Andrew McFarland

Joanna McFarland
James T. Smith Editors

Alfred Tarski
Early Work in Poland –
Geometry and Teaching
This book is dedicated to Helen Marie Smith, in gratitude for
her advice and support, and to Maria Anna McFarland, as she
enters a world of new experiences.
Andrew McFarland • Joanna McFarland
James T. Smith
Editors

Alfred Tarski
Early Work in Poland—Geometry
and Teaching

with a Bibliographic Supplement

Foreword by Ivor Grattan-Guinness

Editors
Andrew McFarland Joanna McFarland
Páock, Poland Páock, Poland

James T. Smith
Department of Mathematics
San Francisco State University
San Francisco, CA, USA

ISBN 978-1-4939-1473-9 ISBN 978-1-4939-1474-6 (eBook)

DOI 10.1007/978-1-4939-1474-6
Springer New York Heidelberg Dordrecht London

Library of Congress Control Number: 2014945118

Mathematics Subject Classification (2010): 01A60, 01A70, 01A75, 03A10, 03B05, 03E75,
06A99, 28-03, 28A75, 43A07, 51M04, 51M25, 97B50, 97D40, 97G99, 97M30

© Springer Science+Business Media New York 2014

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. Exempted from this
legal reservation are brief excerpts in connection with reviews or scholarly analysis or material
supplied specifically for the purpose of being entered and executed on a computer system, for
exclusive use by the purchaser of the work. Duplication of this publication or parts thereof is
permitted only under the provisions of the Copyright Law of the Publisher’s location, in its
current version, and permission for use must always be obtained from Springer. Permissions for
use may be obtained through RightsLink at the Copyright Clearance Center. Violations are liable
to prosecution under the respective Copyright Law.
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.
While the advice and information in this book are believed to be true and accurate at the date of
publication, neither the authors nor the editors nor the publisher can accept any legal responsibility
for any errors or omissions that may be made. The publisher makes no warranty, express or
implied, with respect to the material contained herein.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.birkhauser-science.com)

 Foreword
by Ivor Grattan-Guinness

1. Context

It is common in the history of ideas to find that the reception and development of some
theories of a historical figure proceeded much more rapidly than others, both during his
lifetime and afterward. As a result a historian can enjoy expounding the “neglected”
contributions of the figure, and search for causes of this oversight. However, it is rather
rare for a body of his work to disappear entirely from sight, especially when it was pub-
lished at the time of its creation and the figure in question has been subject to consider-
able attention both by contemporary co-workers and by historians.

Yet this situation pertains to some of the early work of the great mathematician and
logician Alfred Tarski (1901–1983). It is well known that he worked as a schoolteacher
in Poland from the early 1920s and that he published papers in logic and mathematics
regularly, mostly in Polish or Austrian journals; there is even an edition of English
translations of versions of many of these papers (Tarski [1956] 1983). However, during
this period he also published a substantial number of papers, comments, and problems
in Polish journals oriented around mathematics education; but they are not handled in
the fine biography, Feferman and Feferman 2004. Further, the journals in question
escaped the attention of the German mathematics abstracting journal Jahrbuch über die
Fortschritte der Mathematik. In addition, Tarski coauthored an elementary textbook on
geometry in 1935 that has generally been overlooked. The purpose of the present volume
is to bring this material to light in English translation, and also to amplify the historical
context in which the teacher Tarski produced it in the first place.

There are two trends in the development of logic and the foundations of mathematics
that became of major significance during this period, so that their bearing upon Tarski’s
work are worth seeking. One was the development of nonclassical logics to complement
the classical two-valued logic and, in the excessively polemical case of the intuitionism
of L. E. J. Brouwer, to replace it (Mancosu 1998). Tarski does not seem to have been
concerned with other logics in the papers here. The other was the growing recognition
of the need to work in both logic and mathematics with hierarchies of theories, distin-
guishing mathematics from metamathematics (David Hilbert’s program) and logic from
metalogic (a central feature of Kurt Gödel’s famous paper of 1931 on the incompletability
of first-order arithmetic, which stimulated his friend Rudolf Carnap to coin “metalogic”
that year as a technical term. Tarski would soon contribute notably to this trend by coin-
ing the word “metalanguage” in connection with his semantic theory of truth of the 1930s.
vi Foreword

2. Cleft

One would have thought that since logic was among other things a study of the theory
of deduction and since mathematics deployed deduction in proofs of theorems and
definitions of concepts, the two disciplines would happily live side by side. However, this
has never been the case (as is already evident in, for example, Euclid’s Elements); in
particular, the rise in interest in foundational subjects from the late nineteenth century
onward did not eliminate it at all.

A very important source of maintaining the cleft during the nineteenth century is the
founding from the late 1810s onward of the “mathematical analysis” of real variables,
grounded upon an articulated theory of limits, by the French mathematician Augustin
Louis Cauchy. He and his followers extolled rigor, in particular careful nominal defini-
tions of major concepts and detailed proofs of theorems. From the 1850s onward this aim
was enriched by the German mathematician Karl Weierstrass and his many follow-
ers—they brought in, for example, multiple-limit theory, definitions of irrational num-
bers, and an increasing use of symbols—and from the early 1870s, Georg Cantor and his
set theory. However, none of these developments explicitly drew upon any kind of logic.

This silence continued among the many set theorists who helped to develop measure
theory, functional analysis, and integral equations (Jahnke 2003). Even the extensive
dispute over the axiom of choice (a Tarski favorite) from 1904 onward focused mostly on
its legitimacy as an assumption in set theory and mathematics and on the use of higher-
order quantification (Moore 1982): its need to state an infinitude of independent choices
within finitary logic was a trouble for logicians.

The creators of symbolic logics were exceptional among mathematicians in attending

to logic, but they made little impact on their colleagues. The algebraic tradition with
George Boole, Charles Sanders Peirce, Ernst Schröder, and others from the mid-
nineteenth century was just a curiosity to most of their contemporaries. Similarly, when
mathematical logic developed from the late 1870s, especially with Giuseppe Peano’s
“logistic” program at Turin from around 1890, it gained many followers there (Luciano
and Roero 2010) but few elsewhere. However, followers in the 1900s included the Britons
Bertrand Russell and Alfred North Whitehead, who adopted logistic (including Cantor’s
set theory) and converted it into their “logicistic” thesis that all the “objects” of mathe-
matics could be obtained from it; G. H. Hardy but not many other mathematicians
responded (Grattan-Guinness 2000, chapters 8 and 9). From 1903 onward Russell had
also publicized the mathematical logic and arithmetic logicism put forward from the late
1870s onward by Gottlob Frege, which had gained little attention hitherto even from
students of foundations and did not gain much more in the following decades. Hilbert’s
program of metamathematics attracted several followers at the University of Göttingen
and a few elsewhere; however, its impact among mathematicians was limited even
in Germany.
Foreword vii

The situation in Poland is quite poignant; for soon after the re-creation of the country
after the Great War, major schools of mathematicians and logicians emerged. Moreover,
many of the mathematicians worked in set theory and/or its applications, so that links
with logic could be close. In 1920 the Poles launched in Warsaw a journal, Fundamenta
Mathematicae, to cover both disciplines (Lebesgue 1922). The two coeditors for logic were
Jan ukasiewicz and Stanisaw Leniewski, who had obtained chairs at the University
of Warsaw in 1915 and 1919 respectively after taking doctorates at Lwów. They and their
associates formed the largest community in the world working on logic and related topics
(McCall 1967). I am told that Polish has many properties relevant to logic and set theory,
which may have helped stimulate the interest in the first place. However, not many logic
papers appeared in the journal (Tarski being the most frequent author), and the two
logicians resigned from its board in 1928, with little regret from the mathematicians
(Kuratowski 1980, 33–34).

By contrast, a most praiseworthy feature of Tarski’s work is that he researched in both

logic and branches of mathematics (especially geometry) at the same time, thereby
consciously ignoring the cleft. In particular, the material newly come to light treats not
only logic and set theory but also elementary geometry and common algebra. But he
wrote it mostly in Polish, which will have reduced its market, although some of its
contents appeared elsewhere in papers in German or French. Let us now encounter it
in English guise.
 Preface
Alfred Tarski is regarded as one of the four greatest logicians of all time—the others being
Aristotle, Gottlob Frege, and Kurt Gödel.1 Most notably, Tarski was largely responsible
for designing the infrastructure on which most logical research has been based since 1950.
Using that structure, he achieved major technical results in logic, foundations of geome-
try, and abstract algebra. Outside that framework, Tarski discovered major theorems
in set theory and set-theoretic aspects of geometry, and completed some works about
teaching geometry.

Tarski was born in Warsaw in 1901, and grew up there in a time of turmoil. He
completed doctoral studies in 1924, just after a major expansion of the Polish system of
universities. The economic climate was adverse, as was the growing antisemitism. Tarski
obtained full-time employment as a secondary-school teacher, and worked part-time as
a university assistant and researcher. During the next fifteen years, he gained world
renown in the fields of set theory and logic. On a 1939 lecture tour, he was stranded,
fortunately, in the United States, when the Germans and Soviets invaded Poland. After
some trying years, he secured a professorship at the University of California, Berkeley.
There, according to the eminent Polish philosopher Jan Woleski,
Tarski ... created the great Californian School of logic. He ... had a dominant influence upon
the development of logic after World War II.2

After four decades of service at Berkeley, Tarski died there in 1983.

Much of Tarski’s scientific work has been accessible and has become rather well known.
Three years after his death, his colleagues Steven R. Givant and Ralph McKenzie pub-
lished the remarkably complete Collected Papers volumes and a detailed bibliography.
An excellent biography by Anita B. and Solomon Feferman is available in English and
Polish.3 Tarski took meticulous care to document his Berkeley research program, and that
material is readily available for historical study. But a few of Tarski’s early works have
been difficult to access, and for some of them, hardly any background was even detectible.
This was particularly true for some of his early work about geometry. He had completed
that in Warsaw, which was largely destroyed during World War II. There is almost noth-
ing of his left there, and Tarski’s archive in Berkeley starts in 1939.

1
Corcoran 1991.
2
Woleski 1989, 20.
3
Tarski 1986a; Givant 1986; Feferman and Feferman 2004 and 2009. For descriptions of these works see
section 16.2 and chapter 17.
x Preface

The present book has three main goals:

(1) To publish translations as necessary so that
• Alfred Tarski’s works will all be accessible in English, French, or
German; and
• his geometric works will all be accessible in English.
(2) To provide scientific and cultural background information about the
works translated here: their origin, context, structure, and impact.
(3) To update Givant’s 1986 bibliography of Tarski’s publications, and
include an annotated list of major studies of Tarski’s life and work.

By including ample background material in this book, the editors have heeded an
opinion expressed succinctly by a leading mathematics historian, David E. Rowe:
The type of knowledge mathematicians have produced has depended heavily on cultural, politi-
cal, and institutional factors that shaped the various environments in which they have worked.4

The book stresses the connection between Tarski’s work as a teacher and the subjects
of some of his research. It does not itself pursue other connections between environment
and research, but does aim to provide scholars interested in that kind of inquiry a glimpse
of the background and routes to deeper and broader study. It does not supplant the
Fefermans’ 2004 biography, but fills some gaps in their coverage. In some cases, the
present editors have included significant background that was apparently not familiar
to others who have written about Tarski. This is especially true for the year, 1920, of the
Polish–Soviet War, and for Tarski’s work as a secondary-school teacher and teacher-
trainer.

The present editors expect that further background material specifically about Tarski
will come to light only through historians’ studies of other students, teachers, and
scientists associated with him. This is particularly the case for Tarski’s disrupted first
university year, 1918–1919. Since articles about mathematics or philosophy are usually
edited to exclude material not directly related to their theoretical content, the social
context of Tarski’s work will probably be better revealed by investigating his activity in
teaching and teacher-training. Researchers should adopt a maxim: if Tarski is one of
the top four logicians of all time, he must be one of the top n thinkers, and what he
thought about almost anything should be interesting!

This book is organized into four parts:

Part One . . . . Debut Part Three . . . Teaching
Part Two . . . . Geometry Part Four . . . . Supplement
The supplement contains some translations not directly related to geometry research or
teaching, an update of the 1986 bibliography, and annotated lists of major studies of
Tarski and his work.

4
Rowe 2003, 114.
Preface xi

Readers will find general background material in chapters 1, 3; 4, 8; and 9, 14: the
first and last chapters of its three main parts. Background specific to particular transla-
tions is located in the chapters devoted to them.

Part One of the book, Debut, contains only one translation: Tarski’s first published
paper, A Contribution to the Axiomatics of Well-Ordered Sets, written in 1921 while he
was still a student. Two background chapters describe his life through the completion
of his doctoral study in 1924. They complement the Fefermans’ biography, adding mater-
ial about the turbulent years 1918–1920 in Warsaw. There remain gaps in the story: for
example, it is still not clear why Tarski changed his major interest from biology to logic
and mathematics.

Part Two, Geometry, is devoted to Tarski’s work on equidecomposability. The back-

ground chapter 4 summarizes the elementary theory of area and volume, covered to some
extent in secondary schools. It then considers the measure-theoretic approach that led
Stefan Banach and Tarski to show in 1924 that set-theoretic decomposition alone will not
yield a theory in accordance with intuition: a marble and the earth can each be dissected
into the same finite number of disjoint sets of points, which are congruent in pairs. Their
famous paper is translated in this part, along with Tarski’s 1924 paper that showed that
this counterintuitive result has no analogue in the plane. The latter paper was published
in a journal aimed at secondary-school teachers. Attention is then turned to Tarski’s
work on the more elementary concept of decomposition of plane polygons into subpolygons
whose interiors are disjoint but whose boundaries may overlap. Three elegant little
papers are translated, which appeared during 1931–1932 in journals aimed at secondary-
school students and their teachers. Two of those are by Tarski; the third, by another
schoolteacher, Henryk Moese. Inadequate translations of the three were published
obscurely half a century ago, but the translations in the present book are new. The
subsequent impact of all five papers is traced in chapter 8, which concludes Part Two.

Part Three of the book, Teaching, presents a variety of material. Its first chapter
describes Tarski’s family situation and his teaching in a secondary school and in univer-
sity lectures and research seminars. It complements the Fefermans’ treatment of these
aspects of his life. Tarski’s 1929 report to teachers about an important research
conference and his 1932 suggestions on teaching about circles are translated. During
1930–1932 Tarski published fourteen exercises to challenge teachers and talented
students; they are translated and analyzed here. So are representative sections of his
[1935] 1946 coauthored secondary-school text on geometry. The main portion of the
present book concludes with chapter 14, which describes some of Tarski’s activity during
the 1930s that is background for several translations in the supplement, and leads to his
1939 voyage to the New World.

Part Four, the supplement, begins with chapter 15, which consists of translations of
the eleven remaining works of Tarski that until now were accessible only in Polish. They
are about various subjects not closely related to the earlier chapters, and are all very
short. Each is accompanied by a brief discussion that places it in context and renders it
intelligible. The most significant one is the report of Tarski’s 1930 presentation in Lwów:
xii Preface

the first appearance in print of his celebrated theory of truth. The book concludes with
chapters 16–18, which contain annotated lists that update the 1986 bibliography of
Tarski’s publications and identify major studies of his life and his work.

The material gathered in this book will increase the accessibility of Tarski’s early work
and explain some of its relationships to the intellectual, political, and social milieu of
Poland between the world wars. The present editors hope that it will spur broader
investigation into the connections between mathematics and its cultural setting during
that era. This would be particularly welcome for the connection between mathematical
research and mathematics education, as displayed by Tarski’s research on geometry and
his practice both in teaching secondary-school students and training secondary-school
teachers. This hope is a major reason for including works of such contrasting mathe-
matical sophistication in a single volume of selected translations.

Rough maps on pages xxii and xxiii depict Poland and the surrounding area, with the
international borders of 1914 and 1924—before and after the First World War. Cities
of importance to this book have been identified. For place names in this region, English
versions are used when available. Otherwise, for places that have been in more than one
country, the names used in this book are the official names used in 1924.

In 1936 Poland officially adopted some changes in spelling that affect words quoted
in this book, particularly involving the letters i, j, and y. The editors have tried to adhere
to the spellings in the original texts.

The translations are meant to be as faithful as possible to the originals.5 Bibliographic

references and personal names have been adjusted to conform with conventions of the
present book. Some uses of alternative type styles for emphasis, enunciations, and per-
sonal names have been modified. The only intentional modernizations are punctuation
and occasional changes in symbols, where Tarski’s conflict with others used throughout
this book. Those are discussed in the introductions to the individual translations. As an
aspect of adjusting punctuation, the editors modified the use of white space to enhance
visual organization. All [square] brackets in the translations enclose editorial comments.
These are inserted, usually as footnotes, to document changes in technical terms, to note
or suggest corrections for occasional errors in the originals, to clarify possibly troublesome
translation details, and to explain a few passages that seem opaque.

Polish surnames often have gender-specific suffixes. Thus, the wife or unmarried
daughter of Tarski is surnamed Tarska; the wife and the unmarried daughter of uka-
siewicz are surnamed ukasiewiczowa and ukasiewiczówna, respectively. Married
women often use these names hyphenated with those of their husbands. The order varies;
in this book, the husbands’ surnames come second.

5
The introduction by Magda Stroiska and David Hitchcock to Tarski [1935] 2002 was helpful in planning
the Polish translation process.
Preface xiii

In this book, the term gimnazjum refers only to Polish secondary schools that prepared
pupils for university studies during 1900–1939.

Spaces and diacritical and punctuation marks have been ignored during all alphabet-
izations, particularly for the bibliography and index. Warning: this resulted in alphabet-
ization different from the Polish standard! Throughout the book, capitalization of Polish
titles reflects the conventions of the Chicago 1993 manual, which disagree with those now
taught in Polish schools.

The huge bibliography lists all and only works referred to in the book. Each entry
indicates where citations occur. The author–date system is employed for citations: for
example, Tarski 1986a is a citation for a work published under Tarski’s name in 1986.
The present book mentions more than one author named Tarski; citations that include
this surname alone are references to Alfred Tarski. Sometimes an author is to be inferred
from the context, so that a date alone may also serve as a citation of a work.

Biographical information about more than sixty individuals involved with Tarski is
presented in boxes or notes located in various chapters and cross-referenced, as appropri-
ate, in others. The emphasis is on the years before 1945.6 Sources for the data are
identified by footnotes inside the boxes. The book’s index lists both subjects and persons.
The latter entries include personal dates when known.

The project culminating in the present book started with brief conversations, years
apart, between James T. Smith and Steven R. Givant at Berkeley logic colloquia. Smith
recalled that during 1965–1970, he had heard Tarski speak eloquently to general audi-
ences about the degree of equidecomposability of polygons. Although that material had
been published, it was nevertheless virtually inaccessible. Smith mentioned this at the
fall 2007 meeting of the State of Jefferson Mathematics Congress. Joanna and Andrew
McFarland were also attending. Joanna is a teacher of Polish from Pock, Poland.
Andrew, for whom Polish was also a first language, was on the mathematics faculty at
Sonoma State University. The three discussed the possibility of working together to
republish the equidecomposability papers. The critical resources for collaboration had
converged: interest in the project and facility with logic, mathematics, English, and
Polish. Later, Smith and Givant mused that a few other works from Tarski’s early years
had suffered the same neglect, and concluded that publishing a volume of translations
from Polish might be feasible. The McFarlands, Smith, and Springer editor Ann Kostant
discussed this further. It was decided to include translations of representative sections
of the virtually forgotten [1935] 1946 secondary-school text Geometrja by Tarski and two
coauthors. With these translations all of Tarski’s geometric work would be accessible in
English except the famous 1924 paper, in French, by Banach and Tarski on set-theoretic
equidecomposability of geometric figures. This book’s selection of translations was
rounded out by including that paper as well. Investigation of the background and impact
of these works suggested the utility of including an update of the 1986 Tarski

6
Many of these individuals participated in the Polish underground or clandestine educational system during
World War II. The famous 1944 book by Jan Karski is a gripping first-hand description of those activities.
xiv Preface

bibliography. Springer Science+Business Media (Birkhäuser) agreed to this plan in

October 2009.

For the present book, translations from Polish were drafted by the McFarlands, and
edited jointly with Smith. Translations from other languages were done by Smith. The
McFarlands carried out library and archival research in Poland, particularly on Tarski’s
teaching. Smith did that for material accessible in libraries in the United States. Smith
assembled the background information into the present organization, and the result was
edited jointly. Smith was responsible for the design and composition of the book.

The editors wish to acknowledge professors Edith Mendez and Elena A. Marchisotto
for inspiration to undertake historical studies, and Helen M. Smith for her patience,
generosity, insight, and ingenuity. Alfred Tarski’s son, Prof. Jan Tarski, is especially
recognized for his assistance to the present editors, and for his editorial work on other
publications listed in chapters 16 and 17. On matters of content, the editors are grateful
for the advice and assistance of Sheldon Axler, John Corcoran, Stanisaw Domoradzki,
Steven R. Givant, Jacek Juliusz Jadacki, Anna Jaroszyska-Kirchmann, Andrzej Jerz-
manowski, Anna Kozowska, Witold Kozowski, Renato Lewin, Paulo Mancosu, Antony
Polonsky, V. Frederick Rickey, Janusz Rudziski, Andrzej Schinzel, James R. Shilleto,
and Jan Zygmunt. For help with translation, we are indebted to Grayna Ula Furman,
Arek Goetz, Sergei Ovchinnikov, and Michael Thaler. We are immensely thankful for
the library services provided by the Archiwum Akt Nowych, Archiwum Polskiej Akademii
Nauk, Archiwum Pastwowe Miasta Stoecznego Warszawy, Archiwum Pastwowe w
Lublinie, Archiwum Uniwersytetu Warszawskiego, Biblioteka Narodowa, Centralna
Biblioteka Matematyczna Polskiej Akademii Nauk, Biblioteka im. Zieliskich
Towarzystwa Naukowego Pockiego, Ksinica Kopernikaska, Narodowe Archiwum
Cyfrowe, the United States Holocaust Memorial Museum, the University of California
libraries in Berkeley and Richmond, and for the splendid interlibrary loan services of San
Francisco State University. This book was made possible by the retirement system of that
university.
 Contents
Foreword, by Ivor Grattan-Guinness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix

Illustrations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xix

Maps of Central Europe . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxii

Part One: Debut

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

1 School, University, Strife . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.1 Coming of Age in Warsaw . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2 Well-Ordered Sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15

2 A Contribution to the Axiomatics of Well-Ordered Sets (1921) . . . . . . . . . . . 19

3 Doctoral Research . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Part Two: Geometry

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

4 Area, Volume, Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

4.1 Area and Volume . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.2 The Measure Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
4.3 Equidecomposability in the Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
4.4 The Banach–Tarski Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64

5 On the Equivalence of Polygons (1924) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77

5.1 §1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
5.2 §2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83
5.3 §3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84
5.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90
xvi Contents

6 On Decomposition of Point Sets into Respectively

Congruent Parts (1924) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93
6.1 General Properties of Equivalence by Finite or
Denumerable Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
6.2 The Fundamental Theorems about Equivalence by
Finite Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
A. Euclidean Space with One or Two Dimensions . . . . . . . . . . . . . . . . . . . . 106
B. Euclidean Space of Three (and More) Dimensions . . . . . . . . . . . . . . . . . . 109
C. The Surface of a Sphere . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112
6.3 The Fundamental Theorems about Equivalence by
Denumerable Decomposition . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115

7 Degree of Equivalence of Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
7.2 The Degree of Equivalence of Polygons (1931) . . . . . . . . . . . . . . . . . . . . . . . . 135
7.3 Moese’s Contribution (1932) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
7.4 Remarks on the Degree of Equivalence of Polygons (1932) . . . . . . . . . . . . . . 151

8 Research Threads . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

8.1 Two Threads in Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 160
8.2 Thread: Well-Ordering and Finiteness . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
8.3 Thread: Cardinal Arithmetic and the Axiom of Choice . . . . . . . . . . . . . . . . 161
8.4 Thread: Application of Set Theory to Geometry . . . . . . . . . . . . . . . . . . . . . . 162
8.5 Equidecomposability in Elementary Geometry . . . . . . . . . . . . . . . . . . . . . . . 162
8.6 Equidecomposability in Set Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
8.7 The Measure Problem . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
8.8 Generalizing Cantor–Bernstein . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167

Part Three: Teaching

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169

9 Career and Family . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

9.1 Employment and Marriage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171
9.2 Teaching Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 179
9.3 Teaching Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 184
9.4 Research Seminars . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 194
9.5 Warsaw/Pozna Congress, 1929 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 200
9.6 Organizations and Journals for Teachers . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
9.7 Parametr and Mody Matematyk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204
9.8 Tarski’s Contributions to Parametr and Mody Matematyk . . . . . . . . . . . . . 211
9.9 Tarski’s Coauthored Textbook Geometrja . . . . . . . . . . . . . . . . . . . . . . . . . . . 214
Contents xvii

10 Congress of Mathematicians of Slavic Countries (1929) . . . . . . . . . . . . . . 225

11 Circumference of a Circle (1932) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 229

12 Exercises Posed by Tarski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243

12.1 Diluting Wine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249
12.2 Iteration of the Absolute Value Symbol . . . . . . . . . . . . . . . . . . . . . . . . . . 250
12.3 Decomposition into Factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
12.4 Interesting Identity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253
12.5 Exercise 118 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255
12.6 Equidistant . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 256
12.7 System of Inequalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
12.8 Another System of Inequalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 259
12.9 Cutting a Rectangle Out of a Square . . . . . . . . . . . . . . . . . . . . . . . . . . . . 261
12.10 The Degree of Equivalence of Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . 263
12.11 Postulate about Parallels . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264
12.12 Analytic Geometry of Space . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
12.13 Stereometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 268
12.14 Arranging Two Segments in a Plane . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271

13 Geometry for the Third Gimnazjum Class (1935) . . . . . . . . . . . . . . . . . . . . . 273

13.1 §1 Circle and Disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
13.2 §2 Construction Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
13.3 §3 Relative Positions of a Circle and a Line Lying in a Plane . . . . . . . 281
13.4 §4 Construction Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
13.5 Summary of §5–§24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
13.6 §25 Incommensurable Segments—Irrational Numbers . . . . . . . . . . . . . 292
13.7 §26 Measuring Incommensurable Segments with a Unit . . . . . . . . . . . . 296
13.8 §27 Rational Approximations of Irrational Numbers
—Operations on Irrational Numbers . . . . . . . . . . . . . . . . . . . . . . . . 297
13.9 Summary of §28–§32 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
13.10 §33 Equivalent Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
13.11 §34 On Measuring Areas of Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
13.12 §35 Areas of Similar Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316

14 Teaching and Logic — to the New World . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319

14.1 Teaching . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319
14.2 Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321
14.3 To the New World . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 330
14.4 Epilogue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 336
xviii Contents

Part Four: Supplement

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337

15 Assorted Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 339

15.1 Discussion of ukasiewicz 1925, on Understanding Deduction . . . . . . . 340
15.2 Discussion of Greniewski [1927] 1928, on Action . . . . . . . . . . . . . . . . . . 341
15.3 Discussion of Czeowski [1927] 1928, on Causality . . . . . . . . . . . . . . . . . 343
15.4 ukasiewicz 1928–1929, on Definitions . . . . . . . . . . . . . . . . . . . . . . . . . . 346
15.5 Tarski 1929, on Polish Pensions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
15.6 Tarski 1930–1931, on the Concept of Truth . . . . . . . . . . . . . . . . . . . . . . 356
15.7 Tarski 1932, on Banach’s Measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 363
15.8 Discussion of Ajdukiewicz 1936, on Idealism . . . . . . . . . . . . . . . . . . . . . . 365
15.9 Discussion of Wilkosz 1936, on the Significance of Logic . . . . . . . . . . . . 367
15.10 Discussion of Zawirski 1936b, on Synthesis . . . . . . . . . . . . . . . . . . . . . . . 370
15.11 Discussion of Kokoszyska 1936a, on Relative Truth . . . . . . . . . . . . . . . 374
15.12 Two Letters to Sierpiski, 1946–1947 . . . . . . . . . . . . . . . . . . . . . . . . . . . 376

16 Posthumous Publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385

16.1 Papers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
16.2 Monographs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
16.3 Letters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392

17 Biographical Studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399

18 Research Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411

18.1 1986, 1988 JSL Surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
18.2 Other surveys . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 413

Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421

Permissions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 473

Index of Persons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 477

Index of Subjects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 485

 Illustrations
Portraits section, page
Alfred Tarski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . frontispiece
Stefan Mazurkiewicz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 6
Zygmunt Janiszewski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 6
Tadeusz Kotarbiski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 6
Leon Trotsky as the Devil . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 11
Alfred Teitelbaum (Tarski) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 13
Stanisaw Leniewski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 13
Kazimierz Twardowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.2, 18
Kazimierz Pasenkiewicz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 32
Aleksander Jaboski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 34
Mordchaj Wajsberg . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 34
Kazimierz Ajdukiewicz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 37
Wacaw Sierpiski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1, 48
Stefan Banach . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2, 56
Felix Hausdorff . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4, 68
Kazimierz Kuratowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4, 72
Henryk Moese . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1, 131
Maria Tarska and Others in the 1919–1920 Polish–Soviet War . . . . . . . . . . . 9.1, 175
Ugo Amaldi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2, 182
Federigo Enriques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.2, 182
David Hilbert . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3, 185
Mario Pieri . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.3, 187
Jan ukasiewicz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4, 196
Mojesz Presburger . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4, 196
Andrzej Mostowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4, 198
Wanda Szmielew . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.4, 198
Antoni Marian Rusiecki . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.7, 209
Stefan Straszewicz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.7, 209
Zygmunt Chwiakowski and Others . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9, 218
Wacaw Schayer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9, 220
Maria Kokoszyska and Alfred Tarski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2, 327
Leon Chwistek . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2, 329
Janina Hosiassonówna . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 335
Adolph Lindenbaum . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 335
Tadeusz Czeowski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.3, 345
Witold Wilkosz . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.9, 369
Zygmunt Zawirski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15.10, 373
Alfred Tarski . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch.16, 384
xx Illustrations

Figures

Many figures are not titled, but numbered as in the original versions of translated
material. For example, figure 1 in chapter 2 of this book is adapted from figure 1 in the
original version of Tarski [1921] 2014.
section, page
Alfred Tarski Signature . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . frontispiece
Central Europe in 1914 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxii
Central Europe in 1924 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xxiii
Alfred’s Autumn 1919 –Winter 1920 Enrollment Record . . . . . . . . . . . . . . . . 1.1, 6
1920 Propaganda (3 figures) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1.1, 11
Przegld Filozoficzny Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 2, 20
Figure 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 2, 28
Figures 2–3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 2, 29
To the Bursar, Autumn 1923 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 35
Three Signatures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 3, 38
Przegld Matematyczno-Fizyczny, Masthead . . . . . . . . . . . . . . . . . . . . . . . . . 4.3, 58
Hausdorff’s Classification of Rotations in G . . . . . . . . . . . . . . . . . . . . . . . . . . 4.4, 68
Przegld Matematyczno-Fizyczny, Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch. 5, 78
A Square and an Isosceles Right Triangle with the Same Area . . . . . . . . . . . 5.3, 85
Fundamenta Mathematicae 6 (1924), Cover and Table of Contents . . . . . . . . ch. 6, 94
Parallelograms with the Same Base and Altitude . . . . . . . . . . . . . . . . . . . . . . 7.1, 128
Adaptation of Wojtowicz’s Argument . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.1, 128
Gimnazjum in Kpno, Where Henryk Moese Taught . . . . . . . . . . . . . . . . . . . 7.1, 131
Dividing a Square into Two Polygonal Regions . . . . . . . . . . . . . . . . . . . . . . . 7.1, 134
Figures 1 to 4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2, 136
Figures 5 to 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.2, 141
Figure 1 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3, 145
Figure 2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3, 146
Figure 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3, 148
Figures 4 to 6 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3, 149
Figure 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7.3, 150
Warsaw, 1925: 140–150 Marszakowska Street . . . . . . . . . . . . . . . . . . . . . . . 9.1, 173
51 Koszykowa Street . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1, 178
4 Sukowski Street . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.1, 178
Hilbert’s Foundations of Geometry: Title Page . . . . . . . . . . . . . . . . . . . . . . . 9.3, 185
Pieri’s Point and Sphere Memoir: Polish Translation . . . . . . . . . . . . . . . . . . 9.3, 187
Tarski’s Teaching Load at the University of Warsaw (table) . . . . . . . . . . . . . 9.3, 191
First Congress of Mathematicians of Slavic Countries (group portrait) . . . . 9.5, 201
General Polish Exposition, Pozna, May–September 1929 . . . . . . . . . . . . . . . 9.5, 202
Cover and Contents, Mody Matematyk 2 (1931, issue 3) . . . . . . . . . . . . . . . . 9.7, 206
Masthead and Contents, Parameter 2 (1931, issues 8–10) . . . . . . . . . . . . . . . 9.7, 206
Illustrations xxi

section, page
Advertisement for Geometrja . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9.9, 215
Figure 1: Graph of Function g ” log g n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.4, 254
Figure 2: Equidistant Curve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.6, 256
Figure 3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.7, 258
Figures 4 and 5 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.9, 262
Figures 6 and 7 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.11, 266
Figure 8: Tarski’s Parallel Postulate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.11, 266
Figure 9 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.13, 270
Geometrja Table of Contents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch.13, 275
Geometrja Cover . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ch.13, 276
Figure 1, for §1 * . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.1, 277
Figure 2, for §2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2, 279
Figure 3, for §2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.2, 280
Figures 4 and 5, for §3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3, 281
Figure 6, for §3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3, 282
Figures 7 and 8, for §3 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.3, 284
Figures 9 and 10, for §4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4, 286
Figure 11, for §4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4, 287
Figure 12, for §4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4, 288
Figure 13, for §4 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.4, 289
Figure 75, for §25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.6, 292
Figure 76, for §25 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.6, 294
Figure 77, for §27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8, 298
Figure 78, for §27 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.8, 299
Figures 106 to 110, for §33 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.10, 303
Figures 111 and 112, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 305
Figures 113 to 116, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 306
Figures 117 and 118, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 308
Figures 119 and 120, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 309
Figure 121, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 310
Figures 122 to 124, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 311
Figure 125, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 312
Figure 126, for §34 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.11, 315
Figures 127 to 130, for §35 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13.12, 317
Fencing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.2, 329
Headlines from the Völkischer Beobachter . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 330
Gdynia–America Poster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 332
M/S Pisudski Menu . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 332
Arrival in New York . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14.3, 332

*
This and the following § numbers refer to sections of Geometrja.
xxii Illustrations

—————
300 km
(186 mi) R
St. Petersburg

Moscow •

$$$$$$$
Gdynia• $$$
$$$$$$$$ $$ •Vilnius
•Königsberg
•Minsk

Berlin Pozna
$$$$$$
•Pock
$$$$$$$$$
Cambridge $$$$$$$$$$
•Amsterdam R • • Warsaw RUSSIAN EMPIRE
•
R • Breslau •ódz •Lublin
$$$$$$$
London Göttingen •
•Kiev
•Prague •
GERMAN #$$$
• Cracow •Lwów
Zakopane
R Paris EMPIRE
R
Vienna Budapest
•
AUSTRIAN EMPIRE

Bologna
•

Central Europe in 1914

Illustrations xxiii

—————
300 km
(186 mi) •
St. Petersburg

Moscow R
LITHUANIA

$$$$$$$
Gdynia• $$$
$$$$$$$$ $$ •Vilnius
•Königsberg
•Minsk
POLAND
Berlin Pozna
$$$$$$
•Pock
$$$$$$$$$
Cambridge $$$$$$$$$$
•Amsterdam R • R Warsaw SOVIET
•
R • Breslau •ódz •Lublin UNION
$$$$$$$
London Göttingen •
Cracow •Kiev
GERMANY R Prague • $$$$
• Zakopane•Lwów
R Paris
R
Vienna
R
AUSTRIA Budapest

Bologna
•

Central Europe in 1924

 Part One
Debut
These first three chapters describe Alfred Tarski’s childhood, schooling,
and university studies during a time of political chaos and threats of war.
Chapter 1 includes background for his first paper, A Contribution to the
Axiomatics of Well-Ordered Sets, published while he was still a student.
The paper is translated in chapter 2. Chapter 3 provides background for
Tarski’s doctoral research, supervised by Stanisaw Leniewski.

Tarski’s life in Poland unfolded amid a chaotic vortex of political,

social, economic, and scientific developments. Accounts of events, people,
and ideas were merged from several dimensions to form the linear
sequence of pages of this book. That is reflected in the background
chapters 1 and 3 by the use of boxes interspersed in the main narrative.
They contain biographical sketches of some persons associated with
Tarski, and informational essays about several other topics. Each box
can be read independently: readers are not expected to visit them in
sequence. Cross-references refer to them from the main narrative.
 1
School, University, Strife

The first section of this chapter describes Alfred Tarski’s childhood, schooling, and initial
university studies in Warsaw, the city that in 1918 became the capital of the new inde-
pendent Polish republic. The second section provides background for his first publication,
a paper on set theory and logic that he completed while still a student. That paper is
translated in the next chapter.

1.1 Coming of Age in Warsaw

Alfred Tarski was born Alfred Teitelbaum in January 1901 in Warsaw, Poland, an
oppressed part of the Russian Empire.1 His father, Ignacy, or Izaak, stemmed from War-
saw; his mother, Róa Prussak, from ód. Their families were engaged in the lumber
and textile businesses. The Teitelbaums had two children: Alfred’s brother, Wacaw, was
two years younger. The family lived at 51 Koszykowa Street, apartment 14, on the second
floor of a five-story building that they owned.2 It was located in the center of the city,
about three kilometers from the University of Warsaw. From September 1910 to sum-
mer1915, Alfred attended the State Gimnazjum 4 in Warsaw.3 The family was Jewish,
but secular in outlook. The language at home was Polish; in school, Russian. Alfred was
precocious, particularly in languages: he studied French, German, Latin, and Greek at
school, and after school went to synagogue to learn Hebrew. Alfred translated a German
story into Polish at age twelve as an anniversary gift to his parents. Even earlier, he had
shown interest in politics and social justice.

1
For biographical information supporting this section, unless another source is cited, consult the biographies
by Anita B. and Solomon Feferman (2004) and by Jacek Juliusz Jadacki (2003a). Also note the descriptions
of those works in chapter 17. The Fefermans emphasized personal-interview sources; Jadacki, published
records. For historical and sociological information, consult the works by Norman Davies (1982, volume
2, chapters 18–19), Celia S. Heller (1994), and Richard M. Watt (1979).
2
The spellings of the names are from Tarski [1918] 2014, translated in section 16.4. Jadacki (2003a, 143)
verified ownership from 1930 data. Jadacki and the Fefermans (2004, 8) claimed that 15 Koszykowa had
been destroyed in the 1940s. But it still stands as 51a Koszykowa: see Goliska, Porbska, and Srebrny
2009a and 2009b. It is pictured on page 176 of the present book.
3
Tarski [1918] 2014: rzd gimnazyum. A Polish secondary school that prepared students for eventual
university study was called a gimnazyum or gimnazjum. The latter term is used in this book to avoid
confusion with various types of secondary schools of other times and places.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 3
DOI 10.1007/978-1-4939-1474-6_1, © Springer Science+Business Media New York 2014
4 1 School, University, Strife

For more than a century, Poland had been partitioned between Russia on the east, Ger-
many to the west, and Austria to the south. Thoughts of Polish unification and independ-
ence had long spawned agitation and intrigue. World War I broke out in August 1914 for
other reasons. Its conclusion four years later resulted in a unified Polish republic, but
almost by accident. Many Polish military organizations were formed in 1914, and within
two years, nearly two million Polish men were involved. Poles had enjoyed greater free-
dom in the Austrian partition, and the Germans made promises, so most of the Polish
armies served, under the leadership of Józef Pisudski, with Austria and Germany against
Russia. Russia’s allies Great Britain and France regarded the question of Polish inde-
pendence as an internal Russian matter. Pisudski played each side against the other.
International posturing and intrigue increased in intensity. Amid this clamor, Alfred
became a Polish nationalist.

In September 1915 Alfred transferred to the small, elite Mazowieckie Gimnazjum.4 Its
faculty were highly educated scholars, including two with doctorates in philosophy from
the University of Lwów: Stefan Frycz and Bogdan Nawroczyski. Alfred’s favorite sub-
ject was biology; his teacher, Stanisaw Przylecki, had recently earned a medical degree
in Zurich. Alfred’s brilliance impressed both his teachers and his fellow students. That
summer, the Germans entered Warsaw, which they would occupy until the end of the war.
They instituted a number of reforms immediately, including permission for Alfred’s school
to switch its instruction from Russian to Polish.

Since 1870 the University of Warsaw had functioned as a Russian institution, serving
the Imperial establishment. Most Polish students had to attend university abroad. After
student boycotts and the outbreak of the war, the Russians closed the university alto-
gether in 1914 and moved its faculty back to the homeland. The German occupation
supported its autumn 1915 reopening as a Polish university. The new Polish faculty were
assembled from various institutes in Warsaw, from universities in the other partitions
of Poland, and from exile abroad. The university expanded rapidly during the war years,
from one thousand to more than four thousand students. Most belonged to the urban
middle or upper class, from central Poland; about 75% had to work to offset expenses.
About 65% were male; and about 25%, Jewish. During that time the philosophical faculty
grew to about forty, of all ranks.5 According to the mathematician Kazimierz Kuratowski,
who was a student in Warsaw at that time,
... the restored institutions of higher education were ... a fulfilment of the dreams of
many generations, the attainment of the goal of a persistent struggle for Polish educa-
tion. Therefore, beside young students in classes one could see adult representatives
of the Warsaw intelligentsia, for whom a direct contact with the restored Polish uni-
versities and colleges was a deep emotional experience.
... the atmosphere in which the institutions of higher education in Warsaw began their
work ... released a great creative potential ... which produced a surprising development
in many branches of science, including Polish mathematics.

4
Tarski [1918] 2014: Szkoa Ziemi Mazowieckiej.
5
Garlicki 1982, 49, 53, 314–315, 343. See also Manteuffel 1936, 156–175.
1.1 Coming of Age in Warsaw 5

Two mathematicians, the topologists Zygmunt Janiszewski and Stefan Mazurkiewicz,

gathered several others, including Samuel Dickstein, Stefan Kwietniewski, and the logi-
cian Jan ukasiewicz, to start building the mathematics faculty.6

At the Mazowieckie Gimnazjum, Alfred prepared to enter the university. He was

graduated in June 1918 with “excellent, 5” marks in all subjects covered that year:
Polish German History Logic Mathematics Physics
Latin French Civics Hygiene Cosmography

His mathematics courses had included analytic geometry but not calculus. Two of his
nine classmates were too ill to complete the year. Typhus was raging, due to the huge
influx of refugees into Warsaw, overcrowding, and poor sanitation. For example, 775 new
cases were reported in the city during the week ending 23 February. Nearly three-fourths
of these were among the Jewish population; the fatalities, about 9% of the victims, were
most common among those over forty, and twice as prevalent among Christians.7

Alfred’s graduation picture, on page 13, shows his intensity. In October 1918, he
enrolled in a broad university curriculum, intending to concentrate in biology.8 The war
ended officially in November. The Germans released Pisudski, whom they had impris-
oned for a year when he stopped cooperating. Immediately, he assumed command of the
Polish government in Warsaw. But armed conflict continued in the German part of
Poland to the west. The university suspended classes for the academic year 1918–1919
and urged all its students to join the Polish army; about half did so. Alfred was declared
unfit for service, but his studies were interrupted by the continuing Polish struggle for
independence. The present editors do not know what occupied Alfred during the rest of
1918–1919.9

In 1918, Janiszewski published a plan for realizing in the discipline of mathematics

the great creative potential that had stemmed from the fulfilment of dreams for restored
institutions. According to Kuratowski,
One of the principal means suggested ... for attaining that end was the concentra-
tion ... in a relatively narrow field of mathematics ... one in which Polish mathemati-
cians had common interests and ... achievements which counted on a world scale. This
field comprised set theory together with topology, and the foundations of mathematics
together with mathematical logic.

6
Kuratowski 1980, 20, 25–26, 28. ukasiewicz served in the university administration (ukasiewicz[1953]
1994, 133). For portraits and biographical sketches of Janiszewski and Mazurkiewicz, see pages 6 and
14; for ukasiewicz, see section 9.4; a biographical sketch of Kwietniewski is in section 9.3.
7
Szkoa Ziemi Mazowieckiej [1918] 1927. Jadacki 2003a, 141. The typhus report Goodall 1920 presents
a vivid account of conditions in Warsaw. Tuberculosis was a comparable threat (Wynot 1983, 340).
8
Tarski [1918] 2014, translated in section 16.4. Tarski 1924f.
9
Garlicki 1982, 341; Manteuffel 1936, 26–28. The archive document Warsaw 1918–1919 has top and bot-
tom parts. The top, from the university secretariat, 8 November 1918, certified that Alfred was student
number 2909 of the Philosophical Faculty (Wydzia) and said that he should give it to the military and
return it when he was released from service. The bottom, from a military doctor on 5 February 1919, said
“By reason of §1D and age, the requirement not applying, [Alfred] was declared unfit for military service.”
 Alfred’s Autumn 1919 –Winter 1920 Enrollment Record

Stefan Zygmunt Tadeusz

Mazurkiewicz Janiszewski Kotarbiski
around 1930 around 1915 in 1933
1.1 Coming of Age in Warsaw 7

Janiszewski, ukasiewicz, and Mazurkiewicz would undertake that development. They

were joined by the Polish mathematician of greatest world note, who had come to Warsaw
that year: Wacaw Sierpiski, a specialist in set theory.10

Alfred must have caught a glimpse of this new direction during the disrupted year
1918–1919. Reenrolling for the academic year 1919–1920, he would continue some of his
general scientific studies. But his new emphasis was mathematics and logic.11 The atmo-
sphere was electric:
... students could be met more often at political rallies and meetings than in university
lecture rooms or laboratories.12
Withstanding the distraction, Alfred signed up for thirty-one hours of classes per week.
The corresponding entries of his enrollment record are displayed on the facing page.13
They show that Alfred attended
lecture/exercises courses by
• Mazurkiewicz on differential calculus
• Janiszewski on analytic geometry, and
• Sierpiski on set theory;
lectures by
• Sierpiski on determinants and linear equations;
exercises with
• Stanisaw Leniewski on foundations of mathematics; and
lectures by
• Stefan Piekowski on experimental physics,
• Tadeusz Kotarbiski on elementary logic, and
• Leon Petraycki on sociology.
The stamp near the bottom of the left-hand page is the bursar’s receipt for payment of
76 Polish marks. Alfred was charged no laboratory fees, and only for the nineteen hours
of lectures checked in the third column. (Footnotes will lead to biographical sketches and
portraits of many of Alfred’s teachers.)14

10
Janiszewski’s proposal ([1918] 1968) was published by the Mianowski Fund, which supported many aca-
demic activities described in this book. For more information about the fund, consult a box in section
9.3. Kuratowski 1980, 29–31. For a portrait and biographical sketch of Sierpiski, see section 4.1.
11
Givant (1991, 28) mentioned this change but presented no background for its occurrence in 1918–1919.
12
Garlicki 1982, 341.
13
Tarski 1924f. The legend at the top reads “Semestr zimowy. Roku akad. 1919/20.” When Alfred’s enroll-
ment booklet was issued in 1918, academic years consisted of winter and summer semesters. In 1919–
1920, the university converted to three trimesters: autumn, winter, summer ( jesie, zima, letni). Its
documentation placed data for the first two trimesters in the space for the former winter semester. The
headings identify columns for lecturers’ names, lecture titles, hours, tuition, bursar’s certification, and
lecturers’ signatures and dates to certify enrollment and attendance. Summer data are on the following
pages. Jadacki (2003a, 142) reported some of this information, but inaccurately.
14
Biographical sketches of Kotarbiski and Leniewski are on page 9; portraits, on pages 6 and 13.
8 1 School, University, Strife

During the summer trimester in 1920, Alfred continued with twenty-nine hours of
classes per week: with Mazurkiewicz on differential and integral calculus and analysis;
with Sierpiski on number theory and measure theory; with Leniewski on foundations
of set theory; with Piekowski on physics, with laboratory; with Kotarbiski on logic
and Francis Bacon’s methodology; and in his philosophy seminar. Alfred paid a labora-
tory fee of 10 Polish marks and 116 Polish marks for tuition.

The German withdrawal, the Russian Revolution, the collapse of Austrian hegemony,
various independence movements, and the internal politics of the new Polish nation
brought about strife and chaos in Eastern Europe. Polish armies fought six wars in three
years. Eventually the Polish and Soviet armies confronted each other, under military and
political control of Pisudski and Leon Trotsky. (See the box on page 10.) Poland was
immersed in propaganda to stir up support and recruit troops for the war. A million Poles
were marshalled along the front. Battles moved back and forth for months. In summer
1920 a huge Soviet advance threatened the very outskirts of Warsaw. Upper-class citizens
left town, the rest were mobilized but remained strangely calm, and
in July, the students again adopted a resolution to join the army en masse ... for the
second time lectures were suspended.15

The posters depicted on page 11 give an idea of the atmosphere in Warsaw. Alfred
served with a military supply and medical unit.16 His professors Leniewski, Mazur-
kiewicz, and Sierpiski had been working with the military on a project that had decoded
the Soviet army’s communications. With this advantage, the Poles routed the invaders
on 18 August, in a battle known as the Miracle of the Vistula.17 Pisudski’s troops quickly
pushed the invaders out of Poland. This victory brought to a close the chaotic Polish
struggle for unified, independent nationhood. Poles settled down to work, to put into
action the many plans set out during the previous years, and to let Polish culture blossom.

Alfred forwarded his 1919 certificate of military leave to the university administration
with a note signed “Alfred Tajtelbaum.” It seems to be his first recorded use of that Polish
spelling. His patriotic fervor might have been damped somewhat had he realized the
extent to which the war propaganda had heated Poland’s antisemitism. The new Polish
government actually imprisoned some units of volunteers from Jewish communities.18

15
N. Davies 1982, volume 2, 294–295. Garlicki 1982, 341.
16
Feferman and Feferman (2004, 26) reported that Alfred’s service was in 1918, but the earlier account in
this section contradicts that. Requesting enrollment in autumn 1919, Alfred submitted the first form
in Tarski 1919–1920. He described two attachments: his certificate of military leave with a clause
indicating unsuitability for military service, and a certificate recognizing community service. In autumn
1920, Alfred submitted the second form, indicating that a certificate of military service was attached.
17
Nowik 2004–2010, volume 1, 20–27, 39, 231–233, and volume 2, 77 and section IX.5. Volume 1 also serves
as a guide to the confused historiography of this subject, and volume 2 displays numerous signed examples
of decoding. See also Czy 1990, Peposki 1995, 42–43, and the overview Bury 2004. (Much information
about the 1920 war became accessible to the public only after the Communist regimes fell around 1990.)
According to Gazeta Warszawska 1920, Kotarbiski and ukasiewicz also volunteered for military service.
For more information about the professors, see boxes on pages 9 and 14 and in sections 4.1 and 9.4.
18
Tarski 1919–1920. Jadacki 2003a, 143. Polansky 2012, 53–54; Heller 1994, 51 and 314.
1.1 Coming of Age in Warsaw 9

Tadeusz KotarbiĔski was born in 1886 in Warsaw, then a part of the Russian Empire. His father
was a professional artist; both parents were musically gifted. After completing school in Warsaw
in 1906, Tadeusz entered the University of Lwów. He studied logic with Jan ukasiewicz, then turned
to philosophy with Kazimierz Twardowski and earned the doctorate in 1912. Kotarbiski then
returned to Warsaw to teach classics in a gimnazjum and lecture on cultural subjects to the public.
He was a cofounder of the Warsaw Philosophical Institute. In 1919 he was appointed professor of
philosophy at the University of Warsaw. He was a major inspiration for Alfred Tarski, and became
a leader of world significance among analytic philosophers. In 1929 Kotarbiski became dean of the
Faculty of Humanities. His activity in liberal political circles increased as well. During World War
II he taught clandestinely. Afterward he married his former student Dina Sztejnbarg-Kamiska, who
had been a neighbor of Tarski’s parents. She became known as the philosopher Janina Kotarbiska.
Kotarbiski served for four years as rector at the University of ód, then during 1951–1961 as
professor of philosophy at Warsaw. During 1948–1978 he was president of the Polish Philosophical
Society, and during 1957–1962, of the Polish Academy of Sciences. Kotarbiski died in 1981.*

Stanisaw LeĞniewski was born in 1888 near Moscow, to Polish parents. His father was a railroad
engineer. His mother died soon after his birth, and his father remarried. Stanisaw was educated
in Siberian schools, German universities, and then the University of Lwów, where he earned the doc-
torate in 1912 under the supervision of Kazimierz Twardowski. He was also inspired by the work
of Jan ukasiewicz. Leśniewski’s dissertation included the germ of much of his later work, in par-
ticular, insistence on extreme precision of language, and distinction between language and metalan-
guage. Around 1913, Lesniewski married the well-to-do Zofia Prewysz-Kwinto; they had no children.
After research, teaching, and socialist political activity in Moscow, St. Petersburg, and Warsaw, Leś-
niewski was awarded the venia legendi by the University of Warsaw in 1919 and appointed to its chair
of philosophy of mathematics. During the next year, with colleagues Stefan Mazurkiewicz and
Wacaw Sierpiski, he helped the Polish army decode Soviet military communications. Their work
was instrumental in the decisive 1920 defeat of Soviet invading forces in the suburbs of Warsaw.
Leniewski initiated several fundamental logical ideas used in the analysis of language—for example,
semantic categories. He became known for his obsessive precision and perfectionism in philosophical
writing and discussion, and his sharp criticism of the work of most others, who did not attain that
standard. His long-time Warsaw colleague Tadeusz Kotarbiski reported these characteristics:
There was either the ultimate yes or the ultimate no, there was an aversion to half-measures,
loathing of pettiness ... a bent for sudden elations, sharp turns, radical breaking off of friendships,
vehement antipathy towards insincere feelings, adherence to principles and intolerance of
exceptions, the talent and tendency to carry on endless discussions ... .
Mazurkiewicz wrote that Leniewski
as a researcher had a high sense of and need for monumentality, and made no [small] contribu-
tions or fragments, but an idiosyncratic system of foundations of logic and mathematics.
That system, whose three parts were called mereology, protothetic, and ontology, is still studied, but
it has never come into common use. Kotarbiski explained that Leniewski’s
teaching was never aimed at a mass audience: on the contrary, he preferred to deal with the most
exquisitely intelligent.
Leniewski’s lectures generally attracted only tiny audiences, and Alfred Tarski was Leniewski’s
only doctoral student. Nevertheless, Leniewski “attached great importance to his university lectures
and he lectured almost entirely about his own work.” Kotarbiski remembered his colleague’s
research habit of walking to and fro deep in thought, all in vaporous clouds of nicotine. Leniewski
died of lung cancer at age 51, in May 1939. Assessing his contribution to science will probably be a
subject of discourse for years to come.†
————————
— —————————
*Kotarbiski 1977, Wo- † For more information, consult the box on page 10 and these works, particularly
leski 1990b, Goliska- the pages indicated: Betti 2004, 272; Jadczak 1993b, 312, 316; Kotarbiski 1966,
Pilarek et al. 2009a 156, 160; Leniewski 1992, xii; Mazurkiewicz 1939; Pasenkiewicz 1984, 4.
10 1 School, University, Strife

Polish –SS oviet War. During 1792–1914 the region called Poland was divided between three empires.
Russia controlled the center and parts in the east, including Warsaw; Austria controlled parts in the
south, including Cracow and Lwów; and Germany controlled parts in the west and north. (See the
1914 and 1924 maps on pages xxii–xxiii.) The eastern areas included many minorities, particularly
Lithuanians, Belarusians, Ukrainians, and Jews.
The eastern campaigns of World War I started with Russian invasions of German and Austrian
territory in August 1914. Within a year, however, the Germans had driven the Russians entirely out
of Poland. The Russian Revolution began in March 1917. The Bolsheviks seized power in October,
and withdrew from hostilities in March 1918. The German and Austrian regimes began to collapse
that summer, and the war officially ended in November 1918. Within a week Poland declared inde-
pendence, and General Józef Pisudski became its chief of state. Thousands of refugees entered the
country from the east, and there was large-scale migration, particularly of Jews, into Warsaw. Polish
armed forces fought within the country to secure internal control, and against Ukrainian and
Czechoslovak armies to secure its boundaries. Even the 1919 Versailles peace treaty left the question
of Polish borders unsettled.
The Polish army recognized the advantages of intercepting and decoding enemy messages, and
recruited university mathematicians to help. Stanisaw Leniewski, Stefan Mazurkiewicz, and
Wacaw Sierpiski played vital roles in that project.
Pisudski aspired to leadership of a community of Slavic border nations: Lithuania, Belarus, and
Ukraine. Strong opponents favored expanding Poland itself into those regions. The Bolsheviks
aspired to restore Russia’s prewar borders and to spread revolution to Germany.
During 1919 Pisudski led the Polish army into the Soviet Union. By May 1920 the Poles controlled
Vilnius, Minsk, and Kiev. These places suffered greatly during 1918–1920, changing hands several
times. Because of the great distances, however, the Poles could not sustain their effort. Moreover,
the British and French stopped supporting Poland. Even though the Bolsheviks still did not have
complete control of their own country, their army, under Leon Trotsky, drove the Poles back, deep
into Poland, in summer 1920. The Soviet army, controlled by political commissars, was highly
effective and disciplined. The high visibility of some Jews among those leaders, especially Trotsky,
inflamed the antisemitism that was rife in Poland.
Warsaw remained surprisingly calm under threat of conquest. The Soviet army penetrated within
twenty-five kilometers to the east and north and reached the Vistula River. The city mobilized totally
to support the Polish army.
Like the Polish army a few weeks earlier, the Soviet armies were overcome by the great distances.
Moreover, the Poles intercepted and understood all Soviet communications. In August 1920, Pisud-
ski’s army used that intelligence to locate and attack a weak point. They routed the invaders and
soon drove them back through Poland and German East Prussia all the way to Lithuania. In Septem-
ber, the Soviet army began to disintegrate, and sued for peace. This great victory, the Miracle of the
Vistula, put an end to Bolshevik ambitions toward the West.
There was no great celebration in Warsaw. Exhausted, Poles just resumed the slow work of construct-
ing their new country. During 1918–1920 its boundaries were established as shown on the 1924 map,
except that Vilnius and its surrounding area did not become Polish until 1922. Antisemitism con-
tinued to rise: Poland even imprisoned some of its own Jewish volunteer units after the war. And
its incorporation of large territories in the East gave it a large population of resentful minorities.
On the facing page are depicted three large colorful 1920 posters. The captions translate their texts.
Similar propaganda was spread throughout Poland to stir up support and recruit troops for the war.
These display common themes. Trotsky is caricatured stereotypically as a bright-red Jewish devil.
The sky above the ruined city on the right burns red, and Bolsheviks are characterized as anti-
Christian. Contrast the red devil with the manly grenadier, advancing the Polish shield to protect
the future of his country! *
——————————————
*For more information see page 8, footnote 17; N. Davies 1982, chapter 13, and 1972; Szczepaski
1995; and Watt 1979, chapter 6. The Trotsky poster’s artist was named Skabowski (Fuchs 1921, 280).
 1920 Propaganda

BOLSHEVIK FREEDOM
_
THE BOLSHEVIKS PROMISED TO
give you peace,
give you freedom,
give you land,
work and bread.
THEY BASELY CHEATED, AND
unleashed a war with Poland.
Instead of freedom, the fist.
Instead of land, requisition.
Instead of work, misery.
_ Instead of bread, hunger.

To Arms! This Is How a Polish Village

Occupied by the Bolsheviks Looks

To Arms! Save the Fatherland!

Always Think of Our Future.
12 1 School, University, Strife

In October 1920 the university reopened, and Alfred returned to his studies, perhaps
even with greater excitement and vigor. He continued in the same vein, with courses from
Leniewski on foundations of arithmetic and on algebra of logic, Mazurkiewicz on analytic
geometry, and with Sierpiski on higher algebra and on set theory. ukasiewicz had
returned to the faculty after serving during 1919 as the first Polish minister of higher
education,19 and Alfred enrolled in his seminars and courses on philosophical logic. It is
possible to discern three intellectual threads emerging from Alfred’s studies during his
first two years at the university: logic, set theory, and measure theory. They would
extend far into his research career. Repeatedly during 1920–1924, Alfred participated
in the seminars of Kotarbiski, Leniewski, and ukasiewicz. These professors shared
a similar background: they were all students of Kazimierz Twardowski, the founder of
the famed Lwów–Warsaw school of logic.20

In 1921, while still a student, Alfred published his first research paper: A Contribution
to the Axiomatics of Well-Ordered Sets. A subtitle indicated that it stemmed from his
work in Leniewski’s seminar.21 Its mathematical content is discussed in the next section,
and it is translated in its entirety in chapter 2. The paper appeared in Przegld filozo-
ficzny, a philosophical journal founded in 1897 and published at that time by the Warsaw
Philosophical Institute, with the support of the Mianowski Fund.22 The journal’s cover
is displayed on page 20. Its table of contents listed Alfred with the surname Tajtelbaum,
with the Polish spelling that he had adopted the previous year.

