor Grzegorzyk on the foundational research in Philosophy was also significant.
More details and references to Grzegorczyk’s detailed bibliography can be found in
the paper by Krajewski in this volume.
6 Conclusions
The picture of revival of the Warsaw School of Foundations after the WWII that we
present in this notes, raises a variety of issues. The first, and the most important one refers
to the role of an individual in the scientific process. Certainly, the developments in Warsaw
after 1945, were dominated by the contributions and the personality of Professor Andrzej
Mostowski. The two other individuals treated in this text, Professor Helena Rasiowa and
Professor Andrzej Grzegorczyk were able to contribute to the area of Foundations, but
in the reported period, practically all foundational research in Warsaw (possibly with the
exception of the work by Professor Wanda Szmielew group) was originally motivated
by Professor Mostowski. He not only conducted his own research and motivated his
collaborators, but also initiated research of both Professors Rasiowa and Grzegorczyk.
This raises the question of dependence of such scientific events (revival of a scientific
school) on the individuals. Of course, it is difficult to hypothesise, but what would happen
if Professor Mostowski did perish during the German occupation? Would the Warsaw
school revive?
While one may wonder if such questions are meaningful, the many disasters during
the second half of twentieth century and also more recent violent developments in various
parts of the world raise the issue on the dependence of scientific “schools” on the
availability and contributions of individuals. Maybe the events and developments of the
Warsaw School of Foundations tells us something and suggest blueprint for revival in
analogous situations.
Acknowledgements The author is grateful to Andrzej Jankowski, Stanisław Krajewski, Marcin

Mostowski and Andrzej Skowron for their corrections and other helpful suggestions.
References
Andrzej Mostowski
1. Axiom of choice for finite sets. Fundam. Math. 33, 137–168 (1945)
2. On definable sets of positive integers. Fundam. Math. 35, 81–112 (1947)
3. Mathematical logic, An Academic Course. Mathematical Monographs, vol. 18. Warsaw-Wrocław
(1948) (in Polish)
4. Sentences undecidable in formalized arithmetic, An exposition of the theory of Kurt Gödel. Studies
in Logic and the Foundations of Mathematics, North-Holland Publishing, Amsterdam (1952)
5. (with Kuratowski, K.) Set theory. Mathematical Monographs, vol. 27. Polish Scientific Publishers,
Warsaw-Wrocław (1952)
6. (with Robinson, R.M. and Tarski, A.) Undecidability and essential undecidability in arithmetic.
In: Tarski, A., Mostowski, A., Robinson, R.M.: Undecidable theories. Studies in Logic and the
Foundations of Mathematics, Part II. North-Holland Publishing, Amsterdam (1953)
7. (with Ehrenfeucht, A.) Models of axiomatic theories admitting automorphism. Fundam. Math. 43,
50–68 (1956)
8. On a generalization of quantifiers. Fundam. Math. 44, 12–36 (1957)
9. (with Grzegorczyk, A. and Ryll-Nardzewski, Cz.) The classical and the ω-complete arithmetic. J.
Symb. Log. 23, 188–206 (1958)
672 V. W. Marek
10. Thirty years of foundational studies. Lectures on the development of mathematical logic and the study
of the foundations of mathematics in 1930–1964. Acta Philosophica Fennica 17, 1–180 (1965)
11. Constructible sets with applications. Studies in Logic and the Foundations of Mathematics. Polish
Scientific Publishers, Warsaw and North-Holland Publishing, Amsterdam (1969)
12. Zygmunt, J.: Bibliography of Andrzej Mostowski. In: Ehrenfeucht, A., Marek, V.W., Srebrny, M.:
Andrzej Mostowski and Foundational Studies. IOS Press, Amsterdam (2007)
Helena Rasiowa
1. (with Sikorski, R.) A proof of the completeness theorem of Gödel. Fundam. Math. 37, 193–200 (1950)
2. (with Sikorski, R.) Mathematics of metamathematics. Wydawnictwo Naukowe PWN, Warsaw (1963)
3. An algebraic approach to non-classical logics. Studies in Logic and the Foundations of Mathematics,
vol. 78, Wydawnictwo Naukowe PWN/North-Holland, Warszawa/Amsterdam (1974)
4. Algebraic models of logics. Warsaw University (2001)
5. Bartol, W., Orłowska, E., Skowron, A.: Helena Rasiowa, 1917–1994. Modern Logic 5(3), 231–247
(July 1995)
6. Jankowski, A., Skowron, A.: Helena Rasiowa (1917–1994). In: Á. Garrido and U. Wybraniec-
Skardowska (eds.) The Lvov-Warsaw School. Past and Present, Studies in Universal Logic. https://
doi.org/10.1007/978-3-319-65430-0_48
Andrzej Grzegorczyk
1. Some Classes of Recursive Functions. Rozprawy Matematyczne Nr IV, Instytut Matematyczny PAN,
Warsaw (1953)
2. Zarys logiki matematycznej. Biblioteka Matematyczna tom 20, PWN, Warsaw (1961) (English version:
An Outline of Mathematical Logic. Reidel-Holland (1974))
3. Axiomatizability of geometry without points. Synthèse XII(2/3), 228–235 (1960)
4. A theory without recursive models. Bulletin de l’Académie Polonaise des Sciences, Série Math. 10,
63–69 (1962)
5. A philosophically plausible formal interpretation of intuitionistic logic. Indagationes Mathematicae
XXVI, 596–601 (1964)
6. Krajewski, S.: Andrzej Grzegorczyk, a logician par excellence. In: Á. Garrido and U. Wybraniec-
Skardowska (eds.) The Lvov-Warsaw School. Past and Present, Studies in Universal Logic. https://doi.
org/10.1007/978-3-319-65430-0_50
V. W. Marek ()
Department of Computer Science, University of Kentucky, Lexington, KY, USA
Andrzej Mostowski and the Notion
of a Model
Wilfrid Hodges
Abstract Model theory became an independent discipline within logic during the first
half of the 1950s. Andrzej Mostowski made several distinctive contributions to this
development through papers of his. Also his 1948 textbook of logic covers material in the
foundations of model theory, and in 1966 he published a survey book with several chapters
on model theory. We examine his choice of technical terms and concepts during this
period, and we discuss a criticism made by Abraham Robinson of the coverage of model
theory in the 1966 book. On this basis we draw some conclusions about Mostowski’s aims
and attitudes, which were often different from those of other pioneers in the field.
Keywords Model · Interpretation · Metamathematics · Mostowski · Tarski ·

Abraham Robinson
Mathematics Subject Classification (2000) Primary 03A05, 03C99
1 The Emergence of a New Discipline
The official birthday of model theory has to be the publication of Alfred Tarski’s paper
[47] in 1954, which began:
Within the last years a new branch of metamathematics has been de-
veloping. It is called the theory of models and can be regarded as a part (1)
of the semantics of formalized theories. [47, p. 572]
But we can see propaganda for the new discipline, as yet unnamed, 4 years earlier in the
addresses by Abraham Robinson [36] and Tarski [45] to the International Congress of
Mathematicians in 1950. One theme of the propaganda was that the new discipline would
provide new tools that algebraists and other mathematicians could use within their own

674 W. Hodges
disciplines. Thus Robinson:
. . . contemporary symbolic logic can produce useful tools—though by

no means omnipotent ones—for the development of actual mathematics,
(2)
more particularly for the development of algebra and, it would appear,
algebraic geometry. (Robinson [36, p. 694])
Likewise Tarski [45, p. 717] spoke of applications ‘which may be of general interest to
mathematicians and especially to algebraists’.
The new discipline absorbed various earlier pieces of work by Veblen, Löwenheim,
Skolem, Gödel, Mal’tsev and others, going back to the beginning of the twentieth century.
Some of the early contributors to the new discipline were seen at the time, and are
still seen, as primarily model theorists (at least within logic—some like Mal’tsev and
Robinson had mathematical interests outside logic). Besides Mal’tsev and Robinson
this group includes Henkin, Vaught and Fraïssé. But Andrzej Mostowski, though he
was a major figure in early model theory, was never primarily a model theorist. Like
Feferman (his junior by 15 years), Mostowski had interests that ranged across the whole
of logic, including set theory and the study of Gödel’s incompleteness theorem. For
example one of his papers with ‘models’ in the title [30, 1953] is mainly quoted for its
proofs that Zermelo-Fraenkel set theory and first-order Peano arithmetic are not finitely
axiomatisable.
The creation of a new mathematical discipline is always a challenge for historians
of mathematics, to understand where the new discipline came from, what were the forces
that drove it in the direction that it took, and what the creators of the discipline understood
themselves to be doing.
As far as I know, Mostowski never published a reflective account of the aims of model
theory, or of his own aims in this area. Of course we can infer something about his aims
from the problems that he chose to work on. But in this paper I explore two other routes
into his thinking. The first is his choice of terminology—not just the words that he chose
to use, but the concepts that he chose to give names to. The second is an issue raised by
Robinson, that Mostowski’s historical book [33] (1966) might not be entirely objective in
its account of model theory. We will assess Robinson’s criticism; to the extent that it is
sound, it throws some light on Mostowski’s understanding of what model theory is.
My wife and I met Mostowski at a party at Richard Montague’s house in Los Angeles
in summer 1967; we spent some time talking with him and found him genial and friendly.
My wife recalls that we met him again a few days later at a picnic on the Dockweiler State
Beach, and on that occasion he burned his feet on the hot sand. I was sorry I never got to
know him better.
I warmly thank the editors of this volume for their kind invitation to submit this
chapter, Jan Woleński for suggesting Mostowski as a topic, and Barbara Bogacka for
checking my translations of some of Mostowski’s Polish. Of course none of these people
are responsible for my errors. A copy of Mostowski’s book [28] came into my hands too
long ago for me to remember where it came from; my apologies to anybody I should be
thanking for it.
Andrzej Mostowski and the Notion of a Model 675
2 Mostowski’s Writings in Model Theory
We briefly review those parts of Mostowski’s work that relate most directly to model
theory.
There are three papers that had a major impact on model theory. The first is the
paper [29] (1952) ‘On direct products of theories’, which analysed the sentences true
in a cartesian power in terms of the sentences true in the factors. The paper [10] of
Feferman and Vaught owes its existence largely to Feferman’s study of ideas in [29] ([10,
p. 58], [8, p. 37])—though Feferman may have learned them directly from Tarski rather
than through reading the paper. (Possibly Tarski’s reluctance to accept Feferman’s early
work on this topic as of dissertation standard was based on a concern about whether
Feferman had added enough to what was already in [29]; see p. 211 in [9].) The second
is the paper [7] (1956) ‘Models of axiomatic theories admitting automorphisms’, written
jointly with Mostowski’s student Andrzej Ehrenfeucht. This paper introduced the notion
of indiscernibles, which quickly became an indispensable tool of both model theory and
set theory. The third is the paper [32] (1957) ‘On a generalization of quantifiers’; this
paper created one of the main strands of research in generalised model theory, with
applications spilling over into linguistics [34].
Mostowski’s earlier paper [27] (1947) ‘On absolute properties of relations’ seems to
have had little impact, because it was written around a rather obscure and specialised
question. But the paper contains a version of Henkin models of second-order model
theory, 3 years before Henkin published his own account; and on his page 34 Mostowski
comes close to formulating the notion of an elementary extension, 9 years before Tarski
and Vaught [49] published the definitive definition.
The paper [31] (1956) ‘Concerning a problem of H. Scholz’ was one of the earliest
papers in a difficult area of work relating computation theory to the model theory of finite
structures. See [6] for a recent assessment of the field and Mostowski’s contribution to it.
The paper [5] (1978) ‘The elementary theory of well-ordering—a metamathematical
study’ was a reconstruction by Doner and Tarski, after Mostowski’s death, of work that
Mostowski did with Tarski around 1940, applying the (syntactic) method of elimination
of quantifiers to the first-order theory of well-orderings. The paper itself is not model-
theoretic, but like other early work in quantifier elimination, its results were of great
interest to model theorists. Mostowski counted Tarski as the actual supervisor of his
doctoral thesis (1938) in set theory, though formally Kuratowski was the supervisor since
Tarski was not a professor at the time (Krajewski and Srebrny [22, p. 5]).
The paper [30] (1953) ‘On models of axiomatic systems’ applies the model-theoretic
notion of satisfaction. We will study its use of the word ‘model’.
Mostowski also wrote two books which reviewed mathematical logic as a whole. One
of these was his Polish textbook [28] Logika Matematyczna: Kurs Uniwersytecki of 1948,
which competes with Kleene’s more advanced Introduction to Metamathematics for the
role of the last major pre-model-theoretic textbook of logic. The other was a historical
survey Thirty Years of Foundational Studies [33, 1966]. The second of these books has two
chapters on ‘theory of models’, and both books contain material close to the foundations
of model theory.
676 W. Hodges
3 Background and Notation
We will be discussing the formalisation of several informal notions. It will be helpful

to have a convention for distinguishing the informal versions from the formal ones. I
will distinguish the formal versions by writing them with small capitals. Thus we talk
informally of assigning ‘interpretations’ to symbols, but Tarski’s formal version of this
notion is called an ‘INTERPRETATION’.
Several important papers in foundations of mathematics in the period 1910–1940 have
the following setting, which for convenience we can call the ‘archetypal pattern’. A formal
language L is described, and explanations are given for the meanings of the symbols
of the language. Some of the symbols have logical meanings, for example conjunction.
Other symbols, called the ‘nonlogical symbols’ or the ‘primitives’, have meanings that
come from the topic under discussion. The meanings of the nonlogical symbols add up to
a description of a structure M with a given universe or domain of elements; in a modern
usage we call M the ‘standard interpretation’ of the language L or the theory T . A set T
of sentences of L is presented as sentences that are true of M under the given explanation
of their meanings. The paper studies the structure M by analysing logical properties of
the set T . Often T , called a ‘theory’, is presented as the main topic of study. Probably the
best-known example of this general format is Gödel’s paper [11] on the incompleteness
of first-order Peano arithmetic.
This format came under strain as scholars asked new questions. Two particular areas
of strain are worth noting at once. The first is that in the archetypal pattern, the nonlogical
symbols of the language L have a fixed set of meanings that determine a particular
structure. But sometimes one wants to talk about two or more structures that make
the same sentences true. So it became necessary to have a way of detaching the fixed
meanings from the nonlogical symbols.
The second area of strain was that logicians became increasingly interested in the
justification of metatheoretical arguments. So these arguments should be formalised, and
some axioms for them set down. But then there was a question of what to formalise, and
in what language.
One example of these strains is the paper of Löwenheim [23], which shows that if
a first-order sentence is true in some structure, then it is true in some structure with at
most denumerably many elements. In today’s terminology, Löwenheim takes a structure
M and constructs a substructure N of M that satisfies some of the same sentences of L
as M did. But Löwenheim’s proof sometimes descends into obscurity, because he has no
explicit notion of a substructure; see Badesa [1], particularly his sections 6.1 and 6.2, for
documentation of this.
The word ‘model’, in a sense relevant to model theory, begins to appear in German
mathematical writing of the mid 1920s, in order to handle the situation where a structure
M is introduced and then a second structure N is constructed from M. The second
structure is called a model. One of the earliest examples of this usage is Hermann Weyl’s
discussion [54] (written in 1925) of geometric ‘models’ for proofs of consistency. For
example on his pages 30 and 31 we read of a ‘Modell’ of Lobachevsky’s geometry
within Euclid’s geometry; here Euclid’s geometry forms the standard interpretation of
the primitives ‘point’, ‘line’ etc., and the model is the nonstandard interpretation of these
expressions that proves the consistency of Lobachevsky’s axioms.
Another example of this usage, again from 1925, is in von Neumann’s paper [53] on
‘models’ of set theory. Von Neumann supposes that we have a set—call it T —of axioms
for set theory, and he shows how to construct, within the universe of sets described by T ,
a denumerable ‘model’ (his word) of the axioms. In order to carry out the construction,
he describes a second system of axioms—call it U —and claims ‘there obviously exists a
smallest’ system of sets satisfying U . To support the claim, he describes a procedure
for ‘constructing’ the system in infinitely many steps (p. 407f in van Heijenoort [52]).
If this procedure is to be justified in a formal system, what formal system should be
used? Could it be the axiom system T presented by von Neumann himself in [53]? Or
should it be some other more powerful system? And exactly what calculations need to be
represented in the formal system? Besides formalising the construction of , do we also
need to formalise a proof that a system of sets satisfying U also satisfies T ? And if we
do, would it be enough to show how to formalise this claim separately for each axiom, or
must we have a single formal proof covering all of them?
The moral of the von Neumann example is that a piece of metatheory may have
different formalisations, not all equivalent. We will see in Sect. 6 below that even when
we know what parts of the metatheory we want to formalise, there may be more than one
way of choosing concepts to do the required job.
4 Some of Tarski’s Proposals
The second quarter of the twentieth century saw attempts by various people, both to apply
the archetypal pattern to a wider range of problems, and to improve the formalisation
of the pattern. Tarski made a number of contributions. In [41] (1933) he showed how to
define, within the setting of the archetypal pattern, the notion ‘φ is a true sentence of
L’. This notion makes sense, given that in the archetypal pattern the expressions of L
all have appropriate fixed meanings, so it is determinate whether any given sentence of
L is true or not. This definition was Tarski’s original ‘truth definition’. Along the way,
Tarski also defined ‘satisfaction’ in the following sense. Suppose φ(x̄) is a formula of
L with free variables x̄, and we assign meanings to the variables in x̄. Then it makes
sense to ask whether the assigned meanings make φ(x̄) true, in other words, whether they
‘satisfy’ φ(x̄). Tarski analysed what set-theoretic content the assigned meanings would
need to have in order for us to give a formal definition of SATISFACTION; the resulting
definition defines when an assignment of set-theoretic objects to the free variables counts
as satisfying the formula. To formalise the definition, Tarski introduced a second theory
T (a ‘metatheory’), which would contain an exact copy of T but also enough set theory
to formalise the syntax of L and carry out some definitions by induction on the complexity
of formulas of L.
This was useful work in itself, but no help for dealing with the problem of alternative
structures that make the same set of sentences true. In [17] I documented how progressive
advances in the aims of metamathematics forced Tarski to adapt his truth definition step by
step, until eventually he had the model-theoretic form which he published with Vaught in
[49]. Already in 1933 Tarski could handle the case of two structures, one a substructure of
the other, so that he was equipped to formalise Löwenheim’s argument discussed above.
678 W. Hodges
But this case is in a way degenerate, because the relations etc. of the substructure agree
with those of the larger structure, so that all that is needed to specify the substructure is a
formula expressing its domain.
In 1936 [44] Tarski adapted the truth definition to allow new meanings to be assigned
to the nonlogical symbols of L. His idea was to consider an assignment α of appropriate
set-theoretic objects to the nonlogical symbols of L, and a sentence φ of L. He would
replace the nonlogical symbols in φ by distinct variables, thus getting a formula ψ. The
assignment α was defined to be a MODEL of φ if it satisfied ψ, where the assignment is
carried over from the nonlogical symbols to the variables put in place of them. For the
particular purposes of the paper [44], Tarski wanted to talk about all possible assignments
of meanings, so that the MODEL could in fact be exactly the same as the original
assignment of meanings to nonlogical symbols in the theory T . But otherwise Tarski
followed Weyl and von Neumann in using the expression MODEL for a new assignment
of meanings.
The model-theoretic truth definition of Tarski and Vaught [49], which we are told
was already available by 1952 or 1953 [49, p. 82, footnote], dropped this rigmarole of
replacing the symbols by new variables, and assigned the meanings (or rather the set-
theoretic objects representing them) directly to the nonlogical symbols. It was no longer
assumed that the nonlogical symbols came with preassigned meanings.
In [51] Tarski defined a notion that he called INTERPRETATION. (The book went
through several revisions. In the 1994 edition the definition appears on page 114 in §37,
but in the earliest editions the definition is in §33.) It relates two theories, say T and T in
languages L and L respectively. Like a MODEL of T , an INTERPRETATION of T involves
an assignment β to the nonlogical symbols of L. But instead of assigning set-theoretic
objects that convey meanings, β assigns to each nonlogical symbol of L an expression of
L . For each formula φ of L we construct a formula φ β of L by replacing each nonlogical
symbol of L by the expression assigned to it by β. We call β an INTERPRETATION of T
in T if for each sentence φ of T , φ β is provable from T .
Tarski gives some simple examples of INTERPRETATIONS. He takes T to be a theory
expressing that ∼ = is an equivalence relation on the set S. Then for example let T be a
theory of the arithmetic of rational numbers, with a symbol Q for the rational numbers
and a symbol Z for the integers. We can write a formula x ≡ y of L which expresses
that x and y are in Q and the difference x − y is in Z. Then let β be the assignment
that assigns Q to S and ≡ to ∼ =. Assuming that T is strong enough to allow us to prove
that ≡ is an equivalence relation on Q, the assignment β is an INTERPRETATION of T
in T . One can also construct geometric examples that interpret hyperbolic geometry in
euclidean geometry, representing the Klein-Beltrami model of hyperbolic geometry as an
INTERPRETATION in Tarski’s sense.
In Tarski’s terminology, INTERPRETATIONS and MODELS are really quite different
kinds of thing. What they have in common is that they both consist of assignments of
things to the nonlogical symbols of L. But the things assigned, and the condition for the
assignment to be a MODEL or an INTERPRETATION, are quite different. The notion of
an INTERPRETATION is purely syntactic: the assignment assigns strings of symbols to
symbols, and the condition for the assignment to be an INTERPRETATION is that certain
things are formally provable from a given theory. By contrast for a MODEL the assignment
assigns set-theoretic objects, and the condition for the assignment to be a MODEL involves
the notion of satisfaction; it requires that certain things are true, not that they are provable.
As we see from the example of the Klein-Beltrami model, the names ‘interpretation’
and ‘model’ have sometimes been applied to the same things. On p. 114f of the 1994
edition of [51] Tarski makes some remarks that seem to be intended to show how an
INTERPRETATION in his sense could sometimes be regarded as a MODEL in his sense.
He notes that if L has a standard interpretation, then the expressions assigned by β all
have meanings determined by this interpretation, and we can think of β as assigning
these meanings rather than the expressions. Then if the sentences of T are true for the
standard interpretation, anything provable from T will be true too, and one can infer
that the assignment of meanings (rather than expressions) is a MODEL of T in T . These
remarks are correct, but Tarski may have created some confusion by making them. In
general MODEL and INTERPRETATION are different notions, and neither is a special case
of the other. The jump that Tarski describes from INTERPRETATION to MODEL is not just
a change of viewpoint; it needs a substantial mathematical proof.
In [46] (1953), p. 20f, Tarski gives another definition of INTERPRETATION, which
agrees with the one above but removes the assumption that the theories have standard
interpretations. Also instead of physically altering the formula φ to φ β , Tarski achieves the
same effect by adding the assignment β in the form of explicit definitions of the nonlogical
symbols of L in the theory T . With this new definition of INTERPRETATION Tarski’s
attempt above to bring MODELS and INTERPRETATIONS together loses its purchase.
(Strictly the new definition is of ‘INTERPRETABLE IN’, but Tarski still speaks of an
‘INTERPRETATION’, as in the footnote on his p. 22.)
5 ‘Theories’
With this much background in place, we can begin to look at Mostowski’s papers,
starting with [29] ‘On direct products of theories’. This paper considers a structure A
that is a cartesian power B I of a structure B. (Mostowski also considers a variant of
cartesian power.) Mostowski asks how the first-order theory of A can be calculated from
information about what sentences are true in B.
Mostowski introduces the symbol T for the set of first-order sentences true in the
structure B, and he writes T I (where I is the index set of the power) for the set of first-
order sentences true in A. The sets T and T I are called ‘theories’. This notation ‘T I ’
implies that T I depends only on T and I . But this creates a problem: might T I not
also depend on B? For example T might also be the set of sentences true in some other
structure B ; how do we know that (B )I satisfies the same sentences as B I ?
In fact one of the main results of the paper is that we do know exactly this: T I is
determined by T and I , or even (as Mostowski points out) by T and the cardinality of I .
But it seems as if Mostowski has begged the question by assuming this result when he
sets up his terminology.
680 W. Hodges
Closer inspection shows that no question is begged, but Mostowski is using an archaic
terminology. At the beginning of the paper he tells us
Elementary mathematical theories are always concerned with certain

functions defined in a set I (called the universe of discourse of the the– (3)
ory) and certain relations with the common domain I. [29, p. 1]
In other words, a theory always has a standard interpretation. What Mostowski means
when he talks of ‘products of theories’ is exactly the same as what the modern reader
would express by talking of products of structures. Mostowski is working in the archetypal
pattern.
Of course Mostowski can choose to restrict the word ‘theory’ to theories that come
with a certain standard interpretation, provided that he makes it clear that he is doing
this (as in fact he does). But how sensible is this, at a date when other researchers in the
area are freely talking about ‘all the models of a set of axioms’ (as e.g. in Robinson [35]
(1951), p. 36)? Mostowski’s use of terminology suggests that he is simply not aware that
other researchers are using model theory as a framework for studying axiomatic classes
in algebra or other parts of pure mathematics.
A glance at what Mostowski says about ‘elementary theories’ in his textbook [28] of
1948 gives no reassurance at all. Under the head ‘elementary theory of groups’ Mostowski
writes:
The specific constants [i.e. nonlogical symbols] of this theory (apart from
the signs of equality and inequality) are L and σ . The constant L is of
type (), so it is the name of a set which in this theory we call the group. (4)
The constant σ is of type (, , ), which means that it is the name of a
ternary relation . . . [28, p. 234]
He really does seem to be telling us that for purposes of logic there just is one group!
It seems to me that we should draw the conclusion that at least by December 1949—
the date when [29] was submitted—Mostowski had not bought into the propaganda that
Robinson and Tarski would present for model theory at the 1950 International Congress.
At that date it was not one of his aims to provide tools for algebraists. In fact no move in
that direction appears in any of his later model-theoretic papers either. To my eye, none of
them contain anything that invites the description ‘application of model theory to algebra’.
When he came to write the historical book [33], Mostowski did find himself reviewing
other logicians’ work in this area, and he seems to have adopted a more relaxed notion of
‘theory’. But in [33] he gives no definition of ‘theory’, or any indication of how it differs
from ‘axiomatic theory’ or from ‘set of sentences’, or any sign that he is now using it in a
different meaning from his earlier papers.
6 ‘Models’
In his textbook [28] of 1948 and his paper [30] ‘On models of axiomatic theories’
(published in 1953) Mostowski introduces five different definitions of MODEL. For
convenience we can call the two definitions in [28] definitions (48:1) and (48:2), and the
three definitions in [30] definitions (53:1), (53:2) and (53:3). In brief they are as follows.
(48:1): This definition is on page 270 of [28]. It defines ‘MODEL of T in T ’, and it
agrees with Tarski’s definition of ‘INTERPRETATION of T in T as in [51].
(48:2): This is on page 356 of [28]. It defines ‘SEMANTIC MODEL of T ’, and it agrees
with Tarski’s definition of ‘MODEL of T ’ in [44].
(53:1): This definition is on page 136 of [30]. It defines ‘MODEL OF THE FIRST
KIND of T in T ’. The definition is the same as (48:1), except that Mostowski
requires T to contain enough set theory to code up the syntax of T , so that the
requirements ‘φ β is provable in T ’ take the strong form of saying that in T
we can define φ β from φ and β, and T proves a statement expressing that for
every φ in T , φ β is true.
(53:2): This definition is on page 142 of [30]. It defines ‘MODEL OF THE SECOND KIND
of T in T ’. It relates to (48:2) in the same way as (53:1) relates to (48:1);
in other words, Mostowski requires that T contains enough notions—and in
particular enough set theory—to allow us to define satisfaction of formulas of
L, and then prove that every axiom of T is true under the given assignment.
(Strictly Mostowski says that formulas expressing the truth ‘hold’, not that they
are provable from T . But then his argument uses the assumption that ‘the
existence of a real model of the second kind is provable’ in T , so the effect
is the same.)
(53:3): This definition is on page 149 of [30]. It defines ‘MODEL OF THE THIRD KIND
of T in T ’ like (53:2), but using arithmetic instead of set theory.
Let me make a few remarks first about (48:2) and (53:2), which correspond to Tarski’s
MODEL .
In [28] Mostowski proves Gödel’s completeness theorem, using the notion of SAT-
ISFACTION but without mentioning models. For example he says that every first-order
sentence is either satisfiable in the natural numbers or formally refutable (this paraphrases
Theorem 7 on his p. 353). Then he moves on to the downward Löwenheim-Skolem, and
this is where he introduces the definition (48:2). In fact the form of the Löwenheim-
Skolem theorem that he proves is that a satisfiable theory has an at most denumerable
model, and he proves it by deducing it from the form of Gödel’s theorem that says that a
syntactically consistent theory is satisfiable in the natural numbers. So his version of the
downward Löwenheim-Skolem theorem needs no new concepts beyond the ones that he
has already used for Gödel’s theorem. Then why does he choose this place to introduce
the notion of SEMANTIC MODEL? One possible guess is that he has at the back of his
mind a form of the Löwenheim-Skolem theorem which says that a theory which is true in
a standard interpretation has an at most denumerable model; in general the denumerable
model will not be the model given by the standard interpretation, so we are changing the
interpretations of the nonlogical symbols. If this is right, then Mostowski is harking back
to the 1920s notion of ‘model’.
682 W. Hodges
In (53:2) Mostowski arranges the formal definition of satisfaction in T in such a

way that there is no mention of replacing nonlogical symbols by variables. In effect,
Mostowski anticipates the Tarski-Vaught definition of satisfaction in [49]. But this fact
is well hidden in the technical details, and quite possibly none of Mostowski’s readers
realised that he had cleared away the syntactic complexities of Tarski’s definition in [44].
Mostowski’s aim in (53:2) was to prove a mathematical fact about Zermelo-Fraenkel
set theory, taking this set theory to be T . The proof proceeds by taking the language
L of T as an object language, not as a metalanguage for expressing metamathematical
properties of L. This quietly undermines Tarski’s efforts to keep mathematics and
metamathematics distinct. A few years later Robinson created nonstandard analysis in
a similar way, by treating the language in which we do mathematics as one of the
objects that we handle in the mathematics that we do; there is a breakdown of levels.
Mathematicians appreciate this sort of move, which discovers new mathematical facts by
looking at familiar things from an unfamiliar point of view.
Tarski certainly appreciated good mathematics, but it’s hard to imagine moves like
these of Mostowski and Robinson coming from Tarski himself. Feferman [9, p. 223]
makes the interesting remark that Robinson ‘had a certain looseness of presentation that
annoyed Tarski’. It never struck me that Robinson was a careless mathematician, and I
wonder if the remark has to do with Robinson’s willingness to ignore the ‘right’ way of
looking at things, as for example in nonstandard analysis. If that be so, then Mostowski’s
definition (53:2) is loose in much the same way.
Tarski by contrast had a programme to tidy up metamathematics by giving formally
correct definitions of the needed concepts, and it was his view that the question ‘Which
are the needed concepts?’ has an objectively right answer. In a telling passage near the
beginning of [40, p. 112] he says:
In geometry it was a question of making precise the spatial intuitions acquired

empirically in everyday life, intuitions which are vague and confused by their
very nature. Here [i.e. in metamathematics as opposed to geometry] we have
to deal with intuitions more clear and conscious, those of a logical nature
(5)
relating to another domain of science, metamathematics. To the geometers the
necessity presented itself of choosing one of several incompatible meanings,
but here arbitrariness in establishing the content of the term in question is
reduced almost to zero.
This view is likely to appeal to philosophers who regard conceptual analysis as one of
the basic tools of mathematical foundations—a view encouraged by Frege’s analysis of
number and Turing’s analysis of computability. But it’s also a view that stands in the
way of the kinds of conceptual sleight of hand that Mostowski and Robinson exploited.
In this respect Mostowski and Robinson were the mathematicians and Tarski was the
philosopher.
Turning to (48:1) and (53:1), an obvious question is why Mostowski blurs the
distinction between syntax and semantics by using the word MODEL in these two cases.
Again we see Mostowski disregarding distinctions that Tarski put in place. But there are
some further things to be said about these cases. I have discussed and documented Tarski’s
position on the issues elsewhere (chiefly [18], [19] and [20]), so I beg leave to give the
conclusions rather than the supporting evidence.
Mostowski notes in [30] that:
Models of the first kind [i.e. for (53:1)] are the ones with which one has to do
in the usual proofs of consistency and of independence of axiomatic systems (6)
[30, p. 138].
This comment closely matches the applications that he gives in [28] for models of the
kind (48:1); these are the main contents of his Chapter XI on ‘Methodological questions’.
One case that he discusses in Chapter XI is Padoa’s method for proving independence of
concepts within a theory.
In 1900 Padoa gave a loose description of a procedure for showing that in a formal
theory T with nonlogical symbols R0 , . . . , Rn , the concept expressed by R0 is not
definable in terms of the concepts expressed by R1 , . . . , Rn . The procedure was to give
two different interpretations of T which agree on all of R1 , . . . , Rn but disagree on R0 .
Suppose we want to make Padoa’s method formally correct. How should we proceed?
For example in terms of the notions introduced by Tarski, should we treat the ‘interpreta-
tions’ of T as INTERPRETATIONS or as MODELS? When Tarski considered the question
in the 1920s and 1930s [42], he came down on the side of INTERPRETATIONS. In his
reading, Padoa’s procedure was to give a second theory T and two INTERPRETATIONS
β, γ of T in T , such that β and γ assign the same expressions to R1 , . . . , Rn , but it
is a theorem of T that the expressions assigned to R0 by β and γ are not equivalent.
On this account, Padoa’s method is purely syntactic. There is no reference to MODELS or
SATISFACTION anywhere in it. Mostowski [28, pp. 283–291] follows this account.
Today I think most logicians would say that Tarski captured the essential mathematical
content of Padoa’s method, but he threw away the intuition behind it. That intuition is
better captured by formalising Padoa’s procedure in terms of MODELS, as we normally
do today. The issue came to a head in 1953 when Beth [2] proved that in first-order
logic Padoa’s condition is both necessary and sufficient for R0 to be not definable from
R1 , . . . , Rn in T . Beth expressed the condition model-theoretically, though for technical
reasons he used cut-free derivations in his proof. When he sent his proof to Tarski, Tarski
responded through his student Feferman that Beth had misconstrued a syntactic theorem
as a semantic one, and he should rewrite so as to remove the reference to models. (Fefer-
man later thought he remembered saying the opposite to this. But the correspondence is
available in Van Ulsen’s doctoral thesis; I quote the relevant passage in [20].)
To return to the obvious question mentioned earlier: why did Mostowski blur the dis-
tinction between syntax and semantics by using ‘model’ for Tarski’s INTERPRETATIONS?
One naturally asks why these five notions are all called ‘model’. In [28] Mostowski says
nothing to answer this question, but a footnote on page 356 does call attention to the clash
of terminology. In the footnote Mostowski says
We need to distinguish the concept of model defined thus from the

concept of a model of one theory in another, as defined in Chapter XI (7)
(§2 p. 270). This is the reason for using here the term semantic model.
The puzzled reader might well insist that it would be a better reason for not using the
expression ‘model’ for both these notions.
684 W. Hodges
The case of Padoa’s method discussed above suggests a reason for using ‘model’ in
all these cases, namely that Mostowski realised that Tarski’s INTERPRETATIONS and
Tarski’s MODELS were to some extent solutions of the same problems. They could both
be used to formalise earlier informal metamathematical discussions involving variation of
interpretations of symbols. So Mostowski could naturally see all his definitions of MODEL
as related tools in a general logical toolkit. If this is correct, then Mostowski chooses his
terminology more on the basis of possible mathematical applications than on the basis
of conceptual analysis. Here again, comparing him with Tarski, Mostowski is the more
typical mathematician.
There is a feature of [30] that supports this reading. As he introduces each of his three
definitions of MODEL, Mostowski lists some ‘general facts’ about it. The lists are given
in similar formats for the three kinds of MODEL, ‘for comparison with other notions of
model’. (For (58:1) this is on p. 138f, for (58:2) on p. 142 and for (58:3) on p. 208.) The
effect is as if Mostowski is providing a set of tools together with notes on where the tools
can appropriately be used.
Mostowski is probably responsible for one more use of the word ‘model’, namely its
use to mean ‘structure’. This use appears already in [27] (1947) where he speaks of an
‘absolute model’ without any reference to any theory that it is a model of. In [33] he
often uses ‘model’ for ‘structure’—which puts an extra burden of interpretation on the
reader who has to work out whether or not he means ‘model of’ some salient theory. In
mitigation it should be added that when he wrote [27] the word ‘structure’ was not yet
in use among model theorists; at that date one usually said ‘system’, which is even more
open to confusion. Among model theorists the word ‘structure’ came into common use
in the late 1950s, probably under the influence of Robinson and Bourbaki. But whatever
the merits of this use of the word ‘model’, it once more shows Mostowski disregarding
Tarski’s careful analysis of concepts.
7 Thirty Years of Foundational Studies
In 1964 Mostowski gave a series of lectures ‘on the development of mathematical logic
and of the study of foundations of mathematics in the years 1930–1964’ at a summer
school in Vaasa, Finland. Two years later he published a revised version of these lectures
[33]. He remarks in his Foreword that he hopes to convey ‘some of the enthusiasm with
which I witnessed the creation of theories reported on in the following pages’. The lectures
certainly live up to that hope.
They can also serve another purpose for today’s reader. Round about 1960 a consensus
was forming that mathematical logic should be classified under the main heads of Proof
Theory, Set Theory, Model Theory and Recursion Theory. For most of Mostowski’s career
there was no such classification. He draws the divisions in quite different places, and as
a result he makes connections between different areas of logic in ways that a modern
reader may find fresh and stimulating. Just to pick one example at random, the chapter on
‘Semantics’ begins with the truth definition and finishes with speedup theorems in proof
theory.
About his own contributions, Mostowski’s account in this book is modest to a fault.
His paper with Ehrenfeucht on indiscernibles [7] is in the bibliography but not mentioned
in the text; the papers [29] and [32] are not even in the bibliography. He mentions his
paper [30] (at [33, p. 142]), but only to record that its proof that Zermelo-Fraenkel set
theory is not finitely axiomatisable contained a mistake, and the result should be credited
to Richard Montague [26]. (I haven’t been able to find out what mistake he has in mind,
or whether it is significant.)
8 Robinson’s Complaint
Joseph Dauben in his biography of Abraham Robinson [4] records that Robinson was
severely critical of the treatment of model theory in Mostowski’s book [33]. The criticism
was made privately in a letter to Gerald Sacks. Robinson wrote:
This term [‘model theory’] was indeed coined by Tarski in the early
fifties and this is where Mostowski in his “Thirty years of Foundational
Studies” (according to which I apparently started my career in 1963)
places the beginning of the subject.
However, if you were to look at my “On the Metamathematics of Alge–
bra” you will find that it contains not only algebraic applications but
also the general framework of model theory (e.g. the general scheme of (8)
classes of sentences versus classes of models). At the same time I do not
wish to belittle Henkin’s influence on later developments.
In any case, I am not surprised to observe, again and again, that Tarski
has trained his students (and that includes Mostowski) to see history in
the way he wants them to. ([4, p. 450f], quoting a letter to Sacks dated
8 June 1972)
Dauben [4, p. 449] describes this as an ‘uncharacteristic letter’. Certainly it’s uncomfort-
able. But I was glad that Dauben included it, because it does represent a side of Robinson
that I and other people witnessed in the early 1970s. Possibly it was an early sign of the
illness that took him soon afterwards.
Let it be said straight away that Robinson’s specific complaint about [33]—that
Mostowski places the start of Robinson’s career in 1963—is not true. The first of
Robinson’s papers listed in Mostowski’s bibliography is [37] from 1955. On his pages
125f Mostowski describes the contents of this paper, noting that one of Robinson’s ‘most
important applications’ of these results, in a paper of 1959, ‘could hardly be obtained’ by
Tarski’s own preferred methods.
686 W. Hodges
But it is true that Mostowski’s book shows no awareness that Robinson made
any contributions before 1955. These contributions include his address [36] to the
International Congress of Mathematicians in 1950, reporting the main results of his PhD
thesis [35] submitted in 1949. Robinson had submitted a paper to the Congress, but Tarski
as chair of the Logic section had intervened to elevate Robinson to an invited lecturer;
in order to do this Tarski had had to argue for allowing four invited papers in logic, as
opposed to the three allowed to other subjects [4, p. 170]. Mostowski must have been
aware that Robinson was an invited speaker at the International Congress, since he was
present at the Congress himself [4, p. 171]; but apparently Robinson’s paper didn’t register
with him.
In fact Mostowski does mention in [33] two of the innovations that appear in
Robinson’s PhD thesis, but he attributes neither of them to Robinson.
One of these innovations is the use of Steinitz’s theorem to prove the completeness
of the theory of algebraically closed fields of a given characteristic ([33, p. 123],
[35, p. 60]). This was one of the more startling ideas that Robinson presented to the
International Congress in 1950 [36]. Robinson showed that any two algebraically closed
fields of the same characteristic are respectively elementarily equivalent to fields that
have transcendence degree ℵ0 and hence are isomorphic by Steinitz’s theorem. Vaught
later gave a simpler and more general form (‘Vaught’s test’) to this argument by pointing
out that every denumerable first-order theory that is categorical in some uncountable
cardinality and has no finite models is complete—an adjustment that Robinson himself
praised for its ‘remarkable efficacy’ [38, p. 11]. Robinson’s result could be recovered
by deducing from Steinitz’s theorem that the theory of algebraically closed fields of a
given characteristic is categorical in every uncountable cardinality. Mostowski attributes
Vaught’s test to Vaught but gives no attribution for the use of Steinitz.
With hindsight we can see that Robinson’s use of Steinitz’s theorem was a major
step forwards in model theory. It showed that model theorists could apply algebraic
embedding or isomorphism results to prove facts about first-order definability. This very
quickly became, and remains still, one of the main themes of the subject. But it took
some time for the novelty to be appreciated. For example Henkin, in his review of [35] in
1952, mentioned Robinson’s result that the theory of algebraically closed fields of a given
characteristic is complete, but added the dubious claim that
the techniques underlying these derivations . . . have been obtained earlier by

(9)
others. [12, p. 206]
As in Sect. 9 below, Henkin might have been better advised not to write reviews of
publications that he regarded as in competition with his own work.
The second innovation that appears in [35] and is mentioned by Mostowski without
an attribution to Robinson is the notion of the formally defined class of all models of a
given set of first-order sentences. Mostowski [33, p. 119] writes this class as E(X), where
X is the set of sentences. People who know Tarski’s work on the truth definition might
reasonably expect that the notion E(X) appears in papers of Tarski. But in fact Tarski had
considerable misgivings about using this notion in mathematics, as opposed to having it
available as an informal notion of metamathematics. It doesn’t appear at all in [47] where
we would certainly expect to find it; and in [45] it appears only for the case where X is a
single sentence (p. 710). As far as I know, free-wheeling mathematical use of the notion
E(X) is found first in Robinson’s PhD thesis [35] (see his p. 36f), which is a passage that
Robinson himself refers to in the letter quoted by Dauben. (See [21] end of §2 for some
further discussion of this point. There may also be earlier mathematical arguments that
explicitly use this notion in papers of Mal’tsev or the doctoral thesis of Henkin; I have not
checked these in detail.)
In practice historical surveys always leave out something, and there are plenty of other
things that Mostowski could have mentioned but didn’t. I would just say here that his lack
of interest in the applications of model theory to algebra and other disciplines of pure
mathematics made it less likely that he would appreciate the significance of Robinson’s
use of Steinitz’s theorem. Likewise a relative lack of interest in questions of conceptual
analysis would make it less likely that he would appreciate the fine details of Tarski’s
views on metamathematical definitions. There is not the slightest reason to attribute
either of these features of [33] to how Tarski ‘trained his students’ (as Robinson’s letter
suggests).
9 Mostowski’s Attributions in General
An unspoken implication of Robinson’s complaint is that Mostowski attributed too much

to Tarski. Mostowski does say that
The systematic development of model theory was initiated by Tarski in the