The journal was devoted to material from the whole field of philosophy. For example,
the other papers published in the same issue with Alfred’s had to do with philosophy of
law, experimental psychology, history of philosophy, and developmental psychology. Their
authors were established scholars, in their thirties; but Alfred was only twenty, just a
student. The next page after the title, not shown here, lists the editors: Twardowski and
Marjan Borowski. Soon after publication, Borowski wrote to Twardowski,
... I inform you—discreetly—that Przegld filozoficzny doesn’t have many papers of any real
worth in the editorial office. Warsaw coryphaei write little, being afraid of Leniewski!—
although the “scourge of God” has also risen upon him in the person of his pupil, Tajtelbaum.23

19
ukasiewicz [1953] 1994, 133. The prime minister was the internationally famous pianist Ignacy
Paderewski.
20
See Tarski 1924f. Givant 1999 describes these and other threads in detail. ukasiewicz earned the
doctorate in 1902; Kotarbiski and Leniewski, in 1912. Alfred’s philosophy teacher at the Mazowieckie
Gimnazjum, B ogdan N awroczyĔski , had also been a student of Twardowski; he later became professor
of education at the University of Warsaw. For biographical information on Twardowski and a portrait,
see page 18.
21
Tarski [1921] 2014. The present editors looked for other papers of that era that might claim such a rela-
tionship to the seminar, but found none.
22
Philosophical Review, Warszawski Instytut Filozoficzny. For information about the Mianowski Fund,
consult a box in section 9.3.
23
Borowski 1922. A coryphaeus is a leader of a dramatic chorus. Borowski had also been a student of
Twardowski.
 Alfred Teitelbaum
(Tarski) in 1918

Stanisaw Leniewski
around 1915
14 1 School, University, Strife

Zygmunt Janiszewski was born in Warsaw in 1888; his father was a financier. After completing
school in 1907 in Lwów, then part of the Austrian Empire, Zygmunt studied at Zurich, Munich,
Göttingen, and Paris. At Zurich he displayed his social talent by organizing a support group for Polish
students. In 1911, he earned the doctorate in Paris with a dissertation on point-set topology super-
vised by Henri Lebesgue. Janiszewski produced several significant works on that subject, particularly
a simplification of the Jordan curve theorem, and authored some influential articles explaining
modern abstract mathematics to educated Poles. Janiszewski taught briefly in Warsaw and Lwów,
then joined Józef Pisudski’s Polish Legion for service during 1914–1915 against the Russians. He
soon left it, refusing to sign an oath of loyalty to Germany and Austria. When the University of
Warsaw reopened as a Polish university in 1915, Janiszewski was selected as one of its initial
professors. His seminal [1918] 1968 report guided development of a world center of mathematical
research in Poland, through specialization in set-theoretic topology, logic, and foundations of
mathematics. He was a principal founder of the journal Fundamenta Mathematicae, devoted to those
subjects. Janiszewski died at age 31 in the influenza epidemic in January 1920. According to his
student Bronisaw Knaster, “For Janiszewski teaching was a mission and the student a com-
rade ... .” Janiszewski donated all his possessions to educational charities.*

Stefan Mazurkiewicz was born in 1888 in Warsaw, then part of the Russian Empire. His father was
a noted attorney. After graduating from secondary school in 1906 in Cracow, Stefan attended
university courses there, in Munich, in Göttingen, and then briefly in Lwów, where he earned the
doctorate in 1913 with a dissertation on area-filling curves, supervised by Wacaw Sierpiski. Mazur-
kiewicz began a very extensive research career in probability theory, topology, and analysis.
In 1915 Mazurkiewicz became the youngest of the founding mathematics faculty of the newly
reconstituted Polish University of Warsaw. He played a major role in developing the Warsaw school
of mathematics, particularly by leading frequent faculty meetings, formal and informal, on current
research work and on strategies for expanding that activity in the future. Alfred Tarski enrolled in
his courses nearly every semester of his Warsaw studies.
Although best known for theoretical results, Mazurkiewicz’s mathematical activity had major applied
aspects. In Göttingen he had earned a diploma on insurance mathematics. In 1919, together with
his Warsaw colleagues Sierpiski and Leniewski, he began work with the Cipher Bureau (Biuro
Szyfrów), the Polish military agency devoted to decoding enemy communications. In 1920, the group
broke the code used by the invading Soviet army, enabling General Józef Pisudski to avert catastro-
phe through a decisive victory at the very approaches to Warsaw. Mazurkiewicz continued working
with the Bureau as a consultant at least until 1930. The Bureau’s later success in breaking German
codes played a major role in World War II.
Mazurkiewicz married twice. In 1920 he reportedly fought a duel with a military officer over an affair
of the heart. After a divorce, Mazurkiewicz’s former wife married Jan Kowalewski, the officer who
headed the Cipher Bureau in 1920. Mazurkiewicz served several terms, about ten years in all, as dean
of the Faculty of Mathematics and Science and as prorector of the university. Nevertheless, he kept
up a stream of research in topology, analysis, and the foundations of probability theory. Most of his
eight or so doctoral students became leading researchers; two of them—Kazimierz Kuratowski and
Bronisaw Knaster—play roles in this book.
During the 1939–1944 German occupation, Mazurkiewicz taught clandestinely and helped his col-
leagues plan the postwar resurgence of Polish mathematics. But he became seriously ill and died in
a hospital near Warsaw a month after the armistice. He had been reconstructing a destroyed manu-
script for a major work on probability theory; it was published a decade later, in 1956.†

————— —————
*Knaster 1973; Kuratowski † Pawlikowa-Broek 1975. Kuratowski 1981, 62, and 1980. See also
1980: 158–163. the references in footnote 17 on page 8, especially Czy 1990.
1.2 Well-Ordered Sets 15

1.2 Well-Ordered Sets

This section is devoted to the technical background and impact of Alfred Tarski’s first
research paper, A Contribution to the Axiomatics of Well-Ordered Sets, which is translated
in its entirety in chapter 2. It was published in 1921 under the surname Tajtelbaum.
Alfred did not change that until 1924. Nevertheless, this section, on his mathematics and
its presentation, will use the name Alfred Tarski.

Tarski’s paper is about the interrelationships between various basic properties of well-
ordered sets. Georg Cantor had introduced those in 1883: the elements of a well-ordered
set should be generated by a process that
• selects some first element, and
• always identifies a unique next element, no matter how many have already
been generated, provided only that it has not yet generated all elements.
That description is somewhat vague, particularly in its allusion to time via the words
“already” and “not yet.” But it reflects how mathematicians of that era considered a set
and its elements simultaneously, and it shows the centrality of the “unique next” idea
in Cantor’s conception. Cantor evolved this notion, replacing that description in 1897
with the notion of a well-ordering for a set M. This should be an antireflexive antisym-
metric transitive binary relation R on M such that
• M has an element m such that m R s for every s  M different from m;
• if M contains an upper bound for a subset S —that is, an element u  M
such that s R u for every s  S —then it contains an upper bound v for S
such that no upper bound u for S satisfies the condition u R v.
It is easy to see that these latter two conditions are equivalent to the one most commonly
used today: M must be nonempty, and each nonempty subset S of M must contain an
element m such that m R s for every s  S different from m.24

Tarski’s methodology was that of postulate theory. During the 1880s and 1890s,
mathematicians had become increasingly aware of the usefulness of applying the axiom-
atic method precisely in presenting a theory:
• stating clearly the notions that will be left undefined,
• defining solely from these primitive notions all other terms to be used,
• stating clearly the principles that will be left unproved, and
• deriving all theorems of interest solely from these axioms and the primitive
notions and definitions.

24
Cantor 1883, §2; [1895–1897] 1952, §12, and note 16. The words “already” and “not yet” in this
paragraph approximate Cantor’s use of Sukzession.
16 1 School, University, Strife

Postulate theory evolved to facilitate investigation of the selection of primitive notions

and axioms and the phrasing of definitions and theorems.25 It stressed their economy,
simplicity, gracefulness, and ease of use. These are often conflicting criteria: for example,
making individual axioms simpler often requires increasing their number, and simpler
primitive notions, like some primitive hand tools, are often awkward to use. Several
applications of postulate theory involved similar details: studying the order of points on
a line, that of the natural numbers, that of infinite ordinal numbers, and other types of
ordering closely related to those. By about 1900, these studies had become the theory of
ordered sets, a coherent part of postulate theory. Tarski’s paper lies there.

Tarski denoted by A1 , A 2 , and A 3 the familiar trichotomy, antisymmetry, and

transitivity axioms that characterize a strong 26 linear ordering R of a set Z. He then
considered the following sentences:

(~U I Z) ( I =
/ U _ (} a  U )(~ u  U ) ¬ u R a ) (B)
(~U I Z) ( I =
/ U _ (} a  U )(~ u  U ) ( u =
/ a _ a R u) ) (C)
(~U I Z) ( I =
/ U _ (}! a  U )(~ u  U) ¬ u R a ) (D)
(~U I Z) ( I =
/ U _ (} a  U )(~ t, u  U ) ( t R a & u R a _ t = u )) 27
(E)
(~U I Z) ( I = / Z _ (} a  U )(~ u  U ) ( u =
/ U= / a _ ¬ u R a )) (F)
Tarski showed that equivalent axiom systems result by adding B, C, E, or F to
{ A1 , A 2 , A 3 }, and those systems characterize well-ordering relations R. Moreover, the
latter two systems—with E or F —are independent, whereas the former two—with
B or C —are not. Finally, he proved that these systems are all equivalent to that
consisting of the single axiom D.

Tarski’s paper was reviewed in detail by Leon Chwistek, a noted artist and philosopher,
in the Jahrbuch über die Fortschritte der Mathematik. The previous paragraph closely
echoes his report. Then a secondary-school teacher of mathematics, Chwistek would later
become professor at the University of Lwów and a leader in Polish logic.28

Tarski’s axiom D is close to Cantor’s original 1883 definition of a well-ordered set.

Cantor did not stress the properties stated by A1 , A 2 , and A 3 , but did emphasize the
existence and uniqueness of the next element to be selected from those not yet generated,
which can be construed as the elements of the set U.

25
Consult Russell 1903 for an overview of early work, and Scanlan 2003 for a retrospective from the point
of view of modern logic.
26
An analogous characterization of a weak linear ordering would employ the transitivity axiom with
“dichotomy” and “weak antisymmetry” axioms. Analogous results would ensue.
27
The symbol }! in D stands for “there exists a unique.” The terminal clause in E following the universal
quantifier ~ says that at most one element t of U satisfies t R a.
28
Chwistek omitted the condition U = / Z in Tarski’s axiom F. For more information on Chwistek, consult
the biographical sketch in section 14.2.
1.2 Well-Ordered Sets 17

The most involved arguments in Tarski’s paper are the proofs that the axiom systems
with E and F are equivalent to the others. Tarski included no references to the literature,
and only one less-formal reference, attributing a certain logical method to Jan uka-
siewicz. That is Tarski’s sole discussion in this paper of a logical method, independent
of the mathematics under consideration.

Tarski revisited this subject in his 1924c abstract Sur les principes de l’arithmétique
des nombres ordinaux (transfinis). There he presented, without proof, a system of six
independent axioms for the notion of ordinal number in Zermelo set theory without the
axiom of choice. Here and there in later discussions of well-ordered sets, Tarski’s axioms
appear, usually without attribution. For example, Patrick Suppes used the system
{ A1 , B} in his 1960 set-theory text.29 The most important impact of Tarski’s paper is not
its mathematical content, but its reflection of personal style. It displays Tarski’s practice,
in lectures and most research papers, of providing extreme detail in proofs, and of knead-
ing the formulations of definitions and axioms to achieve great concision without sacrific-
ing grace. He probably acquired that habit from his teachers Tadeusz Kotarbiski,
Stanisaw Leniewski, and ukasiewicz: recalling those times, the historian of logic
Józef M. Bocheski (1994, 7) attributed this trait to their common teacher, Kazimierz
Twardowski.

29
In 1960, 75–76, Suppes derived A2 , A3 , C from A1 , B.
18 1 School, University, Strife

Kazimierz Twardowski
around 1918

Kazimierz Twardowski was born in 1866 in Vienna, to an aristocratic family from the part of Poland
in the Austrian Empire. He attended an elite school in Vienna, then studied mathematics, science,
and philosophy at the University of Vienna, with a year at Leipzig and Munich. Twardowski earned
the doctorate in Vienna in 1892, worked for a while in insurance and as a writer, and soon obtained
the venia legendi. The guiding light of his studies was the renowned Franz Brentano, who had
recently left the priesthood and resigned from the regular faculty. His associate, Robert von Zimmer-
mann, officially supervised Twardowski’s dissertation.
Twardowski was appointed professor of philosophy at the University of Lwów in 1895, and soon
built an extremely strong program of scientific philosophy there, which became the Lwów–Warsaw
School. According to the historian of philosophy Arianna Betti,
Twardowski laid emphasis on “small philosophy,” namely on the detailed, systematic analysis of
specific problems—including problems from the history of philosophy—characterised by rigor and
clarity, rather than on the edification of whole philosophical systems and comprehensive world-
views.
Most of Alfred Tarski’s philosophy teachers at Warsaw were students of Twardowski. Twardowski
was wildly popular: his lectures were often held in a concert hall, and one year at seven in the
morning at a cinema near the university! He spent great effort to establish an infrastructure for
philosophy in Lwów, for example, various societies, a psychology laboratory, and the journal Ruch
filozoficzny. Betti noted,
All these activities cost him a lot of time: in fact, Twardowski’s choice to be most of all an
educator and an organizer left him very little time for academic writing.
Twardowski retired in 1930 and died in Lwów in 1938.†

†Betti 2011.
 2
A Contribution
to the Axiomatics of
Well-Ordered Sets (1921)

This chapter is devoted to an English translation of Alfred Tarski’s first published paper,
Przyczynek do aksjomatyki zbioru dobrze uporzdkowanego, [1921] 2014, written while
he was still a student at the University of Warsaw using the surname Tajtelbaum.1 It
appeared in volume 24 of the journal Przegld filozoficzny. This is its first translation.
Background for the paper and a summary are provided in sections 1.1 and 1.2.

The translation is meant to be as faithful as possible to the original. Its only inten-
tional modernizations are punctuation and some changes in symbols where Tarski’s
conflict with others used throughout this book. As an aspect of adjusting punctuation,
the editors greatly increased use of white space to enhance visual organization of the
paper. [Square] brackets in the Polish original have been changed to braces or parenthe-
ses. All [square] brackets in the translation enclose editorial comments. Those are
inserted, usually as footnotes, to indicate or suggest corrections for occasional errors in
the original, and to explain a few passages that seem obscure.

1
Alfred’s changes of surname from Teitelbaum and to Tarski are discussed in section 1.1 and chapter 3.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 19
DOI 10.1007/978-1-4939-1474-6_2, © Springer Science+Business Media New York 2014
Journal Containing the First Paper of Alfred Tarski (Tajtelbaum)
 ALFRED TAJTELBAUM

———

A Contribution to the Axiomatics

of Well-Ordered Sets
From the Seminar of Professor Stanisaw LeĤniewski
at the University of Warsaw

———

According to the traditional definition accepted in set theory, a set Z is ordered with
respect to a relation R if and only if the following three “order axioms” are satisfied:
A1 For all x and y, if x and y are distinct elements of the set Z, then
x has the relation R to y, or else y has the relation R to x. (Or, in
an equivalent formulation, if x and y are distinct elements of the set
Z and x does not have the relation R to y, then y does have the
relation R to x.)
A2 For all x and y, if x and y are elements of the set Z and x has the
relation R to y, then y does not have the relation R to x.
A3 For all x, y, and t, if x, y and t are elements of the set Z, and x has
the relation R to y, and y has the relation R to t, then x has the
relation R to t.

I shall write
xRy instead of “x has the relation R to y”
x RUy instead of “x does not have the relation R to y”
x=
/ y instead of “x is different from y”
x=y instead of “x is not different from y” [that is,]
“x is identical with y.”

A relation R with respect to which a given set Z is ordered is often called a prece-
dence relation; instead of “x R y,” we read “x precedes y.” Axiom A 2 is called the axiom
of antisymmetry; axiom A 3 , the axiom of transitivity.

As is well known, we call an ordered set well-ordered if and only if each of its nonempty
subsets has a first element. Thus, in order to obtain a system of axioms for a well-ordered
set [ Z ], we append to the three order axioms a fourth “well-ordering axiom,” which in
precise formulation takes one of the following shapes:
22 2 Contribution to the Axiomatics of Well-Ordered Sets

B. For every U, if
1. U is a set,
2. for every x, if x is an element of the set U, then x is an ele-
ment of the set Z, and
3. for some k, k is an element of the set U,
then for some a,
1. a is an element of the set U,
2. for every y, if y is an element of the set U, then y does not
precede a.
C. For every U, if
1. U is a set,
2. for every x, if x is an element of the set U, then x is an ele-
ment of the set Z, and
3. for some k, k is an element of the set U,
then for some a,
1. a is an element of the set U,
2. for every y, if y is an element of the set U different from a,
[that is] y =
/ a, then a precedes y.
We see that these two axioms differ in their two distinct, inequivalent interpretations of
the term “first element.”

On the other hand, it is nearly obvious that axiom systems { A1 , A 2 , A 3 , B} and

{ A1 , A 2 , A 3 , C } are equivalent: from the first of these it is possible to deduce axiom C
as a theorem; from the second, axiom B.

Indeed, let us consider the first system. Every set U satisfying the hypothesis of
axiom C also satisfies the identically phrased hypothesis of axiom B; thus it has an
element a such that if y is an element of the set U, then y does not precede a. If in
addition y is different from a, then a precedes y, according to axiom A1 . In other
words, axiom C is satisfied.

Similarly, it is straightforward to show that axiom B can be deduced from axioms

A 2 and C (and the so-called theorem of antireflexivity of the relation R, which follows
from axiom A 2 ).

However, neither the first nor the second axiom system for a well-ordered set is an
independent axiom system. In the first system, axioms A 2 and A 3 can be deduced from
the rest; in the second, A1 and A 3 .

In fact, to prove axiom A 2 from the axiom system { A1 , B} let us consider a set U
consisting of any two elements x and y of the set Z that satisfy the hypothesis x R y
of axiom A 2 . The set U satisfies the hypothesis of axiom B; therefore, it has an element
that no element of U precedes. Since x R y, this element is not y. Therefore, it must
2 Contribution to the Axiomatics of Well-Ordered Sets 23

be x, and because y is an element of the set U, it follows that y RUx, which is exactly
what was to be proved.2

To prove axiom A 3 [from the axiom system { A1 , B}], let us consider a set U
consisting of any three elements x, y, and t of the set Z that satisfy the conditions
x R y and y R t. Again, the set U, satisfying the hypothesis of axiom B, has an element
that no element of this set precedes. Since x R y and y R t, [this element] is neither y
nor t. Therefore, it must be x, and thus
t RUx. (1)
On the other hand,
t=
/ x. (2)
Indeed, y R t entails t RUy by virtue of the previously proven axiom A 2 . Thus x precedes
y, while t does not precede y, [and] therefore x and t are distinct. But if conditions
1 and 2 are satisfied, then x stands in the relation R to t by virtue of axiom A1 ,
Q.E.D.

To prove axiom A1 from the axiom system { A2 , C }, let us consider a set U consisting
of two distinct elements x and y of the set Z. (Should the set Z not contain two
distinct elements at all, axiom A1 would clearly be satisfied,3 since its hypothesis would
not hold for the given set.) The set U satisfies the hypothesis of axiom C, [and]
therefore has an element that precedes every element of the set U that differs from [it].
This element can be x or y, and therefore x R y or y R x, Q.E.D.

To prove axiom A 3 [from the axiom system { A 2 , C }], let us first of all observe that
the so-called theorem of antireflexivity of the relation R follows from the axiom of
antisymmetry:
T. For all x and y, if x and y are elements of the set Z, and x R y, then
x= / y (or, in an equivalent formulation, for every x, if x is an element
of the set Z, then x RUx).
Let us now consider a set U consisting of any three elements x, y, and t of the set Z
that satisfy the hypothesis of axiom A 3 . The set U has an element that precedes every
element of this set that differs from it. This element is not y, since x R y implies y =
/
x and y RUx, by virtue of axiom A 2 and theorem T. Neither is it t, since y R t entails
t=/ y and t RUy. Therefore, this element is x. On the other hand, however, t = / x
because both x R y and t RUy hold. Therefore x precedes t, as an element different
from it in the set U, which is exactly what was to be proved.

Thus, to define a well-ordered set, each of the two axiom systems { A1 , B} and
{ A2 , C } is sufficient.

2
[This paragraph actually proves A 2 from B alone.]
3
[In error, Tarski wrote A 2 in place of A1 in this sentence.]
24 2 Contribution to the Axiomatics of Well-Ordered Sets

Formulation of an axiom that by itself is equivalent to each of the previous systems

is not difficult. Such, for example, is the following axiom:
D. Every nonempty subset U of the set Z has one and only one element
that none of its elements precedes.
More precisely: for every U, if
1. U is a set,
2. for every x, if x is an element of the set U, then x is an ele-
ment of the set Z, and
3. for some k, k is an element of the set U,
then for some a,
1. a is an element of the set U,
2. for every y, if y is an element of the set U, then y does not
precede a,
3. for every t, if t is an element of the set U different from a,
then for some b, b is an element of the set U that precedes t.
I shall give a proof of the equivalence of axiom D with the axiom system { A1 , B}.

Axiom D directly implies axiom B. (It suffices to disregard the third part of the
conclusion.)

But to prove axiom A1 [from axiom D], let us consider a set U consisting of any two
elements x and y of the set Z that satisfy the conditions x = / y and x RUy. The set
U satisfies the hypothesis of axiom D, [and] therefore has an element that no element
of the set U precedes. From axiom B, and thus also from axiom D, as I already proved
above, axiom A 2 can be deduced, and therefore also theorem T on antireflexivity. That
implies y RUy: that is, the element y of the set U is just the one that no element of this
set precedes. On the other hand, x = / y; therefore, according to the third point in the
conclusion of axiom D, x is not such an element, and some element of the set U
precedes it: x R x or y R x. But x RUx (according to theorem B), and therefore y R x,
Q.E.D.

And conversely, axiom D follows from the axiom system { A1 , B}. Indeed, every set
U satisfying the hypothesis of axiom D also satisfies the identically phrased hypothesis
of axiom B; therefore, it has an element a that no element of the set U precedes. This
element is the only one having that property: in fact, if t is an element of the set U,
/ a, then a R t according to axiom A1 , and thus element
then t RUa; if, in addition, t =
t does not satisfy the conclusion of axiom B.4

Thus, axiom D, just like axiom systems { A1 , B} and { A 2 , C }, suffices for defining
a well-ordered set.

4
[The last clause of this sentence seems unnecessary to the proof.]
2 Contribution to the Axiomatics of Well-Ordered Sets 25

Similar definitions cannot, however, have greater significance, in view of the role that
the theory of well-ordered sets plays in modern set theory, being only a part of the theory
of ordered sets in general. In my opinion, a more interesting problem is replacing axiom
B or C by a “well-ordering axiom,” logically weaker than them, which, together with
the three ordering axioms, would form an independent axiom system equivalent to each
of the previous systems.

I shall give two formulations of such an axiom.

E. Every nonempty subset U of the set Z has an element that at most one
element of the subset U precedes.
Precisely: for every U, if
1. U is a set,
2. for every x, if x is an element of the set U, then x is an ele-
ment of the set Z, and
3. for some k, k is an element of the set Z,
then for some b,
1. b is an element of the set U,
2. for all y and t, if y and t are elements of the set U that
precede b, then y is not different from t ( y is identical to t,
y = t).
F. Every nonempty proper subset U of the set Z has an element that no
element different from it in the subset U precedes. (In this axiom the
term “first element” thus occurs with a third meaning, weaker than what
is expressed by axiom B.)
More precisely: for every set U, if
1. U is a set,
2. for every x, if x is an element of the set U, then x is an ele-
ment of the set Z,5
3. for some k, k is an element of the set U, and
4. for some l, l is an element of the set Z, and l is not an ele-
ment of the set U,
then for some a,
1. a is an element of the set U,
2. for every y, if y is an element of the set U different from a,
then y does not precede a.
In this way, I have obtained two new axiom systems for a well-ordered set: { A1 , A 2 ,
A 3 , E } and { A1 , A 2 , A 3 , F }.

5
[At this point there is an editorial error in the reprint of Tarski 1921 in the 1986a Collected Papers: omis-
sion of the clause x jest elementem zbioru Z, which corresponds to the last clause of condition 2. See
footnote 8 on page 28.]
26 2 Contribution to the Axiomatics of Well-Ordered Sets

I turn to the proof of the equivalence of each of these systems with any of the previous
systems: for example, with the system { A1 , B}. We know already that from this last
system, axioms A2 and A3 can be deduced. What remains to be proved is that (1)
axioms E and F follow from the system { A1 , B} as theorems, (2) from each of the new
systems axiom B can be deduced.

The first problem is completely straightforward: axioms E and F follow directly from
axiom B, weakening it. Indeed, if every subset U of the set Z has an element a that
no element in the subset precedes, then it also has an element that at most one element
in the subset precedes: such an element is in fact exactly that element a. Therefore,
axiom E is proved. Similarly, from axiom B follows axiom F: if every subset of the set
Z has a first element, then every proper subset has one as well; if this element is such
that no element of the subset precedes it, then certainly, no element different from that
one precedes it, which is exactly what was to be proved.

I now give proofs that axiom B can be deduced (I) from the system { A1 , A 2 , A 3 , E },
(II) from the system { A1 , A 2 , A 3 , F }.

I. Every set U satisfying the hypothesis of axiom B also satisfies the hypothesis of
axiom E, [and] thus has an element b that at most one element of U precedes. There
are two cases to consider:
1. There does not exist an element of the set U that precedes b. Then,
clearly, element b satisfies the conclusion of axiom B: no element of
the set U precedes it.
2. There does exist an element of the set U —call it a —that precedes
b. Then a satisfies the conclusion of axiom B.
Indeed, if any y is an element of the set U, then either y = a or y =
/ a. If y = a, then
y RUa (by virtue of theorem T, which follows from axiom A 3 ). If y = / a on the other
hand, then y RUb (by axiom E ), [and] therefore y = b or b R y (by axiom A1 ). If y =
b, then a R y (according to the definition of element a), [and] thus y RUa (by axiom
A2 ). Alternatively, if b R y, then since aRb, the axiom of transitivity ( A 3 ) yields
a R y, and thus also y RUa. Thus, element a indeed satisfies the conclusion of axiom
B: no element of the set U precedes it.

II. [This proof extends all the way to a matching editorial comment in brackets on the
next page.] A set U satisfying the hypothesis of axiom B might satisfy the hypothesis
of axiom F or not satisfy it—specifically, it might not satisfy the third premise.

If that is satisfied (that is, if the set U is a nonempty proper subset of the set Z), then
[U ] has an element a such that if y is an element of the set U and in addition y = /
a, then y RUa. But if y = a, then also y RUa, and therefore the element a satisfies the
conclusion of axiom B.
2 Contribution to the Axiomatics of Well-Ordered Sets 27

On the other hand, if the set U does not satisfy the third premise of axiom F, then
the set U is not different from the set Z (it is the improper subset of the set Z). I shall
consider here two possibilities.
1. The set U, and thus the set Z, has only one element: denote it by k.
Since k RUk (by virtue of theorem T), axiom B is satisfied.
2. The set U has more than one element. Let one of these elements be l.
I take under consideration the set W consisting of all elements of the set
U with the exception of l (x is thus an element of W if and only if x
is an element of the set U and in addition x =/ l).

This set certainly satisfies the hypothesis of axiom F, [and] thus has an element a
such that if y is an element of the set W, then y RUa (by virtue of theorem T, the
restriction “if y =
/ a” does not apply). It remains to consider, however, the relationship
of element a to that one element of the set U that is not an element of the set W: that
is, to l. Since a = / l by virtue of axiom A1 , two possibilities then occur here:
a R l or l R a.

If a R l, then l RUa (by virtue of axiom A 2 ), [and] therefore element a satisfies the
conclusion of axiom B.

On the other hand, if l R a, then element l satisfies the conclusion of this axiom.
Indeed, if y is an element of the set U, then either y = l or y is an element of the
set W.

In the first case, y RUl (by virtue of Theorem T ). Alternatively, if y is an element

of the set W, then y RUa, [and] therefore, by axiom A1 , either a = y or a R y. If a =
y, then l R y, [and] so y RUl (by axiom A 2 ).

On the other hand, if a R y, then l R a and a R y both occur, [and] therefore l R y (by
virtue of axiom A 3 ), and thus y RUl, Q.E.D.

Having exhausted all the possibilities, we have in general proved axiom B. [This ends
the proof that started after the matching editorial comment in brackets on the previous
page.]

Next I come to the proof of the independence of the axioms in both systems { A1 , A 2 ,
A 3 , E } and { A1 , A 2 , A 3 , F }. In order to prove the independence of an axiom in a given
system, it suffices to give an interpretation satisfying all the axioms in the system with
the exception of the one whose proof of independence is in question. I will give in turn
proofs of the independence of each of the axioms, in parallel for each system.

(1) Axiom A1 . As the set Z I choose the set of natural numbers, and define a relation
R for the elements of this set in the following way: x R y if and only if x < y – 1. Axiom
A1 is not satisfied: for example, 1 = / 2 but also 1 RU2 and 2 RU1. On the other hand,
the remaining axioms are satisfied. [For] A 2 : if x < y – 1, then y > x + 1, [and] thus
28 2 Contribution to the Axiomatics of Well-Ordered Sets

y > x – 1, y RUx. [For] A 3 : if x < y – 1 and y < z – 1, then x < z – 1. Finally, E and
F: the conclusions of these axioms are in fact satisfied by the smallest number in the
given subset.

The simplest interpretation is the set consisting of two consecutive natural numbers,
for example 1 and 2, with the previous definition of the relation R; it satisfies axioms
A 2 and A 3 because for no values of the variables x, y, and t are the hypotheses of these
axioms satisfied.6

(2) Axiom A 2 . I choose the same set as before and define x R y if and only if x < y
or x = y (x  y). Axiom A 2 is not satisfied: x R y and y R x occur together only if x =
y. On the other hand, [these] axioms are satisfied: A1 (if x = / y, then x < y or y < x,
therefore x R y or y R x); A 3 [similarly]; and E and F —these last two are satisfied,
as before, by the smallest number belonging to the given subset.

The simplest interpretation is the set consisting of a single number with the previous
definition of the relation R.7

(3) Axiom A 3 . As the set Z, I choose a

set consisting of three distinct points on a b
circle, following one another 8 in a given
direction: for example, points a, b, and c
(figure 1). The relation R between the
elements of this set I denote with the aid of a
arrows, as indicated in the figure; thus
a R b, b R c, and c R a.

Axiom A 3 is not satisfied: a R b, b R c,

and simultaneously a RUc. On the other
hand, the remaining axioms are satisfied: A1
c
and A 2 are direct consequences of the defi-
nition of the relation R; and for E and F,
it is easy to check separately each of the Figure 1
seven nonempty subsets of the set Z.

(4) Axioms E and F. As an interpretation, any ordered set that is not well-ordered
will do: for example, the set of rational numbers, ordered according to size.

6
[This paragraph seems unrelated to the rest of the paper. Its relation R is empty; axioms B, E, and
F are thus valid, but not axioms C and D. Tarski’s use of simplest (najprostsz) is unclear: an interpre-
tation with a single element and empty R satisfies all the axioms except A1 .]
7
[This paragraph also seems unrelated. Its relation R satisfies all the axioms except A 2 , B, D.]
8
[At this point there is an editorial error in the reprint of Tarski 1921 in the 1986a Collected Papers:
insertion of the clause x jest elementem zbioru Z, unrelated to the surrounding text. See footnote 5 on
page 25.]
2 Contribution to the Axiomatics of Well-Ordered Sets 29

In this way, I have carried out the proof of the independence of the axioms in both
systems.

It is worthwhile to note that axioms E and F are not equivalent, and neither of them
follows from the other. In order to prove this, it suffices to give two such interpretations
that would, alternatively, satisfy one of the axioms while not satisfying the other.

Thus, if to the set, already mentioned, of three points a, b, and c on the circle we
append the center o of the circle and specify new relationships oRa, oRb, and oRc
(figure 2), we obtain a set that does not satisfy axiom F: for example, the subset U
consisting of points a, b, and c does not satisfy it. This set, on the other hand, satisfies
axiom E: in fact, if a subset contains point o, this point satisfies the conclusion of axiom
E; otherwise, any element of the subset satisfies this conclusion, because at most one
element of the subset bears the relation R to it.

a o
d e

Figure 2 Figure 3

On the other hand, a set consisting of two points d and e, in which we have specified
the relationships d R d, d R e, e R d, and e R e (figure 3), does not satisfy axiom E: in
fact, the improper subset of the set does not satisfy it. But [this set] satisfies axiom F:
every subset consisting of a single element satisfies this axiom, and every proper subset
of the given set is composed of only one element.

Neither the system of order axioms, nor any of the systems obtained that define a well-
ordered set, is a relatively weakest system, that is, one in which it is not possible to obtain
an axiom system equivalent to the given system by replacing any axiom by an axiom
weaker than it without changing the remaining axioms.

Thus, in each of these systems that include axioms A 2 and A 3 it is possible to replace
axiom A 2 by the weaker axiom T, of antireflexivity. The system obtained is equivalent
to the previous one, because A 2 follows from axioms T and A 3 . And yet in this way we
will not obtain relatively weakest systems.
30 2 Contribution to the Axiomatics of Well-Ordered Sets

A certain theorem of Prof. Jan ukasiewicz gives us a method for obtaining such
systems. By virtue of this theorem, if {T1 , T2 , ... ,Tn } is an independent system of axioms,
then a relatively weakest system equivalent to the given one is the following system,
written in symbolic form:
T1 (1) T1
T1 H T2 (2) if T1 , then T2
T1 & T2 H T3 (3) if T1 and T2 , then T3
 
T1 & T2 & $ $ $ & Tn –1 H Tn (n) if T1 , T2 , ... , and Tn –1 , then Tn .9

In this way, we would obtain the following axiom system for ordering:
A1 A1 H A 2 A1 & A 2 H A 3 .
Moreover, the well-ordering axiom would take on one of two forms, easily proved
equivalent:
A1 & A 2 & A 3 H E or if axioms A1 , A 2 , and A 3 are satisfied,
A1 & A 2 & A 3 H F then axiom E or F is satisfied.
However, I do not give a complete formulation of the axioms constructed in this way,
because despite rewording I was not able to put them into an aesthetic form.

9
[This theorem is easy to prove. The present editors have not been able to locate any other source for it.]
 3
Doctoral Research

The account of Alfred Tarski’s life begun in chapter 1 continues here.1 This chapter
describes the latter years of Alfred’s studies at the University of Warsaw, and various
other aspects of his life during 1921–1924. These include some personal details, his part-
time employment as a teacher during his student years, his doctoral research, and his
early participation in professional meetings. It provides a setting for both the detailed
mathematics of Part Two of the book, and for Tarski’s professional career during
1924–1939 as a researcher and secondary-school teacher and teacher trainer, which are
the subject of Part Three.

Chapter 1, section 1.1, described Alfred’s exciting first year of courses, 1919–1920; the
Soviet invasion in summer 1920, the Polish victory that August—the Miracle of the
Vistula—in the very suburbs of Warsaw; and Alfred’s subsequent re-enrollment for
autumn 1920, with courses from Stanisaw Leniewski, Jan ukasiewicz, Stefan Mazur-
kiewicz, and Wacaw Sierpiski. Three intellectual threads in Alfred’s studies had taken
form during those semesters: logic, set theory, and measure theory. They would extend
far into his research career.

From autumn 1920 though autumn 1923, Alfred enrolled in many of the same classes
as the future politician and logician Kazimierz Pasenkiewicz. Decades later, Pasenkiewicz
described 2 the atmosphere that they shared:
In the year 1920–1921, around ten thousand persons enrolled at the University of Warsaw.
This number exceeded by many times the didactic possibilities of the university. In the lecture
halls and laboratories it was very crowded; but not in all. Not many students enrolled in the
theoretical directions, especially in mathematics, philosophy, and logic. ...
This was a period of great liberalism and tolerance. The bureaucratic requirements were
reduced to the most essential. ... The course of studies was very liberal. The only condition for
receiving credit for a semester was to inscribe in a register a definite number of lecture-hours
for subjects lectured in the department, and to obtain from the professor a confirmation of
attendance. Very often these signatures were obtained through the janitors.
[continued on the next page]

1
For biographical information supporting this chapter, unless another source is cited, consult the biogra-
phies by Anita B. and Solomon Feferman (2004) and Jacek Juliusz Jadacki (2003a). Also note the
descriptions of those works in chapter 17. The Fefermans emphasized personal-interview sources; Jadacki,
published records. For historical and sociological information, consult the works by Norman Davies (1982,
volume 2, chapters 18–19), Celia S. Heller (1994), and Richard M. Watt (1979).
2
Pasenkiewicz 1984, 2–3. A portrait and biographical sketch of Pasenkiewicz are on page 32. The following
quotation and those on pages 33 and 36 were translated by Jan Tarski, then lightly edited by the present
editors to conform with the conventions of this book.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 31
DOI 10.1007/978-1-4939-1474-6_3, © Springer Science+Business Media New York 2014
32 3 Doctoral Research

[quotation continued from the previous page]

There were no required lectures, nor examinations. One could, however, pass a discussion
section for an attended lecture course. For a passed discussion section one received a confirma-
tion with a grade.
After attending six semesters, one could take the final examination. Its successful result
gave a diploma of completion of university studies and the right to lecture in secondary schools.
The procedure for obtaining doctorates remained unchanged. It was necessary to obtain a
positive evaluation of the doctoral dissertation and to pass through a traditional routine.

Kazimierz Pasenkiewicz
in 1920

Kazimierz Pasenkiewicz was born in 1897 in Kiev, and schooled there. He fought with the Russian
army in World War I, then against it in the Polish–Soviet War of 1920, in which he lost a leg. Pasen-
kiewicz then entered the University of Warsaw to study mathematics and logic, and attended lectures
and seminars alongside Alfred Tarski. Pasenkiewicz earned the doctorate there in 1933, with a disser-
tation supervised by Tadeusz Kotarbiski. He continued that study and research until World War
II. During that war he worked with the Polish underground and with socialist organizations. After-
ward he became very active in the Communist Party, continuing until its dissolution in 1991. During
1948–1968 Pasenkiewicz served as lecturer, professor, and dean in the philosophy faculty at the Uni-
versity of Cracow. He was an expert on the logical work of Leon Chwistek. Pasenkiewicz died in 1995
in Cracow.*

* Kutta 1997.
3 Doctoral Research 33

During the period from winter 1921 through autumn 1923, Alfred enrolled in courses
following much the same pattern as that of his previous semesters:
• in nearly every term, two or three courses or seminars with Leniewski;
• in nearly every term, courses from Mazurkiewicz—on analytic geometry,
Jordan continua, topology of the plane, and entire functions;
• in most terms, seminars with ukasiewicz and his courses on logic and
philosophy;
• from winter 1921 through winter 1922, courses from Sierpiski on higher
algebra and analysis;
• an autumn 1921 course on mechanics from Czesaw Biaobrzeski;
• in winter 1922 and autumn 1923, courses from Kazimierz Kuratowski on
topology and set theory.3
Alfred was thus continuing to explore the logic and set theory threads that would extend
into his research career. Moreover, with the courses on topology and continua he added
a new strand that, with his previous study of measure theory, would become a thread of
applications of set theory to geometry.

Pasenkiewicz enrolled in the lectures of Sierpiski and Mazurkiewicz that involved

set theory, those of Leniewski on logic, some lectures on physics, and lectures by
ukasiewicz on history of philosophy. Pasenkiewicz recalled,
To the lectures on set theory there came then a fair number of students. These were the times
of Émile Borel, Henri Lebesgue, Ernst Zermelo. The theory of sets was rather fashion-
able. ... To the [very popular] lectures on theoretical physics of Biaobrzeski from Kiev came
students of mathematics and of philosophy; those were the times of Albert Einstein and the
atom.
In contrast, few came to the lectures of Leniewski ... regularly three persons: Jan Drewnowski,
Aleksander Jaboski, and I. ... A few came irregularly, among them Alfred Tarski. He sat in
the last row and read newspapers. After a lecture, or during intermission, he conferred with
the professor; he did not enter into conversations with fellow students. During 1922–1923,
Leniewski acquired a few new students: there appeared Adolf Lindenbaum, Mordchaj Wajs-
berg, and a few others. Tarski livened up. At the seminars he had the possibility to evaluate
those who spoke; he took interest in some of these.
[Tarski] appeared from time to time also at the lectures of Sierpiski, primarily ... when he had
some matters to speak with him about. Tarski did not come to [the lectures on physics or]
history of philosophy. He was consistent in his interests, did not distract himself. ... In his
surroundings Tarski did not notice those who did not show particular abilities or interests in
logic. On the other hand, he valued people with whom he shared these interests. In discussions
with them he did not take on attitudes or a position of authority.4

3
Tarski 1924f.
4
Pasenkiewicz 1984, 2–3. For information about Lindenbaum, see a box in section 14.3; for the other stu-
dents mentioned, see page 34. The present editors suggest the word question in place of evaluate in the
second paragraph of this translation by Jan Tarski.
34 3 Doctoral Research

Aleksander Jaboski Mordchaj Wajsberg

in 1935 around 1925

Jan Franciszek Drewnowski was born in Moscow in 1886. He attended lectures and seminars at
the University of Warsaw during the early 1920s, alongside Alfred Tarski, and earned the doctorate
in philosophy there in 1927, supervised by Tadeusz Kotarbiski. Drewnowski had studied finance
and economics as well. He worked in industry and served as a government official both before and
after World War II. Drewnowski was a member of the Cracow Circle of Catholic philosophers, and
published in both economics and analytical philosophy. He died in Warsaw in 1978.*

Aleksander Jab oĔski was born in 1898 in Woskresenówka, in eastern Poland, then in the Russian
Empire. He served Poland as an engineer officer during the 1920 Polish–Soviet War. Afterward,
he worked as a professional musician while studying at the University of Warsaw. He attended
lectures alongside Alfred Tarski. Jaboski earned the doctorate at Warsaw in physics in 1930, and
received the venia legendi in 1934. As an engineer officer again in 1939, he was interned in the Soviet
Union, left with the army of General Wadysaw Anders via Central Asia and the Middle East, then
served as a lecturer in Scotland. After the Second World War, he helped build the Department of
Physics at the University of Toru. He was a world authority on the physics of light. Jaboski died
in 1980 in Skierniewice, in central Poland.†

Mordchaj Wajsberg was born in 1902 in oma, in central Poland, then in the Russian Empire.
His schooling was interrupted by World War I, then by military service in the 1920 Polish–Soviet War.
He finally completed secondary school in oma in 1923, and entered the University of Warsaw, to
study mathematics and logic. He attended lectures and seminars with Alfred Tarski. Wajsberg soon
began a stream of research results, concentrating on axiomatics of three-valued logic and modal logic.
This area was intimately tied to the work of Tarski and of Jan ukasiewicz, who supervised the
research for Wajsberg’s master’s and doctoral degrees, awarded in 1928 and 1931. Wajsberg pioneered
the use of algebraic and model-theoretic techniques to study those logics. In 1933 he began work as
a secondary-school teacher, first in the eastern Polish town of Kowel (now Kovel, in Ukraine), then
in oma. There he extended his techniques to include general multivalued and intuitionistic logic.
Wajsberg perished in the Holocaust, probably during the Nazi massacre of Jews in œoma in 1942.‡
_______________ _______________
*Majdaski and Lekka-Kowalik 2001. Note: there was † Frckowiak 1998
another scholar of the same era with a similar name. ‡ Surma 1977
3 Doctoral Research 35

The tuition costs entered in Alfred’s enrollment

record reveal a major ordeal for Polish citizens in
the early 1920s: hyperinflation. The entries for Autumn Tuition Costs
autumn 1919 were pictured in section 1.1. Figures Trimester (Polish Marks)
for subsequent autumn trimesters, shown in the 1919 76
table at right, illustrate exponentially increasing 1920 600
costs for 1919–1922, followed by a year of hyper- 1921 2,800
exponential growth. The bursar’s certificate corre- 1922 14,500
sponding to that last line is pictured below. 1923 32,300,000
Poland’s currency soon collapsed. In 1924, the gov-
ernment instituted reforms, and the new Bank Pol-
ski introduced the currency used today, the zoty.5

To the Bursar
Tajtelbaum, Alfred; Student 2909;
Philosophical Faculty; Koszykowa 51-14
Winter 1923–1924

Topology of the Plane 3 hr Mr. Mazurkiewicz

Entire Functions 2 hr ̵̵ ̵̵
Theory of Sets 3 hr K. Kuratowski
Topology 2 hr ̵̵ ̵̵

Payment:
Mk. 19 000 000
  
Library ̵̵ 8 800 000
St. K. Ch. ̵̵ 4 500 000
Total Mk. 32 300 000
22 January 1924

5
Tarski 1924f, 8–11, and 1921–1926. It would be difficult to make precise year-to-year comparisons, because
different aspects of tuition were billed in different ways. Moreover, Alfred evidently obtained partial
tuition waivers for spring 1921 and spring 1922. But the exponential growth outweighs such details. For
information about the Polish economy, see N. Davies 1982, volume 2, 307; Watt 1979, 202–207; and
Wynot 1983, 46–47. Zoty is a Polish word for gold; the English spelling is zloty. The abbreviation St.
K. Ch. in the caption means Studencka Kasa Chorych (Student Health Fund).
36 3 Doctoral Research

Alfred continued his development as a researcher in logic, guided by ukasiewicz,

Leniewski, and Tadeusz Kotarbiski. All three had studied with Kazimierz Twardowski
in Lwów. In 1920 yet another student of Twardowski had arrived in Warsaw to enrich
its research climate: Kazimierz Ajdukiewicz.6 At an April 1921 meeting of the Logical
Section of the Warsaw Philosophical Institute, Alfred presented his first public scientific
address, entitled On the Notion of Proof ( In Response to the Dissertation of K. Ajdukie-
wicz). No copy of its text is known, but Alfred wrote later that it contained the first dis-
cussion of the deduction theorem, a logical tool that would play a fundamental role in his
organization of metamathematics. In December 1921 Alfred presented another paper
to the same group: On the Problems of Extensionality in Logistic and in Ontology. No
copy of this text is known either, but the title suggests that it pertained to Leniewski’s
developing system of foundations of mathematics, the subject of Alfred’s doctoral
research.7

At the age of twenty, Alfred was learning the etiquette of scientific discourse. Pasen-
kiewicz reported about Leniewski’s autumn 1920 seminar:
The base of the seminar’s activities was the book by Louis Couturat, Algebra of Logic. The
criticism of this book was shattering. At one point Tarski stood up and asked whether it was
at all worthwhile to busy oneself with this. Leniewski felt a bit slighted, but asked with
humor, “Do you think that my seminar is a waste of time?” Tarski sat down.

Pasenkiewicz attended the meetings at the Institute, too. Tarski’s patience was improv-
ing, but it had limits:
Tarski was unusually efficient intellectually. At the meetings ... a few times papers were
read ... by Tarski. ... I admired ... [him for answering] the voices in the [subsequent] discussions
without notes, but in the order in which they were presented. ... [Unlike another speaker, who]
answered to all the raised issues with equal attention and friendliness, Tarski, on the other
hand, [responded] only to those which deserved an answer.8

Alfred still lived with his parents, and would do so until he married. According to his
biographers Anita B. and Solomon Feferman, he felt a need to establish some financial
independence. Around 1920, Alfred found a job teaching geometry at a girls’ secondary
school. After two years’ service, according to his account many years later, he was fired
for being Jewish. In 1921 Alfred obtained an appointment to teach logic at the National
Pedagogical Institute. This organization had been founded in 1920 as an adjunct of the
[continued on page 38]

6
For more information about Ajdukiewicz and Twardowski, see the box on page 37 and one in section 1.2.
7
The two talks were listed in Warszawski Instytut Filozoficzny 1921–1922 as A. Tajtelbaum, O pojciu
dowodu (z powodu rozprawy K. Ajdukiewicza) and O zagadnieniach ekstensjonalnych w logistyce i ontolo-
gii. The dissertation mentioned in the parenthetical phrase was [1921] 1966, which Ajdukiewicz had sub-
mitted for the venia legendi. According to Coniglione and Betti 2001, and Betti 2008, 61, that dissertation
introduced to Poland the structural definitions of proof, theorem, consequence, logical theorem, and logical
consequence that Alfred would later organize precisely and forcefully. See Tarski [1930] 1983b, 32.
8
Pasenkiewicz 1984, 5–6. The seminar date is from the enrollment record, Tarski 1924f. Couturat [1908]
1918 had recently been translated from French into Polish by ukasiewicz and Bronisaw Knaster, sup-
ported by a grant from the Mianowski Fund.
3 Doctoral Research 37

Kazimierz Ajdukiewicz
around 1920

Kazimierz Ajdukiewicz was born in 1890 in Tarnopol, then in the Austrian Empire (now Ternopil
in Ukraine). His father was a government official. He was schooled in Cracow, then in Lwów, and
entered the university there in 1908. He studied with Jan ukasiewicz and Wacaw Sierpiski, and
earned the doctorate in 1912 with a thesis on Kant’s philosophy of space, under supervision of
Kazimierz Twardowski. Tadeusz Kotarbiski and Stanisaw Leniewski also earned doctorates from
Twardowski that year. Ajdukiewicz qualified for a teaching credential, but spent the next year in
Göttingen, where he could study the work of David Hilbert, Edmund Husserl, and Leonard Nelson.
During World War I and its aftermath, he served in the Austrian and Polish armies, taught in a
gimnazjum, and married Twardowski’s daughter, a classical philologist. During this time Ajdukiewicz
wrote his influential [1921] 1966 booklet on the methodology of deductive science, which won him
the venia legendi and served as a basis of Alfred Tarski’s later work in that area. During 1922–1928,
Ajdukiewicz served in Lwów as gimnazjum teacher and university dozent, and in Warsaw as philoso-
phy teacher at the same gimnazjum as Tarski and as professor at the university. He was appointed
to the permanent faculty at the University of Lwów in 1928 and promoted to full professor in 1934.
Ajdukiewicz was the originator of many concepts and techniques now familiar in the philosophy of
science; he adapted and extended Tarski’s approach to several areas of philosophical logic. During
World War II, he worked as a teacher during the Soviet occupation, and as a clerk and clandestine
teacher under the German oppression. Afterward, he became professor at the University of Pozna,
where he served as rector during 1948–1952. From 1953 on he edited the journal Studia Logica.
In 1955 Ajdukiewicz returned to Warsaw as head of the Division of Logic of the Polish Academy of
Sciences and professor at the university. He retired from the latter position in 1961, and died sud-
denly in 1963.*

* Giedymin 1978, Raabe 1926.

38 3 Doctoral Research

[continued from page 36]

University of Warsaw through the efforts of ukasiewicz, then a university and govern-
ment administrator. Its purpose was to train schoolteachers; its instructors were selected
and supervised very seriously by university faculty.9 Alfred continued with the institute
for about four years. During that same period, he also taught at Zofja Kalecka’s Gimna-
zjum, a girls’ school that emphasized mathematics and physical science.10

During 1922–1924 Alfred was baptized a Roman Catho-

lic, and he changed his surname legally to the Polish-
sounding Tarski. At right are his signatures from this
period. These acts may reflect a change in religious con-
viction, a desire to be more Polish, or an assessment of the
disadvantages of being Jewish. Jewish scientists from
Poland have painted a bleak picture of the situation facing
Tarski there in academia. Alfred’s decision has been dis-
cussed at some length in the literature; writers disagree
about his motivation. In the early 1970s, he told the
Israeli logician Menachem Magidor,
I considered myself, at that point, to be Polish, culturally and nationally; and I didn’t care
about religion at all, so it had nothing to do with religious belief.

Alfred’s former Berkeley colleague John Corcoran recently reported:

Tarski and I talked about religion quite a bit. It was easy and natural. ... Tarski never revealed
a wavering from the atheistic humanism he came to as a young man ... He never mentioned
that he had ever converted to Catholicism; I think that “conversion” was never of any inner
significance to him.11

9
New educational regulations permitted schools to hire teachers after an oral examination, a sample lesson,
and some university studies or completion of a program at the institute (Manteuffel 1936, 163). This
permitted Alfred to teach at a secondary school, and created a demand for instruction at the institute.
10
The biographies by Steven R. Givant (1999, 50), Jacek Juliusz Jadacki (2003a, 142–143), and the Fefer-
mans (2004, 54) disagree about the details of this paragraph. It follows Givant’s most closely. Only Givant
mentioned the first school, with no name. Jadacki corrected the others about the name of the institute:
Pastwowy Instytut Pedagogiczny. The Fefermans described the institute as a women’s school whose
headmistress fired Tarski. But its director, Pawe Sosnowski, described it in 1923 as coeducational; he
listed “A. Tarski” as instructor for logic during 1921–1922 (Sosnowski 1923, 78, 81). Alfred must have
notified him of the impending name change just before that publication. Tarski was still a staff member
there in autumn 1923 (Zagórowski 1924–1926, volume 1, 106); Jadacki suggested that Alfred remained
so until the institute closed in 1925. Only Jadacki, among the biographers, mentioned the Kalecka
Gimnazjum; a Jewish school, it was not likely the one that fired Alfred. (That is confirmed in Zagórowski
1924–1926, volume 1, 151, and Shneiderman 1995.)
11
The signatures are from Tarski [1918] 2014, 1919–1920, and [1924] 2014a; the first change was discussed
in section 1.1. See Feferman and Feferman 2004, 38–40, 269; this includes a reproduction of the appli-
cation for change of name, Tarski [1924] 2007. See also Jadacki 2003a, 142–145, Woleski 1995a, note
7, and Woleski 2008. The autobiographies of physicist Leopold Infeld and probabilist Mark Kac record
their impressions as students in Cracow and Lwów during 1920–1935 (Infeld 1980, chapters 1–2; Kac
1985, 28). Heller (1994, chapter 6) discusses assimilation and conversion in depth. Corcoran 2011a,
2011b.
3 Doctoral Research 39

Alfred was awarded the doctorate in March 1924, as recommended by Leniewski.12

His dissertation, On the Primitive Term of Logistic, was published in Przegld filozoficzny
in 1923, under the surname Tajtelbaum-Tarski. The term logistic referred to a system
of logic whose features included material equivalence @ and perhaps other Boolean
connectives, propositional variables p, q, ... , variables f, g, ... for truth functions, and
universal and existential quantifiers over these variables. A truth function f is one that
maps truth values to truth values, such that if p @ q is valid, then so is f ( p) @ f (q).13
Alfred showed, roughly speaking, that negation and conjunction can be defined in terms
of the equivalence connective @ and universal quantifier ~ by regarding ¬ p and
( p & q) as abbreviations for ( p @ (~q) q) and ~f ( p @ ( f ( p) @ f (q) )). This can be
14
justified intuitively, but the steps that suggest themselves do not all have counterparts
in Leniewski’s system. Alfred’s achievement was to adapt and fit such definitions into
the system; he had to use a more complicated one for conjunction. Alfred immediately
published a two-part French translation of his dissertation in the journal Fundamenta
Mathematicae during 1923 and 1924. He published the first part under the surname
Tajtelbaum; the second, as Tajtelbaum-Tarski.15 From that time on, he published all his
work as Alfred Tarski.

Leniewski worked on his logical system through the 1920s, and finally published its
details at the end of the decade. In good humor, he emphasized his debt to Tarski:
The system of the foundations of mathematics I have constructed owes a number of important
improvements to Alfred Tarski ... “my” Doctor in the year 1924. ... I will endeavor to show them
explicitly; but because of the nature of things I cannot show properly all of Tarski’s occasional
critical remarks, which undermined this or that link of my theoretical conceptions at the
different stages in the building of my system, and all the subtle and sympathetic counsel and
often impalpable suggestions, from which I had the opportunity to profit in numerous conversa-
tions with Tarski.16

Tarski regarded Leniewski’s system as equivalent to other better-known systems, but

referred to its use in just one more publication, his 1927 report about geometry based on

12
The official certificates are displayed in Jadacki 2003, 168; the official critics are identified there on page
144 as ukasiewicz and Sierpiski. For a detailed description of the presumably similar doctoral defense
of mathematician Jerzy Neyman in Warsaw the same year, see Reid 1983, 53.
13
Truth values are values of propositional variables. See Church 1956, §§06, 28, for more information on
this sort of logical system.
14
To see this, suppose first that p, q should both be valid; that would imply validity of p @ q,
f ( p) @ f (q), and p @ ( f ( p) @ f (q) ), no matter what f might be. Should p be valid but q not, then
letting f denote the identity function, f ( p) @ f (q) and hence p @ ( f ( p) @ f (q) ) would be invalid. If
p were invalid, then letting f be any constant truth-function, f ( p) @ f (q) would be valid and thus
p @ ( f ( p)@ f (q) ) invalid, no matter what q might be. The definition of negation works because
(~q) q is invalid. The remaining connectives and the existential quantifier can then be defined as in
elementary logic.
15
See Tarski 1923a, 1923b, and 1924d.
16
Leniewski [1927–1930] 1992, 180. Leniewski included a translation of the 1927 Polish version of this
quotation in the German original version of Leniewski [1929] 1992, 414. His Polish and German words
for impalpable were nieuchwytny and ungreif bar: elusive and ungraspable.
40 3 Doctoral Research

the notion of solid body. Arianna Betti has analyzed Tarski’s major scientific debt to
Leniewski, which would appear most clearly in Tarski’s work of the 1930s:
... among the things which Tarski was taught by Leniewski were ... how to analyse quotation
marks, the language/metalanguage distinction, the idea that truth is language-relative, the
notion of a closed language, and that natural language is such a language, namely that natural
language is universal.17

During 1922–1923, Alfred had enrolled only in lectures and seminars on logic and
philosophy, ostensibly concentrating on his doctoral research project in logic. Alongside
those studies, however, he pursued intellectual threads that had evidently captivated him
even earlier: set theory and its applications to geometry. As early as 1919–1920, he had
studied geometry and calculus with Zygmunt Janiszewski and Stefan Mazurkiewicz, and
set theory and measure theory with Wacaw Sierpiski. Tarski had continued with the
latter two, studying analysis and algebra, and during spring 1922 and autumn 1923, he
also attended Kazimierz Kuratowski’s topology courses. Sierpiski influenced Tarski to
study in detail Felix Hausdorff’s 1914a book Grundzüge der Mengenlehre.18 Sierpiski
had just published his major1918 survey of the use in set theory and analysis of the axiom
of choice, the basic principle according to which,
for any family F of nonempty sets, there is a set that contains exactly one
element from each member of F.
Moreover, for several years, Sierpiski and Mazurkiewicz had been studying seemingly
paradoxical constructions in set theory. For example, in a 1914 paper, they described two
disjoint unbounded subsets A and B of the plane, each of which is congruent to the
union A F B. Their excitement about such results surely infected their lectures. Tarski
immersed himself in that research area, too. His work would involve its most advanced
aspects.

One filament of this set-theory thread involved delicate consideration of the axiom of
choice. Tarski established striking relationships between this idea and that of the finite-
ness of a set, and discovered that certain other set-theoretic propositions, some quite
familiar, were equivalent to the axiom. For example, in 1914a, Hausdorff included the
theorem that for any infinite set A, there is a one-to-one correspondence between the
elements of A and those of the set A × A of all ordered pairs of elements of A. The proof
made essential use of the axiom of choice. Tarski showed that some use of that principle
was inescapable, because that theorem actually implies the axiom.19 Tarski published this
work simultaneously with his dissertation, and continued to work on this thread in
collaboration with Adolf Lindenbaum. Their joint papers would start appearing in 1926.

17
Lindenbaum and Tarski 1926, 299; Tarski [1927] 1983. Betti 2004, 278–279. Concerning the relationship
between Leniewski and Tarski, see also Betti 2008. For a brief account of Tarski’s research in logic up
to 1939, see section 8.1.
18
See Feferman and Feferman 2004, 48, and Tarski 1924f. For biographical information about Sierpiski
and Hausdorff, consult boxes in sections 4.1 and 4.4. Tarski always recommended Hausdorff 1914a to
his own students, even decades later.
19
Hausdorff 1914a, 127. Tarski 1924b and 1924e, 150. Lindenbaum and Tarski 1926. This theorem about
A × A is due to Gerhard Hessenberg (1906, 108).
3 Doctoral Research 41

That research is not considered further here, because it is relatively accessible and lies
beyond the scope of this book.