(10)
early fifties’ [33, p. 119]
This is true in the sense that Tarski pointed the efforts of leading members of the Berkeley
group into this area of logic during the 1950s. He did this partly by pressing some
questions (e.g. can elementary equivalence be defined without reference to satisfaction?),
and partly by giving basic definitions (e.g. truth in a model, elementary extension). Tarski
himself proved very few mathematical results of model theory—far fewer than Robinson.
But the one result that is regularly credited to him, namely the Łoś-Tarski theorem on
formulas preserved in substructures, was proved in the early 1950s [48, 1954], and it
certainly set a trend.
On p. 121 of [33] Mostowski attributes to Tarski [45] (1950) the notions of submodel
and extension. This is correct in the sense that Tarski included these notions in a list
of basic notions in [45]. But the notions were already in wide use, not least in Garrett
Birkhoff’s notion of subalgebras [3, 1935]. Earlier than Tarski’s paper, Robinson had
given careful definitions of substructures and extensions of structures in his PhD thesis
[35, pp. 65–68]. But very likely Mostowski was unaware of the contents of [35].
Mostowski also attributes the notion of elementary equivalence to Tarski [45]. This
is one place where Mostowski gives Tarski less credit than Tarski is entitled to: Tarski
had already defined elementary equivalence in [43] (1936). Tarski implies on p. 283 of
[50] that he had a ‘correct and precise’ definition of the notion as early as 1930. What
Tarski claims in [45] is a mathematical (as opposed to metamathematical) definition of
elementary equivalence; today this distinction is probably invisible to most logicians.
One other case is worth mentioning, because it shows Mostowski dealing with an
attribution that was sensitive for some people. In [13] Henkin credited Mal’tsev [25] with
688 W. Hodges
giving the first completeness proof for first-order logic with arbitrarily many symbols. In
Henkin’s later paper [14] about the history of Henkin’s own proof of this result, Mal’tsev
doesn’t even get a mention. What happened between these two papers of Henkin was that
Henkin and Mostowski together wrote a review [15] of work of Mal’tsev, in which they
commented on the proof of the completeness theorem in [25]. Mal’tsev uses the Skolem
normal forms φ of sentences φ, more precisely the normal forms that add new relation
symbols. The two reviewers comment that Mal’tsev uses the fact that every model of φ
expands to a model of φ ; Skolem himself claimed only that if φ has a model then so
does φ . They remark ‘This stronger result does not seem to be formulated explicitly in
the literature, although it can be discerned by a careful reading of the usual proofs of
Skolem’s theorem’ [15, p. 56f]. For what it’s worth, when I included this result of Skolem
in my Model Theory text [16, Theorem 2.6.5, p. 63] I included the stronger statement as
part of the theorem; it never occurred to me that it wasn’t obvious from the proof.
Gaps like this in proofs are an embarrassment. I think usually we size up whether
we believe the author knows what he or she is doing, and if the answer is Yes then we
credit the theorem—though we may still point out the gap. This is exactly how Mostowski
handles the issue in [33]:
The possibility of applying the completeness theorem to such problems was

first pointed out by Malcev . . . who also published the first proof of the
(11)
theorem independent of the cardinality of [the set of symbols]. (Malcev’s proof
was not entirely correct but his mistake can easily be corrected.) [33, p. 56]
Mostowski’s statement is faultless.
10 Conclusion
Andrzej Mostowski saw model theory mainly as a source of tools for answering questions
raised in metamathematics, not as a source of new tools for other mathematical disciplines
such as algebra. In this respect his view was quite different from those expressed by
Robinson and Tarski in their 1950 manifestos.
As a corollary, Mostowski was insensitive to some features that were characteristic
of model theory from 1950 onwards. These include the use of embeddings as a tool for
analysing definability properties of structures. Tarski shared this insensitivity (cf. [21]
§3(a)), and it set Mostowski and Tarski apart from Robinson who pioneered this use of
embeddings.
But at the same time, Mostowski and Robinson were alike in that their work was
driven by the need to find mathematical tools to solve certain problems, and not by any
programme of conceptual analysis such as we find in Tarski. In fact Mostowski’s use of
the word ‘model’ shows little regard for Tarski’s conceptual concerns.
Robinson saw the account of model theory in [33] as playing down Robinson’s own
contributions to the subject. There is truth in his criticism—though he exaggerated the
point. But the main cause seems to have been Mostowski’s own lack of interest in the
more algebraic aspects of the subject. Where Mostowski had reason to take an interest in
the historical details, he comes across as careful and scrupulously fair. Possibly he gave
too much credit to Tarski; but if he did, he was certainly not the first or last student to
exaggerate the achievements of his own doctoral supervisor.
Very few new mathematical disciplines are created by a single person, and certainly
model theory was not one of them. But without Mostowski’s contributions, model theory
could easily have evolved in a different direction from the one that it took. As often
happens in mathematics, some of his best ideas were adopted and used in ways that he
could hardly have anticipated; he played a much more significant role than is suggested
by his own modest account in [33].
References
1. Badesa, C.: The Birth of Model Theory: Löwenheim’s Theorem in the Frame of the Theory of
Relatives. Princeton University Press, Princeton (2004)
2. Beth, E.W.: On Padoa’s method in the theory of definition. Ned. Akad. Wet. Proc. Ser. A 56 (=
Indagationes Mathematicae 15), 330–339 (1953)
3. Birkhoff, G.: On the structure of abstract algebras. Proc. Camb. Philol. Soc. 31, 433–454 (1935)
4. Dauben, J.W.: Abraham Robinson: The Creation of Nonstandard Analysis, A Personal and Mathe-
matical Odyssey. Princeton University Press, Princeton (1995)
5. Doner, J.E., Mostowski, A., Tarski, A.: The elementary theory of well-ordering—a metamathematical
study. In: Macintyre, A., Pacholski, L., Paris, J. (eds.) Logic Colloquium ’77, pp. 1–54. North-
6. Durand, A., Jones, N.D., Makowsky, J.A., More, M.: Fifty years of the spectrum problem: survey and
new results. Bull. Symb. Log. 18(4), 505–553 (2012)
7. Ehrenfeucht, A., Mostowski, A.: Models of axiomatic theories admitting automorphisms. Fundam.
Math. 43, 50–68 (1956)
8. Feferman, S.: A fortuitous year with Leon Henkin. In: Manzano, M., Sain, I., Alonso, E. (eds.) The
Life and Work of Leon Henkin, pp. 35–40. Birkhäuser, Heidelberg (2014)
9. Feferman, A.B., Feferman, S.: Alfred Tarski: Life and Logic. Cambridge University Press, Cambridge
(2004)
10. Feferman, S., Vaught, R.L.: The first order properties of products of algebraic systems. Fundam. Math.
47, 57–103 (1959)
11. Gödel, K.: Über formal unentscheidbare Sätze der Principia Mathematica und verwandter Systeme I.
Monatsh. Math. Phys. 38, 173–198 (1931)
12. Henkin, L.: Review: A. Robinson, On the metamathematics of algebra. J. Symb. Log. 17, 205–207
(1952)
13. Henkin, L.: Some interconnections between modern algebra and mathematical logic. Trans. Am.
Math. Soc. 74(3), 410–427 (1953)
14. Henkin, L.: The discovery of my completeness proofs. Bull. Symb. Log. 2(2), 127–158 (1996)
15. Henkin, L., Mostowski, A.: Review: A. Mal’cev, on a general method for obtaining local theorems in
group theory. J. Symb. Log. 24, 55–57 (1959)
16. Hodges, W.: Model Theory. Cambridge University Press, Cambridge (1993)
17. Hodges, W.: What languages have Tarski truth definitions? Ann. Pure Appl. Logic 126, 93–113 (2004)
18. Hodges, W.: Tarski’s theory of definitions. In: Patterson, D.E. (ed.) New Essays on Tarski and
Philosophy, pp. 94–132. Oxford University Press, Oxford (2008)
19. Hodges, W.: Tarski on Padoa’s method: a test case for understanding logicians of other traditions.
In: Chakraborti, M.K., et al. (eds.) Logic, Navya-Nyāya and Applications: Homage to Bimal Krishna
Matilal, pp. 155–169. College Publications, London (2008)
20. Hodges, W.: From choosing elements to choosing concepts: the evolution of Feferman’s work in
model theory. In: Jäger, G., Sieg, W. (eds.) Feferman on Foundations. Outstanding Contributions to
Logic. Springer, Switzerland (2018)
690 W. Hodges
21. Hodges, W.: A short history of model theory. Appendix in: Button, T. and Walsh, S. Philosophy and
Model Theory. Oxford University Press, Oxford, pp. 439–476 (2018)
22. Krajewski, S., Srebrny, M.: On the Life and Work of Andrzej Mostowski (1913–1975). In:
Ehrenfeucht, A., Marek, V.W., Srebrny, M. (eds.) Andrzej Mostowski and Foundational Studies, pp.
3–14. IOS Press, Amsterdam (2008)
23. Löwenheim, L.: Über Möglichkeiten im Relativkalkül. Math. Ann. 76, 447–470 (1915) (Tr. as:
On possibilities in the calculus of relatives. In: Van Heijenoort, J.: From Frege to Gödel. Harvard
University Press, Cambridge (1967), pp. 228–251)
24. Mal’cev (= Mal’tsev), A.I.: The Metamathematics of Algebraic Systems, Collected Papers: 1936–
1967. North-Holland, Amsterdam (1971) (Tr. Wells, B.F. III)
25. Mal’tsev, A.I.: Untersuchungen aus dem Gebiete der mathematischen Logik. Matematicheskie Sb.
1(43), 323–336 (1936) (Tr. as: Investigations in the realm of mathematical logic. In: Mal’cev (=
Mal’tsev), A.I.: The Metamathematics of Algebraic Systems, Collected Papers: 1936–1967. North-
Holland, Amsterdam (1971) (Tr. Wells, B.F. III), pp. 1–14)
26. Montague, R.: Semantical closure and non-finite axiomatizability I. In: Infinitistic Methods (editors
unnamed), pp. 45–69. Państwowe Wydawnictwo Naukowe, Warsaw (1961)
27. Mostowski, A.: On absolute properties of relations. J. Symb. Log. 12, 33–42 (1947)
28. Mostowski, A.: Logika Matematyczna: Kurs Uniwersytecki. Monografie Matematyczne, vol. 18.
Uniwersytet of Warsaw, Warsaw (1948)
29. Mostowski, A.: On direct products of theories. J. Symb. Log. 17(1), 1–31 (1952)
30. Mostowski, A.: On models of axiomatic systems. Fundam. Math. 39, 133–158 (1953)
31. Mostowski, A.: Concerning a problem of H. Scholz. Zeitschrift für Mathematische Logik und
Grundlagen der Mathematik 2, 210–214 (1956)
32. Mostowski, A.: On a generalization of quantifiers. Fundam. Math. 44, 12–36 (1957)
33. Mostowski, A.: Thirty Years of Foundational Studies: Lectures on the Development of Mathematical
Logic and the Study of the Foundations of Mathematics in 1930–1964. Blackwell, Oxford (1966)
34. Peters, S., Westerståhl, D.: Quantifiers in Language and Logic. Clarendon Press, Oxford (2006)
35. Robinson, A.: On the Metamathematics of Algebra. North-Holland, Amsterdam (1951) (This is an
almost exact transcript of Robinson’s PhD thesis, Birkbeck College London, 1949)
36. Robinson, A.: On the application of symbolic logic to algebra. In: Proceedings of the International
Congress of Mathematicians, Cambridge, Massachusetts 1950, vol. 1, pp. 686–694. American
Mathematical Society, Providence (1952) (Reprinted in: Robinson, A.: Selected Papers of Abraham
Robinson, Vol. 1: Model Theory and Algebra. Keisler, H.J. et al. (eds.) North-Holland, Amsterdam
(1979), pp. 3–11)
37. Robinson, A.: Ordered structures and related concepts. In: Heyting, A. (ed.) Mathematical Interpre-
tation of Formal Systems, pp. 51–56. North-Holland, Amsterdam (1955)
38. Robinson, A.: Complete Theories. North-Holland, Amsterdam (1956)
39. Robinson, A.: Selected Papers of Abraham Robinson, Vol. 1: Model Theory and Algebra. Keisler,
H.J. et al. (eds.) North-Holland, Amsterdam (1979)
40. Tarski, A.: Sur les ensembles définissables de nombres réels I. Fundam. Math. 17, 210–239 (1931)
41. Tarski, A.: Pojȩcie prawdy w jȩzykach nauk dedukcyjnych. Prace Towarzystwa Naukowego Warsza-
wskiego, Wydział III Nauk Matematyczno-Fizycznych 34, Warsaw (1933) (Revised and expanded
version tr. as: The Concept of Truth in Formalized Languages. In: Tarski, A: Logic, Semantics,
Metamathematics, 2nd edn. Hackett, Indianapolis (1983) (Tr. Woodger, J.H. (ed.) and introduced
by Corcoran, J.), pp. 152–278)
42. Tarski, A.: Einige methodologische Untersuchungen über die Definierbarkeit der Begriffe. Erkenntnis
5, 80–100 (1935) (Tr. as: Some methodological investigations on the definability of concepts. In:
Tarski, A: Logic, Semantics, Metamathematics, 2nd edn. Hackett, Indianapolis (1983) (Tr. Woodger,
J.H. (ed.) and introduced by Corcoran, J.), pp. 296–319)
43. Tarski, A.: Grundzüge des Systemenkalküls. Fundam. Math. 25, 503–526 (1935) and 26, 283–
301 (1936) (Tr. as: Foundations of the calculus of systems. In: Tarski, A: Logic, Semantics,
Metamathematics, 2nd edn. Hackett, Indianapolis (1983) (Tr. Woodger, J.H. (ed.) and introduced
by Corcoran, J.), pp. 342–383)
44. Tarski, A.: O pojȩciu wynikania logicznego. Przegla̧d Filozoficzny 39, 58–68 (1936) (Tr. as: On the
concept of logical consequence. In: Tarski, A: Logic, Semantics, Metamathematics, 2nd edn. Hackett,
Indianapolis (1983) (Tr. Woodger, J.H. (ed.) and introduced by Corcoran, J.), pp. 409–420)
45. Tarski, A.: Some notions and methods on the borderline of algebra and metamathematics. In:
Proceedings of International Congress of Mathematicians, Cambridge Massachusetts 1950, vol. 1,
pp. 705–720. American Mathematical Society, Providence (1952)
46. Tarski, A., Mostowski, A., Robinson, R.M.: Undecidable Theories. North-Holland, Amsterdam
(1953)
47. Tarski, A.: Contributions to the theory of models I. Indag. Math. 16, 572–581 (1954)
48. Tarski, A.: Contributions to the theory of models II. Indag. Math. 16, 582–588 (1954)
49. Tarski, A., Vaught, R.L.: Arithmetical extensions of relational systems. Compos. Math. 13, 81–102
(1956)
50. Tarski, A: Logic, Semantics, Metamathematics, 2nd edn. Hackett, Indianapolis (1983) (Tr. Woodger,
J.H. (ed.) and introduced by Corcoran, J.)
51. Tarski, A.: Introduction to Logic and to the Methodology of Deductive Sciences. Oxford University
Press, New York (1994) (Expanded and revised translation of Polish original, 1936)
52. Van Heijenoort, J.: From Frege to Gödel. Harvard University Press, Cambridge (1967)
53. Von Neumann, J.: Eine Axiomatisierung der Mengenlehre. J. Reine Angew. Math. 154, 219–240
(1925). Correction ibid. 155, 128 (1925) (Tr. as: An axiomatization of set theory. In: Van Heijenoort,
J.: From Frege to Gödel. Harvard University Press, Cambridge (1967), pp. 393–413)
54. Weyl, H.: Riemanns geometrische Ideen, ihre Auswirkung und ihre Verknüpfung mit der Gruppenthe-
orie. Springer, Berlin (1988) (Written 1925)
W. Hodges ()
Herons Brook, Sticklepath, Okehampton, UK
All Quantifiers Versus the Quantifier All
Abstract In courses of logic for general students the general and existential quantifiers
are the only ones distinguished from among all possible quantifier expressions of the
natural language. One can argue that other quantifiers deserve mention, even though
there are good reason for emphasizing the familiar ones: namely, they are the simplest,
the universal quantifier is a counterpart of the operation of generalizing, the number of
nested quantifiers is a good measure of logical complexity, and the expressive power of
the general quantifier and its dual is considerable.
Yet, even in the teaching about these two simplest quantifiers it has not been resolved
how to indicate the realm to which a given quantifier refers. The methods range from the
Fregean assumption that they refer to the totality of objects in the world to the restricted
quantifiers to many sorted logic. It turns out that these approaches are not fully equivalent,
because the sorts are usually assumed to be nonempty, which results in a problem similar
to the well-known issue with non-emptiness of names in syllogistics.
Logicians have studied various generalized quantifiers. It is, however, unclear how
to treat the quantifier “many” and similar heavily context-dependent ones. They are not
invariant under isomorphisms so no purely logical or mathematical treatment seems
applicable. How else can one characterize the context-independent quantifiers among
all possible quantifiers corresponding to quantifier expressions in natural language? The
following thesis on quantifiers is proposed:
(Principal Thesis) Context-independence = definability in terms of the universal quan-
tifier.
This thesis provides an additional reason for distinguishing the universal quantifier from
among all other quantifiers: it suffices for defining all context-independent ones
Keywords Quantifier · Generalized quantifiers · Quantifier’s range · Definability ·

Context-independence
Mathematics Subject Classification (2000) Primary 03C80; Secondary 00A30, 00A35,

03A05, 03B10, 03B65

694 S. Krajewski
1 The Universal Quantifier and Its Dual
The introduction of quantifiers to logical systems constituted an essential progress with

respect to the calculus of Boolean connectives—even though the propositional calculus
was finally formulated as a system at about the same time as was the predicate (functional)
calculus. The wider system, including the quantifiers “for all” (∀) and “there exists” (∃),
was also a far reaching strengthening of syllogistics, the celebrated system that up to the
nineteenth century was seen as the core of logic, and that can be seen as an ancient form
of a fragment of the predicate calculus.
In logic, from Aristotle to Frege to mid-twentieth century predicate logic, only two
quantifiers were incorporated: the general and the existential. They are still the only ones
taught in general logic courses. Because in classical logic ∃ is the dual of ∀, that is, ∃ =
¬∀¬, we can say that only the universal quantifier is added. (In some non-classical logic,
e.g., the intuitionistic logic, we need to retain the two quantifiers.)
From a certain natural point of view, namely the approach based on the linguistic
realities, it is not clear why the general and existential quantifiers are the only concepts
distinguished from among all possible quantifier expressions of the natural language. In
natural language there are dozens quantifier expressions, that is, expressions that state or
estimate the number of objects of a certain kind, or the size of a collection, or compare
sizes, etc. They include phrases like ‘all’, ‘always’, ‘nowhere’, ‘almost never’, ‘most’,
‘infinitely many’, ‘many’, ‘from time to time’, ‘a few’, ‘quite a few”, ‘several’, ‘just one’,
‘at least one’, ‘an overwhelming part of’, ‘as many as’, ‘roughly as many as’, and many,
many more, including statements like “More girls study programming than boys learn
boxing.” In mathematics, some other quantifier expressions are used, for example ‘there
are finitely many’, ‘there are uncountably many’, ‘the set of . . . is dense in . . . ’, and the
phrases like ‘almost all’, ‘a negligible amount’ are given various precise meanings in
specific mathematical theories.
For a long time logic did not recognize the rich realm of quantifiers or at least
did not perceive it as belonging to the domain of logic. What could be the reason for
the distinguished role of the familiar quantifiers? Let us try to argue from a logical
perspective.
First, simplicity. ‘All things’ corresponds to the full set—either of all things or of all
things in our universe of discourse. ‘At least one thing’ corresponds to the notion of non-
empty set, or is the negation of being the empty set.
The two standard quantifiers are the simplest ones. At the same time, and this is the
second reason, we can see the general quantifier as an abstract counterpart of the operation
of generalization, our important mental faculty. (Existential quantifier is, as mentioned
above, its dual.) This operation can be seen as basic: in Richard Epstein’s Critical thinking
[1] only the operation of generalization is mentioned, the quantifiers are not.
The third and fourth reasons are given after the following Digression and a fifth reason
emerges at the end of the paper.
All Quantifiers Versus the Quantifier All 695
2 A Digression on the Range of the Familiar Quantifiers
Whereas the quantifiers ∀ and ∃ are familiar now, their use causes some problems that are
not trivial, especially in the educational context, when general students are taught. One
problem is: what is the range of quantifiers, the realm to which a given quantifier refers?
Another problem appears when a standard answer is given to the first one: how to indicate
the range?
Frege introduced quantifiers in his framework, in which everything, literally: every
thing, was included. For Frege, and similarly for his successors like Russell, (early)
Wittgenstein or Quine, there is only one value-range for quantifiers, namely “all the
actually existing individuals” (cf. Hintikka [7, p. 30] and Peters-Westerståhl [17, p. 40])
or even “all the conceivably existing individuals” (as in Russell [18]). That assumption
seems, however, unsatisfactory. There are at least two main reasons for dissatisfaction.
First, it seems that in order to apply this approach the world must be perceived as a
collection of things. In particular, ‘always’ is expressible by ∀ only if time is seen as
composed of things such as moments or segments. In addition, only timeless relations are
naturally dealt with (cf. Epstein [2] and [3], Appendix A). Second, to generalize over
everything seems odd. In practice, we almost always mean a specified limited range.
Today we rarely share Frege’s ontology, but we all continue to use his formalism (in
a modified form, of course). The contemporary prevailing approach to this formalism
is, however, vastly different from his. The world is complex, and, usually, in a given
moment we consider only some objects, that is, we specify fragments of the whole
world. We have overwhelmingly adopted the model theoretic approach: models vary,
and generality means “for all elements of an intended range.” The formalism remains
but its interpretation is different: logic is no more about “the world” but rather about
various “possible worlds”, or models. It was Tarski who helped convince logicians to
study truth in models. In Hintikka [7] it is also stressed that this new, model-theoretic
approach overcomes the difficulty inherent in Frege’s approach, namely, how to identify
the basic simple things, the “urindividuals”.
Having agreed on the limited range, we need to express that in the symbolism we use
to deal with quantifiers. There are two traditional ways of expressing the restriction on the
range of variables used inside (logical) formalism. One is the use of variables of different
sorts. The other is the use of restricted, or relativized, quantifiers. As is well known, it is
easy to express relativized quantifiers by unrestricted ones: for any predicates A, B, C
(∀ x)C(x)A(x) ≡ (∀ x)(C(x) → A(x)), (∃ x)C(x)A(x) ≡ (∃ x)(C(x) ∧ A(x)).
In terms of a variable t, assumed to satisfy C(t), we simply have (∀ t)A(t) and (∃ t)A(t).
Is there a difference between the two methods? Of course, the approach is different: the
restrictions can be seen as imposed from outside the system in the case of the language
with different sorts of variables, while they are an optional part within the system in
the case of relativisation. Still, at the first glance it may seem that they are formally
equivalent. But not quite—there is a subtle difference. We normally assume that all sorts
are nonempty—similarly to the assumption that the universe is nonempty, or that in each
model the universe of the model is nonempty. This assumption is not made about the
predicate C(x).
696 S. Krajewski
Sometimes the possible emptiness of C is harmless. Explicit restrictions on the

quantifiers preserve the validity of de Morgan’s laws:
¬(∀ x)C(x)A(x) ≡ (∃ x)C(x)¬A(x), ¬(∃ x)C(x)A(x) ≡ (∀ x)(C(x)¬A(x).
What about the other tautologies that are so useful in manipulating quantifier prefixes?
Using different sorts brings no harm. In contrast to that, relativisation may cause a
problem! It seems to me that while the story with (some) Aristotelian syllogisms being
valid only under the assumption that all the terms are non-empty is very well known,
similar limitations concerning relativized quantifiers are not generally known. It is easy
to see (as was remarked in Krajewski [11]) that the following theorem holds.
Remark on relativized quantifiers:
The following formulas are valid under relativisation to an arbitrary C (we assume that
in A the variable x is not free):
(∀ x)(A ∨ B(x)) ≡ A ∨ (∀ x)B(x),
(∀ x)(A → B(x)) ≡ A → (∀ x)B(x),
(∃ x)(A ∧ B(x)) ≡ A ∧ (∃ x)B(x),
(∀ x)(B(x) → A) ≡ (∃ x)B(x) → A.
The following formulas remain valid only when relativized to nonempty C (we assume
that in A the variable x is not free):
(∀ x)(A ∧ B(x)) ≡ A ∧ (∀ x)B(x),
(∃ x)(A ∨ B(x)) ≡ A ∨ (∃ x)B(x),
(∃ x)(A → B(x)) ≡ A → (∃ x)B(x),
(∃ x)(B(x) → A) ≡ (∀ x)B(x) → A.
In other words the relativisations of the above tautologies are not valid, they are sometimes
false when C is interpreted as an empty set, but the following formulas are valid:
(∃ x)C(x) → [(∀ x)C(x)(A ∧ B(x)) ≡ A ∧ (∀ x)C(x)B(x)],
(∃ x)C(x) → [(∃ x)C(x)(A ∨ B(x)) ≡ A ∨ (∃ x)C(x)B(x)],
(∃ x)C(x) → [(∃ x)(A → B(x)) ≡ A → (∃ x)B(x)],
(∃ x)C(x) → [(∃ x)C(x)(B(x) → A) ≡ (∀ x)C(x)B(x) → A].
3 The Power of ∀ and ∃
The third reason for the distinguishing of ∀ and ∃ from among all possible quantifiers
has to do with logical complexity. The number of nested quantifiers is a good indicator
of logical complexity. The quantifiers ∀ and ∃ provide a great measure of complexity if
the number of alternating nested quantifiers is counted. The realization of this possibility
gave rise to the Kleene-Mostowski hierarchy, classifying the sets obtained from recursive
sets by a series of projections and complements. (See Kleene [10], Mostowski [15].)
Then other similar growing chains of ever more complicated objects were established,
e.g., the analytic hierarchy. From such a perspective these simple familiar quantifiers
look like anything but trivial. It is also of interest that neither Aristotle nor other pre-
modern logicians considered nested quantifiers. If teaching about quantifiers is limited to

formulas, and especially tautologies, with one quantifier or at most two, as is still done
in coursed for general students, the matter looks rather trivial and it remains unclear why
the quantifiers are needed. The power of quantifiers, even the simplest ones, is seen only
when several are combined. This brings us to the next reason for the distinguishing of ∀
and ∃.
The fourth reason emerges when one realizes that these quantifiers bring much more
expressive power than it would seem at first. When the standard additional machinery
available in logic is employed many new quantifiers can be defined. Some of them can
be easily defined within first order logic, for instance the numerical quantifiers: “there are
exactly n”, in short ∃!n , “there are more than n”, in short ∃>n , and their combinations (like
“there are three or four”), etc.
In higher order logics and in set theory many more quantifiers can be defined.
Definitions in mathematics are expressed in a technical language of a given branch, but
logicians have been able to express these definitions in the language of logic. Thus, for
instance, “there are infinitely many” cannot be defined in the first order logic, but can be
defined in the second order logic. The Henkin quantifier, the first example of a branching
quantifier, namely “for every x there exists y, and independently of that for every z there
exists t such that R(x, y, z, t)”, also goes beyond first order logic (see Henkin [6] and
Krynicki et al. [13]), even though it reflects such a way of using the familiar quantifier
expressions corresponding to ∀ and ∃ that can be found in natural language; this quantifier
is easily defined in second order logic: “there exist functions f , g such that for every
x and for every z R(x, f (x), z, g(z))”. The phrase “there are uncountably many” also
defines a quantifier but it makes sense only in reference to a background set theory. It was
unexpected that this quantifier can be recursively axiomatized. (See Keisler [8].) There
are many more examples of mathematical quantifiers. They suggested to mathematical
logicians the concept of a “generalized quantifier”.
4 Generalized Quantifiers in Logic
Generalized quantifiers were introduced to logic by Mostowski [16]. The formula

(Qx)ϕ(x) is satisfied in a model M=(M, . . .) iff the set {a: M| ϕ[a]} belongs to the
family of subsets of M that serves as the interpretation of Q. (Thus ∀ is interpreted as
{M} and ∃ as the family of all non-empty subsets of M.)
This notion was useful but was not sufficient for many formulations that are used in
natural language. Mostowski quantifiers are all of type <1>. The sentence “More girls
study programming than boys learn boxing” cannot be analyzed in logic with type <1>
quantifiers only. A more general definition was introduced by Lindström [14] who allowed
quantifiers of an arbitrary type <n1 ,. . . ,nk > that bind more variables and apply to several
formulas, and in a model M are interpreted as relations between subsets of M (in the case
of monadic quantifiers of type <1,1,. . . ,1>) or, more generally, relations between relations
on M.
Both Mostowski and Lindström were mathematicians so they made an important
assumption which obviously seemed necessary to them: they consider only the quantifiers
698 S. Krajewski
that are invariant with respect to isomorphism. Formally, if M ∼

= M’ then
M | (Qx1 , . . . , xk )(ϕ1 , . . . , ϕn ) iff M’ | (Qx1 , . . . , xk )(ϕ1 , . . . , ϕn ).
The assumption in the case of monadic quantifiers amounts to the fact that only
the size of the sets defined by the quantified formulas matters (cf. Peters-Westerståhl
[17] or Westerståhl [20]). The assumption that logic should be completely topic-neutral
constitutes the reason for admitting into logic only the quantifiers invariant under
isomorphism. Other mathematical properties can be defined by isomorphism-preserving
quantifiers. Yet they are not sufficient for some quantifiers commonly used in natural
language.
It is clear that logic is poorly equipped, if at all, to deal with many from among the
quantifier expressions listed above. For example, the concept “many” is different from the
more logical quantifiers and seems hardly definable in general since its meaning depends
on the situation in which the term is used. It is context-dependent. Peters and Westerståhl
call it “strongly” context-dependent and some authors call it intensional. (See Peters-
Westerståhl [17, p. 213].) To evaluate a sentence with such a context-dependent quantifier
we need an appropriate understanding of the world, or at least of the appropriate fragment
of the world. Logic itself is not sufficient. To know whether it is true or not that many
women at my university are pregnant or that many have been in the Himalayas, we need
to know how many women of a given age are, on average, pregnant, and how many go to
the Himalayas.
It is similarly with quantifier expressions like “a few”, “several”, “a huge number”,
“rarely”, “often”, etc., and even more obviously, with “surprisingly many”, “almost
everyone”, “virtually nowhere”, etc.
I believe that it should always be taught in general logic courses that many concepts
can be defined by the simplest quantifiers, but at the same time the student should be
made aware that many natural language constructions cannot. This seems to be generally
ignored by teachers of logic. For example, in the otherwise comprehensive textbook by
Andrzej Grzegorczyk [5] the other quantifiers are not even mentioned.
This postulate seems to be loosely connected to a remark by Gödel, possibly his only
recorded statement on the present topic. According to Wang [19, p. 266], Gödel said:
“Even though predicate logic is distinguished there are also other notions, such as many,
most, some (in the sense of plurality), and necessity.”
Despite the initial impression that the quantifier “many” is not definable, one could
try to define it formally, or to model it, by adding a variable σ and defining “many” as
more numerous than (the interpretation of) σ . This new variable can be either a numerical
one, interpreted as a cardinal number, or a set variable, interpreted as a certain set S. Then
“many x’s (satisfying ϕ)” is defined as having more members than S, or as the requirement
that the cardinality of the set of the values of x that satisfy the interpretation of ϕ is larger
than the cardinality of S. The set S depends on the context; it is chosen specifically for
each interpretation.
The problem with this attempt is that the definition of “a few” is the same, only with
“<” instead of “>”. And the phrase “more than a few” is formalized exactly as is “quite a
few” and “many”. And do we normally identify “many” with “more than a few”? Hardly.
The above remarks should be easy to understand, but an example can still be helpful.
A certain number, say 7, can play the role of both delimitations in the same discourse. For
example, if exactly 7 students among the 20 students in my Warsaw university class have
read more than ten books in their lifetime and 7 are pregnant, I would say that it is true
that “a few read books” and “many are pregnant”. (Incidentally, I believe that there could
exist schools somewhere in the world in which 7 pregnant among 20 students would be
seen as “few”, and 7 readers among them would be considered “many”.)
So everything depends on the context and introducing σ is of no help. Only the context
counts.
We can still maintain that the logical content is better explained, when this formaliza-
tion is made. The important feature—and a problem from a normal logical perspective—is
that “many” defined as “more than σ ” is not invariant with respect to isomorphisms. To
continue our example, (G, P) ∼ = (G, R), where G is the class, P is the set of pregnant
students in the class, R is the set of book readers in the class, but the sentence “there are
many x that ϕ ” is true in one and false in the other interpretation.
Dealing with context-dependent quantifiers one can wonder how many contexts there
are. Infinitely many? This seems probable, at least in the case of a quantifier “many”.
Is this the reason we are unable to pin them down? If only finitely many contexts were
possible, a fixed number, then perhaps we could give a definition by listing all the cases.
Notice that if a finite but practically unmanageable number of contexts has to be taken
into account then the quantifier is still undefinable by us. However, a sufficiently strong
intelligence, or even robot, could perhaps do that. The problem is analogous to the
problem whether a computer can handle the natural language. The hopes of some early
pioneers of Artificial Intelligence that computers would speak as humans were naive. Yet
in restricted settings, where contexts can be comprehensively listed, it is perfectly possible
to have computers “speak.”
5 Characterizing Context-Independent Quantifiers
It seems that context-independence means that any extralogical terms referring to some
specific fragments of the world are irrelevant for the understanding of the formula. The
topic covered in the statement is of no consequence, only logic counts. Thus
(1) Context-independent quantifiers = Topic neutral quantifiers
We can still maintain that a definition of (a quantifier) being context-independent is
needed. This is clear in specific cases, but can a general definition be given? What is
needed is a criterion—indeed, a context-independent criterion—for context-independence
of quantifiers (or perhaps even more generally, context-independence as such). The idea
is, of course, quite simple: there is no need for any specific knowledge about the world.
Yet, one could say, to understand the Magidor-Malitz quantifier one certainly needs some
non-trivial knowledge. It is, however, a purely logical knowledge (in the broad meaning
of logic), different from the knowledge of the features of the world, physical or social,
that are relevant for the specific situation. We might even try to say that what is needed
for understanding the context-independent quantifiers is the familiarity with merely the
necessary features of the world. One could ignore its contingent aspects.
700 S. Krajewski
It has been noticed above that when linguistic quantifier expressions are reconstructed
within logic the requirement of context-independence is formulated as invariance with
respect to isomorphisms. Thus, we get another thesis:
(2) Context-independent quantifiers = Quantifiers invariant under isomorphisms
Before another thesis proposing a characterization of context-independence of quantifiers
is attempted let us consider the meaning of being a thesis in this context. Church’s Thesis
is the best known example of a thesis identifying a formal concept with an intuitive one.
The mathematical concept of recursive function is identified with the intuitive concept
of effectively computable function. For a long time, the general conviction was that
such a thesis can be justified by various arguments, but there is no way to prove its
correctness because the intuitive concept is too vague to be part of a proof. However in
recent decades there have been various attempts (in particular by Robin Gandy, Wilfried
Sieg, Yuri Gurevich) to prove the identification. Namely, a proper analysis of the intuitive
concept of computability can provide principles that make possible a demonstration that
a function satisfying them must be recursive. There are more examples of similar theses,
for instance “the Cantor-Dedekind thesis” that real numbers are defined by the appropriate
set theoretic constructions. (For a discussion of Church’s Thesis and the other examples
as well as references to literature see, e.g., Krajewski [12].)
In the case studied in the present paper, it is the context-independence applied to
quantifiers that is the intuitive notion we want to characterize.
In addition to topic-neutrality and invariance under isomorphisms we can try look at
the ways the quantifier can be defined. It seems that whatever definition is formulated
it cannot be expressed without taking some specific logic into account. This is because
quantifiers are logical objects. They function inside a logical framework. On the other
hand, it would be hard to emphasize the logical nature but ignore any specific logic. The
way out of the dilemma can be as follows: the defining property is assumed to make
sense in whatever logic it is formulated. For instance, the phrase “ϕ(x, y) defines a well-
ordering” defines a type <2> quantifier, whether in second order logic or in set theory.
Any quantifier Q can give rise to a “logic” L(Q). Then Q is trivially definable in this
logic. To avoid this triviality, let us call a logic basic if it is first order, second order, nth
order, type theory or set theory. Hence the following thesis
(3) A quantifier is context-independent iff it is definable in some basic logic.
Because the common part of all such logics, as far as quantification is concerned, is the
universal quantifier ∀, we can reformulate the thesis as
(3’) A quantifier is context-independent iff it is ∀-definable in some (basic) logic.
Since we admit definability either in first order or second order or higher order logic or in
(formalized) set theory, and the general quantifier appears in each of these logics we can
say in short:
(4) A quantifier is context-independent iff it is definable in terms of ∀,
or briefly,
(Principal Thesis) Context-independence = definability in terms of ∀.
It is seen that the position of the general quantifier, or rather of our two familiar
quantifiers, ∀ and ∃, is vindicated. This is the fifth—in addition to simplicity, the faculty
of generalization, the measuring of complexity, and the expressive power—and rather
unexpected reason for distinguishing ∀: in the presence of the appropriate amount of
logical machinery but with no generalized quantifiers ∀ suffices to define all context-
independent quantifiers. Thus the power of the universal and existential quantifiers is
claimed to be even stronger than it seemed on the basis of the definability of so many
quantifiers by ∀. According to the Principal Thesis, the power of ∀, at least in relation to
quantifiers, extends to the whole realm of context-independence.
Let us repeat that in each basic logic the universal quantifier is included, so definability
in terms of ∀ is really the same as definability in (predicate) logic. The Principal Thesis
says that not only definability of generalized quantifiers in terms of ∀ gives context-
independence but also that context-independent generalized quantifiers are so definable.
Each specific example of a quantifier has been defined (in the proper logic) in terms of ∀;
the Thesis states the generalization to all possible quantifiers.
6 Formalism-Free Definition of ∀-Definable Quantifiers?
There exists an alternative way of looking at the Principal Thesis. If we agree to it then we
can treat the Thesis as the proposal to characterize the definability of quantifiers with the
use of ∀ (in basic logics) as context-independence, that is, a feature formulated without
the recourse to a specific syntactic machinery used in definitions. This brings to mind the
problem of formalism-free characterization of concepts.
The issue of “formalism freeness” has been introduced by Gödel who commented
on the fact that all formal definitions of computable functions give the same class of
functions. Therefore, even though each definition requires some specific formalism, we
have been able to isolate an important class of functions in a formalism-free manner.
Computability is formalism-free, and Gödel [4] proposed to look for a similar grasping of
definability and other notions. Some developments in mathematical logic, notably work
done in model theory by Shelah and Zilber, can be seen as going in this direction—see
Kennedy [9]. According to Shelah, in model theory, conceived as a tower, “the higher
floors do not have formulas or anything syntactical at all.” (Kennedy [9, p. 355])
In our case, the Principal Thesis gives the formalism-free characterization of the class
of quantifiers definable in some standard (predicate) logic, that is, using some formalism.
In each of these logics we have the quantifier ∀, so one can say that this is the class of
quantifiers definable in logic by ∀. The class consists of context-independent quantifiers.
This characterization is formalism-free.
References
1. Epstein, R.L.: Critical Thinking. Wadsworth/Thomson Learning, Belmont (1998)

2. Epstein, R.L.: The metaphysical basis of logic. Manuscrito XXII(2), 133–148 (1999)
702 S. Krajewski
3. Epstein, R.L.: Predicate Logic. The Semantic Foundations of Logic. Wadsworth/Thomson Learning,
Belmont (2001)
4. Gödel, K.: Remarks before the Princeton bicentennial conference on problems in mathematics. In:
Feferman, S., et al. (eds.) Kurt Gödel: Collected Works, Vol. II: Publications 1937–1958, pp. 150–
153. Oxford University Press, Oxford (1990)
5. Grzegorczyk, A.: An Outline of Mathematical Logic. Reidel, Holland (1974)
6. Henkin, L.: Some remarks on infinitely long formulas. Infinitistic Methods, pp. 167–183. Pergamon
7. Hintikka, J.: Selected Papers, vol. 2. Lingua Universalis vs. Calculus Ratiocinator: An Ultimate
Presupposition of Twentieth-Century Philosophy. Kluwer, Dordrecht (1997)
8. Keisler, H.J.: Logic with the quantifier “there exist uncountably many”. Ann. Math. Log. 1, 1–93
(1970)
9. Kennedy, J.: On formalism freeness: implementing Gödel’s 1946 Princeton bicentennial lecture. Bull.
Symb. Log. 19(3), 391–393 (2013)
10. Kleene, S.C.: Recursive predicates and quantifiers. Trans. AMS 53, 41–73 (1943)
11. Krajewski, S.: Remark on quantifiers and their relativisations. Acta Univ. Wratislaviensis 23, Logika,
37–41 (2005)
12. Krajewski, S.: Is Church’s Thesis Unique? In: Olszewski, A., Brożek, B., Urbańczyk, P. (eds.)
Church’s Thesis. Logic, Mind and Nature, pp. 113–135. Copernicus Center Press, Kraków (2014)
13. Krynicki, M., Mostowski, M., Szczerba, L. (eds.): Quantifiers: Logic, Models and Computation, vols.
I and II. Kluwer, Dordrecht (1995)
14. Lindström, P.: First order predicate logic with generalized quantifiers. Theoria 32(3), 186–195 (1966)
15. Mostowski, A.: On definable sets of positive integers. Fundam. Math. 34, 81–112 (1947)
16. Mostowski, A.: On a generalization of quantifiers. Fundam. Math. 44, 12–36 (1957)
17. Peters, S., Westerståhl, D.: Quantifiers in Language and Logic. Clarendon Press, Oxford (2006)
18. Russell, B.: The Principles of Mathematics. Allen and Unwin, London (1903)
19. Wang, H.: A Logical Journey. From Gödel to Philosophy. MIT Press, Cambridge (1996)
20. Westerståhl, D.: Generalized quantifiers. In: Stanford Encyclopedia of Philosophy. http://plato.
stanford.edu/entries/generalized-quantifiers/
S. Krajewski ()
Helena Rasiowa (1917–1994)
Andrzej Jankowski and Andrzej Skowron
The only way to rectify our reasoning is to make them as tangible as

those of the Mathematicians, so that we can find our error at a glance,
and when there are disputes among persons, we can simply say: Let us
calculate, without further ado, to see who is right.
G.W. Leibniz [6]
Abstract This is a biogram of Professor Helena Rasiowa (1917–1994) one of the leading
representatives of logicians from Warsaw. She was not only the great scientist but also the
great human being. Rasiowa influenced numerous researchers from all over the world,
especially by her results in algebraic logic, as well as by her great contribution for the
mathematical community in all respect. She is also co-founder of the Pawlak-Rasiowa
School of Artificial Intelligence (AI) (Jankowski and Skowron, Andrzej Mostowski and
Foundational Studies, pp 106–143, 2008).
Keywords Logic in Warsaw · Helena Rasiowa · Pawlak-Rasiowa School of Artificial

Intelligence (AI) · Metamathematics · Algebraic methods in logic · Foundations of
computer science · Logic in computer science and in AI
Mathematics Subject Classification (2000) Primary 03Fxx, 03Gxx, 03Bxx, 06Dxx,