Soon after his graduation, Tarski applied to the dean of the faculty, the chemist Wiktor
Lampe, for permission to pursue further study in mathematical sciences, particularly
theoretical physics. His letter is translated in chapter 16, section 4. Tarski’s signature
included the title Dr.; it is reproduced on page 38. Tarski signed up for nine physics and
astronomy courses in spring 1925 and five in spring 1926. There was evidently no tuition
charge; he must have been going to sample those offerings.20

In 1925, probably on the basis of his 1924b research on the definition of finiteness,
Tarski was awarded the venia legendi by the faculty: this allowed him to use the title
docent and to give courses, receiving no salary but simply a fraction of his students’ fees.
For a year starting in February 1925, Tarski gave a course on deductive logic at the Free
University of Poland, a private postsecondary institution in Warsaw; the course met for
two hours each week.21

During 10–13 May 1923, Tarski had performed one of the duties of an advanced
student: he had served as secretary of the logic section of the First Polish Philosophical
Congress in Lwów. He did not present a paper, but participated in the discussions after
talks on the principle of contradiction, type theory, axiomatization of physics, truth
functions, and the logic of adjectives. At that meeting he met a new Lwów professor of
mathematics, Stefan Banach, who was also fascinated by the intricacies of set theory and
geometry.22 They began a collaboration that led to their famous 1924 paper On Decompo-
sition of Point Sets into Respectively Congruent Parts. Part Two of the present book is
devoted to Tarski’s research on decomposition of point sets, and chapter 6 contains a full
translation of that paper.

20
Tarski [1924] 2014a; Tarski 1922–1926, documents 18 RP 2909 and 20 RP 2909. Jadacki 2003a, 144.
21
Wolna Wszechnica Polska 1925, 8, 25. Many other University of Warsaw faculty taught part-time at the
Free University; this was apparently Tarski’s only involvement with it. For his service at the University
of Warsaw, see section 9.3.
22
Polski Zjazd Filozoficzny 1927, 266, 284–294, 361. Tarski’s comments were not recorded. For bio-
graphical information about Banach, consult a box in section 4.2.
 Part Two
Geometry
Part Two of this book is devoted to a single thread of Alfred Tarski’s research: applica-
tions of set theory to geometry. Five research papers are translated here, all concerned
with the problem of decomposing two geometric figures of equal measure into equal finite
numbers of components that are congruent in pairs. In the first two papers, both
published in 1924, the components may be sets of any sort, but they must all be disjoint.
Chapter 5 contains the translation of Tarski’s paper On the Equivalence of Polygons, in
which the original figures are constrained to be polygons in the plane. In chapter 6, the
translation of the famous paper by Tarski and Stefan Banach, this restriction is relaxed:
the figures need only be bounded and contain interior points, and all dimensions are
considered. Both of these results depend on the most advanced results in set theory at
that time. The remaining three papers make up chapter 7, and include one by Henryk
Moese that is inextricably tied to Tarski’s work. These results are cast in the framework
of secondary-school geometry: the components must be polygons, and only their interiors
must be disjoint. With these translations and others published elsewhere, all of Tarski’s
geometric research has become accessible in English.

Chapter 8 summarizes some other filaments of this thread of Tarski’s research, which
can be described as technical results in set theory, and which are accessible elsewhere in
French or German. It discusses the impact of all this work on the mathematics of
subsequent years. Chapter 4, the first chapter of Part Two, provides the mathematical
background for the papers translated here, and discusses some details of their original
publication. The subject matter of Part Two is all related to the subjects that Tarski
taught in his position as secondary-school teacher and teacher trainer. That activity is
the subject of Part Three of this book.

Tarski’s research in Poland unfolded amid a chaotic vortex of political, social, economic,
and scientific developments. Accounts of events, people, and ideas were merged from
several dimensions to form the linear sequence of pages of this book. That is reflected
in the background chapter 4 by the use of boxes interspersed in the main narrative. They
contain biographical sketches of some persons associated with Tarski, and informational
essays about several other topics. Each box can be read independently: readers are not
expected to visit them in sequence. Cross-references refer to them from the main
narrative.
 4
Area, Volume, Measure

The synthetic approach to the theory of area and volume initiated by Euclid is described
in the first section of this chapter. The area a of a polygonal region P is computed by
decomposing P into a finite number of polygonal components with disjoint interiors,
which can be reassembled to form a rectangle R with unit base: a is then the altitude
of R. The volume of a polyhedral region can be reckoned in a similar way, but for that
it is necessary to use some form of Eudoxus’s method of exhaustion. Perfected over the
centuries, these methods are employed even today in many secondary schools. This area
theory provides the mathematical background for the research papers by Alfred Tarski
([1931] 2014a and [1932] 2014d) and Henryk Moese ([1932] 2014) on the degree of
equivalence of polygons, which are introduced and translated in chapter 7. The intended
audience for those papers was in fact talented and highly motivated secondary-school
students and their teachers. The earlier research papers by Tarski ([1924] 2014b) and
by Stefan Banach and Tarski ([1924] 2014), translated in chapters 5 and 6, are concerned
with decomposition of geometric figures into more general components that are entirely
disjoint. They require more complicated set theory and the notion of measure, which is
a generalization of area and volume. The necessary additional background is described
in section 4.2; the papers themselves are summarized in sections 4.3 and 4.4.

4.1 Area and Volume

Studying geometry in school, in informal discussions at university, in teaching his own

students, and in training teachers, Alfred Tarski confronted the theories of area and
volume of geometrical figures. There are several very different ways to develop them.
In his paper translated in chapter 5, Tarski referred first to the approach expounded in
a popular text by the noted Italian mathematicians Federigo Enriques and Ugo Amaldi.1

In that presentation, two polygonal regions are called equivalent if they are unions of
the same finite number of polygonal subregions with disjoint interiors, such that corre-
sponding subregions are congruent. This relation is reflexive, symmetric, and transitive.
Parallelograms with congruent bases and altitudes are equivalent. That is visually appar-
ent if the bases are superimposed and the edges opposite them overlap. Using the Archi-
medean axiom, Enriques and Amaldi reduced to this case the one in which the opposite

1
Tarski [1924] 2014b (see page 79); Enriques and Amaldi [1903] 1916, chapter 5. For biographical informa-
tion about Enriques and Amaldi, and portraits, see section 9.2.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 45
DOI 10.1007/978-1-4939-1474-6_4, © Springer Science+Business Media New York 2014
46 4 Area, Volume, Measure

edges are collinear but do not overlap. From there they pursued a chain of arguments
to show that every polygonal region is equivalent to a rectangle with a prescribed base.2

The simplest result in the opposite direction is that equivalent parallelograms P and
Q with congruent altitudes have congruent bases. To prove this, Enriques and Amaldi
assumed the contrary, that the base of P might be congruent to a proper part of that of
Q. On that assumption, they showed how to construct within Q a parallelogram PU con-
gruent to P, which would then be equivalent to the whole parallelogram Q. At the
analogous step in the classical development, Euclid had noted that this situation would
contradict the (vague) common notion, “the whole is greater than the part,” and thus he
rejected the contrary hypothesis. The Italian mathematics teacher Antonio De Zolt anal-
yzed this argument, and devised a more precise geometric assumption to complete it:
If we divide a given polygon into polygonal parts, then the union of all but one
of these parts is not equivalent to the given polygon.
Enriques and Amaldi stated and used that as an axiom.3

From this basis, Enriques and Amaldi derived various familiar results in the theory
of area, including the Pythagorean theorem and its converse. Like Euclid, they avoided
defining area as a function with numerical values. They would only do so three chapters
later, after using the Archimedean and continuity axioms to establish the use of real
numbers to measure lengths of segments: the area of a polygonal region P should be
the length of the altitude of a rectangle with base of length 1 and equivalent to P.

Soon after citing Enriques and Amaldi [1903] 1916, Tarski suggested that readers
should compare the related exposition in David Hilbert’s book Grundlagen der Geometrie.4
Without using any Archimedean or continuity axiom, Hilbert set up an arithmetic of
segments. He constructed an area function by defining the area of a triangular region
to be half the product of an edge and the corresponding altitude, and that of a polygonal
region P to be the sum of the areas of the subregions in a triangulation of P. This
required proving first that the familiar triangle-area formula does not depend on which
edge is selected as the base, and that any two triangulations of P determine the same
area sum. The former results from consideration of similar triangles; the latter requires
a complicated but elementary process that Hilbert only sketched. Equivalent polygonal
regions clearly have equal areas: that implies De Zolt’s axiom.

2
The Archimedean axiom says that given two segments S and T, one can always find a segment U
consisting of congruent copies of S laid end to end such that T is congruent to a subsegment of U.
Details of the arguments depend on those of the definition of polygonal region. The present editors have
this definition in mind: a triangular region is the union of a triangle and its interior; a polygonal region
is a union of a finite family T of triangular regions whose intersections consist of common edges or ver-
tices. Such a family T is called a triangulation of the region.
3
Euclid [1908] 1956, volume 1, 155, 336 (proposition 39); De Zolt 1881 and 1883. The displayed wording
is from Enriques and Amaldi [1903] 1916, chapter 5; a requirement that the interiors of the parts be dis-
joint was evidently implicit in their notion of divide. For information about De Zolt, see the box on
page 47.
4
Hilbert [1899] 1922, chapter 4.
4.1 Area and Volume 47

Antonio De Zolt was born in 1847 in Conegliano, a town north of Venice, then part of the Austrian
Empire. He earned the laureate at the University of Turin in 1872. De Zolt taught for forty-four
years at the Reale Liceo Parini in Milan, and was a long-time member of the Italian Mathesis society.
He published several mathematical works, including 1881 and 1883 booklets on the theory of area
and volume. De Zolt died in Milan in 1926.*

* Tricomi 1962, 45.

Following Euclid, Hilbert also introduced a weaker version of the equivalence rela-
tion: two polygonal regions A and B are equicomplementable 5 if there exist equivalent
polygonal regions Ai and Bi for i = 1, ... , n, such that A and all the Ai have disjoint
interiors, B and all the Bi have disjoint interiors, and the union of A and all the Ai
is equivalent to that of B and all the Bi . It is visually apparent that parallelograms with
congruent bases and altitudes are equicomplementable. Clearly, equicomplementable
polygonal regions have equal areas. Hilbert showed that the converse is true as well,
using an argument based entirely on his planar incidence, order, congruence, and parallel
axioms, without appealing to the Archimedean principle.

Hilbert’s argument that polygonal regions with equal areas are equicomplementable
can be modified easily to show that they are in fact equivalent, provided the Archimedean
axiom is used to show that parallelograms with congruent bases and altitudes are equiv-
alent. The first known statement of this result was a question posed in 1814 by William
Wallace; it was solved the same year by John Lowry. However, it is often called the
Bolyai–Gerwien theorem, referring to the work of the Hungarian and Prussian mathema-
ticians Bolyai Farkas and P. Gerwien in the 1830s.6 Hilbert showed that proving it
actually requires appeal to the Archimedean principle.

Much of this area theory has an analogous counterpart for volumes of polyhedral
regions. But the argument that polyhedra with equal volumes should be equicomplement-
able fails. Hilbert did not discuss that in the 1899 first edition of Grundlagen der
Geometrie, but posed it the next year as the third of his famous list of problems for the
twentieth century: to find inequivalent tetrahedra with equal base areas and congruent
altitudes. His student Max Dehn did so, using advanced algebraic methods. Hilbert cited
Dehn’s work in all later editions of the Grundlagen.7

5
Hilbert’s term was inhaltsgleich; for equivalence, he used Flächengleichheit. In later editions, he
employed other words.
6
Lowry and Wallace 1814; F. Bolyai [1832] 1904; Gerwien 1833a. This theorem is discussed in detail in
Bartocci 2012, 29–39, and in Amaldi [1900] 1914, §3. It could have been proved easily in Enriques and
Amaldi [1903] 1916 by noting that if polygonal regions P and Q have the same area, then their equivalent
rectangles with a specified base must have the same area and hence the same altitude: they must be
congruent, and so P and Q must be equivalent. In 1833b, Gerwien derived an analogous theorem for
spherical polygons.
7
Hilbert [1900] 2000. Dehn 1901–1902. For a particularly simple proof, consult Benko 2007, §2.
48 4 Area, Volume, Measure

These considerations show that advanced mathematical techniques can yield deep
results about the theories of area and volume. As noted in the previous chapter, Tarski
had acquired the background to consider them in that light. His university course work
had provided essential background for the mathematics that he would soon develop, in
collaboration with Stefan Banach, for investigating advanced aspects of those theories.
Tarski had studied advanced geometry and analysis with Zygmunt Janiszewski and Stefan
Mazurkiewicz; topology with Kazimierz Kuratowski; and higher algebra and analysis,
set theory, and measure theory with Wacaw Sierpiski. Sierpiski evidently influenced
Tarski to study in detail Felix Hausdorff’s 1914a book Grundzüge der Mengenlehre.
Sierpiski had recently published a survey of the use of the axiom of choice in set theory
and analysis. For a decade he and Mazurkiewicz had been considering seemingly paradox-
ical constructions in set theory. For example, they had discovered two disjoint unbounded
subsets A and B of the plane, each of which is congruent to the union A F B —this
shows that De Zolt’s axiom does not necessarily hold when the restriction to polygonal
regions is weakened. Their excitement about such results surely influenced those who
attended their lectures. Tarski would soon immerse himself in this research area. His
work would involve the most advanced results in measure theory.8

Wacaw Sierpiski
in 1928

8
See Feferman and Feferman 2004, 48, and Tarski 1924f. Tarski always recommended Hausdorff 1914a
to his own students, even decades later. Sierpiski 1918; Mazurkiewicz and Sierpiski 1914. For infor-
mation about Hausdorff, Janiszewski, Kuratowski, Mazurkiewicz, and Sierpiski, consult the boxes on
pages 69, 14, 72, 14, and 49, respectively; their portraits are displayed near those pages.
4.1 Area and Volume 49

Wacaw SierpiĔski was born in 1882 in Warsaw, which was then in the Russian Empire. The son
of a physician, he attended Russian-language schools and was graduated in 1899 from the highly
regarded Fifth Gymnasium. He then entered the University of Warsaw, also a Russian institution,
studied with the number theorist Georgy F. Voronoy, and won a prize for his first research project.
Sierpiski earned a degree there in 1904 and became a teacher of mathematics and physics at a girls’
secondary school. He participated in the school strikes associated with the 1905 Revolution, resigned,
moved to Cracow—then in the Austrian Empire—and entered the university there to complete work
for the doctorate under Stanisaw Zaremba. Sierpiski returned to Warsaw to teach in a Polish-
language school permitted as a concession by the government. He continued research in number
theory, and became very active in the movement that would soon establish academic and scientific
institutions for an independent Poland.

In 1908–1909 Sierpiski became a docent at the University of Lwów, also in the Austrian Empire.
There he switched his emphasis to set theory, and gave one of the very first university courses on
that subject. He took Austrian citizenship and was appointed professor in 1910. That same year,
he married Anna Kazimiera Leniewska, who hailed from Belarus; they had one child, their son
Mieczysaw. During the next years, the first three of Sierpiski’s many monographs appeared, on
number theory, on irrational numbers, and on set theory—one of the first comprehensive works on
that subject. He brought Zygmunt Janiszewski to Lwów as his assistant, and supervised the doctoral
research of Stefan Mazurkiewicz. At the outbreak of war in 1914, Sierpiski and his wife were
visiting her family, and were interned in Moscow. There he met and began a long research collabora-
tion with Nikolai N. Luzin on set theory.

Immediately after the war, Sierpiski returned to Lwów, then soon became professor at the new
Polish university in Warsaw. With Janiszewski, Mazurkiewicz, and several others, he had been plan-
ning to make Poland a world center of mathematical research, specializing in logic, topology, and
foundations of mathematics. During the next few years, they succeeded. Alfred Tarski was one of
their first students. During 1919–1920, Sierpiski joined other mathematicians in a successful project
to decode enemy military communications. They played a major role in the decisive defeat of the
Bolshevik army in summer 1920 on the banks of the Vistula River only a few miles from Warsaw.*

During 1920–1939 Sierpiski continued research, publication, and academic leadership at a frantic
pace. He was a pioneer in set theory, and one of the most prolific of all mathematicians: he published
more than seven hundred research papers and fifty books. For his research on equidecomposability,
Tarski cited as principal background source the 1923 second edition of Sierpiski’s set theory text.†
Later French and English versions have served as standard references for several generations of
mathematicians. By 1939 Sierpiski was world-renowned.

That year, the German invasion and occupation destroyed the Polish academic community. Sierpiski
survived as a minor office worker, but continued research and gave private clandestine mathematics
courses. Late in the war, his own home and library were destroyed, and he was exiled to Cracow.
After the war, Sierpiski returned to Warsaw as a leader in the reconstruction of Polish academic
and mathematical institutions. He continued research, and received many honors, national and
worldwide. Sierpiski died in Warsaw in 1969.‡

* See section 1.1 of the present book for more details and references about this period.
† Tarski [1924] 2014b, translated in chapter 5: see page 79.
‡ Fryde 1964; Kuratowski 1974; Kuratowski 1980, 167–173; Schinzel 1974, 2009.
4.2 The Measure Problem 51

4.2 The Measure Problem

Questions about area, volume, and measure were in the air not only of Tarski’s circle in
Warsaw, but also among the mathematicians in Lwów. In 1919, as part of his application
for the venia legendi at the Polytechnic School in Lwów, Antoni omnicki had published
a paper on axioms for magnitude. Using techniques of postulate theory, he investigated
the interdependence of some familiar statements involving a system of three relations
y, {, { —for example, the transitive and trichotomy laws. There are many common
interpretations of these symbols. For example, the statement A { B & B y C might
mean that geometric figures A and B should be equidecomposable and B equidecom-
posable with a proper part of figure C, or that propositions A and B should be logically
equivalent and B a logical consequence of proposition C. In recent decades, mathemati-
cians had considered questions about this sort of statement in many contexts.9

In January 1922, omnicki, by then a professor, presented a teacher-training course,

On Equivalence of Plane Figures, and was preparing a new edition of his secondary-school
geometry text, which would be published a year later. That same month, at a meeting
of the Polish Philosophical Society in Lwów, he presented a research paper on this sub-
ject.10 He claimed that with the equidecomposibility interpretation, all required theorems
about area can be derived from a certain system of axioms of the sort mentioned in the
previous paragraph, which included De Zolt’s axiom in the form A y B _ A { / B.
omnicki noted that the remaining axioms in that system, with the propositional interpre-
tation, are all valid in logic, but De Zolt’s is not. Therefore, De Zolt’s axiom is independ-
ent of the others, and deriving it from geometric principles should require a method
different from those used to derive the others.

Among the attendees who discussed omnicki’s paper was his assistant, Stefan Banach,
a self-educated but already noted mathematician who would be awarded the doctorate
from the University of Lwów later that year. Soon, Banach considered this subject from
the point of view of measure theory, and investigated the family M of point sets to which
De Zolt’s postulate could apply. (By the example of Mazurkiewicz and Sierpiski 1914
mentioned in section 4.1, some restriction is necessary.) The next paragraphs display the
connection with measure theory.

Suppose relations { and y are defined between members of a family M of point

sets, so that
A, B  M & A y B _ (} C  M )( A { C & C ‚ B & B – C  M ), (1)

9
Among the works cited in omnicki 1919 was a major 1916 paper by Jan ukasiewicz. Apparently, neither
of those scholars was aware at that time of the similar work of the Peano school in Italy a generation
earlier: for example, Bettazzi 1890. For more information on omnicki, see the box on page 53.
10
omnicki 1923; Maligranda 2008, 105. Describing his teacher-training course, omnicki surveyed the
material covered in section 4.1 and cited for background the Italian source Enriques and Amaldi [1903]
1913 and its Polish translation [1903] 1916.
52 4 Area, Volume, Measure

and a real-valued function m, applicable to all members of M , enjoys the following

analogues of familiar properties of the measures of length, area, and volume:
A, C  M & A { C _ mA = mC (2)
DM & D=
/ I _ mD > 0 (3)
C, D  M & C  D = I _ m( C F D) = mC + mD (4)
Property (4) is called finite additivity because it entails that the sum of the measures of
any finite number of disjoint sets is the measure of their union. De Zolt’s axiom for the
family M can then be derived by contraposition: given A, B  M such that A y B, find
C by property (1) and let B – C = D, so that D  M , D = / I, C  D = I, C F D = B,
and by (2) to (4),
mA = mC < mC + mD = m(C F D) = mB;
/ mB and by (1), A {
thus, mA = / B.

In his pioneering lectures early in the century, Henri Lebesgue had introduced a
central focus of measure theory, the measure problem: to assign, to all bounded subsets
of the line, nonnegative numbers called their measures, such that two congruent sets
always have the same measure, the unit interval has measure 1, and the measure of the
union of any finite or infinite bounded sequence of disjoint sets is the sum of their individ-
ual measures. (This last requirement, called full additivity, is stronger than the finite
additivity mentioned in the previous paragraph.) Analogous problems can be stated for
higher-dimensional point sets. Such a measure would coincide with the length, area, or
volume of any set for which one of those notions had already been defined. Clearly, any
solution of the measure problem would yield a verification of De Zolt’s axiom for the
family M of all point sets. However, that verification would use only finite, not full,
additivity.

Lebesgue had defined for each dimension a fully additive measure function that applied
to all bounded measurable point sets, which was sufficient for many arguments in anal-
ysis. Almost immediately, using the axiom of choice, Giuseppe Vitali showed that no fully
additive measure function can apply to all bounded point sets in any dimension. Through
an even more complicated construction, Felix Hausdorff showed that no finitely additive
measure can apply to all bounded point sets in any dimension greater than two. The cor-
responding statements for dimensions one and two remained undecided.11

In a 1923 paper, Banach settled those two cases of the measure problem that Hausdorff
had left open: finitely additive measures m applying to all bounded point sets are indeed
possible in dimensions one and two. This result is discussed in detail in the following
paragraphs.

11
Lebesgue 1904, 103; Vitali 1905; Hausdorff 1914a, section 10.1 and its appendix, 399–403, 469–472.
See also Chatterji 2002, §4, and the boxes on pages 67 and 75 about Hausdorff’s “paradox” and a non-
measurable set.
4.2 The Measure Problem 53

Antoni omnicki was born in 1881 in Lwów, then in the Austrian Empire. His father was a
secondary-school teacher. Antoni completed schooling there and studied at the University of Lwów,
earning the doctorate in 1903 with a dissertation on hypergeometric functions. Until 1919 he taught
in secondary schools. He obtained a government stipend to study at Göttingen during 1913–1914,
and began publishing research on axiomatics and in analysis. On that basis he won the venia legendi
at the Polytechnic School at Lwów, and became a docent, then professor there in 1921. Stefan Banach
served briefly as his assistant while completing doctoral studies at the university. omnicki became
an expert on mathematical cartography and radio-navigation, and wrote popular secondary-school
texts. He was a gifted organizer, and served several years in the administration of the Polytechnic.
In 1930 omnicki won eight months’ paid research leave to visit mathematicians in Rome, Paris,
Göttingen, and Berlin. With many other leading citizens, he was murdered by the Germans in 1941,
immediately after they invaded Lwów.*

* Jakimowicz and Miranowicz 2007, 90; Maligranda 2008.

Alfred Tarski and Stefan Banach met in Lwów during May 1923, at the First Polish
Philosophical Congress.12 Each one interested in applications of set theory to geometry,
they began a collaboration. In his [1924] 2014b paper, Tarski showed that two polygonal
regions A and B are equivalent precisely when they have the same area. Equivalent
means that A and B can be divided into the same finite number of subfigures such that
corresponding subfigures of A and B are congruent. Departing from the conventions
of elementary geometry, Tarski allowed the term “figure” to encompass all plane point
sets, and required the subfigures to be entirely disjoint. The collaborative paper Banach
and Tarski [1924] 2014 pursued this notion of equivalence in much greater detail, par-
ticularly in three dimensions. It contained a very startling theorem, often termed
“paradoxical”: two balls with any radii whatever are equivalent—for example, the earth
and a marble. Chapters 5 and 6 contain the first published English translations of Tarski
[1924] 2014b and Banach and Tarski [1924] 2014.

An essential step in Tarski’s argument in [1924] 2014b involved appeal to the following
result in Banach 1923:
Theorem III*. To each bounded set A of points in the plane, one can assign
a nonnegative real number m( A), so that if A and B are bounded point
sets, then
(1) m( A) = m(B) if A = M[B] for some isometry M,
(2) m( A F B) = m( A) + m(B) if A and B are disjoint,
(3) m( A) is the Lebesgue measure of A if A is Lebesgue measurable.
Property (1) is called invariance under the group of isometries. Property (2) is finite
additivity. Banach’s theorem III* was preceded by his analogous theorem I, for subsets
of the circumference of a circle. These are cases of the Lebesgue–Hausdorff measure prob-
lem mentioned earlier.13

12
See the last paragraph of chapter 3. For information about Banach and a portrait, see pages 57 and 56.
13
Tarski [1924] 2014b, §2: see section 5.2. Banach 1923, 30–31.
54 4 Area, Volume, Measure

Tarski noted hyperbolically that the proof of Banach’s theorem used “the entire
apparatus of contemporary mathematical knowledge ... .” 14 It did require virtually all of
measure theory, and the rudiments of functional analysis. An essential step in Banach’s
argument was his theorem 16, an early version of what is now called the Hahn–Banach
theorem; for its proof he used well-ordering and recursion on infinite ordinals, where
later mathematicians would use a set-theoretic maximal principle. Thus, the proof of
Banach’s theorem required the axiom of choice.15 Tarski and others investigated various
aspects of Banach’s theorem in several later papers (see sections 8.6–8.8). Nevertheless,
Wacaw Sierpiski remarked in 1954, “the demonstration is still very troublesome.” 16
Banach’s proof of his theorem I, about a circle, is understandable. It has been presented
a bit more clearly by Adriaan C. Zaanen, preceded by an excellent exposition of the
required measure theory and analysis. Sierpiski was probably lamenting the fact that
Banach had specified a construction for the finitely additive function m in his theorem
III*, about the plane, but no argument that it satisfies his requirements (1) to (3). The
boxes on this page and the next contain an outline of such an argument, based on material
in the books by Zaanen and by Stan Wagon.17

Banach’s Theorem III*: outline of proof. For the required framework in measure theory and the
Hahn–Banach theorem, consult Zaanen 1958. This discussion is based on Zaanen’s proof (§28) of
Banach’s theorem I, his remark that it can be generalized to a partial proof of theorem III*, and on
ideas in Wagon 1993, chapter 10, especially theorems 10.4(a), 10.4(e), and 10.8. In turn, Wagon’s
discussion was based on Banach 1923, Neumann 1929, and Mycielski 1979.

For any dimension n, consider the set T = { x  Ün : ~i (0  xi < 1)}; regard it as an n-torus, with
the left and right ends of the xi intervals identified when convenient. Let B be the vector space
of all bounded real-valued functions on T. The subspace L of its Lebesgue-measurable functions
is invariant under all translations W of T: f % W  L whenever f  L. The Lebesgue integral ³, a
nonnegative linear functional on L, is also invariant under all such W: ³T ( f % W) = ³T f for all f  L.

Given f  B and c1 , ... , ck  T, note that the set A( f, k, c1 , ... , ck ) of all averages (1/ k) 6 j f ( x + cj)
for x  T is bounded because f is. Define
Jf = inf sup A( f, k, c , ... , c ).
c1 , ... , ck  T x  T 1 k

Following Banach 1923, Zaanen showed that the functional J is nonnegative and subadditive: for
every f  B, if f ( x)  0 for all x  T, then Jf  0; and for all f, g  B and any scalar t  0,

J( f + g)  Jf + Jg , J(t f ) = t Jf.
[continued]

14
Tarski [1924] 2014b, introduction (chapter 5, page 79).
15
Banach 1923, 19. For the Hahn–Banach theorem, see Zaanen 1983, 112.
16
Sierpiski 1954, 97.
17
Zaanen 1983, chapters 1–9, especially §28; Wagon 1993, chapter 10. The argument in the last paragraph
of the box can be used in any dimension. Applied just before the specialization to the case m = 2, with
Q in place of QU, it would yield an additive extension of P to the family of all bounded subsets of m,
invariant under translations and under reflections across hyperplanes perpendicular to the coordinate
axes. In the case m = 1, that is theorem I of Banach 1923.
4.2 The Measure Problem 55

[continued]
Moreover, ³T f  Jf for all f  L. By the Hahn–Banach theorem, ³ can be extended to a nonnegative
linear functional I on B such that If  Jf. Zaanen used this inequality and properties of the
functional J to show that I is invariant under all translations of T.

Tesselate Ün with copies of T translated parallel to the coordinate axes. Any bounded S I Ün
is the union of finitely many disjoint pieces Sj , each one the intersection of S with the image of
T under a unique translation Wj of Ün. Let Q(S) be the sum of the values Ifj , where fj is the
characteristic function of the set Wj–1 [Sj ]. It is straightforward to show that Q is finitely additive.
If S is Lebesgue measurable, then so are each set Sj and function f j , and hence
Ifj = ³fj = P(Wj–1 [Sj ]) = P(Sj ),
the Lebesgue measure; and since that is finitely additive, Q(S) = P(S). A similar argument shows
that this finitely additive extension Q of P to the family of all bounded subsets of Ün is invariant
under each translation M of Ün. (Subdivide S more finely, so that each set M[Sj] also lies wholly
within some tesellation cell.)

Now set the dimension n = 2. Let C be the circle with circumference 1 about the origin. The
previous discussion, with n = 1, implies the existence of a nonnegative linear functional H on the
space of all bounded real-valued functions on C, invariant under rotations of C and agreeing with
³ on Lebesgue-measurable functions. In particular, if f is a constant function on C, then Hf = c.

For each t  C, let Ut be the counterclockwise rotation about the origin that maps the point
 1/(2S), 0 to t. For each bounded S I Ü2, define fS : C  Ü by setting fS (t) = Q( U–1
t [S]); then let
QU(S) = HfS . If S is Lebesgue measurable, then so is U–1 t [S], and hence Q( Ut [S]) = P(S), the
–1

Lebesgue measure. Thus, fS is constant and QU(S) = HfS = P(S): that is, QU extends P. Its finite
additivity follows from that of Q and the linearity of H. Invariance of QU under translations follows
from the fact that the translations W form a normal subgroup of the isometry group of the plane:
for each bounded S I Ü2 and each t  C,
fW [S] (t) = Q( U–1
t W[S]) = Q( Ut W Ut Ut [S]) = Q( Ut [S]) = fS (t)
–1 –1 –1

t W Ut . Invariance of QU under rotations M about the

because Q is invariant under the translation U–1
origin follows from that of H under M–1 : for each S and t,
1
t M[S]) = Q U M1 ( t ) [ S ]
fM [S] (t) = Q( U–1 = fS(M–1(t)),
and therefore
QU(M[S]) = HfM[S ] = H( fS % M–1) = HfS = QU(S).
Invariance under these rotations and all translations implies that QU is invariant under all direct
isometries.

One more step yields an extension Q2 of P that is invariant under all plane isometries. Choose
a reflection V across some line. For each bounded S I Ü2, set
Q2(S) = ½ QU(S) + ½ QU(V[S]).
That Q2 is finitely additive, extends P, and is invariant under V follows straightforwardly. Invari-
ance under direct isometries M follows from the fact that they also form a normal subgroup:
Q2(M[S]) = ½ QU(M[S]) + ½ QU(V M[S]) = ½ QU(M[S]) + ½ QU(V M V V[S])
= ½ QU(S) + ½ QU(V[S]) = Q2(S),
because QU is invariant under the direct isometries M and V M V.
Stefan Banach in 1919
4.2 The Measure Problem 57

Stefan Banach was born in 1892 in Cracow, then part of the Austrian Empire.* His parents were
Stefan Greczek, a soldier and office worker, and Katarzyna Banach, a maid. Several months after
his birth, the parents entrusted the boy’s care to a foster mother, Franciszka Powa, who ran a
successful laundry business. Greczek provided financial support until 1910, when Stefan Banach was
graduated from Cracow’s classical Henryk Sienkiewicz Gimnazjum. A family friend had given him
French lessons during those years, and he had taught himself real analysis. Banach then moved to
Lwów to study engineering at its Polytechnic School. He had to support himself by tutoring mathe-
matics, and consequently required four years to finish a two-year program. At the onset of war in
1914 he returned to Cracow. Physically disqualified from military service, he worked in construction,
continued tutoring, and pursued his private study of mathematics. Banach married ucja Braus, a
secretary, in 1919. They had one child, a son born in 1922.

By chance, in 1916 Banach met the mathematician Hugo Steinhaus, who recognized his mathematical
talent. In 1918 Banach began publishing original research results, particularly involving measure
theory, and in 1919, he helped found the Cracow Mathematical Society. Steinhaus soon became a
professor at the University of Lwów, and secured for Banach a position as academic and personal
assistant to the mathematician Antoni omnicki at the Lwów Polytechnic. Although Banach lacked
a university degree, he was allowed to submit a dissertation to the university, which awarded him
the doctorate in 1922. That work introduced the structures now called Banach spaces, and presented
their fundamental properties, including the famous Banach fixed-point theorem. A period of intense
research led to Banach’s 1923 paper on the measure problem, a position at the University of Lwów,
and almost immediate promotion to full professor. Banach and Steinhaus founded the Lwów school
of mathematical research. Banach met Alfred Tarski at a conference in Lwów. A year later, they
published their famous 1924 paper on seemingly paradoxical decompositions of solid bodies, translated
in chapter 6.

During the 1920s and 1930s, Banach continued major research activity and teaching in real analysis
and mathematical physics. A series of successful school and university textbooks, some written with
Wacaw Sierpiski and Wodzimierz Stoek, provided extra income.† Banach’s [1931] 1932 monograph
on functional analysis made him famous as the founder of that discipline. Several of his doctoral
students themselves achieved worldwide reputations in that and related fields.

After the Soviet Union invaded Poland in 1939, Banach served as dean at the university. He escaped
the assassination of leaders when the Germans invaded in 1941, but lost his livelihood, and for three
years survived only as a flea-feeder for Rudolf Weigl’s laboratory in Lwów, which manufactured
typhus vaccine for the Germans. Infected fleas were enclosed in a matchbox taped to his body with
a screened opening next to the skin. They would feed for a week and then were harvested for the anti-
bodies.‡ This and hunger destroyed his health. After the war he tried to return to professional life,
but died of lung cancer in 1945.

* Consult Kaua 1996, and Jakimowicz and Miranowicz 2007 for further information, and Duda 2009, which
corrects some errors. See also Albiski 1976.
† For example, Banach, Sierpiski, and Stoek 1933.
‡ Baumslag 2005, 133–134; Szybalski 1999; Waszyski 1996. On the one hand, Weigl collaborated closely
with the Nazis and used coerced human subjects. On the other, his flea-feeders were exempt from close Nazi
scrutiny, and he managed to save from extermination many Lwów intellectuals and Jews, including some other
noted mathematicians. After the war he was honored by the State of Israel with the designation “Righteous
among the Nations.”
58 4 Area, Volume, Measure

4.3 Equidecomposibility in the Plane

In 1924, the year of his doctorate, Alfred Tarski published two abstracts and five research
papers. One of the abstracts was related to his paper on well-ordered sets, translated in
chapter 2. The other began a new research thread, axiomatics of arithmetic. One of the
papers was directly related to his dissertation. Two others were concerned with deep
questions in pure set theory.18 The remaining two papers, applications of set theory to
geometry, are translated in chapters 5 and 6. The latter, his joint paper with Stefan
Banach on surprising counterintuitive decompositions of solid figures, quickly became
famous.

This section is concerned with the former: the paper Tarski [1924] 2014b, On the
Equivalence of Polygons, translated in chapter 5. It appeared in volume 2 of Przegld
matematyczno-fizyczny. That journal, published by the commercial firm Ksinica-Atlas,
had been founded the previous year by editors Wadysaw Wojtowicz and Stefan
Straszewicz. At a time when the relationship of schools and society was undergoing sig-
nificant transformation, they aimed
to establish a link between secondary schools and higher education and research. On the one
hand, changes in basic concepts and theories of science cannot be without influence on
secondary-school methods and curriculum. On the other, nothing raises the teaching level in
schools as much as a teacher’s scientific investigations, however modest in scope. ... Our second
goal, no less important, will be to elicit an exchange of thoughts between teachers ... 19

Mathematical-Physical Review

Scientific and Pedagogical Quarterly

Published in Warsaw
under the Editorship of
W. Wojtowicz and S. Straszewicz

18
The five publications just mentioned are Tarski 1924c, 1924a, 1924d, 1924b, and 1924e, respectively, all
in French. The first was discussed briefly in section 1.2, and the last three in chapter 3.
19
Wojtowicz and Straszewicz 1923. Wojtowicz was a senior member of the Polish mathematics community;
Straszewicz, a new professor at the Warsaw Technical University. This and other Polish mathematics-
education journals of the day are discussed in sections 9.6 and 9.7. The latter includes a portrait and
biographical sketch of Straszewicz. For more information on Wojtowicz see a box in section 9.3.
4.3 Equidecomposibility in the Plane 59

Besides Tarski’s paper, this volume of the journal included articles by

• government educator Antoni M. Rusiecki, on methods for teaching arithmetic;
• recent Lwów physics doctorate Leopold Infeld, on presenting the theory of
similar triangles in secondary school;
• the noted Italian mathematician Giuseppe Peano, on presenting n-place
logarithms without using irrational numbers;
• Lwów professor Hugo Steinhaus, on reasoning about plane areas;
• and several other teachers, professors, and laboratory scientists.20
There were also reviews of physics and mathematics texts from elementary to research
level. Tarski’s article was probably the most advanced in this volume. He selected
material from his joint work with Banach that, with patient and utterly clear exposition,
would be accessible to teachers and students and introduce them to new and advanced
aspects of the material they dealt with in classes.

Tarski’s paper is about the decomposition of plane figures. According to the theory
presented in elementary geometry texts and described in section 4.1, two polygonal
regions have the same area if and only if they are unions of the same finite number of
polygonal subregions with disjoint interiors such that corresponding subregions are con-
gruent. Tarski’s goal was the same result for a different notion of decomposition: two
polygonal regions should have the same area if and only if they are unions of the same
finite number of disjoint subsets such that corresponding subsets are congruent. Tarski
permitted subsets of any sort, not just polygonal regions; but he imposed the stronger
requirement that these subsets themselves be disjoint, not just their interiors. Tarski’s
illustrated discussion of the difference between these notions of decomposition is at the
top of page 85 of the present book.

As mentioned earlier, Tarski assumed familiarity with the elementary geometry

described in section 4.1. Because he employed complicated dissections into arbitrary
disjoint subregions, he made sophisticated use of elementary set theory. Although com-
mon in the mathematical research of the previous two decades, these techniques had been
presented in only a few monographs. Tarski cited just one for this background material:
Sierpiski 1923, the recent, greatly expanded second Polish edition of a book first pub-
lished in 1912. For his own study, Tarski had been also been using Hausdorff 1914a.
Reporting on all the set-theory texts available at that time, Walter Purkert has written,

20
Rusiecki 1924—Rusiecki plays a major role later in this book: see section 9.7. Infeld 1924—Infeld became
an influential physicist both in Poland and worldwide; for a description of his situation as schoolteacher,
see Infeld 1980, book I. Peano 1924—late in his long career at the University of Turin, Peano was heavily
involved in mathematics education: see Marchisotto and Smith 2007, subsection 5.1.5. Steinhaus 1924.
60 4 Area, Volume, Measure

... even through the first decade of the twentieth century no independent textbook had
appeared that systematically and comprehensively presented the entire subject of set theory
as then understood ... . Even the little Polish book published in 1912 by Wacaw Sierpiski
cannot be considered such a work. ... It was above all intended for beginners. Finally, with
Hausdorff 1914a the first textbook was available that treated systematically all subjects then
regarded as part of set theory. ... Felix Hausdorff’s work is also the last such book, at least in
a commonly understood language.21

Tarski also used one recent advanced result in set theory, Banach’s 1923 solution of the
measure problem in dimensions one and two. Tarski noted that its proof had required
“the entirety of the apparatus of contemporary mathematical knowledge ... .” 22 That result
was described in detail in the previous section. But Tarski’s paper used only the result,
not Banach’s argument.

After defining two point sets A and B to be equivalent—abbreviated A { B —if they

are unions of the same finite number of disjoint subsets, such that corresponding subsets
are congruent, Tarski proved some elementary properties of this notion. For the one
troublesome property, transitivity of equivalence, he presented the usual “double net-
work” argument in complete detail. He showed easily that disjoint unions of equal num-
bers of pairwise equivalent sets are equivalent. From Banach’s theorem he inferred the
existence of a finitely additive function m that assigns, to all subsets of the plane,
nonnegative real numbers called their measures so that congruent sets have equal
measures and the measure of a polygonal region is its area. Half of Tarski’s target result
followed directly from this: by the finite additivity of m and its invariance under
congruence, two equivalent polygonal regions have equal measures, and thus equal areas.

Tarski then turned to the converse result, that two polygonal regions with equal areas
should in fact be equivalent. He used sophisticated set-theoretic manipulation, but not
the advanced mathematics, including the axiom of choice, required to justify Banach’s
theorem. The crucial parts of his argument are two lemmas:23 if A is a plane set with
an interior point, and B is a plane set disjoint from A, then A and A F B are equiva-
lent if
(1) B consists of a single point, or
(2) B is a segment, with or without one or both endpoints.

21
Feferman and Feferman 2004, 47. Purkert 2002, 47–49. Hausdorff treated both general set theory and
point-set theory. Purkert considered the 1913 book by Arthur Schönflies somewhat comparable to Haus-
dorff 1914a, but noted that for Schönflies, general set theory was just one of many parts of mathematics,
whereas Hausdorff regarded it as the foundation for the whole discipline.
22
Tarski [1924] 2014b, §2: see page 79 of the present text.
23
Closely related to, but not identical with, lemmas I and II in section 5.2, these results have been rephrased
to facilitate this brief summary.
4.3 Equidecomposibility in the Plane 61

These can be extended recursively to apply when B is a finite union of such points and
segments. Now, if two polygonal regions V and W have the same area, then they are
equivalent in the elementary sense by the theory in section 4.1; each is thus the union
of the same finite number of polygonal subregions with disjoint interiors, and correspond-
ing subregions are congruent. The interiors of corresponding subregions must also be
congruent, so the unions A and B of the interiors of the subregions of V and of W
must be equivalent. Therefore,
V = A F (V – A) { A { B { B F (W – B) = W
by the recursive extension of lemma 2, because each of V – A and W – B is the union
of finitely many segments.

Tarski’s proofs of the two crucial lemmas used the same technique, as follows. First,
given a set A with an interior point, choose a disk K I A, a point p0 on its circumfer-
ence, and a rotation U about its center through an irrational number of degrees. For
k = 1, 2, ... let pk = U k ( p0 ). To prove lemma 1, translate the single point in B to p0 ,
map the set C of all points pk to the set C – { p0 } by the rotation U, and apply the
identity mapping to the rest of A; these mappings show that A { A F B. To prove lemma
2, first assume that B is an open segment congruent to R0 , the radius of K through
p0 , without its endpoints. Map B to R0 by an appropriate direct motion, map the union
C of all the sets U k [R0 ] to the set C – R0 by the rotation U, and apply the identity
mapping to the rest of A: again, A { A F B. The cases in which B contains one or both
of its endpoints can then be covered by referring to lemma 1. The general case is covered
by applying this one recursively: divide B into a finite number of segments, each
sufficiently short.

Presumably to make contact with contemporary presentations of elementary geometry

(see section 4.1), Tarski arranged the argument outlined in the previous two paragraphs
to include a proof of an analogue of De Zolt’s axiom in this context; that was not neces-
sary for the main points just emphasized.

Toward the end of his paper, Tarski noted that the theory of polyhedral volume is quite
different: some three-dimensional statements analogous to major results in plane geom-
etry are false. He alluded to the 1901–1902 result of Max Dehn already mentioned in
section 4.1—not directly, but by citing a section about it in Ugo Amaldi’s major [1900]
1914 expository paper, On the Theory of Equivalence.
62 4 Area, Volume, Measure

Next, Tarski considered the area of a disk C. He referred again to Amaldi [1900] 1914
for a proof that C is not equivalent in the elementary sense to any polygon P. Just what
sort of subregions Amaldi allowed is not clear, but his argument was essentially this: if
C and P were the unions of equal finite numbers of subregions with disjoint interiors,
and all corresponding pairs of subregions were congruent, then for the subregions of P
the sum of the lengths of boundary segments that are concave from within should be
equal to the analogous sum for the convex segments, because the boundary segments of
P itself are straight; but for the subregions of C the concave sum would exceed the
convex sum by the circumference of C, contradicting their congruence with the corre-
sponding subregions of P. Amaldi’s argument evidently stemmed from work of the
Hungarian mathematician Móricz Réthy about equivalence of figures bounded by curves
of a more general nature, and Amaldi described some controversy concerning that.24 By
allowing arbitrary subregions in the paper under discussion, Tarski sidestepped part of
this difficulty. He posed the question, can this theory be extended to arbitrary plane
regions bounded by closed curves?

In conclusion, Tarski posed another question, his circle-squaring problem,25 which

became famous and remained open for decades:
Can a disk and a polygonal region with equal areas be equivalent in Tarski’s
sense?

As mentioned in section 4.1, questions about equidecomposability in the elementary

sense attracted attention of Euclid in ancient times, of William Wallace, John Lowry,
Bolyai Farkas, and P. Gerwien in the early 1800s, and of Réthy, David Hilbert, and Dehn
around 1900. It has since spawned a large literature, including the works of Tarski and
Henryk Moese described and translated in chapter 7.26 The set-theoretic version of this
notion emerged in the early twentieth century via Hausdorff, Stefan Mazurkiewicz, and
Sierpiski, and solidified in the 1924 papers of Banach and Tarski. An even larger and
more varied literature stems from those. In major part, it is concerned with extending
the theory of “paradoxical” decompositions in solid geometry as presented in Hausdorff
1914a and in the Banach–Tarski paper discussed in more detail in section 4.4 and
translated in chapter 6.

24
Amaldi [1900] 1914, §6–7. Réthy 1891; see also Kötter 1891.
25
Tarski also published this question as Tarski 1925a, in the problem section of Fundamenta Mathematicae.
The Collected Papers volume Tarski 1986a included the wrong text for it. Here is the complete original:
38) Un carré et un cercle dont les aires sont égales peuvent-ils être décomposés en un nombre
fini de sous-ensembles disjoints respectivement congruents? Problème de M. Tarski.
26
Tarski [1931] 2014a and [1932] 2014d and Moese [1932] 2014.
4.3 Equidecomposibility in the Plane 63

Later studies of set-theoretic decomposition of plane figures, particularly those con-

cerned with the circle-squaring problem, really stem from Tarski’s paper translated in
chapter 5.27 That work was not reviewed in the Jahrbuch über die Fortschritte der Mathe-
matik, probably because that publication accorded low priority to journals intended
primarily for teachers. In any case, Przegld matematyczno-fizyczny was not a familiar
journal, nor its language widely understood. In the subsequent literature, Tarski [1924]
2014b has been mentioned only rarely, and then obliquely. For example, in his elegant
little book on this subject, Sierpiski merely suggested that readers compare Tarski [1924]
2014b when following a reference to Banach and Tarski [1924] 2014. Tarski’s use of a
rotation through an angle with irrational degree measure is echoed precisely in Herbert
Meschkowski’s well-known book on geometry problems, but without mentioning Tarski.
Each of these authors stated the circle-squaring problem, but without specific reference
to Tarski.28

For a survey of related results in the six decades after Banach’s and Tarski’s work,
consult sections 8.6–8.8 and the literature cited there, particularly the excellent exposition
published in 1993 by Stan Wagon. In 1990, to considerable acclaim, Miklós Laczkowicz
published an affirmative solution to the circle-squaring problem, making heavy use of
results in graph theory and number theory. Remarkably, his proof showed that any two
polygonal or circular regions with the same area are the unions of the same finite number
n of subsets such that corresponding subsets are related by translations. No rotations
or reflections are required! Laczkowicz estimated that with his method, n  10 50. For
the background and methods of proof of this result, and related open questions, consult
another excellent exposition, Gardner and Wagon 1989, entitled
At Long Last, the Circle Has Been Squared!

27
That problem is not mentioned in Banach and Tarski [1924] 2014.
28
Sierpiski 1954, 260. See section 5.3 of the present book and compare Meschkowski 1966, 139–141. Its
title Ist die Quadratur des Kreises lösbar? notwithstanding, Menger 1934 does not mention Tarski’s
problem explicitly.
64 4 Area, Volume, Measure

4.4 The Banach –Tarski Theorems

This section describes the 1924 paper by Stefan Banach and Alfred Tarski, translated in
chapter 6: On Decomposition of Point Sets into Respectively Congruent Parts. It appeared
originally in volume 6 of Fundamenta Mathematicae.

This journal was founded as part of Zygmunt Janiszewski’s plan for establishing
Poland as a world center for mathematical research. He had proposed
to establish a journal, a strictly scholarly one, devoted primarily to one branch of mathematics,
in which we have many truly creative and distinguished mathematicians. ... This journal ...
would accept articles in each of the four languages recognized as international in mathema-
tics. ... A publication such as this ... would find readers everywhere. ... The very existence and
distribution of such a journal published in Warsaw would bear testimony to our cultural life.29

The specified branch was foundations of mathematics, together with set theory, including
its applications. These aims were controversial. Clearly they were nationalistic, but the
insistence on publishing in international languages, not Polish, could be interpreted as
counter to that. And it was not clear to some that enough material would be submitted
to ensure high quality of the published papers.30 The first volume of the journal appeared
in 1920, published by Pastwowe Wydawnictwo Naukowe (the State Scientific Publisher)
with Janiszewski as editor-in-chief. It was the very first specialized mathematics journal.
One of its twenty-four articles was concerned with pure set theory; the others, with appli-
cations in general topology, measure theory, and real analysis. The eight authors were
all Polish: this volume served to introduce them to the world. Five were established pro-
fessors; the other three were emerging scholars. A problem section in this volume pre-
sented ten questions on these same subjects, all of which led to substantial later research.
Their proposers included two additional young Polish scholars and a Russian.31 Unfor-
tunately, Janiszewski died during the 1920 influenza epidemic. Stefan Mazurkiewicz and
Wacaw Sierpiski succeeded him in the editorship.

Besides the Banach–Tarski paper, the 1924 volume 6 of Fundamenta Mathematicae

included Tarski’s 1924b study of set-theoretic definitions of finiteness, and twenty-two
additional articles and three problems whose subject matter was distributed like those
in volume 1. Their seventeen authors included four from Russia, two from the United
States, and one each from France and Germany. The journal had quickly become
international, premier in its field. It is still publishing in 2014.32

29
See section 1.1. Janiszewski [1918] 1968, 116–117. The languages were English, French, German, and
Italian.
30
Kuratowski 1980, 32–37.
31
The third problem, due to the Russian Mikhail Suslin, became famous. For example, see the conclusion
of section 15.12. The problem solution cannot be derived from standard set theory. An appendix to the
celebratory 1935 “second edition” of volume 1 of Fundamenta Mathematicae reviewed later work arising
from these articles: see Braun, Szpilrajn, and Kuratowski 1935.
32
For further information about Fundamenta Mathematicae, consult Duda 1996.
4.4 The Banach–Tarski Theorems 65

The Banach–Tarski paper extended to the context of general set theory the underpin-
nings of the decomposition theory for planar point sets presented in Tarski [1924] 2014b,
the paper translated in chapter 5. The planar theory showed that results familiar in
elementary geometry about decompositions of polygonal regions into equal finite numbers
of subregions with disjoint interiors such that corresponding pairs of subregions are
congruent, continue to hold when the notion of decomposition is modified. The
Banach–Tarski paper is about the decomposition of arbitrary point sets into equal finite
numbers of disjoint subsets such that corresponding pairs of subsets are congruent. These
sets are assumed only to lie in a space S equipped with a metric: a real-valued function33
on S × S. Sets are called congruent if they correspond under a transformation of S that
preserves the metric. In parallel, Banach and Tarski developed the corresponding theory
for decompositions into denumerable infinities of subsets.

In contrast to the planar case, the results for finite decompositions in Euclidean spaces
of three and more dimensions are surprising. Banach and Tarski first presented Felix
Hausdorff’s startling “paradoxical” decomposition of the surface S of a sphere into four
disjoint parts: one that is countable and another that is congruent to the remaining two
and to their union. This is the construction that Hausdorff had used to show that the
measure problem has no solution in three or more dimensions (see section 4.2 and the
box on page 67). By elaborating this construction, Banach and Tarski discovered for gen-
eral figures in three or more dimensions results that are utterly counterintuitive. In
particular, they showed that any two bounded point sets with nonempty interiors are
finitely equidecomposable in the manner just described: a ball the size of an atom is equi-
decomposable with one enclosing our galaxy.34 By Banach’s positive solution of the
measure problem for lower dimensions, the analogous statements for the line and the
plane are false. But Banach and Tarski did prove the corresponding statements for denu-
merably infinite decompositions of linear and planar sets.

In their introduction these authors emphasized that use of the axiom of choice was cru-
cial in proving these results, even for the special cases of polygonal and polyhedral regions.
Tarski was then heavily involved in other research specifically about that axiom: his
1924e paper On Some Theorems Equivalent to the Axiom of Choice had appeared in the
preceding volume of Fundamenta Mathematicae. The axiom was controversial, and some
mathematicians regarded the counterintuitive results of Hausdorff, Banach, and Tarski
as evidence against it. For example, the eminent French mathematician Émile Borel
railed about Hausdorff ’s construction:

33
Banach and Tarski developed much of this theory without assuming additional properties of the space
S and its metric.
34
According to a conversation reported in Moore 1982, 264, Banach and Tarski independently discovered
instances of this theorem, but the general result was due to Tarski.
66 4 Area, Volume, Measure

Thus if one designates by a, b, c the probabilities that a point of S should belong to

A, B, C and admits that the probability that a point should belong to a set E does not change
under rotation, ... one has the contradictory equations a + b + c = 1, a = b + c, a = b, a = c.
This contradiction has its origin in the application ... of [Ernst] Zermelo’s axiom of
choice. ... The paradox arises because A is not defined, in the logical and precise sense of the
word defined. If one scorns precision and logic, one is led to contradictions.35

But the axiom was just as heavily involved in showing the absence of “paradoxical” results
in one and two dimensions. Reflecting on the proofs, Banach and Tarski even suggested
that it was more so. Decades later, logicians investigated deeply the extent to which the
axiom was required: chapter 13 of Stan Wagon’s 1993 monograph is devoted to that
investigation.

Banach and Tarski began §1 of their paper by defining the relations = ~ of congruence,
= =
and f and d of equivalence by finite and by denumerable decomposition, and describ-
ing their elementary properties. They presented theorems in pairs: for example, theo-
rems 1 and 1U stated analogous properties of the relations = f and =d . To guide readers
in constructing proofs, they presented several in detail, including proofs of the transitivity
of = f and of the equivalence of disjoint unions of equal finite numbers of pairwise
equivalent sets. This section of their paper constitutes a kit of tools for gracefully
describing and analyzing complicated mappings in later arguments. For example, directly
applying a hypothesis A = f B would involve specifying subdivisions of A and of B and
the separate isometries under which they correspond. That would lead to an awkward
argument if a previously specified subset C of A had to be related to B. From this
hypothesis, however, their corollary 7 would provide in one step a proper subset D of
B equivalent to C.

Theorems 8 and 8U are analogues of the familiar Cantor–Bernstein theorem of set

theory, in the new contexts of equivalence by finite and denumerable decomposition. Like
the proof of the familiar theorem, those of the new results are surprisingly complicated.
Banach noticed that analogous theorems are useful in other contexts, too. Therefore, he
published a generalization separately in 1924, about relations with two properties that
he called D and E. The Banach–Tarski paper merely referred to that. For a statement
and proof of Banach’s theorem about D, E-relations, see the box on page 70. Various
familiar corollaries of the Cantor–Bernstein theorem also have analogues in the context
of D, E-relations.
[The narrative continues after the box on page 70.]

35
Borel 1914, 255–256. Hausdorff concluded instead that probabilities cannot be measured in this way.
Just as Hausdorff’s construction was included at the very end of his 1914a treatise, so are these the last
words of Borel 1914.
4.4 The Banach–Tarski Theorems 67

Hausdorff ’s “Paradox.” In 1914, Felix Hausdorff reported* a startling decomposition of the surface
S of the unit sphere into four disjoint parts Q, A, B, C: the first is countable, and the remaining three
all congruent, and congruent to B F C as well. This is often termed “paradoxical”; it implies that
no finitely additive extension m of the familiar area function, which is invariant under rotations of
S, can assign nonnegative values—measures—to all subsets of S. To see this, note first that such
an m would have to be monotonic: X I Y I S _ m( X )  m( X ) + m(Y – X ) = m(Y ). If m(Q) > 0,
a contradiction would ensue: there is a rotation U of S such that the sets Un[ Q] for all natural num-
bers n are disjoint; but they all have the same measure, and therefore the measure of a union of more
than m(S)/m(Q) of them would exceed m(S). To find such a U, choose an axis a through the
origin but not intersecting Q, and a rotation U about a through an angle incommensurable with
the differences in longitude, with respect to a, between points of Q. By the congruences and
additivity, m( A) = m(B F C) = m(B) + m(C) = m( A) + m( A): that is, m( A) = 0. The contradiction
m(S) = m(Q) + m( A) + m(B) + m(C) = m(Q) + 3 m( A) = 0 then shows that no such m can exist.
As a corollary, there can be no solution of the three-dimensional measure problem discussed in section
4.2, for otherwise, the measure m( X ) of an arbitrary subset X I S could be computed as three times
the “volume” of the union of the radii through the points in X.

To construct the decomposition, Hausdorff worked in three-dimensional coordinate geometry with

conventional x, y, z axes. He considered 180( and 120( rotations M about the z-axis and \ about
an axis in the xz-plane inclined at an angle T from the z-axis. These generate a group G of rotations
represented by formulas such as L, M, \, \ 2, products of any two of those save the first,
\ 1 M \ 2 M " M \ n with m1 , ... , mn = 1 or 2, and others like the last, preceded and/or succeeded by
m m m

M. By meticulously considering the matrices corresponding to these formulas, Hausdorff determined

that unless T should belong to a particular countable set of angles, G will be a free group: no two
of these formulas will represent the same rotation. He then selected any T outside that set.
– – –
Hausdorff presented the following algorithm to partition G into three disjoint subsets A , B, C . First,
– – –
put L  A and M, \  B and \  C. If the formula for a rotation U  G begins with \ or \ 2, put
2
– – – – –
MU in B or A according to whether U  A or not. If the formula begins with M, put \ U in B, C,
– – – – – – –
or A, and \ U in C, A , or B, according to whether U  A , B, or C. This classifies products of
2

any two of M, \, \ 2 as shown in Hausdorff’s table reproduced on the facing page. That classification
– – – –
can be used with a simple recursive argument to show that { MU : U  A } = B F C, { \U : U  A } =
– – –
B, and { \ U : U  A } = C.
2

Let Q be the (countable) set of intersections of S with the axes of the nontrivial rotations in G.
The family of all orbits Pu = { U( u) : U G } for u  S – Q partitions that set. Hausdorff appealed
to the axiom of choice to form a set M consisting of a single element from each orbit, then defined
– – –
the desired sets A, B, C as the unions of the images U[M] for U  A, B, C, respectively. Each pair
of these sets is disjoint. To see this, suppose u  A  B, for example; then u = U( v) = V(w) for some
– –
v, w  M, U  A, and V B ; therefore, v and w would belong to the same orbit, and hence coincide,
so that U( v) = V( v), v = U–1V( v), and thus v  Q, contradicting M I S – Q. From the construction
– – –
of A, B, C, it follows that M[ A] = B F C, \[ A] = B, and \ 2 [ A] = C: for example,
– –
\[ A] = { \U( u) : u  M & U  A } = { U( u) : u  M & U  B } = C.
Thus, A, B, C are all congruent, and congruent to B F C as well.

* Hausdorff 1914a, appendix, 469–472; 1914b. Hausdorff achieved this result at the very time his 1914a book
was going into print. The main points of his proof are presented here. For further details, consult Stromberg
1979 and Chatterji 2001. For a biographical sketch of Hausdorff, see the box on page 69.
68 4 Area, Volume, Measure

Felix Hausdorff
around 1918

– L M\M
A M \, M \ 2, \ 2M $$$
–
B M, \ M\ M, \ M\, \ M\
2 2
$$$
– \ 2
\M
C \ M\, \ M\
2 2 2
$$$

Hausdorff’s Classification36 of Rotations in G

36
Hausdorff 1914a, 472. See the box on page 67. The original row labels were A, B, C. The order of the
products has been reversed here and in that box, to agree with today’s practice.
4.4 The Banach–Tarski Theorems 69

Felix Hausdorff was born into a wealthy Jewish commercial family in 1868 in Breslau, which was
then in Prussia, but now is called Wrocaw, in Poland. He grew up in Leipzig. Although the family
was seriously religious, Hausdorff became a serious and critical agnostic. His first major interest was
music; he started to study composition. However, during university studies in Leipzig, Freiburg,
and Berlin, he switched to astronomy. Hausdorff earned the doctorate in 1895, then spent a year
in military service, and two as mathematician for the observatory back in Leipzig. He habilitated
there in astronomy, to be qualified to teach. His first publications, in that subject, were of little
consequence. In 1899 Hausdorff married Charlotte Goldschmidt, the daughter of a physician. She
was Lutheran; their only child, a daughter, was born in 1900.*

During these years Hausdorff’s scientific interests changed to mathematics. He published research
on probability theory, including the first explicit treatment of conditional probability and some funda-
mental applications to insurance problems. Then he entered the field of set theory and topology,
which would be his main concern for many years. In 1901 Hausdorff taught one of the first courses
ever in set theory, and started a research program that would provide much of the structure of these
new subjects. He was appointed professor in 1901, but outside the regular system of ranks, and
against major, explicitly antisemitic, opposition.

Hausdorff was very active in artistic and literary circles. Under the pseudonym Paul Mongré he
published philosophical studies, poetry, two novels, and a popular play that satirized the aristocratic
honor code. The main German encyclopedia after World War I featured him equally as mathematician
and writer.† He pursued research and teaching as an amateur—teaching was financially unnecessary.
Never comfortable at Leipzig, and realizing that his scientific future required a conventional career
path, he won regular professorships at Bonn in 1910, then at Greifswald, on the Baltic, in 1913.

There he completed his 1914a book on set theory, one of the first on that subject. A recent appraisal
characterized it as “a pioneering achievement that paved the way for the development of modern
mathematics.” Its appendix presented the complicated “paradoxical” decomposition of a three-
dimensional point set that provided the negative solution to Henri Lebesgue’s measure problem in
dimensions beyond two and led to even more striking decompositions in Banach and Tarski [1924]
2014.‡

Greifswald was isolated, and Hausdorff sometimes its only mathematician. After World War I he
turned somewhat to research in analysis. His 1919 work on the dimension and outer measure of point
sets led decades later to the development of fractal geometry. He returned to Bonn in 1921. There
he applied measure theory in his 1923 course on axiomatic probability theory. Hausdorff’s research
papers and lectures were famous for clarity and elegance, and his colleagues treasured his critical
but genial nature. During the 1920s and 1930s he published new editions of his set theory book,
changing with the times but undiminished in influence.

Hausdorff retired in 1935. Reluctant to leave his familiar surroundings, he made no major attempt
to emigrate, but remained in Bonn as his colleagues and former students departed, died in battle,
or just disappeared. Finally, under threat of deportation to a death camp, Hausdorff, his wife, and
her sister committed suicide together in 1942.

* Czy 1994, 1. For further information, consult Dierkesmann et al. 1967, Mehrtens 1980, and Purkert 2008.
† Mon gré means my taste. Grosse Brockhaus 1931.
‡ Purkert 2008, 45. Hausdorff 1914a is described in section 4.3, and the “paradox” in the box on page 67.
Banach and Tarski [1924] 2014 is translated in chapter 6.
70 4 Area, Volume, Measure

D, E - Relations. In 1924, Stefan Banach introduced* a new general set-theoretic notion and a related
modification of the familiar Cantor–Bernstein theorem. This enabled major simplifications to the
decomposition theory presented in Banach and Tarski [1924] 2014 and translated in chapter 6.
Similar benefits would accrue in pure set theory and topology. Banach considered two properties that
might be satisifed by relations R between sets:
D. whenever A R B, there exists a bijection M : A  B such that X R M[ X ] for each X I A;
E. if A1  A 2 = I = B1  B 2 and A1 R B1 and A 2 R B 2 , then ( A1 F A 2) R (B1 F B 2).

Theorem (Cantor–Bernstein–Banach). Should R have property D and there exist AU I A and

BU I B such that A R BU and AU R B, then there will exist A1 , A 2 I A and B1 , B2 I B such that
A1  A 2 = I = B1  B 2 , A1 F A 2 = A, B1 F B 2 = B, A1 R B1 , and A 2 R B 2 . Thus, if R should also
have property E, then A R B.
Proof. By property D, there exist bijections M : A  BU and \ : AU  B such that X R M[ X ] and
Y R \[Y ] whenever X I A and Y I AU. Set C0 = A – AU and Cn +1 = \ –1M[Cn ] for each natural
number n, then let C be the union of all of the sets Cn . It is straightforward to show that A – C I
AU and \[A – C] = B – M[C]. Finally, set A1 = C, A 2 = A – C, B1 = M[C], and B2 = B – M[C]. The
restriction of M is a bijection from A1 to B1 and that of \, from A 2 to B2 .

* Banach 1924. The proof given here is adapted from Wagon 1993, 25.

[continued from page 66]

With theorems 10 and 10U, Banach and Tarski departed from intuitively comfortable
material. They showed that if a set A is equivalent by decomposition to a union A F Bk
for each k = 1, ... , n, then A is equivalent to A F B1 F ... F Bn . That result is analogous
to one about cardinal equivalence of infinite sets. Under the same analogy, the familiar
theorem that 2 m = 2 n implies m = n for all cardinals m and n corresponds to
theorems 11 and 11U: if disjoint pairs of sets A1 , A2 and B1 , B2 are each equivalent by
decomposition, and so are their unions, then A1 , B1 are equivalent. In 1922, Sierpiski
had provided for the theorem on cardinals a proof that appeared to be generalizable; in
1924, Kazimierz Kuratowski modified that to prove a corresponding result in the context
of the D, E-relations of Banach 1924. To prove theorem 11, Banach and Tarski referred to
Kuratowski 1924. For a statement and proof of the result that they used,37 see the box
on page 71. Theorem 11, in turn, is used just once, much later, in the proof that two point
sets on the surface of a sphere are equivalent if they contain interior points.

Section 1 of the Banach–Tarski paper concludes with several results relating equide-
composability and certain families of “small” sets in a Euclidean space: the families B
of bounded sets, N of nowhere dense sets, P of sets with zero Peano– Jordan content,
F of sets of first category, and L of sets with zero Lebesgue measure.38 Each of these
families contains every subset of and any set congruent to any of its members. Thus, if

37
Although Banach and Tarski described Kuratowski’s theorem in terms of D, E-relations, they made no
use of that idea in their proof.
38
For these concepts see Birkhoff 1948, chapter 11; compare the German and French terminology in Haus-
dorff 1914a, §§7.8, 10.2–10.3, and in Kuratowski 1958, §36.
4.4 The Banach–Tarski Theorems 71

a set A belongs to one of them and is equidecomposable with a set B, the subsets of
A and B in the decomposition all belong to that family. Moreover, these families are
all closed under formation of finite unions, and F and L , under denumerably infinite
unions. Thus the set B will belong to the same family as A, as long as, for families
B , N , and P, the sets A and B are finitely equidecomposable.

Section 2A of the Banach–Tarski paper contains its main results for dimensions one
and two. These had been presented for polygons already in Tarski [1924] 2014b, described
in section 4.3 and translated in chapter 5. The first main result is theorem 16: bounded
Lebesgue-measurable subsets of the line or plane, if equivalent by finite decomposition,
must have the same measure. Its proof depended on theorem III* of Banach 1923,
described in the box on page 54, and thus made essential use of the axiom of choice.
Banach and Tarski noted that its converse is false. In fact, Hausdorff had defined a
[The narrative continues on page 73.]

Kuratowski’s Theorem. Kazimierz Kuratowski provided* a technical result for the theory of decom-
position of point sets presented in Banach and Tarski [1924] 2014 and translated in chapter 6: given
E = M F N = P F Q, M  N = I = P  Q, and bijections M : M  N and \ : P  Q,
there exist disjoint M1 , M2 , M3 , M4 and disjoint Q1 , Q2 , Q3 , Q4 such that
M = M1 F M2 F M3 F M4 , Q = Q1 F Q2 F Q3 F Q4 ,
Q1 = M1 , Q2 = \[M2 ], Q3 = M[M3 ], Q4 = \ M[M4 ].
To prove this Kuratowski first extended M and \ to involutory permutations of E: each extension
is the union of the given function and its inverse. Next he considered for each x  E the set Cx of
all images of x under elements of the permutation group generated by the extended M and \. Each
Cx intersects each of M, N, P, Q. The family of all sets Cx is a partition of E. By the axiom of choice,
there is a function that associates to each Cx a member of Cx  M —call it the chosen one. Each
set Cx consists of these members:

x3 = M \ M(x0 )
x2 = \ M(x0 )
x1 = M(x0 )
x0 —the chosen member
x–1 = \(x0 )
x–2 = M \(x0 )
x–3 = \ M \(x0 )

By a recursive argument, Kuratowski verified that if xk = x l then k and l have the same parity
and x k + j = x l+ j for all integers j. Now consider an arbitrary m  M. It belongs to a unique set
Cx , with chosen member x 0 , and thus m = xk for some k. Kuratowski partitioned M as follows:
if k is odd and m  Q, set m  M1 ; if k is even and xk+1  Q, set m  M3 ;
if k is odd and m Õ Q, set m  M2 ; if k is even and xk+1 Õ Q, set m  M4 .
Then he defined Q1 to Q4 as in the statement of this result. It is straightforward to show that these
sets have the desired properties.

* Kuratowski 1924.
 Kazimierz Kuratowski
around 1915

Kazimierz Kuratowski was born in 1896 in Warsaw, then part of the Russian Empire, the son of a
prominent lawyer. Kazimierz completed secondary school there in 1913, then started engineering
studies at the University of Glasgow. He was home for vacation in August 1914 when World War
I began. That disrupted his studies, but a year later, he was able to enter the new Polish University
of Warsaw, to study mathematics and logic. His first published work stemmed from the logic seminar
of Jan ukasiewicz, but in 1917 he began to work with Stefan Mazurkiewicz and Zygmunt Janiszew-
ski on topology; the latter mentored his doctoral research. Wacaw Sierpiski assumed that role after
Janiszewski died in 1920. After earning the doctorate the next year, Kuratowski was soon appointed
docent. During the next years he originated many concepts now fundamental in set theory, general
topology, and graph theory, including the now familiar definition of ordered pair, formulation of a
maximal principle equivalent to the axiom of choice, axiomatization of topology in terms of the closure
operator, and a simple characterization of planar graphs. Alfred Tarski attended his lectures and
collaborated with him over many years.
In 1927 Kuratowski was appointed full professor at the Lwów Polytechnic University. In 1933 he
published a text, Topologie I, that served for several decades as a standard source. The next year,
he returned to the University of Warsaw and soon became head of its mathematics program. He
helped plan further expansion of Polish mathematical research, but that was thwarted by World War
II. During the German occupation he taught in the clandestine academic system. Afterward, he was
instrumental in the rebirth of Polish mathematics, and helped found the Polish Institute of Mathe-
matics, which he served as director during 1948–1967. It is now part of the Academy of Sciences.
Author of the 1980 history of Polish mathematics that is a principal source for the present book,
Kazimierz Kuratowski was a mathematical statesman of world stature. He died in 1980.*

*Arboleda 1990, Kuratowski 1981.

4.4 The Banach–Tarski Theorems 73

[continued from page 71]

certain subset G of the closed unit interval I that is a countable union of closed
intervals, and showed that it has Lebesgue measure less than one.39 Its complement F =
I – G is thus nowhere dense, with positive measure m. An interval with length m is
not nowhere dense; by the previous paragraph, it cannot be equivalent to F by finite
decomposition. Further, the Cartesian product F × I has planar measure m, but is
nowhere dense and thus not equivalent by finite decomposition to any rectangle with
area m.

Lemma 17 in §2A of the Banach–Tarski Paper is the crucial step in proceeding from
equivalence by decomposition in the elementary sense to the set-theoretic decomposition
employed there: if A is a planar set with an interior point, and B is the union of a finite
number of segments, then A= f A F B. This entails that a polygonal region and its interior
are equivalent by finite decomposition. Immediately afterward, the authors remarked
that an analogous one-dimensional result can be obtained by replacing the words planar
and segments by linear and points. They must have intended that to be theorem 18,
because there is no result with that number! In fact, it is lemma I in Tarski [1924] 2014b
and section 5.3; a proof is provided there.40 Section 2A reaches the same conclusion as
Tarski [1924] 2014b: two polygonal regions are equivalent by (set-theoretic) finite
decomposition if and only if they have the same area.

Section 2B of the Banach–Tarski paper contains its main results for dimensions greater
than two. The authors presented them for dimension three, but described an easy way
to generalize them to higher dimensions. Their argument started with Hausdorff’s
“paradoxical” 1914a decomposition of the surface of a sphere into four disjoint parts: a
countable set and three congruent parts, one of which is also congruent to the union of
the remaining two. (See the box on page 67.) The authors then divided the solid sphere
S into five disjoint parts: the center p and four others, B, C, D, E, obtained by adjoining
to the surface parts the half-open radii containing their points. In summary,
S = B F E F { p} F D F C E is a union of countably many half-open radii.
DFC= ~B= ~C= ~D B, E, { p}, D, and C are disjoint.
Of the uncountably many rotations of S, only countably many map one of those radii
to another. Tarski and Banach let U be one that does not, so that
~ U[E] ‚ B F C F D.
E=
Then they applied corollary 7, mentioned earlier, to determine a set G and a point q
such that
U[E] =
f G ‚ C q  G – C.

39
Theorem 16 is a generalization of theorem 7 in Tarski [1924] 2014b (section 5.2); their proofs are the
same. Hausdorff 1914a, 418.
40
Lemma 17 is the pair of lemmas II, III of Tarski [1924] 2014b (section 5.3). The proofs there are more
detailed than those in the Banach–Tarski paper.
74 4 Area, Volume, Measure

Finally, they defined

A1 = B F E F { p} A2 = D F G F { q}.
From the five lines of formulas just displayed and the transitivity of equivalence by finite
decomposition, Banach and Tarski derived the remarkable lemma 21, A1 = =
f S f A2 :
every solid sphere S contains two disjoint subsets, each equivalent to S by finite
decomposition.

Again applying corollary 7 and a corollary of theorem 8, the analogue of the Cantor–
Bernstein theorem mentioned earlier, Banach and Tarski quickly deduced lemma 22:
every solid sphere S is equivalent by finite decomposition to its union with another
sphere congruent to S. Using theorem 10, the tool presented earlier for handling finite
unions, they deduced lemma 23: every bounded set containing a solid sphere S is
equivalent to S by finite decomposition. Finally, they presented their culminating result,
any two sets that contain interior points are equivalent by finite decomposition: they
must contain congruent solid spheres, to which they are equivalent by lemma 23.

In §2C of their paper, Banach and Tarski derived a theorem for the surface S of a
sphere completely analogous to their main three-dimensional result: any two subsets of
S with interior points are equivalent by finite decomposition. The arguments are similar,
with two exceptions. First, lemma 27, which says that S can be decomposed into two
disjoint sets, each equivalent to S, is both stronger and easier to prove than the analo-
gous lemma 21 for three dimensions, because the simpler definition A 2 = S – A1 is
available. Second, lemma 30, which says that every subset of S with an interior point
is equivalent to S itself, has no analogue in three dimensions. Its proof required
Gerwien’s theorem that two spherical polygons on S are equivalent if they have the same
area, and the theorem of Kuratowski that was presented as a tool in §1 of their paper.41

Section 3 of the Banach–Tarski paper had two goals:

• theorem 35—in any dimension, any two sets with interior points are equiva-
lent by denumerable decomposition; and
• to relate decomposibility questions to Lebesgue measure.
Their first step toward theorem 35 was to recall the main feature of Hausdorff’s con-
struction of a nonmeasurable set: the half-open unit interval A is the union of a denum-
erable infinity of disjoint subsets equivalent to each other by finite decomposition. (See
the box on page 75.) Applying an affine transformation to this A shows that every half-
open interval A has that property. Next, Banach and Tarski applied that result in a
straightforward manner to prove lemma 32: the union of a denumerable infinity of
disjoint intervals each congruent to a single interval is equivalent to that interval by
denumerable decomposition. They presented that result in one dimension, then indicated
how to generalize it to higher dimensions, where an interval is a Cartesian product of one-
dimensional intervals.

41
Gerwien 1833b; see Boltyanski 1978, §§7–8, for a related discussion. For Kuratowski’s theorem, see the
box on page 71.
4.4 The Banach–Tarski Theorems 75

Nonmeasurable Set. Early in the twentieth century, Henri Lebesgue posed the measure problem:
to assign, to all bounded subsets of the line, nonnegative numbers called their measures, such that
two congruent sets always have the same measure, the unit interval has measure 1, and the measure
of the union of any finite or infinite bounded sequence of disjoint sets is the sum of their individual
measures. In an obscure 1905 publication, Giuseppe Vitali showed that no measure for all bounded
subsets is possible. A decade later, Felix Hausdorff published a version of Vitali’s argument, as
follows (1914a, 401–402).
Let D be any irrational number in the half-open unit interval A. The sets
Px = { ( x + kD) mod 1 : k  ] }
for x  A form a partition of A. By the axiom of choice, there is a subset A 0 I A containing exactly
one element of each Px . For each k  ] define
A k = { ( x + kD) mod 1 : x  A 0 }.
These sets also partition A. The elements of A k are obtained from those of A 0 by adding either
kD or kD – 1; therefore, A 0 and A k are either congruent or equivalent by decomposition into pairs
of mutually congruent subsets. That is, A is the union of a denumerable infinity of disjoint subsets
A k , equivalent to each other by finite decomposition. Lebesgue’s requirements would entail that
these A k should all have the same measure m. If m = 0, then A would have measure zero,
contrary to assumption. Thus m > 0, which would contradict the finiteness of the measure of A.

Banach and Tarski quickly reached their target, theorem 35, as follows. Considering
a tiling of the entire space E, they showed that the union of the sets in the previous
paragraph is equivalent to E by denumerable decomposition. Next, they noted that any
set A with an interior point must include a nonempty interval, which is equivalent to
the entire space. By Banach’s adaptation of the Cantor–Bernstein theorem (see the box
on D, E-relations on page 70), A must be equivalent to E. It follows immediately that
any two sets with interior points must be equivalent by denumerable decomposition.

In their concluding discussion of measure-theoretic considerations, Banach and Tarski

used the extension of Lebesgue-measure theory formulated by Hausdorff to accommodate
unbounded sets. They defined two point sets A and B to be almost equivalent by
denumerable decomposition—abbreviated A = p B —if denumerably equivalent sets can
be obtained by removing subsets with Lebesgue measure zero from A and B. They
proved that this reflexive and symmetric relation on point sets is also transitive.42

Their main result about this notion, theorem 41, states that in any dimension, any two
sets with positive Lebesgue inner measure are almost equivalent by denumerable decom-
position. Aiming toward this, Banach and Tarski first proved lemma 37: a Lebesgue-
measurable set A and an open set B of equal measure can be approximated arbitrarily
closely (with regard to their measures) by closed subsets that are equivalent by finite

42
Hausdorff 1914a, 416. Zaanen 1983, chapters 1–9, is an excellent general introduction to this material.
The letter p in the abbreviation stands for the French word presque for almost.
76 4 Area, Volume, Measure

decomposition.43 They iterated that process, approximating each of A and B by unions

of sequences of disjoint subsets, each corresponding pair of which are equivalent by finite
decomposition. They showed that the relative complements of these unions have measure
zero, and hence that A and B are almost equivalent by denumerable decomposition.
They applied this result in the case in which B is the entire space E, and made some
minor adjustments, to prove lemma 39: to any bounded set A with positive Lebesgue
measure corresponds a set C of measure zero such that A F C and E are equivalent
by denumerable decomposition. Some further adjustments led to their main results,
lemma 40 and theorem 41: every set A with positive Lebesgue inner measure is almost
equivalent to E by denumerable decomposition, and hence any two such sets are almost
equivalent to each other.44

According to the final theorem 42 of the Banach–Tarski paper, any two sets are almost
equivalent by denumerable decomposition into Lebesgue-measurable components if and
only if they are Lebesgue measurable with the same measure. The authors presented no
proof of that result, but indicated that it could be obtained by analyzing the previous
arguments.

43
Their demonstration used the regularity theorem: any Lebesgue-measurable set A can be approximated
by an open set G J A and a closed set F I A such that the measure of G – F is smaller than any
previously selected positive number. This stems from the fact that the measure of A is simultaneously
the supremum of the (outer) measures of the closed polygonal or polyhedral subsets of A and the infimum
of the (inner) measures of the open sets containing A. See Hausdorff 1914a, 408–413.
44
Wagon has presented a different proof that applies just as well to subsets of the surface of a sphere (1993,
theorem 9.17, 140–141).
 5
On the Equivalence
of Polygons (1924)

This chapter contains an English translation of Alfred Tarski’s paper O równowanoci

wieloktów, [1924] 2014b, written as he was completing his doctoral studies. It appeared
in volume 24 of the journal Przegl d matematyczno-fizyczny. This is its first translation.
A description of the journal, background for the paper, and a summary are provided in
sections 4.1–4.3.

The translation is meant to be as faithful as possible to the original. Its only inten-
tional modernizations are punctuation and some changes in symbols, where Tarski’s con-
flict with others used throughout this book. Bibliographic references and some personal
names have been adjusted to conform with the conventions used here. The original paper
sometimes employed barely discernible a u g m e n t e d l e t t e r s p a c i n g to emphasize
a phrase.1 In some cases the translation uses italics instead; in others, this emphasis has
been suppressed. As an aspect of adjusting punctuation, the editors greatly increased use
of white space to enhance visual organization of the paper. All [square] brackets in the
translation enclose editorial comments. Those are inserted, usually as footnotes, to
indicate changes in notation and explain passages that seem obscure.

1
In the letter translated in section 15.12, Tarski conveyed his dislike of that style.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 77
DOI 10.1007/978-1-4939-1474-6_5, © Springer Science+Business Media New York 2014
Journal Containing Alfred Tarski’s Paper
On the Equivalence of Polygons
ALFRED TARSKI

On the Equivalence of Polygons

In elementary geometry,2 we call two polygons equivalent if it is possible to divide them

into the same finite number of respectively congruent polygons not having common
interior points. In the theory of the equivalence of polygons, the following statement,
usually accepted without proof in elementary geometry and sometimes called De Zolt’s
axiom, plays a fundamental role:
If polygon V is a part of polygon W, then these polygons are not equivalent.
As is well known, David Hilbert showed 3 that the preceding statement can be proved with
the help of axioms usually cited in elementary geometry textbooks. Because of the
difficulty of that proof, however, one does not make use of it in a secondary-school class.

Relying on De Zolt’s axiom, among others, it is possible in the theory of mensuration

to prove the following theorem, which provides a necessary and sufficient condition for
the equivalence of two polygons:
In order for polygons V and W to be equivalent, it is necessary and sufficient
that they have equal areas.

The question arises, do the [italicized] formulations of both statements above remain
true sentences if equivalence is understood in a broader sense than it usually is in elemen-
tary geometry: that is to say, if 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.

In the present article I show that this question ought to be given an affirmative answer.
It is also interesting that the proofs of both of these very straightforward statements, the
first of which may seem almost obvious, rely on results obtained by Prof. Stefan Banach
with the aid of the entire apparatus of contemporary mathematical knowledge: in par-
ticular, with the help of the so-called axiom of choice.4

2
The definitions and theorems of elementary geometry to which I refer in the present article can be found,
for instance, in the textbook Enriques and Amaldi [1903] 1916.
3
See Hilbert [1899] 1922.
4
Banach 1923 [discussed in section 4.2 of the present book]. All those notions and principles of set theory
to which I refer in the present article—just a few—are contained in the book Sierpiski 1923.
80 5 On the Equivalence of Polygons

Notation

By means of letters p, q, s, ... I denote points, while the letters A, B, K, P, V, ... [denote] geometric
figures—that is, point sets.
The symbol A F B denotes the union of sets A and B: that is, the set consisting of all those
points that belong either to set A or to B. The notion of the union of sets can be extended with
ease to an arbitrary finite number of components; it is even possible to consider the union of all
sets that are terms of a certain infinite sequence. We use the symbols
n ˆ
*
k 1
Ak and *A ,
k 1
k

respectively, as well.5
The symbol A – B denotes the difference of sets A and B: that is, the set consisting of all those
points of the set A that do not belong to B.
As an expression that sets A and B are identical—that is, that they have all points in com-
mon—I write A = B. To express that the sets A and B are disjoint—that is, that they do not
have any points in common—I will write A ][ B. Finally, to express that the set A is a proper
part of set B —that is, that every point belonging to A also belongs to B, but not conversely
—I will write A ‚ B.6
In this article I do not distinguish a point p from the set consisting solely of that same point
p. In this way, for example, the symbol
n

*p
k 1
k

denotes the set consisting of points p1 , p 2 , ... , pn .

§1

I begin by recalling the familiar definition of the congruence of two arbitrary geometric
figures, based on the notion of equal distances between two pairs of points. (This notion
should, of course, either be defined earlier or assumed as a primitive notion.)

Definition 1. Point sets A and B are congruent— A = ~ B —if between their points
a perfect (one-to-one) correspondence can be established that satisfies the following
condition: if p and q are arbitrary points of set A, while pU and qU are their corre-
sponding points in set B, then the distances between the pairs of points p and q, and
pU and qU, are equal.

In my following discussions, I will assume familiarity with elementary properties of the

congruence relation.7

5
[For union Tarski used the Polish equivalent of the English term sum, and for F and A he used + and
6, respectively.]
6
[In the original, Tarski used the symbol G for is a proper part of. He employed it only once.]
7
[The correspondence between p, q, ... and pU, qU, ... need not be a direct isometry; it may reverse orienta-
tion. Moreover, Tarski did not require that it be a restriction of an isometry of the entire plane.]
5.1 §1 81

Definition 2. Point sets A and B are equivalent— A { B —if there exist sets
A1 , A 2 , ... , An and B1 , B 2 , ... , Bn , where n [is] a natural number, that satisfy the
conditions
n n
(a) A= *A
k 1
k and B = *B ,
k 1
k

(b) ~ B whenever 1  k  n,
Ak = k

(c) A k ][ A l and Bk ][ B l whenever 1  k < l  n.

To express that the figures A and B are not equivalent I shall write A { / B. In the
following five theorems, I shall present several elementary properties of the relation of
equivalence.

Theorem 1. If A = ~ B, then A { B. In particular, an arbitrary point set A satisfies

the condition A { A.

Theorem 2. If A { B, then B { A.

Both of those theorems follow directly from definition 2.

Theorem 3. If A { B and B { C, then A { C.

Proof. For the proof I shall apply a method similar to the one that we use in the proof
of an analogous theorem in elementary geometry, the so-called “method of double
networks.”
In view of the equivalence of sets A and B as well as of B and C, there exist point
sets A1 , A 2 , ... , A n and B1 , B 2 , ... , Bn , as well as B1U, B 2U, ... , BmU and C1 , C 2 , ... , Cm that
satisfy all the conditions of definition 2. Let us denote by Bk, l the set of all those points
that belong simultaneously 8 to Bk and BlU. Since every point in set Bk belongs to one
of the sets BlU, [where] 1  l  m, and conversely, it is thus easy to check that
m
(1) Bk = *B
l 1
k, l when 1  k  n,

n
(2) BlU = *B
k 1
k, l when 1  l  m.

In addition, according to condition (c) of definition 2, we have

(3) Bk, l ][ Bk1 , l1 whenever k =
/ k1 , or k = k1 but l =
/ l1 .

In accordance with condition (b) of definition 2, figures A k and Bk are congruent

[when] 1  k  n. From (1), (3), and the general properties of congruence, we thus infer

8
Of course, the possibility is not excluded that some of the sets Bk, l may be empty: that is, that they should
not contain any points. A small modification to the proof would permit removing such sets from our
consideration.
82 5 On the Equivalence of Polygons

with ease the possibility of dividing each of the sets Ak into parts A k, 1 , A k, 2 , ... , A k, n that
satisfy the conditions
m
(4) Ak = *A
l 1
k, l when 1  k  n,

(5) ~ B when 1  k  n and 1  l  m,

A k, l = k, l

(6) Ak, l ][ Ak1 , l1 whenever k =

/ k1 , or k = k1 but l =
/ l1 .
Similarly, from the congruence of figures BlU and Cl [when] 1  l  m, [and] from (2)
and (3), follows the possibility of analogous division of each of the sets Cl into parts
C1, l , C2, l , ... , Cn, l :
n
(7) Cl = *C
k 1
k, l when 1  l  m,

(8) ~ B when 1  k  n and 1  l  m,

C k, l = k, l

(9) Ck, l ][ Ck1 , l1 whenever k =

/ k1 , or k = k1 but l =
/ l1 .
Since according to condition (a) of definition 2 the set A is the union of sets
A1 , A 2 , ... , A n , and set C [is] the union of sets C1 , C 2 , ... , Cm, we may thus conclude from
(4) and (7),
n m
(10) A = ** A
k 1l 1
k, l ,

m n n m
(11) C = **
l 1k 1
Ck , l = ** C
k 1l 1
k, l .

Moreover, from (5) and (8) we immediately obtain

(12) A = ~ C when 1  k  n and 1  l  m.
k, l k, l

Equations (10) and (11) show that each of the sets A and C can be divided into n#m
parts, which in view of (6) and (9) have no points in common, and in accordance with (12)
are respectively congruent to each other. Therefore, according to definition 2, A { C,
Q.E.D.

Theorems 1–3 express that the relation of equivalence is reflexive, symmetric, and
transitive.

Theorem 4. If
(1) ~ B or A = B ) whenever 1  k  n,
A k { Bk (possibly A k = k k k

(2) A k ][ A l and Bk ][ B l whenever 1  k < l  n,

n n
then * Ak {
k 1
*B .
k 1
k

Proof. [To facilitate typesetting, those two unions will be denoted by A and B,
respectively.] Taking theorem 1 into account, we can restrict ourselves in the proof to
5.2 §2 83

considering the hypothesis that A k { Bk for each value of k [such that] 1  k  n. Now
according to definition 2, for each pair of sets A k and Bk there exist sets
Ak,1 , Ak,2 , ... , Ak, mk and Bk,1 , Bk,2 , ... , Bk, mk that satisfy the conditions
mk mk
(1) Ak = *l 1
Ak, l and Bk = *B
l 1
k, l when 1  k  n,

(2) ~ B when 1  k  n and 1  l  m ,

A k, l = k, l k

(3) Ak, l ][ Ak1 , l1 and Bk, l ][ Bk1 , l1 whenever k =

/ k1 , or k = k1 but l =
/ l1 .

From (1) we immediately obtain

n n mk n n mk
(4) A= *
k 1
Ak = **
k 1l 1
Ak, l and B = *
k 1
Bk = ** B
k 1l 1
k, l .

From statements 2–4 it follows that the point sets A and B can be divided into the same
finite number of respectively congruent parts without common points. From this, in
accordance with definition 2, A { B, Q.E.D.

The preceding theorem can be expressed in words in the following way:

If two given point sets can be divided into the same finite number of respectively
equivalent parts having no common points, then these sets are equivalent.

Theorem 5. If sets A and B consist of the same finite number of points, then A { B.
For the proof, it suffices to note that according to definition 1, two arbitrary points are
congruent figures, hence definition 2 can be applied directly.

§2

I turn now to the proof of theorem 6, which can be regarded as a generalization of De

Zolt’s axiom. The proof will rely on the theorem of Banach mentioned already in the
introduction, which for our purposes can be adequately formulated in the following way.

Banach’s Theorem. Each point set A that is part of any polygon can be assigned some
nonnegative real number m( A), called the measure of that set. Moreover, the following
conditions are satisfied:
(1) if A = ~ B then m( A) = m(B),
(2) if A ][ B then m( A F B) = m( A) + m(B),
(3) if W is a polygon then m(W ) is its area.9

9
To the word area one ought to append throughout the words, in relation to some square, chosen as the unit
of area. [Banach’s theorem and its proof are discussed in section 4.2.]
84 5 On the Equivalence of Polygons

Theorem 6. If V and W are polygons and V ‚ W, then V {

/ W.
Proof. Suppose, contrary to the conclusion of the theorem, that
(1) V { W;
then there would exist point sets A1 , A 2 , ... , An and B1 , B 2 , ... , Bn that satisfy all the
conditions of definition 2 [with point sets V, W in place of A, B].
Further, in accordance with Banach’s theorem, let us assign to every bounded planar
set—that is, part of any polygon—its measure. Condition 2 of the theorem mentioned
can be extended with ease, by applying the principle of mathematical induction, to a sum
of an arbitrary finite number of sets having no common points. In view of conditions (a)
and (c) of definition 2, we infer from this that
n n
(2) m(V ) = ¦
k 1
m( Ak ), m(W ) = ¦ m( B ) .
k 1
k

From condition 1 of Banach’s theorem and condition (b) of definition 2 we obtain

(3) m( A k ) = m(Bk) whenever 1  k  n.
Equations (2) and (3) entail immediately
(4) m(V ) = m(W ).
Therefore, according to condition 3 of the cited theorem, polygons V and W must have
equal areas, which contradicts the assumption of our theorem, since V is a proper part
of W.
Assumption (1) thus leads to a contradiction, and we must accept that V {
/ W, Q.E.D.

Reasoning in an analogous way, [we can] prove the more general

Theorem 7. If V and W are polygons with different areas, then V {

/ W.

§3

We now take up the proof of the theorem converse to the one just presented. First of all,
we note that despite what might at first glance be supposed, this theorem does not follow
directly from an analogous theorem of elementary geometry. I will illustrate this circum-
stance with a straightforward example.

Let V be an arbitrary square and W, an isosceles right triangle with base twice as
long as the edge of the square. Having equal areas, V and W are thus equivalent in the
sense of elementary geometry. In fact, each of these polygons can be divided into two
5.3 §3 85

triangles without common interior points,

respectively congruent. (See the figure.)
From this subdivision, however, the sub-
division that would satisfy definition 2 can-
not be obtained via a direct route. Although
the interiors of the [smaller] triangles are in
fact congruent, nevertheless the [parts] that stand out in the figure are broken lines, [and]
since these unions of the boundaries of the triangles have different length, it is not hard
to demonstrate that they are not equivalent in the sense that we established in the pres-
ent article.10

The proof of the theorem that interests us will rely on several lemmas.

Lemma I. If A is a plane set having interior points,11 whereas set B consists of a finite
number of points, and A ][ B, then A { A F B.
Proof. Certainly there exists some disk K that is part of the set A; let us denote its
center by s. Let us choose some positive irrational number D and some point p 0 lying
on the circumference of the disk K. For each natural number k let us denote by pk the
point resulting from the rotation of point p 0 about point s through an angle whose
degree measure is the number k # D (or a number differing from k # D by a multiple of
360) —and we always carry out the rotation in some specified direction. From this, since
the angle of D degrees is incommensurate with a full angle, we infer with ease that no
two points pk and pl with different indices are identical.
Let n be the number of points in the set B. Set
n 1
(1) BU = *p
k 0
k,

ˆ
(2) C = *p
k 0
k ,

ˆ
(3) CU = *p
k n
k ,

(4) D = A – C.
From (1) to (4) and the definition of the points pk we immediately obtain
(5) A = C F D = BU F CU F D,
(6) A F B = B F C F D.

10
[If the correspondence of the interiors of the triangles were extended by somehow subdividing the
segments shown on the left and rearranging the parts to form those on the right, the total length of the
left-hand segments would equal that on the right. But they differ: 4 + 2 =
/ 3 + 2 2.]
11
We call point p an interior point of a plane set A if there exists a disk with center p that is a part of the
set A.
86 5 On the Equivalence of Polygons

According to theorem 5, since each of the sets B and BU consists of n points, this
statement follows:
(7) B { BU.
With ease, we also convince ourselves that
(8) C =~ CU.

In fact, if the set C is rotated about an angle of n # a degrees, then it covers the set CU.
In other words, if we assign to an arbitrary point pk of the set C the point pk+ n of the
set CU, then we define a perfect correspondence between the points of these sets, that
satisfies the conditions of definition 1.
Statements 5 to 8 show that the sets A and A F B can be divided into the same finite
number of parts, respectively equivalent, or even congruent or identical. It is also easy
to check, [by] relying on statements 1 to 4, by the way of specifying the points pk , and
[by] the hypothesis of the theorem, that no two of the three parts into which we divide
each of these sets have common points. From this, in accordance with theorem 4, we infer
that A { A F B, Q.E.D.

Lemma II. If A is a plane set having interior points, while the set B consists of all
points of some segment except at most the end points, and A ][ B, then A { A F B.
Proof. The ideas behind the proofs of lemmas I and II are similar to each other. Let
us denote by G the length of the segment from which B differs by at most the absence
of the end points. Certainly there exists a natural number n large enough that some
disk K that has a radius of length equal to G/n is part of the set A.
Clearly, the set B can be divided into n segments of length G/n without common
interior points, and two of these might have only one endpoint each. Let us denote the
interiors of these segments by C0 , C1 , ... , Cn –1 and set
n 1
(1) C= *C
k 0
k,

(2) D = B – C.
It is easy to see that D is a set consisting of a finite number of points.12
Let us choose some positive irrational number D [and] denote by C0U the interior of
some radial segment of the disk K; when k is an arbitrary natural number, [denote]
by CkU the set formed by rotating the set C0U through an angle of k # D degrees about the
center of the disk K in a certain specified direction. As in the proof of lemma I, we
convince ourselves that no two of the sets CkU and ClU with different indices have
common points. Set

12
[This number is] equal to n + 1, n, or n – 1, depending on whether the set B has both endpoints, or
just one, or, lastly, does not have any.
5.3 §3 87

n 1
(3) CU = *C
k 0
k,

ˆ
(4) E = *C
k 0
k,

ˆ
(5) EU = *C
k n
k ,

(6) F = A – E.
From (1) to (6), we immediately obtain
B = C F D, E = CU F EU, A= E F F = CU F EU F F,
from which [follow]
(7) A F D = CU F D F EU F F,
(8) A F B = C F D F E F F.
As the interiors of segments of the same length G/n, the sets Ck and CkU are congruent.
Therefore, from (1) and (3) we infer
(9) C { CU.
Furthermore, reasoning as in the proof of lemma I, we reach the conclusion that
~ EU.
(10) E =
Finally, as it is not difficult to be convinced, no two of the sets CU, D, EU, F nor of
C, D, E, F have common points. In view of this, we can apply theorem 4; by virtue of
statements 7 to 10 we have
(11) A F D { A F B.
On the other hand, the set D, as we already noticed, consists of a finite number of points.
Therefore, according to lemma I,
(12) A { A F D.
From (11) and (12) it follows, in accordance with theorem 3, that A { A F B, Q.E.D.

Lemma III. If A is a plane set having interior points, while B [is] the union of a finite
number of segments, and A ][ B, then A { A F B.
Proof. It is nearly obvious that the set B can be regarded as a union of a finite number
of segments B1 , B2 , ... , Bn without common interior points. Let us set B1U = B1 , and
when 2  k  n denote by BkU the set differing from the segment Bk in at most the
absence of one or two endpoints and that of the points belonging to any of the segments
B1, B2 , ... , Bk–1 [—that is,]
88 5 On the Equivalence of Polygons

k 1
Bk Bk  *B .
l 1
l
13

We obtain
n
(1) B= *B ,
k 1
k

(2) BkU ][ BlU whenever 1  k < l  n.

On the other hand, there certainly exist sets A1 , A 2 , ... , A n having interior points and
also satisfying the conditions
n
(3) A= *A ,
k 1
k

(4) A k ][ A l whenever 1  k < l  n.

In fact, if we divide into n circular sectors some disk K that is part of the set A, denote
by A1 , A 2 , ... , An–1 the interiors of all but one of these sectors, and set
n 1
An = A – *A ,
k 1
k

then we will at that time obtain sets with the desired properties.
In view of (1), (3), and the condition A ][ B given in the hypothesis of the theorem,
we have
(5) A k ][ Bl whenever 1  k  n and 1  l  n.
Thus, we can assert with ease that every pair of sets A k and Bk , where 1  k  n,
satisfies the conditions of lemma II. Therefore,
(6) Ak { Ak F BkU when 1  k  n.
From statements (2), (4), and (5) we conclude further that
(7) Ak F BkU ][ Al F BlU whenever 1  k < l  n.
Moreover, from (1) and (3) also follows
n n
(8) A= *A ,
k 1
k AFB= *
k 1
Ak « Bk .

In accordance with (6) to (8) the sets A1 , A 2 , ... , A n and A1 F B1U, A 2 F B2U, ... , A n F BnU,
[whose unions14 are] sets A and A F B, satisfy all the conditions of theorem 4. Thus,
we finally obtain A { A F B, Q.E.D.

Lemma III now enables us [to give] a direct proof of the theorem converse to theorem 7.

13
[This sentence and the next might not fully explain the first sentence of the proof. One can apply mathe-
matical induction as follows: if B is the union of a finite number of segments without common interior
points, and C is a segment, then C – B is also such a union, and B F C = B F (C – B).]
14
[In the original, the phrase here in brackets was vague: w stosunku do.]
5.3 §3 89

Theorem 8. If V and W are polygons with the equal areas, then V { W.

Proof. As is well known, polygons V and W are equivalent in the sense of elementary
geometry. Thus, they can be divided into the same number of polygons having no
common interior points. Let V1 , V 2 , ... , Vn and W1 , W 2 , ... , Wn be the interiors of the
polygons obtained as a result of such a division. Certainly we have
(1) V = ~ W when 1  k  n,
k k

(2) Vk ][ Vl , and also Wk ][ Wl , whenever 1  k < l  n.

According to definition 2 we infer from (1) and (2) that
n n
(3) *
k 1
Vk { *W .
k 1
k

Set
n n
(4) A=V– *
k 1
Vk , B=W– *W .
k 1
k

From this we immediately obtain

n n
(5) A ][ *
k 1
Vk , B ][ *W ,
k 1
k

and
n n
(6) V=AF *V ,
k 1
k W=BF *W .
k 1
k

In view of (4) it is easy to see that A and B are broken lines, the unions of the
boundaries of the polygons that we obtained by the subdivision of V and W; each is thus
the union of a finite number of segments. Moreover, since the sets
n n

*
k 1
Vk and *W
k 1
k

certainly have interior points, after applying lemma III [and] in accordance with (5) and
(6), we thus obtain
n n
(7) *
k 1
Vk { A F *V
k 1
k = V,

n n
(8) *
k 1
Wk { B F *W
k 1
k = W.

From statements (3), (7), and (8), according to theorem 3, it follows that V { W, Q.E.D.

Theorems 7 and 8 immediately entail

Conclusion 9. In order for polygons V and W to be equivalent, it is necessary and

sufficient that they have equal areas.

Theorem 6 and conclusion 9 settle the question posed at the beginning of the present
article.
90 5 On the Equivalence of Polygons

The question arises here whether the statements are true [that are] analogous to the
theorems proved in this article but relate to polyhedra instead of polygons. As it happens,
such statements are false. Specifically, the following theorem can be proved:
Two arbitrary polyhedra are equivalent.
The proof of this statement is complicated enough not to include it here: it is contained
in a joint article by Banach and me, entitled On Decomposition of Point Sets into Respec-
tively Congruent Parts.15

We are easily made aware of how greatly the above theorem contradicts our intuitions,
if we consider so much as the following conclusion that flows from it:
An arbitrary cube can be divided into a finite number of parts without common
points, which then can be rearranged to form a cube with an edge twice as long.
The theorem becomes even more striking when we recall that, as Max Dehn showed,16
even two polyhedra with equal volumes may not be equivalent in the sense of elementary
geometry.

In conclusion, I pose here the following problem, which as far as is known, is to this
day not settled:
Can theorem 8 be extended to arbitrary plane regions bounded by closed curves?
Specifically, can a disk and a polygon with equal areas be equivalent in the
sense of definition 2? 17

Summary18
On the Equivalence of Polygons

In elementary geometry two polygons (or polyhedra) are called equivalent by decomposition if they
can be decomposed into the same finite number of respectively congruent polygons (or polyhedra)
that have no common interior points. In the theory of equivalence of polygons the following
theorem, sometimes called De Zolt’s axiom, plays an important role:
1. Two arbitrary polygons, one of which is [properly] contained in the other, are never
equivalent by decomposition.

15
[Banach and Tarski [1924] 2014, translated in chapter 6, with background and summary in section 4.4.]
16
Compare Amaldi [1900] 1914, §11, 161–172. [Tarski failed to mention the author, Ugo Amaldi. See also
Dehn 1901–1902.]
17
A disk and a polygon with equal areas are not equivalent in the sense of elementary geometry: compare
Amaldi [1900] 1914, §§6–7, 151–157. [See the previous footnote. This problem, known as Tarski’s circle-
squaring problem, was published separately as Tarski 1925b. It has since been solved, affirmatively: see
the discussion in section 4.3 of the present book. The wrong text was printed for Tarski 1925b in the
Collected Papers volume Tarski 1986a; the original text is reproduced and translated in section 4.3.]
18
[In the original, the summary was in French.]
5.4 Summary 91

Starting from this principle, one establishes the following theorem, which presents a necessary
and sufficient condition for the equivalence of polygons:
2. In order that two polygons should be equivalent by decomposition, it is necessary
and sufficient that they should have equal areas.

In my note I envisage the notion of equivalence in a sense more general than that of elementary
geometry: two point sets (thus, in particular, two polygons or polyhedra) are termed equivalent
by decomposition if they can be decomposed into the same finite number of respectively congruent
arbitrary point sets that have no common points.
I prove that, even admitting this definition of equivalence, theorems 1 and 2 remain valid.
In demonstrating the cited theorems I rely on results obtained by Banach in measure theory
(Banach 1923). Establishing theorem 2, I also make use of the following lemma:
P being the interior of a polygon and Q the point set obtained from P by adding a
finite number of segments, the sets P and Q are equivalent by decomposition.
It is interesting to remark that in attributing to equivalence the sense established in this note,
theorems 1 and 2 may not be extended to polyhedra. This results from the following theorem,
which perhaps seems paradoxical:
Two arbitrary polyhedra (with equal volumes or not) are equivalent by decomposition.
This theorem is demonstrated in the note Banach and Tarski [1924] 2014.19

19
[Summarized in section 4.4 of the present book and translated in full in chapter 6.]
 6
On Decomposition of
Point Sets into Respectively
Congruent Parts (1924)

This chapter contains an English translation of the paper Sur la décomposition des
ensembles de points en parties respectivement congruents, [1924] 2014, by Stefan Banach
and Alfred Tarski. It appeared in volume 6 of the journal Fundamenta Mathematicae.
This is its first translation. Its best-known result is often called the Banach–Tarski
paradox: any two balls with different radii can be decomposed into the same finite
number of disjoint, respectively congruent parts. Background for the paper and a sum-
mary are provided in sections 4.1–4.4. Section 8.6 discusses its relationship with related
later research by Tarski and others.

This translation adheres closely to the terse French of the original. A few set-theoretic
symbols have been replaced by modern equivalents.1 Some abbreviations have been
spelled out, and some uses of alternative type styles for emphasis, enunciations, and per-
sonal names have been modified. Punctuation, paragraph breaks, and page layout have
been streamlined or redesigned to conform to the style of the present book. Bibliographic
references and some personal names have been adjusted to conform with the conventions
used here. All [square] brackets in the translation enclose editorial comments. Those
are inserted, usually as footnotes, to note further editorial changes and correction of
errors in the original, and occasionally to provide additional information.

1
The original symbols G, H, +, 6  ×, 0, ( ), and U for subset, superset, binary and general union,
intersection, empty and singleton sets, and interior have been replaced by I, J, F, A , , I, { }, and (,
respectively.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 93
DOI 10.1007/978-1-4939-1474-6_6, © Springer Science+Business Media New York 2014
 TREŚĆ TOMU VI.
Str.
W. Sierpiski: Sur une propriété des ensembles ambigus . . . . . . . . . . . . . . . . . . . . 1
S. Kempisty: Sur les fonctions approximativement discontinues . . . . . . . . . . . . . . . 6
A. Khintchine: Über einen Satz der Wahrscheinlichkeitsrechnung . . . . . . . . . . . . . 9

S. Mazurkiewicz: Remarque sur un théorème de M. Mullikin . . . . . . . . . . . . . . . 37
W. Sierpiski: Un exemple effectif d’un ensemble mesurable (B) de classe D . . . 39
A. Tarski: Sur les ensembles finis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
H. Lebesgue: Sur le théorème de Schoenflies . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96

R. L. Moore: Concerning the common boundary of two domains . . . . . . . . . . . . 203
R. L. Wilder: On the dispersion sets of connected point-sets . . . . . . . . . . . . . . . . 214
P. Urysohn: Über ein Problem von Herrn C. Carathéodory . . . . . . . . . . . . . . . . . 229
S. Banach: Un théorème sur les transformations biunivoques . . . . . . . . . . . . . . . 236
C. Kuratowski: Une propriété des correspondances biunivoques . . . . . . . . . . . . 240
S. Banach et A. Tarski: Sur la décomposition des ensembles de points
en parties respectivement congruentes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244


Fundamenta Mathematicae 6 (1924)

Portions of Cover and Table of Contents
 On Decomposition of Point Sets into
Respectively Congruent Parts
By

Stefan Banach (Lwów) and Alfred Tarski (Warsaw)

In this note we study the notions of equivalence of point sets by finite decomposition and
by denumerable decomposition. Two point sets situated in a metric space are called equiv-
alent by finite or denumerable decomposition just in case they can be decomposed into
one and the same finite number, or into a denumerable infinity, of disjoint, respectively
congruent parts.

The principal results contained in the present article are the following:
In a Euclidean space of at least three dimensions two arbitrary sets, bounded
and containing some interior points ( for example, two spheres with different
radii), are equivalent by finite decomposition.
An analogous theorem holds for sets situated on the surface of a sphere; but the
corresponding statement concerning Euclidean spaces of one or two dimensions
is false.
On the other hand,
In a Euclidean space of any dimension two arbitrary sets (bounded or not),
containing interior points, are equivalent by denumerable decomposition.

The demonstration of the preceding theorems rests on results of Felix Hausdorff,

Giuseppe Vitali, and Stefan Banach2 about the general problem of measure. Thus, they
make use of Ernst Zermelo’s axiom of choice. The role that this axiom plays in our
reasoning seems to us to merit attention.

We have in mind, indeed, the two following theorems that result from our research.
I. Two arbitrary polyhedra are equivalent by finite decomposition.
II. Two different polygons, one of which is contained in the other, are never
equivalent by finite decomposition.3
Now, we do not know how to demonstrate either of these two theorems without appealing
to the axiom of choice: neither the first, which perhaps seems paradoxical, nor the second,
which is in full accord with intuition. Moreover, analyzing their demonstrations, one can
see that the axiom of choice intervenes in the demonstration of the first theorem in a form
much more restricted than it does in the second.

2
Hausdorff 1914a, 401, 469; Vitali 1905; Banach 1923, 30–31.
3
This theorem can be regarded as a generalization of the theorem, familiar in elementary geometry, some-
times called De Zolt’s axiom. See Tarski [1924] 2014b [translated in chapter 5].
96 6 Decomposition of Point Sets into Respectively Congruent Parts

§1
General Properties of Equivalence by
Finite or Denumerable Decomposition

The arguments in this section are valid for point sets situated in an arbitrary space, on
which is imposed the single hypothesis that to each pair (a, b) of points there should
correspond a real number U(a, b) called the distance between points a and b. Only
corollaries 14–15U concern Euclidean space.

Definition 1. Point sets A and B are congruent,

A=~ B,

if there exists a function M that transforms A to B bijectively and satisfies this

condition: for two arbitrary points a1 and a2 of the set A,
U(a1 , a2 ) = U ( M(a1 ), M(a2 )) .

In what follows we suppose that the elementary properties of the notion of congruence
are known.

Definition 2. Point sets A and B are equivalent by finite decomposition,

A = B, f

if there exist sets A1, A 2 , ... , An and B1, B2 , ... , Bn that fulfill the following conditions:
n n
I. A = *k 1
Ak and B = *B ,
k 1
k

II. A k  A l = I = Bk  B l when 1  k < l  n,

III. A k ~
= Bk whenever 1  k  n.

Definition 2U. Point sets A and B are equivalent by denumerable decomposition,

A = B, d

if there exist sets A1, A2 , ... , A n , ... and B1, B2 , ... , Bn , ... that fulfill the following
conditions:
ˆ ˆ
I. A = *k 1
Ak and B = *B ,
k 1
k

II. A k  A l = I = Bk  B l when k =
/ l,
III. A k ~
= Bk for all natural numbers k.

In the theorems 1–15U that follow we establish the elementary properties of the notions
just introduced, without limiting ourselves just to those that are useful to us later. To
each theorem concerning equivalence by finite decomposition corresponds a theorem on
6.1 §1 General Properties of Equivalence 97

equivalence by denumerable decomposition; the demonstrations of the corresponding

theorems being completely analogous, we limit ourselves to giving only one of them here.

~ B, then A = B.
Theorem 1. If A = B, or even A = f

~ I, one
This is an immediate consequence of definition 2. Taking into account that I =
deduces further,

~ B, or even A = B, then A = B.
Theorem 1U. If A = B, A = f d

Theorem 2. If A = =
f B, then B f A.

Theorem 2U. If A = =
d B, then B d A.

Those theorems result immediately from definitions 2 and 2U.

Theorem 3. If A = = =
f B and B f C, then A f C.

Proof. Let n n
A= *A
k 1
k and B= *B
k 1
k (1)

m m
B= *B
l 1
l and C= *C l (2)
l 1
be the decompositions of sets A and B, and of B and C, respectively, that satisfy
conditions I–III of definition 2. We set
Bk , l = Bk  BlU when 1  k  n and 1  l  m; (3)
from (1)–(3) it follows immediately that
m
Bk = *B
l 1
k, l whenever 1  k  n, (4)

n
BlU = *B
k 1
k, l whenever 1  l  m. (5)

Moreover, following (3) and condition II of the cited definition one obtains
Bk, l  Bk1 , l1 = I when k =
/ k1 or l =
/ l1 . (6)
Sets A k and Bk being congruent when 1  k  n, one deduces according to (4) and
(6) the existence of sets A k , 1 , A k , 2 , ... , A k , m that satisfy formulas
m
Ak = *A
l 1
k, l whenever 1  k  n, (7)

Ak, l  Ak1 , l 1 = I when k=

/ k1 or l =
/ l1 , (8)
~B
Ak, l = k , l whenever 1  k  n and 1  l  m. (9)
98 6 Decomposition of Point Sets into Respectively Congruent Parts

In the same way, when 1  l  m, congruence of the sets BlU and Cl implies by virtue
of (5) and (6) that there exist sets C1 , l , C2 , l , ... , Cn , l such that
n
Ck = *C
k 1
k, l whenever 1  l  m, (10)
Ck, l  Ck1 , l1 = I when k =
/ k1 or l =
/ l1 , (11)
~C
Bk , l = k , l whenever 1  k  n and 1  l  m. (12)
By virtue of (1), (2), (7), and (10), one concludes that
n m
A= ** A
k 1l 1
k, l , (13)
m n n m
C= **
l 1k 1
Ck , l = ** C
k 1l 1
k, l ; (14)

from (9) and (12) one finally obtains

~C
Ak, l = k , l whenever 1  k  n and 1  l  m. (15)

Formulas (13) and (14) provide a decomposition of sets A and C into a finite number
(equal to n $ m) of parts; following (8), (11), and (15), this decomposition fulfills the
conditions of definition 2. Thus one has
A=
f C, Q. E. D.

In a completely analogous way one can demonstrate the following

Theorem 3U. If A = = =
d B and B d C, then A d C.

In conformity with theorems 1–3U the relations of equivalence by finite and by

denumerable decomposition are reflexive, symmetric and transitive.

Theorem 4. If sets A and B can be decomposed into disjoint subsets,

n n
A= *
k 1
Ak , B= *B
k 1
k ,
in such a way that
Ak =
f Bk whenever 1  k  n,

then
A=
f B.

Proof. The hypothesis of the theorem implies that, for each k such that 1  k  n,
there exists a decomposition of the sets A k and Bk ,
mk mk
Ak = *
l 1
Ak, l , Bk = *B
l 1
k, l , (1)

that satisfies the conditions

Ak, l  Ak1 , l1 = I = Bk, l  Bk1 , l1 when k=
/ k1 or l =
/ l1 , (2)
6.1 §1 General Properties of Equivalence 99

~B
Ak, l = whenever 1  k  n and 1  l  m. (3)
k, l

From (1) one obtains

n mk n mk
A= **
k 1l 1
Ak, l , B= ** B
k 1l 1
k, l . (4)

In conformity with definition 2, formulas (2)–(4) immediately yield

A=
f B, Q. E. D.

Theorem 4U. If sets A and B can be decomposed into disjoint subsets,

ˆ ˆ
§ n n
·
A= *A , k B= *B k ¨ respectively, A *A , k B *B k ¸,
k 1 k 1 © k 1 k 1 ¹
in such a way that
Ak =
d Bk for each natural number k ( respectively, when 1  k  n ) ,

then
A=
d B.

Theorem 5. If A = f B, there exists a function M, defined for all points of the set A,
that fulfills these conditions:
I. the function M transforms A into B bijectively,
II. for every subset C of A, one has C = M(C) .4 f

Proof. Let
n n
A= *A
k 1
k , B= *B
k 1
k (1)

be a decomposition of the sets A and B that satisfies the conditions of definition 2. The
sets A k and Bk being congruent for 1  k  n, definition 1 yields the existence of
functions Mk that map A k to Bk without changing the distances between the points
being transformed. Set
M( p) = Mk( p) in case p  A k , where 1  k  n. (2)
With no trouble, by virtue of the properties of the decomposition (1), one concludes from
this that
the function M maps A into B bijectively. (3)
Now let
CIA (4)
and set

4
For every function M defined for all points of a set A, if C I A, then M(C) denotes the set of all elements
M( p) for which p  C.
100 6 Decomposition of Point Sets into Respectively Congruent Parts

Ck = C  A k whenever 1  k  n. (5)
From (1), (4), and (5) one obtains immediately,
n
C= *C
k 1
k , (6)

Ck  Cl = I when 1  k < l  n, (7)

Ck I A k whenever 1  k  n. (8)
From (3), (4), (6), and (7), it follows that
n
M(C) = * M(C ) ,
k 1
k (9)

M(Ck )  M(Cl ) = 0 when 1  k < l  n. (10)

Finally, it results from (2) and (8), in conformity with the indicated property of the
functions Mk , that
~ M(C ) whenever 1  k  n.
Ck = (11)
k

Formulas (6) and (9) provide a decomposition of the sets C and M(C) that fulfills,
following (7), (10), and (11), all the conditions of definition 2. Thus one has
C=
f M(C). (12)
Conditions (7) and (12) prove that M is the desired function.

Theorem 5U. If A = d B, there exists a function M defined for all points of the set A and
5

fulfilling these conditions:

I. the function M maps A to B bijectively,
II. if C should be an arbitrary subset of A, then C =
d M(C).

Theorems 5 and 5U directly imply the following corollaries.

Corollary 6. If A =
f B and there exists a decomposition of the set A into disjoint
subsets, ˆ
n
§ ·
A= *A k ¨ respectively, A *A k ¸,
k 1 © k 1 ¹
then there also exists a decomposition of the set B into disjoint subsets,
ˆ
n
§ ·
B= *B k ¨ respectively, B *B k ¸,
k 1 © k 1 ¹
such that

Ak =
f Bk whenever 1  k  n ( respectively, for every natural number k ).

5
[Banach and Tarski stated this hypothesis as A =
f B, evidently in error.]
6.1 §1 General Properties of Equivalence 101

Corollary 6U. If A =
d B and there exists a decomposition of the set A into disjoint
subsets,
ˆ
n
§ ·
A= * Ak ¨ respectively, A *A k ¸,
k 1 © k 1 ¹
then there also exists a decomposition of the set B into disjoint subsets,
ˆ
n
§ ·
B= *B k ¨ respectively, B *B k ¸,
k 1 © k 1 ¹
such that

Ak =
d Bk whenever 1  k  n ( respectively, for every natural number k ).