06Exx, 68Txx; Secondary 03G27, 03G05, 03G05, 03B35, 03B45, 03B50, 03B55, 03B60,
03B70, 06D15, 06D20, 06D25, 06E15, 68T15, 68T27, 68T37, 97P20
Helena Rasiowa was born on June 23, 1917. During her school days and also later, she
attended piano classes at the Fryderyk Chopin Academy of Music. In 1938, she was
enrolled at the University of Warsaw (Faculty of Mathematics and Science). Among
her classmates there were well-known personalities like Wanda Szmielew and Roman
Sikorski who, after 1945, significantly contributed to the development of mathematics and
logic in Poland. She had the opportunity to attend lectures of renowned mathematicians
such as Wacław Sierpiński, Kazimierz Kuratowski, Jan Łukasiewicz, and Karol Borsuk.
In 1939, the World War II broke out, and Rasiowa had to interrupt her studies. She
established contact with her professors of mathematics and physics at the University
of Warsaw and resumed her studies at the underground university in German-occupied

704 A. Jankowski and A. Skowron
Warsaw functioning at significant risk in spite of prohibition by the German occupiers

of both secondary and university education. By July 1944, she passed all mandatory
exams required for the title of Master of Science in mathematics and prepared her MSc
thesis devoted to mathematical logic under the academic supervision of Jan Łukasiewicz.
The manuscript of her dissertation was burned during the Warsaw Uprising of 1944.
Helena Rasiowa and her mother were miraculously saved after spending numerous days
in different cellars, covered by the ruins of houses in the Old City in Warsaw that were
destroyed during the bombing. The two women left Warsaw and survived thanks to the
help of other members of their family.
When the war ended, Rasiowa rewrote her MSc thesis, and in 1945 she graduated from
the University of Warsaw with MSc in mathematics. From September 1945 to November
1946, she worked at the Wojciech Górski’s secondary school in Warsaw, pursuing her
graduate studies at the Section of the Philosophy of Mathematics, supervised by Andrzej
Mostowski. On November 1, 1946, she became the assistant lecturer at the University of
Warsaw, where she worked for the rest of her life. During her work at the University of
Warsaw, she was granted various academic degrees and titles, and she gradually promoted
to academic positions like assistant professor, associate professor, and then full professor.
Her first publications, Axiomatisation d’un système partiel de la théorie de la déduction
and Sur certaines matrices logiques, came out in 1947 [7]. In 1950, she received doctorate
in mathematics and science based on her PhD thesis, Algebraic treatment of the functional
calculi of Lewis and Heyting, under the academic supervision of Andrzej Mostowski. The
thesis reflects the main theme of her future research on algebraic methods in logic. In 1956
Rasiowa received her academic degree Doctor of Science in Mathematics (equivalent to
habilitation today) at the Institute of Mathematics of the Polish Academy of Sciences.
For the habilitation degree she submitted two papers, viz., Algebraic models of axiomatic
theories and Constructive theories. Together these papers formed a dissertation entitled
Algebraic models of elementary theories and their applications. Between 1954 and 1957
she held a post of Associate Professor there, jointly with an analogous position at the
University, and became Professor in 1957 and subsequently Full Professor in 1967.
During that time, she wrote about 20 different papers coauthored by Roman Sikorski,
Andrzej Mostowski, Andrzej Białynicki-Birula, and Jerzy Łoś. In 1957, she was granted
the title of a Professor by the Central Certifying Board and in 1967, the title of a Full
Professor by the Council of State.
As an academic, Helena Rasiowa represented, with pride and dignity, the Polish
school of logic, which was popularized during the interwar period by such scientists
as Jan Łukasiewicz, Stanisław Leśniewski, Kazimierz Ajdukiewicz, and Alfred Tarski.
Her research from the 1950s (including her PhD dissertation, habilitation thesis and
numerous other publications, co-authored by other well-known scientists with whom she
worked at that time) led to a deeper understanding of the possibilities related to the use
of algebraic and topological methods in logic. These concerned both the classical and
non-classical predicate calculus (including intuitionistic logic, modal logics, and many-
valued logics). In the academic year of 1949/1950, Helena Rasiowa participated in the
lectures by Stanisław Mazur, devoted to “constructive” analysis. At that time, she became
familiarized with the results of the research on “constructive” models of computation in
analysis. That research was initiated between 1936 and 1939 as part of the collaboration
Helena Rasiowa (1917–1994) 705
between Stefan Banach and Stanisław Mazur. Later on, Helena Rasiowa and Andrzej
Grzegorczyk [4] published lecture notes out of that lectures.
The interest in algebraic methods of logic (including their applications in theorem
proving) and models of computations for identifying constructive fragments of mathe-
matical theories were naturally connected with the research of logical foundations of the
emerging computer science and studies in the field of AI (e.g., methods of automatic
theorem proving). As a result in 1960s, a group of researchers around Professor Rasiowa
started conducting studies in such fields as automatic theorem proving, algorithmic
logic and, later in 1970s, program logic, approximate reasoning, and other domains of
foundations of computer science. For approximately 30 years, that is till the end of her life,
Professor Rasiowa continued that research together with her students and collaborators.
Her work helped to open many important projects related to applications of logic in the
foundations of computer science. Moreover, together with Zdzisław Pawlak, she founded
a scientific journal, known as Fundamenta Informaticae, devoted to the foundations of
computer science. From the very beginnings of that journal in 1977 till the end of her life,
she was its editor-in-chief. Thanks to her unimaginably hard scientific and organizational
work, Fundamenta Informaticae managed to gain international renown. She was an active
Collecting Editor with Studia Logica (since 1974) and Associate Editor of the Journal of
Approximate Reasoning (since 1986).
Her devotion to work bore fruits in the fields of AI and computer science, resulting
in the creation of an entire school of students and collaborators, who still continue the
research initiated by Rasiowa and Pawlak. The school is sometimes referred to as the
Polish Pawlak-Rasiowa School of AI [5].
The establishment and development of the School emphasize considerable contribu-
tions of Professor Rasiowa in the development of research devoted to the application of
mathematical logic in Computer Science and AI. Much earlier than other mathematicians
of her generation, she realized that computer science may be a great inspiration for logic.
She also paid attention to the need of research on logical foundations of computer science
by actively participating in their development. By 1960s she envisioned the significance
of logic for the development of theoretical computer science and AI. She was aware
of the facts that numerous scientific problems of theoretical computer science may be
solved by methods developed in logic, and logical research in computer science may
constitute an important source of inspiration for the development of logic. Helena Rasiowa
inspired to conduct intensive, fruitful research on program logics, methods of reasoning
with incomplete information and logical calculi for AI systems. As the head of the
Section of Mathematical Logic at the Institute of Mathematics (University of Warsaw),
she conducted seminars and research projects devoted to these fields.
It is worth recalling that one of the first great scientists to propose the research
on relationships between algebraic systems and logical systems was George Boole
(who referred to the ideas formulated by, e.g., Leibniz). The works by Boole and his
successors led to the development of the concept of Boolean algebra and helped to
establish its relationship with the classical predicate calculus. This relationship is naturally
based on an algebraic construction introduced by Lindenbaum and Tarski, based on the
equivalence classes generated by formulas. The Stone’s representation theorem applied to
the Lindenbaum–Tarski algebra contributed to the development of research on topological
properties of models for logic.
The second very important algebraic construction, which defines the relationships
between logic and algebra, was initiated by the research of Łukasiewicz and Post. This
approach treats formulas as algebraic functions over particular algebras. Key algebraic
methods in the research concerning intuitionistic logic and modal logic were developed
by Marshall Stone, Alfred Tarski, and John McKinsey. Helena Rasiowa became actively
involved in this research area. Together with Roman Sikorski, she developed the first
algebraic proof of the Gödel’s completeness theorem for the classical predicate calculus.
Next, she used an algebraic method to prove analogous theorems for the predicate calculus
of intuitionistic logic and modal logics. This enabled her to obtain algebraic proofs of a
number of other important theorems concerning both classical logic and numerous non-
classical logics. Since that moment, numerous logicians and mathematicians in the world
have started to use analogous algebraic methods in their research on logical systems.
The first monograph by Helena Rasiowa, The Mathematics of Metamathematics, which
she co-authored with Roman Sikorski, was published in 1963 [10]. It contains a systematic
and comprehensive overview of the results related to the algebras of logical calculi
(including classical, intuitionistic, modal, and positive logic). The monograph exerted a
profound influence on numerous logicians in the world. Let us quote professor Melvin
Fitting [3]:
[. . . ] the ideas were of a sort I had never come across before, and I was enchanted. This way of
using algebra, producing known results in classical logic, then applying similar techniques to non-
classical logics to get new results|it all seemed like magic. Profound results fell out so effortlessly,
it seemed. How could one read this book and remain unaffected?
In the Rasiowa–Sikorski monograph we may find, e.g., a very interesting proof of the
Rasiowa–Sikorski lemma, based on a popular (among the representatives of the Polish
School of Mathematics) method of proving existential theorems by means of the Baire
category theorem. This method was used by such well-known mathematicians as Stefan
Banach, Hugo Steinhaus, and Stanisław Mazur. At the same time, we become familiarized
with numerous interesting topological properties of models for the classical predicate
calculus. Numerous generalizations of the Rasiowa–Sikorski lemma to other algebraic
structures (e.g., Post’s algebras, certain types of distributive lattices in the form of the
Rauszer–Sabalski lemma) were created. Moreover, there appeared versions of the proof
of the Rasiowa–Sikorski lemma that did not used the Baire category theorem.
In The Mathematics of Metamathematics, algebraic models for the predicate calculus,
introduced by Rasiowa and Sikorski, also are presented. These include Boolean models
for the classical predicate calculus and Heyting models for the intuitionistic predicate
calculus. Dana Scott and Robert Solovay used the concept of Boolean models to simplify
a well-known Cohen’s proof of the independence of the axiom of choice, along with
Cantor’s continuum hypothesis. In this simplification, the key role is played by both
Boolean models and the Rasiowa–Sikorski lemma itself [2]. In 1973, Denis Higgs showed
that there are some strong relationships between algebraic semantics of Rasiowa and
Sikorski and the semantics based on topos of sheaves [2]. Later, these relationships were
independently developed by Dana Scott and his students. This led to further modifications
of the Cohen’s method and its generalizations in topos theory.
Another well-known monograph of Helena Rasiowa, Algebraic Approach to Non
Classical Logics [8], presents algebraic theories for a wide class of non-classical logics.
It encompasses positive and classical implicational logics, logics weaker than the positive
implicational logic, minimal logic, positive logic with semi-negation, constructive logics
with strong negations and Post’s logics. The aim of the book was to identify a class of
logics that is as broad as possible and for which a general algebraic theory could be
formulated. The developed theory consists of tools for proving significant theorems about
the logics from this class simultaneously for the entire class, and not—as it was treated
earlier—separately for each logic.
Helena Rasiowa is the author of about one hundred scientific works and the academic
textbook, viz., The Introduction to Contemporary Mathematics, which gained popularity
both in Poland and abroad. Her works and monographs helped to educate whole
generations of logicians in Poland and abroad. In different countries, one can meet
scientists who after studying the monographs and works of Professor Helena Rasiowa
proudly call themselves her students, despite the fact that they have never met her in
person. She educated many generations of students and young academic researchers
and was an academic supervisor of nearly 20 PhDs, of which several obtained titles
of professors. This includes: Michael Bleicher (1961, Tulane University), Vladimir G.
Kirin (1966, University of Zagreb), Andrzej Salwicki (1969), Nguen Cat Ho (1971,
Vietnam Academy of Sciences), Cecylia Rauszer (1971), Ewa Orłowska (1971), Graźyna
Mirkowska (1972), Maria Semeniuk-Polkowska (1972), Wiktor Bartol (1973), Antoni
Kreczmar (1973), Jerzy Tiuryn (1975), Lech Banachowski (1975), Anita Wasilewska
(1975), Michał Krynicki (1976), Bolesław Szymański (1976), Dimiter Vakarelov (1977,
Sofia University), Bogdan Sabalski (1977), Halina Przymusińska (1979), and Leszek
Rudak (1986).
The extremely high position of Helena Rasiowa within the international scientific
environment resulted in numerous invitations from the best universities and research
centers in Europe and both Americas. She gave lectures, including frequent plenary
lectures, during numerous international congresses and conferences. She also chaired
many scientific sessions.
During her last 2 years of life, Helena Rasiowa, not succumbing to the aggravating
disease, continued working on her new monograph, Algebraic models of logics, of which
she is the author of eight chapters [9]. In June 1994, she started the preparation of a
special edition of Studia Logica, entitled Reasoning with Incomplete Information, and, in
particular, prepared the list of its topics.
The readers are referred to [1] for additional details on the scientific achievements of
Helena Rasiowa, in particular for the full list of her publications.
It should be emphasized that despite of her significant scientific activity, Professor
Rasiowa did not refuse taking various roles that were important for the scientific
and academic life. The university was her second home. She was a person of great
perseverance. In retrospect, it is amazing to realize how she was able to meet all these
challenges.
From 1955 to 1958, she was the Director of the Extramural Studies at the Institute of
Mathematics (University of Warsaw). From 1964 to 1970, she headed the Section of the
Foundations of Mathematics at the University of Warsaw and from 1970 till the end of
her life, she was the head of the Section of Mathematical Logic at the same university. For
more than 15 years she was the Dean of the Faculty of Mathematics, Computer Science
and Mechanics (1958–1960, 1962–1966, 1968–1978). She also represented the Faculty
Council of the Faculty of Mathematics, Computer Science and Mechanics to the Senate of
the University of Warsaw. Helena Rasiowa actively participated in the scientific life and
performed a number of important functions in academic organizations and associations
both in Poland and abroad. From 1961 to 1968, she was a Scientific Secretary of the
Institute of Mathematics of Polish Academy of Sciences, and later, of its board member.
From 1970 to 1972 she was the member of the Committee of Mathematical Assessors at
the Central Council of Higher Education and from 1968 to 1972 she was a Chairperson
of the Committee of Mathematical Assessors for Teacher’s Colleges. Professor Rasiowa
was active in the Polish Mathematical Society (PMS). In PMS, she was a member of
the board in the Warsaw Chapter, responsible for the popularization of mathematics and
training of teachers. She was appointed as the President of the Warsaw Chapter of PMS
twice (1957–1958, 1963–1964) and she was also the Secretary of the Society (1955–1956)
and its Vice-President (1958) as well. In the Association for Symbolic Logic, she was a
Council member (1958–1960) and a member of the Executive Committee for Foreign
Affairs (1972–1974). In 1972, she was an Alternate Assessor (1972–1975) and Assessor
(1975–1979) in the Division of Logic, Methodology and Philosophy of Science of the
International Union of History and Philosophy of Sciences. For several years, she was the
Chairperson of the Scientific Council of Computational Center by the Polish Academy
of Sciences and later of the Institute of the Foundations of Computer Science of the
Academy, as well.
Thanks to her active participation, efforts aimed at the creation of the Polish Associ-
ation for Logic and Philosophy of Science were undertaken between 1980–1981. This
association is the continuation of the Polish Logic Society, founded on April 22, 1936
by Jan Łukasiewicz and Alfred Tarski. The creation of the society, whose aim was to
“practice and nurture logic and methodology, along with their history, didactics and
possible application,” was connected with a great development of logic in the interwar
Poland. The Polish Logical Society was the second organization of this type in the world
(after the Association for Symbolic Logic in the USA). After the war, the society was not
reactivated. The efforts aimed at the creation of a similar society in 1980–1981 turned
out futile as the authorities refused to register it. During the IX International Congress
of Logic and Philosophy of Science in Uppsala in 1991, a group of Polish congress
participants met to initiate the establishment of the association. On December 10, 1991, a
constituent meeting of the Polish Association for Logic and Philosophy of Science took
place.
It is possible to identify numerous areas of contemporary research, which are directly
or indirectly inspired by the works of Helena Rasiowa. These include algebraic methods
in logic, deductive systems like tableau, mathematical foundations of computer science,
and artificial intelligence (including approximate reasoning and interactive granular
computations).
Helena Rasiowa was an extremely hard-working, friendly and cheerful person, who has
been always kind and sensitive to human problems with great intelligence. She was always
ready to help, not only as a scientist, but also as a human being in real-life problems. In
each, even most difficult situation, she tried to do as much good as possible—this was her
life motto as an academic teacher, devoted wholeheartedly to logic and Polish science.
She was able to share the love for her children, Krystyna Kijewska and Zbigniew Raś,
with the love for mathematics.
Professor Helena Rasiowa was a Franciscan tertiary, a fact which, during her life,
was known only to some of her loved ones. The faith combined with a passion for
mathematics, enabled her to survive the most difficult moments, like the war or the very
last stage of her life, when the incurable disease began to destroy her. This was a great
lesson for her students about how to behave in the most difficult situations.
Acknowledgements Authors are grateful to Victor Marek and Soma Dutta for their corrections and other
helpful suggestions.
References
1. Bartol, W., Orłowska, E., Skowron, A.: Helena Rasiowa, 1917–1994. Modern Log. Int. J. Hist. Math.
Log. Set Theory Found. Math. 5(3), 231–248 (1995)
2. Bell, J.L.: Set Theory Boolean-Valued Models and Independence Proofs. Oxford Logic Guides, vol.
47. Oxford University Press, Oxford (2011)
3. Fitting, M.: A tribute to professor Helena Rasiowa. In: Orlowska, E. (ed.) Logic at Work. Essays
Dedicated to the Memory of Helena Rasiowa, pp. v–viii. Physica, Heidelberg (1998)
4. Grzegorczyk, A., Rasiowa, H.: Computable analysis. Rozprawy Matematyczne, vol. 33, pp. 1–111.
PWN, Warsaw (1963)
5. Jankowski, A., Skowron, A.: Logic for artificial intelligence: the Rasiowa–Pawlak school perspective.
In: Ehrenfeucht, A., Marek, V., Srebrny M. (eds.) Andrzej Mostowski and Foundational Studies,
pp. 106–143. IOS Press, Amsterdam (2008)
6. Leibniz, G.W.: Dissertatio de Arte Combinatoria. Leipzig (1666)
7. Rasiowa, H.: Axiomatisation d’un système partiel de la théorie de la déduction and Sur certaines
matrices logiques. C. R. Soc. Sci. Lett. Varsovie, C1 III, 22–37 (1947)
8. Rasiowa, H.: Algebraic Approach to Non Classical Logics. Studies in Logic and the Foundations of
Mathematics, vol. 78. North Holland, Amsterdam (1974)
9. Rasiowa, H.: Algebraic Models of Logics. Warsaw University, Warsaw (2001)
10. Rasiowa, H., Sikorski, R.: The Mathematics of Metamathematics. Monografie Matematyczne, vol.
41, 1st edn. (1963), 2nd edn. (1968), 3rd edn. (1970). PWN, Warsaw
A. Jankowski
The Dziubański Foundation of Knowledge Technology, Warsaw, Poland
A. Skowron ()
Institute of Mathematics, University of Warsaw, Warsaw, Poland
Systems Research Institute, Polish Academy of Sciences, Warsaw, Poland
Post Algebras in the Work of Helena
Rasiowa
Ewa Orłowska
Abstract A survey of some classes of Post algebras is given including the class of plain
semi-Post algebras, Post algebras of order m, m>1, as its particular instance, Post algebras
of order ω+ , and Post algebras of order ω + ω∗ . Representation theorems for each of the
classes are given. Some examples of the algebras in the classes are constructed.
Keywords Ordered algebraic structures · Lattice · Boolean algebra · Post algebra ·

Representation theorem · Many-valued logic
Mathematics Subject Classification (2000) Primary 03G20, 06D25
Emil Post doctoral dissertation [24] (see also Post [25]) provided a description of an n-
valued, n > 1, functionally complete algebra. The notion of Post algebra was introduced
in Rosenbloom [43]. Then was the paper by Wade [53] and in Rousseau [44, 45] an
equivalent formulation of Post algebra was given which became a starting point for an
extensive research. Various generalizations of Post algebras inspired by computer science
have been proposed. In 1970-ies Helena Rasiowa and Tadeusz Traczyk run a seminar on
Post algebras in the Department of Mathematics of the University of Warsaw, gathering
the participants both from the Polish universities as well as guests and PhD students from
abroad. This paper is a survey of major classes of Post algebras which were the subject
of research of Helena Rasiowa at that time and were studied by the participants of the
seminar. The present paper is based on the earlier paper included in:
Iturrioz, L., Orłowska E., Turunen E. (eds.): Atlas of Many-valued Logics. Mathemat-
ics Report 75 of the Department of Information Technology of the Tampere University of
Technology, 2000.
The Atlas was elaborated by the participants of the COST Action 15 “Many-valued
Logics for Computer Science Applications”, 1994–1999.

712 E. Orłowska
1 Plain Semi-Post Algebras
These algebras were introduced and investigated in Cat Ho [1], Cat Ho and Rasiowa [2–
4]. Let (T , ) be a poset. A subset s of T is an ideal provided that s = ∅ and for all t ∈ s
and w ∈ T , if w t then w ∈ s. Let ET be the set of ideals of T together
with the empty
set ∅. Clearly, T ∈ ET. It is known that any s ∈ ET is of the form s = {s(t) : s(t) ⊆ s},
where s(t) = {w ∈ ET : w t}. The system (ET, ⊆) is a complete lattice, where join
and meet are set-theoretical union and intersection, respectively.
An abstract algebra
(P) P = (P , ∪, ∩, →, ¬, {dt : t ∈ T }, {es : s ∈ ET})
where ∪, ∩, → are 2-argument operations, ¬, dt for t ∈ T are unary operations and es for
s ∈ ET are 0-argument operations (constants) is a plain semi-Post algebra (psP-algebra)
of type T provided that the following conditions are satisfied:
(p0) (P , ∪, ∩, →, ¬) is a Heyting algebra with the zero element 0 = e∅ and the unit
element 1 = eT ,
for any a, b ∈ P
(p1) dt (a ∪ b) = dt a ∪ dt b,
(p2) dt (a ∩ b) = dt a ∩ dt b,
(p3) dw dt a = dt a,
(p4) dt es = 1 if t ∈ s, otherwise dt es = 0,
(p5) dt a ∪¬dt a = 1,
(p6) a = {es(t ) ∩ dt a : t ∈ T } where is the least upper bound in P .
Let P = (P , ∪, ∩, →, ¬, {dt : t ∈ T }, {es : s ∈ ET}) be a psP-algebra of type (T , ). By
BP we denote the set of elements of P of the form dt a, t ∈ T .
Proposition 1.1
(a) BP is closed under the operations ∪, ∩, →, ¬ of P .
(b) The algebra BP = (BP , ∪, ∩, →, ¬, 1, 0) is a Boolean algebra.
Let CP be the set of all complemented elements in the distributive lattice (P , ∪, ∩). Then
Proposition 1.2
(a) CP is closed under the operations ∪, ∩, →, ¬ of P .
(b) The algebra Cp = (CP , ∪, ∩, →, ¬, 1, 0) is a Boolean algebra.
(c) For every a ∈ CP , dt ¬a = ¬dt a, t ∈ T .
Note that BP and CP do not always equal. Consider a poset (T , ) such that T = {a, b, c}
and = {(b, a)}. Then ET = {∅, {b}, {c}, {b, c}, {a, b}, T }, BP = {∅, T }, and CP =
{∅, T , {c}, {a, b}}.
Proposition 1.3 (Epstein Lemma) For any set {aj : j ∈ J } of elements in P it holds

(a) a = P {aj : j ∈ J } iff for every t ∈ T dt a = BP {dt aj : j ∈ J },

(b) a = P {aj : j ∈ J } iff for every t ∈ T dt a = BP {dt aj : j ∈ J },

where P , P , BP , BP denote infinite joins and meets in the algebras P and BP ,
respectively.
Post Algebras in the Work of Helena Rasiowa 713
Proposition 1.4

(a) dt (a → c)
= {dw a → dw c : w t}
(b) dt ¬a = {¬dw a : w t}
(c) dw a dt a whenever w t, for any w, t ∈ T
(d) a b iff dt a dt b for all t ∈ T
(e) ew et iff w ⊆ t, for any w, t ∈ ET
It follows that every psP-algebra of type (T , ) uniquely determines a set of infinite meets
of P

M(P ) = {dw a → dw c : w } : t ∈ T .
Observe that for any sets s , s ∈ ET there exists the relative pseudo-complement s → s
defined by

s → s = {s ∈ ET : s ∩ a ⊆ s }
and the pseudo-complement ¬s defined by

¬s = s → ∅ = {s ∈ ET : s ∩ s = ∅}.
Clearly, s → s , ¬s ∈ T .
Proposition 1.5
(a) For any poset (T , ), the system (ET, ∪, ∩, →, ¬, T , ∅), where ∪, ∩ are set-
theoretical operations of union and intersection, respectively, and →, ¬ are defined
as above, is a Heyting algebra with the unit element T and zero element ∅.
(b) Given a psP-algebra P = (P , ∪, ∩, →, ¬, {dt : t ∈ T }, {es : s ∈ ET}), let EP =
{es : s ∈ ET}. Then (EP, ) is a poset isomorphic to (ET, ⊆).
Condition (b) follows from Proposition 1.4(e).
Example 1.1 An important example of a psP-algebra is the following algebra, referred to
as a basic psP-algebra:
(ET, ∪, ∩, →, ¬, {dt : t ∈ T }, {es : s ∈ ET})
where (ET, ∪, ∩, →, ¬) is the Heyting algebra defined above and the operations dt , t ∈
T , and es , s ∈ ET, are defined by:
es = s, in particular e∅ = ∅ and e∅ = T ,
dt s = T if t ∈ s, otherwise dt s = ∅.
714 E. Orłowska
Proposition 1.6 The basic psP-algebra is functionally complete, that is any n-argument
operation f : ETn → ET, n = 0, 1, . . . , is definable with the operations of this algebra.
Given a Boolean algebra B = (B, ∪, ∩, →, ¬, 1B , 0B ) and a poset (T , ) by a
descending T -sequence of elements of B we mean an indexed family (bt )t ∈T of elements
of B such that w t in T implies bt bw in B (for the sake of simplicity we denote
the Boolean ordering of B with the same symbol). We say that B and T satisfy condition
(erpc) of existence of relative pseudo-complement if
(erpc) For any two descending T -sequences b = (bt )t ∈T , c = (ct )t ∈T of elements of B
there exists B {bw → cw : w t} for all t ∈ T .
Example 1.2 We present a psP-algebra PT (B) of type T determined by a Boolean algebra
B = (B, ∪, ∩, →, −, 1B , 0B ) such that B and T satisfy condition (erpc). The universe
P (B) of PT (B) is the set of all descending T -sequences of elements of B. We define a
partial ordering on P (B) as follows. Let b = (bt )t ∈T and c = (ct )t ∈T be any elements
of P (B). Then
b c in P (B) iff bt ct in B for all t ∈ T .
The system (P (B), ) is a lattice with join and meet defined by
b ∪ c = (bt ∪ ct )t ∈T , b ∩ c = (bt ∩ ct )t ∈T .
Since B and T satisfy (erpc), for any b, c in P (B) there exists the relative pseudo-
complement b → c and

b → c = (xt )t ∈T , where xt = {bw → cw : w t}.
For every s ∈ ET we define
es = (xt )t ∈T , where xt = 1B if t ∈ s, otherwise xt = 0B .
Moreover, we put
dw b = (xt )t ∈T , where xt = bw for every t ∈ T ,

¬b = (xt )t ∈T , where xt = {−bw : w t}.
It is easy to verify that the algebra
PT (B) = (P (B), ∪, ∩, →, ¬, {dt : t ∈ T }, {es : s ∈ ET})
defined above is a psP-algebra of type (T , ).

Proposition 1.7 Let a Boolean algebra B and a poset (T , ) satisfying (erpc) be given.
Let P be the algebra PT (B) defined as in Example 1.2. Then the algebra BP (see
Proposition 1.4) is isomorphic to B.
Example 1.3 A particular instance of the algebra defined in Example 1.2 is a set algebra
obtained by taking the field of all subsets of a set as the respective Boolean algebra. Let
U be a nonempty set and let B(U ) be the field of all subsets of U . We have 1B(U ) = U
and 0B(U ) = ∅. For any poset (T , ), B(U ) and T satisfy condition (erpc). Let P (B(U ))
be the set of all descending T -sequences of sets from B(U ). The ordering on P (B(U ))
is the set inclusion. The algebra PT (B(U )) = (P (B(U )), ∪, ∩, →, ¬, {dt : t ∈ T }, {es :
s ∈ ET}) defined as in Example 1.2 is a psP-algebra of type (T , ). The infinite joins in
the axiom (p6) are set unions.
Proposition 1.8 (Representation Theorem) Let P = (P , ∪, ∩, →, ¬, {dt : t ∈ T },
{es : s ∈ ET}) be a psP-algebra of type (T , ). If T is denumerable and either well-
founded or the set M(P ) is denumerable (in particular if P is denumerable), then for
any denumerable set Q of infinite joins and meets in P there exists the field B(U ) of all
subsets of a nonempty set U and a monomorphism h from P into PT (B(U )) preserving
all the operations in Q.
2 Post Algebras of Order m
The first axiom system for the algebras characterising Post’s m-valued logics, for
a finite m greater than 2, was presented in Rosenbloom [43]. He called them Post
algebras. The axiomatisation was then simplified in Epstein [9] and Traczyk [48].
Traczyk proved the equational definability of the class of Post algebras. Over the
years the theory of Post algebras and several generalizations of these algebras have
been developed. Here we define Post algebras of order m as a particular case of psP-
algebras.
Let (Tm , ) be a poset such that Tm = {1, . . . , m − 1}, where m is a natural number
greater than 2, and is a natural ordering in Tm . Then ETm = {∅, s(1), . . . , s(m − 1)},
where s(t) = {w ∈ Tm : w t}. Clearly, (ETm , ⊆) is isomorphic to {0, 1, . . . , m −
1} with the natural ordering. Hence, we can identify these two posets and assume that
constants es are indexed with elements from {0, 1, . . . , m − 1}.
By a Post algebra of order m we mean a psP-algebra of type (Tm , ).
It can be easily shown that this definition is equivalent to the standard definition of
Rousseau [44, 45].
Example 2.1 A classical example of a Post algebra of order m is an m-element Post
algebra such that P = {e0 , . . . , em−1 }, and for i, j ∈ {0, 1, . . . , m − 1} the operations
in P are defined as follows:
(ex1) ei ∪ ej = emax(i,j ) ,
(ex2) ei ∩ ej = emin(i,j ) ,
(ex3) ei → ej = 1 if i j , otherwise ei → ej = ej ,
(ex4) ¬ei = ei → 0,
(ex5) di ej = 1 if i j , otherwise di ej = 0.
716 E. Orłowska
Proposition 2.1
(a) e0 → a =
em−1 ,
(b) es → a = {dt a ∩ et : t s} ∪ ds a,
(c) em−1 → a = a,
(d) a → es = es ∪ ¬ds+1a, for s = 0, . . . , m − 2,
(e) a → em−1 = em−1 .
We define disjoint operations cs for s ∈ {0, 1, . . . , m − 1} as follows:
(c1) c0 a = ¬d1 a = ¬a,
(c2) cs a = ds a ∩ ¬ds+1a for s ∈ T \ {m − 1},
(c3) cm−1 a = dm−1 a.
We clearly have
cs a ∩ ct a = eo for s = t.
Any element a of P has the following disjoint representation:

(c4) a = {ct a ∩ et : t ∈ T }.
Theorems analogous to Propositions 1.1–1.8 hold and constructions from Examples 1.2
and 1.3 carry over to the case of type (Tm , ). Algebras presented in Example 1.1 can
be identified with those defined in Example 2.1. Moreover, the algebras BP and CP
corresponding to a Post algebra of order m coincide.
The m-valued Post logic is a propositional logic with binary connectives ∧, ∨, →,
unary connectives ¬, Dt for t ∈ T , and propositional constants Es for s ∈ {0, 1, . . . , m −
l}. The algebraic semantics for the logic is determined in the standard way by the class
of Post algebras of order m. A Hilbert-style axiomatisation of m-valued Post logic and
its completeness with respect to the algebraic semantics is presented in Rasiowa [30].
The main results on m-valued Post logic include: Model existence theorem (Rasiowa
[31]), Craig interpolation theorem (Rasiowa [31]), Herbrand theorem (Perkowska [23]).
Applications of the m-valued Post logic are concerned with the theory of programming.
An algorithmic logic based on m-valued Post logic is developed in Perkowska [23].
Post algebras and logics of any finite type (T , ) are considered in Nour [16]. They
are also treated in Konikowska, Morgan and Orłowska [13].
3 Post Algebras of Order ω+
Let (Tω , ) be a poset such that Tω = ω is the set of natural numbers and is the natural
ordering of natural numbers. Then ETω = {∅, s(1), s(2), . . . , Tω }. Clearly, (ETω , ⊆) is
isomorphic to {0, 1, 2, . . . , ω} with the natural ordering. Hence, we can identify these two
posets and assume that constants es are indexed with elements from {0, 1, 2, . . . , ω}.
A Post algebra of order ω+ is a psP-algebra of type (Tω , ).
An example of a Post algebra of order ω+ can be defined in a way similar to that
developed in Example 2.1.
Theorems analogous to Propositions 1.1–1.8 hold and constructions from examples 1.2
and 1.3 carry over to the case of type (Tω , ). Moreover, the algebras BP and CP
corresponding to a Post algebra of order ω+ coincide.
Representation theory for Post algebras of order ω+ has been also developed in
Maksimova and Vakarelov [14], Rasiowa [39].
A Hilbert-style axiomatisation of ω+ -valued Post predicate logic and its completeness
with respect to the algebraic semantics is presented in Rasiowa [34]. The other results on
ω+ -valued Post logic include: Kripke style semantics (Maksimova and Vakarelov [15],
Vakarelov [52]), Herbrand theorem and a resolution-style proof system (Orłowska [17–
19]), relational semantics and a relational proof system (Orłowska [21]).
Applications of the ω+ -valued Post logic are concerned with the theory of program-
ming. An algorithmic logic based on ω+ -valued Post logic is developed and investigated
in Rasiowa [36–38].
4 Post Algebras of Order ω + ω∗
These algebras are introduced and investigated in Epstein and Rasiowa [11, 12]. Let T =
{1, 2, . . . , −2, −1} and E = {0, 1, 2, . . . , −2, −1}. A Post algebra of order ω + ω∗ is an
algebra of the form (P) in Sect. 1 satisfying axioms (p0)–(p6), where in (p0) 0 = e0 and
1 = e−1 , and the following

(p7) d1 a = d−1 a ∪ {ds a ∩ ¬ds+1 a : 1 s −1} pivot elimination axiom
(p8) (a → b) ∪ (b → a) = 1.
The axiom (p7) says that an element e such that et e (dt e = 1) for all positive t and
e < et (dt e = 0) for all negative t does not exist.
Propositions analogous to Propositions 1.1–1.4 hold for Post algebras of order ω + ω∗ .
Moreover, the algebras BP and CP corresponding to a Post algebra of order ω + ω∗
coincide.
Example 4.1 A most natural example of a Post algebra of order ω + ω∗ is a linear Post
algebra of order ω + ω∗ defined as follows:
P = {es : s ∈ E}, and the operations in P are defined with

conditions analogous to (ex1)–(ex5) from Example 2.1.
Disjoint operations in Post algebras of order ω + ω∗ can be defined with conditions

analogous to (cl), (c2), (c3) from Sect. 2 by replacing m − 1 with −1. Then any element
a of P has a disjoint representation given by condition (c4).
In Post algebras of order ω + ω∗ one can define arithmetic-like operations in the
following way.
718 E. Orłowska
The successor sa (the predecessor pa) of an element a of P is an element given by the

following disjoint representation

sa = {ct a ∩ et +1 : t ∈ E}

pa = {ct a ∩ et −1 : t ∈ E}
provided that either of these exist.

The inverse −a of an element a is given by the disjoint representation

−a = {ct a ∩ e−1 : t ∈ E}
provided that it exists.

Addition and multiplication operations have disjoint representations as follows

a+b = {ct (a + b) ∩ et : t ∈ E}

where for each t ∈ T the infinite join ct (a + b) = {ci a ∩ cj b : i + j = t} exists,

a·b = {ct (a · b) ∩ et : t ∈ E}

where for each t there is the finite join ct (a · b) = {ci a ∩ cj b : ij = t}.
Proposition 4.1 A Post algebra of order ω +ω∗ with inverse, addition and multiplication
is a commutative ring with unit, where the ring zero is e0 and the ring unit is e1 .
These rings have the characteristic 0.
For a descending T -sequence X = (Xt )t ∈T of sets from the field B(U ) of all subsets
of a nonempty set U we define

X+ = {Xt : t positive}

X− = {Xt : t negative}.
It can be shown that the algebra of descending T -sequences X = (Xt )t ∈T of sets from
B(U ) such that X+ = X− , with the operations defined as in Example 1.2, is a Post
algebra of order ω + ω∗ .
Representation theorem for Post algebras of order ω + ω∗ has the following form.
Proposition 4.2 (Representation Theorem) For every denumerable Post algebra P of
order ω + ω∗ there is a monomorphism h of P into a Post set algebra of order ω + ω∗
whose elements are descending T -sequences X = (Xt )t ∈T of sets from the field B{U )
of all subsets of a nonempty set U such that X+ = X− . Moreover, h preserves a given
denumerable set Q of infinite joins and meets of P.
Applications of the logic are concerned with approximation reasoning. An approximation

reasoning to recognise a subset S of a nonempty universe U is understood as a process of
gradual approximating S by
subsets of U S ⊆ S1 ⊆ S2 ⊆ . . . which cover S

and subsets . . . ⊆ S−2 ⊆ S−1 ⊆ S which are contained in S.
Then the approximations of set S are defined as follows:

S+ = {St : t positive}

S− = {St : t negative}.
In Epstein and Rasiowa [12] a characterisation of sets S such that S + = S − is given.