Corollary 7. If A = f B, then to each subset C of A corresponds a subset D of B

subject to conditions
I. C=
f D,

II. if C =
/ A, then D =
/ B.

Corollary 7U. If A = d B, then to each subset C of A corresponds a subset D of B

subject to conditions
I. C=
d D,

II. if C =
/ A, then D =
/ B.

Theorems 8 and 8U, which we shall establish now, will play an important role in the
arguments of the following subsections.

Theorem 8. If A1 I A, B1 I B, A = = =
f B1 , and B f A1 , then A f B.

Theorem 8U. If A1 I A, B1 I B, A = = =
d B1 , and B d A1 , then A d B.

Proof. In conformity with theorems 4–5U, the two relations under consideration in this
work, equivalence by finite and by denumerable decomposition, possess properties (D) and
( E) defined in Banach 1924, 236. Therefore theorems 8 and 8U are just immediate conse-
quences of the theorem 3 established in that note, which concerns all relations possessing
those two properties.

Corollary 9. If A J B J C and A = = =
f C, then A f B and B f C.

Corollary 9U. If A J B J C and A = = =

d C, then A d B and B d C.

Those corollaries follow directly from the preceding theorems if one replaces there A1
by B as well as B1 by C and applies in turn theorems 1 and 3, or 1U and 3U, respectively.
102 6 Decomposition of Point Sets into Respectively Congruent Parts

Theorem 10. If A =
f A F Bk whenever 1  k  n, then
n
A=
f A F *B .
k 1
k

Proof. We consider two cases.

(a) [The case in which] the sets A, B1 , B2 , ... , Bn are disjoint. We shall proceed by
induction. The theorem being evident for n = 1, we suppose it is true for n = nU and
prove that it holds as well for n = nU + 1. Thus one has
n
A=
f A F *B ,
k 1
k (1)

A=
f A F BnU+ 1 , (2)
n
A  BnU+ 1 = I = BnU+ 1  *B .
k 1
k (3)

In conformity with theorem 4, one obtains from (1) and (3),

n 1
A F BnU+ 1 =
f A F *B .
k 1
k (4)

From (2) and (4) it follows immediately, by virtue of theorem 3, that

n 1
A=
f A F *B ,
k 1
k Q. E. D.

(b) The general case. We set

k 1
§ ·
B1U = B1 – A, BkU = Bk – ¨ A « *B ¸ l whenever 2  k  n. (5)
© l 1 ¹
From this one concludes without pain,
n n
AF *
k 1
Bk = A F *B ,
k 1
k (6)

A F BkU I A F Bk whenever 1  k  n. (7)

Now we apply corollary 9, replacing there A by A F Bk , B by A F BkU, and C by A.
One obtains from (7) and the hypothesis of the theorem,
A=
f A F BkU whenever 1  k  n. (8)
By virtue of (5), the sets A, B1U, B2U, ... , BnU are disjoint. Case (a) already established, one
deduces from (8) and (6) that
n n
A=
f *
k 1
Bk = A F *B .
k 1
k

Theorem 10 is thus completely demonstrated.

6.1 §1 General Properties of Equivalence 103

Theorem 10U. If A =
d A F Bk when 1  k  n, then
n
A=
d A F *B .
k 1
k

Theorem 10U can be extended to the case of a denumerable infinity of summands [sets
Bk ] ; but that would require a special demonstration, rather more complicated.

The theorem to which we now pass is all-important in the arguments of §2C.

Theorem 11. If A1 = = =
f A 2 , B1 f B2 , A1 F A 2 f B1 F B2 , and A1  A 2 = I = B1  B2 , then

A1 =
f B1 .

Proof. Kazimierz Kuratowski has established a general theorem concerning reflexive,

symmetric, and transitive relations that possess properties (D) and ( E).6 In case
A1 F A 2 = B1 F B2 , theorem 11 follows immediately from that. To pass to the general
case, we note that by virtue of corollary 6, the set A1 F A 2 splits into two disjoint subsets,
A1 F A 2 = B1U F B2U ,
so that
B1U =
f B1 , B2U =
f B2 . (1)
Since B1 =
f B2 , one obtains from (1), following theorem 3,

B1U =
f B2U ;

by virtue of the preceding case, one can thus conclude that

A1 =
f B1U. (2)
Formulas (1) and (2) immediately yield
A1 =
f B1 , Q. E. D.

Theorem 11U. If A1 = = =
d A 2 , B1 d B2 , A1 F A 2 d B1 F B2 , and A1  A 2 = I = B1  B2 , then

A1 =
d B1 .

In corollaries 12 and 12U we shall give an easy generalization of the preceding theorems.

Corollary 12. If A1 , A 2 , ... , A2 n as well as B1 , B2 , ... , B2 n should be disjoint sets, and

I. A1 = =
f A k and B1 f Bk when 1  k  2 ,
n

2n 2n
II. *
k 1
Ak =
f *B ,
k 1
k

then
A1 =
f B1 .

6
Kuratowski 1924, 243. See the demonstration for theorems 8 and 8U.
104 6 Decomposition of Point Sets into Respectively Congruent Parts

Proof. By virtue of theorem 11, the corollary is true in case n = 1. To apply the
principle of induction, we suppose that the corollary holds for n = nU and prove that it
holds also for n = nU + 1. Thus one has7
A1 = =
f A k and B1 f Bk when 1  k  2
nU+1
, (1)
n 1 n 1
2 2

*
k 1
Ak =
f *B.
k 1
k (2)

We set
2n 2 n 1 2n 2 n 1
AU = * Ak , A2 = * Ak , BU = * Bk , B2 = *B . k (3)
k 1 k 2n 1 k 1 k 2n 1

Following theorem 4, from (1)–(3) one concludes immediately,

AU =
f A2, BU =
f B2, AU F A2 =
f BU F B2. (4)
Sets AU and A2 as well as BU and B2 being disjoint, one deduces from theorem 11
according to (4),
AU =
f BU. (5)
Formulas (1), (3), and (5) prove that the sets A1 , A 2 , ..., A2n and B1 , B2 , ..., B2n satisfy the
hypothesis of the theorem. In conformity with our assumption one thus obtains
A1 =
f B1 , Q. E. D.

Corollary 12U. If A1 , A 2 , ..., A 2n as well as B1 , B2 , ..., B2n should be disjoint sets, and
I. A1 = =
d A k and B1 d Bk when 1  k  2 ,
n

2n 2n
II. *
k 1
Ak =
d *B ,
k 1
k

then
A1 =
d B1 .

Theorem 13. If A =
f B and A belongs to a class K of sets that satisfies conditions

I. when X  K and Y  K one has X F Y  K,

II. when X  K and Y I X one has Y  K,
~ X one has Y  K,
III. when X  K and Y =
then the set B also belongs to the class K.
Proof. In conformity with definition 2, let
n n
A= *
k 1
Ak , B= *B
k 1
k (1)

7
[Evidently in error, Banach and Tarski wrote n1 for nU after this in the proof. The translation uses nU.]
6.1 §1 General Properties of Equivalence 105

be a decomposition of sets A and B into disjoint, respectively congruent parts. Accord-

ing to (1), condition II of the hypothesis of the theorem implies
A k  K whenever 1  k  n.
From this, and from condition III, follows
Bk  K whenever 1  k  n. (2)
Since condition I can be extended by a simple induction to the case of a union of an
arbitrary finite number of sets, one concludes by virtue of (1) and (2) that
B  K, Q. E. D.

Theorem 13U. If A =
d B and A belongs to a class K of sets that satisfies conditions
ˆ
I. when Xn  K for every natural number n, one has *X
n 1
n  K,
II. when X  K and Y I X, one has Y  K,
~ X, one has Y  K,
III. when X  K and Y =
then the set B also belongs to the class K.

The two preceding theorems directly imply the following corollaries: A and B being
sets situated in a Euclidean space with an arbitrary number of dimensions, one has

Corollary 14. If A = 8
f B and A is nowhere dense, then B is also nowhere dense.

Corollary 14U. If A =d B and A belongs to the first category in the sense of Baire, then
B also belongs to the first category.

Corollary 15. If A =f B and A is a set measurable in the Peano–Jordan sense, with

measure zero, then B is also a set measurable in the same sense, with measure zero.

Corollary 15U. If A =d B and A is a Lebesgue-measurable set with measure zero, then

B is also a Lebesgue-measurable set with measure zero.

In an analogous way one proves that if A =

f B and A is a bounded set, then B is also
bounded.

8
A (point) set A is called a boundary set if it contains no interior point. A set A is called nowhere dense
[non-dense in the original] if the set consisting of all the points of A as well as the accumulation points
of A is a boundary set. A set A is said to belong to the first category in the sense of Baire if it is the union
of denumerably infinitely many nowhere dense sets.
106 6 Decomposition of Point Sets into Respectively Congruent Parts

§2
The Fundamental Theorems about
Equivalence by Finite Decomposition

In the arguments of this section as well as the following one we consider point sets
situated in a Euclidean space with a finite number of dimensions.

A. Euclidean Space with One or Two Dimensions

The most important theorem of this subsection is theorem 16, which establishes a condi-
tion necessary for two Lebesgue-measurable linear or planar point sets to be equivalent
by finite decomposition.

Theorem 16. If A and B should be bounded Lebesgue-measurable linear or planar sets

and A =
f B, then m( A) = m(B).
9

Proof. By a theorem of Stefan Banach,10 one can assign to each bounded set A situated
in Euclidean space with dimension one or two a nonnegative real number f ( A) in such
a way that the following conditions should be satisfied:
I. if A =~ B, then f ( A) = f (B),
II. if A  B = I, then f ( A F B) = f ( A) + f (B),
III. if A is Lebesgue-measurable, then f ( A) = m ( A).
In conformity with definition 2, the hypothesis of the theorem implies the existence of
sets A1 , A 2 , ... , An and B1 , B2 , ... , Bn that satisfy these formulas:
n n
A= *A ,
k 1
k B= *B ,
k 1
k (1)

~ B whenever 1  k  n,
Ak = (2)
k

A k  A l = I = Bk  B l whenever 1  k < l  n. (3)

By virtue of condition I and formula (2) one obtains
f ( A k) = f (Bk) whenever 1  k  n. (4)
Condition II can be extended via an easy induction to the case of an arbitrary finite
number of disjoint summands. One can thus conclude from (1) and (3),
n n
f ( A) = * f ( A ),
k 1
k f (B) = * f (B ) .
k 1
k (5)

Formulas (4) and (5) immediately yield

9
m( A) denotes the Lebesgue measure of a set A (linear measure if one is considering a one-dimensional
space, planar measure in the two-dimensional case, and so on).
10
Banach 1923, 30–31 [discussed in section 4.2].
6.2 §2 Fundamental Theorems about Finite Decomposition 107

f ( A) = f (B). (6)
Following the hypothesis of the theorem, one finally deduces from III and (6) that
m( A) = m(B), Q. E. D.

The converse of the preceding theorem is not valid. This follows directly from corollary
14: two point sets A and B, where the one is nowhere dense and the other is not, can
have the same Lebesgue measure even though they are not equivalent by finite decom-
position. Nevertheless, the converse does hold in a particularly simple case, notably in
the case of polygons. To establish this, we prove first the following

Lemma 17. Should A be a planar set that is not a boundary set,11 and B be the union
of a finite number of segments, then A =
f A F B.

Proof. We consider two cases.

Case (a): A  B = I. Let C be a disk satisfying the formula
C I A. (1)
Evidently, the set B can be decomposed into a finite number of (not necessarily disjoint)
segments, each of which has length less than that of a radius of C:
n
B= *B .
k 1
k (2)

We consider segment Bk for an arbitrary k such that 1  k  n. Let D1 be a segment

congruent to Bk and situated on a radius of C, but not containing the center of the
circle. We choose an angle D incommensurable with a right angle and, for each natural
number n, designate by Dn + 1 the segment obtained by turning segment D1 through
angle n # D about the center of the disk (in a fixed sense).
We set11.5
ˆ
E= *D ,
n 1
n (3)

ˆ
F= *D
n 2
n, (4)

G = A – E. (5)
Since E I C, one obtains immediately from (1) and (3)–(5) the following decomposition
of sets A and A F Bk :
A = G F F F D1 , A F Bk = G F E F Bk . (6)
Evidently,
~ G,
G= ~ E,
F= ~B ,
D1 = (7)
k

11
See page 105, footnote 8.
11.5
[The starting indices of the following two unions were incorrectly specified in the original paper.]
108 6 Decomposition of Point Sets into Respectively Congruent Parts

because F is obtained from E by rotation through angle D. Finally, one deduces easily
from (3)–(5) as well as from the indicated property of angle D (incommensurability with
a right angle) that formulas (6) effect a decomposition of the sets A and A F Bk into
disjoint subsets. Then, in conformity with (6), (7), and definition 2,
(8) A=
f A F Bk .

The same reasoning being valid for each segment Bk when 1  k  n, one can apply
theorem 10. According to (2), one obtains
A = A F B. f

Case (b): the general case. Apparently, the set A – B is not a boundary set. Since
( A – B)  B = I, by virtue of (a) one concludes that
A–B =
f ( A – B) F B = A F B.

This formula and the evident inclusion

AFBJAJA–B
imply, following corollary 9, that
A=
f A F B, Q. E. D.

Thanks to a remark of Adolf Lindenbaum, one can state a theorem for linear sets,
analogous to the preceding lemma, by replacing the term segment by point.

Theorem 19. If polygons A and B have the same area, one has
A=
f B.

Proof. As one knows, polygons A and B are equivalent by decomposition in the sense
of elementary geometry. That is, one can decompose them into the same number of
respectively congruent polygons without common interior points. Let A1 , A 2 , ... , A n and
B1 , B 2 , ... , Bn be the interiors of these polygonal parts.12 Evidently,
n n

*
k 1
Ak =
f *B
k 1
k . (1)

Since the sets

n n
A– *A
k 1
k and B – *B k 1
k

are composed of a finite number of segments, one concludes, by applying the preceding
lemma, that
n n
§ n
· n

*A =
k f *A k F ¨A *A k ¸ = A and similarly *B = B.
k f (2)
k 1 k 1 © k 1 ¹ k 1

Following theorem 3, one immediately obtains from (1) and (2),

A = B, f Q. E. D.

12
[Banach and Tarski used here the term polygones partiels.]
6.2 §2 Fundamental Theorems about Finite Decomposition 109

Theorems 17 and 19 directly imply

Corollary 20. For two polygons to be equivalent by finite decomposition, it is necessary

and sufficient that they should have the same area.

B. Euclidean Space of Three (and More) Dimensions

The arguments of this part concern space of three dimensions. To extend the obtained
results to a space of n > 3 dimensions, it would be necessary to consider, instead of
spheres, the sets of all points ( x1 , x 2 , ... , xn ), where these (rectangular) coordinates satisfy
the conditions
( x1 – a1 )2 + ( x 2 – a 2 )2 + ( x 3 – a 3 )2 = U 2,
b  xk  c whenever 3 < k  n,
where a1 , a 2 , a 3 , U, b, and c are constants.

To establish the principal result of this work, theorem 24, we shall first demonstrate
some lemmas.

Lemma 21. Every sphere S contains two disjoint subsets A1 and A 2 such that
S= =
f A1 and S f A 2 .

Proof. According to the famous theorem known as Hausdorff ’s paradox,13 one can
decompose the surface of the sphere S into four disjoint subsets BU, CU, DU, and EU,
where EU is a denumerable set and sets BU, CU, and DU satisfy the formulas
~ CU F DU, BU =
BU = ~ CU =~ DU.

Let p be the center of the sphere S. We designate by B, C, D, and E the unions of

all the radii of the sphere S, excluding the center p, whose endpoints belong to BU,
CU, DU, and EU, respectively. In this way one evidently obtains a decomposition of the
sphere S into five disjoint parts,
S = B F C F D F E F { p}, (1)
subject to conditions
~ C F D,
B= (2)
~C=
B= ~ D. (3)
As far as the set E is concerned, we shall use here only the following property, already
noted by Felix Hausdorff:
there is a proper subset F of B F C F D such that E = ~ F. (4)

13
Hausdorff 1914a, 469.
110 6 Decomposition of Point Sets into Respectively Congruent Parts

It is easy to convince oneself of this, by making the sphere S turn suitably about one of
its axes.

From (2) and (3), without effort, one obtains

B=
f B F C, BFC=
f B F C F D,

from which [follows], by virtue of theorem 3,

B=
f B F C F D. (5)
We set
A1 = B F E F { p}. (6)
By theorem 4, formulas (1), (5), and (6) yield
S=
f A1 . (7)
On the other hand, it results immediately from (3) and (5) that
C=f B F C F D, (8)
D = B F C F D;
f (9)
applying corollary 7, one deduces from (4) and (8) the existence of a set G that satisfies
the formulas
F = =
f G, which yields E f G ; (10)
G I C and G =
/ C. (11)
Conforming with (11), let
q  C – G, (12)
and then let us set
A 2 = D F G F { q}. (13)
Sets C and D being disjoint, one deduces from (11) and (12) that sets D, G, and
{ q} are disjoint, too. Now, the sets { p} and { q} being evidently congruent, one
concludes according to (1), (9), (10), and (13),
S=
f A2 . (14)
Finally, one easily obtains
A1 F A 2 I S and A1  A 2 = I . (15)
Formulas (7), (14), and (15) prove that A1 and A 2 are sets such as those sought.

Lemma 22. If S1 and S2 should be congruent spheres, then

S1 =
f S1 F S2 .

Proof. Conforming with the preceding lemma, let A1 and A 2 be sets subject to the
conditions
6.2 §2 Fundamental Theorems about Finite Decomposition 111

S1 =
f A1 , S1 =
f A2 , (1)
A1 F A 2 I S1 and A1  A 2 = I. (2)
By virtue of (1) and the hypothesis of this lemma,
S2 =
f A2 .

Corollary 7 then implies the existence of a set B such that

B I A2 , (3)
=
B f S2 – S1 . (4)
According to theorem 4, one concludes easily from (1)–(4) that
A1 F B =
f S1 F (S2 – S1 ) = S1 F S2 , (5)
A1 F B I S1 I S1 F S2 . (6)
Formulas (5) and (6) yield immediately, by virtue of corollary 9,
S1 =
f S1 F S2 , Q. E. D.

Lemma 23. If a bounded set A, situated in Euclidean space of three dimensions,

contains a sphere S, then
A = S. f

Proof. Since A is a bounded set, one can obviously decompose it into n (not neces-
sarily disjoint) subsets,
n
A= *B ,
k 1
k (1)

each of which, when 1  k  n, fulfills the condition

Bk is contained in a sphere Sk congruent to S. (2)
By virtue of the preceding lemma,
S=
f S F Sk whenever 1  k  n,

which, according to theorem 10, implies that

n
S=
f S F *S .
k 1
k (3)

On the other hand, one obtains from (1), (2), and the hypothesis of this lemma,
n
SIAISF *S .
k 1
k (4)

By reason of corollary 9, formulas (3) and (4) directly imply

A=
f S, Q. E. D.
112 6 Decomposition of Point Sets into Respectively Congruent Parts

The lemma just demonstrated now permits us to establish

Theorem 24. If two arbitrary sets A and B, situated in Euclidean space of three
dimensions, are bounded and are not boundary sets, then
A = B. f

Proof. Let S1 and S2 be spheres contained in A and B, respectively; evidently, one

can assume that
S1 =
f S2 . (1)
By virtue of lemma 23 one obtains
A= =
f S1 and B f S2 . (2)

According to theorems 1 and 3, one concludes immediately from (1) and (2) that
A=
f B, Q. E. D.

Thus one sees in particular that two spheres with different radii are equivalent by finite
decomposition, whereas, as we have proved earlier, two circles are equivalent only when
their radii are equal. This essential difference between spaces of two and three dimen-
sions is intimately related to the fact that the problem of measure has a positive solution
in the first case and negative in the second.

C. The Surface of a Sphere

Theorem 31, which is fundamental in this part of our investigation, shows from the view-
point of equivalence by finite decomposition that the surface of a sphere behaves in a way
fully analogous to three-dimensional space.

Lemma 25. If A and B are sets situated on the surface of the same sphere, A is not
a boundary set (with respect to this sphere) and B is composed of a finite number of arcs
of great circles, then
A = A F B. f

The proof of this is completely analogous to that of lemma 17.

Lemma 26. If spherical polygons A and B, situated on the surface of the same sphere,
have equal areas, then
A = B. f

The proof is based on the preceding lemma and does not differ from that of theorem
19; one utilizes the familiar theorem according to which two spherical polygons, situated
on the surface of the same sphere and possessing equal areas, are equivalent by finite
decomposition in the sense of elementary geometry.14

14
Gerwien 1833b. [This citation was incorrect in the original.]
6.2 §2 Fundamental Theorems about Finite Decomposition 113

Lemma 27. The surface S of any sphere can be decomposed into two disjoint subsets
A1 and A 2 such that one has S = =
f A1 and S f A 2 .

Proof. By reasoning as in the demonstration of lemma 21, one proves the existence
of sets A1U and A 2 satisfying formulas
S=
f A1U, S=
f A2 , (1)
A1U F A 2 I S and A1U  A 2 = I. (2)
We set
A1 = S – A 2 ; (3)
from (2) and (3) one obtains without pain,
S = A1 F A 2 , A1  A 2 = I , (4)
S J A1 J A1U. 15
(5)
By virtue of (1) and (5) one concludes again, by applying corollary 9, that
S= =
f A1 and S f A 2 . (6)
Formulas (4) and (6) show that A1 and A 2 are sets such as those desired.

With the aid of corollary 6, this lemma is generalized by an easy induction in the follow-
ing way:

Lemma 28. If n is an arbitrary natural number, the surface S of any sphere can be
decomposed into n disjoint subsets A1 , A 2, ... , A n such that S =
f A k whenever 1  k  n.

Lemma 29. If n is an arbitrary natural number and the surface S of a sphere is

decomposed into 2 n congruent spherical polygons B1 , B2 , ... , B 2 n without common inter-
ior points, then S =
f B1 .

Proof. We set
k1
B1U = B1 , BkU = Bk – *B
l 1
l whenever 2  k  2 n . (1)

One evidently obtains a decomposition of S into 2 n disjoint subsets:

2n
S= *B .
k 1
k (2)

Since each set BkU with 1  k  2 n contains some interior points (with respect to the
surface S) and the set Bk – BkU is composed of a finite number of arcs of great circles
(which might be reduced to a single point), one concludes, according to lemma 25, that
BkU =
f BkU F (Bk – BkU) = Bk . (3)
There results from this immediately, by virtue of the hypothesis of the theorem,
B1U =
f BkU whenever 1  k  2 .
n
(4)

15
[Banach and Tarski omitted the U from A1U here.]
114 6 Decomposition of Point Sets into Respectively Congruent Parts

On the other hand, conforming with lemma 28, let

2n
S= *A k 1
k (5)

be a decomposition of the sphere S into some disjoint subsets such that

S= =
f A k , and thus A1 f A k , whenever 1  k  n. (6)
By (2) and (4)–(6) the sets A1 , A 2, ... , A 2 n and B1U, B2U, ... , B 2 n fulfill all the conditions
of corollary 12; thus one obtains
A = BU.
1 f 1 (7)
Formulas (1), (6), and (7) imply immediately that
S=
f B1 , Q. E. D.

Lemma 30. If the set A, situated on the surface S of a sphere, is not a boundary set
(with respect to this sphere), then A =
f S.

Proof. One proves easily that the set A contains a spherical polygon A1 whose area
is 4 S U2 / 2 n, where U designates the length of a radius and n is a sufficiently large
natural number. We decompose S into 2 n congruent polygons without common interior
points: n
2
S= *B .
k 1
k

According to the preceding lemma, one obtains

S=
f B1 . (1)
The spherical polygons A1 and B1 having the same area, one concludes by applying
lemma 26 that
A1 =
f B1 . (2)
From (1) and (2) this results immediately:
S=
f A1 . (3)
On the other hand,
S J A J A1 . (4)
Conforming with corollary 9, formulas (3) and (4) yield
A = S, f Q. E. D.

Lemma 30 established, the proof of the fundamental theorem 31 is evident:

Theorem 31. If point sets A and B, situated on the surface of the same sphere, are
not boundary sets (with respect to this surface), one has
A=
f B.
6.3 §3 Fundamental Theorems about Denumerable Decomposition 115

§3
The Fundamental Theorems about
Equivalence by Denumerable Decomposition

The arguments of this section concern Euclidean spaces with an arbitrary number n of
dimensions; but to fix ideas, we shall operate in the space of n = 1 or n = 2 dimensions.

Lemma 32. If A1 , A 2, ..., A m , ... are disjoint n-dimensional intervals16 congruent to a

single interval A, then
ˆ
A=
d *A
m 1
m.

Proof. Consider the case of dimension n = 1. As Felix Hausdorff has shown17 (using
an idea of Giuseppe Vitali), one can decompose every segment into a denumerable infinity
of disjoint subsets equivalent in pairs by finite decomposition. Let these be the decomposi-
tions of the segments A, A1, ..., A m , ... :
ˆ
A= *B ,
k 1
k (1)
ˆ
Am = *B
k 1
m, k for each natural number m. (2)

All these segments being congruent, one can evidently suppose that
B =~B for all natural numbers k and m,
k m, k

from which follows

Bk =
f Bm, l for all natural numbers k, l, and m. (3)
One concludes from (2) that
ˆ ˆ ˆ
*A
m 1
m = **B
m 1k 1
m, k . (4)

Since any double series can be transformed by the method of diagonals into a simple
series, formulas (1) and (4) furnish decompositions of the sets
ˆ
A and *A
m 1
m

into denumerable infinities18 of disjoint parts, which according to (3) are respectively
equivalent by finite decomposition. By virtue of theorem 4U one deduces from this that

16
A point set is called an n-dimensional interval if it is composed of all points ( x1 , x2 , ..., xn ) subject to the
condition a  xk  b whenever 1  k  n, where a and b are constants.
17
Hausdorff 1914a, 401. Strictly speaking, Hausdorff decomposes not the whole segment, but the segment
without one endpoint. However, this objection, which one can avoid, has only an insignificant influence
on the arguments that will follow.
18
[The original was singular here: “(1) et (4) fournissent une décomposition ... en une infinité ... .”]
116 6 Decomposition of Point Sets into Respectively Congruent Parts

ˆ
A=
d *A
m 1
m,

which proves the theorem for the case of linear space.

To pass to the case of n-dimensional space for n > 1, it suffices to replace the points
of each decomposed segment by (n – 1)-dimensional intervals perpendicular to it.

Lemma 33. If A1 , A 2 , ..., A m , ... are disjoint n-dimensional intervals congruent to each
other, and E denotes the entire n-dimensional space, then
ˆ
E =
d *A
m 1
m.

Proof. Consider the case n = 2. The plane E can be decomposed easily into a
denumerable infinity of (not necessarily disjoint) squares,
ˆ
E= *B
m 1
m, (1)

such that
~ B for every natural number m.
Am = (2)
m

Set
m1
C1 = B1 , Cm = Bm – *B
k 1
k for each m  2. (3)

According to (1) and (3) one easily obtains

ˆ
E= *C
m 1
m, (4)

Bm J Cm for each natural number m. (5)

Formulas (2) and (5) evidently imply the existence of sets D1 , D2 , ..., Dm , ... satisfying
the formulas
~D ,
Cm = (6)
m

Am J Dm for each natural number m. (7)

Sets C1 , C2 , ..., Cm , ... as well as D1 , D2 , ..., Dm , ... being disjoint, one concludes from (4)
and (6) in conformity with definition 2U that
ˆ
E =
d *D
m 1
m. (8)

From (7) one also deduces

ˆ ˆ
EJ * A J*D
m 1
m
m 1
m. (9)

By virtue of corollary 9U, formulas (8) and (9) immediately yield

6.3 §3 Fundamental Theorems about Denumerable Decomposition 117

ˆ
E =
d *A
m 1
m, Q. E. D.

Lemma 34. If the set A, situated in n-dimensional space E, is not a boundary set, then
A=d E.

Proof. Suppose as before that n = 2. The set A evidently contains a square AU. Let
A1 , A 2 , ..., Am , ... be disjoint squares, congruent to AU and situated in the same plane
E: the existence of such squares in the plane is manifest. By applying the two preceding
lemmas, one immediately obtains
ˆ ˆ
AU =
d *
m 1
Am and E =
d *A
m 1
m,

from which, by virtue of theorem 3U, follows

AU =
d E.

Since at the same time,

E J A J AU,
one concludes, conforming to corollary 9U, that
A=
d E, Q. E. D.

Lemma 34 established, one immediately deduces from it the fundamental theorem of

this section, namely

Theorem 35.19 If sets A and B, situated in a Euclidean space of an arbitrary number

of dimensions, are not boundary sets, one has
A=
d B.

Now we shall generalize the notion of equivalence by denumerable decomposition, by

introducing the following definition:

Definition 3. Point sets A and B are almost equivalent by denumerable decomposition,

A=
p B,

if there exist sets A1 , A 2 , B1 , and B2 satisfying the following conditions:

I. A = A1 F A 2 , B = B1 F B2 , A1  A 2 = I = B1  B2 ,
II. A1 =
d B1 ,

III. A 2 and B2 are Lebesgue-measurable 20 and m( A 2 ) = m(B2 ) = I.

19
A particular case of this theorem has been announced in Sierpiski 1918, 142. [ There, A and B were
squares of different sizes.]
20
In the arguments that will follow, we suppose that the notion of measure has been extended to unbounded
sets. See Hausdorff 1914a, 416.
118 6 Decomposition of Point Sets into Respectively Congruent Parts

The relation just defined is evidently reflexive and symmetric; we shall prove that it
is also transitive.

Theorem 36. If A =p B and B = =

p C, then A p C.

Proof. Let
A = A1 F A 2 , B = B1 F B2 , (1)
B = B1U F B2U, C = C1 F C2 , (2)
respectively, be decompositions of sets A and B, and of B and C, satisfying the con-
ditions of definition 3. From (1) and (2) stem immediately the following formulas:
B1 = ( B1  B1U) F ( B1  B2U), B1U = ( B1U  B1) F ( B1U  B2 ) .
Since A1 = =
d B1 and B1U d C1 , one then concludes from corollary 6 that sets A1 and C1
can be decomposed into disjoint parts,
A1 = AU F A 3 , C1 = CU F C3 , (3)
such that
AU =
d B1  B1U, CU =
d B1U  B1 , (4)
A = B  B U,
3 d 1 2 C = BU  B .
3 d 1 2 (5)
Set
A2 = A 2 F A 3 , C2 = C2 F C3 ; (6)
according to (1)–(3) and (6), one obtains
A = AU F A2, C = CU F C2, (7)
and from this one can easily be convinced that sets AU and A2 as well as CU and C2
are disjoint. From (4) one deduces further,
AU =
d CU. (8)
Finally, one can prove that sets A2 and C2 have measure zero. Conforming to the
properties of decompositions (1) and (2) one has, in effect,
m( A 2 ) = m(B2 ) = m(B2U) = m(C2 ) = 0. (9)
Since B1  B2U I B2U and B1U  B2 I B2 , it follows from this that
m(B1  B2U) = m(B1U  B2 ) = 0. (10)
Applying corollary 15U, according to (5) and (10) one concludes that
m( A 3 ) = m(C3 ) = 0, (11)
and from (6), (9), and (11) one finally obtains
m( A2) = m(C2) = 0. (12)
According to definition 3, formulas (6), (8), and (12) imply that
A =p C, Q. E. D.
6.3 §3 Fundamental Theorems about Denumerable Decomposition 119

The fundamental theorem on denumerable equivalence manifestly entails the following

consequence:
If sets A and B, situated in a Euclidean space, are not boundary sets, then A =
p B.

We propose to give in theorem 41 a generalization of this proposition.

Lemma 37. If A and B are sets situated in an n-dimensional Euclidean space, A is

Lebesgue-measurable, B is an open set, and m( A) = m(B), then to each positive real
number G correspond two closed sets A1 and B1 such that
I. A1 I A and B1 I B,
II. A1 =f B1 ,

III. m( A1 ) = m(B1 ) > m( A) – G.

Proof [for the case] n = 2. According to a familiar theorem in measure theory,21 there
certainly exists a closed bounded set AU satisfying the formulas
AU I A and m( A) > m( AU) > m( A) – G . (1)
Since m( AU) < m(B), one can prove the existence of squares C1 , C2 , ... , Cm and
D1 , D2 , ... , Dm that satisfy the following conditions (Ck( and Dk( denote the interiors of
squares Ck and Dk, respectively):
the sets C1(, C2(, ... , Cm( as well as D1(, D2(, ... , Dm( are disjoint, (2)
~ D (which implies C( =
Ck = ~ D () whenever 1  k  m, (3)
k k k
m m
AU I *C
k 1
k and *D
k 1
k I B. (4)

From (1), (2), and (3) follows the existence of a closed set A1 such that one has
A1 I AU I A, (5)
m( A1 ) > m( A) – G, (6)
m m
A1 I *C
k 1
D
k , and thus A1 = *
k 1
A1 ª CkD . (7)

Conforming to (3) and (7), let E1 , E2 , ... , Em be sets enjoying the following properties:
E = ~ A  C( whenever 1  k  m, (8)
k 1 k

Ek I Dk( whenever 1  k  m, (9)

and set
m
B1 = *E .
k 1
k (10)

According to (4), (9), and (10) one evidently has

21
[The regularity theorem: see Hausdorff 1914a, 408–413, or footnote 43 in section 4.4 of the present book.]
120 6 Decomposition of Point Sets into Respectively Congruent Parts

B1 I B. (11)
By virtue of (2), (7), (9), and (10), one obtains
A1 =
f B1 . (12)
Finally, the sets A1  C1(, A1  C2(, ... , A1  Cm( being closed (as intersections of closed
sets: A1  Ck( = A1  Ck ), one concludes following (8) and (10) that the sets E1 , E2 , ... , Em
and B1 are also closed, and that
m( A1 ) = m(B1 ) . (13)
Formulas (5), (6), and (11)–(13) prove that A1 and B1 are the desired sets.

Lemma 38. If A and B are point sets situated in a Euclidean space, A is Lebesgue-
measurable, B is open, and m( A) = m(B), then A =p B.

Proof. We shall define two infinite sequences { A n } and {Bn } of sets recursively in
the following way:
I. A1 and B1 are closed sets contained in A and B, respectively, and
satisfying conditions
A =B , 1 f 1 (1)
m( A1 ) = m(B1)  ½ m( A) . (2)
II. If n is an arbitrary natural number, then A n +1 and Bn +1 are
closed sets such that
n n
A n +1 I A – *A
k 1
k and Bn +1 I B – *B
k 1
k ,

and satisfying the formulas

A n+1 =
f Bn+1 , (3)
§ n
·
m( A n+1 ) = m(Bn+1)  ½ m ¨ A  * Ak ¸ . (4)
© k 1 ¹
Based on lemma 37 one proves22 by an easy induction the existence of all terms of the
sequences { A n } and {Bn }. That is indeed evident for n = 1; and if sets
A1 , A 2 , ... , A n and B1 , B2 , ... , Bn exist, one obtains easily from (2) and (4),
n n
A= *
k 1
Ak is Lebesgue-measurable, B– *B
k 1
k is open, (5)

§ n
· § n
·
m¨ A  *A k ¸ = m ¨ B  * Bk ¸ . (6)
© k 1 ¹ © k 1 ¹
By virtue of the lemma mentioned, conditions (5) and (6) immediately imply the existence
of closed sets A n+1 and Bn+1 satisfying (3) and (4).

22
[Evidently in error, Banach and Tarski cited lemma 36.]
6.3 §3 Fundamental Theorems about Denumerable Decomposition 121

Therefore, one can conclude

lim m( A n ) = lim m( Bn ) = 0 ,
n‘ˆ n‘ˆ

from which, by reason of (4) and (6),

§ n
· § n
·
lim m ¨ A  * Ak ¸ = lim m ¨ B  * Bk ¸ = 0. (7)
n‘ˆ
© k 1 ¹ n‘ˆ © k 1 ¹
Set
ˆ
AU = *A ,
k 1
k A2 = A – AU, (8)

ˆ
BU = *B ,
k 1
k B2 = B – BU. (9)

Evidently, one has

A = AU F A2, B = BU F B2, AU  A2 = BU  B2 = I . (10)
From (7), (8), and (9), one deduces directly,
m( AU) = m( A), m(BU) = m(B),
from which follows
m( A2) = m( B2) = 0 . (11)
By virtue of (4), (8), and (9), one obtains finally, by applying theorem 4U,
AU =
d BU. (12)
In conformity with definition 3, formulas (10)–(12) immediately yield
A=
p B, Q. E. D.

Lemma 39. If a bounded set A situated in a Euclidean space E is Lebesgue-measurable

and has positive measure, then there exists in the same space a set C of measure zero
such that
A F C = E. d

Proof. Since m( A) > 0, there evidently exists a bounded open bound set B whose
measure is equal to that of A. Conforming to the preceding lemma, one can conclude that
A=
p B.
23

Thus, let A1 , A 2 , B1 , and B2 be sets fulfilling the conditions of definition 3:

A = A1 F A 2 , B = B1 F B2 , A1  A 2 = B1  B2 = I , (1)
A1 =d B1 , (2)
m( A 2 ) = m(B2 ) = 0. (3)

23
[In error, Banach and Tarski wrote E in place of B here.]
122 6 Decomposition of Point Sets into Respectively Congruent Parts

Further, let C and D be bounded sets satisfying the formulas

~ B and D =
C= ~A , (4)
2 2

(C F D)  ( A F B) = I; (5)
sets A and B being bounded, the existence of sets C and D is evident.
By virtue of (1), (2), (4), and (5) one obtains easily, by applying theorem 4U,
A F C = A1 F A 2 F C =
d B1 F B2 F D = B F D. (6)
The set B F D not being a boundary set, one deduces from lemma 34 that
BFD=
d E. (7)
By reason of theorem 3U, formulas (6) and (7) immediately yield
AFC=
d E.

Since in addition the set C, following (3) and (4), has measure zero, lemma 39 is com-
pletely demonstrated.

Lemma 40. If a set A situated in Euclidean space has positive Lebesgue inner measure
(finite or not), then A =
p E.

Proof. The set A evidently contains a bounded Lebesgue-measurable subset AU with

positive measure. Let C be a set with measure zero fulfilling the conditions of lemma
39 with respect to AU. Therefore,
AU F C =
d E, (1)
m(C) = 0 . (2)
Since
AU F C I A F C I E ,
one concludes from (1), by virtue of corollary 9U, that
AFC=
d E.

From this, following corollary 6U, it follows that the space E can be decomposed into two
disjoint sets,
E = E1 F E2 (3)
such that
E1 =
d A, E2 =
d C – A. (4)
From (2) and (4) one deduces easily, applying corollary 15U, that
m(E2 ) = 0 . (5)
Set
A1 = A, A2 = I , (6)
which implies
6.3 §3 Fundamental Theorems about Denumerable Decomposition 123

A = A1 F A 2 . (7)
Formulas (3) and (7) provide a decomposition of the sets A and E that, one proves
quickly by virtue of (4)–(6), satisfies the conditions of definition 3. Therefore, one has
A=
p E, Q. E. D.

Theorem 36 and the lemma just demonstrated immediately imply the following

Theorem 41. If sets A and B, situated in a Euclidean space with an arbitrary number
of dimensions, have positive Lebesgue inner measures (finite or not), then
A=
p B.

It should be remarked that in the decompositions furnished by the fundamental

theorems of this work some Lebesgue-nonmeasurable sets must necessarily occur. Indeed,
one sees that two sets are equivalent by decomposition (finite or denumerable) into
Lebesgue-measurable sets only under the condition that they should have the same
measure. As far as almost equivalent sets are concerned, one has the following

Theorem 42. For two point sets situated in a Euclidean space with an arbitrary number
of dimensions to be almost equivalent by denumerable decomposition into Lebesgue-
measurable (or even closed) sets, it is necessary and sufficient that they should have the
same measure.24

This theorem is deduced easily from lemma 38 and theorem 36, by analyzing their
demonstrations.

24
In the same order of ideas one can establish the following theorem: if A and B should be sets situated
in a Euclidean space with an arbitrary number of dimensions, if A is Lebesgue-measurable, B is open,
and m( A) < m(B), then the set A is equivalent to a subset of B by denumerable decomposition into
Lebesgue-measurable sets.
 7
Degree of
Equivalence
of Polygons

The elementary approach to the theory of area, described in section 4.1, includes defini-
tions of polygonal region, of equivalence of a pair V, W of regions under decomposition
into the same finite number n of respectively congruent subregions with disjoint
interiors, and of the area of a polygonal region. It includes the Bolyai–Gerwien theorem:
polygonal regions are equivalent if and only if they have the same area. The 1924 papers
of Alfred Tarski and Stefan Banach, translated in chapters 5 and 6, extended that theory
to include analogous results with a different, set-theoretic, definition of equivalence.

Tarski returned to the elementary theory in the early 1930s with two papers devoted
to the degree of equivalence of a pair V, W with the same area: the smallest possible num-
ber n of pairs of congruent subregions in such a decomposition. The papers appeared
in journals addressed to gimnazjum teachers and students. The papers’ background
involves Tarski’s employment as a schoolteacher and in training teachers, which will be
discussed later, in chapter 9. Tarski’s first paper posed a question that was immediately
answered and published by a reader, gimnazjum teacher Henryk Moese; his work is
essential for understanding Tarski’s second paper. Section 7.1 of the present chapter is
concerned with publication details and with the mathematics in these three papers.

Draft translations, by Izaak Wirszup, of those three papers were published informally
in 1952 by the University of Chicago. Beyond that, these papers have attracted little
professionial attention. But Tarski presented this material in wonderful lectures to the
public and to high-school students. One of the present editors, James T. Smith, attended
those in Berkeley during the 1960s and in Regina, Saskatchewan, in 1970. Because this
mathematics is so elegant and was only barely accessible, and because Tarski was evi-
dently reluctant to disseminate those drafts further, the Tarski and Moese papers are
translated anew here, in sections 7.2 to 7.4.1

1
The drafts constituted the booklet Tarski and Moese 1952. The new translations are designated Tarski
[1931] 2014a, Moese [1932] 2014, and Tarski [1932] 2014d.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 125
DOI 10.1007/978-1-4939-1474-6_7, © Springer Science+Business Media New York 2014
126 7 Degree of Equivalence of Polygons

7.1 Introduction

The three papers by Alfred Tarski and Henryk Moese that are translated in chapter 7
appeared in two different journals:
7.2 Tarski On the Degree of Equivalence of Polygons Mody matematyk 1 (1931)
7.3 Moese Contribution to A. Tarski’s Problem “On Parametr 2 (1931–1932)
the Degree of Equivalence of Polygons”
7.4 Tarski Further Remarks about the Degree of Parametr 2 (1931–1932)
Equivalence of Polygons
Distributed together, the journals formed part of a project to enhance the quality of
mathematics instruction in Polish schools, described in section 9.7. The main journal,
Parametr, was addressed to gimnazjum teachers; its supplement, Mody matematyk
(Young Mathematician), targeted their particularly interested students.

A Hebrew translation of the first paper, by Dov Jarden, was published in 1951 in the
Israeli journal Riveon lematematika. The next year, the University of Chicago published
a draft English translation of all three papers, by Izaak Wirszup.2 The first paper was
reprinted in Polish in 1975 in the Polish journal Delta.

Tarski began his paper On the Degree of Equivalence of Polygons (section 7.2) by
defining the notions of polygonal region and equivalence by finite decomposition and
presenting nontrivial examples. His style in addressing his student audience was less
formal than that in his research papers: for example, he left the requirement of disjoint
interiors to readers’ interpretation of the verb decompose.3 For polygonal regions W, V
that are equivalent— W { V —Tarski defined the degree V(W, V ) of equivalence to be
the smallest number of subregions into which each can be decomposed, so that all corre-
sponding pairs of subregions are congruent.4

Tarski listed four elementary properties of this function V. He omitted proofs, but
suggested that the most involved of these,
W { U & V { U _ V(W, V )  V(W, U ) $ V(V, U ), (property 4)
could be proved by imitating the proof of the transitive law for {.5 Next, Tarski defined
the diameter G( P) of a polygonal region P as the maximum distance between two points

2
The Hebrew translation is Tarski [1931] 1951. The English drafts constituted the booklet Tarski and
Moese 1952, which was reproduced from typescript. The present editors do not know the circumstances
of those publications. For information about Jarden and Wirszup, see the box on page 130.
3
In Polish, podzieli.
4
Two doctoral students of Felix Bernstein at the University of Halle–Wittenberg had started an investiga-
tion of this subject. Hans Brandes (1907) considered an analogous notion of degree of equivalence in which
only triangular subregions are allowed; he used it to analyze the complexity of proofs of the Pythagorean
theorem. Paul Mahlo (1908) derived some more-general results and applied them to the same problem.
Kitizi Yanigahara (1927) extended this approach and reported related work of Wilhelm Süss. Tarski did
not mention any of this earlier research. For more information, see Pambuccian 2004.
5
See theorem 3 in section 5.1.
7.1 Introduction 127

of P. He gave a detailed proof of the conditional inequality

G(W )
W { V & W convex _ V(W, V )  , (property 5)
G(V )
then left to his readers its generalization to pathwise connected regions W. Tarski noted
that such inequalities can be used to establish upper and lower bounds of V(W, V ), and
that any example decomposition also provides an upper bound. For several cases Tarski
used these bounds to completely determine V(W, V ). In others, his upper bound was
larger than his lower bound, and thus the exact value of V(W, V ) remained unknown.

The remainder of the paper was devoted to theorems about a new function W, defined
for each positive real number x as follows: W( x) = V(W, V ), where W is a unit square
and V is a rectangle with edge lengths x and 1 x .

Tarski encouraged his readers to establish for themselves the first major result,

W( x)  2 + M x 2  1 N for every x  1.6 (theorem III)

He suggested examining the demonstrations of two related theorems in elementary
textbooks: for example, the well-known text Outline of Elementary Geometry for Use in
Secondary Schools, by Wadysaw Wojtowicz.7 Analyzing those proofs would, Tarski
claimed, lead students quickly to remarkable new mathematics. As a further enticement,
he indicated slyly that this study had led him to the puzzling decompositions in figure
2 of his paper (page 136). His arguments are presented in the following three paragraphs.
Their sophistication suggests the level of involvement that Tarski expected of his student
readers.

Tarski’s first hint was Wojtowicz’s figure 148, adapted below. It shows that two paral-
lelograms ABCD and ABCUDU with the same base AB and altitude h can each be
decomposed into n subregions such that corresponding subregions are congruent.
Denote by g the altitude, with respect to the same base, of the intersection of those two
parallelograms, shaded in the figure; then n = 1 + M h gN . The case shown, in which
n = 5, is typical for the situation in which h > g and h g is not an integer. When g does
divide h, the details at the top of the figure become simpler, but the same formula holds.
When h  g, the edges opposite the common base overlap, and the formula yields n = 2,
the correct value.8

6
For every real number x, M x N denotes the ceiling of x: the smallest integer greater than or equal to x.
7
Wojtowicz [1919] 1926, §§177, 191, 192. For information about Wojtowicz, see a box in section 9.3.
8
The Archimedean axiom justifies replicating the shaded triangle upward by steps until a copy overlaps
the line CD. Most authors, instead, note that ABCD and ABCUDU would be obviously equidecomposable
if they overlapped, then move CD rightward by steps, maintaining an overlap each time, until it collides
with CUDU. Wojtowicz’s method facilitates counting the subregions in the decomposition.
128 7 Degree of Equivalence of Polygons

D C DU CU

A B

Parallelograms with the Same Base and Altitude

The second hint referred to a construction in Wojtowicz’s §§191–192, which prepared

for his proof of the Pythagorean theorem. The next figure is an adaptation for the
situation at hand. Given a unit square W = BCCUBU and a number x > 1, extend CCU
past C to point A so that ABC is a right triangle with leg BC of length 1 and
hypotenuse AB of length x. Locate point AU on ACU so that ABBUAU is a parallelogram,
and point X where lines BC and AUBU intersect. (The figure shows the case in which
AU lies between A and C and, consequently, X between B and C; a similar argument
is required when AU lies between C and CU.) Finally, locate points D and E on the
line AUBU so that V = ABDE is a rectangle. By straightforward computation,

1 x,
1
BD has length BX has length .
x2  1

CU

C 1
W
E AU X D BU

A x B

Adaptation of Wojtowicz’s Argument

Square W and parallelogram ABBUAU have the same base BBU and altitude 1. By the
argument of the previous paragraph, they can each be decomposed into n subregions
such that corresponding subregions are congruent, where n = 1 + M h gN and h and g
are the lengths of BC and BX. Moreover, their triangular intersection BXBU is a
component of each decomposition. Split that into two right triangles BXD and BDBU,
and notice that the latter is congruent to triangle AEAU. Thus, square W and rectangle
V are each decomposed into 2 + M h gN subregions, so that corresponding subregions are
congruent. Finally, as desired,
7.1 Introduction 129

W( x) = V(W, V )  2 + M h gN = 2 + M x 2  1 N .

The next results were much easier. The simplest is an improvement over the previous
inequality for the case in which x is an integer n:
W(n)  n for all integers n > 0. (theorem IV)
To see this, note that in this case,9 each of W and V can be decomposed into n rectan-
gles with edge lengths 1 and 1 n , hence V(W, V )  n. Further, for every x > 0, integer
or not,
1
G(W ) = 2 , G(V ) = x2  ,
x2
and by property (5),

G(W ) x4  1
W( x) = V(W, V )  = . (theorem V)
G(V ) 2 x2
Theorems IV and V yield 1.4  W(2)  2, and thus W(2) = 2; similarly, W(3) = 3. How-
ever, these results do not alone determine W(4). In the problem section of the same
journal issue, Tarski formally proposed finding a proof that W(4) = 4, a known result.10

Tarski concluded this paper with more exercises and two conjectures. Two exercises
are simple applications of theorem III. A third, W(21 4)  4, would require a new example,
which Moese would describe in the next paper. Tarski’s first conjecture, that W(n) = n
for every positive integer n, was suggested by its known instances for n  4. It would
soon be confirmed by Moese. The second conjecture—that W( x)  3 for every positive
x=/ 1 2 , 1, 2 —would in fact be refuted by Tarski himself in the third of these papers.

Moese began his paper, A Contribution to A. Tarski’s Problem “On the Degree of
Equivalence of Polygons,” by confirming Tarski’s conjecture that W(n) = n for all positive
integers n.11 As just noted, Tarski had already shown that W(n)  n. To derive the
opposite inequality, Moese introduced two ideas: strips of width x between parallel lines,
and VW ( x),12 the minimum number of such strips whose union contains a specified
bounded region W. This enabled him to substitute for the detailed analysis of the
geometry of the unit square V that of the simpler geometry of its inscribed disk and the
corresponding hemisphere. It is easy to see that W(n)  VV ( x) = VK ( x), where x = 1 n
and K is a disk with radius 1 2 . Noting that the surface area of the portion of the
hemisphere over K that lies over one of these strips depends only on x, he showed that
VK ( x) = M 1 x N . It follows that W(n)  n.

9
Generalize Tarski’s figure 7 in section 7.2, page 141, where n = 3.
10
This problem is Tarski [1931] 2014e: see section 12.10.
11
For more information about Moese, see the box on page 130. His paper is translated in section 7.3.
12
Moese’s slightly different notation for this expression did not make the dependence on W explicit.
130 7 Degree of Equivalence of Polygons

Dov Jarden was born in 1911 in Motol, then in the Russian Empire (now Motal’, in Belarus). He
studied in Jewish schools there and in Warsaw, earned a rabbinical degree, and emigrated to Israel
in 1934. At the Hebrew University of Jerusalem he earned master’s and doctoral degrees in mathe-
matics and Hebrew studies, respectively. In 1946 he founded the journal Riveon lematematika.
Jarden authored, translated, and edited many works on geometry, number theory, the Hebrew
language, and medieval Jewish history. He taught in secondary schools and universities in Israel and
the United States. Jarden died in Jerusalem in 1986.*

Izaak Wirszup was born in 1915 in Vilnius, then under occupation by the German Army. He began
university studies there during the 1930s, particularly with the mathematician Antoni Zygmund.
After World War II, he rejoined Zygmund at the University of Chicago, earned the doctorate there
in 1955, became a faculty member, and was appointed full professor there in 1965. Wirszup was
influential in mathematics education both locally and nationwide, and remained active until his death
in 2008.†

Henryk Moese was born in 1886 in the Maopolska region of southern Poland. He began a teaching
career around 1910, probably at a private gimnazjum in Kolbuszowa, a small town in southeastern
Poland, then in the Austrian Empire. In 1919 he moved from there to rem, a small city in western
Poland, to become a teacher of mathematics, physics, and geography at the public General Józef
Wybicki Boys’ Gimnazjum. By 1929 he had become its assistant director. His workload consisted
of twenty-four hours per week in classes, and twelve hours of other duties. Moese’s namesake, prob-
ably his son, entered the first (youngest) class of that school for the year 1927/1928, and continued
there through the fourth class, in 1930/1931.
In 1930 Moese was appointed director of the Liceum Ogólnoksztacce, in Kpno, a small city in
the southwest, at that time near the German border. The younger Henryk enrolled there a year later.
A publicly funded gimnazjum with a classical emphasis, it enrolled about 275 students—about one-
fourth, young women. Moese was responsible for a faculty of about twelve, and taught mathematics
himself for six hours a week to the upper three classes. The teaching loads of the other faculty aver-
aged about twenty-seven hours per week. The school is pictured on page 131; Moese served there
until 1933. During that time he was also involved in scouting. That year, Moese evidently moved
on. In September 1935 he was serving as director of the Mikoaj Kopernik Gimnazjum in Toru, the
historic city on the Vistula halfway between Warsaw and the Baltic. The portrait on page 131 shows
him in his study there.
In a 1929 commemorative school report, Moese published a study of an intriguing arithmetic
puzzle, the “seven sevens problem.” During the early 1930s he contributed materials for publication
in the journals Parametr and Mody matematyk, which were aimed at Polish gimnazjum teachers and
their most interested students (see section 9.7). These included the paper described in this section
and translated in section 7.3, and many smaller items.
The present editors have not yet traced the elder Moese’s life before or after the period covered
in this sketch. The younger Henryk was probably the Polish teacher and scholar born in Wysoka in
southeastern Poland in 1917, who completed gimnazjum studies in Toru in 1935 and earned a doc-
torate in philosophy from the university there in 1955.‡

*Yarden 2006. † Chicago 2008.

‡ Zagórowski 1924–1926, volume 1, 275, and volume 2, 231. rem 1928, 7, 11, 13; 1929, 88; 1930, 61; 1931,
12, 58. Kpno 1931, 30, 33; 1932, 31–32, 73; 1933, 46, 65. Kurzawa and Nawrocki 1978, 129 (“Moese” is mis-
spelled there). L. Sobociski 1935. Moese 1929. Rusiecki and Straszewicz 1932, 154. Toru 1995, 478. For
more information about Polish secondary schools of this era, see chapter 9.
 Gimnazjum in Kpno
Where Henryk Moese Taught
from 1930 to 1933

Henryk Moese
in His Study
in Toru, 1935
132 7 Degree of Equivalence of Polygons

In his second paper, Tarski reported that Zenon Waraszkiewicz had been the first to
prove that W(n) = n for all positive integers n, but with an argument too complex for
publication in Parametr. Moese proved it independently with the simpler argument just
described.

Zenon Waraszkiewicz was born in 1909 in Warsaw, then part of the Russian Empire. His parents
were schoolteachers. The family took refuge in Odessa during World War I, then returned to Warsaw.
Zenon completed gimnazjum studies there in 1926, and entered the University of Warsaw to study
mathematics. He earned a master’s degree in 1930, and the doctorate in 1932, supervised by Stefan
Mazurkiewicz. Waraszkiewicz continued research in point-set topology and analysis, and earned the
venia legendi in 1937. Until World War II he taught in Warsaw secondary schools, served as assistant
at the Warsaw Polytechnic University, and as dozent at the University of Warsaw. During the
German occupation, Waraszkiewicz taught in the Polish clandestine schools. In 1945 he became a
professor at the new University of ódz, but died there that same year.*

*Derkowska 2001; Tatarkiewicz 2003.

Moese’s next result, W(x)  1 + M xN for all real x  1, strengthened the upper bound
2 + M x 2 – 1 N in Tarski’s earlier paper.13 That required only an example decomposition
of the unit square and an x by 1 x rectangle into that number of components. Moese
merely presented a diagram14 for the case M xN = 4; readers may discover his reasoning
by reconstructing that figure. The paper continued with some further results about
values of W, which can be used to solve exercises posed in Tarski’s earlier paper. More-
over, a major result in Tarski’s next paper would rest on Moese’s final result: for any
positive integers n and p,
§ 1·
W¨n  ¸  n + 1. (theorem 4)
© p¹
Moese concluded with two conjectures, the first of which concerned that theorem:
W( x) = n + 2 for every positive integer n and every x such that n < x  n + 1, except
when x = n + 1 p for some positive integer p, in which case W( x) = n + 1.

13
See theorem III on page 142; it is straightforward to show that the new bound is always less than or equal
to the earlier one.
14
Figure 4 on page 149.
7.1 Introduction 133

In his paper Further Remarks about the Degree of Equivalence of Polygons (section 7.4),
Tarski defined the width Z( F ) of a plane figure F to be the smallest width of any strip
that contains F, provided there is one.15 He then proved that if F is contained in some
strip, contains a disk of the same width, and is the union of subsets C1 , ... , Cn , then
n
Z( F )  ¦Z
i 1
Ci . (theorem A)

For the preliminary case in which F is a disk with diameter d = Z( F ), he suggested

that readers imitate a proof in Moese’s preceding paper. Indeed, each Ci is contained
in some strip with width Z(Ci ); the portion of the hemisphere H over F that lies over
this strip has surface area 1 2 S d Z(Ci ); and the sum of these surface areas cannot be less
than that of the entire hemisphere, which is 1 2 S d 2 = 1 2 S d Z( F ). Tarski provided the
remaining details of the proof of theorem A, then quickly derived his next result: if W
and V are equivalent polygonal regions and W satisfies the condition of theorem A, then
Z(W )
V(W, V )  . (theorem B)
Z(V )
Tarski noted that when applicable, theorem B can provide a larger lower bound for
V(W, V ) than property 5 in his previous article.16 Thus, it would be useful to generalize
it, requiring that W be merely convex or connected rather than contain a disk of the
same width. Such a generalization would follow from an analogous one of theorem A.
The present editors are not aware of any such generalizations in the literature.

In Tarski’s earlier paper, property 5 entailed theorem V, a lower bound for the function
values W( x). He showed that the discussion in his second paper leads easily to a much
better lower bound: W( x)  M xN whenever x  1. Combining that with Moese’s upper
bound, Tarski concluded that for every x  1,
W( x) = M xN or 1 + M xN . (theorem E)
Consequently, for any positive integers n and p,
§ 1·
W¨n  ¸= n + 1. (theorem F)
© p¹
Indeed, this value of W is n + 1 or n + 2 by theorem E, and Moese had shown
already 17 that it cannot exceed n + 1. Moese had conjectured that W( x) = MxN only when
x = n + 1 p for some positive integers n and p.

15
Tarski used the notation S(F) for the width of F; in the present book, Z(F) is used to avoid confusion
with the mathematical constant S.
16
See page 127 in the present section.
17
See page 132 in the present section.
134 7 Degree of Equivalence of Polygons

In his final theorem H, Tarski confirmed this for the case 1 < x  2: that is, if
1 < x  2, then a unit square Q can be decomposed into connected polygonal subregions
A and B that can be rearranged to form a nonsquare rectangle R x with edges of length
x and 1 x only when x = 1 + 1 p for some positive integer p. Tarski attributed this
result to Adolf Lindenbaum and Zenon Waraszkiewicz, but claimed that the proof was
too subtle to include in this paper. The figure on the next page shows that a decomposi-
tion is possible for x = 1 + 1 p . Tarski suggested that their proof entailed a result
stronger than theorem H: the only way to decompose Q into two subregions A and
B that can be rearranged to form Rx for any x is to make both A and B congruent
to a region A p , for some positive integer p, as shown in that figure. This confirms the
conjecture of Henryk Moese that concluded his [1932] 2014 paper, translated in
section 7.3.18

Tarski noted that if the restriction on x in theorem H could be removed, study of the
function W would be rather complete. But he did mention several different directions
in which this subject could be developed further. Later, Tarski himself studied the effect
of limiting the number of edges of the polygons allowed in the dissections—his work made
essential use of theorem H.19 The present editors are not aware of any other studies.
Tarski’s paper concluded with a selection of exercises, of various levels of difficulty.

–


Ap Ap

1
p steps of height –
p

Bp

–
Bp
1
p + 1 steps of width –—
p+1

Dividing a Square into Two Polygonal Regions

That Form a ( p+1) p by p ( p+1) Rectangle ( p = 4)

18
See also the abstract, Lindenbaum [1937] 1938. The present editors have seen a promising sketch of an
entirely elementary proof by J. Shilleto (2013).
19
See section 8.5.
7.2 The Degree of Equivalence of Polygons (1931) 135

7.2 The Degree of Equivalence of Polygons (1931)

This section contains an English translation of Alfred Tarski’s paper O stopniu równo-
wanoci wieloktów, [1931] 2014a, which appeared in volume 1 of the journal Mody
matematyk. Aimed at particularly interested gimnazjum students, this journal was dis-
tributed with another one, Parametr, addressed to their teachers.20 A draft English
translation by Izaak Wirszup was published informally by the University of Chicago as
Tarski and Moese 1952. The present translation was carried out independently, but used
Wirszup’s for confirmation.

This translation is meant to be as faithful as possible to Tarski’s original text. Its only
intentional modernizations are punctuation, and some changes in symbols, where Tarski’s
conflict with others used throughout this book. A bibliographic reference has been
adjusted to conform with the conventions of the present book. All [square] brackets in
the translation enclose editorial comments inserted for clarification, often as footnotes.
As an aspect of adjusting punctuation, the editors increased white space to enhance visual
organization.

DR. ALFRED TARSKI (Warsaw)

On the Degree of Equivalence of Polygons

In this article I want to discuss some concepts, belonging entirely to the realm of elemen-
tary geometry, which until now have been investigated hardly at all.

As is well known, we call two polygons W and V equivalent, expressing this with the
formula W { V, if they can be divided into the same number of respectively congruent
polygons.21 This subdivision of equivalent polygons into congruent parts is not unique:
two equivalent polygons can be divided into congruent parts in various ways, with respect
to the number as well as the form of these parts.

We explain this with an example: figure 1, and figure 2 as well, show that a square
with edge a and a rectangle with edges 5 4 a and 4 5 a are equivalent to each other,
but their subdivisions in the two figures are quite distinct.

In connection with this observation, a question arises in a natural way: what is the
least number of respectively congruent parts into which two given equivalent polygons
can be divided? We want to touch upon a problem of exactly this type in Parametr.

20
For further information about these journals, consult sections 7.1 and 9.7.
21
[This notion of equivalence is discussed in section 4.1. It is different from that used in chapters 5 and
6. The two notions are compared succinctly on page 59.]
 W
W2 W3 W4 W5 V
V5
V4
W1 V1
V3
V2

Figure 1

W
V
W3

W2 V2 V3
W1 V1

Figure 2

Figure 3 Figure 4
7.2 The Degree of Equivalence of Polygons (1931) 137

With this goal, we adopt the following definition. The degree of equivalence 22 of two
equivalent polygons W and V shall be the least natural number n satisfying this
condition: each of the polygons W and V can be subdivided into n polygons in such
a way that the polygons obtained by subdivision of W are respectively congruent to the
polygons obtained by subdivision of V. We shall denote the degree of equivalence of
polygons W and V by the symbol V(W, V ).

A certain remark ought to be made here. It is convenient in the present considerations

to give the word “polygon” a meaning broader than that which is used at the beginning
of instruction in elementary geometry. Specifically, here we call a polygon in the broader
sense a plane figure that is the composition of a finite number of polygons in the ordinary
sense of that word. For example, the figure composed of rectangles W2 and W4 in figure
1 is a polygon in the broader sense, and so is the figure composed of both of those rectan-
gles and the quadrilateral W3 in figure 2 as well. We note in passing that the extension
of the concept of polygon is extremely useful in the whole theory of the equivalence of
polygons. Without this extension, many arguments from this theory [that are] encoun-
tered in elementary textbooks err from a lack of precision.

Applying the broader meaning to “polygon” presents a certain difficulty in properly

defining the notion of congruence. We restrict ourselves here to the following visual
explanation: two polygons in the broader sense— like geometric figures of any kind—are
congruent if one of them can be “laid upon” the other (without changing the relative
positions of the component parts of either of them) in such a way that they “coincide.”
For example, the polygon shown in figure 3 is not congruent to the polygon shown in
figure 4, but is equivalent to it.

Until now we knew very little about the degree of equivalence of polygons. We give
here as examples a few elementary properties of this concept.
1. For any equivalent polygons W and V,
V(W, V ) = V(V, W ).
2. In order for V(W, V ) = 1, it is necessary and sufficient that polygons W
and V be congruent; in particular, for any polygon W, we have
V(W, W ) = 1.
3. If the polygon W can be divided into polygons W1 and W2 and polygon
V into polygons V1 and V2 in such a way that W1 { V1 and W2 { V2 ,
then
V(W, V )  V(W1 , V1) + V(W2 , V2) .

22
As far as we [Tarski] know, this concept was introduced by Dr. Adolf Lindenbaum (Warsaw) who together
with the author of this article established some of the properties of this concept.
138 7 Degree of Equivalence of Polygons

4. If W { U and V { U, then
V(W, V )  V(W, U ) # V(V, U ) .
Properties 1–3 are self-evident. Also, justification of property 4 does not present difficul-
ties to those among the Readers who are aware of the application of the so-called double-
subdivision-network method in the proof of the theorem that two polygons equivalent to
a third are equivalent to each other.

For the formulation of the next property the concept of the diameter of a polygon is
necessary. The diameter of a polygon W, designated by G(W ), shall be a longest one of
the segments joining two points of the polygon W. It is easy to show that every polygon
has a diameter; there may be many congruent diameters.
5. If W and V are equivalent polygons, and in addition W is a convex
polygon, then
G(W )
V(W, V )  .
G(V )
Proof. Let us apply reasoning by contradiction.23 Specifically, suppose that contrary
to the conclusion of [property] 5,24
G(W )
V(W, V ) = n < . (1)
G(V )
It follows immediately that
G(W )
G(V ) < . (2)
n

In accordance with the definition of diameter there can be found in polygon W two
points A0 and A n that are the endpoints of the diameter G(W ). Divide [segment]
A 0 A n into n congruent subsegments 25 and let A1 , A 2 , ..., A n –1 be the subdivision points.
Each of the segments A k A k+1 (where 0  k < n) is congruent to an nth part of the
diameter G(W ); therefore, in view of (2) we have A k A k+1 > G(V ). From this we conclude,
moreover, that
A k A l > G(V ) (3)
for any distinct natural numbers k and l lying between 0 and n [inclusive].

According to the definition of degree of equivalence, from (1) it follows that polygons
W and V can each be subdivided into n [polygonal] parts that are respectively congru-
ent. Let W1 , W2 , ..., Wn be the polygons obtained by subdivision of W, and V1 , V2 , ..., Vn
be polygons respectively congruent to them, obtained from subdivision of V.

23
[Tarski’s term was rozumowanie apagogiczne: apagogic reasoning.]
24
[Here and twice afterward, Tarski referred to property 5 as “theorem 5.”]
25
[Tarski’s term was czci: parts.]
7.2 The Degree of Equivalence of Polygons (1931) 139

As we know, points A 0 and A n belong to the polygon W. Since W is by assumption

a convex polygon, all points of the segment A 0 A n , particularly A 1 , A 2 , ..., A n –1 , belong
to W. In this way, we distinguished n + 1 points A 0 , A 1 , ..., A n in polygon W and at
the same time we subdivided this polygon into n parts W1 , W2 , ..., Wn . We infer, there-
fore, that at least two of the indicated points belong to the same part: for instance, points
A k and A l , [with] k = / l, belong to the part Wm .

Since the polygons Wm and Vm are congruent, we can certainly find in polygon Vm
two points Bk and Bl such that segment Bk Bl is congruent to segment A k A l. Points
Bk and Bl , belonging to Vm , must also belong to V; consequently, segment Bk Bl
cannot exceed the diameter of polygon V. Replacing Bk Bl with the congruent segment
A k A l , we obtain the formula
A k A l  G(V ), (4)
where k and l are two distinct natural numbers lying between 0 and n [inclusive].
In view of the evident contradiction between (3) and (4), we must reject hypothesis (1)
and accept [property] 5 as proved.

[Property] 5 may be generalized, replacing the condition “W is a convex polygon” by

the condition “W is a connected polygon” (that is, a polygon, any two points of which
can be joined by a broken line, all of whose points belong to this polygon). The proof of
this generalized theorem, which requires a slight modification of the original proof, we
leave to the Reader.26

Using the definition of degree of equivalence and the properties of this notion given
above, it is possible to investigate the degree of equivalence in reference to different
concrete pairs of equivalent polygons. In general, we are able to find only certain upper
and lower bounds for the degree of equivalence of each particular pair of polygons.

We obtain an upper bound immediately in those cases in which we have a drawing

establishing the equivalence of polygons W and V by decomposition into respectively
congruent parts: if in each of those the number of parts is n, then by virtue of the
definition of degree of equivalence, we will have
V(W, V )  n.
Also, in establishing an upper bound, we can sometimes use properties 3 and 4.

For a lower bound we have first of all the trivial bound V(W, V )  2, which by virtue
of property 2 follows for any pair of polygons W and V that are equivalent but not
congruent. It is significantly more difficult to obtain a stronger lower bound; for the
moment, we have available only property 5.

Only in a few cases has it been possible to obtain a lower bound coinciding with an
upper bound, and thus to determine exactly the degree of equivalence of the polygons.

26
[ Let p be a broken line in W from A to A n. Let B0 = A, Bn = A n, and when 0 < i < n, let Bi be an
intersection of p with the perpendicular to the diameter G(W ) at Ai . Then Bk Bl  Ak Al  G(W )/n
whenever k = / l, and Tarski’s argument with B0 , ... , Bn in place of A 0 , ... , An yields the result.]
140 7 Degree of Equivalence of Polygons

We give here a few examples.

A. Let W and V be the square and rectangle, respectively, in figure 1 or in figure

2. Figure 1 yields V(W, V )  5. On the other hand, from figure 2 we obtain
a stronger bound: V(W, V )  3. According to property 2 we have V(W, V )  2;
in this case, property 5 does not give a better bound. Thus, finally,
2  V(W, V )  3.
The question which of the numbers 2 and 3 is the value for V(W, V ) remains
open.
B. Let V be a rectangle with edges 5 4 a and 4 5 a (as in the previous example)
and let U be a rectangle with edges 5 2 a and 2 5 a. Figure 5 gives V(U, V) 
2. Since, on the other hand, we have V(U, V )  2 according to property 2 or
5, we finally obtain
V(U, V ) = 2.
C. Let W, V, U be the figures described in examples A and B. According to
property 4 we have
V(W, U )  V(W, V ) # V(U, V ).
As we showed in [examples] A and B, V(W, V )  3 and V(U, V ) = 2; therefore,
V(W, U )  6. In this case, however, instead of applying property 4, it is better
to base the argument directly on figure 6, which yields V(W, U )  4. By
property 2, finally, we have
2  V(W, U )  4.
The question of finding the exact value of V(W, U ) again remains open.
D. Let W be a square with edge a and let V be a rectangle with edges 3 a and
1 3a. From figure 7 it can be seen that V(W, V )  3. On the other hand,
applying property 5, we obtain
G(W )
V(W, V )  .
G(V )
It is easy to see that the diameters of the rectangles are their diagonals; in
view of this,

G(V ) = a 82 9 , G(W ) = a 2 , and therefore V(W, V )  41 9 > 2.

Since V(W, V ) is a natural number, the inequality V(W, V ) > 2 can be replaced
by the inequality V(W, V )  3. The upper and lower bounds coincide, and
consequently we have
V(W, V ) = 3.
We content ourselves with these examples.
 V

V2 U

V1 U1 U2

Figure 5

W
W3 W4

U W2

U4
U1 U2 W1
U3

Figure 6

W3

V W2

V1 V2 V3 W1

Figure 7
142 7 Degree of Equivalence of Polygons

Referring to the first sentence of the present article, we repeat once more that in this
part of elementary geometry nearly everything is left to do. A whole series of attractive
topics to develop arises here, from which we extract the following.

Let Q be a square with edge a, while P should be a rectangle with edges x # a and
1 x # a, where x is any positive real number. Polygons Q and P are clearly equivalent.
It is easy to see that their degree of equivalence is a function of x ; we shall denote it by
the symbol W(x). Thus, we set
V(Q, P) = W(x).
The topic whose development we would keenly recommend would be a precise investiga-
tion of the function W(x).

In the following theorems, we present some of the properties known to us.

I. The function W(x) is defined for all positive numbers and takes on as its
values only positive whole numbers.
II. W(x) = W(1 x ) for every x > 0.
These are direct consequences of the definition of the function W(x) and of the definition
of the degree of equivalence.

Moreover, denoting by the symbol M xN the least integer n not less than a given real
number x (and therefore satisfying the formula n – 1 < x  n), we have

III. W(x)  2 + M x 2  1 N for any x  1.

We shall not give a proof of the above theorem. We remark only that this proof can be
obtained by analyzing the proofs of two known theorems from the theory of equivalence
of polygons, namely (1) the theorem of equivalence of two parallelograms whose bases
and altitudes are respectively congruent, [and] (2) the theorem that the square con-
structed on a leg of an arbitrary right triangle is equivalent to the rectangle constructed
from the hypotenuse and from the projection, on the hypotenuse, of the leg under consid-
eration.27 We recommend that the Reader give a complete proof for the theorem discussed,
or at least for a few of its special cases, such as 28

W(11 3 )  3 W(2 1 4 )  4 W ( 10 )  5.
Theorem III establishes a certain upper bound for the function W(x). In some cases
in which x is a rational number, it is possible to obtain other, often stronger, bounds for
the function in question. Moreover, in establishing these bounds it is not necessary to

27
Compare the proofs of both these theorems in the textbook Wojtowicz [1919] 1926, §177 and §§191–192.
(By an analysis of the proof of the second of these theorems, we obtained figure 2, given earlier.)
28
[Section 7.1 includes this proof of theorem III. Henryk Moese soon improved this result, in the paper
Moese [1932] 2014 translated in section 7.3. The first and last special cases are straightforward applica-
tions, but the middle one required a new example subdivision, also provided by Moese.]
7.2 The Degree of Equivalence of Polygons (1931) 143

resort to subdividing the rectangles into figures that are not rectangles (in contrast with
theorem III).

Omitting here the general case, we give the following easy theorem:
IV. W(n)  n for every positive integer n.
This follows from the following observation: a square with edge a can be subdivided into
n congruent rectangles with edges a and 1 n # a, from which a rectangle with edges
n # a and 1 n # a can then be constructed (compare figure 7 for n = 3).

An easy proof of [the following] theorem is based on property 5 of the degree of

equivalence (compare example D given above):

x4  1
V. W(x)  for every x > 0 .
2 x2

We are unable at this point to strengthen in significant measure the lower bound for
the function W(x) established in the previous theorem.29 However, a certain strengthen-
ing, devoid of greater significance, is presented by the following theorem, which we give
here without proof:

x4
W(x)  for x  1 .
2 x2  1

Finally, we note a direct conclusion from theorem V:

VI. W(x)  +ˆ as x  +ˆ.

With the help of the above theorems, we can calculate the values of the function
W(x) for certain values of the argument—a few, anyway. Thus, we have W(1) = 1, W(2) =
W( 1 2 ) = 2, and W(3) = W( 1 3 ) = 3 (concerning that last formula, compare example D).
By a slightly different method it may be shown that W(4) = W(1 4) = 4: the justification
for this formula we leave to the Reader.30

On the other hand, establishing the value of the function W(x) for other values of x,
even for integer values x  5, still presents difficulties. In particular, we are unable at
this point to prove the following theorem,31 which seems highly probable:
W(n) = n for any positive integer n.

29
[A proof is included in section 7.1. Tarski did improve this result significantly with theorem C of the later
paper Tarski [1932] 2014d translated in section 7.4.]
30
[In the problem section of the same issue of Mody matematyk, Tarski posed the proof of W(4) = 4. This
is an immediate consequence of a general theorem in the paper Tarski [1932] 2014d translated in section
7.4, but Tarski suggested another proof. See section 12.10.]
31
[This conjecture was soon proved by Moese and others: see section 7.3.]
144 7 Degree of Equivalence of Polygons

As another example of a theorem32 unproved at this point, but probable, we quote the
following statement:
W(x)  3 for every positive x different from 1 2, 1, and 2.
This statement, together with theorem III, would permit calculation of values of the
function W(x) for infinitely many values of the argument. In fact, we would have W(x) =
3 for every x satisfying the inequalities 1/ 2  x  2 and x = / 1.

From the remarks above it follows clearly that we are at the present moment still very
far from precise knowledge of the course of values of the function W(x).

The editors encourage Readers to send in results of further investigations of topics

touched on in the above article.

7.3 Moese’s Contribution (1932)

This section contains an English translation of Henryk Moese’s paper Przyczynek do prob-
lemu A. Tarskiego: “O stopniu równowanoci wieloktów,” [1932] 2014, which appeared
in volume 2 of the journal Parametr. The journal’s target audience was gimnazjum
teachers and their most serious students. Moese was a gimnazjum teacher in Kpno, a
small city in southwestern Poland.33 In this paper he solved a problem posed in the [1931]
2014a paper of Alfred Tarski mentioned in the title, which is translated in section 7.2.
Moese’s work was fundamental for Tarski’s subsequent [1932] 2014d paper, translated
in section 7.4. For that reason, and because of its previous inaccessibility, it is included
here in this volume devoted to Tarski. A draft English translation by Izaak Wirszup was
published informally by the University of Chicago as Tarski and Moese 1952. The present
translation was carried out independently, but used Wirszup’s for confirmation.

This translation is meant to be as faithful as possible to Moese’s original text. Its only
intentional modernizations are punctuation, and some changes in symbols, where Moese’s
conflict with others used throughout this book. Some uses of alternative type styles for
emphasis have been modified. Bibliographic references have been adjusted to conform
with conventions of the present book. All [square] brackets in the translation enclose
editorial comments inserted for clarification, often as footnotes. As an aspect of adjusting
punctuation, the editors increased white space to enhance visual organization.

32
[Tarski did write theorem ( twierdzenie), but he refuted this conjecture in his later paper Tarski [1932]
2014d, translated in section 7.4.]
33
For further information about Parametr, consult sections 7.1 and 9.7. For information about Moese, see
the box on page 130.
HENRYK MOESE (Kpno)

A Contribution to the Problem of A. Tarski,

“On the Degree of Equivalence of Polygons”

In connection with the article On the Degree of Equivalence of Polygons by Dr. Alfred
Tarski,34 I would like to prove here some theorems concerning the function W( x). On
questions of notation, I refer Readers to the cited article by Tarski.

Theorem 1. W(n) = n for every natural 35 number n.

Proof [which continues to page 148]. The part of the plane contained between two
parallel lines at mutual distance x, we call simply an x-strip.36 Figure 1, with its hatched
part, sufficiently explains the definition of x-strip.

Figure 1

We define now a certain auxiliary function Vw ( x). Let W be an arbitrary bounded

plane figure (that is, [one that] can be covered by a polygon). To cover this entire figure
with x-strips, a certain number of strips must be used. The number of strips needed to
cover the figure depends on the way of placing the strips on the figure. It can be shown
easily that there is a least number of strips that can cover the entire figure. This mini-
mum number of strips we denote precisely by the symbol Vw ( x).37

Lemma I. If W is a disk with radius r, then

ª 2r º
Vw ( x) = « » for 0 < x  2 r.
«x»
(Here, M 2 r / xN denotes the least integer n not less than the real number 2 r / x .)

34
[Tarski [1931] 2014a, translated in section 7.2.]
35
[For Moese, the natural numbers were evidently 1, 2, 3, ... .]
36
[Moese’s term was pas x.]
37
[The subscript in Vw is not a reference to the polygon in question—in this case W —but rather the first
letter of the Polish word wielokt—polygon.]
146 7 Degree of Equivalence of Polygons

II
III
I
IV

x V
$
AU
O

x
A
I x
II x
III x
IV

Figure 2

[Hemisphere over a disk with center O and radius r]

Proof. Let us construct a hemisphere on the given disk (see figure 2).38 Point O is the
common center of the disk and hemisphere; the radii of the disk and hemisphere equal r.
It is possible to establish the following correspondence between points belonging to the
given disk and those lying on the surface of the hemisphere: let us draw from an arbitrary
point of the disk A a line perpendicular to the plane of the disk; this line will intersect
the surface of the hemisphere at point AU, and we call points A and AU corresponding
points.
Each point of the disk corresponds to just one point lying on the surface of the hemi-
sphere, and vice versa. Points of the disk lying in the interior of an x-strip will correspond
to points of the surface of the hemisphere lying in the interior of the spherical x-strip.
However many x-strips are used to cover the disk, that many spherical x-strips cover the
corresponding surface of the hemisphere.
In figure 2, the disk was covered by four x-strips ( x = r 2 ) running parallel, adjacent
to one another and not overlapping each other, just as the four spherical x-strips ( x =
r 2 ) running parallel, adjacent to one another but not overlapping each other, cover the
surface of the hemisphere. The x-strips V and II have common points interior to the
disk; similarly, the spherical x-strips V and II have common points. These common
parts are hatched in figure 2.

38
[Moese’s verb was opiszmy—let us describe. His term koo—circle—has been translated here as disk.
Figure 2 has been slightly simplified.]
7.3 Moese’s Contribution 147

Now, instead of counting the least number of x-strips necessary to cover the whole
disk, we will do the analogous task for the surface of the hemisphere. In both cases, on
the basis of the correspondence established above, the result of the calculation must be
the same. The solution of this problem for the hemisphere, however, is much easier than
that for the disk.
The area of a spherical x-strip (on the hemisphere) is Sr x. For a given hemisphere,
this area depends solely on the altitude of the spherical x-strip, and not on the location
of the strip on the hemisphere. Therefore, all spherical x-strips of the hemisphere have
the same area. It follows from this that the least number of spherical x-strips that cover
the entire hemisphere cannot exceed
ª 2 Sr 2 º ª 2r º 39
« Srx » —that is, « x ».
« » « »
Nor can it be less than M 2 r / xN . When 2 r / x = n (where n is an integer and positive),
the least number of spherical x-strips that can cover the entire hemisphere must be n.
Moreover, these strips cannot overlap each other, and thus they must run parallel to each
other and adjoin each other. [On the other hand] if 2 r / x is not an integer, then in
covering the hemisphere with spherical x- strips, we have more freedom; the strips will
overlap each other, but fewer than M 2 r / xN strips are not enough to cover the hemisphere
completely.
Therefore, from the correspondence just established, it follows that
ª 2r º
Vw ( x) = « » Q. E. D.
«x»

For the function Vw ( x) let us adopt the following notation:

Vk w ( x) for a square, Vk ( x) for a disk.40
Let us suppose that the edge of the square and the diameter of the disk are equal to a
(and thus the radius of the disk is r = a 2.

Lemma II. Vk w ( x) = Vk ( x) = M a / x N when 0 < x  a.

Proof. From lemma I it follows that to cover a disk with the least number of x-strips
it suffices to arrange these strips parallel to each other so that [they are] adjacent to each
other, starting from the circumference of the disk. (See figure 3.) We arrange the x-strips
parallel to the side a of the square. Then it is apparent that the same least number of
strips that covers the whole disk covers the square as well; and thus it must be [the case
that] Vk w ( x) = Vk ( x), [ Q. E. D.]

39
[This part of Moese’s original paper was unclear, probably an editorial fault. The area of the hemisphere
is 2 S r 2. Covering the hemisphere with n = M 2r/ xN parallel adjoining nonoverlapping strips confirms the
displayed claim. The displayed material originally included the first sentence of the next paragraph.]
40
[These subscripts are the first letters of the Polish words kwadrat—square—and koo—circle.]
148 7 Degree of Equivalence of Polygons

a x

Figure 3

Lemma III. W( x)  Vk w ( x # a) when 0 < x  1.

Proof. It is clear that [when] 0 < x  1, each of the W( x) respectively congruent
polygons into which a square with side a and an equivalent rectangle with sides x # a
and a x may be subdivided must be wholly covered by an ( x # a)-strip. It follows from
this that it cannot be [the case that] W( x) < Vk w ( x # a), Q. E. D.

Let us return now to the proof of theorem 1. Let us suppose that 1 x = n, where n
is a positive integer; then on the basis of the lemmas, it must be [the case] that W(n)  n.
Since on the other hand we have W(n)  n, therefore at last,
W(n) = n Q. E. D.

Theorem 2. W( x)  n + 2 when n < x  n + 1: in other words,

W( x)  1 + M xN when x  1.
This theorem strengthens the upper bound of the function W( x) given in theorem III of
the cited article Tarski [1931] 2014a.41
The proof follows directly from figure 4, drawn for M xN = 4.

§ n 1· § n ·
Theorem 3. W ¨ ¸ = W¨ ¸ = 2 for every natural number n.
© n ¹ © n 1¹
Again, sufficient proof of this theorem is given by figure 5, where n = 4.

41
[Translated in section 7.2.]
 I

II

III IV

I II III V
IV

Figure 4

I I

II

II

Figure 5

Figure 6
150 7 Degree of Equivalence of Polygons

§ 1·
Theorem 4. W¨n  ¸  n + 1, where n, k are any positive integers.
© k¹
This theorem is a generalization of theorem 3.
[Proof.] The truth of this theorem is made clear by figure 6, drawn for n = 3, k = 4.
Figure 6 shows that W(3 1 4 )  4.

And [figure 7 illustrates] yet another trick: polygon II should be understood in the
broader sense (see the remark in Tarski’s [1931] 2014a article42 ). Figure 7 yields
W(1 2 5 )  3. More such tricks could be given; they all become insignificant, however,
insofar as they do not give a stronger bound for the function W(x) than theorem 2. For
x = 1 2 5 , theorem 1 gives the same bound as figure 7.

III

I
II

II

III I

Figure 7

The exercises suggested by Tarski [in connection with his theorem III] now would have
the following solutions:
W(1 1 3 ) = 2, W(2 1 4 )  3, W ( 10 )  5, and moreover, W ( 15 )  5.

In view of the above, I dare to mention the following probable theorem—

W( x) = n + 2 when n< xn+1
—with the following exception:

42
[Translated in section 7.2.]
7.4 Remarks on the Degree of Equivalence (1932) 151

§ 1·
W¨n  ¸ = n + 1,
© k¹
where n and k are any natural numbers. The question of proof remains open.43 Perhaps
some Reader might prove the above theorem for n = 1, or even find a trick, like the
previous ones, that would disprove the theorem in question.

For n = 2,44 the following theorem seems to me very likely:

In case that the square can be divided into just two polygons, so that these
polygons in a different arrangement can form an equivalent rectangle, both of
these parts of the subdivision of the square must be right-angled polygons
(interior angles 90( or 270() congruent to each other.

7.4 Remarks on the Degree of Equivalence (1932)

This section contains an English translation of Alfred Tarski’s paper Uwagi o stopniu
równowanoci wieloktów, [1932] 2014d, which appeared in volume 2 of the journal
Parametr.45 The journal’s target audience was gimnazjum teachers and their most serious
students. The paper continued the discussion begun in Tarski [1931] 2014a and Moese
[1932] 2014, translated in sections 7.2 and 7.3. A draft English translation by Izaak
Wirszup was published informally by the University of Chicago as Tarski and Moese
1952.46 The present translation was carried out independently, but used Wirszup’s for
confirmation.

The translation is meant to be as faithful as possible to Tarski’s original text. Its only
intentional modernizations are punctuation, and some changes in symbols, where Tarski’s
conflict with others used throughout this book. Some uses of alternative type styles for
emphasis, enunciations, and personal names have been modified. Bibliographic references
and personal names have been adjusted to conform with conventions of the present book.
All [square] brackets in the translation enclose editorial comments inserted for clarifica-
tion, often as footnotes. As an aspect of adjusting punctuation, the editors increased white
space to enhance visual organization.

43
[In his [1932] 2014d paper, translated in section 7.4, Tarski claimed that the proof for the case n = 1
was known, but too involved to include there, and that all other cases remained open. See the discussion
at the end of 7.1.]
44
[Moese probably meant n = 1 here.]
45
For further information about Parametr, consult sections 7.1 and 9.7.
46
The original paper and that draft were reprinted in Tarski’s 1986a Collected Papers, volume 1, 595–602.
Some serious typographical errors were introduced in those editions.
DR. ALFRED TARSKI (Warsaw)

Remarks on the
Degree of Equivalence of Polygons

In an article published in Mody matematyk,47 I posed several problems concerning the

degree of equivalence of polygons. Evidently, the article was written with a “lucky hand”:
the topic that I touched upon aroused the interest of several mathematicians. Thanks
to their investigations, various hypotheses that I suggested were either established or dis-
proved. And as to the main problem proposed in the cited article, about a complete
investigation of the function W( x), at the present moment not much is still lacking for
a definitive solution.

In the following remarks, I wish to set down all the results obtained up to the present
moment concerning the function W( x), referring to the preceding article by Henryk Moese
in this issue of Parametr and to some investigations by Adolf Lindenbaum and Zenon
Waraszkiewicz 48 with which I am acquainted but which have not yet been published; and
in addition, to extract from them certain facts of a more general nature. Moreover, I
intend to propose a few further problems in the same field.

Recall that we call the degree of equivalence of two equivalent polygons W and V,
symbolically V(W, V ), the least natural number n satisfying the following condition:
polygons W and V can [each] be divided into n respectively congruent polygons. In
particular, the degree of equivalence of a square with edge a and a rectangle with edges
x # a and 1 x # a is denoted by the symbol W( x).

We shall need the concept of width for a given geometric figure. As is well known, that
part of a plane bounded by two parallel lines carries the name strip; the distance between
these lines we call the width of the strip. The width of the narrowest strip covering a
plane figure F we shall call the width of figure F, and we shall denote it by the symbol
Z(F ).49 It is not hard to show that every figure F that can be covered by any strip at all
possesses a definite width. For example, the width of a rectangle is equal to the length
of the smaller of its sides; the width of an equilateral triangle is its altitude; and of a
circle, its diameter. It is also obvious that if figure F is part of figure G, then Z(F ) 
Z(G); if figures F and G are congruent, then Z(F ) = Z(G).

47
Tarski [1931] 2014a [translated in section 7.2].
48
[For more information on Moese and Waraszkiewicz and on Lindenbaum see the boxes on pages 130 and
132 and in section14.3. With their names Tarski used the honorifics pan, pan magister, and pan doktor,
respectively. Pan means Mr.; magister indicated that Waraszkiewicz had a master’s degree but no doctor-
ate yet.]
49
[Tarski used S for width; this translation uses Z to avoid confusion with the numerical constant S.]
7.4 Remarks on the Degree of Equivalence (1932) 153

It is possible to justify the following theorem:

A. If a figure F contains in itself, as a part, a disk with diameter equal to

the width of the figure (for example, if figure F is a disk or a parallelo-
gram) and if, moreover, we subdivide this figure into any n parts
C1 , C2 , ..., Cn , then
Z(F )  Z(C1 ) + Z(C2 ) + $$$ + Z(Cn ).
Proof. For the case in which figure F is a disk, the proof is an almost word-for-word
repetition of the proof of lemma I in the cited article by Moese.
For the general case, we argue in this way. In figure F inscribe a disk K of the same
width. Some of the parts C1 , C2 , ..., Cn of figure F must have points in common with the
disk K: for simplicity, let these be C1 , C2 , ..., Cp , where p  n. Further, for k =
l, 2, ... , p, let CkU be the common part of figure Ck and the disk K. It is easy to see that
the disk K was subdivided into parts C1U, C2U, ..., CpU; therefore, as we already know,
Z(K )  Z(C1U) + Z(C2U) + $$$ + Z(CnU).
On the other hand, Z(CkU)  Z(Ck ) for k = 1, 2, ... , p; therefore,
Z(K )  Z(C1 ) + Z(C2 ) + $$$ + Z(Cp )
 Z(C1 ) + Z(C2 ) + $$$ + Z(Cn ), Q. E. D.

The proof that all parallelograms, in particular, satisfy the assumptions of the above
theorem does not present the Reader any difficulties.50

From theorem A we derive an important consequence concerning the degree of

equivalence:

B. If W is a polygon containing in itself a disk as a part, with diameter

equal to the width of the polygon (for example, if W is a parallelogram),
and V [is] any polygon equivalent to it, then
Z(W )
V(W, V )  .
Z(V )
Proof. According to the definition of degree of equivalence, polygons W and V can
be subdivided into V(W, V ) = n respectively congruent polygons: W into W1 , W2 , ...,
Wn , and V into V1 , V2 , ..., Vn . According to theorem A,
Z(W )  Z(W1 ) + Z(W2 ) + $$$ + Z(Wn ).
Moreover, obvious relations hold:

50
[The proof requires a rather tedious school-geometry argument. Tarski proposed it as part of exercise
4 at the end of this paper.]
154 7 Degree of Equivalence of Polygons

Z(W1 ) = Z(V1 )  Z(V ), Z(W2 ) = Z(V2 )  Z(V ),

... , Z(W n ) = Z(Vn )  Z(V ).
Therefore, Z(W )  Z(V ) # n = Z(V ) # V(W, V ), and thus

Z(W )
V(W, V )  , Q. E. D.
Z(V )

It is worthwhile to compare the theorem proved a moment ago with theorem 5 in my

previous article.51 Both of these theorems establish certain lower bounds for the degree
of equivalence:
G(W ) Z(W )
V(W, V )  and V(W, V )  .
G(V ) Z(V )
In many instances the second inequality gives a much better result. For example, if W
is a rectangle with edges 16 and 12, and V, a rectangle with edges 192 and 1, then from
theorem 5 we obtain52

1922  12
V(V, W )  = 92.1625 ,
162  122
and thus V(W, V )  10; however, theorem B gives V(W, V )  12. On the other hand, one
ought to notice that theorem 5 has a significantly wider scope of applicability than
theorem B: the latter, we can at the present time prove only for certain special polygons
(not even for triangles), while in theorem 5 the assumption that W be a convex or even
connected polygon is sufficient. For this purpose it would be worthwhile to return to
theorem A: should it be possible to extend that theorem to arbitrary convex figures, we
would immediately obtain the desired strengthening of theorem B.

In theorem V of my article,53 I gave a certain lower bound for the function W( x); thanks
to theorem B this bound can be considerably strengthened:

C. W( x)  M xN for every number x  1.

Proof. Let W be a square with edge a, while V [should be] a rectangle with edges
x # a and 1 x # a , [where] x  l. Then V(W, V ) = W ( x), Z(W ) = a, and Z(V ) =
1 x # a ; and thus, according to theorem B, W( x)  a / ( 1 x # a ) = x. Moreover, since W( x)
is a whole number, and the symbol M xN denotes the least whole number  x, then
W( x)  M xN , Q. E. D.

51
[Tarski [1931] 2014a, translated in section 7.2. There, that result is refered to as property 5.]
52
[In the original, the first term of the following inequality was misprinted as W(V, W ).]
53
[Theorem V is different from theorem (or property) 5.]
7.4 Remarks on the Degree of Equivalence (1932) 155

An upper bound that I had established in theorem III [of that article] has also been
strengthened. This is theorem 2 in Moese’s article:54

D. W(x)  1 + M xN for every number x  1.

Recalling that W( x) is a whole number, we derive immediately from theorems C and

D the following conclusion:

E. For every number x  1, either W( x) = M xN or W( x) = 1 + M xN.

In view of the above theorem, for every value x  1 we are able to establish the value
of W( x) within accuracy 1.55 It remains only to investigate for which values x the
function W( x) takes on each of its two possible values. A partial answer to this question
we find in theorem

F. If x is a number of the form x = n + 1 p where n and p are natural

numbers  1, then W( x) = M xN = n + 1.

Theorem F disproves one of my conjectures, according to which W( x) should be  3

for all values x except 1 2 , 1, and 2. On the other hand, as a particular case of this
theorem, a confirmation of another hypothesis, also suggested by me, can be obtained:

G. W(n) = n for every natural number n.

It seems probable that numbers x of the form n + 1 p are the only numbers satisfying
the condition W(n) = M xN . At this point, we are able to prove this conjecture only for the
case x  2:

H. If 1 < x  2, then in order that W( x) = M xN , it is necessary and sufficient

that x be of the form x = 1 + 1 p , where p is any natural number.

I shall not include the proof here. It is somewhat complicated and requires some subtle
methods of reasoning. From the proof of this follows, among other things, that in the case
in which a square can be subdivided into two parts from which a rectangle can be con-
structed, the only possible method of subdivision is that described by Moese in the proof
of theorem 3 in his article.56

If it should be possible to remove the condition x  2 from the assumptions of the

above theorem, the problem of the function W( x) would finally be solved. However,

54
[Moese [1932] 2014, translated in section 2.5 of the present book. Tarski referred here to a figure 1, but
that seems incorrect.]
55
[In the original this last phrase was z dokadnoci do 1.]
56
[This conjecture was proposed formally at the end of Moese’s article. Tarski referred here to a figure 2,
but that seems incorrect. For more about theorem H and its consequences see section 8.5.]
156 7 Degree of Equivalence of Polygons

reflecting on the challenges thus far and on the proof of theorem H [suggests that] this
problem does not belong among the completely easy ones.

I must now devote a few words to the question of “author’s priority” with respect to
the results presented here. The question is somewhat complicated, as is usual in these
situations, when one and the same group of problems interests several people simulta-
neously, and in addition some of them collaborate with each other. Entering into play
were a few Warsaw mathematicians; Moese investigated these problems completely
independently. The earliest result chronologically is theorem G. Waraszkiewicz proved
it first; the same result was obtained later, although completely independently, by Moese.
The ideas in both proofs are fairly closely related. Waraszkiewicz’s reasoning is somewhat
more complicated, and for this reason it will not appear in Parametr. By analyzing the
proofs of theorem G, I came to the more general theorems A, B, and C. (By the way,
theorem C, even if not clearly formulated by Moese, is contained in his lemmas II and III.)
Theorems D and F come from Moese: F in a somewhat weaker formulation with the
symbol  instead of = in the conclusion. However, they were obtained independently
in Warsaw as well: Lindenbaum proved theorem F, and D [was proved by] Bronisaw
Knaster and the author of these remarks. The results obtained bring to light the last
hypothesis concerning the course of values of the function W( x), which I wrote about
here. A certain special case of this hypothesis was confirmed by Lindenbaum and Warasz-
kiewicz, to whom we owe theorem H.

As Lindenbaum noticed, all the results so far obtained for the function W( x) still hold
when, in the definition of this function, the square is replaced by any rectangle.

In conclusion, I propose a few more problems, closely connected with the concept of
degree of equivalence.

Besides the function W( x), several analogous functions may be defined, whose courses
of values are not yet exactly known. For example, two parallelograms with equal bases
and altitudes present one of the simplest examples of equivalent polygons. The degree
of equivalence of such parallelograms depends, among other things, upon the magnitude
of their angles. Let us suppose for simplicity that one parallelogram is a square and the
second has an acute angle M and denote the degree of their equivalence by the symbol
T(M). On the one hand using theorem B and on the other hand by analyzing the proof
of the theorem on the equivalence of parallelograms given in elementary geometry
textbooks,57 it can be shown that
Mcsc MN  T(M)  1 + Mcot MN .
It is also not hard to prove that in some cases both of the indicated bounds for the function
T(M) coincide, and the value of the function is thus uniquely determined; in others,
however, these bounds differ by 1. One is then concerned with establishing the value
of the function T(M) in these situations, in which Mcsc MN < 1 + Mcot MN .

57
See Wojtowicz [1919] 1926, §177, [and page 128 in section 7.1 of the present book].
7.4 Remarks on the Degree of Equivalence (1932) 157

The functions W( x) and T(M) remain in a close relationship to a certain more general
concept, namely the degree of irregularity of a polygon. If we agree to consider a square
as the most “regular” polygon, then the degree of equivalence of a polygon W and a
square equivalent to it might be called the degree of irregularity of this polygon, symbol-
ically U(W ): the greater U(W ), the less regular the polygon W. If in particular, W is
a rectangle with sides a and b, where a  b, then

U(W ) = W ab .

If W is a parallelogram with base and altitude equal to a and an acute angle M, then
U(W ) = T(M). It would be an interesting thing to know the general properties of the new
concept.

Finally, there are many problems concerning the number of methods for [achieving]
the best subdivision of equivalent polygons. Let W and V be equivalent polygons and
let V(W, V ) = n. The question is, how many methods can be used to subdivide these poly-
gons into n respectively congruent polygons? (We regard two subdivision methods as
distinct if the subdivision nets are not congruent to each other.) We know situations in
which there is only one method of subdivision. For example, as I mentioned in connection
with theorem G, this happens when W is a square with side 1 and V [is] a rectangle
with sides ( p +1) / p and p / ( p +1) , where p is any natural number. Again, in other cases
there are many methods of subdivision. For example, if W is a square with edge 1 and
V [is] a rectangle with edges 7 5 and 5 7 , then we know at the present time three
different methods. Two of those are mentioned in the article by Moese,58 and I mentioned
a third in my previous article.59 It would be an interesting thing to establish which
methods of subdivision are possible in this or other special cases.

Exercises

1. What is the width Z(W ) and the diameter G(W ) in the case that polygon W is
(1) a triangle, (2) a trapezoid, (3) a kite? 60

2. Determine the width Z(F ) of a figure F that is a sector of a disk with radius r
and central angle M. Investigate the change in Z(F ) with respect to M and sketch
its graph (0(  M  360() .

3. [Suppose] A, B, and C are the three vertices of an equilateral triangle with edge
a. With point A as center, draw the arc BC that is less than a semicircle, and do
the same with points B and C. Let W be the “circular polygon” bounded by these

58
Remarks in connection with figure 5 in Moese [1932] 2014 [section 7.3, page 149].
59
Tarski [1931] 2010a, figure 2 with its dimensions altered [section 7.2, page 136].
60
[A kite is a quadrilateral with two pairs of adjacent congruent edges; Tarski’s term for it was deltoid.]
158 7 Degree of Equivalence of Polygons

three arcs. Calculate the width Z(W ) and diameter G(W ). Repeat the exercise
with any n-sided regular polygon, where n is [an arbitrary] odd natural number.

4. Show that every parallelogram R contains within itself, as a part, a disk with diam-
eter Z(R). Check whether this theorem is [also] valid for a kite, a triangle, and a
trapezoid.

5. Show that the area s(W ) of any polygon W satisfies the inequality
s(W ) < G(W ) # Z(W ) .

6. If W is a convex figure contained in the plane then

s(W )  1 2 # G(W ) # Z(W )
( s —area of the figure, G —diameter, Z —width) .
Moreover, in the above formula we will have equality if and only if W is a triangle.61

7. [Suppose] W is a square with edge a, while V [is] a rectangle with edges 21 16 and
16 21 .
Give at least three distinct methods for subdividing these two polygons into
three respectively congruent parts. Investigate all possible methods for such a
subdivision.

8. [Suppose] W is a square with edge a, while V [is] a parallelogram with base and
altitude equal to a and an acute angle M. Show that
M csc MN  V(W, V )  1 + M cot MN .
Show that the numbers M csc MN and 1 + M cot MN differ by at most 1, and investi-
gate for which values of M they are equal.

9. [Suppose] W is a parallelogram with edges 100 and 87 and acute angle 88(,
and V [is] an equivalent parallelogram with base 100 and acute angle 10(.
Find V(W, V ) .

61
Exercise 6 was suggested by A. L. from Warsaw [probably Lindenbaum].
 8
Research Threads

Previous chapters have included translations of research papers published by Alfred

Tarski and others in 1921, 1924, and 1931–1932. They were accompanied by extensive
background discussions of their topics and of Tarski himself and others involved in his
career. The present chapter will give the reader a sense of the impact of that research,
and it will provide references to closely related subsequent developments by Tarski and
by others who continued his investigations.

The emphasis is on work in Poland that has not been widely discussed elsewhere. This
chapter mentions only briefly some of Tarski’s research in logic, which brought him great
fame but is not emphasized in this book because it is already readily accessible. That
discussion is connected with some short translations included in the supplementary
chapter 15.

In 1999 Steven R. Givant published Unifying Threads in Alfred Tarski’s Work, a

wonderful biographical study of Tarski based on their collaboration and personal associa-
tion during 1973–1983. Givant showed how Tarski, throughout his career, would return
again and again to various research problem areas, each time advancing them, producing
longer and longer threads of results. Some of these threads are composed of very tightly
intertwined filaments. That pattern provides an organization for the study of Tarski’s
research. Some of the threads continued long after Tarski relocated to the United States
in 1939. Others started anew there, sometimes splitting off and becoming independent
from threads started earlier in Poland.

The sections of this chapter will follow threads and filaments of Tarski’s research that
began during 1920–1931:
Two Threads in Logic 1920–
Thread: Well-Ordering and Finiteness 1921–
Thread: Cardinal Arithmetic and the Axiom of Choice 1924–
Thread: Application of Set Theory to Geometry (with four filaments)
Equidecomposability in Elementary Geometry 1931–1932
Equidecomposability in Set Theory 1924–
The Measure Problem 1930–
Generalizing Cantor–Bernstein 1927–

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 159
DOI 10.1007/978-1-4939-1474-6_8, © Springer Science+Business Media New York 2014
160 8 Research Threads

8.1 Two Threads in Logic

One thread of Alfred Tarski’s research in logic began with his 1923 doctoral dissertation,
supervised by Stanisaw Leniewski. Tarski showed that material equivalence and
universal quantification over propositional variables could constitute the primitive notions
of Leniewski’s system of logic. It is described in more detail in chapter 3, page 39.
Leniewski acknowledged the importance of that work and of Tarski’s less formal
contributions, but Tarski did not pursue them further. For information on developments
influenced by this thread of Tarski’s research, readers should consult works on
Leniewski’s logic.

Partly in collaboration with his former teacher Jan ukasiewicz, Tarski developed a
second thread, a framework for research in logic and for its presentation, which has since
become a standard in the mathematical world. Already in 1920, as a Warsaw student,
Tarski presented a talk that included the deduction theorem, which would be fundamental
for that framework. (See chapter 3, page 37.) Two small collaborations of ukasiewicz
and Tarski are described and translated in sections 15.1 and 15.4: Tarski’s discussion
of ukasiewicz’s 1925 paper On a Certain Way of Understanding the Theory of Deduction,
and his contribution to ukasiewicz’s 1928–1929 paper On Definitions in the Theory of
Deduction.1 As a university dozent, Tarski worked closely with ukasiewicz’s research
seminar, and in 1929 officially became ukasiewicz’s assistant, supervising student
research. From this activity stemmed several publications on the fundamental ideas of
logic, notably Tarski’s [1930] 1983b paper On Some Fundamental Concepts of Metamathe-
matics. Tarski developed his celebrated definition of true sentence as part of this program.
His first publication on that subject, On the Concept of Truth in Reference to Formalized
Deductive Sciences ([1930–1931] 2014), is discussed and translated in section 15.6.2
Several key ideas in Tarski’s approach stemmed also from his association with Leniewski.

During the 1930s Tarski published expanded versions of those two works, which
became the core of his framework for logic. They have become particularly well known
through their translations in Tarski’s [1956] 1983 book Logic, Semantics, Metamathe-
matics: Papers from 1923 to 1938. A great expansion of research in model theory and
formal semantics after World War II was based in large part on that framework. Chapter
18 of the present book lists surveys of Tarski’s work and influence in these areas.

Occasionally Tarski published comments on subjects in philosophical logic. Sections

15.2, 15.3, 15.8, and 15.10 contain discussions and translations of his 1927 comments on
the theories of action and causality, and his 1936 comments on idealism and metaphysics.

1
Tarski [1925] 2014 and ukasiewicz 1925; ukasiewicz [1928–1929] 2014.
2
A smaller contribution on this subject, Tarski’s commentary on the paper Kokoszyska 1936a, is translated
and discussed in section 15.11.
8.3 Thread: Cardinal Arithmetic and the Axiom of Choice 161

8.2 Thread: Well-Ordering and Finiteness

Another thread, on the notions of well-ordering and finiteness, also began while Alfred
Tarski was a student. His 1921 [2014] paper A Contribution to the Axiomatics of Well-
Ordered Sets, translated in chapter 2, stemmed from the seminar of Stanisaw Leniewski,
but Wacaw Sierpiski seems to have had greater influence on this thread of Tarski’s
work. Tarski’s 1924c abstract Sur les principes de l’arithmétique des nombres ordinaux
(transfinis) summarized a presentation to the Warsaw section of the Polish Mathematical
Society on axioms for the ordinal number system; it has a similar flavor. So does Tarski’s
major 1924b paper Sur les ensembles finis, which is concerned with various possible defini-
tions of the notion of finiteness and the relationship between their equivalence and the
axiom of choice. This highly readable paper, Tarski’s Habilitationsschrift, earned for him
the venia legendi—qualification to teach at a university.3

During the late 1930s, Tarski and his student Andrzej Mostowski returned to develop
this thread much further, within the framework that Tarski had developed for logical
research. World War II and the Cold War prolongued the delay. In 1949 Mostowski
and Tarski published an abstract about their work, Arithmetical Classes and Types
of Well-Ordered Systems. A major study by John E. Doner, Mostowski, and Tarski
finally appeared in 1978, after Mostowski’s death: The Elementary Theory of Well-
Ordering—A Metamathematical Study. Doner had earned the doctorate in 1968 under
Tarski’s supervision in Berkeley.

8.3 Thread: Cardinal Arithmetic and the Axiom of Choice

As noted in chapter 3, the axiom of choice was a major interest of Alfred Tarski’s teacher
Wacaw Sierpiski, and it came to play a major role in some of Tarski’s own research
threads. During the mid-1920s, Tarski published three papers on theorems related to
cardinal arithmetic and to the axiom:
• Sur quelques théorèmes qui équivalent à l’axiome du choix
• Quelques théorèmes sur les alephs
• Communication sur les recherches de la théorie des ensembles4
The last was a collaboration with Adolf Lindenbaum. The publications in this thread are
accessible, and thus beyond the scope of the present book. For further information on
them, consult the 1988 survey by Azriel Lévy.

3
Jadacki 2003a, 144–145.
4
Tarski 1924e, Tarski 1925b, and Lindenbaum and Tarski 1926.
162 8 Research Threads

8.4 Thread: Application of Set Theory to Geometry

This thread would connect Alfred Tarski’s research in set theory and logic to his interest
in school mathematics. It includes his [1924] 2014b paper On the Equivalence of Polygons
and the famous [1924] 2014 Banach–Tarski paper, which are described in sections 4.3
and 4.4, and translated in full in chapters 5 and 6. This thread also includes the more
elementary articles on the degree of equivalence of polygons, featured in chapter 7.

Another thread of Tarski’s geometric studies, axiomatics, began around 1927 and
continued into the 1950s.5 Readily accessible now, that work is not covered in this book.
For information about that thread, see the surveys by Lesaw W. Szczerba (1986) and by
Tarski and Steven R. Givant (1999).

The following sections discuss four filaments of this applied-set-theory thread:

• equidecomposability in elementary geometry (1931–1932)
• equidecomposability in set theory (1924–)
• the measure problem (1930– )
• generalizing the Cantor–Bernstein theorem (1927– )

8.5 Equidecomposability in Elementary Geometry

This elementary-geometry filament probably took root during the mid-1920s, when Alfred
Tarski began teaching the subject. But it surfaced only later, with his [1931] 2014a and
[1932] 2014d papers on the degree of equivalence of polygonal regions P and Q with
the same area: the smallest integer n such that P and Q can each be subdivided into
n polygonal subregions with disjoint interiors such that pairs of corresponding subregions
are congruent. A problem left open in the first of these papers was solved by Henryk
Moese in the paper [1932] 2014, and that work was discussed and utilized in Tarski’s
second paper. In a [1937] 1938 oral presentation, Adolf Lindenbaum proved the final
theorem H of Tarski’s second paper, which had been attributed to him and Zenon
Waraszkiewicz; but he did not publish details. Tarski’s first paper on this filament was
reprinted in 1975 and translated and published in Hebrew in Tarski [1931] 1951. Draft
English translations of all three of these Polish articles constituted the obscure publica-
tion Tarski and Moese 1952. Beyond that, these works have had little impact.6 New
translations and a discussion of all three papers are included in chapter 7 and section 9.8,
with the hope of bringing them to light.

5
The geometric-axiomatics thread may have begun as early as 1924: see the first paragraph of Tarski [1924]
2014b, §1, translated in section 5.1.
6
The present editors have found only one reference by another author to this material: Howard W. Eves
(1963–1965, volume 1, chapter 5) mentioned inequality III of Tarski [1931] 2014a, crediting Tarski without
citing any specific source or mentioning the strengthened version of the inequality in Tarski [1932] 2014d.
Eves may have learned about inequality III orally, for he also credited Walter B. Carver and did not men-
tion that the inequality was incorrectly typeset in the draft translation Tarski and Moese 1952. The pres-
ent editors have not been able to confirm Carver’s role.
8.6 Equidecomposability in Set Theory 163

Tarski himself continued this research to some extent. For example, some open
conjectures about the degree of equivalence can be stated in the form, if n is a natural
number, then )(n), where )(n) is a formula with a variable n, which would become
a sentence of the elementary algebra of real numbers if n were replaced by a constant
representing a specific natural number. In the report A Decision Method for Elementary
Algebra and Geometry, Tarski suggested that applying his method serially to many cases
)(1), )(2), ... would either refute such a conjecture or provide considerable intuitive
evidence for its validity.7

Around 1970 one of the present editors, James T. Smith, more than once heard Tarski
lecture on this subject. A principal object of consideration was the function W( x), the
degree of equivalence of the unit square Q and an x by 1 x rectangle P. Tarski also
considered the analogous number Wm( x) that results by limiting the subregions to
polygons with at most m edges. Clearly, Wm( x)  W( x). Applying his own 1930b meta-
mathematical result about the undefinability of the concept of natural number within
elementary real arithmetic, Tarski showed that for each m there must exist x such that
Wm( x) > W( x): there is no uniform bound on the complexity of the polygons required to
decompose P and Q each into a minimal number of subregions with correponding com-
ponents congruent. His argument made essential use of the final theorem H of his [1932]
2014d paper.8

In 2011 Tom M. Apostol and Mamikon A. Mnatsakanian considered the extended

problem of equidecomposability of polygonal regions with equal areas and perimeters,
requiring boundary pieces to correspond as well as interior pieces. Their methodology
does not overlap Tarski’s.

8.6 Equidecomposability in Set Theory

Alfred Tarski’s [1924] 2014b paper On the Equivalence of Polygons, and the famous 1924
paper On Decomposition of Point Sets into Respectively Congruent Parts by Stefan Banach
and Tarski, began a filament of Tarski’s research that tied together his interests in the
axiom of choice and in elementary geometry. These works are concerned with a notion
of decomposition different from the elementary one: decomposition of arbitrary point sets
into arbitrary subregions, which are required to be entirely disjoint.

These papers have had enormous impact. During the next half century, Tarski and
others returned repeatedly to questions raised there, developing whole new areas of
mathematics. The intricate underlying constructions by Felix Hausdorff, the intricate
reasoning employed by Banach and Tarski, and their startling conclusions have posed

7
Tarski [1948] 1957, 4–5.
8
Tarski [1930] 1983b, 134; see also [1948] 1957, 61. Theorem H is discussed in section 7.1. Recently, J.
Shilleto (2012) noted that if Theorem H were stated to include more of what Tarski’s suggested proof
actually entailed, then it would imply Tarski’s result about Wm directly, without appealing to metamathe-
matical results.
164 8 Research Threads

attractive challenges for expository writers—certainly for the editors of the present book.
A notable early example was Karl Menger’s 1934 lecture to a lay audience in Vienna. In
the same vein, but much more detailed, is the book Wapner 2005. Sally and Sally 2007
is a sophisticated text aimed at students of various levels; its chapter 5 is an excellent
introduction to this problem area. Stromberg 1979 is an elementary but extremely
detailed mathematical account of the “paradoxical” Banach–Tarski theorem. For a more
advanced approach, and to survey the large literature stemming from the Hausdorff and
Banach–Tarski results, the best source is the monograph Wagon 1993. Other surveys,
emphasizing Hausdorff’s legacy, include Schreiber 1996 and Czy 1994, chapter 1.

The 1990 solution by Miklós Laczkowicz of Tarski’s circle-squaring problem, more than
six decades after its statement in Tarski [1924] 2014b, has been discussed already at the
end of section 4.3. Trevor Wilson (2005) discovered a notable strengthening of Laczko-
wicz’s result and the Banach–Tarski paradox.

The Banach–Tarski paper sparked immediate questions about the generality of its
results and the most effective ways to present them. Such questions, in turn, led to new
research areas. Some further results are included in section 5 of Tarski’s 1926 joint paper
with Adolf Lindenbaum, already mentioned in section 8.3 in connection with Tarski’s set-
theoretic research. Others contributed too, particularly Johann (John) von Neumann.
In 1929, Neumann recast the measure-theoretic arguments of Hausdorff and Banach to
emphasize the role of the group of transformations under which the measures should be
invariant. For any point set E, any I I E, and any group G of transformations of E,
he defined an [E, I, G ]-measure to be an additive function P from the power set of E
to the set of all nonnegative real numbers, invariant under all transformations in G , such
that P(I) = 1. He showed that an [ E, I, G ]-measure exists if and only if a [ G, G, G ]-
measure exists. This result allows measure-theoretic questions to be shifted from the
geometric context to the group-theoretic; sometimes that is a simplification. Today, a
group G is called amenable when such a measure exists.9 The theory of amenable groups
underlies all subsequent work in this area. The first milestone in that theory was stated
in Tarski’s 1929–1930 abstract on additive functions:
Adapting to the notions introduced by J. v. Neumann in his recent work ... this result can be
generalized in the following way: ... in order for an [ E, I, G ]-measure to exist it is necessary
and sufficient that I not be equivalent to its half under any finite decomposition relative to
the group G .

By its half Neumann had meant any subset H I I that is equivalent to its complement,
I – H. Thus, “paradoxical” decompositions of point sets arise when, and only when, the
measure problem fails to have a solution.10 Much later, this result played a pivotal role
between the two main parts of Stan Wagon’s monograph: he claimed, “Tarski’s theorem

9
Neumann 1929, 78–82. Neumann called such groups measurable (messbar). For more information,
consult Wagon 1993, chapter 10, and de la Harpe 2004.
10
Tarski 1929–1930, 117. That preliminary report was submitted to the Warsaw Society of Sciences and
Letters on 2 May 1929 by Sierpiski. There are several apparent errors in statements of definitions and
theorems, perhaps due to mistranslation into French.
8.7 The Measure Problem 165

is not as well known as it deserves to be.” 11 Tarski 1929–1930 contained no proofs; the
proof for this theorem would not appear in any form until nearly a decade had passed.

After relocating to the United States, Tarski maintained interest in questions about
set-theoretic equidecomposability, and he inspired work by others. His [1946–1947] 2014
letters to Wacaw Sierpiski, translated in full in section 15.12, explain some of that
activity. For example, his new Berkeley colleague Raphael M. Robinson studied the
Banach–Tarski decompositions of certain point sets S into k + l disjoint pieces, of which
the first k can be “reassembled” to constitute all of S, as can be the last l pieces. In
1947, Robinson showed that for a solid sphere S, the minimum number k + l of pieces
is five (k = 3 and l = 2), one of which may consist of a single point; for a spherical
surface S, the minimum is four.

8.7 The Measure Problem

The remaining two filaments of Alfred Tarski’s research under consideration here elabor-
ated results of Stefan Banach that underlay their original papers on equidecomposability.
One of these filaments has to do with the measure problem: to assign, to all bounded
subsets of the line, nonnegative numbers called their measures, such that two congruent
sets always have the same measure, the unit interval has measure 1, and the measure
of the union of two disjoint sets is the sum of their individual measures. Analogous
problems can be stated for higher-dimensional point sets. Banach’s 1923 solution of the
measure problem in dimensions one and two was described in detail in section 4.2. Felix
Hausdorff’s 1914 “paradox” had shown that the problem has no solution in higher dimen-
sions. Their work started a new area of mathematics: existence and properties of additive
set functions invariant under various transformation groups. Tarski’s interest in this
area emerged as a series of major works in the 1930s. In his 1929–1930 abstract Sur les
fonctions additives dans les classes abstraites et leur application au problème de la mesure,
Tarski announced that a modified problem, without the requirement of invariance under
congruence, thus for the trivial transformation group, always has a positive solution: for
each set E there is a nonconstant additive function P from the power set of E to the
two-element set {0, 1} such that P({ x}) = 0 for each x  E. Tarski proved this in his
1930a paper Une contribution à la théorie de la mesure by constructing P as the charac-
teristic function of a prime filter in the power set.12 The proof can be regarded as a
streamlining of Banach’s original argument, but to build the filter, Tarski still used well-
ordering and recursion with ordinal numbers, rather than a maximal principle. This line
of inquiry was explored in great detail in the two-part paper Tarski 1939–1945, Ideale
in vollständigen Mengenkörpern.