Post algebras of order ϑ, where ϑ is an arbitrary ordinal number are introduced and
investigated in Przymusinska [26–29].
Acknowledgment Support from the National Science Centre project DEC-2011/02/A/HS1/00395 is

gratefully acknowledged.
References
1. Cat-Ho, N.: Generalized post algebras and their applications to some infinitary many-valued logics.
Diss. Math. 57, 1–76 (1973)
2. Cat-Ho, N., Rasiowa, H.: Semi-Post algebras. Stud. Logica. 46(2), 147–158 (1987)
3. Cat-Ho, N., Rasiowa, H.: Plain semi-Post algebras as a poset-based generalization of Post algebras
and their representability. Stud. Logica. 48(4), 509–530 (1989)
4. Cat-Ho, N., Rasiowa, H.: LT-fuzzy logics. In: Zadeh, L.A., Kacprzyk, J. (eds.) Fuzzy Logic for
Management of Uncertainty, pp. 121–139. Wiley, New York (1992)
5. Chang, C.C., Horn, A.: Prime ideals characterisation of generalized Post algebras. Proc. Symp. Pure
Math. 2, 43–48 (1961)
6. Dwinger, Ph.: Notes on Post algebras I, II. Indag. Math. 28, 462–478 (1966)
7. Dwinger, Ph.: Generalized Post algebras. Bull. Acad. Pol. Sci. Math. 16, 559–563 (1968)
8. Dwinger, Ph.: A survey of the theory of Post algebras and their generalizations. In: Dunn, J.M.,
Epstein, G. (eds.) Modern Uses of Multiple-Valued Logic, pp. 53–75. Reidel, Dordrecht (1977)
9. Epstein, G.: The lattice theory of Post algebras. Trans. Am. Math. Soc. 95, 300–317 (1960)
10. Epstein, G., Horn, A.: P-algebras. An abstraction from Post algebras. Algebra Univers. 4, 195–206
(1974)
11. Epstein, G., Rasiowa, H.: Theory and uses of Post algebras of order ω + ω∗ , Part I. In: Proceedings
of the 20th International Symposium on Multiple-Valued Logic, Charlotte, NC, pp. 42–47 (1990)
12. Epstein, G., Rasiowa, H.: Theory and uses of Post algebras of order ω + ω∗ , Part II. In: Proceedings
of the 21th International Symposium on Multiple-Valued Logic, Victoria, pp. 248–254 (1991)
13. Konikowska, B., Morgan, Ch., Orłowska, E.: Relational formalisation of arbitrary finite valued logics.
Log. J. Interest Group Pure Appl. Logic 6(5), 755–774 (1998)
14. Maksimova, L., Vakarelov, D.: Representation theorems for generalized Post algebras of order ω+ .
Bull. Acad. Pol. Sci. Math. 22(8), 757–764 (1974)
15. Maksimova, L., Vakarelov, D.: Semantics for ω+ -valued predicate calculi. Bull. Acad. Pol. Sci. Math.
22(8), 765–771 (1974)
720 E. Orłowska
16. Nour, A.: Etude de modeles logiques. Extensions de la logique intuitionniste. PhD dissertation,
University Joseph Fourier, Grenoble (1997)
17. Orłowska, E.: The Herbrand theorem for ω+ -valued logic. Bull. Acad. Pol. Sci. Math. 25(11), 1069–
1071 (1977)
18. Orłowska, E.: The resolution principle for ω+ -valued logic. Fundamenta Informaticae 2, 1–15 (1978)
19. Orłowska, E.: Resolution systems and their applications II. Fundamenta Informaticae 3, 333–362
(1980)
20. Orłowska, E.: Mechanical proof methods for Post logics. Log. Anal. 110–111, 173–192 (1985)
21. Orłowska, E.: Post relation algebras and their proof system. In: Proceedings of the 21th International
Symposium on Multiple-Valued Logic, Victoria, pp. 298–305 (1991)
22. Perkowska, E.: Herbrand theorem for theories based on m-valued logics. Bull. Acad. Pol. Sci. Math.
19, 893–899 (1971)
23. Perkowska, E.: On algorithmic m-valued logics. Bull. Acad. Pol. Sci. Math. 20, 717–719 (1972)
24. Post, E.: Introduction to a general theory of elementary propositions. Bull. Am. Math. Soc. 26 (1920)
25. Post, E.: Introduction to a general theory of elementary propositions. Am. J. Math. 43, 163–185 (1921)
26. Przymusińska, H.: Craig interpolation theorem and Hanf number for ϑ-valued infinitary predicate
calculi. Bull. Acad. Pol. Sci. Math. 28, 207–211 (1980)
27. Przymusińska, H.: Gentzen type semantics for ϑ-valued infinitary predicate calculi. Bull. Acad. Pol.
Sci. Math. 28, 203–206 (1980)
28. Przymusińska, H.: On ϑ-valued infinitary predicate calculi. Bull. Acad. Pol. Sci. Math. 28, 193–198
(1980)
29. Przymusińska, H.: Some results concerning ϑ-valued infinitary predicate calculi. Bull. Acad. Pol. Sci.
Math. 28, 199–202 (1980)
30. Rasiowa, H.: A theorem on existence of prime filters in Post algebras and the completeness theorem
for some many-valued predicate calculi. Bull. Acad. Pol. Sci. Math. 17, 347–354 (1969)
31. Rasiowa, H.: Ultraproducts of m-valued models and a generalization of the Löwenheim-Skolem-
Gödel-Malcev theorem for theories based on m-valued logics. Bull. Acad. Pol. Sci. Math. 18, 415–420
(1970)
32. Rasiowa, H.: The Craig interpolation theorem for m-valued predicate calculi. Bull. Acad. Pol. Sci.
Math. 20, 341–346 (1972)
33. Rasiowa, H.: Formalised ω+ -valued algorithmic systems. Bull. Acad. Pol. Sci. Math. 21, 559–565
(1973)
34. Rasiowa, H.: On generalized Post algebras of order ω+ and ω+ -valued predicate calculi. Bull. Acad.
Pol. Sci. Math. 21, 209–219 (1973)
35. Rasiowa, H.: An Algebraic Approach to Non-classical Logics. North Holland and Polish Scientific
Publishers, Amsterdam, Warsaw (1974)
36. Rasiowa, H.: Extended ω+ -valued algorithmic logic (a formalised theory of programs with recursive
procedures). Bull. Acad. Pol. Sci. Math. 22, 605–610 (1974)
37. Rasiowa, H.: Many-valued algorithmic logic as a tool to investigate programs. In: Epstein, G.,
Dunn, J.M. (eds.) Modern Uses of Multiple–Valued Logic, pp.77–102. Reidel, Dordrecht (1977)
38. Rasiowa, H.: Algorithmic logic. Multiple-valued extension. Stud. Logica. 38(4), 317–325 (1979)
39. Rasiowa, H.: Topological representations of Post algebras of order ω+ and open theories based on
ω+ -valued Post logic. Stud. Logica. 44(4), 353–368 (1985)
40. Rasiowa, H.: Towards fuzzy logic. In: Zadeh, L.A., Kacprzyk, J. (eds.) Fuzzy Logic for Management
of Uncertainty, pp. 5–25. Wiley, New York (1992)
41. Rasiowa, H.: Axiomatization and completeness of uncountably-valued approximation logic. Stud.
Logica. 53(1), 137–160 (1994)
42. Rasiowa, H., Epstein, G.: Approximation reasoning and Scott’s information systems. In: Proceedings
of the 2nd International Symposium on Methodologies for Intelligent Systems, North Holland, pp.
33–42 (1987)
43. Rosenbloom, P.C.: Post algebras, I. Postulates and general theory. Am. J. Math. 64, 167–188 (1942)
44. Rousseau, G.: Logical systems with finitely truth values. Bull. Acad. Pol. Sci. Math. 17, 189–194
(1969)
45. Rousseau, G.: Post algebras and pseudo-Post algebras. Fundam. Math. 67, 133–145 (1970)
46. Saloni, Z.: Gentzen rules for the m-valued logic. Bull. Acad. Pol. Sci. Math. 20, 819–826 (1972)
47. Traczyk, T.: Some theorems on independence in Post algebras. Bull. Acad. Pol. Sci. Math. 10, 3–8
(1963)
48. Traczyk, T.: An equational definition of a class of Post algebras. Bull. Acad. Pol. Sci. Math. 12,
147–149 (1964)
49. Traczyk, T.: On Post algebras with uncountable chain of constants. Algebras and homomorphisms.
Bull. Acad. Pol. Sci. Math. 15, 673–680 (1967)
50. Traczyk, T.: Prime ideals in generalized Post algebras. Bull. Acad. Pol. Sci. Math. 16, 369–373 (1968)
51. Urquhart, A.: An interpretation of many-valued logic. Z. Math. Logik Grundlagen der Mathematik
19, 111–114 (1973)
52. Vakarelov, D.: Lattices related to Post algebras and their applications to some logical systems. Stud.
Logica. 36, 89–107 (1977)
53. Wade, L.I.: Post algebras and rings. Duke Math. J. 12, 389–395 (1945)
54. Zabel, N.: New techniques for automated deduction in many-valued first order, finite and infinite
logics. PhD thesis, Institut National Polytechnique de Grenoble (1993)
E. Orłowska ()
National Institute of Telecommunications, Warszawa, Poland
Andrzej Grzegorczyk, a Logician Par
Excellence
Abstract A short biography of Andrzej Grzegorczyk (1922–2014) is presented, listing

his main accomplishments in logic, mentioning his philosophical views, followed by a list
of all his books and a selection of main papers. (This paper is based on the biographies
listed in Part III of the references section.)
Keywords Andrzej Grzegorczyk · Logic · Undecidability · Rationalism ·

Anti-psychologism
1 Life
Andrzej Grzegorczyk was born in Warsaw on August 22, 1922, as the only son of Piotr, a
historian of Polish literature, and Zofia who was a physician. He obtained his high school
certificate in 1940, already during World War II. He studied physics and philosophy at
the clandestine Polish university in Warsaw. He was first attracted to logic before the war
when he heard on the radio a lecture of Jan Łukasiewicz on Stoic logic.
After the war Grzegorczyk lived in Cracow, where he obtained MA in philosophy, and
in 1946 he returned to Warsaw, where he became an assistant to Władysław Tatarkiewicz
and a secretary of Przeglad ˛ Filozoficzny (Philosophical Review). He began an intensive
study of logic and the foundations of mathematics, in part because this was the safest area
of philosophy during the Communist rule. In 1950, he obtained PhD at the University
of Warsaw. His dissertation, On Topological Spaces in Topologies without Points, was
written under the supervision of Andrzej Mostowski. He was admitted to the Institute of
Mathematics of the Polish Academy of Sciences, where after Habilitation he became a
“docent” in 1953, an associate professor in 1961, and a full professor in 1972. Until 1968,
he also lectured at the University of Warsaw. In 1974, Grzegorczyk moved to the Institute
of Philosophy of the Polish Academy of Sciences, and became the head of its Ethics
Group in 1982. He retired in 1990. From 1999 to 2003 he served as the chairman of
the Committee of Philosophy of the Polish Academy of Sciences. It is worth mentioning

724 S. Krajewski
that after a series of lectures at the University of Amsterdam in 1965, he was offered a
permanent position there, but he refused since he did not want to leave Poland.
Often involved in organizing scholarly activities, he headed the 1973 Logical Semester
that inaugurated the International Mathematical Center (the Banach Center) of the Polish
Academy of Sciences. At that time it provided a rare occasion for extensive contacts of
mathematicians from the Soviet bloc with those from the West. Cooperation with Russia
and Ukraine was important for Grzegorczyk, particularly after 1990. In 1995–1997 he led
a special project, “One Hundred Years of the Lvov-Warsaw School”, which culminated in
conferences in Lvov and Warsaw. Grzegorczyk participated in many other conferences in
logic and philosophy. He served as an assessor in the Executive Committee of the Interna-
tional Union of History and Philosophy of Science, the Division of Logic, Methodology
and Philosophy of Science, and in 1979 he was elected to the Institut International de
Philosophie. He received two honorary doctorates: from University of Clermont-Ferrand
(2010) and from Jagiellonian University in Kraków (2013). He died in 2014.
In 1953 Andrzej Grzegorczyk married Renata Majewska, who would become a
professor of Polish language and literature at the University of Warsaw. They have two
children and six grandchildren.
In addition to his scholarly activities, Grzegorczyk was also a writer and was
involved in public activities. During the Communist period, he signed some petitions
protesting against limitations of freedom. Ha had life-long interest in the ideology of non-
violence, a method of fighting without violence. Quite early he showed serious interest in
environmental issues: in the 1970s, he popularized the warnings of the Club of Rome.
Grzegorczyk was a devout Roman Catholic, but he was a very independent Christian
and was often critical of the official Church policies. He was also open to other faiths and
churches, especially the Russian Orthodox Christianity. What is even more important,
he was on good terms with intellectuals of various ideological convictions, religious and
atheist, pro-Communist and anti-Communist.
2 Logical Accomplishments
Grzegorczyk’s best known papers belong to logic and the foundations of mathematics.
One of his early papers, the interpretation of Lesniewski’s ontology as Boolean algebra
without 0 (see [P5]), suggested, according to him, that Leśniewski’s calculus of names
added nothing interesting to logic.
In widely quoted 1953 paper Some Classes of Recursive Functions [2], he described
and investigated classes of recursive functions obtainable by superposition, restricted
recursion and the operation of restricted minimum (f is defined form g and h by: f(n) =
the smallest x<h(n) such that g(n,x)=0) from some initial functions containing addition,
in the next step multiplication, then raising to power, then superpowering, etc. This
leads to a subrecursive hierarchy, that is, the so-called Grzegorczyk hierarchy. It forms
a strictly increasing infinite sequence of classes of functions such that its union is equal to
the class of primitive recursive functions. The third class of the Grzegorczyk hierarchy
is identical with the class of elementary functions, definable as the smallest class of
functions containing addition and subtraction, and closed with respect to superposition
as well as restricted summation and restricted multiplication.
Andrzej Grzegorczyk, a Logician Par Excellence 725
Grzegorczyk continued the investigations in computable analysis that had been

initiated by Banach and Mazur (see [P4, P6, P7, P9]). In particular, he co-edited (with
Rasiowa) Mazur’s posthumously published book Computable Analysis. Grzegorczyk also
gave various definitions of computable real numbers. In this way he wanted to apply
the concept of effectiveness used in the arithmetic of natural numbers in mathematical
analysis. He also studied (see [P16]) computable functionals of higher types.
In 1958 Grzegorczyk co-authored (with Mostowski and Ryll-Nardzewski) a fundamen-
tal paper “The classical and the omega-complete arithmetic” ([P10], also [P12]) about
second-order arithmetic, which treats it as formalized in first-order logic, but makes it
possible to speak about natural numbers as well as their sets. Due to the introduction of
the omega-rule, the 11 -relations are representable in it.
Working on computability, decidability and undecidability, Grzegorczyk considered
various proofs of undecidability using recursively enumerable sets which are not recursive
(see [P8]), and proved that elementary topological algebra, that is, Boolean algebra
with closure, is undecidable, because arithmetic is interpretable in it (see [P3]). He also
showed (see [P14]) that the calculus of combinators has no recursive model. His most
recent contribution, “Undecidability without arithmetization” (2004, see [P26]), deals
with the theory of concatenation of texts (conceived as a primitive notion), introduced
by Tarski in the 1920s. The theory of concatenation was shown by Grzegorczyk to be
undecidable; therefore, it can replace arithmetization in metamathematics. Computability
is replaced here by the effective recognizability of texts. The arithmetical relativization
of quantifiers can be replaced by the relativization of expressions to subexpressions of a
given expression.
Grzegorczyk, like other Polish logicians and mathematicians, for example, Tarski,
Mostowski, Rasiowa, Sikorski, was interested in intuitionistic logic and constructive
mathematics. Whereas in his work on constructivism he looked at constructive mathe-
matics from the point of view of classical methods, he believed that intuitionists may
“perhaps penetrate more deeply into the meaning of logical concepts than classical logic
does.” Grzegorczyk’s semantics for intuitionistic logic introduced in “A philosophically
plausible formal interpretation of intuitionistic logic” (1964) (see [P15, P17]) is similar to
the much better known construction by Kripke. It is based on the phrase “in my inquiry,
I am forced to assert the sentence A at the moment t,” and the notion of strong assertion,
that is, an assertion that may not be abandoned at a later moment.
A modal interpretation of Grzegorczyk’s semantics for intuitionism leads to the system
named by Boolos S4.Grz, that is, the modal system S4 plus the formula
((A ⇒ A) ⇒ A) ⇒ A,
called Grzegorczyk’s axiom. This system has applications in the provability logic.
Grzegorczyk considered logic as belonging to the foundations of mathematics. In this
field he worked according to Tarski’s program: every formal method is admissible in
foundational research, independently of whether it is finitist, constructive or completely
infnitistic.
726 S. Krajewski
3 Views
Among his ten philosophical books there is a study Logic—a Human Affair (1997),
and (all in Polish): Schemata and the Human Being (1963), Philosophy in the Time of
Challenge (1979), Ethics in Internal Experience (1989), Life as a Challenge. Introduction
to Rationalistic Philosophy (1993), Europe, Discovering the Sense of Existence (2001),
as well as a book of short literary forms Moral Stories (1986). He also published over
60 papers on mathematical and philosophical logic and over a hundred articles on ethics,
religion, social issues.
Grzegorczyk approved Łukasiewicz’s statement that logic is the morality of speech and
thought. Logic was conceived by Grzegorczyk in a broad sense as including semiotics
and the methodology of science. According to him, logic in this sense constitutes a
basic ingredient of European rationalism (see [18]). Since his youth he was convinced
that everything can be expressed precisely and logically. He believed that European
rationalism brought an exceptional intellectual and cultural success, and therefore must
be protected. Teaching logic and proliferating its standards is the way.
For Grzegorczyk, logic is a human affair. He maintained that semantic relations are al-
ways of someone and for someone. This leads to the acceptance of psychologism in logic.
He also claimed that anti-psychologism inevitably leads to idealism. Grzegorczyk argued
that paradoxes should not be interpreted as showing that our language is inconsistent, but
rather that our concepts and conceptual systems are limited. For example, the Grelling
antinomy shows that there exists a set of expressions which cannot be correctly named
(see [P22]) and the Liar paradox demonstrates the existence of problems about which
nobody can think consistently, sincerely, and in a fully conscious way. In this approach,
semantic antinomies appear as laws of thinking (see [17]).
Grzegorczyk also studied axiomatic geometry based on the concept of solid (see
[P13]). He continued some works by Tarski, but he also motivated his work by referring
to reism, a view formulated by Tadeusz Kotarbiński who claimed that individual spatio-
temporal concrete things are the only elements of the furniture of the world. Grzegorczyk
was acquainted with reism by Henryk Hiż, his clandestine teacher of logic. Grzegorczyk
was always strongly attracted to this approach (see [P1, P2, P19, P24]), and believed that
reism is in principle a good ontology for natural science. On the other hand, he clearly
saw difficulties and limitations of reism in mathematics, particularly in set theory since it
makes talking about infinite sets problematic.
Grzegorczyk was a Christian and believed that Christianity represents the values of
European rationalism. According to him, the history of Christianity shows the sense of
the world by revealing the realm of the sacred and transcendence. He also admitted the
presence of the highest values in other religions or ideologies. Humanity is not reducible
to biological facts. European rationalism is open to moral values. Jesus’s testimony
represents the highest pattern of morality. At the same time, Grzegorczyk was ready to
(verbally) accept the need for some use of force in order to promote rational standards.
Grzegorczyk was strongly attracted to the attitude of radical non-violence, as repre-
sented by Mahatma Gandhi or Martin Luther King. He helped organize symposia on this
issue. He was uncompromising in recommending compromise in all conflicts. We should
talk with everyone, he claimed, also with a terrorist. He was also one of the first people
in Poland accepting ecological challenges; he proposed limitations on consumption and

opposed waste, although such appeals sounded highly abstract in Poland of the 1970s.
He considered moral issues and social challenges to be so important in the contem-
porary world that he criticized as immoral the focus, so common among the brightest
minds, on abstract intellectual questions which can have no impact on social problems.
His move from the Institute of Mathematics to the Institute of Philosophy and Sociology
was partly motivated by this frustration. He felt he participated in the malaise. After all, he
avowed that for him mathematical problems in formal logic were an “obsession, addiction,
narcotic.”
He believed that we, humans, need solidarity of all people, independently of where they
live. He even appealed to the UN in the 1970s proposing to establish a law that everybody
is entitled to help every person who is in a worse situation, wherever the other person
lives. In particular, richer people or societies should help poorer ones, the stronger should
protect the less powerful, the better educated should do something for the less educated,
etc.
Grzegorczyk’s religiosity and his understanding of religion provided an extension or
correction to his views on the supremacy of logic in all reasoning, including the moral
debates. In 1974 he wrote a poem in which it was stated that logic is powerless in dealing
with religious matters: “I do not refute those who say You don’t exist. I agree with the
friends who say that Your concept is contradictory. . . only those have a pure concept of
God who don’t have it at all. . . . ”
4 Influence
Grzegorczyk will remain known due to his logical achievements: the Grzegorczyk
hierarchy, results about undecidability, second-order arithmetic, the S4Grz system of
modal logic, semantics for intuitionistic logic, his concatenation theory. Also his course
in mathematical logic An Outline of Mathematical Logic (in Polish 1961, in English 1974,
see [6, 8]) was for a long time the only detailed textbook in Polish in which logical
calculus, model theory and recursion theory were presented in a balanced fashion, and
influenced many students of mathematics and philosophy in Poland. His textbook on the
foundations of arithmetic An Outline of Theoretical Arithmetic (1971, in Polish, see [7]),
in which he simultaneously developed arithmetic and exposed its logical basis, was much
less influential. He popularized recursive functions and the problems of computability and
decidability not only in Poland but also in France (see [3, 4]).
Grzegorczyk was productive but he was always working alone as it was never easy to
work with him. He established no school, and hardly had students. He supervised only
two PhD dissertations, one in logic and one in ethics. The one in logic was mine, but as a
matter of fact I received little from him; I owed my logical education primarily to Andrzej
Mostowski.
Grzegorczyk was not a member of any ideological group, and in no circle was
he considered fully “one of our guys”. At the same time he was widely respected
for being absolutely honest and sincere. Whereas he was recognized as an important
728 S. Krajewski
logician, Grzegorczyk’s philosophical and axiological views did not become important
or influential. His other works were rarely noticed. There was a notable exception: in
1987 he obtained a literary prize for the book “Moral stories”.
In addition to particular formal achievements two general main messages are expressed
in Grzegorczyk’s work: first, the utmost importance of logical thinking, second, the all-
human solidarity. I believe that for all who knew him he can easily remain the best
available personification of logic, or the individual embodying the ideal type of a logician.
This is a rather ambivalent tribute, even among logicians.
References
In Part I all Grzegorczyk’s books are listed, and in Part II a selection of his articles. A
fairly comprehensive list is available in the paper A1 mentioned below.
Books
1. Logika popularna [Popular Logic, in Polish]. PWN, Warsaw (1955). (2nd ed. (1958); 3rd ed. (1961);
Czech trans. (1957); Russian trans. (1965))
2. Some Classes of Recursive Functions. Rozprawy Matematyczne Nr IV. Instytut Matematyczny PAN,
Warsaw (1953)
3. Zagadnienia rozstrzygalności [Problems of Decidability, in Polish]. PWN, Warsaw (1957)
4. Fonctions Récursives. Gauthier-Villars, Paris (1961)
5. Schematy i człowiek [Schemata and Human Being, in Polish]. Biblioteka Wiȩzi. Znak, Kraków
(1963)
6. Zarys logiki matematycznej [An Outline of Mathematical Logic, in Polish]. Biblioteka Matematy-
czna tom 20. PWN, Warsaw (1961) (subsequent improved editions up to 6th ed. (1984))
7. Zarys arytmetyki teoretycznej [An Outline of Theoretical Arithmetic, in Polish]. Biblioteka Matem-
atyczna tom 37. PWN, Warsaw (1971) (2nd ed. (1983))
8. An Outline of Mathematical Logic. Reidel, Dordrecht (1974). English version of Zarys logiki
matematycznej [An Outline of Mathematical Logic, in Polish]. Biblioteka Matematyczna tom 20.
PWN, Warsaw (1961) (subsequent improved editions up to 6th ed. (1984))
9. Filozofia czasu próby [Philosophy in the Time of Challenge, in Polish]. Édition du Dialogue, Paris
(1979) (2nd ed.: PAX, Warsaw (1984))
10. Próba treściowego opisu świata wartości i jej etyczne konsekwencje [An Attempt of a Contetual
Description of the World of Values, and Its Ethical Consequences, in Polish]. Zakład Psychologii
PAN, Ossolineum, Wrocław (1983)
11. Moralitety [Moral Stories, in Polish]. PAX, Warsaw (1986)
12. Etyka w doświadczeniu wewnȩtrznym [Ethics in Internal Experience, in Polish]. PAX, Warsaw
(1989)
13. Mała propedeutyka filozofii naukowej [Small Introduction to Scientific Philosophy, in Polish] PAX,
Warsaw (1989)
14. Życie jako wyzwanie. Wprowadzenie w filozofiȩ racjonalistyczna˛ [Life as a Challenge. Introduction
in Rationalistic Philosophy]. IFiS PAN, Warsaw (1993) (2nd ed. (1995))
15. Життя як виклик [Life as a Challenge, in Ukrainian, with modifications]. Scholar, Warsaw—
Lvov (1997). Transl. and commentaries by Борис Домбровский, Олег Гiрний
16. Жизнь как вызов [Life as a Challenge, in Russian, with modifications]. Вузовская Книга,
Moscow (2000). Transl. by Виктор Макаренко
17. Logic—a Human Affair. Scholar, Warsaw (1997)
18. Europa—odkrywanie sensu istnienia [Europe, Dicovering the Sense of Existence, in Polish]. Fun-
dacja Akademii Teologii Katolickiej dla Wspierania Rozwoju Chrześcijańskich Nauk Społecznych
i Społecznej Gospodarki Rynkowej, Warsaw (2001)
19. Psychiczna osobliwośćczłowieka [Psychological Peculiarity of Man, in Polish] Scholar, Warsaw
(2003)
Selected Papers and Articles
P1. Próba ugruntowania semantyki jȩzyka opisowego. Przeglad ˛ Filozoficzny 44, 330–371 (1949)
P2. The pragmatic foundations of semantics. Synthèse 8, 300–324 (1950–1951)
P3. Undecidability of some topological theories. Fundam. Math. 38, 137–152 (1951)
P4. Elementary definable analysis. Fundam. Math. 41, 311–338 (1954)
P5. The systems of Leśniewski in relation to contemporary logical research. Stud. Logica. 3, 77–97
(1955)
P6. Computable functionals. Fundam. Math. 42, 168–202 (1955)
P7. On the definition of computable functionals. Fundam. Math. 42, 232–239 (1955)
P8. Some proofs of undecidability of arithmetic. Fundam. Math. 43, 166–177 (1956)
P9. On the definitions of computable real continuous functions. Fundam. Math. 44, 61–71 (1957)
P10. Mostowski, A., Ryll-Nardzewski, C.: The classical and the omega-complete arithmetic. J. Symb.
Log. 23, 188–206 (1958)
P11. Some approaches to constructive analysis. In: Constructivity in Mathematics, pp. 43–61. North
P12. Mostowski, A., Ryll-Nardzewski, C.: Definability of sets in models of axiomatic theorems. Bulletin
de l’Académie Polonaise des Sciences, Série Math. 9, 163–167 (1961)
P13. Axiomatizability of geometry without points. Synthèse XII(2/3), 228–235 (1960)
P14. A theory without recursive models. Bulletin de l’Académie Polonaise des Sciences, Série Math. 10,
63–69 (1962)
P15. A philosophically plausible formal interpretation of intuitionistic logic. Indag. Math. XXVI, 596–
601 (1964)
P16. Recursive objects in all finite types. Fundam. Math. 54, 73–93 (1964)
P17. Assertions depending on time and corresponding logical calculi. Compos. Math. 20, 83–87 (1968)
P18. The moral basis for peace: the absolute value of the human individual. Dialect. Hum. 1(1), 19–28
(1974)
P19. On certain formal consequences of reism. Dialect. Hum. 1, 73–80 (1977)
P20. Classical, relativistic and constructivist ways of asserting theorems. In: Woleński, J. (ed.) Philosoph-
ical Logic in Poland, pp. 19–31. Kluwer, Dordrecht (1994)
P21. Non-violence: wychowanie do negocjacji, demokracji i współistnienia. In: Edukacja wobec wyzwań
XXI wieku, pp. 57–92. Komitet Prognoz Polska 2000 PAN, Warszawa (1996)
P22. The paradox of Grelling and Nelson presented as a veridical observation concerning naming. In:
Kijania-Placek, K. (ed.) The Lvov-Warsaw School and Contemporary Philosophy, pp. 183–190.
Kluwer, Dordrecht (1998)
P23. Is Antipsychologism Still Tenable? In: Woleński, J., Köhler, E. (eds.) Alfred Tarski and the Vienna
Circle, pp. 109–114. Kluwer, Dordrecht (1999)
P24. Reizm i analiza logiczna (Spojrzenie w epoce postmodernizmu). In: Polska Filozofia Analityczna,
pp. 31–41. Toruń, Poland (1999)
P25. The Vocation of Europe. Dialogue and Universalism IX(5–6), 11–41 (1999)
P26. Undecidability without Arithmetization. Stud. Logica. 79(2), 163–230 (2005)
730 S. Krajewski
Selected Articles on Andrzej Grzegorczyk
A1. Krajewski, S.: Andrzej Grzegorczyk (1922–2014) [in Polish]. Stud. Semiotyczne XXVIII–XXIX,
63–88 (2015)
A2. Krajewski, S., Woleński, J.: Andrzej Grzegorczyk: Logic and Philosophy. Fundam. Inf. 81(1–3),
1–10 (2007)
S. Krajewski ()
A Mystery of Grzegorczyk’s Logic
of Descriptions
Joanna Golińska-Pilarek and Taneli Huuskonen
It is important to approach the deeds of previous generations with the

right attitude, accepting the fact that their actions were an actual step
of their development. Regardless of any alternative ways in which
reality could have evolved, we must not belittle their thoughts and
actions, but rather augment them, keeping in mind all that our
ancestors could not have taken into account. So, let us not claim that
we are building a new philosophy on new principles, as we are not
essentially wiser than our predecessors, but let us utilize the whole
record of the past, carefully considering their whole evolution
(biological and cultural). Their experiences form the basis of our
growth.
Andrzej Grzegorczyk
Abstract In 2011, Andrzej Grzegorczyk formulated Logic of Descriptions (LD), a new

logical system in which the classical equivalence has been replaced with the descriptive
equivalence. Two sentences are descriptively equivalent whenever they describe the same
state of affairs. Grzegorczyk’s logic LD is built from the ground up by revising the
axioms of classical propositional logic and rejecting those that do not correspond to
the intended interpretation of the descriptive equivalence as the connective expressing
equimeaning relations between sentences. Grzegorczyk’s last paper, which introduced in
detail philosophical motivations of LD and its axiomatization, has become an inspiration
for investigating the properties of LD and its various modifications. In this paper we
present the basics of Grzegorczyk’s logic LD and then we survey the recent results on LD
that have shed light on mysterious properties of the Grzegorczyk’s descriptive equivalence
connective.
Keywords Logical connectives · Paradoxes of classical logic · Descriptive

equivalence · Equimeaning · Grzegorczyk’s logic of descriptions
Mathematics Subject Classification (2000) Primary 03A05; Secondary 03-02, 03B60

732 J. Golińska-Pilarek and T. Huuskonen
1 Introduction
Andrzej Grzegorczyk devoted the last 5 years of his life to the study of a new logical
system avoiding weaknesses and paradoxes of the classical logical connectives. His
main goal was to build a logic determined by the logical structure and properties of
descriptions of the world that people use in everyday life. Grzegorczyk emphasized very
strongly the extremely positive role played by formal and metamathematical studies in the
development of science in the twentieth century. He appreciated the particularly powerful
impact of classical logic on the development of information technology and artificial
intelligence. Yet at the same time, he felt that the time had come to reject the current
paradigm and to propose another way of describing the meaning of logical concepts, not
yet used by researchers, and based on the analysis of the role of human language in its
entirety to the challenge of the human condition.
The main philosophical assumption of Grzegorczyk’s standpoint was that in the human
description of the cognized world’s phenomena, the roles of negation, conjunction, and
disjunction differ significantly from those of implication and equivalence. Negation,
conjunction, and disjunction are very primitive and have clear intuitive descriptive
meaning, while the classical implication and equivalence are derivative and have no
intuitively plausible sense. Furthermore, it is exactly implication and equivalence that
are responsible for some paradoxical laws of classical logic, such as “false implies
everything”, “truth is implied by anything” and “all true sentences are logically equivalent
to each other”.
As a consequence, states Grzegorczyk, we are forced to accept that among all the
logical connectives exactly negation, conjunction, and disjunction—together with the
equimeaning connective (or descriptive equivalence) expressing the assertion that two
descriptions have the same meaning—are well suited as the primitive concepts of a new
logic. Indeed, those four connectives are crucial and necessary for descriptive practice,
that is, for the way people actually describe reality. As descriptions and descriptive
equivalences among them have became crucial for Grzegorczyk’s approach, he called
his new logical system the Logic of Descriptions, or LD for short.
The first exposition of Grzegorczyk’s new logic, its philosophical motivations and
assumptions was published in 2011 in the paper [5], cf. [6]. In the paper, Grzegorczyk
proposed a number of axioms and rules that the equimeaning connective (descriptive
equivalence), denoted by ≡, should satisfy and he posed a number of open problems,
in particular whether the new connective ≡ is different than the classical equivalence.
Grzegorczyk’s study on the logic of descriptions was quickly joined by other researchers,
in particular the authors of this paper. In 2012 in the paper [3], we published the
first results on the formal properties of LD, showing that the descriptive equivalence
connective is essentially different than the classical one and the logic itself is indeed new.
Further results are presented in the forthcoming paper [4].
Grzegorczyk’s Logic of Descriptions 733
Andrzej Grzegorczyk was informed about our results, he actively participated in

seminars and meetings devoted to the logic of descriptions, where we had lively
discussions about Grzegorczyk’s motivations and the adequacy of his axioms. Toward
the end of his life, Andrzej Grzegorczyk was working on a deeper exposition of the whole
philosophical system underlying the logic LD. He was writing a book concerned with a
complete philosophical theory of acquisition of logical connectives, making use of results
from empirical sciences and linguistics. Unfortunately, professor Grzegorczyk did not
finish this work before his death in 2014.
The aim of the present paper is to survey the current research results on Grzegorczyk’s
Logic of Descriptions LD and some of its variants. The paper consists of three sections and
conclusions. In Sect. 2, we present the basics of the logic LD, its language, axiomatization
and semantics. Section 3 is concerned with the exposition of the formal properties of
LD, in particular those that we consider to be the most amazing and striking. In Sect. 4
we discuss some extensions and modifications of the logic LD. The paper ends with the
conclusions section, in which we list a few open problems, among others.
2 The Logic of Descriptions LD
The logic LD is a propositional logic. The vocabulary of LD consists of the following

pairwise disjoint sets of symbols:
• V = {p0 , p1 , p2 , . . .}—an infinite countable set of propositional variables,
• {¬, ∧, ∨, ≡}—propositional operations of negation ¬, conjunction ∧, disjunction ∨,
and descriptive equivalence ≡.
The set of LD is defined in a standard way as the smallest set that contains all the
propositional variables and is closed under the propositional operations of LD. We will
use the following four shorthand notations for LD-formulas:
df
• (p → q) = (¬p ∨ q) (classical implication)
df
• (p ↔ q) = (p → q) ∧ (q → p) (classical equivalence)
df
• (p ⇒ q) = (p ≡ (p ∧ q)) (descriptive implication)
df
• (p ⇔ q) = [(p ⇒ q) ∧ (q ⇒ p)] (quasi descriptive equivalence)
The logic LD has been defined by a Hilbert-style axiomatization given in [5]. It consists
of 17 axioms and 4 rules of inference. The axioms are as follows:
(Ax0) ¬(p ∧ ¬p)
(Ax1) p≡p
(Ax2) ¬¬p ≡ p
(Ax3) p ≡ (p ∧ p)
(Ax4) p ≡ (p ∨ p)
(Ax5) (p ∧ q) ≡ (q ∧ p)
(Ax6) (p ∨ q) ≡ (q ∨ p)
(Ax7) (p ∧ (q ∧ r)) ≡ ((p ∧ q) ∧ r)
(Ax8) (p ∨ (q ∨ r)) ≡ ((p ∨ q) ∨ r)
(Ax9) (p ∧ (q ∨ r)) ≡ ((p ∧ q) ∨ (p ∧ r))

(Ax10) (p ∨ (q ∧ r)) ≡ ((p ∨ q) ∧ (p ∨ r))
(Ax11) ¬(p ∧ q) ≡ (¬p ∨ ¬q)
(Ax12) ¬(p ∨ q) ≡ (¬p ∧ ¬q)
(Ax13) (p ≡ q) ≡ (q ≡ p)
(Ax14) (p ≡ q) ≡ (¬p ≡ ¬q)
(Ax15) (p ≡ q) ⇒ ((p ≡ r) ≡ (q ≡ r))
(Ax16) (p ≡ q) ⇒ ((p ∧ r) ≡ (q ∧ r))
(Ax17) (p ≡ q) ⇒ ((p ∨ r) ≡ (q ∨ r))
The axioms express basic properties of conjunction, disjunction and negation with
respect to the descriptive equivalence. Note that only one axiom does not involve the
connective ≡, namely the axiom (Ax0), which reflects a kind of consistency of the logic.
The intended meanings of the other axioms are:
• Descriptive equivalence ≡ represents an equivalence relation: (Ax1), (Ax13), and
(Ax15).
• Extensionality property of ≡: (Ax14) through (Ax17).
• Idempotency of conjunction and disjunction: (Ax3) and (Ax4).
• Commutativity of conjunction and disjunction: (Ax5) and (Ax6).
• Associativity of conjunction and disjunction: (Ax7) and (Ax8).
• Distributivity of conjunction over disjunction: (Ax9).
• Distributivity of disjunction over conjunction: (Ax10).
• Involution of negation: (Ax2).
• De Morgan laws for negation: (Ax11) and (Ax12).
The rules have the following forms:
ϕ ≡ ψ, ϕ ϕ(p0 , . . . , pn )
(MPE) (Sub)
ψ ϕ(p0 /ψ0 , . . . , pn /ψn )
ϕ, ψ ϕ∧ψ
(∧1 ) (∧2 )
ϕ∧ψ ϕ, ψ
where ϕ, ψ, ψ0 , . . . , ψn are any LD-formulas and p0 , . . . , pn are propositional variables

with the additional restriction that the rule (Sub) applies only to axioms. The rules (Sub),
(∧1 ), (∧2 ) are standard in classical logic. However, it should be emphasized that instead
of the classical Modus Ponens rule, the logic LD has a similar inference rule but with
respect to the descriptive equivalence operator. Note also that LD does not include any
rule for introduction or elimination of disjunction and negation.
The provability of a formula is defined in LD in a standard way. Thus, a formula
ϕ is said to be provable in LD (# ϕ for short) whenever there exists a finite sequence
ϕ1 , . . . , ϕn of LD-formulas, n ≥ 1, such that ϕn = ϕ and each ϕi , i ∈ {1, . . . , n}, is an
axiom or follows from earlier formulas in the sequence by one of the rules of inference.
If X is any set of LD-formulas, then ϕ is said to be LD-provable from X (X # ϕ for
short) whenever there exists a finite sequence ϕ1 , . . . , ϕn of LD-formulas, n ≥ 1, such
that ϕn = ϕ and for each i ∈ {1, . . . , n}, ϕi is an axiom or ϕi ∈ X or ϕi follows from
earlier formulas in the sequence by one of the rules of inference.
Observe that if we interpret ≡ as the classical equivalence, then the LD-axioms are
classical tautologies and all the rules preserve classical validity. It means that one of the
possible models of LD is the two-element Boolean algebra of classical propositional logic.
Therefore, p ≡ ¬p is not provable in LD, and hence the logic LD is consistent in the sense
that it does not entail all formulas. Moreover, as shown in [4], the logic LD is actually
paraconsistent, that is, even a contradiction in LD does not entail all formulas. The proof
of this fact uses semantics for LD.
The first sound and complete semantics for LD was introduced in [3]. Then, after some
weakening and improvement of the LD-models, the strong soundness and completeness
of LD with respect to the class of paraconsistent LD-models was proved in the paper
[4]. Generally, models of the logic LD are based on the so-called Grzegorczyk algebras
satisfying some further conditions. Now, following presentations given in [3] and [4], we
present the definition of LD-models in detail.
A structure (U, ⊕, ⊗) is said to be a distributive bisemilattice whenever the following
hold, for all a, b, c ∈ U and for any 1 ∈ {⊗, ⊕}:
• a 1 b = b 1 a, (commutativity of ⊗, ⊕)
• a 1 (b 1 c) = (a 1 b) 1 c, (associativity of ⊗, ⊕)
• a 1 a = a, (idempotency of ⊗, ⊕)
• a ⊕ (b ⊗ c) = (a ⊕ b) ⊗ (a ⊕ c), (distributivity of ⊗ over ⊕)
• a ⊗ (b ⊕ c) = (a ⊗ b) ⊕ (a ⊗ c). (distributivity of ⊕ over ⊗)
A de Morgan bisemilattice is a structure (U, ∼, ⊕, ⊗) such that (U, ⊕, ⊗) is a distributive
bisemilattice and for all a, b ∈ U , the following hold:
• ∼∼a = a,
• ∼(a ⊕ b) = ∼a ⊗ ∼b.
A Grzegorczyk algebra is a structure (U, ∼, ⊕, ⊗, ◦) such that (U, ∼, ⊕, ⊗) is a de
Morgan bisemilattice and for all a, b, c ∈ U , the following hold:
• a ◦ b = b ◦ a,
• a ◦ b = ∼a ◦ ∼b,
• a ◦ b = (a ◦ b) ⊗ ((a ◦ c) ◦ (b ◦ c)),
• a ◦ b = (a ◦ b) ⊗ ((a ⊕ c) ◦ (b ⊕ c)),
• a ◦ b = (a ◦ b) ⊗ ((a ⊗ c) ◦ (b ⊗ c)).
Fact 2.1 A structure (U, ∼, ⊕, ⊗, ◦) is a Grzegorczyk algebra if and only if the following
conditions hold, for all a, b, c ∈ U :
(LD1) a ◦ b = b ◦ a,
(LD2) a ◦ b = (a ◦ b) ⊗ ((a ◦ c) ◦ (b ◦ c)),
(LD3) a ◦ b = ∼a ◦ ∼b,
(LD4) a ◦ b = (a ◦ b) ⊗ ((a ⊕ c) ◦ (b ⊕ c)),
(LD5) a ◦ b = (a ◦ b) ⊗ ((a ⊗ c) ◦ (b ⊗ c)),
(LD6) a ⊕ b = b ⊕ a,
(LD7) a ⊕ (b ⊕ c) = (a ⊕ b) ⊕ c,
(LD8) a ⊕ a = a,
(LD9) a ⊗ b = b ⊗ a,
(LD10) a ⊗ (b ⊗ c) = (a ⊗ b) ⊗ c,
(LD11) a ⊗ a = a,
(LD12) a ⊗ (b ⊕ c) = (a ⊗ b) ⊕ (a ⊗ c),
(LD13) a ⊕ (b ⊗ c) = (a ⊕ b) ⊗ (a ⊕ c),
(LD14) ∼(a ⊕ b) = ∼a ⊗ ∼b,
(LD15) ∼(a ⊗ b) = ∼a ⊕ ∼b,
(LD16) ∼∼a = a.
It is worth to emphasize the following fact:
Fact 2.2 Boolean algebras, Kleene algebras, and de Morgan algebras are Grzegorczyk
algebras.
The converse of the above does not hold. The class of Grzegorczyk algebras contains
subclasses that form bases for semantics of various non-classical logics of different types.
This shows how different from other well known non-classical logics the logic LD is.
Grzegorczyk algebras are a base for structures of LD.
Definition 2.3 A paraconsistent LD-structure is a structure of the form
U, ∼, ⊕, ⊗, ◦, D!, where
• U, D are non-empty sets such that D U ,

• U, ∼, ⊕, ⊗, ◦! is a Grzegorczyk algebra,
• for all a, b ∈ U , the following hold:
– a ⊗ b ∈ D if and only if a ∈ D and b ∈ D,
– a ◦ b ∈ D if and only if a = b,
– ∼(a ⊗ ∼a) ∈ D.
A paraconsistent LD-structure is called a classical LD-structure if there is no a ∈ U such
that a ⊗ ∼a ∈ D. In what follows, paraconsistent LD-structures will be also referred to
as LD-structures or LD-models.
Definition 2.4 Let M = (U, ∼, ⊕, ⊗, ◦, D) be an LD-structure. A valuation on M is
any mapping v : V → U such that for all LD-formulas ϕ and ψ:
• v(¬ϕ) = ∼v(ϕ),
• v(ϕ ∧ ψ) = v(ϕ) ⊗ v(ψ),
• v(ϕ ∨ ψ) = v(ϕ) ⊕ v(ψ),
• v(ϕ ≡ ψ) = v(ϕ) ◦ v(ψ).
A formula ϕ is said to be satisfied in M by a valuation v if and only if v(ϕ) ∈ D. It is
true in M whenever it is satisfied in M by all the valuations on M, and it is LD-valid if
it is true in all paraconsistent LD-structures.
Definition 2.5 Let X and ϕ be a set of LD-formulas and a single LD-formula, respec-
tively. The formula ϕ is a semantic consequence of X, denoted by X |LD ϕ, if for every
paraconsistent LD-structure M and every valuation v in M such that M, v | X, it holds
that M, v | ϕ.
In [3] soundness and completeness with respect to the classical LD-structures is proved,
while in [4] it has been proved that LD is strongly sound and complete with respect to the
paraconsistent LD-structures. Hence, the following holds:
Theorem 2.6 (Soundness and Completeness of LD) For every LD-formula ϕ the
following conditions are equivalent:
1. ϕ is LD-provable.
2. ϕ is LD-valid.
3. ϕ is true in all classical LD-structures.
Theorem 2.7 Let X and ϕ be a set of LD-formulas and a single LD-formula, respectively.
Then, the following conditions are equivalent:
1. X #LD ϕ.
2. X |LD ϕ.
The above theorem provides a way of proving that the logic LD is paraconsistent in the
following sense:
Definition 2.8 A logic is paraconsistent (or contradiction-tolerant) iff there are formulas
ϕ, ψ such that ϕ ∧ ¬ϕ # ψ.
In [4] the following theorem is proved:
Theorem 2.9 The logic LD is paraconsistent. In particular, p ∧ ¬p #LD q.
Proof Let M = (U, ∼, ⊕, ⊗, ◦, D) be a paraconsistent LD-structure defined as:
U = {0, 1, 2},
∼a = 2 − a,
a ⊕ b = max(a, b),
a ⊗ b = min(a, b),

1, if a = b,
a◦b =
0, otherwise;
D = {1, 2}.
Now, let v(p) = 1, v(q) = 0. Then, it is easy to show that v(p ∧ ¬p) = 1 ∈ D, but
v(q) = 0 ∈/ D. Hence, the logic LD is paraconsistent. *
,
The result on the semantics shows that LD has much in common with the non-Fregean
logic SCI introduced by Suszko in [7], cf. [8]. Indeed, the logic LD can be seen as
non-Fregean in the sense that it rejects the main assumption of classical Fregean logic,
according to which sentences with the same truth value have the same denotations.
Although the logics LD and SCI share the language and have similar philosophical
motivations, they differ considerably in the formalization and they are different logics,
as we will see.
3 Properties of LD
The first results on provable and unprovable formulas presented in [3] have shown in
particular that: (1) the logic LD is new, in the sense that its class of valid formulas cannot
be identified with valid formulas of any other well known non-classical logic, (2) the
descriptive equivalence and implication connectives are different from the classical ones,
(3) neither the absorption nor the boundedness laws are provable in LD.
In [3] the following has been proved:
Theorem 3.1 The following formulas are not provable in LD:
1. (ϕ ≡ ψ) ↔ (ϕ ↔ ψ)
2. (ϕ ⇒ ψ) ↔ (ϕ → ψ)
3. (ϕ ≡ ϕ) ≡ (ψ ≡ ψ)
4. (ϕ ∨ (ϕ ∧ ψ)) ≡ ϕ
5. (ϕ ∧ (ϕ ∨ ψ)) ≡ ϕ
6. (ϕ ∨ ¬ϕ) ≡ (ψ ∨ ¬ψ)
7. (ϕ ∧ ¬ϕ) ⇒ ψ
8. ϕ ⇒ (ψ ∨ ¬ψ)
9. ¬(ϕ ≡ ¬ϕ)
Proof By way of example, we will show that the absorption laws, i.e., formulas 4 and 5
of the theorem, are not provable in LD.
Let M = (U, ∼, ⊕, ⊗, ◦, D) be a paraconsistent LD-structure such that U =
{0, 1, 2, 3}, D = {2, 3} and the operations are defined as follows:
∼ 0 1 2 3 ◦ 0 1 2 3 ⊗ 0 1 2 3 ⊕ 0 1 2 3
3 2 1 0 0 3 0 0 0 0 0 0 0 0 0 0 0 3 3
1 0 3 0 0 1 0 1 1 0 1 0 1 2 3
2 0 0 3 0 2 0 1 2 3 2 3 2 2 3
3 0 0 0 3 3 0 0 3 3 3 3 3 3 3
It can be verified that the above tables indeed define a Grzegorczyk algebra. Then, the
absorption laws do not hold, as 1 ⊕ (1 ⊗ 0) = 1 ⊗ (1 ⊕ 0) = 0 = 1. Therefore, the
formulas (p ∨ (p ∧ q)) ≡ p and (p ∧ (p ∨ q)) ≡ p are not satisfied in M by a valuation
v such that v(p) = 1 and v(q) = 0, which by Theorem 2.6 means that these formulas are
not provable in LD. Proofs for other formulas can be found in [3]. ,
*
Observe the following particular consequences of Theorem 3.1:

• The descriptive equivalence and implication are different from the classical ones
(unprovability of formulas 1 and 2).
• True descriptive equivalences may be not identical (unprovability of formula 3).
• The absorption laws are not theorems of LD (unprovability of formulas 4 and 5).
• True sentences are not necessarily descriptively equivalent (unprovability of formula
6).
• False does not imply everything (unprovability of formula 7).
• Not everything implies the truth (unprovability of formula 8).
• Negation and disjunction do not behave in a classical way, which is clearly manifested
by the unprovability of formula 9. Indeed, ¬(p ≡ ¬p) is not a theorem of LD, while
it can be easily observed that the formula p ≡ ¬p is not satisfiable in any LD-model
and, on the other hand, the formula ¬(p ≡ ¬p) ∨ (p ≡ ¬p) is valid in LD.
This shows that LD can be seen as a logic that avoids the classical paradoxes of
implication. On the other hand, LD shares many of the classical laws.
Theorem 3.2 The following formulas are LD-valid:
1. ϕ ∨ ¬ϕ
2. (ϕ ∧ (ϕ → ψ)) → ψ 1
3. (ϕ ≡ ψ) ⇒ (ϕ ⇔ ψ)
Moreover, (ϕ ⇔ ψ) #LD (ϕ ≡ ψ) and the following rules are admissible in LD:
ϕ ≡ ψ, ψ ≡ θ ϕ
(tran) (∨)
ϕ≡θ ϕ∨θ
The proofs of the above properties can be found in [3] and [4].
The LD-validity of the formula (ϕ ≡ ψ) ⇒ (ϕ ⇔ ψ) and the fact that (ϕ ⇔ ψ) #LD
(ϕ ≡ ψ) connect the descriptive equivalence ≡ with the quasi descriptive equivalence
operator ⇔. Recall that it is defined as follows:
df
(p ⇔ q) = [(p ⇒ q) ∧ (q ⇒ p)]
As in classical logic, it could be expected that the descriptive equivalence can be
expressed as the conjunction of two implications, which would mean that the descriptive
and quasi descriptive equivalences are indistinguishable in LD. However, as shown in [4]
it does not hold, since the following theorem is true:
Theorem 3.3 The following formula is not LD-valid:
(p ≡ q) ≡ (p ⇔ q).
1 Recall that ϕ → ψ is short for ¬ϕ ∨ ψ.

In [4] the following specific formulas have been discussed:

[(p ≡ q) ∧ (q ≡ r)] ⇒ (p ≡ r) (AxT)
((p ≡ q) ∧ p) ≡ ((p ≡ q) ∧ q) (AxD)
The formula (AxT) expresses a strong form of transitivity of ≡. The formula (AxD)
is called the Delusion Axiom in reference to an example used to illustrate its intuitive
content. Consider someone who believes that Vladimir Putin and Donald Trump are the
same person, just wearing different masks. Such a deluded person should accept that the
following two statements say the same:
(α) Putin is the president of Russia
(β) Trump is the president of Russia
Consequently, she would accept that the following statements are descriptively equiva-
lent:
1. α ≡ β and α.
2. α ≡ β and β.
The acceptance of the identity of these claims above means evaluating a conjunction
involving a delusion of identity as if the identity were true, which is expressed by axiom
(AxD).
In [4] is has been shown that neither (AxT) nor (AxD) is a theorem of LD.
Theorem 3.4 The following formulas are not provable in LD:
1. [(p ≡ q) ∧ (q ≡ r)] ⇒ (p ≡ r)
2. ((p ≡ q) ∧ p) ≡ ((p ≡ q) ∧ q)
There has been some controversy over the last three axioms of LD:
(Ax15) (p ≡ q) ⇒ ((p ≡ r) ≡ (q ≡ r))
(Ax16) (p ≡ q) ⇒ ((p ∧ r) ≡ (q ∧ r))
(Ax17) (p ≡ q) ⇒ ((p ∨ r) ≡ (q ∨ r))
One of their intended meanings is to express the extensionality principle for descriptions:
Sentences that have equal meaning (have the same content or are the same descriptions) are
interchangeable in all possible contexts.
Axioms (Ax15), (Ax16), (Ax17) seem to express this property in a stronger form.
Their adequacy has been deeply discussed in [2] and the conclusion presented there is
rather negative. Instead a new very weak Minimal Grzegorczyk Logic MGL has been
introduced with a rule expressing the extensionality principle instead of the axioms.
However, this account goes beyond the scope of this paper, so the interested reader should
consult [2].
On the other hand, in the paper [4] other ways of expressing the extensionality principle
have been explored. We start this discussion with the formal definitions.
Definition 3.5 The Weak Extensionality Principle (WEP) is the following metarule:
ϕ≡ψ
,
ϑ(p/ϕ) ≡ ϑ(p/ψ)
where ϕ, ψ, and ϑ are arbitrary formulas.

Definition 3.6 The Strong Extensionality Principle (SEP) is the claim
# (ϕ ≡ ψ) ⇒ (ϑ(p/ϕ) ≡ ϑ(p/ψ)),
where ϕ and ψ are arbitrary formulas, and ϑ(p) is a formula in which p actually occurs.
Definition 3.7 The Grzegorczyk Extensionality Principle (GEP) is the statement
# (ϕ̄ ≡ ψ̄) ⇒ (ϑ(p1 /ϕ1 , . . . , pn /ϕn ) ≡ ϑ(p1 /ψ1 , . . . , pn /ψn )),
where ϕ1 , . . . , ϕn , ψ1 , . . . , ψn are arbitrary formulas, ϑ is a formula whose propositional

variables are contained in {p1 , . . . , pn }, and ϕ̄ ≡ ψ̄ is short for (ϕ1 ≡ ψ1 ) ∧ . . . ∧ (ϕn ≡
ψn ).
Observe that SEP implies the WEP and the GEP implies the WEP for a given logic. In [4]
the following have been proved
Theorem 3.8
1. Neither the SEP nor the GEP holds for LD.
2. The WEP holds for LD.
In the next section we will discuss extensions of LD with axioms (AxT) and (AxD) as well
as some other modifications of LD, together with their properties and mutual relationships.
4 Extensions and Modifications of LD
In the previous section we discussed the most important properties of LD. In particular,
we mentioned that the transitivity axiom (AxT) and the delusion axiom (AxD) are not
provable in LD. Hence, it was quite natural to consider extensions of LD with these
axioms, which were deeply studied in the paper [4].
Logics obtained by adding (AxT) and (AxD) to the axiomatization of LD are denoted
by LDT and LDD, respectively. The notions of provability in LDT and LDD are defined in
a similar way as in LD. Models of LDT and LDD are defined as paraconsistent LD-models
that satisfy the semantic counterparts of axioms (AxT) and (AxD), respectively. Then, the
notions of satisfaction, truth and validity in LDT and LDD are defined in a standard way.
Theorem 4.1
1. #LDD (AxT).
2. #LDT (AxD).
3. #LDT (p ≡ q) ≡ (p ⇔ q).
The above theorem is proved in [4]. Note that the last formula is not provable in LD, while
it is in LDT. Thus, LDT seems to be a better candidate than LD to compare Grzegorczyk’s
approach with various other non-classical logics, which are usually defined mainly in
terms of implication.
In [4] the following is proved:
Theorem 4.2
1. #LD p ⇒ (p ∨ q).
2. #LDT p ⇒ (p ∨ q).
3. #LDD p ⇒ (p ∨ q).
The above theorem allows us to show that LD, LDT, and LDD are different from several
non-classical logics. Recall that p → (p ∨ q) is provable in intuitionistic logic and
relevance logics T, E, R, EM, and RM, see e.g.,[1]. Hence, if we identify the descriptive
implication ⇒ with the implication of the other logic, then LD, LDT, and LDD are
different from any of the aforementioned non-classical logics. Moreover, the descriptive
equivalence ≡ cannot be identified with a necessary equivalence in any class of Kripke
frames since (p ↔ (p ∧ (p ∨ q))) is true in all frames.
Definition 4.3 Let L and L be logics. A logic L is said to be:
• not stronger than L , L ≤ L for short, whenever all formulas valid in L are valid in L ,
• equal with L , L = L for short, L ≤ L and L ≤ L,
• weaker than L , L < L for short, whenever L ≤ L and L = L ,
• uncomparable with L whenever L ≤ L and L ≤ L.
Comparing provable formulas in the logics LD, LDT, and LDD, we can state the
following:
Fact 4.4 LD < LDT < LDD.
As noted in the previous section, the question whether the last three axioms of LD
adequately express the extensionality principle has caused some controversy. This has
led to some modifications of LD in which these three axioms have been substituted with
alternative forms. The most important ones, explored in the paper [4], are the logics LE
and LDS obtained by replacing the axioms (Ax15), (Ax16), (Ax17) with the following
formulas, respectively:
(Ax15)LE ((p ≡ q) ∧ (p ≡ r)) ≡ ((p ≡ q) ∧ (q ≡ r))
(Ax16)LE ((p ≡ q) ∧ (p ∧ r)) ≡ ((p ≡ q) ∧ (q ∧ r))
(Ax17)LE ((p ≡ q) ∧ (p ∨ r)) ≡ ((p ≡ q) ∧ (q ∨ r))
(Ax15)LDS ((p ≡ q) ∧ (r ≡ s)) ⇒ ((p ≡ r) ≡ (q ≡ s))

(Ax16)LDS ((p ≡ q) ∧ (r ≡ s)) ⇒ ((p ∧ r) ≡ (q ∧ s))
(Ax17)LDS ((p ≡ q) ∧ (r ≡ s)) ⇒ ((p ∨ r) ≡ (q ∨ s))
Table 1 The semilattice of

Grzegorczyk’s logics of LDD
descriptions
LDT
LDS LD LE
Table 2 Extensionality LD LDT LDD LE LDS

principles in Grzegorczyk’s
logics of descriptions WEP + + + − +
SEP − + + − −
GEP − + + − +
The axioms for LDS are essentially copied from Suszko’s Sentential Calculus of Identity,
substituting the descriptive implication for the classical one.
Below we list some properties of the logics under consideration. Their proofs can be
found in [4].
Theorem 4.5
1. The formula (p ≡ q) ≡ ((p ≡ q) ∧ (q ≡ q)) is provable in LE, but not in LD.
2. LE ≤ LDD.
3. LE ≤ LDT.
4. LE is uncomparable with LD and LDT.
5. LDS is paraconsistent.
6. LDS and LD are uncomparable.
7. LDS < LDT.
8. The GEP holds for LDS.
9. The SEP holds for LDT.
10. The WEP does not hold for LE.
All these results lead to the following:
Theorem 4.6
1. LD < LDT < LDD
2. LDS < LDT < LDD
3. LE < LDD
Hence, we have the following picture of dependencies among the logics LD, LDT, LDD,
LE, LDS (Table 1):
We end this section with two summary tables (Table 2):
Table 3 Validity of some special axioms

(AxT) (AxD) (p ≡ q) ≡ (p ⇔ q) LD LE LDS
LD − − − + − −
LDT + − + + − +
LDD + + + + + +
LE − ? ? − + −
LDS − − ? − − +
5 Conclusions
We have discussed the logic LD of descriptions together with its extensions LDT and
LDD as well as the modifications LE and LDS. We have presented a strongly sound and
complete semantics for LD, and used it to show that LD is a paraconsistent logic. Then,
we have listed examples of classical laws which are provable and not provable in LD. We
have reported what extensionality properties the logic LD has. Finally, we have presented
formal properties of extended and modified versions of LD and stated the relationships
among them (Table 3).
Still there are many open problems concerning the logics in question. Below we list a
few of them. In what follows, L stand for any of LD, LE, LDS, LDT, or LDD.
1. Is L decidable? If so, what is its complexity?
2. Is L equivalent to a previously known paraconsistent logic? If not, how does it relate
to them?
3. How is a logic obtained by replacing ≡ with ⇔ in L related to L?
4. The last three axioms of LE express a property that resembles extensionality. Can this
property and its relationship with actual extensionality be formulated in an informal,
intuitive way?
References
1. Anderson, A.R., Belnap, N.D.: Entailment. The Logic of Relevance and Necessity, vol. I. Princeton
University Press, Princeton (1975)
2. Golińska-Pilarek, J.: On the minimal non-Fregean Grzegorczyk’s logic. Stud. Logica. 104(2), 209–234
(2015)
3. Golińska-Pilarek, J., Huuskonen, T.: Logic of descriptions. A new approach to the foundations of
mathematics and science. Stud. Log. Grammar Rhetor. 27(40), 63–94 (2012)
4. Grzegorczyk’s non-Fregean logics and their formal properties. In: Urbaniak, R., Payette, G. (eds.)
Applications of Formal Philosophy: The Road Less Travalled, Logic, Argumentation & Reasoning,
vol. 14. Springer (2017)
5. Grzegorczyk, A.: Filozofia logiki i formalna LOGIKA NIESYMPLIFIKACYJNA [Philosophy of Logic
and Perceptive Equivalence]. Zagadnienia Naukoznawstwa 4(190), 445–450 (2011)
6. Grzegorczyk, A.: Errata to Grzegorczyk, A.: Filozofia logiki i formalna LOGIKA NIESYMPLIFIKA -
CYJNA [Philosophy of Logic and Perceptive Equivalence]. Zagadnienia Naukoznawstwa 4(190),
445–450 (2011). Zagadnienia Naukoznawstwa 4(190) (2011)
7. Suszko, R.: Non-Fregean logic and theories. Analele Universitatii Bucuresti, Acta. Logica. 9, 105–125
(1968)
8. Suszko, R.: Semantics for the sentential calculus with identity. Stud. Logica. 28, 77–81 (1971)
J. Golińska-Pilarek () • T. Huuskonen

Roman Suszko: Logician and Philosopher
Mieczysław Omyła
Abstract The paper presents basic information about life and main scientific achieve-
ments of Roman Suszko (1919–1979). He is known as one of the first logicians
who applied the model theory to non-mathematical problems, particularly to study
development of knowledge. The article is divided into five sections: 1. Life, 2. Main
papers, 3. Views, 4. Influences, 5. Summary.
Keywords Roman Suszko · Logical material · Diachronic logic · Non-Fregean logic
1 Life
Roman Suszko was born on 9th of November 1919 in Podobora near Cieszyn in Poland
(at present—Czech Republic).
Roman Suszko started his studies at Poznań University in 1937. In 1939, with outbreak
of Second World War, he was displaced to Cracow, where he spent the whole of the
occupation period; he worked as a night guard, telephonist and—at the same time—he
attended the clandestine study classes of physics, mathematics and philosophy under the
guidance of professors of Jagiellonian University. In 1945 he obtained a title of Master of
Arts in Philosophy for the work The achievements of polish logic promoted by Professor
Zygmunt Zawirski.
In 1946–1953 he worked at the Chair of the Theory and Methodology of Science,
Faculty of Mathematical-Natural Sciences, University of Poznań. In 1948 he was awarded
a title doctor for the thesis On Normal Systems and Few Questions of Basic Logic written
under the supervision of Professor Kazimierz Ajdukiewicz. The thesis was published in
the form of two articles On Analytical Axioms and Logical Rules and From the Theory of
Definitions. In 1951 he was awarded a postdoctoral degree in logic also on the Faculty
of Mathematical-Natural Sciences, University of Poznań for the dissertation Canonic
Axiomatic systems, Studia Philosophica IV, Poznań 1951. In 1953 he moved to Warsaw
and started to work at the Faculty of Philosophy at the University of Warsaw. From the
very beginning of his work in Warsaw he had close contact with the Polish Academy

748 M. Omyła
of Sciences, at the beginning only with the Group of Algebra headed by Jerzy Łoś at
the Institute of Mathematics, and then with the Group of Logic headed by Kazimierz
Ajdukiewicz at the Institute of Philosophy and Sociology PAN.
In 1959 he was awarded a title of associate professor. In 1960 he was elected to
the position of a Dean of the Faculty of Philosophy at the Warsaw University. In 1960
he moved to the Institute of Philosophy and Sociology of PAN, where in 1966–1969
he was a head of the Group of Logic Institute of Philosophy and Sociology Polish
Academy of Sciences. In 1967–1969, 1970–1973 he worked as a professor at the Stevens
Institute of Technology in Hoboken, New Jersey, USA. After returning from the USA in
1973 till his death he worked at the Group of Logic Institute Philosophy and Sociology
Polish Academy of Sciences. Besides scientific and teaching activity Suszko played also
important role as a publisher: he was one of the founders of Studia Logica—journal on
logic with international coverage—and till the end he was a member of its Programming
Board.
He died on cancer on 3 June 1979 in Warsaw.
2 Main Papers
Roman Suszko’s works were published in prestigious international journals of mathe-

matics, logic and philosophy, such as: “Fundamenta Mathematicae”, “Journal Symbolic
Logic”, “Colloquium Mathematicum”, “Synthese”, “Theoria”, “Logique et Analyse”,
“Archiv für anathematise Logik und Grundlagen-Forschung ”, “Studia Logica” and
others.
From the scientific achievement of Suszko we select those publications, which
influenced greatly the choice scientific interest and investigation of many logicians,
philosophers and mathematicians, both in Poland and abroad.
[1] Canonic axiomatic systems. Studia Philosophica IV, 301–330 (1951)
[2] On the extending of models. II: Common extensions (together with J. Łoś). Fund.
Math. XLII, 343–347 (1955)
[3] Formalna teoria wartości logicznych I [A formal theory of logical values I]. Studia
Logica VI, 145–237 (1957)
[4] On the extending of models. IV: Infinite sums of models (with J. Łoś). Fund. Math.
XLIV/1, 52–60 (1957)
[5] Remarks on sentential logics (with J. Łoś). Indagationes Mathematicae 20, 177–183
(1958)
[6] Syntactic structure and semantical reference II. Studia Logica IX, 63–91 (1960)
[7] Wykłady z logiki formalnej. Cz˛eść I [Lectures on formal logic]. PWN, Warsaw
(1965). Ed. B. Stanosz
[8] Formal logic and the development of knowledge. In: Problems in the Philosophy of
Sciences, International Colloquium London, pp. 210–222. North-Holland, Amster-
dam (1965)
[9] Ontology in the Tractatus of L. Wittgenstein. Notre Dame Journal of Formal Logic
9, 7–33 (1968)
Roman Suszko: Logician and Philosopher 749
[10] Non-Fregean logic and theories. Analele Universitatii Bucuresti, Acta Logica
11, 105–125 (Russian translation of that work is in the collection of articles
Neklassičeskaja logika, Moscov 1970)
[11] Identity connective and modality. Studia Logica XXVII, 7–39 (1971)
[12] Abstracts Logics (with D.J. Brown). Dissertationes Mathematicae CII, PWN,
Warsaw (1973)
[13] Abolition of the Fregean Axiom. Lecture Notes in Mathematics 453, 169–239
(1975)
[14] The Fregean axiom and Polish mathematical logic in the 1920s. Studia Logica 36,
376–380 (1977)
[15] The reification of situations. In: Woleński, J. (ed.) Philosophical Logic in Poland,
pp. 247–270. Kluwer Academic Publishers (1994). Translation from Polish by T.
Stazeski
Suszko’s scientific work is very diverse and strictly connected with his philosophical
views, and above all—with his views on philosophy of language and of logic.
In his doctoral thesis Suszko considered with so called “problem of logic without ax-
ioms”. He refers—from one side—to the theory of language presented by K. Ajdukiewicz
in O znaczeniu wyrażeń (On the meaning expressions), according to which one can among
the semantic rules of a language distinguish: axiomatic, deductive and empirical rules,
and—from other side—he refers to works written by Jaśkowski and Gentzen, which dealt
with replacing logical axioms with sets of inference rule.
According to Suszko the rules offered by them are not “proper inference rules”,
because they let to infer infinitely many logical theorems without any assumptions.
According to Ajdukiewicz they are at the same time axiomatic inference rules.
Another important thread in Suszko’s work was the problem of connections between
logical matrixes and sentential calculi. This subject appear again in many Suszko’s
works written at different age, devoted to both abstract and specific logics. In the
work Remarks on sentential logics Łoś and Suszko formulated the general notion of
strong adequacy of a matrix with respect to abstract sentential calculi and formulated
general conditions of existence of such a matrix for any calculus. The work is still
quoted in almost all publications on sentential calculus. Suszko in cooperation with J.
Łoś wrote certain number of works on models for first-order predicate languages. The
results obtained in those works Suszko used later in his research on formal aspects
of development of knowledge. According to Suszko certain epistemological problems,
which were investigated traditionally in an intuitive way, contemporary logic can search
in a strict and formal way. It is possible thanks to the fact that logical semantics—
and especially the model theory—is a branch of logic. In Suszko’s opinion logic is
strictly connected with epistemology, and he considered the mathematical model theory
as a formal part of the theory of knowledge. Application of logic to the research on
development of knowledge was called by Suszko “diachronic logic”. He devoted to the
diachronic logic certain number of publications.
Introducing non-Fregean logic was considered by Suszko as his most important
achievement. The name “non-Fregean logic” is derived from the fact that Suszko
rejected Frege’s assumption, according to which all true sentences have one common
semantic correlate, just as all false sentences have one common semantic correlate.
Those assumption Suszko called “Frege’s axiom” and more strictly—“semantic version of
750 M. Omyła
Frege’s axiom”. The common correlate of true sentences is usually interpreted as: truth,
being or universe of a model, which means: all objects that satisfy any true sentence.
Analogically the common correlate of all false sentences is usually interpreted as: false,
non-being, all objects that belong to the universe of model and that satisfy any false
sentence. The term non-Fregean logic was introduced by Suszko in [10].
The “non-Fregean period” (1966–1979) was the most creative period of Suszko’s life.
He wrote 36 scientific works and promoted seven doctors. However for him it was also the
time of tragedy, because during this period his 15-years-old son was killed in an accident
and his 24-years-old stepson died in tragic circumstances.
3 Views
In Roman Suszko’s logical writings there are many remarks and reflections on the idea of
logic which is closely related to his work in formal logic.
According to Suszko the subject of logical investigations are any conceptual construc-
tions that came into being as the result of the cognition of the world. The totality of
those constructions is called by Suszko logical material. The state of logical studies at
any time is determined by the logical material available as well as the research tools
at the hand. Among various research tools used for studying logical material Suszko—
in agreement with the trends of contemporary logic—gives priority to mathematical
instruments, especially he prefers the set-theoretical, algebraic and topological methods.
According to Suszko logic—and especially semantics—is not the science devoid of
assumptions but, on the contrary, logic has at its fundament certain general and schematic
knowledge of reality, such as: there are objects, which has properties and which are related
one with another; there are certain states of affairs and some of them occur, while other do
not occur etc. In Suszko’s opinion logic gains those knowledge from the set theory. Since
the set theory contains certain ontological presumptions of logic, Suszko used to call this
theory “formal ontology”. What is essential in this view, is that there are certain general,
structural, and—at the same time—formal properties of the world to be investigated by
ontology.
Suszko took the view, that formal languages being designed and studied by logic
are not free creations, but abstracts derived from common languages and languages
of particular sciences. They may also constitute hypothetical presumptions of those
languages. We investigate the consequences of those presumptions within logic. “Science,
cognitive process and natural language, which play essential role in this process, are frame
of reference for formal logical research.” wrote Suszko in his Formal theory of logical
values I. According to Suszko logic investigates language by means of all formal tools,
i.e. by means of any mathematical method available.
Suszko distinguished two basic categories of expressions: sentences and names. All
other categories have no any independent meaning but are designed for construction
of more complex expressions. Those auxiliary expressions were called by Suszko
“formators”; there are formators that do not bind variables (e.g. connectives, predicates)
and formators that bind variables (such as quantifiers and description operators). Suszko
was convinced that there is certain logico-philosophical parallelism, and namely: names
Roman Suszko: Logician and Philosopher 751
correspond to objects; sentences correspond to situations; and formators correspond to

functions. That is why Suszko used to quote J. Bocheński: “Syntax mirrors ontology”.
Suszko believed that there are certain semantical principles which connect the
syntactical construction rules with theirs object references. Those principles allows us
to introduce the notion of a model for any language which is formalized in a standard
way. The philosophical views that inspired his logical work are here presented.
4 Influences
Roman Suszko had great influence over the choice of investigation problems and over
the way of scientific investigations carried out by many philosophers, logicians and
mathematicians, both in Poland and abroad.
Joint work Suszko’s and Łoś’s [5] was continued by R. Wójcicki and his school. Also
the works [2], [4], [8] served as a source of inspiration for the logicians who applied the
model theory to the methodology of empirical sciences.
Suszko’s reflections on the connections between natural languages and formal lan-
guages were continued among others by: B. Stanosz, A. Nowaczyk, U. Wybraniec-
Skardowska, J. Pogonowski. Those reflections were continued in such works as:
Stanosz, B., Nowaczyk, A.: Logiczne podstawy j˛ezyka (in Polish) [Logical Founda-
tions of language]. Ossolineum (1976)
Wybraniec-Skardowska, U.: On language adequacy. Studies in Logic, Grammar and
Rhetoric 40(53). In: Hensel, M., Poczobut, R. (eds.) Cognitive Systems, pp. 257–292
(2015)
Wybraniec-Skardowska, U.: On the Denotation of Quantifiers. In: Omyła, M. (ed.)
Idee logiczne Romana Suszki (in Polish) [Logical ideas of Roman Suszko], pp. 89–
119. Warsaw (2001)
In turn work [6] devoted to the parallelism between language and reality, and work
[3] in which Suszko ascribes to sentences logical values different than true and false,
serves as a reference base for G. Malinowski, M. Omyła, B. Wolniewicz, A. Wójtowicz.
Suszko inspired with non-Fregean logic and abstract logic logicians in USA working at
the Stevens Institute of Technology: S.L. Bloom, D.J. Brown, J.D. Kagan, A. Michaels,
R. Quackenbush and others, and in Poland: W. Dzik, G. Malinowski, M. Omyła,
J. Pogonowski, J. Zygmunt. Non-Fregean logic is also subject of study for younger
generation logicians: T. Huuskonen, J. Golińska-Pilarek, P. Łukowski, A. Wójtowicz,
J. Wesserling. With the course of years it was turned out that the abstract logic which
was grown out from the considerations on characteristics of non–Fregean logic and on
its place in the spectrum of logical calculi found its application in universal algebra.
J. Czelakowski, W. Dzik, W. Dziobiak, A. Wroński and others—in Poland—and W. Blok,
D. Pigozzi, J.M. Font—abroad—are working on abstract logics inspired by non-Fregean
logic.
752 M. Omyła
After Suszko’s death three monographs at least were published which related mean-
ingfully to his ideas and which—one can say—fulfill his program:
Omyła, M.: Zarys logiki niefregowskiej (in Polish) [Outline Non-Fregean Logic].
Warsaw (1986)
Wójtowicz, A.: Znaczenie nazw a znaczenie zdań, w obronie ontologii sytuacji (in
Polish) [Meaning name and meaning sentences]. Semper, Warsaw (2007)
Dzik, W.: Unification Types in Logic. Silesian University, Katowice (2007)
Very good discussion of the work [13] constitutes the review:
Malinowski, G., Zygmunt, J.: R. Suszko “Abolition of the Fregean Axiom”. Erkenntnis
12, 369–380 (1978)
Full bibliography of Suszko and its description can be found in:
Omyła, M., Zygmunt, J.: Roman Suszko (1919–1979): Bibliography of the Published
Work with an Outline of His Logical Investigations. Studia Logica 43, 421–441 (1984)
Comparison between diachronic logic and non-Fregean logic can be found in the work:
Omyła, M.: Roman Suszko—from diachronic logic to non-Fregean logic. In:
Krajewski, W. (ed.) Polish Philosophers of Science and Nature in the 20th Century,
pp. 153–161. Amsterdam – New York (2001)
5 Summary
Suszko’s logical ideas were a source of inspiration in the choice of problems and in the
ways of carrying out logical investigations in the second half of twentieth century and in
first decades of twenty-first century, both in Poland and in the world.
The majority of polish logicians of the second half of twentieth century remained
under his scientific influence and his irresistible personal charm. Almost on all scientific
conferences devoted to logic taking place in Poland, beginning from Suszko’s death till
today, both speakers and commentators refer to Suszko’s ideas as great inspiration for
other logicians.
M. Omyła ()
Faculty of Law and Administration, Cardinal Stefan Wyszyński University in Warsaw, Warszawa, Poland
From Formal Theory of Knowledge
to Non-Fregean Logic
Mieczysław Omyła
Abstract The terms: ‘diachronic logic’ and ‘non-Fregean logic’ we owe to Roman
Suszko. He called ‘diachronic logic’ an application of classical logic to study of the
development of knowledge. But Non-Fregean logic is a logical calculus obtained from
the classical logic by adding identity connective and axioms for it. The main goal of the
paper is to proof that the non-Fregean logic is a continuation of diachronic logic. The
article is divided into following parts: 1. Diachronic logic, 2. Non-Fregan logic, which
contain 2.1. Introduction, 2.2. Axiomatic form of non-Fregean Logic, 2.3. Properties of
non-Fregean logic, and Bibliography.
Keywords Diachronic logic · Development of knowledge · Non-Fregean logic
Mathematics Subject Classification (2000) Primary 03B60, 03B65, 03B99
Roman Suszko was logician who united in their research the mathematical form of
logic with its philosophical content. One of his work starts with the following epigraph:
“Abstract mathematics may be a thorough philosophy”. This epigraph may be interpreted
in various ways, but the most natural is that we can solve certain philosophical problems
in mathematical way, by creating at the beginning their formal representation, and next—
by solving formal problems related with them. Solving those related formal problem
constitutes at the same time looking for an answer to the initial philosophical question.
In practice the formulation of a given philosophical problem in a formally strictly and
adequate way is a considerable achievement as such. As a matter of fact we restrict
ourselves to creating certain pattern or formal model which has a proper philosophical
interpretation.
During 1957–1968 Suszko devoted certain number of works to the formal analysis of
development of knowledge; this analysis was done with the aid of the models theory for
the classical predicate logic calculus.
Suszko called the research on development of knowledge “diachronic logic”.
Hence diachronic logic is application of the classical models theory for first-order
predicate languages to research development of knowledge.
Suszko used here eloquent terminology offered earlier by Ferdinand Saussure in the
monography Cours Lingistique Gènèrale, where it was offered to distinguish synchronic

754 M. Omyła
linguistic from a diachronic one. Suszko defined “synchronic logic” as syntax and
semantics of formal languages where classical logic is in force, and “diachronic logic”—
as application of synchronic logic to research on development of knowledge. He devoted
to those issues several articles in 1957–1968 [7, 9].
In 1968–1979 Suszko worked extremely intensely on non-Fregean logic, which
he himself invented. This logic was inspired by Wittgenstein’s Tractatus Logico-
Philosophicus. Suszko invented this logic because he thought that the ontology of
Tractatus—considered as certain philosophical theory—goes beyond the formal means
that are available in the framework of the classical first-order predicate calculus.
In the present work I am trying to show, that research on non-Fregean logic is
continuation of earlier research on diachronic logic.
1 Diachronic Logic
A lot of effort was devoted in the 1950s of twentieth century to problems on the border
between formal logic and theory of knowledge. In Suszko’s opinion the contemporary
logic can describe certain aspects of development of knowledge in a precise way and can
throw new light on traditional epistemological questions.
Theory of knowledge explores the epistemological opposition < S, O > where S is a
subject of knowledge and O is an object of knowledge. Suszko represented the subject of
knowledge S by the sequent
(*) (L, Cn, A, T ),
where:
(i) L is a formalized version of language used by the subject of knowledge,
(ii) Cn is operation of consequence defined on language L,
(iii) A is a set of analytical axioms formulated in language L,
(iv) T is set of sentences of the language L accepted by the subject of knowledge S.
Every sentence of language L represents certain thought, and every predicate of language
L represents certain notion at the disposal of subject S. Operation of consequence Cn
defined on language L, is the totality of logical thinking rules. Set of analytical axioms A
contains logical and extra-logical thinking principles, which are at the disposal of subject.
As far as the set of accepted sentences T is concerned we assume that Cn(T ) = T ,
because subject of knowledge usually does not know all logical consequences accepted
by himself and does not know all of his own assumptions. It is assumed also that T −
Cn(A) = ∅, i.e. that the subject accepts certain sentences that are not true in an analytical
way.
In order to study in a formal way the epistemological opposition < S, O >, the reality
which is the object of knowledge has also to be defined in a strict way. For a finite subject
the world as a the whole reality R is never an object of knowledge; at every moment t
subject of knowledge sees only certain fragment of the world, which at the moment t
becomes an object of knowledge Ot for subject S. Suszko calls the period during which
given subject studies one and the same fragment of the world “epoch in development of
From Formal Theory of Knowledge to Non-Fregean Logic 755
knowledge”. If during the given period t certain fragment of knowledge constitutes an

object of knowledge, subject attaches to this world-fragment certain language in order to
speak about it. Speaking loosely, the object of discursive knowledge O on a given stage
of development of knowledge constitutes such a fragment of reality R, that may be caught
by the net of notions being at the disposal of the subject, i.e. such a fragment, which serves
as a model for language L used by the subject. For the sake of simplicity Suszko assumes
that the language used by subject of knowledge is certain first-order predicate language L,
and subject of knowledge is represented by certain intended model of language, it means
by the structure of the following type:
M =< U, d(C1 ), d(C2 ), . . . >,
where: U is universe of language L i.e. it is a set, from which the nominal variables of the
language take their values, and d(Ck ) is denotation of the extra-logical constant Ck .
Selected objects from the set U , or certain sets of objects which represent adequate
properties of objects and relations and relations between objects or—possibly—functions
defined on the set of objects U serve as denotations of extra-logical constants of the given
language. Therefore epistemological opposition
< S, O > is represented by Suszko as the following:
(**) < (L, Cn, A, T ), M >,
where (L, Cn, A, T ) represents a subject of knowledge and model M of the language
L constitutes a formal representation of object of knowledge. One of the consequences
of the assumption, that the subject is equipped with the language L, which is—with the
first-order predicate language is the following: we represent the object of knowledge with
the intended model of the language.
For a given epistemological opposition we mark the set of true sentences of the
language L in model M as V er(L, M). Let us notice that we have here to do with a
double relativisation of the notion of truth: to the language and to the model.
Because the language L is a language defined by the structure of the model M, we can
mark this set shorter with the symbol V er(M).
Research on the formal epistemological opposition (**) is a matter of synchronic logic.
In turn research on changes of this opposition in time belongs to diachronic logic.
Development of knowledge consists—according to apt formulation of Suszko—in
gaining more and more amount of truths about wider and wider object of knowledge.
Within the framework of diachronic logic a development of knowledge is represented by
transformation of epistemological oppositions
< (L, Cn, A, T ), M > / < (L∗ , Cn∗ , A∗ , T ∗ ), M ∗ >
in such a way, that the set of sentences which are simultaneously true and which are
accepted by the subject at the next stage of knowledge contains the set of sentences which
were true and were accepted by the subject at the previous knowledge stage, i.e.
T ∩ V er(M) ⊆ T ∗ ∩ V er(M ∗ ).
756 M. Omyła
Presenting the basic ideas of diachronic logic Suszko used eloquent terminology, which—
among others—was a result of modification of Ajdukiewicz’s terminology used in works
Ajdukiewicz [1, 2].
Let us present briefly those terminology. During every given period of development
of knowledge t subject of knowledge sees certain fragment of the really existing reality
R, which is called by Suszko “the world-layer in period t” and which is represented in
the system of diachronic logic by certain model Mt . The period during which the subject
sees one and the same layer of the world is called by Suszko an “epoch in development
of knowledge”. While initial research on object of knowledge, i.e. of a given world-
layer, subject attaches the language which fits for speaking about this fragment of the
world. Language Lt tailored to speak about the world-layer in epoch t is called by Suszko
“conceptual apparatus in epoch t”.
Let < (Lt , Cn, At , Tt ), Mt > be the epistemological opposition in a given epoch t.
Suszko uses further the following terminology: the set of sentences T , accepted by the
subject of knowledge is called by him the picture of the world in epoch t and in turn the
set Tt ∩ V er(Mt ) is a true fragment of world-picture in epoch t, or “real world-knowledge
in epoch t”. Suszko calls the set Cn(Tt ) a “world-perspective in epoch t” or “potential
world-knowledge in epoch t” and the set of sentences Cn(Tt ) ∩ V er(Mt ) constitutes than
“true fragment of world-perspective in epoch t”.
Suszko distinguishes two main types of knowledge development:
(1) evolutionary, by which the object of knowledge does not change, what means that
the subject of knowledge sees the same world-layer and the syntactic structure of the
language remains also unchanged.
(2) revolutionary, by which the object of knowledge does change.
The evolutionary process of knowledge consists above all in the situation where the
sequence of the sets of sentences accepted by subject of knowledge T , T ∗ , T ∗∗ , . . .
contains more and more true sentences about the same object of knowledge. It happens
that during the evolutionary development of knowledge the extra-logical principles of
thinking change, i.e. set of axioms A becomes set of axioms A∗ . Suszko considers
following two cases of that kind:
(i) A = A∗ and Cn(A) = Cn(A∗ ),
(ii) A ⊆ A∗ and Cn(A) = Cn(A∗ ).
In the case (i) the systematization of axioms takes place. In the case (ii) we have to do
with reinforcement of axioms. The case (ii) embraces the following sub-case:
(L, Cn, A, T )/(L, Cn, A∗ , T ∗ )/(L, Cn, A∗∗ , T ∗∗ )
which consists in fact that certain sentence α the given language L, is initially not
accepted, i.e. α ∈ T and the development of knowledge goes in such a way, that at the
beginning α becomes a non-analytical theorem, α ∈ (T ∗ − Cn(A∗ )), and at the next stage
of knowledge this sentence becomes one of the analytical sentences of given language, i.e.
α ∈ Cn(A∗∗ ); according to Suszko this kind of development of knowledge was noticed
by conventionalists. Evolutionary development of knowledge corresponds with what T.S.
Kuhn calls in his book The structure of Scientific Revolutions, Chicago, 1962 “normal
stage of science development”.
In turn revolutionary development of knowledge consists in the situation, in which the

object of knowledge—represented here by the model of language—change. It means that
the change M/M ∗ takes place. Suszko shows two kinds of such a change:
(2a) The universe of objects U does not change, but new properties of objects and new
relations between them are discovered, and—as a result—new model M ∗ is built,
which constitutes an extension of model M
M =< U, R1 , R2 , . . . , Rn > /M ∗ =< U, R1 , R2 , . . . , Rn , Q1 , Q2 , . . . , Qm >
Such an extension of knowledge object causes an extension of the language L to

language L∗ , which corresponds to the model M ∗ . In this case
L ∩ V er(L∗ , M ∗ ) = V er(L, M).
This condition means, that all sentences of language L which are true in model M
remain to be true also in model M ∗ . Additionally there are in model M ∗ also true
sentences, that were impossible to formulate in language L:
V er(L∗ , M ∗ ) − V er(L, M) = ∅
(2b) Universe of model M =< U, R1 , R2 , . . . , Rn > becomes extended and as a result

the object of knowledge becomes M ∗ =, where U ⊆ U ∗
and U = U ∗ , and relations Ri∗ for i = 1, 2, . . . , n are extensions of—accordingly—
relations Ri . New objects become then known. Model M constitutes sub-model of
model M ∗ , it may happen that V er(L, M) = V er(L, M ∗ ) and certain sentences
that are true in model M may not be true in model M ∗ .
Suszko considered the following as very important: the theory of models can be applied
to formulate precisely and to study everything what traditional theory of knowledge had
studied just intuitively. Suszko was fascinated by the possibility of precise grasping the
traditional issues, although he remained skeptical about the possibility to achieve in such
a way any more important results.
Suszko’s works on diachronic logic constitute one of the first in the world attempts
to apply the theory of models to extra-mathematical questions, and especially—to
philosophical problems. His works initiated applications of the theory of models to
methodological research carried out in Poland. Those research flourished in Poland
especially in 1960s and 1970s of twentieth century.
2 Non-Fregean Logic
2.1 Introduction
During “diachronic logic” period Suszko assumed that subject of knowledge S is equipped
with language L, which was the language of the classical predicate calculus.
758 M. Omyła
One of the consequences of that assumption was that object of knowledge M is a model
the language L, i.e. it is a structure of the type: (U, R1 , R2 , . . . , Rn ). From philosophical
point of view it means that the world is considered as universe of objects which inhere
properties and stay in certain relations. Names of the language L refer to objects which
are elements of the universe of the model M. One-place predicates refer to properties
of objects and many-place predicates refer to relations occurring between objects. The
question arises: what sentences of language L in model M refer to?
Answers this question are in papers Suszko [6, 8]. He introduces there the notion
of generalized denotation for sentence formulas. And namely: any sentence formula
α(x) of language L refers in a model M to all of those objects which satisfy this
formula in the model M, i.e. {x ∈ U : α(x)}. If the formula α(x) is a sentence, it
means, if there are no free variables in the formula, then whether every object of the
universe satisfies it or no object in the universe satisfies it. Hence there are in model
M only two generalized denotations of sentences and all true sentences have the same
generalized denotation and all false sentences have one common generalized denotation.
The generalized denotation of true sentence in model M constitutes universe of model U ,
and the generalized denotation of false sentence is empty set. Because the sum total of
generalized denotations of sentences in model M consists of two elements, both sentence
variables and quantifiers binding those variables are redundant. What is important from
our point of view, Suszko almost from the beginning of his scientific carrier assigned to
sentences not only logical values (truth and false) but also semantic correlates, which he
called “generalized denotations of sentences”.
If set of generalized denotations of sentences equipped in set-theoretical operations
corresponding with logical connective, then algebra semantic correlate sentences will be
isomorphic with two-elements algebra of logical values.
Because of that isomorphism, Frege could supposed that sentences are names of special
objects called “logical values of sentences”.
Under the influence of Tractatus Suszko modified his view and started to consider
situation presented in a sentence as semantic correlate of this sentence. Since Wittgenstein
wrote in Tractatus: 4.03 [. . . ] A proposition communicates a situation to us, and so it must
be essentially connected with the situation. And the connection is precisely that it is its
logical picture.
Besides Suszko was convinced that logic should not impose any quantitative restraints
on the universe of semantic correlates, except the one: there are at least two correlates
of sentences, because correlate of any true sentence is different from a correlate of false
sentence.
The name “non-Fregean logic” is justified by the fact that in this logic there no
theorems asserting how many semantic correlates of sentences there can be.
At the base of non-Fregean logic lies also convictions, that syntactic categories
of linguistic expressions should conform to ontological categories of theirs semantic
correlates. On behalf of that conformity Suszko postulated—after Wittgenstein—that
situations stated by the sentences constitute semantic correlates of these sentences, and
sentential variables take their values from the universe of all situations correlated with a
given language.
Syntax and semantics of non-Fregean logic displays a logic-philosophical parallelism
between language and reality: we have names, functors and quantifiers in language
and—objects, situations and functions in the reality; and functors and quantifiers refer
to certain kind of functions.
However one can’t conclude from above, that every object in a given universe of our
discourse has a name, but one can conclude that every object may be a value of certain
name variable. Similarly not every situation or a state of affairs occurring between objects
of our discourse’s universe may be described with sentences of our language, but if there
are in the language sentential variables, then every of those situations may be a value
of certain sentential variable. Let us notice that sentential variable differ fundamentally
from other kinds of variables, because they are at the same time sentential formulas, and
therefore they enter into logical connections with the rest of sentences and sentential
formulas of a given language. Because of that the logical consequence influences the
interpretation of sentential formulas.
2.2 Axiomatic Form of Non-Fregean Logic
To speak in formal way about the structure of universe of situations and universe of
objects, Suszko introduced to literature of logic languages which he called W-languages
(in honor of L. Wittgenstein). In the alphabet of these languages, there are:
(1) two kinds variables: sentential variables: p, q, r, . . . , and nominal variables:
x, y, z, . . . ; (2) truth-functional connectives: ¬ (negation), ∧ (conjunction), ∨
(disjunction), ↔ (implication), ↔ (equivalence); (3) predicate-letters: P1 , P2 , . . . , Pn ; (4)
function-symbols: F1 , F2 , . . . , Fm ; (5) symbols identity: identity connective and identity
predicate which both symbolized by the sign “≡”; (6) quantifiers: ∀, ∃ binding both kinds
of variables.
Each of the quantifiers may bind both sentential variable or nominal one, depending
on which variable follows directly after it. Analogous the context uniquely determines
whether we have to do with identity connective or identity predicate since the expression
x ≡ p is not a formula of the language discussed. A detailed description of the syntax
of the W-kind languages has been presented in the papers: Bloom [3], Suszko [10, 12]
Operation Cn on L is generated by the Modus Ponens rule and the schemas of logical
axioms. To describe the consequence Cn in W-languages the following notations are
introduced:
Letters: v, w, v1 , w1 , v2 , w2 , . . . will be metalanguage variables denoted depending on
the context, either sentential variables or the nominal variables. By the letters: α, β, γ , . . . ,
will be denote any sentential formulas, by the letters: ζ, ξ, τ, . . . , we denote any nominal
formulas, and finally: φ, ϕ, ψ, . . . , denote sentential formulas or nominal ones, depending
on the context. Symbols α[v/φ] denoted result substitution in formula α(v) for free
variable v the expression φ.
The result of proceeding the formula α by any finite number of universal quantifiers,
i.e. ∀v1 ∀v2 . . . ∀vn α, where n ≥ 0 is called generalization of the formula α. For any set
of sentential formulas X, by Gen(X) will be denoted the set of all generalizations of
formulas in the set X.
The formulas of the form φ ≡ ϕ, are called equations.
760 M. Omyła
The structural version non-Fregean logic in W-language L is introduced by accepting

logical axioms and the only inference rule Modus Ponens.
The logical axioms are those formulas which are generalizations of any formula of the
following sorts:
(A1) Axiom Schemata for truth-functional connective (they are classical)
(A2) Axiom Schemata for quantifiers
(i) ∀vα → α[v/φ]
(ii) α → ∀vα (if v is not free in α)
(iii) ∀v(α → β) → (∀vα → ∀vβ)
(iv) ∃vα ↔ ¬∀v¬α
(A3) Axiom Schemata for identity connective and predicate:
(A3.1) Congruence axioms. All formulas of the form:
(i) ϕ ≡ φ (when φ, ϕ vary in at most bound variables)
for every functor we accept the invariance axiom:
(ii) ϕ1 ≡ φ1 ∧ ϕ2 ≡ φ2 ∧ . . . ∧ ϕn ≡ φn → (ϕ1 , ϕ2 , . . . , ϕn ) ≡ (φ1 , φ2 , . . . , φn )
(iii) ∀v(α ↔ β) → (Qvα ≡ Qvβ), where Q = ∀, ∃.
(A3.2) Special axiom for identity:

ϕ ≡ φ → (α[v/ϕ]) → α[v/φ])
The set logical axioms LA is the sum of three sets: A1, A2, A3 i.e. LA = A1 ∪ A2 ∪ A3.
A set of all the sentential formulas which are derivable from any set X and from logical
axioms in any finite number of steps through the application of MP rule is called theory
and is denoted by Cn(X).
A formula α is called logical theorem non-Fregean logic iff α ∈ Cn(∅).
2.3 Properties of Non-Fregean Logic
If in logical theorems of non-Fregean logic we replace at every place the sign “≡” with
sign “↔”, then we receive theorems of classical logic. It means that the non-Fregean
logic constitutes a generalisation of classical logic and the classical logic constitutes a
reinforcement of non-Fregean logic.
The sentence:
(AF) ∀p∀q[(p ≡ q) ≡ (p ↔ q)]
Suszko called “ontological version of Frege’s axiom”. In non-Fregean theories, in which
(AF) is a theorem the connectives “≡” and “↔” are indistinguishable. The non-Fregean
theorie in which (AF) is a theorem are classical theories expressed in non-Fregean
language. We can also derive classical logic by addition to its theorems the—seemingly
weaker than (AF)—axiom:
∀p∀q[(p ↔ q) → (p ≡ q)]
From the philosophical point of view the most important properties of non-Fregean logic
is its logical bivalence and extentionality. The logical bivalence of non-Fregean logic finds
expression in the following theorem of that logic:
∀p∀q∀r[(p ↔ q) ∨ (q ↔ r) ∨ (p ↔ r)]
In turn the extentionality of this logic finds its expression in the fact, that schemas (2) and:
ϕ ≡ φ → (α[v/ϕ] ≡ α[v/φ])
ϕ ≡ φ → (α[v/ϕ] → α[v/φ])
are schemas of logical theorems. These schemas state that expressions that have the
same semantical correlates are mutually interchangeable in all sentential contexts without
accordingly changing semantic correlate of those contexts (salva identitate) and without
changing logical values of those contexts (salva veritate).
To logical theorems of non-Fregean logic belong theorems:
∃x(x ≡ x), ∃p∃q¬(p ≡ q)
which state accordingly that the universe of objects is non-empty and that universe of
situations contains at least two elements. To logical theorems of non-Fregean logic though
do not belong any conditions, that limit “from above” the number of objects and situations
in universe, what means that for every natural number n the following formulas are not
logical theorems:
(x1 ≡ x2 ) ∨ (x1 ≡ x3 ) ∨ . . . ∨ (xn−1 ≡ xn )

(p1 ≡ p2 ) ∨ (p1 ≡ p3 ) ∨ . . . ∨ (pn−1 ≡ pn )
Non-Fregean logic—as every calculus—can be developed without any philosophical

presumptions. Nonetheless for this logic the ontology of situations contained in Tractatus
served as a fundament. This logic presumes the ontology, according to which there exist
objects, situations and functions. Suszko introduced attention to the fact that the division
all of beings into objects, situations and functions has logical character, i.e. it results the
fact, that we describe the world with languages in which we have names, predicates,
connectives and quantifiers.
Suszko introduced non-Fregean logic because he was convinced that there exist in the
world certain beings and aspects of beings, which can be properly told about with the
aid of sentential variables. In other words, the ontology that underlies non-Fregean logic
contains the view, that it is not enough to consider the world as the universe of objects
only, but we have to consider it also as the universe of possibilities, among which some
of them become realized, i.e.—there are facts. In [11] Suszko wrote: “. . . perceiving an
object x consists of perceiving at least one situation that x is so-an-so”.
It is worth stressing that the importance of non-Fregean logic goes beyond its
ontological applications. Since this logic has initiated research on abstract logics. Those
logics constitutes new chapter in application of algebra to logic. This problems are
discussed among others in monographs Czelakowski [4], Dzik [5].
762 M. Omyła
References
1. Ajdukiewicz, K.: Das Weltbild Und die Begriffsapparatur. Erkenntnis IV, 259–287 (1934)
2. Ajdukiewicz, K.: The scientific world perspective. Erkenntnis V, 22–30 (1935)
3. Bloom, S.L.: A completeness theorem for “theories of kind W”. Stud. Logica 27, 43–55 (1971)
4. Czelakowski, J.: Protoalgebraic Logics. Kluwer, Dordrecht (2001)
5. Dzik, W.: Unification Types in Logic. Silesian University Press, Katowice (2007)
6. Suszko, R.: Syntactic structure and semantical reference II. Stud. Logica 9, 63–91 (1960)
7. Suszko, R.: An abstract scheme of development of knowledge. In: Actes X Congress International
d’Histoire des Sciences Varsovie 1965, pp. 52–55. Ossolineum, Wrocław (1967)
8. Suszko, R.: An essay in the formal theory of extension and of intension. Stud. Logica 20, 7–36 (1967)
9. Suszko, R.: Formal logic and the development of knowledge. In: Colloquium in Philosophy of
Science. London 1965, pp. 210–222. North Holland, Amsterdam (1968)
10. Suszko, R.: Ontology in the Tractatus of L. Wittgenstein. Notre Dame J. Form. Log. 9, 97–121 (1968)
11. Suszko, R.: Non-Fregean logic and theories. Analele Universitatii Bucuresti. Acta Logica 11, 7–33
(1968)
12. Suszko, R.: Identity connective and modality. Stud. Logica 27, 7–39 (1971)
M. Omyła ()
Faculty of Law and Administration, Cardinal Stefan Wyszyński University in Warsaw, Warsaw, Poland
Categories of First-Order Quantifiers
Abstract One well known problem regarding quantifiers, in particular the 1st-order
quantifiers, is connected with their syntactic categories and denotations. The unsatis-
factory efforts to establish the syntactic and ontological categories of quantifiers in
formalized first-order languages can be solved by means of the so called principle of
categorial compatibility formulated by Roman Suszko, referring to some innovative
ideas of Gottlob Frege and visible in syntactic and semantic compatibility of language
expressions. In the paper the principle is introduced for categorial languages generated
by the Ajdukiewicz’s classical categorial grammar. The 1st-order quantifiers are typically
ambiguous. Every 1st-order quantifier of the type k > 0 is treated as a two-argument
functor-function defined on the variable standing at this quantifier and its scope (the
sentential function with exactly k free variables, including the variable bound by this
quantifier); a binary function defined on denotations of its two arguments is its denotation.
Denotations of sentential functions, and hence also quantifiers, are defined separately
in Fregean and in situational semantics. They belong to the ontological categories that
correspond to the syntactic categories of these sentential functions and the considered
quantifiers. The main result of the paper is a solution of the problem of categories of the
1st-order quantifiers based on the principle of categorial compatibility.
Keywords 1st-order quantifiers · Categorial languages · Syntactic categories ·

Denotation · Ontological categories · Denotational semantics · Compositionality ·
Categorial compatibility
Mathematics Subject Classification (2000) Primary 03C07, 03H05; Secondary 03A99
The main problems and their solutions connected with the subject of this paper were presented at the
Logic Colloquium’15 held in Helsinki on August 3–8, 2015, co-located with the 15th International
Congress of Logic, Methodology and Philosophy of Science, CLMPS (see my abstract [35]; see
also [36]).

1 Introduction
Around 1879 Frege and—independently—Charles Sanders Peirce developed a way to

extend sentential logic by introducing symbols representing determiners, such as ‘all’,
‘some’, ‘no’, ‘every’, ‘any’, and so on.
Frege and Peirce used two symbols: the universal quantifier (which we will write ∀)
corresponding roughly to the English ‘all’, ‘every’ and ‘each’ and the existential quantifier
(which we will write ∃) corresponding to the English ‘some’, ‘a’, ‘an’.
In this paper we will consider only standard, Fregean quantifiers ∀ and ∃ of the 1st-
order as individual variable-binding operators. They are used in formulas of predicate
logic of the 1st-order and in formalized languages of elementary theories based on this
logic. Their syntactic role and semantic references, i.e. denotation, extension, created
some problems that have not been satisfactorily solved yet.
In the next part (Sect. 2) I shall partially explicate the problem of quantifiers. In Sect. 4,
I’ll outline some intuitive foundations of my theory of categorial languages which gives
the formal direction for justification of my solution of the problem of quantifiers. It
corresponds to the principle (CC) of categorial compatibility based on some Frege’s ideas
and was formulated by Roman Suszko [24]. The solution of the problem is presented in
Sect. 5.
2 Problem of Quantifiers
The problem of quantifiers is connected with the difficulty pertaining to establishing their
syntactic and semantic categories.
Leśniewski’s theory of semantic/syntactic categories [13, 14], which was improved by
Ajdukiewicz [1] by introducing categorial indices, does not, obviously, solve this problem,
which limits the universal character of the theory.
Leśniewski’s hierarchy of semantic/syntactic category does not include any variable-
binding operators. Leśniewski, in his protothetics and ontology systems, allows only one
operator—the universal quantifier, noting it as parentheses, Ajdukiewicz, on the other
hand, indicates the difficulty of assigning to quantifiers the index s/s or s/ns.
Assigning to them the index s/s, i.e. the category of sentence-forming functors of one-
sentence argument, would mean that the quantifiers belong to the same category as one-
argument connectives, and assigning to them the index s/ns of sentence-forming functors
of one-name and one-sentence arguments would mean that we include them into the same
category as some expressions of indirect speech, e.g. ‘think that’, ‘know that’, etc.
It has been suggested that the categorial grammar, which Bar-Hillel derived from Aj-
dukiewicz’s version of the theory of semantic/syntactic categories, does not satisfactorily
account for the role of bound variables and operators binding them.
Suszko [22, 24] assigns to them the index s//s/n, and thus the index of sentence-forming
functor of the argument, which is a one-argument predicate. In this way, the index, for
example in the sentence ‘∀x(x flows)’ pertains to the entire quantifier-variable pattern
‘∀x(x . . .)’ (see Simons [20]) which corresponds to English word ‘everything’ (see also
Cresswell [4], Simons [21]).
Categories of First-Order Quantifiers 765
Suszko and many other researchers of language syntax treat quantifiers as

expressions independent of the quantifier variable. Generally, researchers avoid
bound variables in attempting to solve the problem, for example by means of
combinators (Curry [5, 6], Curry and Feys [7], see e.g. Simons [20]).
But earlier, Suszko stated that mounting variable-binding operators into a syntactic
scheme requires general principles other than the theory of syntactic/semantic categories.
The principle (CC) of categorial compatibility is one such principle. It allows us to
assign to every expression of a formalized 1st-order language, which possesses an index
symbolizing a syntactic category, a denotation whose ontological category (relative to the
universe U of a given model of the language) is indicated by the same index.
Suszko assumes that
• the denotation of the entire expression ∀x(e(x)), where e(x) is a sentential function
with the free variable x, is ether the logical value 1 (of truth) or the logical value 0 (of
falsity) which belong to the ontological category with the index s, and
• the denotation of the universal quantifier ∀ is the function of generalization which has
the value 1 in only one case, if its argument is the universe U .
The function of generalization belongs to the ontological category with the index
s//s/n because its arguments are any sets belonging to the family P (U ) included into the
ontological categories with the index s/n. In this way the principle (CC) holds although
the principle of syntactic connection (SC) does not hold because no index is assigned to
quantifier variable x, and the scope of the quantifier ∀ (here e(x)) is not one-argument
predicate of the syntactic category with the index s/n.
In the next parts of this paper I explicate both the principle (SC) of syntactic connection
and the principle (CC) of categorial compatibility on the basis of my theory of categorial
languages [30–34] which allows us to give some solutions to the problem of quantifiers.
The essence of the approach proposed here is considering them to be typical syntactic
notions: functors-functions mapping language expressions into language expressions that
correspond to some functions on extralinguistic objects—on denotations of arguments of
these functors.
Let us note that a standard background for research in the field of mentioned quantifiers
assumes treating them as some functions or relations on extralinguistic objects, mostly
functions with index t//t/e (cf. Mostowski [18], Lindström [15], Montague [16, 17],
Nowaczyk [19], van Benthem [27, 28], van Benthem and Westerståhl [29]).
3 Some Intuitive Foundations of the Theory of Categorial

Languages
3.1 Main Ideas of Formalization of Categorial Language
In the paper, formal-logical considerations relate to syntax and extensional semantics of

any language L characterized categorically:
• in the spirit of some ideas of Husserl [12] and Leśniewski–Ajdukiewicz’s theory of
syntactic/semantic category (see Leśniewski [13, 14], Ajdukiewicz [1, 2]),
• in accordance with Frege’s ontological canons [10],

• in accordance with Bocheński’s motto [3]: syntax mirrors ontology, and
• some ideas of Suszko [22–25]: language should be a linguistic scheme of ontological
reality and simultaneously a tool of its cognition.
The paper includes developing and some explications of these authors’ ideas. It also
presents, in a synthetic form, some ideas presented in my papers published in [30–34].
Language L is there defined, if the set S of all well-formed expressions (briefly wfes)
is determined. These expressions must satisfy requirements of categorial syntax and
categorial semantics.
3.2 Categorial Syntax
The categorial syntax of L is connected with generating the set S by the classical
categorial grammar and belonging wfes of S to appropriate syntactic/semantic categories.
A characteristic feature of categorial syntax is that each composed wfe of the set S has a
functor-argument structure, in this sense that, in accordance with the principle originated
by Frege [8], it is possible to distinguish in it its constituent called the main functor, and
the other constituents—called arguments of that functor, yet each constituent of the wfe
has a determined syntactic category.
If e is a functor-argument wfe of S, f is its main functor and e1 , e2 , . . . , en its
subsequent arguments then e can be written in the functional-argument form:
e = f (e1 , e2 , . . . , en ). (e)
In categorial approach to the language L, syntactic categories of wfes of L are

determined by attributing to them, like their expressions, categorial indices of a certain
set I . To every wfe e of the set S is unambiguously assigned a categorial index (type)
iS (e) of the set I ; wfes belonging to the same syntactic category CATa have the same
categorial index a.
Categorial indices were introduced by Ajdukiewicz [1] into logical semiotics with
the aim to determine the syntactic role of expressions and to examine their syntactic
connection, in compliance with the principle of syntactic connection (SC) discussed
below.
The set S of all wfes of L is then intuitively defined as the smallest set including the
vocabulary of L and closed with respect to the principle (SC), which in free formulation
says that
(SC) The categorial index of the main functor of each functor-argument expression of
the language L is formed out of the categorial index of the expression which the
functor forms together with its arguments, as well as out of the subsequent indices
of arguments of this functor.
In the formal definition of the set S it is required that each functor-argument constituent
of the given expression should satisfy the principle (SC).
If the functor-argument expression e = f (e1 , e2 , . . . , en ) is a wfe (it belongs to the set
S), then in accordance to the principle of syntactic connection (SC) the index of its main
functor f formed from the index a of e and successive indices a1 , a2 , . . . , an of successive

arguments e1 , e2 , . . . , en of the functor f , can be written in the following quasi-fractional
form:
iS (f ) = iS (e)/iS (e1 )iS (e2 ) . . . iS (en ) = a/a1 a2 . . . an . (iS )
3.2.1 An Algebraic Structure of Categorial Language
In categorial language L we can distinguish two sets: the set B of all basic wfes of S and
the set F of all functors of S such that
S = B ∪ F and B ∩ F = ∅,
where functors of the set F differ from basic expressions of B that they have indices
formed from simpler ones. If the functor f has the functoral index of the form (iS ), i.e. the
index of the form a/a1a2 . . . an then it belongs to the syntactic category CATa/a1a2 . . . an
and so to the category of functors forming expressions with the index a if their arguments
are n expressions with successive indices a1 , a2 , . . . , an . So the functor f can be treated
as the following partial function defined on wfes of S:
f : CATa1 × CATa2 × . . . × CATan → CATa
mapping of wfes from Cartesian product of syntactic categories CATa1 , CATa2 , . . . , CATan
into the category CATa. Then we have
f ∈ CATa/a1 a2 . . . an = CATaCATa1 ×CATa2 ×...×CATan . (CATf )
In this way we simultaneously can regard the categorial language L as an algebraic

structure L, partial algebra with the carrier S and the set Fo ⊆ F of partial functions on S
(simple functors of L):
L = S, Fo! .
3.3 Categorial Semantics
Categorial extensional semantics is connected with denotations of wfes of S and with

their belonging to an appropriate semantic extensional category. Each constituent of the
composed wfe has determined a semantic extensional category and also a denotation,
and thus—an ontological category (the category of ontological objects). Denotations
(extensions) of wfes of L are sets of object references (references) of wfes of L, objects
of the cognized reality, e.g.: individuals, sets of individuals, states of affairs, operation on
the indicated objects, and the like.
We will concentrate only on referential relationships between expressions of L and
reality to which they refer. We enrich the categorial grammar generating L by the
denotation operation δ regarded as its semantic component. The denotation operation

δ assigns to every wfe of the set S an object of ontological reality ONT describing by the
language L—its denotation belonging to an ontological category. So
δ : S → ONT, (δ)
where ONT is the sum of all ontological categories corresponding to wfes of S.

According to some innovative ideas of Frege [9, 10], Bocheński’s (his famous
motto: syntax mirrors ontology) and Suszko [22–24] who anticipated the research in
categorial semantics and was the first to use categorial indices as a tool for coordination
of expressions and their references, extralinguistic objects, the mutual dependence of
syntactic and semantic formal description of L should be considered by keeping the
principle (CC) of categorial compatibility, based on the compatibility of the syntactic
category of each language expression of L with the ontological category assigned to its
denotation. The principle (CC) of syntactic and semantic, i.e. also ontological categorial
compatibility in Suszko’s formulation can be given by keeping for any wfe e of categorial
language L the relationship:
e ∈ CATι iff δ(e) ∈ ONTι, (CC)
where CATι and ONTι are: the syntactic category and the ontological category, respec-
tively, with the same categorial index ι, and δ is the operation of denotation.
From the principle (CC) it follows that for any e = f (e1 , e2 , . . . , en ) ∈ S with
the main functor-function f ∈ CATa/a1 a2 . . . an satisfying the condition (CATf ) the
following conditions are satisfied:
δ(f ) ∈ ONTa/a1 a2 . . . an = ONTaONTa1 ×ONTa2 ×...×ONTan (ONTf )
and
δ(f (e1 , e2 , . . . , en )) = δ(f )(δ(e1 ), δ(e2 ), . . . , δ(en )). (PCD)
The condition (ONTf ) states that the denotation (object reference) of the main functor
of the composed wfe e of the set S is the set-theoretical function mapping the Cartesian
product of ontological categories ONTa1 × ONTa2 × . . . × ONTan into the ontological
category ONTa and it is defined by means of the condition (PCD) connected with some
Frege’s ideas and called the principle of compositionality of denotation.
3.3.1 An Algebraic Ontological Structure Corresponding to the Partial

Algebra L
The operation δ assigns the following ontological structure RL of a reality corresponding

to language L to the algebraic structure L:
RL = ONT, ONTFo ! ,
where ONTFo is the sum of all ontological categories corresponding to all functors of
the set Fo. The structure RL is a partial algebra similar to the algebra L and the principle
(PCD) is simultaneously the condition of homomorphism of the algebra L into the algebra
RL , i.e.
δ : S, Fo! −−→ ONT, ONTFo ! .

hom
A model of language L is the structure of homomorphic images of components of L,

i.e. the substructure ML = δ(S), δ(Fo)! of the structure RL .
If we distinguish in the set B of basic wfes of S the category CATs of all sentences of
language L, then the notion of truthfulness of any sentence e ∈ CATs in the model ML is
defined as follows:
e is a true sentence in the model ML iff δ(e) ∈ T , (T)
where T is primitive notion of the considered theory intuitively understood either as the
singleton with the true value (in Fregean semantics) or as the set of all states of affairs
that take place (in situational semantics).
4 The Solution of the Problem of Quantifiers of 1st-Order
The unsatisfactory efforts to establish, in the sense of the principle (CC) of categorial
compatibility, the category of quantifiers in formalized 1st-order languages can be solved
by means of notions and statements of the above outlined theory of categorial languages.
Let L1 be any 1st-order formalized language. Let us treat any standard quantifier of L1
as a context-dependent functor of two arguments:
1. a quantifier variable (the variable accompanying this quantifier) and
2. its scope, i.e. a sentential function including as a free variable the same variable as the
quantifier variable.
4.1 Different Types of the 1st-Order Quantifiers and Their

Syntactic Categories
A standard, the 1st-order quantifier is a functor forming a new sentential function (in
particular a sentence of L1 ) in which there occur one free variable less than in the scope
of this quantifier (the variable bound by the quantifier). As such a functor, a quantifier can
be treated as a set-theoretical function relative to the number of free individual variables
occurring in its scope. So, we should not speak of one existential ∃ or one universal
quantifier ∀ but about different types of such quantifiers depending of the number of free
variables in their scope. We will use numerical superscripts in order to point out these
different types of quantifiers.
Let
• Var be the set of all individual variables for L1 , with categorial index n1 ;
• S = S0 —the set of all its sentences, with the categorial index s;
• Sk (k ≥ 1)—the set of all sentential functions in which exactly k free variables occur,
with the index sk .
For example, if α(x1 , x2 , x3 ) ∈ S3 , where x1 , x2 , x3 ∈ Var, then the expressions:
∀3 x2 α(x1 , x2 , x3 ) ∈ S2 ,
∃2 x3 ∀3 x2 α(x1 , x2 , x3 ) ∈ S1 ,
∀1 x1 ∃2 x3 ∀3 x2 α(x1 , x2 , x3 ) ∈ S0
and quantifiers ∀3 , ∃2 , ∀1 belong to different syntactic categories with indices s2 /n1 s3 ,

s1 /n1 s2 , s/n1 s1 , respectively.
More generally, the quantifiers ∀k and ∃k (k ≥ 1) are treated as the functors-functions:
∀k , ∃k : Var × Sk → Sk−1 (S0 = S).
Thus, in accordance to (CATf ), for k > 0 we have
(CAT∀k , ∃k ) ∀k , ∃k ∈ CATsk−1 /n1 sk (s0 = s),
and the principle of syntactic connection (SC) for them is satisfied.

Their denotations and ontological categories should be defined in such a way as to
satisfied the principle (CC) of categorial compatibility (their denotations should belong to
the ontological category ONTsk−1 /n1 sk ) and the principle (PCD) of compositionality of
denotation.
Let the denotation operation for the language L1 be the function d in Fregean, standard
semantics and the function d in the situational, non-standard semantics:
d, d : S(L1 ) → ONT(L1 )
mapping the set S(L1 ) of all wfes of L1 into the set ONT(L1 ) which is the sum of all
ontological categories in the ontological structure RL1 .
We will give here two possible solutions of denotations of quantifiers of the 1st-order
taking into account two different ways of understanding of the denotation of sentences
and sentential functions presented below.
4.2 Denotations of 1st-Order Quantifiers and Their Ontological

Categories
4.2.1 Fregean Semantics
We assume that if U is the universe of individuals in an established model ML1 of L1 , 1

is the value of truth, 0—the value of falsity then
d(x) ∈ {U } = ONTn1 for any x ∈ CATn1 = Var;

d(p) ∈ {0, 1} = ONTs for any p ∈ CATs = S;
k
d(sf) ∈ 2U = ONTsk for any sf ∈ CATsk = Sk (k ≥ 1)
and for any x1 , x2 , . . . , xk ∈ Var and for any sf = α(x1 , x2 , . . . , xk ) ∈ Sk
d(α(x1 , x2 , . . . , xk )) =
{(u1 , u2 , . . . , uk ) ∈ U k | d(α o (x1 /u1 , x2 /u2 , . . . , xk /uk )) = 1},
where α o (x1 /u1 , x2 /u2 , . . . , xk /uk ) is a sentence which we get from sentential function sf
by replacement of its all free variables x1 , x2 , . . . , xk of Var by suitable individual names
of individuals u1 , u2 , . . . , uk of the universe U , i.e. the denotation of sf is the set of all
k-tuples from U k which satisfy this sentential function.
Denotation for the quantifier ∀k of the type k(k ≥ 1) is defined by induction as
follows:
(a) for k = 1 and any α(x) ∈ S1

1, d(x) = U = d(α(x))
d(∀ xα(x)) = d(∀ )(d(x)), d(α(x))) =
1 1
.
0, d(x) = U = d(α(x));
According to (a) the quantifier sentence obtained from any sentential function α(x) by
preceding it with the universal quantifier ∀1 is a true sentence in the established model
ML1 of L1 with the universe of individuals U iff every object of the universe U satisfies
the α(x) which is the scope of ∀1 .
(b) for k = j + 1(j > 0) and any α(x1 , x2 , . . . , x, . . . , xj +1 ) ∈ Sj +1
d(∀j +1 xα(x1 , x2 , . . . , x, . . . , xj +1 )) =
= d(∀j +1 )(d(x), d(α(x1 , x2 , . . . , x, . . . , xj +1 ))) =
= {(u1 , u2 , . . . , uj +1 ) ∈ U j | d(α o (x1 /u1 , x2 /u2 , . . . , x/u, . . . , xj +1 /uj +1 ))= 1
for each u ∈ U }.
According to (b) the denotation of the sentential function sfk−1 ∈ Sk−1 obtained from
the sentential function α(x1 , x2 , . . . , x, . . . , xj +1 ) ∈ Sk (k > 1) by binding the variable
x by the universal quantifier ∀k (k = j + 1 > 1) is the set of all j = (k − 1)-tuples
(u1 , u2 , . . . , uk−1 ) of individuals of U such that all sentences obtained by the substitution
of all j free variables in sfk−1 , respectively, by names of individuals of these tuples and
names of any individuals of U representing x are true; in other words the denotation
of sfk−1 is the set of all such (k − 1)-tuples (u1 , u2 , . . . , uk−1 ) of individuals of U that
for any individual u of U k-tuples (u1 , u2 , . . . , u, . . . , uk−1 ) build from them satisfy the
scope α(x1 , x2 , . . . , x, . . . , xj +1 ) of the quantifier ∀k .
Thus for any k ≥ 1
ONTn1 ×ONTsk
d(∀k ) ∈ ONTsk−1 /n1 sk = ONTsk−1 .
Similarly for d(∃k ):


1, d(x) ∩ d(α(x)) = ∅
d(∃ xα(x)) = d(∃ )(d(x)), d(α(x))) =
1 1
.
0, d(x) ∩ d(α(x)) = ∅;
d(∃j +1 xα(x1 , x2 , . . . , x, . . . , xj +1 )) =
= d(∃j +1 )(d(x), d(α(x1 , x2 , . . . , x, . . . , xj +1 ))) =
= {(u1 , u2 , . . . , uj +1 ) ∈ U j | d(α o (x1 /u1 , x2 /u2 , . . . , x/u, . . . , xj +1 /uj +1 ))= 1
for some u ∈ U }.
According to (a) the quantifier sentence obtained from any sentential function α(x)
by preceding it with the existential quantifier ∃1 is true sentence in the established model
ML1 of L1 with the universe of individuals U iff at least one object of the universe U
satisfies the α(x) which is the scope of ∃1 .
According to (b) the denotation of the sentential function sfk−1 ∈ Sk−1 obtained from
the sentential function α(x1 , x2 , . . . , x, . . . , xj +1 ) ∈ Sk (k > 1) by binding the variable
x by the existential quantifier ∃k (k = j + 1 > 1) is the set of all j = (k − 1)-tuples
(u1 , u2 , . . . , uk−1 ) of individuals of U such that all sentences obtained by the substitution
of all j free variables in sfk−1 , respectively, by names of individuals of these tuples and
the substitution some individual name of u for x are true; in other words the denotation
of sfk−1 is the set of all such (k − 1)-tuples (u1 , u2 , . . . , uk−1 ) of individuals of U that
for some individual u of U k-tuples (u1 , u2 , . . . , u, . . . , uk−1 ) build from them satisfy the
scope α(x1 , x2 , . . . , x, . . . , xj +1 ) of the quantifier ∃k .
Thus, for any k ≥ 1
ONTn1 ×ONTsk
d(∃k ) ∈ ONTsk−1 /n1 sk = ONTsk−1 .
Moreover, the principle (CC) is also valid for ∀k and ∃k in situational semantics.
4.2.2 Situational Semantics
In situational semantic we assume that
d(x) ∈ {U } = ONTn1 for any x ∈ CATn1 = Var;

d(p) ∈ {St} = ONTs for any p ∈ CATs = S,
where St is the set of all states of affairs, St = T ∪ F, T ∩ F = ∅ and T is the nonempty

set of all states of affairs that take place and F —the nonempty set of remaining states of
affairs. Stk ⊂ St is the set of states of affairs with k individuals.
d(sf ) ∈ 2Stk = ONTsk for any sf ∈ CATsk = Sk
and for any x1 , x2 , . . . , xk ∈ Var and for any sf = α(x1 , x2 , . . . , xk ) ∈ Sk
d(α(x1 , x2 , . . . , xk )) =
{s ∈ Stk | s = d(α 0 (x1 /u1 , x2 /u2 , . . . , xk /uk )) for any (u1 , u2 , . . . , uk ) ∈ U k }.
So, if the denotation operation is understood here as the operation d then the
denotations of sentences are states of affairs and the denotation of any sentential function
is the set of all states of affairs that are denotations all sentences represented by the
sentential function.
Denotation for the quantifier ∀k is defined by induction as follows:
d(∀1 xα(x)) = d(∀1 )(d(x)), d(α(x))) ∈ T iff d(α o (x/u)) ∈ T for each u ∈ U ;
d(∀j +1 xα(x1 , x2 , . . . , x, . . . , xj +1 )) =
= d(∀j +1 )(d(x), d(α(x1 , x2 , . . . , x, . . . , xj +1 ))) =
= {s ∈ St | s = d(α o (x1 /u1 , x2 /u2 , . . . , x/u, . . . , xk /uk ))
for each u ∈ U, any (u1 , u2 , . . . , uj +1 ) ∈ U j }.
According to (a) the quantifier sentence obtained from any sentential function α(x) by
preceding it with the universal quantifier ∀1 is a true sentence in an established model ML1
of the language L1 with the universe of individuals U iff every sentence representing this
sentential function is true (because their denotations are states of affairs that take place).
According to (b) the denotation of sentential function sfk−1 ∈ Sk−1 obtained from the
sentential function α(x1 , x2 , . . . , x, . . . , xj +1 ) ∈ Sk (k > 1) by binding the variable x by
the universal quantifier ∀k is the set of all denotations of sentences (intuitively—the set
of all states of affairs describing by these sentences) which can be obtained from sfk−1
by replacing all free variables in it with individual names of any individuals of U ; in
other words, it is the set of all denotations of sentences (all states of affairs) which can
be obtained from α(x1 , x2 , . . . , x, . . . , xj +1 ) by replacement for the variable x binding
by ∀k individual names of any individual of U (of the denotation of this variable) and for
remaining variables in it also individual names of any individuals of U .
ONTn1 ×ONTsk
d(∀k ) ∈ ONTsk−1 /n1 sk = ONTsk−1 .
Similarly for d(∃k ):

d(∃1 xα(x)) = d(∃1 )(d(x)), d(α(x))) ∈ T iff T ∩ d(α(x)) = ∅
d(∃j +1 xα(x1 , x2 , . . . , x, . . . , xj +1 )) =
= d(∃j +1 )(d(x), d(α(x1 , x2 , . . . , x, . . . , xj +1 ))) =
= {s ∈ St | s = d(α o (x1 /u1 , x2 /u2 , . . . , x/u, . . . , xj +1 /uj +1 ))
for some u ∈ U, any (u1 , u2 , . . . , uj +1 ) ∈ U j }.
1 ×ONTsk
d(∃k ) ∈ ONTsk−1 /n1 sk = ONTsONTn
k−1 .
4.3 The Syntactic and Semantic Compatibility of Quantifiers
In our categorial approach to syntax and semantics of the 1st-order formalized lan-
guage L1 its quantifiers have been treated as context-dependent two-argument functors-
functions of different categorial types k > 0 (defined on the set Var of all its individual
variables and the set of all its sentential functions Sk with exactly k free variables) and
with values in the set of sentential functions Sk−1 possessing one free variable less or, in
particular, in the set of sentences S:
∀k , ∃k : Var × Sk → Sk−1 (S0 = S).
Thus, according to the condition (CATf ), quantifiers ∀k , ∃k belong to syntactic

categories:
CATn1 ×CATsk
(CAT∀k , ∃k ) ∀k , ∃k ∈ CATsk−1 /n1 sk = CATsk−1 (s0 = s),
and it means that they satisfy the principle (SC) of syntactic connection.
It was also shown that for the denotation operations:
d, d : S(L1 ) → ONT(L1 )
their denotations, according to the condition (ONTf ), belong to ontological categories:
ONTn1 ×ONTsk
d(∀k ), d(∀k ), d(∃k ), d(∃k ) ∈ ONTsk−1 /n1 sk = ONTsk−1 . (ONT∀k , ∃k )
5 Conclusions
From the conditions (CAT∀k , ∃k ) and (ONT∀k , ∃k ) follow the following conclusions:
1. the 1st-order quantifiers ∀k , ∃k (k > 0) satisfy the principle of syntactic connection
(SC) and the principle of categorial compatibility (CC)
and
2. the problem of standard quantifiers is solved by employing the conceptual apparatus
and statements of the outlined theory of categorial languages.
It should also be noted that
3. in languages with other operators biding variables the problem of their denotations
can be solved in an analogous way,
but
4. for branching quantifiers used in Independence-Friendly logic (see Hintikka [11]) the
outlined here denotational (compositional) semantics does not work.
However,
5. according to Frege’s ideas, the proposed categorial approach to language syntax and
semantics can be developed in the same spirit for formalized languages of higher order
than 1.
6. the proposed approach to semantics of the 1st-order formalized languages of differ
from the standard in the Tarski’s approach [26] and other improved versions; first of all
it refers to the concept of denotation of any language expression instead to the concept
of satisfaction—the crucial ancillary notion in the definition of truth; this notion may
be omitted in the definition of the concept of a true sentence and probably replaced by
the notion of denotation.
References
1. Ajdukiewicz, K.: Die syntaktische Konnexität. Stud. Philos. 1, 1–27 (1935) [English translation:
Syntactic connection. In: McCall, S. (ed.) Polish Logic 1920–1939, pp. 202–231. Clarendon Press,
Oxford (1967)]
2. Ajdukiewicz, K.: Zwiazki˛ składniowe mi˛edzy członami zdań oznajmujacych ˛ [Syntactical relations
between constituents of declarative sentences]. Stud. Filozoficzne 6(21), 73–86 (1960) [First pre-
sented in English at the International Linguistic Symposium in Erfurt, 27 September–2 October 1958]
3. Bocheński, J.M.: On the syntactical categories. New Scholasticism 23, 257–280 (1949)
4. Cresswell, M.J.: Logics and Languages. Mathuen and Co. Ltd, London (1973)
5. Curry, H.B.: Grundlagen der kombinatorischen Logik. Am. J. Math. 52, 509–536, 789–834 (1930)
6. Curry, H.B.: Some aspects of grammatical structure. In: Jakobson, R. (ed.) Structure of Language and
Its Mathematical Aspects, vol. 12, pp. 57–68. AMS, Providence, RI (1961)
7. Curry, H.B., Feys, R.: Combinatory Logic, vol. 1. North Holland, Amsterdam (1958)
8. Frege, G.: Begriffschrift, eine der arithmetischen nachbildete Formalsprache des reinen Denkens.
Halle (1879) [English translation in: Geach, P.T., Black, M. (eds.) Translations from the Philosophical
Writings of Gottlob Frege. Blackwell, Oxford (1970)]
9. Frege, G.: Die Grudlagen der Arithmetik: eine logische-mathematische Untersuchungen über den
Begriff der Zahl. Breslau (1884) [English translation: The Foundations of Arithmetic: A Logico-
Mathematical Enquiry into the Concept of Number. Blackwell, Oxford (1950)]
10. Frege, G.: Über Sinn und Bedeutung. Zeitschrift für Philosophie und pilosophishe Kritik 100, 25–
50 (1892) [English translation in: Feigel, H., Sellars, W. (eds.) Readings in Philosophical Analysis.
Appleton-Century-Crofts, New York (1949), and also In: Beaney, B. (ed.) The Frege Reader, pp. 151–
171. Blackwell, Oxford (1997)]
11. Hintikka, J.: Principles of Mathematics Revisited. Cambridge University Press, Cambridge (1996)
12. Husserl, E.: Logische Untersuchungen. Vol. I, Halle (1900), Vol. II, Halle (1901)
13. Leśniewski, S.: Grundzüge eines neuen Systems der Grundlagen der Mathematik. Fundam. Math. 14,
1–81 (1929)
14. Leśniewski, S.: Über die Grundlagen der Ontologie. Comptes rendus des séances de la Société des
Sciences et des Lettres de Varsovie, Classe II 23, 111–132 (1930)
15. Lindström, P.: First-order predicate logic with generalized quantifiers. Theoria 32, 186–195 (1966)
16. Montague, R.: Universal grammar. Theoria 36, 373–398 (1970)
17. Montague, R.: Formal Philosophy: Selected Papers of Richard Montague. Yale University Press, New
Haven, CT (1974). Ed. and introd. R.H. Thomason
18. Mostowski, A.: On generalization of quantifiers. Fundam. Math. 44, 12–36 (1957)
19. Nowaczyk, A.: Categorial languages and variable-binding operators. Stud. Logica 37, 27–39 (1978)
20. Simons, P.: Combinators and categorial grammar. Notre Dame J. Form. Log. 30 (2), 241–261 (1989)
21. Simons, P.: Languages with variable-binding operators: categorial syntax and combinatorial se-
mantics. In: Jadacki, J., Paśniczek, J. (eds.) The Lvov-Warsaw School—The New Generation.
Poznań Studies in the Philosophy of Sciences and Humanities, vol. 89, pp. 239–268. Rodopi,
Amsterdam/New York (2006)
22. Suszko, R.: Syntactic structure and semantical reference, Part I. Stud. Logica 8, 213–144 (1958)
23. Suszko, R.: Syntactic structure and semantical reference, Part II. Stud. Logica 9, 63–93 (1960)
24. Suszko, R.: O kategoriach syntaktycznych i denotacjach wyrażeń w j˛ezykach sformalizowanych [On
syntactic categories and denotation of expressions in formalized languages]. In: Rozprawy logiczne
[Logical Dissertations to the Memory of Kazimierz Ajdukiewicz], pp. 193–204. PWN, Warsaw
(1964)
25. Suszko, R.: Ontology in the tractatus of L. Wittgenstein. Notre Dame J. Form. Log. 9, 7–33 (1968)
26. Tarski, A.: The semantic notion of truth: and the foundations of semantics. Philos. Phenomenol. Res.
4(3), 341–376 (1944)
27. van Benthem, J.: Essays in Logical Semantics. Reidel, Dordrecht (1986)
28. van Benthem, J.: Quantifiers in the world of types. In: van der Does, J., van Eijck, J. (eds.) Quantifiers,
Logic and Language, pp. 47–61. Stanford University, Stanford, CA (1996)
29. van Benthem, J., Westerståhl, D.: Directions in generalized quantifier theory. Stud. Logica 53, 389–
419 (1995)
30. Wybraniec-Skardowska, U.: Theory of Language Syntax. Categorial Approach. Kluwer Academic,
Dordrecht/Boston/London (1991)
31. Wybraniec-Skardowska, U.: Logical and philosophical ideas in certain approaches to language.
Synthese 116(2), 231–277 (1998)
32. Wybraniec-Skardowska, U.: On denotations of quantifiers. In: Omyła, M. (ed.) Logical Ideas of
Roman Suszko, pp. 89–119. Faculty of Philosophy and Sociology of Warsaw University, Warsaw
(2001)
33. Wybraniec-Skardowska, U.: On the formalization of classical categorial grammar. In: Jadacki, J.,
Paśniczek, J. (eds.) The Lvov-Warsaw School—The New Generation. Poznań Studies in the
Philosophy of Sciences and Humanities, vol. 89, pp. 269–288. Rodopi, Amsterdam/New York (2006)
34. Wybraniec-Skardowska, U.: On language adequacy. Stud. Log. Grammar Rhetor. 40 (53), 257–292
(2015)
35. Wybraniec-Skardowska, U.: Categories of first order quantifiers. Bull. Symb. Log. 22(3), 427–428
(2015). doi: 10.1017/bsl.2016.22
36. Wybraniec-Skardowska, U.: Categorial compatibility of the 1st-order quantifiers. Non Class. Log.
Theory Appl. 8, 134–140 (2016)
The Lvov-Warsaw School: A True
Mythology
Jean-Yves Beziau
Dedicated to Jan Zygmunt, my host during my first stay in Poland,

Wrocław, 1992–93
Abstract I discuss various aspects of the Lvov-Warsaw School: its past, present and
future; its location, evolution, mathematics; the variety of its members. I develop this
analysis on the basis of my 25-year experience with Poland.
Keywords Lvov-Warsaw School · Twardowski · Leśniewski · Łukasiewicz · Tarski ·

Polish logic · Metalogic · Methodology of deductive sciences · Consequence ·
Universal logic
Mathematics Subject Classification (2000) Primary 01A72; Secondary 0303, 03A05,

03B22, 03B45; 03B50; 03B53; 03C95
This paper is a mix between personal recollections and logico-philosophical reflections

about the glorious Lvov-Warsaw School. When supervising the edition and production
of the present book in the series Studies in Universal Logic I started to think about my
experience with Poland and I realized how much I had been connected to this tradition
from the very start of my research up to the development of my present and future projects.
So I decided to develop the topic, to look at all the archives I have, and to investigate more
about the Lvov-Warsaw School.1
1Ihave consulted many books and papers, among them: 4–6, 31, 43, 48, 49, 55, 57, 62, 65–67, 75, 77, 89,
92, 93, 95, 99, 102 and others also quoted in the list of references of this paper, which gathers the main
works on the topic.