In the presentation [1932] 2014a, On Geometric Properties of Banach’s Measure, Tarski

outlined a plan for recasting Banach’s work into a geometric framework that would unfold

11
Wagon 1993, 125, 144.
12
Marshall H. Stone (1938, §5) explained how this idea was articulated independently at this time by Tarski
and several others.
166 8 Research Threads

more naturally from the familiar theory of Peano– Jordan content. Previously published
only in Polish, its abstract is translated in section 15.7. It includes no details at all.
Tarski returned to this project again in his 1938b paper, Über das absolute Mass linearer
Punktmengen; unfortunately, even there the exposition was meager.

During the late 1930s, Tarski collected much of his work in this area into publications
that are really extended abstracts.13 Section 4 of his proofless 1937c paper Über additive
und multiplikative Mengenkörper und Mengenfunktionen is a list of results relating
cardinality considerations to various versions of the measure problem. In his 1938a paper
Algebraische Fassung des Massproblems, Tarski presented the results of his 1929–1930
abstract in much more detail, including his important theorem that “paradoxical” decom-
positions of point sets arise when, and only when, the measure problem fails to have a
solution. The 1938a paper is difficult to read, for two reasons. First, its basic notion is
not the algebra of subsets of a metric space, but an abstract version of the algebra of their
equivalence classes under finite equidecomposability. This allowed Tarski to vary more
freely the family of equivalence classes under consideration, and to omit disjointness
provisions from additivity conditions, but it resulted in an uncomfortable setting for some
otherwise familiar concepts. Second, Tarski included only minimal explanations and only
hints of proofs. One hopes that some mathematician will rework with better exposition
the results in this important paper.

Tarski’s 1938b article presented a notion that he had outlined in his [1932] 2014a
abstract. He termed a set X of real numbers absolutely measurable if the supremum of
the lengths of all segments equivalent to subsets of X by finite decomposition should be
equal to the infimum of the lengths of all segments that contain a subset equivalent to
X. He presented many properties of this notion, and comparisons with more-standard
concepts, but with little exposition and only hints of proofs. For example, a set X is
absolutely measurable just in case all measures satisfying the requirements of Banach’s
version of the measure problem assign it the same value. In that case, this value coincides
with the supremum and infimum just mentioned: Tarski called it the absolute measure
of X. Hugo Hadwiger developed these ideas further during the 1950s and made them
the basis of section 3.5.3 of his 1957 monograph Vorlesungen über Inhalt, Oberfläche und
Isoperimetrie.

Tarski returned to this area of inquiry after he became established at Berkeley. The
broad 1948b study Measures in Boolean Algebras, coauthored with his junior colleague
Alfred Horn, generalized and cleaned up some of the steps in Banach’s solution of the
measure problem (see section 4.2 of the present book) and clarified some of the results
in Tarski 1938a. Chapters 13–16 of Tarski’s 1949a book Cardinal Algebras also presented
much material of this sort.14 For further references about the work of others, consult
Wagon 1993.

13
Tarski complained bitterly about his lack of sufficient time during those years, due to his dependence on
two jobs (see section 9.1). Perhaps that explains the lack of exposition in these papers.
14
The abstract Horn and Tarski 1948a covers part of 1948b, but not this part.
8.8 Generalizing Cantor–Bernstein 167

8.8 Generalizing Cantor –Bernstein

In 1924 Stefan Banach published an elaboration of the Cantor–Bernstein theorem that

was used heavily in the Banach–Tarski paper. His formulation (see the box on page 44)
was not particularly graceful nor as useful as it might be. That probably spurred Tarski
to reconsider it several times. In February and December 1927, Tarski and his close
friend Bronisaw Knaster presented three papers on this subject to the Warsaw section
of the Polish Mathematical Society:
• Tarski 1927a Sur quelques propriétés caractéristiques des images d’ensembles
• Tarski 1927b Quelques théorèmes généraux sur les images d’ensembles
• Knaster 1927 Un théorème sur les fonctions d’ensembles
The first just introduced a setting simpler than Banach’s; the others were successive
generalizations of Banach’s result. Tarski later described the third of these papers as
containing joint work that led directly to his further research two decades later.15 The
next year, at the International Congress of Mathematicians in Bologna, Tarski presented
a major survey of results on this subject: Tarski [1928] 1930, Über Äquivalenz der Mengen
in Bezug auf eine beliebige Klasse von Abbildungen. He mentioned several other contribu-
tors, and also indicated that the 1924 theorem of Kazimierz Kuratowski that was used
in the Banach–Tarski paper would fit into the new framework.

Soon after his 1939 voyage to the United States, Tarski presented material from this
research filament at several conferences. He adapted some of these results for use in the
1949a book Cardinal Algebras, published a preliminary abstract 1949b, and finally the
full 1955 paper, A Lattice-Theoretical Fixpoint Theorem and Its Applications. Its main
result, the core of the Cantor–Bernstein theorem, can be stated succinctly: every increas-
ing function on a complete lattice has a fixed point. In more recent years, largely through
the work of Dana S. Scott, Tarski’s fixpoint theorem has become a foundational block in
the theory of computing. In his 2006 survey of Tarski’s impact on that field, Solomon
Feferman noted,
... the influence of Tarski on the semantics of programming languages is so pervasive that to
detail it would require an entire presentation in itself. ... [The publication] most cited in the
computer science literature, namely his lattice-theoretic fixpoint theorem ... is an elegant
abstract formulation of the essential characteristic of definition by recursion.

For further information, consult Feferman 2006 and the literature cited there.16

As the present book went to press, an extraordinarily detailed book-length study of the
mathematics and logic related to this section appeared: Hinkis 2013.

15
Tarski 1955, 286. These papers were abstracts with considerable detail. The latter two were presented
on the same day. Knaster was Tarski’s next-door neighbor, according to the roster Polskie Towarzystwo
Matematyczne 1927, 137, 139, and to Goliska-Pilarek, Porbska-Srebrna, and Srebrny 2009a, A29.
16
Tarski 1949a, chapters 11, 16; 1955, 286. In 1951 Tarski’s doctoral student Anne C. Davis (later Anne
C. Morel) proved a converse: all lattices with this fixed-point property are complete (see Davis 1955).
Feferman 2006, 7.
 Part Three
Teaching
Part Three of this book is devoted to Alfred Tarski’s work as a secondary-school mathe-
matics teacher and teacher-trainer in Poland during 1924–1939. It contains translations
from and commentary on all his publications in that field. The biographical and back-
ground information in chapter 9 concentrates on Tarski’s involvement with secondary
education and on his family.

During 1924–1935, Tarski became involved in efforts to improve mathematics instruc-

tion at the secondary level in Poland. Chapter 10 is a translation of his report to second-
ary teachers about the First Congress of Mathematicians of Slavic Countries, held in
Warsaw in 1929. Tarski concluded that report by noting that such a congress did not
serve the needs of secondary education in mathematics; a separate organization and
activities should be established. Responding to those needs, the journals Parametr and
Mody matematyk were founded during the next two years by Antoni M. Rusiecki and
Stefan Straszewicz. A description of the journals is presented in section 9.7. An article
by Tarski in Parametr on teaching about the circumference of a circle is translated in
chapter 11. Tarski was a major contributor of exercises posed in these journals for consid-
eration by teachers and students. Those are all translated and discussed in detail in
chapter 12.

Tarski published two further articles in Parametr and Mody matematyk, about the
degree of equivalence of polygons with equal areas. These were already included in Part
Two of this book because they can be interpreted as research papers closely related to
other material featured there, as well as enrichment for secondary teachers and students.

In 1935, with coauthors Zygmunt Chwiakowski and Wacaw Schayer, Tarski published
the text Geometrja for use in the third year of secondary school. That book is described
in section 9.9; chapter 13 consists of translations of representative excerpts.
170 1 Part Three

Tarski’s contribution to the enhancement of mathematics instruction in Poland was

limited by his everyday teaching duties, by his preoccupation with the research into logic
and foundations of mathematics that made him famous, and by adverse economic condi-
tions and the onset of World War II, which impaired dissemination of his ideas. On the
other hand, the considerable influence of his secondary-teaching experience on his mathe-
matical and logical research is confirmed by the numerous intellectual filaments that
connect ideas in Part Three of this book to those that dominated the previous Part Two
on geometry, and to his pioneering work on logic.

Part Three concludes with the brief chapter 14, which ties Tarski’s early career as an
educator more closely to his research activity, particularly in logic, and recounts his
fortunate relocation to the United States at the very outbreak of World War II.

The material gathered in Part Three explains some of the relationships of Tarski’s
early work to the intellectual, political, and social milieu of Poland between the world
wars. The present editors hope that it will spur broader investigation into the connection
between mathematical research and mathematics education during that era. This hope
is a major reason for including works of such contrasting mathematical sophistication in
a single volume of selected translations.

Tarski’s life in Poland unfolded amid a chaotic vortex of political, social, economic, and
scientific developments. Accounts of events, people, and ideas were merged from several
dimensions to form the linear sequence of pages of this book. This is reflected in Part
Three by the use of singly outlined boxes interspersed in the main narrative. They
contain biographical sketches of some persons associated with Tarski, and informational
essays about some other topics. Each box can be read independently: readers are not
expected to visit them in sequence. Cross-references refer to them from the main
narrative.

Doubly outlined boxes are used in chapter 12 to distinguish the exercises, written by
Tarski, from their discussions, written by the present editors.
 9
Career and Family

This chapter describes Alfred Tarski’s work as a secondary-school mathematics teacher

and teacher-trainer in Poland during 1924–1935. It also provides background for that
activity, and for all of Tarski’s publications that stemmed from it. The first of those is
translated in chapter 10: his [1929] 2014a report to secondary teachers about the First
Congress of Mathematicians of Slavic Countries, held in 1929 in Warsaw. Tarski noted
that such a congress did not serve the needs of secondary education in mathematics; a
separate organization and activities should be established. The present chapter includes
a description of the journals Parametr and Mody matematyk, which were founded soon
after to meet some of those needs. This provides background for the exercises Tarski
posed to stimulate both teachers and students, for his article on teaching about the
circumference of a circle, and for his two research papers on the degree of equivalence
of polygons with equal area.1 They were all published in those journals, and are translated
in chapters 12, 11, and 7, respectively. Finally, the present chapter provides an overview
of the school geometry text2 that Tarski coauthored in 1935; chapter 13 includes transla-
tions of representative excerpts.

9.1 Employment and Marriage

The account in chapter 3 of Alfred Tarski’s life through age 24 included his attaining some
financial independence through employment as an instructor in logic courses for teachers
and as a mathematics teacher in Warsaw secondary schools.

Warsaw was then midway through a period of furious expansion of education. The city
was growing rapidly:
year 1914 1921 1931
population3 884,500 936,700 1,170,800
Moreover, under Russian oppression until 1915, only about one-third of Warsaw children
in the group aged seven through thirteen had been attending school; for most, schooling
lasted only one or two years. Warsaw’s literacy rate in 1915 was only about 53%. This
situation began to improve under the 1915–1918 German occupation, and the 1921 Polish

1
Tarski [1932] 2014e, [1931] 2014a, and [1932] 2014d.
2
Chwiakowski, Schayer, and Tarski [1935] 1946.
3
Wynot 1983, 93, 107.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 171
DOI 10.1007/978-1-4939-1474-6_9, © Springer Science+Business Media New York 2014
172 9 Career and Family

constitution instituted compulsory schooling for that age group. By 1925, about 63% of
those children were attending, and the literacy rate had risen to 85%. By 1931, these
proportions were 93.5% and 90%. World War I and its aftermath actually caused a decline
in population of this age group during the 1920s; nevertheless, Warsaw had to expend
great effort to achieve that improvement: 4
year 1918/19 1923/24 1929/30
population, age 7–13 150,000 132,000 110,000
number in school 50,000 83,160 102,850

In 1918 the new Polish school system was a chaos inherited from three fallen empires,
incorporating their institutions and a great variety of private schools. Reforms were
undertaken immediately, but were required only in public schools, and began to take
effect in Warsaw only in 1922–1923. Elementary schools might offer as many as seven
years of instruction. Secondary schools overlapped them: the first secondary year
generally paralleled the fifth elementary year. Qualified pupils could switch then to a
gimnazjum: a Polish secondary school that prepared students for university studies. The
full gimnazjum course of study lasted eight years, usually through age eighteen.5 Unlike
elementary-school enrollment, that in secondary schools did not grow during the 1920s,
but hovered around 30,000. Thus, it was probably not easy for anyone to find a good
teaching job at that level during the 1920s. Whereas elementary schools were populated
mostly by students from working-class families, the last years of secondary school were
dominated by those from white-collar and small-business families. A large majority of
secondary-school students attended private schools, whose high tuition made them
inaccessible to working-class families.6, 7

Poland reformed its school system in 1932: elementary schools would offer seven years
of instruction. Qualified students could transfer after the sixth year to a gimnazjum.
Gimnazjums would offer four years of instruction, covering general education—including
Latin—and some practical subjects. New institutions called liceums would offer a final
two years of secondary school.8

4
Konarski 1971, 222. Wynot 1983, 235–237. Many details in this section stem from sparse Polish govern-
ment and school records. Stanisaw Konarski (1971, 215; 1973, 179) reported that most of the archives
were destoyed during World War II and what remains is of little use.
5
Kasperowiczowa 1969, 188; Konarski1971, 221; Sadowska 2001, 71. Although her 2001 book is much
more comprehensive than those two earlier studies, Joanna Sadowska did not mention them.
6
Wynot 1983, 241–244; private secondary-school tuition in 1930 was about 700 zlotys per year.
7
According to Joseph Marcus (1983, 44, 184), the average annual income for a Polish nonfarm worker in
1929 was 2034 zlotys. He claimed that sixty percent of urban households could not afford even the average
annual rent for minimal accommodation, one room and kitchen: 480 zlotys.
8
Kasperowiczowa 1969, 194; Sadowska 2001, 71; Poland 1932. The reforms were gradually phased in.
During 1932–1934, students did not transfer into the first two classes of secondary schools. In fall 1934,
the secondary classes changed to the new numbering system. Starting in 1937–1938, three types of liceums
were instituted, emphasizing classics, natural science, or mathematics and physics. The liceums graduated
only one class before World War II began.
9.1 Employment and Marriage 173

In fall 1925 Tarski began employment as a mathematics teacher at the Third Boys’
Gimnazjum of the Trade Union of Polish Secondary-School Teachers, financed by the
Society of Friends of Polish Secondary Schools in Warsaw.9 It was located at 29 Nowo-
lipski Street, a building that it shared with some apartments. The school had been
founded in 1922 under other auspices, with only a hundred students. At first it covered
only four years—I and IV–VI—and relied on other schools to provide facilities that it
could share during afternoons. The students’ parents soon arranged that it be taken over
by the Union. By fall 1924, the Gimnazjum had acquired its own premises and switched
to the conventional morning schedule. By fall 1925, it covered all eight years and enrolled
three hundred students. In 1932 the school was reconstituted as the Boys’ Gimnazjum
of the Society of Friends of Polish Secondary Schools, under the same director, and
relocated at 150 Marszakowska Street, with entrance on Rysia Street.10 (See the figure
below.) In 1933, the school was renamed in honor of the noted Polish writer Stefan
eromski, who had died in 1925.11 According to the philosopher Karol Martel, who had
been a student there around 1930, the Gimnazjum
... had high academic standards and a liberal-minded faculty, many of whom were assimilated
Jews and socialists. The director, Teofil Wojeski, was the author of a book about eromski
and an active member of the Polish Socialist Party. Tarski immediately found the atmosphere
intellectually, politically, and culturally congenial ... . The students were good and the teachers
serious ... .12

Warsaw, 1925
140–150 Marszakowska Street

Rysia Street enters Marszakowska

midway back on the side opposite
the viewer. The eromski Gimna-
zjum was evidently behind No. 150,
the tall building most distant from
the viewer, with two turrets.

9
The Polish names of these organizations were III Gimnazjum Mskie Zwizku Zawodowego Nauczy-
cielstwa Polskich Szkó rednich and Towarzystwo Przyjaciól Polskiej Szkoy redniej w Warszawie.
10
The Polish word for reconstituted was zlikwidowane. The school had evidently lost its connection with
the trade union. The two school locations, Tarski’s family home, and the university were all in central
Warsaw, at most three kilometers distant from each other.
11
The school was destroyed during World War II; it is not related to a present-day Warsaw school with the
same name. This account of the school stems from Jadacki 2003a, 144–145; Konarski 1973; Petrozolin-
Skowroska et al. 1994; Zagórowski 1924–1926, volume 1, 135, and volume 2, 248; and especially the
1926 article by Henryk Raabe. For more information about Raabe, see the box on page 177.
12
Feferman and Feferman 2004, 56. Wojeski served during 1922–1939, except for one year.
174 9 Career and Family

Tarski met his bride, Maria Witkowska, at that gimnazjum. She had come there a year
later than he, as counselor to the first-year students. Her sister Józefa was a secretary
there.13 The wedding took place in the chancery of the Roman Catholic All Saints Church
in Warsaw, on 24 June 1929. The marriage record lists Alfred’s and Maria’s positions
as docent at the university and as official (in Polish, urzdniczka). Maria’s brother
Antoni, then an army lieutenant stationed in Vilnius,14 attended as witness. Tarski’s
father and mother were listed, but as “Ignacy and Róa, the Tarski couple,” with no
mention of their surname Teitelbaum.15

The marriage record indicates that Maria was from Minsk,16 and twenty-seven: born
in 1902, one year younger than her groom. Her parents, deceased, were Wincenty Wit-
kowski and Maria Janczewska. In their biography of Tarski, Anita B. and Solomon
Feferman wrote,
The gods were good to Tarski when they sent him Maria Witkowska as a life partner.

They have provided a detailed and sympathetic account of an interesting, complicated,

heroic woman whom they knew, respected, and admired. They noted,17
Many of Tarski’s acquaintances had the impression that Maria’s family was gentry ... . How-
ever, the facts according to the Tarski children are that Maria’s parents owned a butcher shop
where they both worked long hours; they may also have had some small land holdings ...
but they weren’t rich.

In his biographical sketch of Tarski, Jacek Juliusz Jadacki provided much more detailed
information about Maria’s background, with an emphasis that at first seems incompatible
with that quotation. The following three paragraphs are based mostly on his account.18

Maria’s maternal grandfather Julian Janiczewski owned a large estate near Minsk.
He was killed in the German attack on that city during the final stages of World War I.
Maria’s paternal grandfather Antoni Witkowski and her father also owned property near
Minsk. For advocating Polish independence, they had once been exiled to Siberia by the
tsarist regime. Besides her older sister Józefa and younger brother Antoni, born in 1900
and 1905, Maria had sisters Agnieszka, Jadwiga, and Helena, born in 1904, 1909, and
1911. Their mother died at age thirty-eight, soon after the birth of her sixth child in
eleven years. At this time, Maria’s father was seventy years old. He then married his

13
Feferman and Feferman 2004, 65; Jadacki 2003a, 148; Raabe 1926. First-year students were about
twelve years old.
14
Long part of part of Russian-held Poland, Vilnius (in Polish, Wilno) was occupied at various times during
1915–1922 by German, Polish, Russian, and Lithuanian forces. It was part of Poland until 1939, when
it was annexed by the Soviet Union. Since 1991 it has been the capital of Lithuania.
15
Jadacki 2003a, 133. Parafia Wszystkich witych [1929] 1940.
16
Minsk (in Polish, Misk) also had been part of Russian Poland. During 1915–1921 it changed hands
several times between Russian, German, and Polish control, suffering grievously. Afterward it became
part of the Soviet Union. It has been the capital of Belarus since 1991.
17
Feferman and Feferman 2004, 64.
18
Jadacki 2003a, 133–137, 148–149. Jadacki’s sketch seems to have been prepared from informal notes
with little editing. It relied on personal contacts with surviving members of Maria’s parents’ families.
9.1 Employment and Marriage 175

first wife’s younger sister Antonina, who became stepmother to Maria and her siblings.
These details are not incompatible with the impressions that the Fefermans reported:
Maria’s forbears were landowning families, but at least one suffered years of ill fortune.

During the 1919–1920 Polish–Soviet War, in front of his family, at age seventy-eight,
Maria’s father, Wincenty, was murdered by Soviet troops. The children were moved to
an orphanage on a collectivized farm. The three eldest sisters quickly joined the Polish
Military Organization—East.19 They safeguarded and transmitted communications,
helped prisoners escape, and transported weapons. Agnieszka was a nurse in a hospital
for Polish prisoners; Maria was treasurer for her unit and assembled explosives. Józefa
became deputy commander of the women’s section, and made several daring and danger-
ous trips across enemy lines to Moscow, in times of chaos and blockade, always without
identification or with false papers. The photograph below shows the three sisters with

Maria Tarska in the 1919 –1920

Polish–Soviet War

From left, the women are Maria,

Józefa, and Agnieszka Witkowska.
The man is Józefa’s fiancé Bolesaw
Zahorski.

19
The name in Polish is Polska Organizacja Wojskowa–Wschód. Organized by Józef Pisudski during World
War I for diversionary and intelligence operations, this secret body persisted in the east for several years
afterward (N. Davies 1982, volume 2, 381). The following description of the sisters’ exploits is based on
Ziemiaski 1933, 94–99, 196–198. Jadacki (2003a, 148) cited this account of the efforts of women in the
organization, but his report differs in some respects.
176 9 Career and Family

fellow soldier Bolesaw Zahorski, who would become Józefa’s husband.20 After the war,
the three sisters could not return to Minsk, in the Soviet Union; they went to Warsaw.
All three were awarded the Cross of Valor medal (Krzyem Walecznych) for heroism.
During 1924–1927, Maria studied law at the University of Warsaw. She and Józefa found
employment at the Third Boys’ Gimnazjum of the Trade Union of Polish Secondary-
School Teachers, where Tarski taught. Maria became the counselor for its first-year
students (aged about twelve), and Józefa became the school secretary.

Soon after their marriage, the Tarski couple moved to an apartment on Sukowski
Street in the oliborz district.21 (See the photograph on page 178.)
Inspired by the Bauhaus ... principles of design, this area of stylish new apartments, flats, and
houses surrounded by greenbelts was considered a choice place to live ... out of the hustle and
bustle of the center of town yet ... only a ten-minute trolley ride to the university.

Tarski made frequent recreational visits to the Tatra mountains in southern Poland,
particularly to the resort town Zakopane. Maria sometimes accompanied him and his
professional associates, although she did not travel elsewhere with him. His excursions
there and abroad included affairs with some female colleagues, evidently with Maria’s
knowledge.
Her way of dealing with her husband’s ... affairs was to accept them as graciously as she could.
The ethos of the culture ... was that husbands would have their mistresses and wives would
look the other way; moreover, in their circle of friends, monogamy was seen as an unnatural
bourgeois convention that scarcely anyone adhered to.

Maria continued university studies during 1930–1935, in education. She also kept up her
martial skills and nerve: in 1931 she took honors in a two-position pistol-shooting match
for Warsaw teachers.22 Sharing two incomes, she and Alfred were enjoying “relatively
good times ... surrounded by friends and like-minded people both at home and at work.”
Nevertheless, Alfred’s
... biggest problem was finding enough time to do research and write papers; his solution, when
the pressure became too great and he absolutely needed uninterrupted time to bring a piece
of work to conclusion, was to resort to ... calling in sick.

Tarski’s gimnazjum substitute during his leaves was Roman Hampel.23

20
For information about Zahorski, see the box on page 177.
21
This paragraph and the next are based mostly on Feferman and Feferman 2004. The quotations are from
pages 66 and 92. They specified Tarski’s address as No. 8, but Jadacki (2003a, 153) said that until late
1934, the address was No. 2, apartment 5. Perhaps the Tarski family moved during that time. See also
Nowicki 1992, 154.
22
Ogniwo 1931, 112. This article identified her employer as Warszawa III Gimnazjum Zwizku Zawodo-
wego Nauczycielstwa Polskich Szkó rednich; it had not yet changed its name to eromski Gimnazjum.
23
Feferman and Feferman 2004, 66; Jadacki 2003a, 145. Concerning Hampel, see the box on page 177.
9.1 Employment and Marriage 177

Roman Hampel was born in 1907 and schooled in Warsaw, when it was in the Russian Empire.
During 1924–1929 he studied mathematics at the university there, obtaining a master’s degree.
During the 1930s he taught in secondary schools and obtained another master’s at Cracow in
education and psychology. In 1951 he became chair of management mathematics and then vice-dean
of the electrical engineering faculty at the Warsaw Polytechnic. Hampel died in 1963.*

Born in Warsaw in 1882 when it was in the Russian Empire, Henryk Raabe became active in socialist
politics around 1902. He earned a doctorate in biology at the University of Cracow in 1915 and
started teaching in secondary schools. Raabe helped found the Trade Union of Polish Secondary-
School Teachers† in 1919 and served as its president during 1922–1927. In 1922 the union took over
the school that became the eromski Gimnazjum. Raabe taught biology there, and Alfred Tarski
joined the faculty in 1925. As president of the union, Raabe edited its journal Ogniwo, which pub-
lished his 1926 account of the school and Tarski’s [1929] 2014 report, translated in chapter 10. Raabe
worked as a union organizer through the 1930s. During the 1939–1941 Soviet occupation, he became
professor at the University of Lwów. After World War II, he participated in the establishment of the
Communist regime in Poland, serving as ambassador to the Soviet Union, rector of the University
of Lublin, and member of parliament. A successful popularizer of science, Raabe died in 1951.‡

Bolesaw Zahorski was born in 1887. A journalist active in the Polish independence movement, he
published under the pseudonym Bolesaw Zygmunt Lubicz, and was imprisoned in Siberia. In 1920
he authored a volume of poetry about service in the Polish–Soviet war. After that war he married
his fellow soldier Józefa Witkowska, who would become Alfred Tarski’s sister-in-law. In 1922, soon
after returning to civilian life, Zahorski died in an accident.§

* Jakubowski 1964. † Zwizek Zawodowy Nauczycielstwa Polskich Szkó rednich ‡ Brzk 1983
§ Roliski 1996, 482.

In summer 1934, Maria was pregnant with the Tarskis’ first child. Alfred spent 30
August through 7 September at conferences in Prague, and was preparing for a prolonged
visit abroad during the next year.24 They gave up their oliborz apartment, and evidently
sublet a room in Alfred’s parents’ apartment. Maria bore their son Janusz Andrzej on
11 December 1934.25 Soon, Alfred left for a research visit to Vienna: a Rockefeller grant
supported him there from January through June 1935. He returned home briefly, then
spent the week of 15 September at the First International Congress for the Unity of
Science in Paris.

24
See section 14.2. Jadacki, probably in error, suggested (2003a, 153) that Tarski may also have studied
in Vienna during 1934.
25
On passport documents in the Archiwum Akt Nowych (Central Archives of Modern Records) in Warsaw,
dated August 1934, and on the 29 November letter Tarski [1934] 2014, Alfred used his parents’ address.
His loan application, Tarski [1935] 2014, indicated that the Tarski family was living in one room pod-
najtym przy wikszej rodzinie—sublet ( from or with) ( a or the) larger family. It is not clear whether
Tarski was emphasizing or downplaying his reliance on his parents. These letters are translated in section
16.3. See also Jadacki 2003a, 153–158. The Fefermans gave Janusz’s birth date incorrectly (2004, 66).
 51 Koszykowa Street
Tarski’s Parents’ Residence

4 Sukowski Street
Tarski’s Residence, 1936–1939
9.2 Teaching Geometry 179

Afterward, the couple, now with a child, realized that their quarters with Alfred’s
parents were impossibly crowded. Because rents were exhorbitant, Alfred applied to the
Ministry of Religious Denominations and Public Education26 for an advance equal to his
university salary for six months to cover a down payment of 3500 zlotys toward purchas-
ing a new apartment in oliborz. He would pay it back in installments over three years.
After a short time, Tarski withdrew the application, noting that he had obtained a loan
from another source. From early 1936, the Tarski family lived at an apartment they
owned, 4 Sukowski Street, apartment 18. It consisted of two and a half rooms, kitchen,
bathroom, and basement.27 This building and Tarski’s parents’ building are contrasted
in the figures on page 178. Some statistics permit comparing the Tarskis’ life with that
of Warsaw residents in general: around 1930, only 38% of the Warsaw population were
housed in quarters of more than two rooms; and repaying that down-payment loan would
have cost the Tarskis more than half of the average annual salary, 2034 zlotys, of a single
Polish nonfarm worker.28

9.2 Teaching Geometry

The context in which Alfred Tarski began teaching geometry at the Third Boys’ Gimna-
zjum of the Trade Union of Polish Secondary-School Teachers included its faculty and
curriculum and the tradition of geometry instruction in which Tarski would craft his own
approach. This section considers these aspects in turn, and emphasizes two geometry
texts on which he relied.
When Tarski arrived at the Gimnazjum in 1925 to teach mathematics, it covered the
full eight years of secondary school, including those that overlapped elementary school.
Its faculty numbered twenty-one, including the headmaster and a priest. Besides Tarski,
only one other teacher, the art instructor, was younger than thirty. There were three
women, who taught languages, social studies, and biology. Two men who had earned doc-
torates taught science. The one other mathematics teacher had an engineering degree.
All faculty members were identified as Christian—fourteen, including Tarski, as Roman
Catholic. The Gimnazjum offered instruction in the following subjects:29
Subjects Number of Teachers
Arts and Gymnastics 4
Biology, Chemistry, and Physics 3½
Latin 3
Polish 2½
French and German 2½
Geography and History 2
Philosophy and Religion 2
Mathematics 1½

26
Ministerstwo Wyzna Religijnych i Owiecenia Publicznego, often abbreviated MWRiOP.
27
The Fefermans suggested that the loan may have been facilitated by the Teachers’ Union. The new
apartment was described in the letter Zakad 1935 from the Social Insurance Institution, which had just
developed a large complex of new apartments in oliborz. That organization still exists.
28
Marcus 1983, 44, 186. It is reasonable to assume, but not verified, that the required 3500 zlotys amounted
to about six months’ university salary for Tarski.
29
Zagórowski 1924–1926 volume 2, iv–vi, 248–249. Teachers with two subjects are counted half in each.
180 9 Career and Family

There was no instruction in Russian! One of the biology teachers, Henryk Raabe,
reported that the school was remodeled in 1926 and “completely met the needs of a
modern school, as far as is possible under the conditions of being located in a residential
building.” His account showed that the faculty remained stable for that year; the
philosophy teacher retired and was replaced by the already well-known philosopher
Kazimierz Ajdukiewicz.30

Evidently, instruction at the Gimnazjum included mathematics each year—both

algebra and geometry. The national standard recommended four hours per week for years
IV–VI and five hours for the final years VII–VIII.31 Tarski taught only mathematics, but
the other mathematics instructor also taught physics. Tarski’s teaching load was
probably all five of these classes: twenty-two hours per week total.32

What comprehensive geometry texts were available for use in a gimnazjum in 1925?
Surveying library catalogs and advertisements in Polish journals for teachers, the present
editors discovered seven such texts, three of which were written originally in Polish:
Originally Polish Translated
omnicki 1923 Enriques and Amaldi [1903] 1916 (from Italian)
Wojtowicz [1919] 1926 Kiselev 1917 (from Russian)
Zydler 1925 Monik 1896 (from German)
Zupani 1918–1921 (from German) 33
Perhaps Tarski based his first gimnazjum classes on one of those texts. Some of them
are notable. Andrei Petrovich Kiselev’s book was originally published in Russian in 1892.
An English edition appeared in 2006–2008; the publisher claimed, “It is by far the most
famous Russian textbook, in all subjects, ever.” Tarski himself probably learned from a
Russian edition when he was in school. Somewhat modernized versions of the texts by
Wadysaw Wojtowicz and Jan Zydler are currently available, too. Tarski referred to
Enriques and Amaldi [1903] 1916 to establish the elementary foundation required for his
own [1924] 2014b research work. In his [1931] 2014a article on equidecomposability of
polygons, Tarski used specific arguments from a later edition of Wojtowicz’s text. These
two texts will now be described and compared.34

30
Raabe 1926. For more information on Raabe and Ajdukiewicz, see the boxes on page 177 and chapter 3.
31
Poland 1931. The present authors do not know what was recommended for years I–III.
32
According to his Berkeley colleague John W. Addison (1983), Tarski sometimes taught as many as twenty-
nine hours per week. In section 9.3 it is shown that during 1930–1939, Tarski’s university teaching load
averaged 5.75 hours per week. See also Givant 1999, 50.
33
Books by Rihard Zupani are often listed with the German spelling, Richard Suppantschitsch. A few
specialized books were available for trigonometry, analytic geometry, and descriptive geometry, which
were covered to some extent in the recommended curriculum. Kalicun-Chodowicki 1925 is an example.
34
Tarski’s references to these texts are on pages 79 and 142 of the present book. For more information on
Wojtowicz, see the box on page 189.
9.2 Teaching Geometry 181

Ugo Amaldi was born in 1875 in Verona. He earned the laureate from the University of Bologna
in 1898 under supervision of Salvatore Pincherle. Amaldi won a professorship at Cagliari in 1903,
then transferred to Modena in 1906, Padova in 1919, and Rome in 1924. There he taught mathemat-
ics in the architecture faculty until 1942, when he joined the science faculty. Amaldi’s main research
lay in the determination and classification of all continuous groups of transformations that depend
on finitely or even infinitely many parameters. In 1901 Amaldi began extending Sophus Lie’s
pioneering work, and he eventually published a complete, detailed exposition for dimensions one
through four. In 1918 Amaldi received a gold medal for this work from the Società Italiana delle
Scienze, detta dei XL, and was elected to the Reale Accademia dei Lincei. Amaldi was best known,
however, for his vital collaborations with Federigo Enriques and Tullio Levi-Civita in producing many
highly readable expositions of elementary mathematics, starting in 1903, and mathematical physics,
starting in 1920. He was sincerely but unostentatiously religious, with a reputation for calmness and
impartiality. His son and daughter both became scientists. Amaldi died in Rome in 1957.*

Federigo Enriques was born in 1871 in Leghorn, to a prosperous family. They soon moved to Pisa,
where he attended secondary school and the Scuola Reale Normale Superiore. He earned the laureate
there in 1891 under supervision of Enrico Betti, then spent five years in postgraduate study and
temporary positions. Enriques won the competition for a permanent chair in Bologna in 1896; he
remained there until 1923. His best-known mathematical work was with Guido Castelnuovo on the
birational classification of surfaces. Enriques also investigated questions in foundations of geometry
and in differential geometry. In 1907 he wrote a major article on foundations for the Encyklopädie
der mathematischen Wissenschaften. Later he collaborated with Castelnuovo on two more such
articles, about algebraic surfaces and birational transformations. Enriques developed close relation-
ships with scholars of many other disciplines. From 1900 on, his work included elementary texts and
books for mathematics teachers, many in collaboration with Ugo Amaldi, and books on history and
philosophy of mathematics and science. Polish translations of some of these works played a major
role in the geometry discussed in the present book. In 1906 Enriques was elected to the Reale Acca-
demia dei Lincei. During 1907–1913 he served as president of the Italian Philosophical Society. He
organized and presided over the 1911 International Congress of Philosophy held in Bologna. In 1923
Enriques was appointed to the chair of higher geometry at the University of Rome. There he founded
the National Institute for the History of Science. He remained in Rome until 1938, when he was
dismissed because he was Jewish. Enriques spent most of his remaining years in hiding. He died
in 1946.†

*Terracini 1958, Viola 1957.

† Castelnuovo 1947, Eisele 1971. Castelnuovo and Enriques 1908, 1914. The present account of Enriques’s
last years is from Castelnuovo 1947. According to Eisele 1971, however, Enriques “retired from teaching”
during those years.
Ugo Amaldi
about 1910

Federigo Enriques
in 1914
9.2 Teaching Geometry 183

Of the texts just listed, Elements of Geometry for Use in Higher Secondary Schools by
Federigo Enriques and Ugo Amaldi probably played the most important role in Tarski’s
early geometry teaching. It was first published in Bologna in 1903. Both authors were
noted research mathematicians, early in their careers, specializing in connections between
algebra and geometry.35 At that time in Italy it was common for professors to spend ser-
ious effort simultaneously on research and on improving instruction in the schools. This
book appeared in many editions, tailored to various audiences. In 1916, Wojtowicz trans-
lated it into Polish, presumably for use in Polish schools.

Enriques and Amaldi followed an axiomatic approach, but tempered it by describing

various physical experiences before codifying them as geometric postulates. These were
phrased in terms of the undefined notions point, line, plane, incidence of these objects,
order of collinear points, and congruence of segments and of angles, using the notion of
set when appropriate. The authors gave great care to formulation and examples of the
basic concepts, and presented proofs of theorems in detail. There are ample illustrations
and hundreds of exercises. A French reviewer wrote,
The authors (through perhaps excessive imitation of the ancient geometers) exclude the idea
of number, or at least postpone its introduction as long as possible, and reason with figures
and geometric magnitudes. Such a method seems somewhat artificial today. ... In sum, the
authors have made logical rigor paramount ... but sometimes that seems to be at the expense
of the natural order and simplicity.36

The authors assumed that their student readers could tolerate and absorb the content
of long, meticulous discussions. They made no allowance for those whose reading skills
had not yet attained the level of highly educated adults. The result was a book that is
mathematically correct and delightful reading for a sophisticated, motivated, independent
learner, such as a professional mathematician.

Wojtowicz evidently perceived this disconnect between the book’s style and its intended
readers: students like those whom Tarski would teach. In 1919 Wojtowicz published his
own text that covered most of the same material and added some new topics. He spent
considerable effort to help readers, by clarifying emphasis and simplifying the language.
According to Stefan Kwietniewski’s brief 1923 review in the Handbook for Self-Education,
Of all the geometry textbooks originally written in Polish, Wojtowicz’s book stands out for its
accuracy and [abundance of ] content.

Nevertheless, Wojtowicz’s style was not different enough to make it a substantial

improvement over Enriques and Amaldi [1903] 1916. It still assumed of its readers a level
of sophistication beyond that of any but exceptionally talented secondary-school students.

This survey of geometry books published for use in Polish-language secondary schools
reveals that from 1900 on, administrators and mathematics teachers could select from
a variety of approaches and textbooks. By 1925, the Polish government had begun to
standardize the curriculum. That process led to new publications and a greater number

35
Enriques and Amaldi [1903] 1913. For more information on those authors, see the box on page 181.
36
Revue de métaphysiques 1904.
184 9 Career and Family

from which to choose. For example, Bazyli Kalicun-Chodowicki, the director of a boys’
gimnazjum in Lwów, published in 1925 a textbook with the full title
Descriptive Geometry for the Higher Classes of Secondary Schools (A Systematic Course
Following the Program of the Ministry of Religious Denominations and Public Education).

Stricter curriculum standards probably improved the overall quality of mathematics

instruction in Poland, but also curtailed the variety of approaches available37 and may
have made it more difficult for the most knowledgeable teachers to experiment with
alternative approaches, such as Tarski’s own axiom system for geometry, discussed in
section 9.4.

9.3 Teaching Teachers

To prepare students for participation in a rapidly evolving society, teachers should not
merely present their subjects as they were themselves taught. In 1914, Stefan Kwietniewski
and Wadysaw Wojtowicz emphasized application of that principle to the teaching of
mathematics in Poland:
One of the most outstanding attributes of modern mathematics is the work in elucidating its
foundations ... . These results ... are so momentous that they must exert a profound influence
on the exposition of the whole of mathematics, from the lowest to the highest levels of teaching.
Today teacher[s] may no longer take the precritical position, allez en avant et la foi vous
viendra. [They] must not only thoroughly understand the newest research in the foundations
of mathematics, but also restructure their lessons from the ground up, in order to make them
most straightforward and in accordance with modern science. This reconciliation ... can only
be established through many trials and the concerted efforts of a whole multitude of people
—scholars as well as educators.38

Through his research, Alfred Tarski had become part of that forward movement. This
section will argue that this principle played an essential role in his university teaching,
both in elementary courses intended for general audiences and in courses offered specifi-
cally for training teachers. Other noted mathematicians had contributed major works
emphasizing the connection between foundational research and mathematics education.
Tarski referred to them in his own research reports, and presumably used them in
preparing his own classes in schools and those at the National Pedagogical Institute, for
preparing other teachers (see chapter 3).

Three of those works are described next: David Hilbert’s Foundations of Geometry,
Mario Pieri’s Elementary Geometry Based on the Notions of Point and Sphere, and the
collection Questions Regarding Elementary Geometry edited by Federigo Enriques.39
These descriptions are followed by a detailed account of Tarski’s efforts in the training
of school teachers.

37
A study is needed of the approaches to secondary mathematics instruction prevalent in the early 1900s,
contrasted with the Polish reforms of the interwar period; but that is beyond the scope of the present book.
38
Kwietniewski and Wojtowicz 1914. The French clause (italicized by the present editors) means go on,
and it will come to you. For more information on Kwietniewski and Wojtowicz, see the box on page 189.
39
Hilbert [1899] 1971, Pieri [1908] 1915, and Enriques [1900] 1914–1917.
David Hilbert
in 1900

Hilbert’s
Foundations of Geometry
Title Page
186 9 Career and Family

In 1899 Hilbert had published Foundations of Geometry. For at least half a century
it would be the standard rigorous presentation of Euclidean geometry.40 Hilbert’s unde-
fined terms were point, line, and plane; incidence of those objects; betweenness of points;
and congruence of segments and of angles. (Enriques and Ugo Amaldi adopted those
primitive terms for their [1903] 1913 text.) Hilbert developed a large portion of Euclidean
geometry in a few pages. He adhered to the style conventional for mathematical research:
rigorous, but without extreme detail. His postulates stated well-known properties of
geometric objects familiar to all. Many were quite succinct. But some were formulated
in terms of undefined properties of defined notions, such as angle congruence, and became
logically complicated when the latter were replaced by their definitions.

Hilbert introduced early the use of points on a selected line as scalar coordinates, and
showed how various geometric properties of points correspond to algebraic properties of
scalars. For example, he showed how versions of Desargues’s theorem are connected with
the scalar addition and multiplication operations. Although that theorem involves only
incidence of points and lines, all in a single plane, and in three dimensions is derivable
from incidence axioms alone, he showed that additional postulates—his congruence
axioms—are required to prove it in plane geometry.

Hilbert’s striking new results, the scope of the geometry he considered, his effective
style, and the placement of his work in the mainstream of the mathematics of the day
made Foundations of Geometry instantly famous. A French translation, [1899] 1900, was
published the next year. Hilbert’s work became identified with the birth of the modern
axiomatic method in mathematics, even though that had been employed already for some
years by Italian mathematicians. Tarski relied on Hilbert’s book to provide the necessary
foundation for his own [1924] 2014b research work and for the [1931] 2014f exercise that
he posed for gimnazjum students and teachers.41

The second foundational work on which Tarski relied for the development of his
approach to geometry was Mario Pieri’s [1908] 2007 memoir Elementary Geometry Based
on the Notions of Point and Sphere.42 An alternative to Hilbert’s foundation, it also
presented its subject axiomatically, but with all its notions and postulates defined and
formulated in terms of the notion point and the relation that holds between points a, b,
and c just when a and b are equidistant from c. The paper’s title reflects Pieri’s exten-
sive use of elementary set theory in developing geometry from his postulates: he defined
the sphere through b centered at c as the set of all points a such that a and b are
equidistant from c. Pieri’s second aim was to foster more extensive use of properties of
spheres in presenting elementary geometry, even in school courses. A third goal, which
Pieri had already pursued for a decade, was to promote the use of transformations in
elementary geometry. Pieri introduced various geometric transformations early through
definitions, and employed them extensively throughout the memoir.

40
Hilbert [1899] 1971: see the bibliography for details about its editions. This paragraph and the two fol-
lowing were adapted from Marchisotto and Smith 2007, subsection 3.10.2, with the publisher’s permission.
41
Tarski’s references to Hilbert’s text are on pages 79 and 264 of the present book.
42
For information about Mario Pieri’s life and mathematics, consult Marchisotto and Smith 2007.
 Mario Pieri
around 1908

Pieri’s
Point and Sphere Memoir
Polish Translation
188 9 Career and Family

Pieri’s presentation of elementary geometry differed greatly from Hilbert’s. Pieri’s

postulates employed, besides his two undefined terms, only the simplest defined notions,
such as sphere. Many of his postulates were not at all succinct, but when phrased solely
in terms of the primitive notions point and equidistance, their logical structures were
simple. Pieri’s exposition departed from the conventional mathematical research style:
he provided virtually every detail of every proof ! Hilbert’s exposition was much more
readable, but Pieri’s was easier to analyze.

In 1915, Pieri’s Point and Sphere memoir was translated into Polish by Stefan
Kwietniewski, one of the original members of the mathematics faculty at the University
of Warsaw. The Mianowski Fund supported its publication. Tarski studied it intensely.
His later colleague Steven R. Givant reported,
Tarski was critical of Hilbert’s axiom system from a logical perspective ... [He] preferred Pieri’s
system, where the logical structure and the complexity of the axioms were more transparent.43

Although the formats of Hilbert [1899] 1922 and Pieri [1908] 2007 differed greatly,
each presented much the same material as a traditional secondary-school geometry course
based on Euclid’s centuries-old approach. Hilbert did adopt an advanced viewpoint to
discuss the reasoning behind his approach, but provided only exceedingly brief explana-
tions. In contrast, the third foundational work on which Tarski relied was Enriques’s
[1900] 1914–1917 book Questions Regarding Elementary Geometry. This was a collection
of articles by thirteen Italian authors, who described various aspects of elementary geom-
etry in considerable detail from an advanced viewpoint. Publication of its translation was
also supported by the Mianowski Fund. The authors of its articles ranged from secondary-
school teachers to leaders in mathematical research. According to its translators Kwiet-
niewski and Wojtowicz, it
can be considered an extensive scientific and didactic commentary on a modern course in
elementary geometry ... . But in addition to this (despite certain shortfalls unavoidable in a
collective work), it is a straightforward and clear introduction to the new research in axioma-
tics, written especially with the needs of teachers in mind.44

Three of the articles were devoted to philosophical and pedagogical questions, five to
fundamental geometric concepts, and seven to the theory of Euclidean constructions and
constructibility. Two very long articles were devoted to non-Euclidean geometry and to
isoperimetric problems. Several of the authors were major figures in Italian mathematics,
and some of these papers even today provide excellent background for theoretical ques-
tions about elementary geometry.

Having described the foundational works on which Tarski relied for developing his
approach to geometry, this section now surveys the contexts in which he presented it:
his university courses and three “vacation courses” for in-service teachers in Poland.

43
Pieri [1908] 1915. Givant 1999, 50. See also Szczerba 1986, 908.
44
Kwietniewski and Wojtowicz 1914. For more information about these men and the Mianowski Fund, see
the box on page 189.
9.3 Teaching Teachers 189

Stefan Kwietniewski was born in 1874 in Warsaw, which was then in the Russian Empire. His father,
Wadysaw (1840–1902), a mathematician, taught in secondary schools and sometimes as docent at
the University of Warsaw. He also translated mathematical texts, and served in academic administra-
tive roles, one with the Mianowski Fund. Stefan’s brother became professor of zoology at Lwów.
Stefan studied in Munich and Göttingen, then earned the doctorate from the University of Zurich
in 1902 with a dissertation on surfaces in a space of four dimensions, supervised by Heinrich
Burkhardt. Afterward, he supported himself by lecturing on and writing about mathematics and
translating mathematics texts. From 1915 to 1939, he regularly gave a geometry course at the
University of Warsaw, but to few students. Withdrawn and solitary, he did not participate in its rich
mathematical life, and never became a professor. His lectures eventually constituted almost the only
evidence of his existence. An avid outdoorsman, he moved to a rural environment in 1939, but died
suddenly in 1940. With support from the Mianowski Fund, Kwietniewski translated several texts
that are important to the mathematics in the present book, and contributed significantly to volumes
1 and 3 of the Heflich and Michalski 1915–1932 Handbook for Self-Education, which introduced many
Poles to advanced mathematics.*

Wadysaw Wojtowicz was born in 1874 in Winiowiec and schooled in ytomierz—then in the
Russian Empire, those places are now in western Ukraine. He studied law at the University of Kiev,
then mathematics and philosophy in Berlin and Zurich. From 1907 on, Wojtowicz taught in and
served as director of several secondary schools. During 1919–1922 he also served in the Ministry of
Religious Denominations and Public Education. Wojtowicz was editor of Wektor and, with Stefan
Straszewicz, of Przegld matematyczno-fizyczny—journals devoted to secondary-school mathematics.
The latter published Alfred Tarski’s [1924] 2014b paper On the Equivalence of Polygons. Wojtowicz
translated into Polish several well-known mathematical works, including those of Federigo Enriques
and Ugo Amaldi, which were important sources for the mathematics described in the present book.
Wojtowicz himself authored a volume of mathematical tables and several popular secondary-school
textbooks. In research reports, Tarski referred specifically to Wojtowicz’s [1919] 1926 geometry text.
Wojtowicz died in 1942 in Warsaw.†

Mianowski Fund. Under Russian rule, education in the Polish language and for the advancement
of the Polish people was heavily suppressed. In 1869 the Russian government closed the only
university so devoted. On the death of its last rector, Józef Mianowski, members of its former
community obtained permission to organize a foundation named for him, ostensibly to pay for
printing treatises. The government required it to be supported solely by private funds, and forbade
it to initiate other activities in science and education. The fund was remarkably successful in raising
support, including ownership of an oil field in the Caucasus. It became Poland’s largest institution
supporting science. The fund financed a series of Polish guidebooks for self-instruction in university-
level subjects and numerous translations of monographs in advanced mathematics and physics. It
published the research journal Prace matematyczno-fizyczne, the journals Przegld pedagogiczny and
Wektor, which served teachers and supported mathematics instruction, and the journal Nauka polska,
devoted to scientific news. These publications substituted for instruction not provided to Polish
students under the Russian regime, and lay well beyond the scope of the Russians’ intention for the
foundation. Several of them were particularly important for the geometry in the present book.
During the first decade of Polish independence, the fund supported seventy percent of the scholars
receiving doctoral degrees in Poland. It was supporting the journal Przegld filozoficzny when Alfred
Tarski’s first paper, Tarski 1921 [2014], was published there. There is no evidence that it supported
Tarski further, but it did fund some of the activity of his colleague Stefan Banach and teachers
Tadeusz Kotarbiski and Stanisaw Leniewski. The Mianowski Fund survives today.‡

*Dobrzycki 1971; Marczewski 1971.

† W. Piotrowski 2003. Tarski’s references to Wojtowicz [1919] 1926 are on pages 142 and 156 of the present book.
‡ Kuzawa 1968, chapter 3; Hübner, Piskurewicz, and Zasztowt 1992.
190 9 Career and Family

In 1925, Tarski had been granted the venia legendi, or right to teach at the university,
probably based on his 1924b paper about definitions of finiteness. From fall 1925 on, the
University of Warsaw hired him each year, always on a temporary part-time basis, to
teach various courses concerned with logic research and with elementary mathematics
from an advanced viewpoint. Many of these were aimed at some combination of general
audiences and future teachers. Tarski began working closely with Jan ukasiewicz and
the students associated with his research program in logic. In most trimesters Tarski
offered a problem seminar (proseminar) and supervised some of those students’ research.
In autumn 1929 he officially became assistant to ukasiewicz.45

Tarski’s university teaching assignments are detailed in the table on page 191.46 In
1919 the university had adopted an academic year of three trimesters:

autumn . . . October–December,
winter . . . . January–March,
summer . . . April– June.

Most of Tarski’s courses lasted all year; exceptions are noted in the table. Except in
1925/1926, each of his classes met once a week; the number of hours is shown. Some
items in the table reflect notes in Jadacki 2003a indicating that scheduled courses were
canceled. During 1925–1929, Tarski’s average university teaching load was 2.6 hours per
week. His appointment as ukasiewicz’s assistant brought an increase: during
1930–1939 his average was 5.75 hours. With difficulty, Tarski managed that on top of
about 22 hours per week teaching at the eromski Gimnazjum. (See section 9.2.)

Tarski’s courses listed in the table with names in roman type were probably offered
principally for students beginning research in logic. Those with names in italics were
probably attended by general audiences, including future and in-service teachers. For
example, Warsaw elementary-school teacher Wacaw Schayer enrolled in Tarski’s courses
Topics in Methodology and Topics in Elementary Geometry during 1930/1931 and 1931/
1932, respectively. He earned a master’s degree, became a school director, and coauthored
with Tarski the secondary-school textbook Geometrja, described and translated in section
9.9 and chapter 13.47 The borderline between those types of courses was probably not
clear,48 but even with that allowance, it is apparent that nearly half of Tarski’s classroom
work for the University of Warsaw supported its role in teacher preparation.

45
Jadacki 2003a, 144–149.
46
The data are from Warsaw University 1925–1939. Jacek Juliusz Jadacki (2003a) has noted that there
were occasional differences between course advertisements and what actually took place. Evidently, the
university adjusted the schedule according to student demand and availability of instructors and funds.
47
For more information about Schayer, see the box in section 9.9, page 221.
48
For example, Mojesz Presburger attended Tarski’s courses of both types during 1925–1929 (Zygmunt
1991, 213).
 Tarski’s Teaching Load at the University of Warsaw

Year Hr Courses Remarks

1925/26 2 Theory of cardinal numbers Fall and winter only.

1926/27 2 Elementary mathematics Origin of Tarski’s axioms for geometry.

Canceled: 2-hr logic research course.

1927/28 2 Logic research seminar Canceled: 4-hr elementary mathematics course

1928/29 2 Arithmetic of real numbers Tarski recorded (1933, 56; [1933] 1983, 205) that
1 Algebra in secondary schools the 1927/28 and 1928/29 seminars included
2 Logic research seminar reports on the completeness of various theories.

1929/30 2 Logic research seminar Canceled: 2-hr elementary geometry course.

1930/31 2 Arithmetic of natural numbers

2 Topics in methodology
2 Exercises and proseminar in logic With ukasiewicz (). Exercises=wiczenia.

1931/32 2 Topics in elementary geometry

2 Methodology of deductive sciences
2 Exercises and proseminar in logic ()

1932/33 2 Topics in elementary geometry Stressing area, volume. Only 1 hr in winter.

2 Methodology of deductive sciences
1 Logic exercises for scientists ()
2 Logic exercises, proseminar () Only 1 hr in winter.

1933/34 2 Theoretical arithmetic

2 Methodology of deductive sciences
1? Logic exerc. for mathematicians (). Fall only.
1 Logic discussion, proseminar (). discussion=konwersatorjum. 2 hr, spring.

1934/35 3 Theory of school algebra Theory=podstawy teoretyczne.

2 Methodology of deductive sciences These courses were all scheduled for the full year
2 Logic and methodology exercises but Tarski spent the winter and spring in Vienna.

1935/36 2 Theory of school algebra

1 Exercises on the theory of algebra
2 Methodology of deductive sciences
1 Logic and methodology exercises

1936/37 2 Theory of arithmetic

2 Methodology of deductive sciences Described in Hi 1971, 234.
1 Logic and methodology exercises

1937/38 2 Teaching elementary mathematics

2 Methodology of mathematics The course title has changed slightly.
1 Logic and methodology exercises

1938/39 2 Teaching mathematics

1 Exercises in elementary math.
2 Methodology of mathematics
1 Logic and methodology exercises
192 9 Career and Family

During 1930–1934 Tarski’s research seminars were described as conducted under

ukasiewicz’s supervision. In January 1934, Tarski was appointed adjunkt in the
philosophy faculty, and those notations were discontinued. Tarski’s presentations of
courses here named seminar, exercises, or discussion probably did not precisely reflect
the distinction in course names, which would have been assigned administratively. They
certainly involved discussion and individual guidance, probably at several levels—from
fundamental exercises in elementary logic courses to problems under investigation by
doctoral-research students. The latter were called travaux dirigés (directed studies) in
Jadacki 2003a.

Tarski offered the course titled Zarys metodologji nauk dedukcyjnych for six years.
He adapted this title for the English translation of his logic text: Introduction to Logic
and to the Methodology of the Deductive Sciences ([1936] 1995). That translated title is
used in the teaching-load table. Tarski probably began each year lecturing at the elemen-
tary level of this marvelously clear text, appealing to a general audience. But after three
trimesters, he was introducing his newly sophisticated students to recent research of the
Polish school of logicians. According to one student, not then at research level, who
attended Tarski’s 1937 general Methodology of Deductive Sciences course,
Almost everything that he said during those lectures was his own, new and surprising. Today
it belongs to the classic material that forms the foundations of metamathematics.49

Evidently, Tarski’s guideline for that course, and probably for his entire career in
Poland, was the principle articulated at the start of the present section: current work
in foundations of mathematics is so momentous that it “must exert a profound influence
on the exposition of the whole of mathematics, from the lowest to the highest levels of
teaching.” 50 For Tarski, there was no clear border between guiding research in logic and
mathematics and offering new results for consideration and use by his community.

During the summers of 1929, 1930, and 1931, Tarski combined teaching and recreation:
he gave lectures during vacation courses for qualified future and in-service secondary-
school mathematics teachers. Sponsored by the Ministry of Religious Denominations and
Public Education, the courses took place, respectively, in Warsaw; in the Baltic resort
city of Puck; and in Nowy Targ, a small city in mountainous southern Poland. Organized
by Stefan Straszewicz, they lasted approximately two weeks in July, and featured many
lectures by leading Polish mathematicians. Their stated goal was
to acquaint students with the modern scientific views of basic mathematical concepts related
to the teaching of school subjects.

The 1930 course in Puck, about which the most detail is available, seems typical. The
following account reveals the high intellectual level of these courses, and reflects the
seriousness with which the Polish government and academic establishment regarded their
task of incorporating recent scientific advances into public education and awareness.51

49
Hi 1971, 234. Henryk Hi later became a noted professor of linguistics.
50
Kwietniewski and Wojtowicz 1914.
51
Rusiecki and Straszewicz 1930–1931; the previous quotation is from page 111.
9.3 Teaching Teachers 193

The course at Puck extended for three or four weeks during July 1930:

Speaker Hours Subject

Wacaw Sierpiski 14 Real analysis
Stanisaw Ruziewicz 14 Algebra
Otton Nikodym 20 Mathematics teaching
Zygmunt Szulczyski 6 Educational psychology
Alfred Tarski 14 Arithmetic or geometry (?)
Stefan Straszewicz 20 Geometry
Stefan Straszewicz 8 Discussions

The lectures were described as covering elements of a subject from an advanced viewpoint,
or as covering selected topics, and they included time for exercises. Sierpiski, the leading
Polish mathematician, is an important figure in the present book. Ruziewicz, a professor
at Lwów, had been his student before World War I. They lectured for two hours a day
during the first and second weeks, respectively. A noted researcher in analysis, Nikodym
published that same year a large work on teaching. Szulczyski was a ministry official.
Tarski was added to the program after its first announcement; his topic is not known,
but in the other two summer courses he lectured on arithmetic and on geometry. Those
other courses totaled 105 and 88 hours; this one would total 96 if Tarski also lectured
for 14 hours. There were forty students, including six women. Some students and speak-
ers brought their families. One participant reported that there were ample opportunities
for seaside recreation, a weekend guided tour of the nearby free city of Danzig with its
monumental brick architecture dating from Hanseatic times, and a tour of an enormous
construction project, the new Polish port at Gdynia.52

The previous discussion has concentrated on Tarski’s effort in training teachers. As

just noted, this effort was intertwined with that for his research-oriented courses. The
latter had a profound effect on the worlds of philosophy and mathematical research. That
aspect of Tarski’s career is not the focus of this book, because his work in logic and
mathematics has been relatively accessible and has become well known. However,
entirely omitting an account of the research aspect of Tarski’s teaching at the University
of Warsaw would be awkward and misleading. Therefore, the following section 9.4 will
sketch this aspect of Tarski’s career. Section 9.5 will return to his work in mathematics
education.

52
Nikodym 1930–1937, volume 1. Rusiecki and Straszewicz 1930–1931, volume 1(6), 224.
194 9 Career and Family

9.4 Research Seminars

Background for Alfred Tarski’s early research as a student at the University of Warsaw
was described in chapters 1 and 3. Section 9.3 emphasized his work in presenting math-
ematics and logic to general audiences, including future and in-service teachers, at the
university and during summers. But about half of the courses that he taught at the uni-
versity were devoted to specialized research. Many of those were connected with his
position as research and teaching assistant to Jan ukasiewicz. As noted in section 9.3,
it is difficult to distinguish his activities with those two types of audience, and Tarski
himself probably did not do so sharply. He introduced general students to new results
that clarified the subjects they were studying. Moreover, his own research often stemmed
from questions encountered in elementary teaching. In a sketch of Tarski’s life, Jan
Zygmunt wrote,
It is obvious that all researchers and intellects of distinction use teaching to announce, above
all, their own theories, results, and opinions. But it is not always the case that teaching is a
source of research inspiration. ... Tarski was inspired by his teaching and his students. When
he taught elementary mathematics, he undertook problems from the foundations of school
geometry and arithmetic in his research; this yielded results on the equivalence of geometric
figures ... and the ... completeness of the elementary arithmetic of real numbers.53

Tarski’s University of Warsaw teaching assignments were detailed in the table on page
191. Courses whose names are not italicized there were offered for students beginning
research in logic. Tarski’s first course, in 1925/1926 on cardinal numbers, may not have
been closely related to his other teaching, but he would certainly have presented some
of his recently published results on that subject (see Tarski 1924e and 1925b).