780 J.-Y. Beziau
1 From Bahía Blanca to Wrocław
1.1 Meeting with Stan Surma at the End of the World
It is like a fair trail: in August 1992 I took a train from Bahía Blanca, Argentina to
Wrocław, Poland. It was an all night train and by no chance I was seated next to Stan
Surma and his son Charles (Fig. 1). We said good bye in Buenos Aires this August 1992
and we were to meet again only in August 2007, 15th years later, for the 2nd World
Congress on Universal Logic (UNILOG) I organized in Xi’an, China. In the mean time I
developed the Universal Logic project, defending a PhD in Paris (July 1995), organizing
the 1st UNILOG in Montreux (March 2005) and launching the journal Logica Universalis
(January 2007), a project rooted in Poland, considering in particular that I coined the
expression “Universal Logic” when in Wrocław in 1992–93.
Stan Surma was the head of the logic group at the Jagiellonian University in Kraków
during the sixties, many of his students became later main logicians in Poland: in
particular Grzegorz Malinowski, Jan Woleński, Jan Zygmunt, Andrzej Wroński, Jerzy
Perzanowski. He founded there in 1965 the journal Prace z Logiki which in 1973 became
Reports on Mathematical Logic. He then escaped communism going first to Black Africa
where he was teaching mathematics in the jungle, then Australia, then New Zealand (In
Xi’an, his son Charles told me this story and we also produced an interview/film with
Fig. 1 JYB with Stan and Charles Surma at the 2nd UNILOG in Xi’an, 2007
The Lvov-Warsaw School: A True Mythology 781
Stan). When in New Zealand he edited The Collected Works of Leśniewski (cf. [50]). A
very important book to maintain alive the Lvov-Warsaw tradition, since Leśniewski is one
of the main figures of this school.
Still in Kraków, Surma “originated an ambitious program of reconstruction of various
results obtained in past in logic but presented in an incomplete or inaccurate way. Surma’s
team was able to offer a complete and rigorous setting for some of Lindenbaum’s, Post’s,
Wajsberg’s and others’ results.” [84, p. 106]. April 27–29, 1973 he organized in Kraków
the XIX edition of the conference of the history of logic with subject the scientific
achievement of Mordchaj Wajsberg (published in the Bulletin of the Section of Logic
the same year, see [69]). Surma also published the collected works of Wajsberg [81], and
later on a paper entitled “The logical work of Mordachj Wajsberg” [71]. Surma wrote also
several papers on Lindenbaum (see [70] and [68]).
But Surma’s work did not restrict to this historical/editorial line of work, he pursued
logical research in the very spirit of the Polish tradition. At the 9th Latin American Sym-
posium on Mathematical Logic which took place August 10–14, 1992 in Bahía Blanca he
gave the talk “Alternatives to the consequence theoretic approach to metalogic” (the corre-
sponding paper was published in the IX SLALM proceedings [72]). This work was much
in the spirit of what I was working on for my PhD at this time. When I created the journal
Logica Universalis, his paper “A Galois connection” was included in the first issue [73].
1.2 The Polish Brazilian Connection
I left Paris in August 1991, where I was doing a PhD in mathematical logic, to work 1 year
with Newton da Costa in São Paulo, Brazil. I was first interested in the work of da Costa
through paraconsistent logic, the topic of my Master in mathematical logic [8]. But when I
came to Brazil we were working on what da Costa calls the Theory of Valuation, a general
theory of logics based on bivaluations, that he started to develop after having provided
non truth-functional bivalent semantics for his systems of paraconsistent logic and other
non-classical logics. This is what I presented at the IXth SLAM in Bahía Blanca.2
In view of extending and generalizing this general theory of logic I started to use the
expression “abstract logic”. There were two reasons do to so: it fits the spirit of modern
mathematics, a general theory of logical systems would be a theory of logical structures,
and a good name for a logical structure would be “abstract logic” in the same way as
“abstract algebra”. The second point is more philosophical, emphasizing abstraction of
the objects. From this perspective my project was at this time to have a PhD with title
From Formal Logic to Abstract Logic.
Then, I discovered the work of Suszko at the library of the Department of Mathematics
of the University of São Paulo. Suszko had himself used the expression “abstract logic”.3
I started to seriously think of going to Poland to know more about all this. I talked with da
2A systematic paper on that was published in Logique et Analyse [32] and a more philosophical note
[33], related to the talk I presented in Bahía Blanca was published in the same proceedings as Surma’s
paper; see also Loparic and da Costa’s paper reprinted in the anthology of universal logic [21] with my
comments.
3 See about this, one of Suszko’s papers reprinted in the anthology of universal logic [21], with comments
by Ramon Jansana and my paper “La logique abstraite au sein de la mathématique moderne” [9].
782 J.-Y. Beziau
Costa and he encouraged me to do so. He had himself developed strong links with Poland,
having been there several times in the 1960s and 1970s and inviting Polish logicians to
come to Brazil: Lech Dubikajtis, Jerzy Kotas, Grzegorz Malinowski, Jerzy Perzanowski.
And Ryszard Wójcicki, who wrote his book Lectures on propositional calculi [82] there:
Although the work on this book was concluded only after my return to Poland, the substantial
part of it have been done during my stay in Brazil sponsored by FAPESP (Fundação de Amparo
a Pesquisa do Estado de Sao Paulo, Brazil, grant no 80/1188–8). I benefited a great deal and in
various ways from the opportunity to have scientific context with my Brazilian colleagues and
friends. My greatest debt has been to Prof. Ayda I. Arruda, at that time the Director of Instituto de
Matemática e Estatística e Ciência de Computacão, Universidade Estadual de Campinas both for
the care she took for creating me excellent conditions for work and for her keen and penetrating
interest in the ideas I discussed in my lectures. Also I own a special debt to Prof. Newton C. A.
da Costa for his invitation to Instituto de Matemática e Estatística, Universidade de São Paulo and
stimulating discussions we held, and to Prof. Elias Alves for his introducing me to people from
the Logical Center of UNICAMP and his assisting me on many occasions.
This book is a overall presentation and synthesis of the general approach to logical
systems mainly develop by Polish logicians, which is sometimes qualified as Polish Logic.
Wójcicki later wrote a second version of this book [83], with more details, which can be
considered as the Bible of Polish logic, internationally published and promoted by Kluwer
(now Springer). I myself prefer the style of the first edition, published by Ossolineum in
Poland, which is nowadays difficult to find.
Newton da Costa himself was born in Curitiba, South of Brazil, which is the second
most important Polish city in the world outside of Poland, after Chicago. And Brazil
also has the second largest community of Polish immigrants after the USA, nearly
two millions. The Polish colony was started by Sebastian Woś (1844–1933) from the
city of Siolkowice (region of Opole). Then, between the two world wars, there was
an important second wave of Polish immigration to Brazil, in particular many Jews.
Famous among them is Leopoldo Nachbin (1922–1993), considered as the most important
Brazilian mathematician, funding member of the IMPA (Institute for Pure and Applied
Mathematics) in Rio de Janeiro, whose PhD was directed by Laurent Schwartz.
1.3 Polish Surroundings
Besides the logical attraction to Poland I was also interested to go to Poland for various
reasons. When I was a student in Paris in the 1980s, two Polish artists were very popular:
Roman Polanski and Witold Gombrowicz. I saw many movies by Polanski in particular
Repulsion (1965) with Catehrine Deneuve and Le Locataire (1976) with Isabelle Adjani.
I also saw him in a theatre on a one man show, interpreting The Metamorphosis by Franz
Kafka. From Gombrowicz I read in that order three books: Cosmos (1965), Ferdydurke
(1937), Possessed (1939) and was quite impressed. I saw the movie adaptation of
Ferdydurke directed by Jerzy Skolimowski, a quite good adaptation (never easy to make a
good movie about a literature masterpiece). Skolimowski is a famous Polish filmaker who
has collaborated with Polanski in the seminal Knife in the water (1962), both being from
the famous Łódź’s film school. I also saw Skolimowski’s very good movie, Torrents of
Spring (1989), adaptation of a novel by Turgenev, with Nastassia Kinski, the daughter of
the legendary actor Klaus Kinksi, also of Polish origin. There are several connections
between Polish logicians and movies. Lindenbaum’s father was a film producer (see
[104]) and Roman Suszko appeared in a mythical post-war Polish movie A trip down
the river (1970) directed by Marek Piwowski (original title in Polish: Rejs).
Gombrowicz was exiled in Argentina and became famous when the French publisher
Maurice Nadeau (1911–2013) got interested in his work. René Goscinny (1926–1977) is
another Polish guy connecting Poland, Argentina and France. Of Polish Jewish origin,
born in France, he spent his youth in Buenos Aires, moved back to France after WWII
and became a cultural icon through comic books: Astérix of course, but also Lucky Luke
and Iznoud (my favorite one). With Astérix, Goscinny promoted the Gauls as a founding
cultural myth of French identity. It became a symbol of French culture in the same way
as Mickey Mouse is a symbol of American culture. This myth is a mix of many things
in particular figures imported from Argentina (cf. Patoruzú). Goscinny was really good at
puns, in particular multilinguistic puns, therefore intraduisible. . . The name of his main
hero Astérix has a double or triple meaning, one of them being king of the stars. The
French mathematical society created in 1973 a journal called Astérisque (Fig. 2), punny
name referring to this symbolic figure in the continuity of the Bourbachic funny naming
tradition, rich of meaning and quite spiritual in the French way (“avoir de l’esprit”)—
one of a few foreign members of Bourbaki was Samuel Eilenberg (co funder of category
theory with MacLane), a Polish Jew.
At the Sorbonne I was a student of Sarah Kofman, also from Polish Jewish origin.
Her father was a Rabbi who was deported during WWII and was beaten to death in
Auschwitz, because he didn’t want to work during the Sabbath, by a Jewish kapo, who
later on became a successful merchant in Paris. She wrote a very nice book about Plato:
Comment s’en sortir, literally: How to get out? This is a book about the notion of “aporia”
(Greek etymology: without a path). She pointed out that in the dialogues, when facing an
“aporia”, a deadlock, by using rational thinking, dialectics, then Plato uses the myth to
go out, telling some stories. She showed very well that at the heart of Plato’s philosophy
there is a constant mix between logos and muthos, a true mytho-logy. Unfortunately, she
thought, the mythical dimension was washed away in Aristotle’s philosophy. She was one
of my favorite teachers and I did a Master thesis with her on Plato’s cave [7].4
On my family side, one of my mother’s uncles emigrated from Marocco to Argentina in
the 1950s and his three daughters all married Polish emigrants there. Moreover the sister
of my mother married a guy of Polish origin, Jean Dybowsky, descendant of the famous
Fig. 2 Astérix and mathematics
4 She later on committed suicide and I dedicated to her a paper I wrote on suicide [11]. When in Poland I
visited Auschwitz and this was a terrific experience.
784 J.-Y. Beziau
African explorer bearing the same name, giving birth to three neo-Polish cousins. Finally,
my father told me several times that my mother (of Swiss citizenship) had some Polish
Jewish ascendancy. I don’t know if it is true and I never inquired to check this information,
because on the one hand if it is true, it is a logical truth, a tautology, since good logicians
are all Polish Jews, on the other hand I don’t feel the necessity to identify myself with
a specific community, subculture, or abelian subgroup. To simply be a rational animal is
already quite complicated!
Anyway, for all these reasons, I had some sympathy for Poland. And after going from
the 1st World (France) to the 3rd World (Brazil), I was curious to discover the 2nd World,
which was a mystery for me. I remember that during an exchange in high school in France
I spent 2 weeks in Bayreuth in 1979 and one of the attractions was to go to watch the iron
curtain (Czech border). On the other side it looked really like another world. I knew very
few things about Poland. In the train from Bahía Blanca to Buenos Aires Stan Surma draw
me a logic map of Poland. As the erudite reader may know it is not possible to go all the
way by train from Argentina to Poland. I took a bus from Buenos Aires to São Paulo, then
a plane to Paris and then again a train to Poland. I stopped a few days in Brazil to meet
again Newton da Costa and in France to visit my family.
I was going to Poland within an exchange program between France and Poland. The
French Ministry of Foreign Affairs sent me a train ticket and I took a train from Paris to
Warsaw on October 1st, 1992 (Fig. 3). It was the train Paris-Moscow, a 48 h trip, Warsaw
being at the middle of the way, a 24 h trip. I was in a compartment with a friendly Polish
professor. Arriving in Warsaw there was nobody from the French Embassy waiting for me
as promised (Fig. 4). I didn’t know where to go. I had lots of baggages, few money and
didn’t speak Polish. Delikatnej sytuacji . . .
Fig. 3 JYB and grandma Alice, Oct 1 1992, France: departure to Poland
Fig. 4 Logical arrival in Warsaw
I met by chance on the platform a French girl who was in the same situation as me and
the Polish professor helped us to call the French Embassy. It was a Friday, late afternoon.
They told us they right now could not help us, but that we could use the diplomatic visitor
flat at the Embassy and that we will talk on Monday. The flat was OK but with only one
bed so we had to sleep on the same bed. And during the weekend I walked around Warsaw
with this girl, I think her name was “Sylvie”. Monday morning we had a meeting with
the staff at the French embassy and they put us on a train to Wrocław. For this girl this
was normal since she was supposed to study at the Art School there. In my case this was
quite a surprise because I was prepared to go to Łódź. When in Brazil I had written to
Wójcicki, he didn’t reply to me, but after some months I received a letter from Grzegorz
Malinowski inviting me to join his group in Łódź (Fig. 5).
786 J.-Y. Beziau
Fig. 5 Malinoswki’s letter inviting me to join his logic group in Łódź

But for some reasons the French Ministry of Foreign Affairs organized my coming to
Worcław. That’s how I arrived in this town I knew nothing about. On the platform was
waiting for me Jan Zygmunt who drove me in his little Fiat to a University House in Plac
Grunwaldzki (cf.[39]).
2 The Atopicity of the Lvov-Warsaw School
The expression “Lvov-Warsaw School” has become canonical. It is rather a proper name
than a definite description. But, like many proper names, it is also improper. It can
reasonably serve as an identification device, we roughly know what we are talking about,
like when we use names such as “Poland”, “Truth” or “Alfred Tarski”. But this does
not mean we know exactly what it is. Identification does not confer identity. Maybe
Poland and Truth do not have a proper identity. The identity of Alfred Tarski is more
palpable, tangible, at least imaginable, paintable, photographical, not (yet!) completely
mythical. The consistency of a phenomenon can be tested by trying to answer the five
basic questions: when, where, what, why, how? Let’s see if LWS passes the test. . .
I will use “LWS” as an acronym/abbreviation for the Lvov-Warsaw School. Sometimes
people use “LWs” or “LW-s”. I prefer to capitalize the School, as they do in Vienna with
the Circle. Note also that in the canonical expression “Lvov-Warsaw School”, “Lvov” has
become a standard English spelling for the name of this city, but the tendency is nowadays
to write “Lviv”.
2.1 A School Without a Location
I progressively discovered that I arrived not only in Wrocław but also in Lvov and
Breslau. Before WWII the city of Wrocław was named “Breslau” and was in Germany.
The Friedriech Wilhelm University of Breslau was an important University frequented
by famous people like the Nobel prizes Erwin Schrödinger, Max Bohr or Friedrich
Bergius. The book by Frege Die Grundlagen der Arithmetik, eine logisch-mathematische
Untersuchung über den Begriff der Zahl was published in Breslau in 1884 by W. Koebner.
After WWII “Breslau” became “Wrocław” and the surrounding region became part of
Poland. On the other hand the city of Lvov and the surrounding region, which was part
of Poland before WWII, became part of Soviet Union. Poland was displaced on the West
(Fig. 6). This displacement was not purely physical, but also humanistical. Poland, like
many countries, has a variable geometry. . .
Wrocław is a sort of transposition of Lvov. Polish people from Lvov moved to
Wrocław. German people of Breslau fled or were expelled “home”, i.e. in what would
be called “DDR”, Deutsche Demokratische Republik, now out of the map, like Prussia,
Atlantis or Gaul.
Wrocław University became a major university of Poland after WWII, Wrocław a
major student town, and also one of the most prosperous Polish towns, in particular due
to his President (or Mayor) Rafał Dutkiewicz, who did study logic at Wrocław University
and that I shortly met during my stay—he was not yet famous (Fig. 7).
788 J.-Y. Beziau
Fig. 6 After WWII Poland goes West—Lvov moves to Wrocław
Fig. 7 Rafał Dutkiewicz logically successful mayor of Wrocław
A symbol of the transposition of Lvov to Wrocław is the Ossolineum foundation. This

foundation was created by the Count Józef Maksymilian Ossoliński (1748–1826) in Lvov.
I used to have lunch at the canteen of the Ossolineum foundation which was one block
from the Department of Logic in the same street, Szewska street. Ossolineum was during
the communist time the main publisher of the Polish Academy of Science, publishing
many books of logic and the journal Studia Logica. When I arrived there everything was
changing. Wójcicki made a deal in 1992 with Kluwer (which later on was incorporated
into Springer) and Jan Zygmunt was the editor-in-chief of Studia Logica. I went to the
Ossolineum office in Wrocław asking if it was possible to buy old issues of the journal. I
succeeded to buy a good number of issues which were in the cellar for a price as low as 2
or 3 cents each.
Wrocław became after WWII one of the most important logic centers in Poland.
Wójcicki describes the situation as follows:
Wrocław was another important center of postwar logic. The most prominent figure in foundations
of mathematics has been Czeslaw Ryll-Nardzewski. He grouped around himself several very
talented young peoples (L. Pacholski, B. Wȩglorz, A. Wojciechowska, and others) but that group
was much more oriented towards foundational than purely logical problems. Also Jerzy Łoś,
known for a number of outstanding theoretical results, notably his famous ultraproduct theorem,
started his academic career in Wrocław. Another eminent Wrocław logician was Jerzy Słupecki.
His most prominent collaborators were Ludwik Borkowski, Witold A. Pogorzelski and Boguslaw
Iwanuś. . . . my own academic biography starts in Wrocław too. [84, pp. 503–504].
Considering all these aspects we can maybe talk about a Lvov-Warsaw-Wrocław

school of logic. But on the one hand Wrocław has not been the only and main center
of logic after WWII in Poland and on the other hand there was a serious break in the
development of logic in Poland after WWII, in particular with the emigration of Alfred
Tarski to the USA where he founded a center for logic in Berkeley (California) who
became during 25 years the most important logic center in the world. Considering this
move, one could also talk of a Lvov-Warsaw-Berkeley school of logic. But this also
would be controversial because few people of the “original” school followed Tarski in
California—one of them was the young Kalicki who tragically died there in a car accident
(see [101]).
Słupecki tried without success to develop a logic school in Wrocław, as explained by
Woleński and Zygmunt:
Since the beginning of his activities in Wrocław, Slupecki supervised graduate students and
doctors, evaluated doctoral and habilitation theses and influenced mature logicians. This circle was
formed by Edward Baluka, Ludwik Borkowski, Edmund Glibowski, Boguslaw Iwanuś, Tadeusz
Kubinski, Jerzy Nowak, Witold A. Pogorzelski, Juliusz Reichbach and others, but it was not a
school as Slupecki would have envisaged it, and such a school that could not develop in Wrocław
mathematical circles, of which he was fully aware. [98]
My host in Wrocław, Jan Zygmunt, was the director for the Department of Logic,
which full and official name was Katedra Logiki i Metodology Nauk. It literally means
Chair of Logic and Methodology of Science. It was part of the Institute of Philosophy
located in a building which was very close to the main building of the university where
there was the beautiful “Aula Leopolodina” and a nice cafeteria to which Jan Zygmunt
used to frequently invite me to chat having a tea.
At the department I had a big office room. On the wall it was written: “Alfred Tarski
and Dana Scott were there in 1955”. I don’t know the exact impact of this on my mind,
but I spent lots of time reading all the papers of Dana Scott on rules and general logic and
wrote in January 1993 the paper “Rules, derived rules, permissible rules and the various
types of systems of deduction” [12]. Later on I decided to include one of these papers
by Scott, “Complenetess and axiomatizibility in many-valued logic”, in the anthology
of universal logic and asked Lloyd Humberstone, one of his former students to write the
790 J.-Y. Beziau
presentation of it (see [21]). This room was the one of Tomasz Skura, who at this time was
in an exchange program in Konstanz, Germany invited by Andre Fuhrmann (who later on
moved to Brazil and became my friend). Other permanent members of the department
were Tomasz Furmanowski and Jacek Hawranek. I also met at this time the East-German
logician Max Urchs who was partially employed at the department. The present director
of the department, Marek Magdziak, was at this time doing his PhD there. There was
a regular seminar to which I presented two talks (as recalled by Zygmunt [39]) and to
which was coming Juliusz Reichbach, a retired logician, who had done some important
work on completeness in the 1950s (cf. [60, 61]) and after that emigrated to Israel but
didn’t succeed to adapt and came back to Poland.
Reichbach was living in the same building as me on Plac Grunwaldzki. It was a
Stalinist building called “the house of science” at the corner of Plac Grunwaldzki and
Curie-Skłolodowskiej street. I used to bump into him on the nearby supermarket where
he was buying vodka. One day I invited him to have a tea in my flat but when he saw my
paintings on the wall (Fig. 8) he was quite afraid and decided not to enter. I had made only
few painting when I was a child in Corsica and then in high school. For some reasons I
started to paint quite a lot when I was living in Wrocław. In some sense this was quite
natural if we think of someone like Leon Chwistek who was both a famous logician and
painter.
I remember that I bought some tubes of paint on the Russian market just below my flat.
At this time the (ex) Soviet Army was leaving the country after many years of occupation
selling all they had, from watches to planes. After some years the department of logic was
Fig. 8 One of the paintings I did when in Wrocław

Fig. 9 Back to Wrocław in 1998: JYB and Valdimir Vasyukov
moved from the town center to the ex-Soviet Military based. Visiting Wrocław again in
1998 I had the opportunity to see the place. At this time the Russian logician Vladimir
Vasyukov was also there. He was a good friend of Jan Zygmunt and also became a good
friend of mine (he invited me to Moscow in 2001 and I went again to Russia in 2003, 2009,
2012, 2016). The father of Vasyukov had been a military officer in activity in Wrocław
and Vladimir spent some years of his life there when he was a child and spoke good Polish
(Fig. 9).
1992–93 was a great time in Wrocław, it was a transition period. After the liberation of
Poland people were optimistic, new bars and restaurants were opening every day. There
were lots of foreigners: business men coming to invest, visitors at the university and
tourists. At the same time the way of life and style of life of the communist time was
still very present. Moreover communism, in particular due to isolation, had frozen things,
so it was like a travel in time, back to the 1960s or 1950s and even the 1930s.
Plac Grunwaldzki was used by the Germans at the end of the war as an airfield:
most of the buildings were demolished for that and furthermore the region was heavily
bombarded by the Russians. After the war it was developed in a “modern” way—after my
departure it was further modernized with a big shopping center (Fig. 10). The department
of mathematics was located at the entrance of Plac Grunwaldzki, near the bridge. I
used to go there at the library, where there was a mix of books from the University of
Breslau and the University of Lvov. There were some logicians there, like Wȩglorz, but I
hardly met/knew him. This was in fact symptomatic of the separation between logic and
mathematics in Poland which, according to the legend, goes back to a dispute between
Sierpiński and Leśniewski. Also, though the department of logic was in the institute of
792 J.-Y. Beziau
Fig. 10 Evolution of Plac Grunwaldzki
philosophy, we nearly had no contact with the other parts of the institute, still under heavy
influence of Marxism.
During my first stay in Poland I visited also some other towns, in particular Łódź
and Kraków which were two other important logic centers in Poland at this time, with
important departments of logic. This was not the case of Warsaw. I went there only a few
times to solve some questions with the French Embassy.
2.2 A School Without a Topic
How can we characterize the topic of the Lvov-Warsaw School? The Vienna Circle (VC)
is strongly associated with “Logical empiricism” and “Logical positivism”. It would
be difficult to find an expression that could systematically be associated with the name
“Lvov-Warsaw School”.
There is a famous paper by Ajdukiewicz entitled “Logistic anti-irrationalism” [3]. So
one may want to put the equation:
LWS = Logical Anti-Irrationalism
that would be parallel to the equation:
VC = Logical Positivism
But the expression “anti-irrationalism” is too much idiosyncratic and its meaning is
not completely clear.5 It is a bit confuse for a school wanting to promote conceptual
clarification. And we may wonder why being so negative, with a double negation. Why
not being purely positive? But then we would have something like “Logical Rationalism”
which is rather pleonastic.
There is another expression, also used by Ajdukiewicz: “Methodology of deductive
sciences”. This expression is the title of his habilitation thesis defended in 1921 in Lvov
[1], and was then systematically used and promoted by Alfred Tarski, in Polish, German,
English and French, in one way or another:
• “Fundamentale Begriffe der Methodologie der deduktiven Wissenschaften. I.” (paper
published in 1930)
• O Logice Matematycznej i Metodzie Dedukcyjnej (book published in Lvov-Warsaw
1936 [76])
• Sur la méthode deductive (talk presented at the 9th International congress of philoso-
phy in Paris in 1937, cf. [78])
• Introduction to logic and the methodology of deductive sciences (book published in
Oxford in 1941. English version of [76])
The interesting thing about Methodology of Deductive Sciences, hereafter MDS, is
that it keeps a philosophical perspective which disappears in the case of expressions like
“Metalogic” or “Polish logic”. The word Metalogic was mainly promoted in Poland, in
particular it was used by Wajsberg. It is a very important word that rightly characterizes
the Polish perspective: on the one hand the distinction between two levels, on the other
hand the generally beyond it; it is not only the study of mathematical reasoning, but
the deductive reasoning of any science. Tarski was also much interested in biology and
physics. It is of course inspired by the Metamathematics of Hilbert and there is a direct
connection since Ajdukiewicz visited Göttingen. However Polish logicians went much
higher at the metalevel, because firstly, as it is well known, and contrarily to Hilbert, they
were promoting no limitation at the metalevel, and secondly they had more imagination
which led to the idea of arithmetization of syntax, crucial to Gödel’s theorem. This was
mainly promoted by Łukasiewicz in the early twenties as recalled by Tarski who explained
that to Gödel during a visit in Vienna (cf. [80]).
5 See our recent paper “Is logical relativity irrational?” in a special issue of Studia Metodologiczne
dedicated to Ajdukiewicz [25].

794 J.-Y. Beziau
Before WWII were already created some Departments of Logic and Methodology
of Science in Poland and this expanded after WWII. Nowadays there are many such
departments in Poland, most of them in philosophy institutes, but not always. Can we
consider that
LWS = MDS
is a key equation explaining everything? This would be the easiest solution, but a
simplification. First it important to notice that between the two wars, LWS was not a
tiny part of the university but the dominating part. Łukasiewicz was twice the rector of
the University of Warsaw. Good also to remember that Twardowski was rector of the
University of Lvov and that Ajdukiewicz, who married his daughter after WWII, became
the rector of the university of Poznań (Jaśkowski was also a rector after WWII: rector of
Nicolas Copernicus University in Toruń).
Few years ago the University of Warsaw created a symbolic monument with four
statues: those of Twardowski, Łukasiewicz, Tarski and Leśniewski. This means that these
four men are considered as the most important figures of the history of the University of
Warsaw (Fig. 11).
On the one hand we could say that at some point MDS was dominating the University
of Warsaw and that it is not anymore the case. Or on the other hand we could say that
LWS was much broader than MDS, which is true in some sense if we consider that LWS
was an nebulous group of philosophers, mathematicians, linguists and artists.
But looking at the Fig. 11 we see another paradox: if we put aside Twardowski, the
bridge between Lvov and Warsaw and considered as the founding father of the LWS (the
figure of a father is typically mythical, before and after Freud), the three other men are
mainly known as logicians. So why not putting the following equation
LWS = Polish Logic?
Here is what Jan Zygmunt wrote about this expression:

The term “Polish logic” was coined by McCall to signal the important contributions to modern
logic by logicians from Poland between the wars. There were several centres of research, of which
Fig. 11 Monument at Warsaw University with statues of Twardowski, Łukasiewicz, Tarski and
Leśniewski
the Warsaw school, which grew out of the earlier Lvov-Warsaw philosophical movement, was the
most significant. Its development was closely connected with the Warsaw school of mathematics,
which gave it its characteristic mathematical bent. . .
The main centres of logic research between these dates were Cracow, Lwów, Poznań, Wilno and,
most importantly, Warsaw. The Warsaw school of logic, founded by Łukasiewicz and Leśniewski,
began as an intellectual offshoot of the Lvov-Warsaw philosophical movement, but quickly
eclipsed it in both quantity and quality of research. The development of the Warsaw school of
logic was closely connected with that of the Warsaw school of mathematics. In 1918 one could
already speak of Warsaw as a fairly strong centre of research in set theory and topology under
the direction of Janiszewski, Mazurkiewicz and Sierpiński. The two schools shared organizational
structures, swapped directors and collaborated on many academic initiatives. This accelerated the
development of both schools and deepened their research in key areas. In later years Kuratowski,
Lindenbaum, Tarski and finally Mostowski made significant contributions to both schools. [103,
p. 6634]
But on the other side there were also philosophers, as described by Wójcicki;
One may also point out that an interwar (1920–1939) formation widely known as the “Polish
School of Logic” overlapped in a very substantial way with another formation known as the
“Lwow-Warsaw School of Philosophy”. The members of the latter not only considered logic to
be the main tool of philosophical analyses but they often contributed to logical investigations by
themselves. [84, p. 498].
So at the end we can have the following picture: a group of three men—Lesńiewski,
Łukasiewicz and Tarski—mainly logicians with on one side some mathematicians, on
the other side some philosophers. In the French version of Wikipedia, the following is
said about Leśniewski : “With Jan Łukasiewicz and Alfred Tarski, who was his only
doctor, he formed a troika which during the decades 1920 and 1930 made the University
of Warsaw one of the most important research centers of mathematical logic in the world.”
This troika scheme is confirmed by the entry on Łukasiewicz in the English version
of Wikipedia: “He remained a professor at the University of Warsaw from 1920 until
1939 when the family house was destroyed by German bombs and the university was
closed under German occupation. He had been a rector of the university twice. In this
period Łukasiewicz and Stanisław Leśniewski founded the Lvov-Warsaw school of logic
which was later made internationally famous by Alfred Tarski who had been Leśniewski’s
student.” And Woleński provided a general picture as follows:
Alfred Tarski (1901–1983), later recognized as one of the greatest logicians of all times, decided
to specialize in logic (he graduated and obtained his PhD under Leśniewski) and became the third
pillar of the Warsaw School of Logic (WSL for brevity). In the 1920s and early 1930s these
Big Three were joined by (I list them in alphabetical order) Stanislaw Jaśkowski (1906–1965),
Adolf Lindenbaum (1904–1941?) (the question marks in the dates indicate that exact data are
uncertain or even unknown), Andrzej Mostowski (1913-1975), Mojżesz Presburger (1904–1943?),
Jerzy Słupecki (1904–1987), Boleslaw Sobociński (1906–1980) and Mordechaj Wajsberg (1902–
1943?). All of them except for Sobociński were mathematicians by training. . . . The WSL as a
working group had eleven members at its peak, that is, around 1937. Is this large or small? Of
course, everything depends on the point of reference. Evaluating from the contemporary point of
view, about a dozen people working together in logic is perhaps not so many. However, if one
looks at this group from a broader international perspective, one should remember that no other
place in the world in which logic was actively done had even one third of this amount. Thus, at the
time Warsaw was the place most populated by professional logicians in the world. . . . The Warsaw
logical community was much larger than the WSL sensu stricto. Some mathematicians, already
mentioned above, like Kuratowski, should be included. [94, p. 35]
796 J.-Y. Beziau
Woleński also note that: “According to Łukasiewicz, logic is an autonomous subject

which is subordinated neither to philosophy nor to mathematics. On this view, logic is no
servant of any other science.” [91, p. 377] So it makes sense to consider that logic is the
heart of the LWS even if we don’t use the expression “Lvov-Warsaw School of Logic” to
make clear that LWS was not reduced to logic or to at least logic reducely considered.
Polish logic can be understood as literally logic made in Poland. But can we way
that there is (or there was) a special way of doing logic in Poland? One may think of
some peculiarity such as Polish notation, the search for the unique axiom or the focus
on propositional logic. But, at the middle of all this, there is the theory of consequence
developed by Tarski, as emphasized by Jan Zygmunt:
A metamathematical theme was present in Polish logic from the early 1920s, beginning with
Ajdukiewicz (1921). Tarski’s efforts allowed investigations of the sentential calculi to be carried
out within an explicitly metamathematical framework. Tarski generalized this framework to a
mathematical theory of two primitive concepts (sentence and consequence) which he called ‘the
methodology of the deductive sciences’. Within this theory he was able to provide a conceptual
apparatus for investigating deductive systems. (Lindenbaum contributed much to this work,
including the widely known Lindenbaum maximality lemma.) [103, p. 6636]
and Ryszard Wójcicki:

Note that already in the early thirties, Tarski, at that time a young docent at Warsaw University,
formed his theory of deductive systems. It was intended to be the most general theory of
logical reasoning. The central notion of Tarski’s theory is a consequence operation, thus an
operation which applied to any set of premises yields the set of all the conclusions derivable
from (or, under semantic interpretation of the consequence operation entailed by) those premises.
Since, depending on one’s logical preferences, logical validity (and hence both derivability
and entailment) can be understood in different ways, Tarski focused his attention on those
characteristics of the consequence operation which are by and large independent of our logical
preferences. Curiously enough the theory of deductive systems, known also as the theory of the
consequence operation, has attracted little attention outside of Tarski’s native country. In this
way the theory of the consequence operation became a Polish specialty, not to say idiosyncrasy,
often treated as some strange, not to say redundant, province of logical investigations. Only now,
after nonmonotonic logic has been invented, the significance of Tarski’s idea of studying the
consequence operation in an abstract way, independent of any rules of inference derived from
a specific system of logic, has become obvious. [84, p. 498].
This theory can be called Polish logic because it was developed mainly in Poland
and it is not well-known outside of Poland. Tarski himself when he moved to California
developed model theory because he had the idea that it was a more general framework and
didn’t work anymore directly or explicitly on the theory of consequence operator. Results
about this theory can be found in the two books of Wójcicki [82, 83] and on the book
Completeness theory for propositional logics by Pogorzelski and Wojtylak, originally
published in 1982 by the Silesian University of Katowice, that we re-edited in 2008 in the
book series Studies in Universal Logic [56]. A seminal work in this line is the monograph
by Łoś, originally written in Polish [51], translated in English by Robert Purdy, but not
yet published.
2.3 A School Which Is Not a School
The Aula Leopolidna of the University of Wrocław is nowadays considered as the baroque
jewel of Poland. “Baroque” is a name of an artistic movement 1590–1725. The name
was given only afterwards by the Swiss historian Jacob Burckardt (1897–1897) in his
book Der Cicerone Eine Anleitung zum Genu der Kunstwerke Italiens published in 1855
(see [29]). The people who had developed the baroque style as a reaction to protestant’s
sobriety, didn’t identify as a movement and didn’t self-baptize them as “baroque”.
Burckardt created the Baroque movement, he did not create the word. Jean-Jacques
Rousseau notably used this word in his Dictionary of Music:
Baroque : Une musique baroque est celle dont l’harmonie est confuse, chargée de modulations et
dissonances, le chant dur et peu naturel, l’intonation difficile, et le mouvement contraint. Il y a bien
de l’apparence que ce terme vient du baroco des logiciens. (A Baroque, or rough music, is that,
whose harmony is confused, filled with modulations and dissonances, its notes hard and unnatural,
the intonation difficult, and the movement constrained. It appears evidently that this term must be
derived from the Baroco of the logicians, William Waring’s translation). [63]
In syllogistic, Baroco is the name of the fourth mode of the second figure, an example
of which would be:
All good logicians are Polish
Some humans are not Polish
Therefore some humans are not good logicians
Strange indeed! Maybe the Lvov-Warsaw School, by its heterogeneity and diversity, can
be considered as baroque. Considering self-consciousness and self-baptism, the situation
is not so baroque because LWS presented itself as such: at the event organized in 1935 in
Paris by Louis Rougier, Ajdukiewicz, in his introductory speech used this expression:
Presque tous les membres polonais de notre Congrès sont disciples de l’école connue sous le nom
d’école de Lwôw et de Varsovie. (Most of the Polish members of our congress are disciples of a
school known as school of Lwôw and Warsaw.) [2]
It was later on used again for example by Zbigniev Jordan [47], but the full
canonization is certainly due to the works and activities of Jan Woleński (in particular:
[86–88]).
With LWS we don’t have something as explicit as with the Bourbaki’s group, the
Surrealist movement or the Vienne Circle. In the two last cases, besides the self-baptism,
there were also manifestos. Chapman wrote the following about the Vienna circle:
Philosophers are not generally known for being team players. Philosophical ideas may sometimes
be attributed jointly to two or more thinkers, but it often turns out that these people disagreed on
fundamental issues, or that they worked in separate countries or even separate centuries. . . The
Vienna Circle was unusual. It consisted of a large but identifiable group of philosophers who met
regularly, collaborated on work of mutual interest, and largely agreed on their conclusions. They
even acknowledged their group identity by coining the term The Vienna Circle, or Der Wiener
Kreis, in the title of a collaborative manifesto. [30, p. 7]
Let us compare this with what Surma wrote about LWS:

A distinctive feature of the school and one of the secrets of its success was the spirit of teamwork.
The mutual collaboration among the members was so close and intimate that it is often hard to
decide who should be credited with which particular results. [71, p. 102]
798 J.-Y. Beziau
But it is more connected with the school of Polish logic. There is a joke, pushing
to the extreme this empathy, claiming that all important results proved by Tarski at this
time are due to Lindenbaum. What is interesting is that this group of people were able
to work together without sharing the same views. Lindenbaum was strongly communist
(see [96, 104]), in a rather fanatic and irrational way, this was not the case of Tarski or
Łukasiewicz.
Woleński wrote that: “Lindenbaum, Presburger, Tarski and Wajsberg were Jews, but
Leśniewski and Sobociński (later also Łukasiewicz) were strongly antisemitic.” [94, p. 38]
As it is known Tarski changed his name and did as much as possible to hidden his Jewish
identity to escape persecution but was inclined to consider Jews has a superior race (see
[38]).
In Wikipedia, we find the following description:
A school of thought (or intellectual tradition) is a collection or group of people who share
common characteristics of opinion or outlook of a philosophy, discipline, belief, social movement,
economics, cultural movement, or art movement. . .
Schools are often named after their founders such as the “Rinzai school” of Zen named after Linji
Yixuan and the Asharite school of early Muslim philosophy named after Abu l’Hasan al-Ashari.
They are often also named after their places of origin, such as the Ionian School of philosophy that
originated in Ionia and the Chicago school of architecture that originated in Chicago, Illinois and
the Prague School of linguistics, named after a linguistic circle found in Prague, or Tartu-Moscow
Semiotic School whose representatives lived in Tartu and Moscow. (Wikipedia, School of thought)
The name “Lvov-Warsaw School” is related to a name of a (bi)location, not a person.

Although, according to the myth Twardowski is the father of LWS, nobody would use the
expression “Twardowski School” instead of or as a synonym to LWS. When the name of
a school is attached to a location, this means that some people gathered in a place, but
shared something which is not limited to the place. In the case of LWS it is clear that a
group of people gathered in Warsaw (Lvov in the shadow) and shared something which
was part of a general rational movement but whose specific identity is connected to logic,
however difficult to characterize.
On the other hand the situation of the group that Tarski developed in California is much
simpler. It can be qualified in various ways: Tarski’s school, Berkeley’s school of model
theory, etc. There is a main figure and a main topic. But funny enough, since 1957, it is
called Group in Logic and the Methodology of Science. . .
3 The Future of the Lvov-Warsaw School
3.1 No Future
One may wonder what the future of the Lvov-Warsaw school is. A simple and direct reply
would be: No Future. We can say that this school ended at the time of the second world
war with:
• the death of Leśniewski (natural death few months before the war),
• the departure of Tarski to United States (just before the war),
• the destruction of Warsaw by the Nazi Germany,
• the invasion of Lvov by the Soviets,

• the departure of Łukasiewicz to Ireland (after the war).
These are five main central symbolic points with many corollaries and side-effects:
The killing of Lindenbaum, Wajsberg and Presburger. The suicide of Witkiewicz. The
departure of Bocheński, Sobociński, Kalicki, etc.
WWII officially started with the invasion of Poland by the Nazi Germany on September
1st 1939. On the other side Poland was invaded by the Soviets on September 17, 1939.
During WWII Poland became a German Nazi death camp. Poland was chosen by the
Nazis to develop the main extermination camps, the most famous being Auschwitz-
Birkenau. The population of Poland before WWII was about 35 millions of people, after,
less than 24. It is the country with the highest percentage of its citizens that died during
WWII. It is difficult to imagine the survival of any school under such circumstances.
Moreover, after the war, Poland was dominated by the Soviet Union during about
40 years. Poland became free again only with the third republic in 1989, result of the
Solidarność movement which played an important role in the final collapse of the Soviet
Union in 1991. When I arrived in Wrocław in October 1992 the Soviet Army was
still there. Things have been changing slowly: the 1st May of 2004 Poland entered the
European Union. I was in Wrocław at this time, after taking part to a meeting of logic in
Karpacz, and there was a great party on the main square. But 13 years after, for better or
for worse, Poland is still not part of the eurozone.
The Soviet built the impressive tower dominating the city of Warsaw named Joseph
Stalin Palace of Culture and Science (Fig. 12), nicknamed the 8th sister because of its
similarity with the seven sister Stalinist skyscrapers in Moscow, being the second highest
one after the main building of Moscow State University. In this tower there was in
Fig. 12 Stalin power in Poland: the palace of culture and science

800 J.-Y. Beziau
particular the department of mathematics of the University of Warsaw, where Andrzej

Skowron, as he recently told me, attended a talk by Newton da Costa in the 1970s.
The result of the Soviet domination over the development of logic in Poland after
WWII was on the one hand isolation from Western Europe and United States, on the other
hand the domination of the Russian school. The best Polish students were sent to Moscow,
this was the case of Surma who went to study with Kolgomorov. Philosophy in the Soviet
sphere was strongly ideologically oriented towards Marxism. This certainly broke the
interaction between logic and philosophy and in Poland logicians concentrated on more
technical matters: algebraic methods applied to the study of non-classical logics (Rasiowa,
see [59]), a general theory of logic (Suszko’s abstract logic, see [45]), foundations of
mathematics (in particular Mostowski, see [54]).6
The Lvov-Warsaw School is very much linked to the 2nd Republic of Poland which
lasted from 1918 to 1939, a 20 year period, described as follows by Jan Zygmunt:
In 1918, after 120 years of subjugation by foreign powers, Poland reappeared on the map of Europe
as a free and independent nation-state. In 1939 the outbreak of the Second World War reversed
these brief gains, bringing down the curtain on the Second Republic and inflicting massive damage
on its economy, culture and learning. [103, p. 6634].
During this interwar period grew a milieu, a society, a world that was almost
completely destroyed, didn’t re-appear and never will. In this period logic developed,
within a cocktail of philosophy, art and mathematics, an extraordinary situation which
does not exist anymore anywhere in the world. This happened in Poland at the beginning
of the twentieth century and also in Austria. Maybe it would make sense to speak about a
Vienna-Lvov-Warsaw school, considering in particular that Twardowski studied in Vienna
and that the interaction between logicians in Vienna and Warsaw was very important
during the 1930s. As we already said Tarski went to Vienna and met Gödel (see [38] and
[80]). The two schools met in Paris in 1935 at the congress organized by Louis Rougier
(cf. [25]). Quine came to Europe and symbolically visited Vienna and Warsaw (see [36]
and [58]).
Such atmosphere does not exist nowadays in other places in the world, in particular not
in United States, where in some sense the center of civilization was displaced after WWII.
Tarski created a school of logic in Berkeley, which became from the 1950s until his death
the main school of logic in the world, but it turned out to be mainly a school of model
theory centered on mathematics. At the start Tarski tried to develop applications of the
axiomatic method to other sciences, in particular to physics. He organized, jointly with
Leon Henkin and Patrick Suppes, an important event in December 1957 at Berkeley (cf.
[44]). But there was no real follow up. Nowadays the main group of people working on
the logic of physics, especially relativity theory, is in Hungary, a group formed by Istvan
Németi.
Other Polish logicians were spread around the world: Sobociński in Notre Dame
(USA), Bocheński in Fribourg (Switzerland), Łukasiewiz in Dublin (Ireland) [64], Surma
in Auckland (New Zealand). Although these people were quite active, for example
Sobociński created Notre Dame Journal of Logic, no one, excepted Tarski, created a group
6 Stan Surma recalls this time in a film we did with him in Xi’an. In Warsaw there were important seminars
directed by Andrzej Mostowski, Helena Rasiowa and Zdisław Pawlak. To know more about that period,
see [37], [85], [90].
Fig. 13 USA won the cold war: Warsaw becoming as beautiful as New York
or a school of logic. The relations between Polish logicians outside of Poland were not
developed in a systematic way. Moreover the relation between these expatriate logicians
with those inside Poland were severely restricted due to the cold war and the iron curtain.
So we cannot talk of a Polish school of logic around the world after WWII.
One may think that the end of Soviet Union has led to a true liberation of Poland, but
this is a fairly naive perspective symbolically illustrated by the statue of liberty at the
entrance of New York (Fig. 13). People say that during the Soviet period in Poland, they
had money but there was nothing to buy, after this period there are many things to buy
but they don’t have money to buy them. You can see a Ferrari in a shopping center but
it is a pleasure only for the eyes. One can also say more generally that during the Soviet
time there was no choice, in the post-Soviet time, there is the freedom to become a slave
(Fig. 14).
It is interesting to remember the etymology of the word “school”, skhole in Greek:
“spare time, leisure, rest ease; idleness; that in which leisure is employed; learned
discussion” (Online Etymology Dictionary.) In the open market society, a researcher has
to become a business man, i.e. someone who is always busy, in particular fundraising
is considered as a major quality and request for a researcher. There cannot therefore be
schools in the original sense, like was the Lvov-Warsaw School.
802 J.-Y. Beziau
Fig. 14 After 1991 Poland went further West
3.2 Logic in Poland 1992–2017
Since the end of the Soviet Union things are changing in Poland. I will describe the
situation from the point of view of my own experience having visited Poland 11 times in
25 years, my first visit being a 14 month stay at the Department of Logic of the University
of Wrocław in 1992–93 and my latest visit being a visit in Warsaw in June 2017 for the
2nd World Congress on Logic and Religion I was co-organizing.
From 1995 to 1999 I was working at the National Laboratory for Scientific Computing
(LNCC) in Rio de Janeiro. During this period I came three times to Poland for three
different conferences:
• April 1997, Karpacz, 2nd Conference Applications of Logic in Philosophy and the
Foundations of Mathematics
• July 1998, Toruń, Stanislaw Jaśkowski Memorial Symposium
• August 1999, Kraków, 11th LMPS (Logic, Methodology and Philosophy of Science)
congress.
The series of conferences “Applications of Logic in Philosophy and the Foundations
of Mathematics” (ALPFM) was launched in 1996 and it happens every year since then. It
generally takes place in a very nice village named Karpacz (now in Szklarska Porȩba) in
the South West of Poland. It has been organized by Jan Zgymunt, Piotr Wojtylak, Janusz
Fig. 15 JYB with Jan Zygmunt in 1997 in Karpacz
Czelakowski, Marcin Selinger, and Tomasz Połacik. I was invited to the first edition but
was not able to take part to it. Two neo-Brazilian friends of mine went there: Marcelo Tsuji
and David Miller. In 1997 I presented at the second edition the talk “Universal Logic in
Perspective” (Fig. 15).
July 15–18, 1998 was organized in Toruń a conference commemorating 50 years of
the publication of the paper by Jaśkowski on discussive logic (cf. [46]). It was a kind of
follow up of the first world congress on paraconsistency organized in Ghent (Belgium)
in 1997. Jaśkowski is considered as one of the two forerunners of paraconsistent logic
(the other one is the Russian logician Vasiliev). Generally Łukasiewicz is not considered
as a forerunner, although he wrote a seminal book on the principle of contradiction in
Aristotle [52], severely criticizing arguments of Aristotle defending this principle. But
Łukasiewicz didn’t present a paraconsistent system of logic. He famously presented a
three-valued system of logic [53] (later on criticized by Suszko [74]). Funny enough the
matrices of this system were later used by Asenjo, da Costa and D’Ottaviano and Priest,
to develop paraconsistent systems of logic.
The work of Jaśkowski on discussive logic has been promoted outside of Poland
mainly by Newton da Costa and his pupils, in particular Lafayette de Moraes (see e.g.
[34]). For this reason, at the occasion of this memorial conference was attributed to
da Costa Nicholas Copernicus University’s medal of merit. Da Costa was not able to
come. He was represented by his colleague Itala D’Ottaviano, with whom he developed
a logic called “J3”—the “J” referring to Jaśkowski, the “3” to three-valuedness. It is
a paraconsistent logic trying to modelize Jaśkowski’s ideas using three-valued logical
804 J.-Y. Beziau
Fig. 16 Jaśkowski Memorial Symposium, Nicholas Copernicus University, Toruń, 1998. JYB, Arthur
Buchsbaum, Lafayette de Moraes
matrices (see [35]). I myself presented at this meeting a talk entitled “The Paraconsistent
Logic Z” [17], presenting yet another modelization, closer to Jaśkowski’s ideas, based on
modalities, inspired by a discussion I had with Arthur Buchsbaum [18] (another Brazilian
logician, who also spent some times at the University of Wrocław through the connection
I established with Jan Zygmunt) (Fig. 16).
In 1999 I went to Poland for the 11th LMPS congress which took place in Kraków
August 20–26. It was my first participation to this series of events—after that I went to
the 12th edition in Oviedo (Spain) in 2003, the 13th edition in Beijing (China) in 2007 and
the 15th edition in 2015 in Helsinki (Finland). The edition in Kraków was by far the best.
In particular because it was the more logical one. This series of event, mainly initiated
and promoted by Tarski (cf. [38]), has become more and more oriented towards history
and philosophy of science, most of the time without any serious logical basis. The event in
Kraków was masterly organized by Jan Woleński. Stanislas Lem (1921–2006), the science
fiction writer, author of Solaris (with was famously adapted in cinema by Tarkowsky in
1972) was invited to give a general audience talk. I met there the late Hartley Salter, with
whom I had discussions which lead to my work on the square of opposition (many papers
and a series of world events, see [22, 26]). I presented at this LMPS meeting a talk entitled
“The Philosophical Import of Polish Logic” (cf. [13]) in which I emphasize the import of
three logical notions which were clarified within the Polish school of logic: structurality,
truth-functionality, extensionality.
In 2000 and 2001 I was at Stanford University working with Patrick Suppes and didn’t
come to Poland. From 2002 to 2008 I was working in Switzerland and during this time I
also came three times to Poland for three conferences:
• May 2003, Karpacz, 8th ALPFM conference
• April 2004, Karpacz, 9th ALPFM conference
• March 2006, Zakopane, Applications of Algebra to Logic and Informatics X.
I was invited to work at the Institute of Logic of the University of Neuchâtel in
Switzerland by Denis Miéville. Neuchâtel is where Jean Piaget was born and his father
was the first rector of the university. Piaget taught in Paris and then built a school and
research center in Geneva. One of his students, Jean-Blaise Grize, was from the region of
Neuchâtel and developed logic and semiology there. He became rector of the University
of Neuchâtel and created an institute of logic and a semiological research center. Miéville
was his main student and he became head of the institute of logic and later on also rector
of the university. Grize had interest for many things but Miéville main interest was about
Leśniewski. He published many papers and book on Leśniewski and became one of the
best specialist of the Polish bear followed by his student Pierre Joray. Knowing my interest
for Polish logic, Miéville invited me to work with them in Neuchâtel.
In 2003 I went to Karpacz with Pierre Joray and Nadine Gessler, another student of
Miéville. I took part to the 8th edition (May 6–10, 2003) of the ALPFM conference
in Karpacz, together with Pierre and Nadine, presenting the talk “A New Four-Valued
Approach to Modal Logic” (later publisher as [20]). The following year (2004) the event
took place April, 26–30, I came again, with Pierre Joray, and presented the talk “Does
logic need axioms?” (see my later paper [19]).
With Joray and Miéville we also went to a meeting in Nancy (November 21–22,
2003) organized by Roger Pouivet on Philosophy in Poland 1918–1939. A book was
subsequently published including my paper about Tarski’s axioms for the consequence
operator [16]. I met there for the first time Katarzyna Gan-Krzywoszyńska, who later on
was my partner for the organization of congresses on universal logic and the square of
opposition.
Sandra Lapointe visited us in Neuchâtel and then invited us to a meeting she organized
September 23–26, 2004, in Montréal, Canada: Logic, Ontology, Aesthetics—The Golden
Age of Polish Philosophy where I presented a lecture entitled “Tarski on Consequence
and Consequence” discussing the difference and relation between Tarski’s operator of
consequence and the notion of consequence he introduced in his famous paper “On the
concept of logical consequence” [77].
I had heard about Zakopane, in particular because of Witkiewicz, but never visited the
town, so I was glad to receive an invitation by Joanna Grygiel to come to the 10th edition
of the workshop Applications of Algebra to Logic and Informatics which took place there
March 6–12, 2006. I presented a talk entitled “Universal Algebra and Universal Logic”.
David Makinson also was there. There was an interesting excursion organized at night in
the mountain on a sledge driven by horses on the snow (Fig. 17).
Since 2010 I am back to Rio de Janeiro as professor of logic of the department of
philosophy of the University of Brazil. And since then I also came three times to Poland:
• February 2012, Poznań
• September 2016, Łódź, 8th International Conference—Non-Classical Logics Theory
and Applications
• June 2017, Warsaw, 2nd World Congress on Logic and Religion.
In January 2012 I was part of an exchange program with the University of Munich
with Matthias Schirn and I decided to come to visit Katarzyna in Poznań. I gave at Adam
Mickiewicz University a talk on the square of opposition and Katarzyna showed me the
place where Suszko was living (Figs. 18 and 19). Suszko’s father was the rector of the
University there after Ajdukiewicz.
In 2016 I was invited to take part to the 8th International Conference—Non-Classical
Logics Theory and Applications, which took place September 5–7, 2016 in Łódź,
806 J.-Y. Beziau
Fig. 17 In Zakopane in 2006, night excursion. Applications of Algebra to Logic and Informatics
Poland. This is a series of events which alternatively takes place in Łódź and Toruń,
two of the most active centers of logic in Poland since WWII. I presented there the
talk “Paraconsistent Logic from A to Z”. The pun is as follows: “A” is the first letter
of “Angers” the place where I discovered paraconsistent logic (see [18]) and “Z” the
last letter of “Łódź” where I was giving the talk and at the same time the name of
a paraconsistent logic I developed and that I named like that because it is based on
possible worlds and “Z” is the last letter of “Leibniz”. The organizer of the event, Andrzej
Indrzejczak, organized for us a special visit to Łódź Jewish cemetery which was very
impressive, with a guide telling us all about the story of Łódź’s ghetto.
During this visit I also did a stop in Warsaw in preparation of the 2nd world congress
on logic and religion that we organized there in June 2017. After a first successful edition I
organized in Brazil with my colleague Ricardo Silvestre in 2015 (cf. [28]), we received the
proposal by Marcin Trepczyński to organize a second edition at the University of Warsaw.
In September 2016 Marcin showed me the projected place of the congress and presented
me to Stanisław Krajewski who was the main organizer of the meeting together with
Marcin and Piotr Balcerowicz. The event was a great success with invited speakers such
as Dov Gabbay, Michal Heller (Templeton Prize), Saul Kripke (I succeed to convince him
to come for his first visit to Poland), Laurent Lafforgue (Fields medal), and Jan Woleński.
There were many participants from all over the world that enjoyed very much Warsaw
and the friendly atmosphere of the meeting (Fig. 20).
Fig. 18 JYB in front of Suszko’s house in Poznań

808 J.-Y. Beziau
Fig. 19 JYB in Poznań with Piotr Leśniewski
Fig. 20 The 2nd World Congress on Logic and Religion. University of Warsaw, June 2017
3.3 The Universal Logic Project
Since now more than 25 years I am developing the Universal Logic project. It is deeply
connected with Polish logic and the Lvov-Warsaw School. Here are five basic steps
corresponding to the development of the project:
• 1993, Wrocław, choice of the expression “Universal Logic”
• 1995, Paris, PhD Recherches sur La Logique Universelle
• 2005, Montreux, 1st World Congress on Universal Logic (UNILOG)
• 2007, Neuchâtel-Basel, launching of the journal Logica Universalis and the book series
• 2018, Vichy, 6th UNILOG, World Logic Prizes, including Alfred Tarski’s Prize
I chose the expression “Universal Logic” when in Wrocław in 1993. I have explained
the choice and meaning of this expression in some other papers (see in particular [14],
[15], [23] and [24]), so I will not enter in details here, just focusing on the essential.
The expression “Universal Algebra” was promoted (not invented) by Garrett Birkhoff.
The central feature of Birkhoff’s approach is the abandonment of any axiom. I was much
inspired by this to develop a general theory of logics, generalizing Tarski’s theory of
consequence operator, going one more step into abstraction, throwing out the axioms.
I crystallized this into the expression Axiomatic emptiness (see [19] and my paper [16]
discussing Tarski’s axioms).7
I presented then a lecture entitled “Universal Logic” at the Logica’94 meeting in Czech
Republic and defended in July 1995 a PhD in mathematical logic at the University of Paris
7 bearing this expression in the title [10]. Jan Zygmunt came to Paris and was a member
of the Jury. I remember that he stayed at the house of a Polish friend of us, Richard Zuber,
Rue Mouffetard. After that I spent time in Rio de Janeiro and California, in particular 2
years at Stanford University with Patrick Suppes.
I was able to systematically start to develop the Universal Logic project when I was in
Switzerland during the period 2002–2008. I submitted a research project on universal
logic to the Swiss Science Foundation (SNF) which was accepted. After organizing
a small workshop at the Universality of Neuchâtel in 2004 I started to work on the
preparation of the 1st UNILOG—World Congress and School on Universal Logic in
Montreux in Spring 2005 (Fig. 21, Fig. 22). The SNF was at this time giving special
support for people from Eastern Europe in particular students so I decided to invite
Katarzyna Gan-Krzywoszyńska from Poznań that I had known in Nancy in 2004 and
she accepted to help us in the organization of the event. She was especially dedicated
and efficient and she subsequently took part to the organization of the further editions
Fig. 21 UNILOG’2005 Montreux—train trip to the Marmot’s Paradise, Katarzyna Gan-Krzywoszyńska

with Andrei Rodin
7I recently defended the idea that anti-classical logic, the complementary of classical logic, can be
considered as a logic in view in particular of the theory of refutation initiated by Łukasiewicz. Anti-
classical logic is a concrete example of a logic obeying none of Tarski’s axioms; See [27].
810 J.-Y. Beziau
Fig. 22 UNILOG’2005, Montreux—excursion on the Lake. Urszula Wybraniec-Skardowska, Janusz

Czelakowski, Saul Kripke, Musa Akrami, Joanna Grygiel, João Marcos
of UNILOG: the 2nd UNILOG in Xi’an, China in 2007, the 3rd UNILOG in Lisbon,
Portugal in 2010, the 4th UNILOG in Rio de Janeiro, Brazil in 2013 and the 5th
UNILOG in Istanbul in 2015. She also helped me to develop other projects such as the
congresses on the square of opposition. We jointly organized a congress on analogy in
Puebla Mexico in 2015. She invited me to write a paper on a special issue of the journal
Studia Metodologiczne dedicated to Kazimierz Ajdukiewicz [25] and prepared the paper
“Personal recollections about JYB by Newton da Costa and others” published in the
second volume of the Festschrift for my 50th birthday [39].
At all UNILOGs we had important participations of Polish logicians: Janusz
Czelakowski (invited speaker at UNILOG’2005), Piotr Wojtylak (tutorial on consequence
operator at UNILOG’2005) and Joanna Grygiel (tutorial on universal algebra for logics at
UNILOG’2005), Stan Surma and Jan Woleński (invited speakers at UNILOG’2007),
Andrzej Wiśniewski (tutorial on erotetic logics at UNILOG’2013) and Beata
Konikowska (invited speakers at UNILOG’2013), Roman Murawski (invited speaker
at UNILOG’2015), Andrzej Indrzejczak (tutorial on cut-elimination at UNILOG’2015).
Urszula Wybraniec-Skardowska, one of the two editors of the present book, took part
to the 1st UNILOG in Montreux and to successive editions, organizing in particular a
workshop on logic and linguistics at UNILOG’2013 together with Marcos Lopes. As a
result they published a special issue of the Journal of Logic, Language and Information
(volume 23, issue 3, 2014). Urszula also took part to the SQUARE events, in particular
the 2014 edition in the Vatican [100].
The forthcoming edition of UNILOG will happen in Vichy in June 2018. The Polish
logician Jerzy Tomasik (who studied logic in Wrocław) was working at the University
Blaise Pascal in Clermont-Ferrand. He knew the late Marcel Guillaume, also from
this University, a friend of Newton da Costa, who helped him to publish his work on
paraconsistent logic in France in the 1960s and was therefore seminal in the world
promotion of paraconsistent logic (see [40–42]). Tomasik was at the UNILOG’2013 in
Rio and during a visit in Vichy I told him about my project to organize the 6th edition
there (He himself organized in Clermont-Ferrand the Logic Colloquium in 1994). He then
presented me to Christophe Rey working at the Vichy Campus of the now unified new
university Clermont Auvergne (UCA), with whom I am now working on the organization
of the 6th UNILOG.
At this edition of UNILOG there will be a session entitled World Logic Prizes. In
2014 I decided to create in Brazil, the Newton da Costa Logic Prize. It is a prize open to
any researcher working in logic in Brazil. People have to submit a non published paper.
The winner won publication of his paper in Logica Universalis and participation to an
international logic event. The first edition of the prize was organized in 2015 and the
winner, Rodrigo Freire, was invited to take part to UNILOG’2015 in Istanbul. I then
decided to work on the promotion of similar prizes in many countries. At UNILOG’2018
(Fig. 23) we will have about 15 countries presenting a winner, in particular the winner of
the Polish Prize of Logic that I suggested to name Alfred Tarski Logic Prize.
The present book is published in Birkhäuser’s book series Studies in Universal Logic
(SUL). I was particularly glad to develop with Birkhäuser a journal and a book series on
Fig. 23 UNILOG’2018
812 J.-Y. Beziau
universal logic, because the collected papers of Tarski were published by Birkhäuser. It is
a four volume book which was released in 1986 [79]. Since many years it is out of print.
We are now working on the re-edition of this book in the SUL series with an additional
fifth volume including letters and unpublished works of Tarski.
Acknowledgements Thanks to Jan Zygmunt, Newton da Costa, Jan Woleński, Roman Murawski,
Katarzyna Gan Krzywoszyńska, Pierre Cartier, Suasan Gómez, Vladimir Vsyukov, Urszula Wybraniec-
Skardowska and Rober Purdy for information, photos, comments and remarks.
References
1. Ajdukiewicz, K.: Z metodologii nauk dedukcyjnych, Lvov (1921)

2. Ajdukiewicz, K.: Allocution d’ouverture VI. In: Actes du congrès international de philosophie
scientifique, Sorbonne, Paris, 1935 Volume I, pp. 19–21. Hermann, Paris (1936)
3. Ajdukiewicz, K.: Logistyczny antyirracjonalizm w Polsce. Przeglad Filozoficzny, vol. 37, pp. 399–
408 [English translation: Logistic Anti-Irrationalism in Poland. Poznań Studies in the Philosophy of
the Sciences and the Humanities, vol. 74, pp. 241–250. Rodopi, Amsterdam (2001)]
4. Balcerzyk, S., Bartol, W., Orłowska, E., Wieczorek, A., Wojciechowska-Waszkiewicz, A.: Jerzy Łoś
1920–1998. Stud. Logica 65, 301–314 (2000)
5. Batóg, T.: Ajdukiewicz and the development of formal logic. In: Sinisi, V., Woleński, J. (eds.)
The Heritage of Kazimierz Ajdukiewicz. Poznań Studies in the Philosophy of Sciences and the
Humanities, vol. 40, pp. 53–67. Rodopi, Amsterdam (1995)
6. Batóg, T., Łuszczewska-Romahnowa, S.: Logic and methodology of science. In: Krajewski, W. (ed.)
Polish Philosophers of Science and Nature in the 20th Century. Poznań Studies in the Philosophy of
Sciences and the Humanities, vol. 74, pp. 53–67. Rodopi, Amsterdam (2001)
7. Beziau, J.-Y.: D’une caverne à l’autre. Master Thesis, Department of Philosophy, University Panthón
Sorbonne (Paris 1), Paris (1988)
8. Beziau, J.-Y.: La logique paraconsistante C1 de Newton da Costa. Master Thesis, Department of
Mathematics, University Denis Diderot (Paris 7), Paris (1990)
9. Beziau, J.-Y.: La logique abstraite au sein de la mathématique moderne. Ruch Filozoficzny 50, 289–
293 (1993)
10. Beziau, J.-Y.: Recherches sur la logique universelle. PhD Thesis, Department of Mathematics,
University Denis Diderot (Paris 7), Paris (1995)
11. Beziau, J.-Y.: O suicídio segundo Arthur Schopenhauer. Discurso 28, 127–143 (1997)
12. Beziau, J.-Y.: Rules, derived rules, permissible rules and the various types of systems of deduction.
In: Hauesler, E.H., Pereira, L.C. (eds.) Proof, Types and Categories, pp. 159–184. PUC, Rio de
Janeiro (1999)
13. Beziau, J.-Y.: The philosophical import of Polish logic. In: Talasiewicz, M. (ed.) Methodology and
Philosophy of Science at Warsaw University, pp. 109–124. Semper, Warsaw (2002)
14. Beziau, J.-Y.: From consequence operator to universal logic: a survey of general abstract logic. In:
Logica Universalis: Towards a General Theory of Logic, pp. 3–17. Birkhäuser, Basel (2005)
15. Beziau, J.-Y.: 13 Questions about universal logic. Bull. Sect. Log. 35, 133–150 (2006)
16. Beziau, J.-Y.: Les axiomes de Tarski. In: Pouivet, R., Rebuschi, M. (eds.) La philosophie en Pologne
1918–1939, pp. 135–149. Vrin, Paris (2006)
17. Beziau, J.-Y.: The paraconsistent logic Z - A possible solution to Jaśkowski’s problem. Log. Log.
Philos. 15, 99–111 (2006)
18. Beziau, J.-Y.: Adventures in the paraconsistent jungle. In: Beziau, J.-Y., Carnielli, W.A., Gabbay,
D.M. (eds.) Handbook of Paraconsistency, pp. 63–80. College Publication, London (2007)
19. Beziau, J.-Y.: What is a logic? - Towards axiomatic emptiness. Log. Investig. 16, 272–279 (2010)
20. Beziau, J.-Y.: A new four-valued approach to modal logic. Log. Anal. 54, 109–121 (2011)
21. Beziau, J.-Y. (ed.): Universal Logic: An Anthology. Birkhäuser, Basel (2012)
22. Beziau, J.-Y.: The new rising of the square of opposition. In: Beziau, J.-Y., Jacquette, D. (eds.)
Around and Beyond the Square of Opposition, pp. 6–24. Birkhäuser, Basel (2012)
23. Beziau, J.-Y.: The relativity and universality of logic. Synthese. Special Issue Istvan Németi 70th
Birthday 192, 1939–1954 (2015)
24. Beziau, J.-Y.: Logical autobiography 50. In: Koslow, A., Buchsbaum, A. (eds.) The Road to
Universal Logic: Festschrift for the 50th Birthday of Jean-Yves Beziau Volume II, pp. 19–104.
25. Beziau, J.-Y.: Is logical relativity irrational? Stud. Metodologiczne. Special Issue Kazimierz
Ajdukiewicz 35, 63–78 (2015)
26. Beziau, J.-Y.: Hartley Slater and false contradictions. South Am. J. Log. 6, 101–107 (2016)
27. Beziau, J.-Y., Buchsbaum, A.: Let us be antilogical: anti-classical logic as a logic. In: Moktefi, A.,
Moretti, A., Schang, F. (eds.) Soyons Logiques/Let Us Be Logical, pp. 1–10. College Publication,
London (2016)
28. Beziau, J.-Y., Silvestre, R.: Logic and religion. Log. Univers. 11, 1–12 (2017)
29. Burckhardt, J.: Der Cicerone Eine Anleitung zum Genu der Kunstwerke Italiens. Schweighauser,
Basel (1855)
30. Chapman, S.: Language and Empiricism - After the Vienna Circle. Palgrave Macmillan, Basingstoke
(2018)
31. Coniglione, F., Poli, R., Woleński, J. (eds.): Polish Scientific Philosophy: The Lvov-Warsaw School.
Rodopi, Amsterdam (1993)
32. da Costa, N.C.A., Beziau, J.-Y.: La théorie de la valuation. Log. Anal. 146, 95–117 (1994)
33. da Costa, N.C.A., Beziau, J.-Y.: La théorie de la valuation en question. In: Abad, M. (ed.) Proceed-
ings of the Ninth Latin American Symposium on Mathematical Logic, pp. 95–104. Universidad del
Sur, Baíha Blanca (1994)
34. de Moraes, L.: Sôbre a lógica discurisva de Jaśkowski. Master Thesis, University of São Paulo, São
Paulo (1970)
35. D’Ottaviano, I.M.L., da Costa, N.C.A.: Sur un problème de Jaśkowski. C. R. Acad Sci. Paris 270,
1349–1353 (1970)
36. Dresner, E.: Quines philosophy of language and Polish logic. Hist. Philos. Log. 20, 79–96 (1999)
37. Ehrenfeucht, A., Marek, W., Srebrny, M. (eds.): Andrzej Mostowski and Foundational Studies. IOS
Press, Amsterdam (2008)
38. Feferman, S., Feferman, A.: Alfred Tarski, Life and Work. Cambridge University Press, Cambridge
(2004)
39. Gan-Krzywoszyńska, K.: Personal recollections about JYB by Newton da Costa and others. In:
Koslow, A., Buchsbaum, A. (eds.) The Road to Universal Logic: Festschrift for the 50th Birthday of
Jean-Yves Beziau Volume II, pp. 1–18. Birkhäuser, Basel (2015)
40. Guillaume, M.: La logique mathématique en sa jeunesse. In: Pier, J.-P. (ed.) Development of
Mathematics 1900–1950, pp. 185–367. Birkhäuser, Basel (1994)
41. Guillaume, M.: Regard en arrière sur quinze annèes de coopération douce avec lécole brésilienne de
logique paraconsistante. Log. Aanal. 39, 5–14 (1996)
42. Guillaume, M.: La logique mathématique en France entre les deux guerres mondiales: Quelques
repères. Rev. Hist. Sci. 62, 177–219 (2009)
43. Hendricks, V.F., Malinowski, J.: 50 years of studia logica: editorial introduction. In: Trends in Logic
50 Years of Studia Logica, pp. 1–10. Springer, Dordrecht (2003)
44. Henkin, L., Suppes, P., Tarski, A. (eds.): The axiomatic method, with special reference to geometry
and physics. In: Proceedings of an International Symposium Held at the University of California,
Berkeley, December 26, 1957–January 4, 1958. North-Holland, Amsterdam (1959)
45. Jansana, R.: Bloom, Brown and Suszko’s work on abstract logics. In: Beziau, J.-Y. (ed.) Universal
Logic: An Anthology, pp. 251–256. Birkhäuser, Basel (2012)
46. Jáskowski, S.: Rachunek zdan dla systemów dedukcyjnych sprzecznych. Stud. Soc. Scientiarum
Torunensis Sect. A I(5), 55–77 (1948)
47. Jordan, Z.: The development of mathematical logic and of logical positivism in Poland between the
two wars. In: Polish Science and Learning, vol. 6. Oxford Oniversity Press, Oxford (1945)
48. Kijania-Placek, K., Woleński, J. (eds.): The Lvov-Warsaw School and Contemporary Philosophy.
Springer, Dordrecht (1998)
814 J.-Y. Beziau
49. Lapointe, A., Woleński, J., Marion, M., Miskiewicz, W.: The Golden Age of Polish Philosophy -
Kazimierz Twardowski’s Philosophical Legacy. Springer, Dordrecht (2009)
50. Leśniewski, S.: In: Surma, S.J., Srzednicki, J.T., Barnett, D.I., Rickey, V.F. (eds.) Collected Works,
2 volumes. Nijhoff - Kluwer, Amsterdam; Polish Academy of Science, Warsaw (1992)
51. Łoś, J.: O matrycach logicznych’. Prace Wrocławskiego Towarzystwa Naukowego Seria B 19, 1–41
(1949)
52. Łukasiewicz, J.: O zasadzie sprzeczności u Arystotelesa. Akademia Umiejȩtniści, Kraków (1910)
53. Łukasiewicz, J.: O logice trójwartościowej Ruch Filozoficny 5, 170–171 (1920)
54. Mostowski, A.: Foundational Studies, Selected Works, vols. 1 and 2. North-Holland, Amsterdam
(1979)
56. Pogorzelski, W., Wojtylak, P.: Completeness Theory for Propositional Logics. Birkhäuser, Basel
(2008)
57. Pouivet, R., Rebuschi, M. (eds.): La philosophie en Pologne 1918–1939. Vrin, Paris (2006)
58. Quine, W.V.O.: The Time of My Life. MIT Press, Cambridge (1985)
59. Rasiowa, H.: An Algebraic Approach to Non-Classical Logics. North-Holland, Amsterdam (1974)
60. Reichbach, J.: Completeness of the functional calculus of first-order. Stud. Logica 2, 245–250 (1955)
61. Reichbach, J.: On the first-order functional calculus and the truncation of models. Stud. Logica 7,
181–220 (1958)
62. Rescher, N.: The Berlin school of logical empiricism and its legacy. Erkenntnis 64, 281–304 (2006)
63. Rousseau, J.-J.: Dictionnaire de la musique. Marc Michel Rey, Amsterdam (1768) [English
translation by William Waring, London, 1779]
64. Schmidt Am Busch, H.C., Wehmeier, K.F.: On the relations between Heinrich Scholz and Jan
Łukasiewicz. Hist. Philos. Log. 28, 67–81 (2007)
65. Sinisi, V., Woleński, J. (eds.): The Heritage of Kazimierz Ajdukiewicz. Poznań Studies in the
Philosophy of Sciences and the Humanities, vol. 40. Rodopi, Amsterdam (1995)
66. Skolimowski, H.: Polish Analytical Philosophy. Routledge and Kegan Paul, London (1967)
67. Sobociński, B.: In Memoriam Jan Łukasiewicz (1878–1956). Philos. Stud. 6, 349 (1956)
68. Surma, S.: Studies in the History of Mathematical Logic. Ossolineum, Wrocław (1973)
69. Surma, S. (ed.): Special issue on Mordchaj Wajsberg. Bull. Sect. Log. 2, 91–96 (1973)
70. Surma, S.: On the origin and subsequent applications of the concept of the Lindenbaum algebra.
In: Cohen, L.J., Łoś, J., Pfeiffer, H., Podewski, K.-P. (eds.) Logic, Methodology and Philosophy of
Science, VI (Hannover, 1979), pp. 719–734. North-Holland, Amsterdam (1982)
71. Surma, S.: The logical work of Mordchaj Wajsberg. In: Srzednicki, J. (ed.) Initiatives in Logic,
pp. 101–115. Matrinus Nijhoff, Dordrecht (1987)
72. Surma, S.: Alternatives to the consequence theoretic approach to metalogic. In: Abad, M. (ed.)
Proceedings of the Ninth Latin American Symposium on Mathematical Logic, pp. 1–30. Universidad
del Sur, Baha Blanca (1994)
73. Surma, S.: A Galois connection. Logica Univers. 1, 209–219 (2007)
74. Suszko, R.: The Fregean Axiom and Polish mathematical logic in the 1920s. Stud. Logica 36, 377–
380 (1978)
75. Szaniawski, K. (ed.): The Vienna Circle and the Lvov-Warsaw School. Nijhoff, Dordrecht (1988)
76. Tarski, A.: O Logice Matematycznej i Metodzie Dedukcyjnej, Atlas, Lvov-Warsaw (1936). Eng.
ed., Introduction to Logic and to the Methodology of Deductive Sciences. Oxford University Press,
Oxford (1941)
77. Tarski, A.: O pojciu wynikania logicznego. Przegld Filozoficzny 39, 58–68 (1936)
78. Tarski, A.: Sur la méthode déductive. Travaux du IXe Congrès International de Philosophie, vol. 6,
Actualités Scientifiques et Industrielles, vol. 535, pp. 95–103. Hermann & Cie, Paris (1937)
79. Tarski, A.: In: Givant, S.R., McKenzie, R.N. (eds.) Collected Papers - 4 Volumes. Birkhäuser, Basel
(1986)
80. Tarski, A.: Drei Briefe an Otto Neurath. Grazer Philosophische Studien 43, 1–32 (1992)
81. Wajsberg, M.: Logical Works (translated and edited by S.J. Surma). Ossolineun, Wrocław (1977)
82. Wójcicki, R.: Lectures on Propositional Calculi. Ossolineum, Wrocław (1984)
83. Wójcicki, R.: Theory of Logical Calculi. Basic Theory of Consequence Operations. Kluwer,
Dordrecht (1988)
84. Wójcicki, R.: The postwar panorama of logic in Poland. In: dalla Chiara, M.L., et al. (eds.) Logic
and Scientific Methods, pp. 497–508. Kluwer, Dordrecht (1997)
85. Wójcicki, R., in collaboration with Zygmunt, J.: Polish logic in postwar period. In: Trends in Logic
50 Years of Studia Logica, pp. 11–33. Springer, Dordrecht (2003)
86. Woleński, J.: Filozoficzna szkoła lwowsko-warszawska. Państwowe Wydawnictwo Naukowe,
Warszawa (1985)
87. Woleński, J.: Filozofia szkoły lwowsko-warszawskiej. Uniwersyte Wrocławski, Wrocław (1986)
88. Woleński, J.: Logic and Philosophy in the Lvov-Warsaw. Kluwer, Dordrecht (1989)
89. Woleński, J.: Polish logic. Log. J. IGPL 2, 93–100 (1992)
90. Woleński, J.: Philosophy inside communism: the case of Poland. Stud. Sov. Thought 43, 93–100
(1992)
91. Woleński, J.: Mathematical logic in Poland 1900 1939: people, circles, institutions, ideas. Mod. Log.
5, 363–405 (1995)
92. Woleński, J.: Essays in the History of Logic and Logical Philosophy. Jagiellonian University Press,
Kraków (1999)
93. Woleński, J.: The achievements of Polish School of Logic. In: Baldwin, Th. (ed.) The Cambridge
History of Philosophy 1870–1945, pp. 401–416. Cambridge University Press, Cambridge (2003)
94. Woleński, J.: Mathematical logic in Warsaw: 1918–1939. In: Ehrenfeucht, A., Marek, W., Srebrny,
M. (eds.) Andrzej Mostowski and Foundational Studies, pp. 30–46. IOS Press, Amsterdam (2008)
95. Woleński, J.: Kazimierz Twardowski and the development of philosophy of science in Poland. In:
Galavotti, M.C., Nemeth, E., Stadler, F. (eds.) European Philosophy of Science: Philosophy of
Science in Europe and the Viennese Heritage, pp. 173–182. Springer, Cham (2014)
96. Woleński, J.: Adolf Lindenbaum. In: Internet Encyclopedia of Philosophy (2017)
97. Woleński, J., Köhler, E.: Alfred Tarski and the Vienna Circle - Austro-Polish Connections in Logical
Empiricism. Kluwer, Dordrecht (1999)
98. Woleński, J., Zygmunt, J.: Jerzy Słupecki (1904–1987): life and work. Stud. Logica 48, 401–411
(1989)
99. Wybraniec-Skardowska, U.: Polish Logic. Some Lines from a Personal Perspective. Publications of
100. Wybraniec-Skardowska, U.: Logical squares for classical logic sentences. Logica Univers. 10, 293–
312 (2016)
101. Zygmunt, J.: The logical investigations of Jan Kalicki. Hist. Philos. Log. 2, 41–53 (1981)
102. Zygmunt, J.: Mojȯzesz Presburger: life and work. Hist. Philos. Log. 12, 221–223 (1991)
103. Zygmunt, J.: Polish logic. In: Routledge Encyclopedia of Philosophy, pp. 6634–6639. Routledge,
London/New York (1998)
104. Zygmunt, J., Purdy, R.: Adolf Lindenbaum: notes on his life, with bibliography and selected
references. Logica Univers. 8, 285–320 (2014)
J.-Y. Beziau ()

Brazilian Research Council, University of Brazil, Rio de Janeiro, Brazil
Ecole Normale Supérieure, Paris, France

The Lvov-Warsaw School. Past and Present. Ángel Garrido - Urszula Wybraniec-Skardowska (Eds.)

Uploaded by

Copyright:

Available Formats

The Lvov-Warsaw School. Past and Present. Ángel Garrido - Urszula Wybraniec-Skardowska (Eds.)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Lvov-Warsaw School. Past and Present. Ángel Garrido - Urszula Wybraniec-Skardowska (Eds.)

Uploaded by

Copyright:

Available Formats

Studies in Universal Logic

Editorial Board Members

Hajnal Andréka (Hungarian Academy of Sciences, Budapest, Hungary)

More information about this series at http://www.springer.com/series/7391

The Lvov-Warsaw School.

ISSN 2297-0282 ISSN 2297-0290 (electronic)

Library of Congress Control Number: 2018943846

Mathematics Subject Classification (2010): 01A72, 03-03

© Springer International Publishing AG, part of Springer Nature 2018

Printed on acid-free paper

Editor-in-Chief Jean-Yves Beziau

Part I Twardowski’s School: The Period of Crystallization of LWS

Introduction. The School: Its Genesis, Development and Significance . . . . . 3

Kazimierz Twardowski: A Great Teacher of Great Philosophers . . . . . . . . 15

Jan Łukasiewicz: A Creator of New Ideas in Logic and a Reinterpreter

Kazimierz Ajdukiewicz: The Cognitive Role of Language . . . . . . . . . . . . 47

On Ajdukiewicz’s Project of the Semantic Theory of Knowledge . . . . . . . . 65

Categorial Grammars and Their Logics . . . . . . . . . . . . . . . . . . . . . . . 91

Material Implication and Conversational Implicature in Lvov-Warsaw

Tadeusz Czeżowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133

Tadeusz Czeżowski’s Theory of Knowledge and Beliefs . . . . . . . . . . . . . . 137

What Is Reasoning? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153

Tadeusz Kotarbiński: Socrates of Warsaw . . . . . . . . . . . . . . . . . . . . . . 165

Agency in a Praxiological Approach . . . . . . . . . . . . . . . . . . . . . . . . . 175

Zygmunt Zawirski: A Creator of New Ideas in Methodology of Science

Zygmunt Zawirski’s Concept of Scientific Metaphysics . . . . . . . . . . . . . . 201

Stanisław Leśniewski: Original and Uncompromising Logical Genius . . . . . 209

Maria Kokoszyńska-Lutmanowa: The Great Polemist . . . . . . . . . . . . . . . 235

Seweryna Łuszczewska-Romahnowa . . . . . . . . . . . . . . . . . . . . . . . . . 241

On Existential Dependence and Independence in the World of Thoughts

Józef I.M. Bocheński . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263

J.M. Bocheński’s Theory of Signs . . . . . . . . . . . . . . . . . . . . . . . . . . . 273

Jan Salamucha (1903–1944) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279

6 Intellectual Formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282

Struve and Biegański: Towards Modern Approach to Logic . . . . . . . . . . . 285

Łukasiewicz and His Followers in Many-Valued Logic . . . . . . . . . . . . . . 301

Tomorrow’s Sea-Battle and the Beginning of Temporal Logic . . . . . . . . . . 329

Alfred Tarski (1901–1983) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 361

Some Philosophical Aspects of Semantic Theory of Truth . . . . . . . . . . . . . 373

Tarski’s Influence on Computer Science . . . . . . . . . . . . . . . . . . . . . . . 391

The Absence of Multiple Universes of Discourse in the 1936 Tarski

Stanisław Jaśkowski: Life and Work . . . . . . . . . . . . . . . . . . . . . . . . . 457

Stanisław Jaśkowski and Natural Deduction Systems . . . . . . . . . . . . . . . 465

Variations on Jaśkowski’s Discursive Logic . . . . . . . . . . . . . . . . . . . . . 485

Czesław Lejewski: Propagator of Lvov-Warsaw Ideas Abroad . . . . . . . . . . 499

Andrzej Mostowski: A Biographical Note . . . . . . . . . . . . . . . . . . . . . . 551

Foundations and Philosophy of Mathematics in Warsaw, the School

Jerzy Słupecki (1904–1987) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 567

Rejection in Łukasiewicz’s and Słupecki’s Sense . . . . . . . . . . . . . . . . . . 575

Bolesław Sobociński on Universals . . . . . . . . . . . . . . . . . . . . . . . . . . 615

Many-Valued Logics in the Iberian Peninsula . . . . . . . . . . . . . . . . . . . . 633

Ontology of Logic and Mathematics in Lvov-Warsaw School . . . . . . . . . . 645

4 Tadeusz Kotarbiński: Reism . . . . . . . . . . . . . . . . . . . . . . . . . . . . 652

Part III The War and Post-War Period

A View of Revival of Mathematical Logic in Warsaw, 1945–1975 . . . . . . . . 665

Andrzej Mostowski and the Notion of a Model . . . . . . . . . . . . . . . . . . . 673

All Quantifiers Versus the Quantifier All . . . . . . . . . . . . . . . . . . . . . . 693

Helena Rasiowa (1917–1994) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 703

Post Algebras in the Work of Helena Rasiowa . . . . . . . . . . . . . . . . . . . . 711