In 1926/1927 Tarski presented a course on elementary mathematics from an advanced

viewpoint, in which he introduced his own axiom system for elementary Euclidean
geometry. He needed a modern system to provide a rigorous basis for Euclid’s classical
results. Tarski objected to the famous system that David Hilbert had introduced in 1899,
because the logical interrelationship of its primitive concepts and postulates was insuffi-
ciently clear. Instead, Tarski adapted a system of Oswald Veblen based on just three
primitive terms: point, a point’s lying between two others, and congruence of the segments
determined by two pairs of points. Using only those terms Tarski phrased a set of very
simple postulates. He then adopted the approach used by Mario Pieri to derive from these
the entire theory, providing all details of the proofs. Tarski used this system himself from
then on, in research and presumably in teaching. But for various reasons, it was not
published in any detail until three decades later. Then it became a standard for his own
and others’ extensive further study of foundations of geometry. Detailed proofs were not
published until the year Tarski died.54

53
Zygmunt 1995, xvi.
54
Tarski and Givant 1999, 175. Hilbert [1899] 1971, Veblen 1911, Pieri [1908] 2007: for a comparison of
those three systems, see section 9.3 and Marchisotto and Smith 2007, section 3.10. Tarski featured his
distinctive version of the parallel axiom in the [1931] 2014f exercise that he posed for consideration by
students and teachers: see section 12.11. Tarski [1957] 1959. Schwabhäuser, Szmielew, and Tarski 1983.
9.4 Research Seminars 195

While teaching the 1926/1927 elementary-mathematics course, Tarski was also working
on an entirely different axiomatization of Euclidean geometry based on the undefined
notions individual, part of, and sphere (in the sense of ball). That system incorporated
aspects of mereology, Stanisaw Leniewski’s part–whole theory; of general topology, also
under intense study at Warsaw; and of Mario Pieri’s axiomatization of elementary geom-
etry, mentioned in section 9.3. Tarski announced his results on this subject at the First
Polish Mathematical Congress in Lwów in September 1927; Tarski [1927] 1983 is an
enlarged English translation of his French article in the congress proceedings. Although
closely related to the material in this book, that work is not discussed further here
because it is already easily accessible.55

During the next year, 1927/1928, Tarski took charge of the seminar on problems in
logic, which included supervision of some of the students’ research. ukasiewicz had
founded the seminar a year earlier. Its exciting agenda has been described by Jan
Zygmunt in his 1991 study of one of the students, Mojesz Presburger. The seminar
participants immediately began producing significant results, and under Tarski’s guidance
continued to do so.56

The results from this seminar that are most significant for this book have to do with
elimination of quantifiers. The following example of this technique, applied to an open
sentence in elementary real arithmetic, is familiar from the quadratic formula:

} x ( x 2 + b x + c = 0) @ b2  4 c.
The existential quantification } x has been eliminated from the open sentence on the
left: it does not appear in the equivalent condition on the right. For an elementary theory
T, such as this arithmetic or Tarski’s system for elementary geometry,57 it is sometimes
possible to find a larger theory TU such that every open sentence of T is equivalent to
an open sentence of TU with no quantifiers. Theories T with this property have many
other important features. Tarski liked to formulate exercises in elementary arithmetic
or geometry that amounted to more complicated examples analogous to the one given
above. He evidently felt that by solving these, students would gain some basic familiarity
with a technique that would later play a major role in mathematics and its applications.

55
In 1948, ukasiewicz’s doctoral student Stanisaw Jakowski published a simplification of Tarski’s
presentation, possibly influenced by Tarski’s research seminar two decades earlier. See also Jakowski
1949. For further development of this subject, consult Gruszczyski and Pietruszczak 2008.
56
ukasiewicz and Tarski [1930] 1983, 38. Research reports by ukasiewicz and Tarski stirred up much
controversy at the 1927 Second Polish Philosophical Congress: see section 14.2. For another glimpse of
the milieu of logic in Warsaw at this time, see Woleski 1995a, 379.
57
An elementary language is one that can be specified using only individual variables and constants, finitary
operations on and relations between them, Boolean connectives, and the universal and existential quanti-
fiers ~ and }. A sentence is an expression of the language that, given an interpretation of its constants,
operators, and relations, is either true or false: all its variables are bound by quantifiers. An open sentence
is an expression that is a sentence or would become one if its unbound variables were bound by initial
quantifiers. (Open sentences are often called propositional functions.) A logical consequence of a set S
of sentences is a sentence that is true in any interpretation in which all members of S are true. A set
T of sentences is a theory if it contains the logical consequences of each of its subsets S.
196 9 Career and Family

Mojesz Presburger in1925

Jan ukasiewicz in 1935

The most memorable result that stemmed directly from Tarski’s 1927/1928 seminar
is Presburger’s application of quantifier elimination to the elementary arithmetic of
natural numbers with addition but without multiplication. This led to an algorithm for
deciding the truth or falsity of all sentences in that theory, and that algorithm has found
applications in both theoretical computer science and in software engineering. Presburger
presented his work in Warsaw at the First Congress of Mathematicians of Slavic Coun-
tries, 23–27 September 1929.58 ,59 By 1930, Tarski himself was able to use the technique
to devise algorithms for deciding the truth or falsity of all sentences in the richer elemen-
tary theories of real arithmetic and geometry. Many other significant new results, on
other areas of logic, originated in the 1927/1928 seminar: for example, the one now
known as the upward Skolem–Löwenheim theorem. Also this year, Tarski began present-
ing publicly his organization of the fundamental concepts of metamathematics.60

58
Presburger [1929] 1991. The Congress is described in the following section, and in Tarski’s [1929] 2014a
article, which is translated in chapter 10.
59
M ojzesz P resburger was born in Warsaw in 1904; his father soon died and he was brought up by his
mother. Mojesz entered the University of Warsaw in 1923, and attended most of Tarski’s lectures and
seminars during 1925–1930. He prepared the monographs Ajdukiewicz 1928 and ukasiewicz [1929] 1963
for publication. Having earned a master’s degree from the university in 1930, he left academia for the
insurance industry. Presburger and his wife were murdered by the Nazis. For further information about
him, consult Zygmunt 1991.
60
For various reasons, Tarski’s decision procedures were published only after many years’ delay, in [1940]
1967 and [1948] 1957. For additional information about the 1927/1928 seminar, see Fundamenta Mathe-
maticae 1934 and Mancosu, Zach, and Badesa 2009, 132–134. See also Zygmunt 2010.
9.4 Research Seminars 197

Jan ukasiewicz was born in 1878 in Lwów, which was then part of the Austrian Empire. His father
was an Army officer and his mother, the daughter of a civil servant. He was reared Roman Catholic.
Graduating from gimnazjum in 1897, ukasiewicz entered the University of Lwów to study mathe-
matics and philosophy. There he earned the doctorate in 1902 under supervision of Kazimierz Twar-
dowski. After three years’ service as tutor and library clerk and one, studying at Berlin and Louvain,
he earned the venia legendi and began teaching in the University of Lwów, offering its first course
in mathematical logic. In 1910 ukasiewicz published a noted book on Aristotelian logic, a subject
to which he returned many times. In 1911 he was appointed professor.
By 1915 ukasiewicz had published about forty books and papers. That year, he became one of
the first professors of the newly reopened Polish University of Warsaw. He held adminstrative
positions there from 1916 to 1918, when he became head of the division of higher education in the
Ministry of Religious Denominations and Public Education. In his address on that occasion he
unveiled the idea of three-valued logic, which had evolved from his earlier work on probability theory.
For the year 1919 ukasiewicz served as that minister in the cabinet of Ignacy Paderewski.
That same year, Stanisaw Leniewski, also a student of Twardowski, was appointed to a second
professorship in logic at Warsaw. Together, ukasiewicz and Leniewski founded the Warsaw school
of logic. Their most illustrious student, Alfred Tarski, earned the doctorate there in 1924. ukasie-
wicz pursued very fundamental questions. He invented prefix notation for logical operators that year,
and his [1929] 1963 introductory logic text served as a standard for many years. In 1926 he organized
a celebrated research seminar, for which Tarski soon became partially responsible; it continued until
1939 and gave rise to much of the logical methodology used worldwide during the next decades.
ukasiewicz served as university rector twice, during 1922–1923 and 1931–1932. In 1928 he
married Regina Barwiska, who stemmed from an aristocratic background.* In 1932 he invited
Heinrich Scholz, the only professor of logic in Germany, to give lectures in Warsaw; this led to
fortunate association during the next years. On Scholz’s recommendation, ukasiewicz received an
honorary doctorate from the University of Münster. The German ambassador presented it at a cere-
mony in Warsaw in December 1938. In February 1939, ukasiewicz traveled to Münster to give a
celebratory lecture. In September 1939, the Germans invaded Poland, bombed Warsaw, and
destroyed everything ukasiewicz owned.
During the German occupation, ukasiewicz worked as a translator in a government archive, and
taught clandestinely. For several years, Scholz provided financial and bureaucratic support for the
ukasiewicz couple at considerable danger to himself. Due to their backgrounds, their opposition
to Bolshevism, their association with the Polish government, and their German connection, the couple
realized that the coming Soviet invasion and conquest would threaten their lives. Scholz arranged
for them to move to Münster in July 1944 just before it, too, was destroyed, by Allied bombing.
After the armistice, ukasiewicz taught logic in a displaced-persons camp and in Brussels. In
March 1946 he accepted an appointment as lecturer at the Royal Irish Academy in Dublin. ukasie-
wicz lived there until he died in 1956. He continued research in logic, finishing the last fifteen of his
hundred-odd publications, including his second book, [1951] 1957, on Aristotelian logic.†

* In some publications, the marriage date is given incorrectly as 1939.

† For further information, consult ukasiewicz [1953] 1994, Schmidt am Busch and Wehmeier 2007, Schreiber
1999, and B. Sobociski 1956.
198 9 Career and Family

Andrzej Mostowski around 1931 Wanda Szmielew in 1938

As shown in the table on page 191, Tarski continued directing the seminar and student
research until he left Warsaw in 1939. He was officially appointed assistant to ukasie-
wicz in 1929. During the 1930s, at least six Warsaw students were conducting research
in logic under direction of ukasiewicz, with support from Tarski.61 In 1934 Tarski was
appointed adjunkt in the philosophy department; his courses were then listed independ-
ently of ukasiewicz. Two students soon began logic research directly with Tarski:
Andrzej Mostowski and Wanda Szmielew. Mostowski’s 1938 doctorate was officially
supervised by Kazimierz Kuratowski, because Tarski was not a professor. Under Tarski’s
guidance Szmielew achieved her first publishable result, [1938] 1947; she finished her
doctorate with him at Berkeley after World War II. Both Mostowski and Szmielew
became world leaders in logic research. For further information about them, see the box
on page 199. Other aspects of Tarski’s research program in the 1930s are discussed
briefly in chapters 8 and 14.

During the period 1927–1939, Tarski himself published papers on nearly fifty different
research projects in logic and mathematics; many of these must have been discussed in
detail in the seminar. Tarski’s lectures and seminar certainly played a major role in the
worldwide development of mathematical logic during those years. It is a tragedy that
virtually no detailed day-to-day record remains of them.

61
Jadacki 2003a. Scholz 1957. According to Rickey 2011, ukasiewicz was responsible for the doctorates
of Mordchaj Wajsberg (1931), Zygmunt Kobrzyski, Stanisaw Jakowski (1932), Bolesaw Sobociski
(1937), and Jerzy Supecki (1938). Czesaw Lejewski studied with ukasiewicz but received the doctorate
in London (1954).
9.4 Research Seminars 199

Andrzej Mostowski was born in 1913 in Lwów, which was then in the Austrian Empire. His father,
a medical doctor, died of typhus the next year. The family then moved to Zakopane, and in 1920 to
Warsaw, where Andrzej’s mother worked for a bank. Andrzej entered the University of Warsaw after
graduating from gimnazjum in 1931. He studied mathematics and logic with Alfred Tarski and other
noted faculty, began research closely related to Tarski’s work in set theory from a decade earlier, and
earned a master’s degree in 1936. A wealthy uncle supported his study in Vienna and Zurich for the
next two years. Returning to Warsaw, Mostowski resumed research with Tarski, extending a method
for proving independence of various principles from the axioms of set theory with Urelemente
(individuals that are not sets). Mostowski’s thesis was published in 1938, and the degree awarded
in 1939, under official supervision of Kazimierz Kuratowski, since Tarski was not a professor.
During World War II, Mostowski worked as a clerk and did clandestine teaching. In 1944 he
married a logician, Maria Matuszewska; and in late 1945, he was appointed professor at the Univer-
sity of Warsaw. Mostowski quickly became a leader there and eventually a world leader in mathemati-
cal logic. He produced many major research papers and monographs, often closely connected with
Tarski’s work. Mostowski died suddenly and prematurely in 1975.*

Wanda Szmielew was born in Warsaw in 1918 to Dawid and Bronisawa Montlak. She was schooled
in Warsaw and married Borys Szmielew soon after graduation; they were divorced in 1954. She began
studies at the University of Warsaw in 1935, particularly with Alfred Tarski and Adolf Lindenbaum.
Her early research was connected with theirs in set theory from a decade earlier. Her first result
([1938] 1947) came quickly, but its publication was delayed by World War II. During that time
Szmielew was employed as a surveyor, taught clandestinely, and continued her research. In 1945 she
obtained employment at the University and Technical Institute in ód. She completed her master’s
degree at Warsaw in 1947 and obtained a position as assistant there. Two years later she traveled
to Berkeley, where she completed research for the doctorate in 1950, the fifth one supervised by
Tarski. In her dissertation, published in 1955, she demonstrated the decidability of the elementary
theory of Abelian groups. Szmielew returned to a professorship at Warsaw, rose through the ranks,
and became the leading figure in foundations of geometry in Poland. Her later research connected
that field with set theory and the foundations of algebra. She died prematurely of a tumor, in 1976.†

* For further information, consult Krajewski and Srebrny 2008.

† ód 1946, Domoradzki 2011. For further information, consult Kordos, Moczyska, and Szczerba 1977.
200 9 Career and Family

9.5 Warsaw/Poznań Congress, 1929

Alfred Tarski’s student Mojesz Presburger presented a now famous result (see the pre-
ceding section) at the First Congress of Mathematicians of Slavic Countries, held in
Warsaw and Pozna, 23–27 September, 1929. The Congress was quite large: most of the
Polish mathematicians with roles in the present book attended. There were about 180
participants, including 35 from outside the country. Some participants are pictured on
page 201. Attendance would have been larger, had the Soviet Academy of Sciences not
declined to participate. In his opening address, at the Warsaw Polytechnic University,
Wacaw Sierpiski said,
[This is] a regional congress. There is no political substratum nor objective. It is only about
facilitating the cooperation of mathematicians living in countries not very distant from each
other, and speaking related languages.

Sierpiski’s understanding of political was naively restrictive: any such meeting would
have political import, particularly so at that time in Eastern Europe, as the Soviets
realized.62

The eight plenary sessions of the congress featured invited talks by highly distin-
guished speakers, half from Poland and half from abroad, on various subjects:
Kazimierz Kuratowski . . . . . . . . . . . . . . . . . . . . . . topology
Stefan Mazurkiewicz . . . . . . . . . . . . . . . . . . . . . . . . topology
Otton Nikodym . . . . . . . . . . . . . . . . . . . . . . . . . . . . mathematics and society
Wacaw Sierpiski . . . . . . . . . . . . . . . . . . . . . . . . . . topology
Abraham A. Fraenkel . . . . . . . Germany . . . . . . . Georg Cantor’s legacy
Leon Lichtenstein . . . . . . . . . . Germany . . . . . . . fluid mechanics
Karl Menger . . . . . . . . . . . . . . . Austria . . . . . . . . dimension theory
Kyrille Popoff . . . . . . . . . . . . . . Bulgaria . . . . . . . integration and mechanics

At least a dozen of the contributed papers were closely related to the mathematics
discussed in this book.

Tarski presented two papers:

Remarks on Some Basic Concepts of Metamathematics,
Cantor’s Hypothesis about the Alephs.63
The first had the same title as his German paper [1930] 1983, completed in March 1930.
There, Tarski indicated that this material stemmed from Tarski 1928, the report of a talk
given in Warsaw a year previously. Tarski’s congress talk was probably adapted from an
intermediate draft. His second talk was probably similar to one on the same subject that
he presented in Vienna in February 1930 and published as Tarski 1931.

62
Sierpiski [1929] 1930, 10. The complete proceedings of the congress are included in Leja 1930. Academi-
cian Nikolai N. Luzin and two other Russians were on the committee that organized the congress. But
its promise was eclipsed by events at home: during early 1929 the Soviet Academy had come under intense
political pressure and investigation for counterrevolutionary bias and activities; a purge began in August,
and by the end of the year eleven percent of its permanent staff had been dismissed (Levin 1988).
63
In Polish, Uwagi o kilku podstawowych pojciach metamatematyki and Hypoteza Cantora o alefach.
9.5 Warsaw/Pozna Congress, 1929 201

1 2 3 4 5 6 7 8 9 10 11

First Congress of Mathematicians of Slavic Countries

Warsaw Polytechnic University, September 1929 64

The Congress offered at least seven scheduled social events. For example,
On 24 September the participants were received at 17:00 by the prime minister, Dr. Kazimierz
witalski, the honorary president of the congress, at the palace of the Prime Minister;
afterward, at 20:00, they attended an evening performance at the opera, with a performance
65
of the Polish ballet Pan Twardowski.
The ballet, written during 1919–1920 by Ludomir Róycki, was based on a Polish folktale
reminiscent of the Faust legend. On the last day of the congress in Warsaw, the partici-
pants attended a reception at the city hall, and many departed at 23:00 on a sleeper train
for Pozna. The next day, after a closing ceremony at the university, they visited the
General Polish Exposition, which had been open in Pozna since May.66

64
The present editors have tentatively identified several in the front row:
1 Janina Hosiassonówna 4 Stefan Mazurkiewicz 7 Jan ukasiewicz 10 Alfred Tarski
2 Antoni M. Rusiecki 5 Kyrille Popoff 8 Leon Lichtenstein 11 Mojesz Presburger
3 Samuel Dickstein 6 Wacaw Sierpiski 9 Kazimierz Kuratowski
For other photographs of the same group, some participants changed positions. On one of those, in the
Tarski Archive at the Bancroft Library in Berkeley, Maria Tarska identified Alfred and Sierpiski.
65
Leja 1930, 28. The ballet had no connection with the philosopher Kazimierz Twardowski who figures
in the present book.
66
The journey to Pozna would have taken about five hours. See the poster and photograph on the facing
page: Jastrzbowski 1929; Powszechna Wystawa Krajowa 1929, 3.
202 9 Career and Family

General Polish Exposition, Pozna

May–September 1929

Readers are invited to compare Tarski’s impressions of these events in chapter 10 and
to note his judgment that a conference of this sort could not meet the needs of secondary
teachers, nor those whose main concern was training them:
including in the program of one and the same assembly issues of the exact sciences and of
didactics puts [didactics] at a disadvantage.

Tarski proposed as an alternative a joint conference of teachers and researchers, “devoted

exclusively to the problems of mathematics in the framework of the educational
system.” 67

67
Tarski [1929] 2014a, translated in chapter 10.
9.6 Organizations and Journals for Teachers 203

9.6 Organizations and Journals for Teachers

In Alfred Tarski’s 1929 report translated in chapter 10 of this book, he argued that
conferences oriented toward mathematical research would not meet the needs of those
teachers and scholars interested mainly in the pedagogy of mathematics. He proposed
a joint effort by both groups, targeted exclusively at pedagogical issues. Was any such
activity underway during 1925–1929?

In several other countries, societies had existed for years to foster interaction between
the research and educational communities in mathematics, for example:
Country Society, year founded Journal
Hungary . . . . . . . Bolyai János Matematikai Társulat, 1891 . . . . Matematikai és physikai lapok
Italy . . . . . . . . . . Mathesis, 1895 . . . . . . . . . . . . . . . . . . . . . . . . . . Periodico di matematica
United Kingdom The Mathematical Association, 1871 . . . . . . . . Mathematical Gazette
United States . . . Mathematical Association of America, 1915 . . American Mathematical Monthly

Each of these published a journal aimed at both communites.

A 1927 Polish teachers’ almanac listed only two organizations serving secondary-school
teachers in general throughout Poland. Neither one addressed mathematics teachers
specifically.68 The larger, with 7000 members, was the Towarzystwo Nauczycieli Szkó
rednich i Wyszych, which published the journal Przegld pedagogiczny.69 Each year
during 1921–1939 it organized a congress of Polish secondary-school teachers. The
smaller organization, with 2000 members, was the Zwizek Zawodowy Nauczycielstwa
Polskich Szkó rednich, the organization that supported the gimnazjum where Tarski
taught. It published the journal Ogniwo, which included Tarski’s [1929] 2014 report.70
In July 1930, this second organization merged with the Zwizek Polskiego Nauczycielstwa
Szkó Powszechnych, an organization of 36,000 teachers in Polish schools at all levels,
to form the Zwizek Nauczycielstwa Polskiego, which still exists. This was the organiza-
tion that invested in the housing development where the Tarski family rented and
purchased their homes from 1930–1939. These latter organizations published the jour-
nals Gos nauczycielski and Ruch pedagogiczny.71 In addition to the journals named here,
several smaller ones were published for various groups of teachers. During 1925–1929
all these journals included brief book reviews, announcements, and news about scientific
congresses and courses. These included items about mathematics, but the journals
presented little specifically about secondary-level instruction in mathematics.

68
Tomczak 1927, 247–248.
69
Society of Teachers in Secondary Schools and Universities. The journal name means Educational Review.
70
Polish Union of Secondary-School Teachers. Ogniwo means The Link.
71
Polish Teachers’ Association. The journal names mean The Voice of Teachers and Educational Trends.
204 9 Career and Family

9.7 Parametr and Mody Matematyk

The previous section noted that unlike some other countries that also led the development
of mathematics in the late 1920s, Poland had no organization specifically for mathematics
teachers. While there were some journals targeted at teachers in general, none was
devoted specifically to mathematics education. The journal Przegld matematyczno-
fizyczny, described in section 4.3, had attempted to address both the mathematics research
and education communities. Edited by Wadysaw Wojtowicz and Stefan Straszewicz, it
published Alfred Tarski’s [1924] 2014b paper on equidecomposability of plane polygons,
which is translated in chapter 5. But that journal continued for only three years.

In 1930, Straszewicz and Antoni M. Rusiecki, an official in the Ministry of Religious

Denominations and Public Education, tried again to fill the need for a journal for
mathematics teachers. They persuaded the Pozna firm Ksigarnia w. Wojciecha (St.
Wojciech’s Bookstore) to underwrite publication of a new journal, Parametr, which the
two would edit. On its first page, the publisher stated its purpose:
The genius of Polish mathematics has in recent years brought [it] to a leading position in world
science. But there are widespread complaints about the unsatisfactory results of mathematics
teaching in Polish schools. We ... recognize the need for repairs in this area of Polish life,
... [and] believe that Parametr will fill the existing gap in Polish educational publishing and
will contribute to raising the level of mathematics education in Poland. The journal will discuss
issues related to the teaching of mathematics in ... schools. It will also contain a section for
older students.

The editors explained,

The publisher, St. Wojciech, considers issuing Parametr ... a public service for the Polish
schools, and does not treat [it] as a source of income ... .

Although Rusiecki was principally employed by the ministry, it evidently played no formal
role in publishing the journal.72

Subscriptions cost fifteen zlotys per year for ten issues: about the cost of an expensive
book. The first year’s volume constituted four hundred pages. Each issue contained four
or five articles, some book reviews—often quite extensive—and lists of new books. There
were sections for professional news, miscellaneous notes, problems for readers to solve,
and solutions. Each issue but one concluded with a summary in Lingua Peano, an arti-
ficial language developed for international scientific discourse around 1900 by the mathe-
matician and linguist Giuseppe Peano.73 The ministry was impressed and encouraged by
the journal. Late in the first year of publication, it described Parametr as

72
Ksigarnia w. Wojciecha 1930; Rusiecki and Straszewicz 1930, 323. For further information about those
editors, consult the boxes on pages 208 and 210. St. Wojciech is often referred to as St. Adalbert.
73
Rusiecki 1931 is a brief introduction to Lingua Peano for readers of Polish. Peano’s 1903 term for the
language was Latino sine Flexione. It is very easy to learn!
9.7 Parametr and Mody Matematyk 205

a journal devoted to mathematics education in schools ... . It is maintained at a high level, and
contains valuable methodological advice. It can be of service to teachers of mathematics in
primary schools and secondary schools and teacher training. It should be found in every library
for school teachers.74

The editors were concerned about the balance between material that would interest
only teachers and that which might attract both teachers and older students. In five of
the nine issues of volume 1, some articles of the latter type were segregated into a section
for youth.75 But some of the articles for teachers, mostly about pedagogy, did contain
examples that might engage students. About one-third of all the articles in the 1930
volume 1 would have interested just teachers.

Beginning with the 1931/1932 volume 2 of Parametr, the editors moved its student-
oriented content, including most of the problem section, to a new journal, Mody mate-
matyk, distributed with Parametr at no additional cost.76 Parametr then contained the
material intended just for teachers. The editors actually increased the amount of material
in Mody matematyk that would interest advanced students as well as teachers. The total
number of pages in the two volumes increased by twenty percent. The journals are
remarkable for their inclusion of so much material that related mathematics to the society
in which their readers lived.

During these first years, the two journals included about one hundred articles.
Rusiecki himself wrote seventeen. Sixty-nine were contributed by authors who apparently
had no university affiliation: most of them were probably teachers. Nineteen were writ-
ten by associates of postsecondary institutions. Only six of these latter authors wrote
more than one article; Tarski contributed three. Example mastheads and tables of con-
tents are shown in the figures on page 206.

After the first year, the problem sections were apportioned rather strangely between
the two journals. Tarski contributed fourteen exercises to Parametr, volume 2, and to
Mody matematyk, volume 1: more than any other contributor except Rusiecki, the editor.
The problem sections are described in detail in chapter 12, and Tarski’s contributions
are all translated and discussed there.

74
Ministerstwo 1930.
75
Dzia dla modziey
76
A student subscription to Mody matematyk alone cost 4 zlotys; group discounts were available (Rusiecki
and Straszewicz 1930, 322).
 =

=
=

= Parametr 2(1931/1932)(8–10) :
Alfred Tarski,
The Theory of the Circumference of a Circle
in the Secondary School,
Remarks on the Degree of Equivalence of
Parametr 2(1931/1932)(2–3) supplement,
Polygons.
Mody matematyk 1(1931/1932)(3) :
Henryk Moese,
Alfred Tarski. A Contribution to Tarski’s Problem “On the
On the Degree of Equivalence of Polygons Degree of Equivalence of Polygons.”
9.7 Parametr and Mody Matematyk 207

These periodicals were very interesting, delightful to look at and read, and should have
provided a significant contribution to mathematics education. But, according to their
editors,
Mody matematyk did not find a resonance as Parametr did. Although issues were circulated
to all public and private schools, only a few schools have responded by ordering a subscription.77

The journals may have suffered from poor marketing. But production problems became
unmanageable as well. Responsibility for publication evidently lay almost entirely on
Rusiecki. He pointed this out several times in editorial notes. Two of those, late in the
third year of publication, are most poignant:
Parameter has survived a serious illness: it was not published for half a year. ... [This] is not
the slightest fault of the publisher. The entire responsibility for the lapse in production falls
on the editorial office. ... The Editor of Parametr serves in the Ministry of Religious Denomina-
tions and Public Education as a ministerial school instructor. At the beginning of the current
school year, he buckled under the workload: in September instructional visits began as well
as participation in the Second Congress of Polish Mathematicians in Vilnius; in October and
November there occurred more than twenty days of travel each month, including organization
of a two-week instructors’ course in Lwów; and finally in December there began intensive work
on a framework of mathematics programs in [public] schools ... in connection with the intended
school reform. Meanwhile, reporting activites and progress on “Mathematics for the fifth class
of elementary school” [is] to be completed.78
For the delay in the issuance of Mody matematyk we sincerely apologize. The Publisher does
not bear any guilt for this. The whole responsibility falls on the editor, who could not manage
to cope with all the current obligations on his time. The editors must admit that despite the
bitterness that they have suffered as a result of not presenting Mody matematyk [on time],
there was also a drop of sweetness: readers’ letters proved that Mody matematyk has friends.79

After the 1931–1932 volumes, the journals ceased publication for several years. Parametr
reappeared in 1939 for a brief run, until the German invasion.80

Straszewicz had already joined several others to found a new journal, Matematyka i
szkoa, published by the Towarzystwo Nauczycieli Szkó rednich i Wyszych.81 Pursuing
much the same goals as Parametr and structured similarly, it published five issues during
1938–1939.

77
Rusiecki and Straszewicz 1932b, 33.
78
Rusiecki and Straszewicz 1932b, 145.
79
Rusiecki and Straszewicz 1931–1932, 97. The last phrase has a double meaning: the journal title means
the young mathematician.
80
For further information about these journals, consult Dubiel 1975, and Dubiel 1990, section 2.2.1, 31–
34. But that report of their publication history is incorrect.
81
For further information about this journal, consult Dubiel 1990, section 2.2.2, 34–35. This organization,
the Society of Teachers in Secondary Schools and Universities, also published the journal Przegld pedago-
giczny to serve teachers of all disciplines.
208 9 Career and Family

In 1948, after the war, Rusiecki and Straszewicz collaborated again to found the journal
Matematyka: Czasopismo dla nauczycieli. It was published by the Polish Mathematical
Society on behalf of the Ministry of Education.82 The new journal’s structure and flavor
clearly descended from that of the earlier ones. Its 1949 second volume included some
solutions of exercises that had been posed in Parametr a decade or more earlier, including
one for Tarski’s rather difficult [1930] 2014c exercise83 about factoring polynomials of the
form ax 4 + b. The same volume began with a paper by Stanisaw Jakowski that inter-
preted, for an audience of teachers, Tarski’s [1927] 1983 axiomatization of Euclidean
geometry based on the notion of sphere. In an interesting way, the journal continued the
tradition of Parametr and Mody matematyk in connecting mathematics with its social
context: the editorial note Iwaszkiewicz 1949a that immediately followed Jakowski’s
paper provided a Marxist commentary! 84

Antoni Marian Rusiecki was born in 1892 in Bodzechów, a village near Kielce in southern Poland,
then part of the Russian Empire. His father was a railroad stationmaster. Antoni was schooled in
Warsaw. He attended the Polytechnic Institute in St. Petersburg, then the University of Kiev, from
which he was graduated in 1916. Rusiecki taught in secondary schools there and in Biaystok. In
1922 he became instructor of mathematics at the Ministry of Religious Denominations and Public
Education in Warsaw. His work involved development and administration, as well as teacher-training
visits throughout the country. The autobiography of the noted mathematician Mark Kac provides
a glimpse of one of those visits, which resulted in Kac’s first publication, as a student in 1931. For
a more detailed account of Rusiecki’s workload, see his lament on page 207. Simultaneously, Rusiecki
taught in several Warsaw schools. During 1930–1931 he and Stefan Straszewicz founded the journals
Parametr and Mody matematyk, devoted to teaching mathematics in the schools.* He himself
provided a substantial portion of their content, which must have required stupendous effort. Editing
these impressive journals rested too heavily on his shoulders. The journals quickly became irregular,
and disappeared during 1932–1939. During World War II, Rusiecki ran a used-book store and taught
in clandestine schools and in a rural elementary school. After returning to Warsaw, he served as
editor at two major book publishers and was a founding editor of the journal Matematyka: Czasopismo
dla nauczycieli (Journal for Teachers). In addition, Rusiecki taught methodology of teaching at the
University of Warsaw and worked in support of the Polish Mathematical Olympiad. He was author
or coauthor of numerous elementary-school mathematics texts and books on methods of teaching,
often in collaboration with Wacaw Schayer or Adam Zarzecki. Rusiecki died in Warsaw in 1956.†

* These two journals, described in detail in the present section, included the material translated in sections
7.2–7.4 and chapters 11 and 12.
† W. Piotrowski 2003; Królikowski 1991; Kac 1985, 4.

82
The title means Mathematics: Journal for Teachers. See its editor’s report, Iwaszkiewicz 1949b. This
handsome journal is still in publication.
83
Tarski’s exercise is translated and discussed in detail in section 12.3.
84
Jakowski 1949, Iwaszkiewicz 1949a. Tarski’s axiomatization is briefly described in section 9.4.
Antoni Marian Rusiecki
around 1935

Stefan Straszewicz
around 1935
210 9 Career and Family

Stefan Straszewicz was born in 1889 in Warsaw, then part of the Russian Empire. His father was
an army officer and his mother, a teacher. He finished secondary school in 1906 in Biaystok then
studied for a year with the Society of Scientific Courses* in Warsaw. He left in 1907 to study in
Zurich, returned in 1911 for a year’s military service, then taught in several Warsaw schools. In 1913
he resumed studies at the University of Zurich, where he earned the doctorate in 1914 with a
dissertation on convex geometry, supervised by Ernst Zermelo. Straszewicz remained there as a
researcher until 1919, and then served in the Polish military during the Polish–Soviet War.
For a few years starting in 1919, Straszewicz held multiple part-time teaching positions at
secondary schools, the National Pedagogical Institute (Pastwowy Instytut Pedagogiczny), the Free
Polish University (Wolnej Wszechnicy Polskiej), and the University of Warsaw. At the latter three,
he was a colleague of Alfred Tarski. At the university, Straszewicz taught a course in mathematics
for natural scientists every year from 1921 to 1934. He earned the venia legendi there in 1926, began
teaching differential geometry as well, and continued mathematical research. By 1935 he had pub-
lished about ten papers on the theories of point sets and curves. Then he decreased his university
activity, and taught only one more course there, elementary geometry, during 1936/1937.
Straszewicz had begun teaching at the Warsaw Polytechnic in 1920, was appointed professor in
its School of Civil Engineering in 1928, and was soon serving that university in administrative roles.
Straszewicz edited Przegld matematyczno-fizyczny, a journal devoted to instruction at the secon-
dary level, during its short life, 1923–1926. This Review published Tarski’s paper that is translated
in chapter 5. In 1930 Straszewicz was a founding editor of the journals Parametr and Mody
matematyk, which served both teachers and students. Described in detail in the present section, those
journals published the works of Tarski translated in sections 7.2–7.4 and chapters 11 and 12.
Unfortunately, they were issued only during 1930–1932 and 1938–1939. In 1937 Straszewicz started
another, similar journal, Matematyka i szkoa.
From 1926 until 1939, Straszewicz chaired the committee on mathematics at the Ministry of
Religious Denominations and Public Education. He became heavily involved nationwide in curricu-
lum development, teacher supervision, and teacher training. He had become a mathematical
statesman with nationwide influence.
Straszewicz coauthored numerous secondary-school texts, including a 1935 geometry book known
for its emphasis on transformations. It competed directly with the [1935] 1946 text by Zygmunt
Chwiakowski, Wacaw Schayer, and Tarski, parts of which are translated in chapter 13 of the present
book.
During World War II, Straszewicz headed the clandestine activity of the Warsaw Polytechnic.
After the war he helped rebuild it, served for three years as its vice-president, and also lectured at
the University of Warsaw. He spearheaded the Polish Mathematical Olympiad during the 1950s and
served as president of the Polish Mathematical Society. Straszewicz died in Warsaw in 1983, a
mathematical statesman of international influence.†

*Towarzystwo Kursów Naukowych, a private organization that provided the only postsecondary scientific
instruction available under the Russian oppression.
† Warsaw University 1921–1939; Piatowicz 2006.
9.8 Tarski’s Contributions to Parametr and Mody Matematyk 211

9.8 Tarski’s Contributions to Parametr and Mody Matematyk

During 1930–1932, Alfred Tarski contributed three articles to the journals Parametr and
Mody matematyk and spurred another one by Henryk Moese.85 In addition, Tarski sub-
mitted fourteen exercises to the journals’ problem sections for students and teachers to
consider. These made him one of the most significant contributors. All of these articles
and exercises are translated in sections 7.2 and 7.4 and chapters 11 and 12 of the present
book. The preceding section gave an overview of the journals. This section briefly
describes the articles, sometimes by referring to discussions elsewhere in this book.
Detailed discussions of the journals’ problem sections and Tarski’s exercises are found
in chapter 12.

Traditionally, students learned in secondary school that any two polygonal regions
W and V with the same area can be subdivided into the same number of polygonal
subregions with disjoint interiors such that all pairs of corresponding subregions of W
and V are congruent. Tarski called the smallest number of subregions for which this
is possible the degree V(W,V ) of equivalence of W and V. His paper On the Degree of
Equivalence of Polygons86 introduced this notion and some of its most general properties,
and then considered some special cases in detail. In particular, for each x > 0, Tarski
defined W( x) = V(W,V ), where W is a unit square and V is a rectangle with edge lengths
x and 1 x . He proceeded to compute some values of W, such as W(1) = 1, W(2) = 2, and
W(3) = 3, and proved that W(n)  n for every positive integer n. Tarski left as exercises
a number of details and the computation of W(4). He concluded by stating some problems
that were yet unsolved, such as whether W(n) = n for every positive integer n. The
methodology of this mathematics is elementary, within the scope of the secondary-school
curriculum. Tarski presented it to students, providing a glimpse of what formulating and
researching a real mathematical problem is like.

Henryk Moese, a secondary teacher and reader of Mody matematyk, took up Tarski’s
challenging unsolved problem and proved that W( n) = n for all positive integers n, using
reasoning only slightly more complicated than Tarski’s. He settled some other questions
and stated a new conjecture that would determine by simple formulas the values W( x) for
all positive real numbers x. Moese’s work was published in the next volume of Mody
matematyk’s parent journal Parametr, which targeted mathematics teachers.87

In the same issue of Parametr, Tarski published a continuation of his first paper on
the degree of equivalence.88 He extended his own and Moese’s work to show that W( x) =
M xN or 1 + M xN for all real x  1. Moese had conjectured that the former value obtains
only when x = n + 1/p for some positive integers n and p. Tarski reported that he

85
Front matter of those journal issues is displayed on page 206 of the present book.
86
Tarski [1931] 2014a, translated in section 7.2.
87
Moese [1932] 2014, translated in section 7.3. For information about Moese, see a box and portrait in
section 7.1.
88
Tarski [1932] 2014d, translated in section 7.4.
212 9 Career and Family

had proved this for the case 1 < x  2, but claimed that the proof was too subtle for that
journal.

Detailed discussions of these three papers on the degree of equivalence are found in
sections 7.1 and 8.5 of the present book. In 1952 Izaak Wirszup drafted English transla-
tions of the papers, and the College of the University of Chicago published them as
typescripts with hand-lettered symbols, only partially edited.89 The present editors have
not been able to determine why that occurred. For decades the original articles and those
drafts were the only form in which this mathematics was accessible, and they were
difficult to find. Complete new translations are included in sections 7.2–7.4 of the
present book.

That same issue of Parametr also contained Tarski’s paper The Theory of the Circum-
ference of a Circle in the Secondary School.90 To his target audience of secondary-school
geometry teachers, he advocated an alternative to the standard definition of the circum-
ference of a circle C and derivation of its basic properties.

Tarski first presented a definition of the circumference of a circle C that he termed

usual.91 Essentially, he proved that
(a) the sequence of perimeters pn of regular polygons with n edges,
inscribed in C, is increasing;
(b) that of the perimeters qn of the corresponding circumscribed polygons
is decreasing;
(c) pn < qn for every n; and
(d) qn – pn approaches zero as n increases.
By a result in the theory of limits, already covered at this stage of the curriculum, there
is a unique number c such that pn < c < qn for every n. This number c is defined to
be the circumference of C. Tarski analyzed its dependence on the radius r of C, defined
the number S so that c = 2 Sr always, and showed how to use his intermediate results
to approximate S using trigonometric tables.

Tarski claimed that this standard argument, which he called the method of limits, has
three disadvantages:

89
Tarski and Moese 1952. For information about Wirszup, see a box in section 7.1.
90
First published in Parametr 2 (1931–1932), 257–267; translated in chapter 11.
91
The present editors have found no clear source for Tarski’s usual definition. It can be extracted easily,
however, from some elementary-geometry texts popular in earlier decades, supplemented by a precise
treatment of the theory of limits. For example, see C. Davies 1890, Book V, 149–162.
9.8 Tarski’s Contributions to Parametr and Mody Matematyk 213

• The proofs of (a) and (b) are cumbersome, involving complicated trigonometry
and algebra.92
• The regular polygons play an unnaturally special role in this definition, so
that this method does not suggest a general way to define the lengths of
other curves.
• The definition seems to use more of the theory of limits than is necessary.

To meet those objections, Tarski presented an alternative approach, which he called

the method of cuts, in the sense of Dedekind cuts:
(d) Properties of polygons already covered at this stage of the curriculum imply
that the perimeter of any simple polygon inscribed in C is smaller than the
perimeter of any polygon circumscribed about C.
(e) By the continuity principle, some number c lies between all those inscribed
perimeters and all the circumscribed perimeters.
(f) The number c will be unique if any sequences pn and qn of inscribed and
circumscribed perimeters can be found such that qn – pn approaches zero as
n increases. Instead of using the perimeters pn and qn of regular polygons
with n edges for all n as in the standard approach, Tarski suggested using
just those of the regular polygons with 2 n edges, for all n. He showed that
the trigonometry and algebra required is much simpler, and can be used in the
same way to approximate S.
Certainly this method of cuts satisfies the first objection in the previous paragraph and
alleviates the second. However, as Tarski noted, both methods use about the same
amount of the theory of limits, just in different order.
Tarski was not breaking new ground here. Steps (d) and (e) of his method of cuts were
used in the text by Federigo Enriques and Ugo Amaldi, with which Tarski was familiar.
Step (f)—doubling the number of edges of the inscribed and circumscribed polygons at
each step of the sequences—had been employed by Archimedes two millennia earlier.93

92
Tarski’s method of proof included demonstration of the inequality
sin E E tan E
< < ,
sin D D tan D
for commensurable D, E with 0( < D < E < 90(. This kind of reasoning was familiar to ancient Greek
geometers, and was handed down to modern times by historians. See Heath 1921, volume 2, 5, 276–283.
93
Enriques and Amaldi [1903] 1916, chapter 7, §§522–532. Tarski cited that text as a recommended source
of geometric lore on the first page of his [1924] 2014b paper, translated in chapter 5. See also Archimedes
[1897] 2002a.
214 9 Career and Family

9.9 Tarski’s Coauthored Textbook Geometrja

During the mid-1920s, as Alfred Tarski began his career as a geometry teacher at the
Third Boys’ Gimnazjum, there was little variety among the Polish textbooks available
for his classes. Only three had originated in Poland. To fill the need, several new ones
appeared during the next decade. In 1935 Tarski collaborated with two colleagues to
publish a textbook, Geometrja dla trzeciej klasy gimnazjalnej (Geometry for the Third
Gimnazjum Class), called simply Geometrja in this chapter. Excerpts are translated in
chapter 13.94

The senior coauthor, Zygmunt Chwiakowski, had earned a doctorate in mathematics.

He was an experienced Warsaw gimnazjum teacher, active in teachers’ organizations.
By 1925 he had begun publishing mathematics textbooks for use in elementary and
secondary schools. The junior coauthor, Wacaw Schayer, began teaching in Warsaw
elementary schools in 1927 and then earned a master’s degree in mathematics from the
University of Warsaw.95 Schayer attended Tarski’s courses Topics in Methodology and
Topics in Elementary Geometry during 1930/1931 and 1931/1932, respectively. Geometrja
was the only schoolbook that Tarski authored. But both Chwiakowski and Schayer con-
tinued publishing textbooks for elementary and secondary schools, many in collaboration
with each other and with other mathematicians mentioned in the present book. After
World War II, Schayer became a nationally prominent politician. Little is known about
Chwiakowski; for example, the present editors do not know how or when he met his
coauthors.96

Geometrja was published in Lwów by the firm Pastwowe Wydawnictwo Ksiek

Szkolnych (National Schoolbook Publisher). An advertisement97 for it is depicted on the
facing page. The text at the top says that the publisher recommended the following
gimnazjum textbooks for the academic year 1937/1938. The other books advertised are
for classes in history, French, German, Latin, and English. The prices are in zlotys.
Geometrja was republished twice, during and just after the war—for further details, see
the box on page 224.

Geometrja adhered to the official curriculum published in 1934 by the Ministry of

Religious Denominations and Public Education. That document contained guidelines
for implementing the major 1932 reform of the Polish school system: the curriculum for

94
Chwiakowski, Schayer, and Tarski [1935] 1946. Textbook availability in the 1920s was discussed in
section 9.2 of the present book.
95
Tarski’s coauthor Wacaw Schayer was a first cousin of S tanisaw S chayer (1899–1941). The latter once
taught mathematics with Chwiakowski in the same Warsaw gimnazjum, was a member of the Warsaw
Society of Sciences and Letters, and founded the Oriental Institute of the University of Warsaw
(Zagórowski 1924–1926, volume 2, 250; Konarski 1994; Jerzmanowski 2013).
96
Chodorowski et al. 2005, 31, 44. For more information on Chwiakowski and Schayer, see the boxes on
pages 219 and 221.
97
Pastwowe Wydawnictwo Ksiek Szkolnych 1937, from page 221 of the 1 September 1937 issue of the
journal Przegld pedagogiczny, published by the Society of Teachers in Secondary Schools and Universi-
ties, which was mentioned in section 9.6.
9.9 Tarski’s Coauthored Textbook Geometrja 215

gimnazjum years I–IV. Geometrja conforms so closely with the specifications that one
wonders whether its authors had a hand in formulating them. But the guidelines do not
suggest the style of presentation or details of the exercise sets: those are the province
of the textbook authors.98 Three books competing with Geometrja were published
the same year: Iwaszkiewicz 1935, omnicki 1935, and Straszewicz and Kulczycki
[1935] 1948.

Unlike many geometry textbooks, Geometrja is not self-contained. Intended only for
the third gimnazjum year, it relied on other textbooks for prerequisite material covered
in the preceding years.99 Its authors assumed that student readers of Geometrja had
already become familiar with various techniques for reasoning, calculating, measuring,
and drawing, and with some of the most basic ideas and theorems of deductive
geometry. Moreover, it left many important topics in geometry to be presented as high
points in the textbooks for the later years of gimnazjum.

 

Advertisement in Przegld pedagogiczny, 1937

98
Poland 1934, 117–123. The reform is often referred to under the name of the minister: Janusz Jdrze-
jewicz. See page 172 for a broader description. Its recommendations for the liceum years V–VI would
not be phased in until 1937. The present editors do not know which mathematicians delineated this
curriculum.
99
In 1934, the year before Geometrja appeared, a different publisher had published a similarly titled text-
book for the preceding year’s course, written by the Lwów mathematician Antoni omnicki. For more
information about him, see a box in section 4.2.
216 9 Career and Family

Among the logical notions prerequisite for Geometrja were

• definition, axiom, use of the word arbitrary;
• proof, including use of exhaustive and of mutually exclusive cases;
• reduction to previously solved problems;
• converse and equivalent statements;
• informal use of inclusion, intersection, and union of sets.
Geometrja also assumed familiarity with these features of elementary geometry:
• points, segments, and lines;
• segment arithmetic, including the axiom of Archimedes;
• oriented angles, angle arithmetic, and degree measure;
• triangles of various sorts, their congruence and anticongruence, and the
basic inequalities;
• erecting and dropping perpendiculars;
• parallels and their use in subdividing a segment into equal parts;
• polygons of various types and their basic properties;
• circles, and intersection of two circles.
No solid geometry is involved in Geometrja.100

The following paragraphs describe the contents of Geometrja: see its full translated
table of contents in the introduction to chapter 13 of the present book. The thirty-five
sections of Geometrja are divided into three parts:
I. Geometric Loci . . . . . . . . . . . . . . . . . . . . . . . §1 –§13
II. Angles in Relation to a Circle . . . . . . . . . . . §14–§23
III. Measuring Segments and Areas . . . . . . . . . §24–§35

Sections §1–§7 and §11–§13 in part I of Geometrja continue (and to some extent,
review) the presentation of very elementary topics begun in a previous geometry course.
These include
• lines and circles and their intersections, with an explicit axiom about them;
• perpendicular bisectors;
• parallel lines.
Three sections—§2, §4, and §7—are devoted to construction problems involving these
notions. Logical principles are repeatedly discussed in geometrical context. Conditions
for existence and uniqueness of solutions of construction problems are a recurring theme.

Section §3 contains another example, rather novel, of this incorporation of logic.

Considering a line g and a circle C with given radius and center, it presents in great
detail equivalent systems of exhaustive and mutually exclusive conditions for relating the
number of intersections of C with g to the relationship of the radius and the distance
from g to the center. In Geometrja, applications of this technique are called closed

100
The title of §3 of Geometrja does use the word plane, but the theory in the book would be valid even if
it stipulated that there should be just one plane.
9.9 Tarski’s Coauthored Textbook Geometrja 217

systems of theorems, and occur several times. Tarski also featured the technique in a
general context under the name Hauber’s law in his well-known logic textbook, written
at the same time.101

The three remaining sections of part I of Geometrja and all of part II can be regarded
as applications of the preceding elementary topics in specific, somewhat more involved,
situations. Sections §8–§10 are devoted to angle bisectors and the circumcircle and
incircle of a triangle; §14–§19, to arcs and chords of a circle. They present the standard
theorems on their subjects. Sections §20 and §21 consist of related construction problems.
Sections §22 and $23, on quadrilaterals inscribed in and circumscribed about a circle, and
on regular polygons, provide distinctive climaxes for this part of Geometrja.

Throughout the first two parts of Geometrja, arithmetic is done with segments, not
with their numerical lengths.102 This approach is evidently part of the prerequisite
material; the authors do not really emphasize it. Sections §1 and §2 are ambiguous in
this regard: distances could be segments or numbers. But the first sentence of §3 (section
13.3) makes it clear:
We know that the distance from a point M to a line PQ ... is the segment MN constructed
from point M perpendicular to PQ.

On the other hand, the first four sections in part III are about segments and numbers.
The first, §24, is a review of arithmetic with rational numbers and its relation to the
manipulation of commensurable segments.

Section §25 shows that the diagonal of a square is incommensurable with its edge, so
that if lengths are to be measured by numbers, irrational numbers must be used. Its
method is remarkable: from the assumption that the ratio r of diagonal to edge should
be rational, the authors derive the equation r 2 = 2 without the usual resort to the
Pythagorean theorem. Then they follow the traditional route to contradiction: no
rational fraction that represents r could be irreducible. This section and the next two
describe in some detail, but informally, the relationships between irrational numbers and
approximating rationals, and culminate with the following criterion for equality:
If for each value of m two irrational numbers J and G have the same approximations
l
__ l+1
and ____
m m
with precision 1/ m , then the numbers are equal to each other: J = G.103

Geometrja makes no attempt to define irrational numbers, but regards them as given and
introduces student readers to methods for handling them.

101
Tarski [1936] 1995, §50. For Hauber’s law see Hoormann 1971 and Hauber 1829, 265, §287.
102
With segment arithmetic, an author regards certain segments as sums of two others, and develops the
required properties of this addition without referring to numerical lengths of segments. An alternative
approach assumes that the required properties of numbers have already been established, and postulates
existence and properties of a correspondence between segments and their numerical lengths.
103
Chwiakowski, Schayer, and Tarski [1935] 1946, §27, property 2, translated in section 13.8 of the present
book.
Zygmunt
Chwiakowski
in 1918

œ Chwiakowski

Teachers at the Jan Zamoyski Gimnazjum, around 1930

9.9 Tarski’s Coauthored Textbook Geometrja 219

Zygmunt Chwiakowski, the senior author of the 1935 secondary-school text Geometrja coauthored
with Wacaw Schayer and Alfred Tarski, was born in 1884. Chwiakowski earned a doctorate, and
in 1914 published a monograph in Russian on functional equations. In 1918, when the photograph
on the facing page was taken, he was teaching mathematics at the Zofia Sierpiski girls’ gimnazjum
in Warsaw. Later, he transferred to the Stanisaw Staszic boys’ gimnazjum, which was directed by
the mathematician Jan Zydler, author of a popular secondary-school geometry text. From 1923 to
1934, Chwiakowski taught at the Jan Zamoyski boys’ gimnazjum on Smolna Street. He was well
liked there, and was described as merry, witty, and a passionate hunter. He was depicted affection-
ately by an anonymous student: the rightmost figure in the caricature reproduced on the facing page.
Chwiakowski was active in the mathematics section of the organization of specialized teachers’ circles
in Warsaw. By 1925, he had begun publishing mathematics textbooks for both elementary and
secondary schools. Simultaneously with Geometrja, and with the same publisher, he coauthored with
Schayer a textbook on elementary arithmetic and geometry. They produced at least twenty such
books during 1935–1939. In fall 1949, Chwiakowski served briefly on the committee that established
Poland’s secondary-school Mathematics Olympiad.*

*Zagórowski 1924–1926, volume 1, 123, and volume 2, 249. Chodorowski et al. 2005, 31, 44. Przegld peda-
gogiczny 1929, 1931. Dobrzyska and Olszewska 1998. Schinzel 2000, 157.

The traditional treatment of similarity occupies §28–§32 of Geometrja. The first of

these sections, entitled A Theorem of Thales, is not about the familiar Thales char-
acterization of angles inscribed in semicircles, but about the proposition that parallel
projection of one line g onto another preserves ratios of lengths of segments of g. For
commensurable segments, this follows from earlier material about parallels. For incom-
mensurable segments, equality of the ratio of two segments and that of their projections
is derived through use of the equality criterion just mentioned. This Thales theorem then
leads in §29 to the familiar AA, SAS, and SSS similarity theorems for triangles. These
sections and the next are replete with example applications and exercises. Section §32
presents a theorem that is required for the area theory that will conclude Geometrja:
If through corresponding vertices in two similar convex polygons we pass all the diagonals, then
they will divide the polygons into the same number of triangles, respectively similar and iden-
tically arranged.104

The concluding three sections of Geometrja are devoted to the area of polygons. Section
§33 merely introduces the notion of sum of polygonal regions with disjoint interiors, and
defines polygonal regions to be equivalent if they are sums of equal numbers of sub-
regions, congruent in pairs. Having established some segment as having unit length, the
next section informally assumes four area axioms:

104
Chwiakowski, Schayer, and Tarski [1935] 1946, §32.
220 9 Career and Family

• To each polygonal region corresponds a number, its area.

• The area of a square with unit edge is 1.
• Congruent polygonal regions have the same area.
• The area of a sum of polygonal regions with disjoint interiors is the sum
of their areas.
On this basis, the area of an a × 1 rectangle with rational a is computed directly. For
irrational a, it then follows from the equality criterion for irrational numbers mentioned
earlier. The area of an a × b rectangle with rational b is then computed directly, and
for irrational b, it follows again from that criterion. The formulas for areas of other
special polygons are then derived as usual. Finally, §35 shows that if r is the ratio of
corresponding edges of two similar triangles, then the ratio of their areas is r 2. The
analogous result for similar polygons in general follows from the §32 result mentioned
earlier.

Wacaw Schayer
in 1924
9.9 Tarski’s Coauthored Textbook Geometrja 221

Wacaw Maria Schayer was born in Warsaw, then a part of the Russian Empire, in 1905. His father,
a lawyer, died two years later. Later, his mother married a mining engineer. In 1918 the family
moved to Sosnowiec, an industrial city in southern Poland. During the 1920 Polish–Soviet War, at
age fifteen, Wacaw served for three months in the Polish army. After his stepfather died in 1922,
Wacaw completed his schooling in Warsaw. He entered the university in 1924 to study mathematics
and physical sciences.
Simultaneously with university studies, Schayer obtained employment as an elementary-school
teacher in the Marymont district, joined the teachers’ union, and became active in socialist politics.
In 1927 he married Genowefa Zajc. Their two daughters, Alicja and Krystyna, were born in 1931
and 1933. During 1930/1931 and 1931/1932, Schayer attended Alfred Tarski’s courses Topics in
Methodology and Topics in Elementary Geometry. In 1932 Schayer became director of the Bolesaw
Limanowski elementary school in the oliborz district, sponsored by the Workers’ Society of Friends
of Children (Robotnicze Towarzystwo Przyjació Dzieci); he also taught in the affiliated gimnazjum.
In 1935 Schayer coauthored the secondary-school textbook Geometrja with Zygmunt Chwiakowski
and Tarski. Excerpts are translated in chapter 13 of the present book. During 1935–1939 Schayer
coauthored an extensive series of elementary-school texts with Chwiakowski, and started another
series with Antoni M. Rusiecki. Schayer continued his involvement with socialist and teachers’
organizations, serving during 1938–1939 as vice-president of the “New Paths” Society for Democratic
Education, founded by Czesaw Wycech.
In September 1939 Schayer moved to Lwów, just days before the Soviet invasion of that region.
He worked there briefly as a teacher, then as editor and translator. In 1941, after the Germans seized
Soviet-occupied Poland, Schayer returned to Warsaw. By that time, the Department of Education
and Culture of the Polish underground government had taken form under Wycech’s leadership.
Under the pseudonym Wiktor Jerzmanowski (the maiden name of his paternal grandmother), Schayer
became its general secretary, and skillfully coordinated the work of many smaller organizations
throughout occupied Poland. He continued teaching at the secretly functioning Limanowski School
and helped organize the Polish People’s Party (Polskie Stronnictwo Ludowe). Schayer’s daughter
Alicja, a thirteen-year-old messenger, was killed in the 1944 Warsaw Uprising. Immediately after
the war, Wycech became minister of education in the provisional government, and Schayer served
during 1945–1947 as vice-minister for schools. Schayer was awarded the Commander’s Cross of the
Order of Polish Rebirth. His son Andrzej Jerzmanowski was born in 1946.
During the postwar era, Schayer maintained a high level of political activity and editorship. He
served as member of Parliament for the United People’s Party in 1952–1956, during 1954–1957 as
vice-minister of agriculture, and in 1958 as vice-minister for education. He died suddenly at age 54
in 1959. Since the war, he had resumed his collaboration with Rusiecki to publish school mathematics
texts; their series continued long after Schayer’s death.*

*For further information about Schayer and the underground government, consult Schayer 1924, Turkowski
1994, Karski 1944, and Redzik 2004. Turkowski reported that Schayer earned a master’s degree in 1931, but
Schayer’s university records show that he was still attending classes in Spring 1932. Some of this information
is from Jerzmanowski 2013. Andrzej Jerzmanowski is now professor of biochemistry and molecular biology
at the University of Warsaw, and a member of the Polish Academy of Sciences.
222 9 Career and Family

Two parts of the plane-geometry curriculum very closely related to the content of
Geometrja were left for inclusion in later geometry courses:
• the Pythagorean theorem;
• precise description of the use of real numbers, particularly to define the
circumference S of a unit circle and other irrational numbers needed for
geometry.
The Pythagorean theorem is easily obtainable from the area theory at the end of Geo-
metrja, and would have been an attractive culmination, but the authors opted neither to
use nor to mention it! It appeared early in the official curriculum for the following year
IV.105 Tarski himself had sketched a compatible approach to the real number system in
the article Tarski [1932] 2014e, translated in chapter 11 of the present book. That paper
related to the curriculum for year V.106

Even so truncated, Geometrja is an interesting and impressive book. Already noted

is the abundance of exercises in its earlier parts. That continues throughout: there are
fifty-five in the sections §33–§35 on area. Second, its selection of 130 figures (for a book
of 108 pages!) is impressively rich and effective. Third, its simplicity of language is start-
ling, especially when contrasted with that of Enriques and Amaldi [1903] 1916. In spite
of that simplicity, Geometrja presents details, avoids false statements, gives convincing
arguments where they are feasible, and states assumptions clearly where they are not.

These qualities of a geometry textbook make a difference to some student readers. For
example, the senior editor of the present book first studied geometry from a popular
American textbook, more comprehensive and polished than Geometrja, but written at the
same time for somewhat similar students. That exciting experience opened up the world
of mathematics for him. But it left some bad tastes. For example, that book related the
area of a region to counting squares on an underlying grid—a good idea—but did not
bother to summarize the properties of this process on which it might base the area
reckoning; nor did it discuss refining the grid. In its plan for a proof of the theorem of
Thales mentioned earlier, the American textbook may even have misled students by
suggesting that any two segments were commensurable! In fact, it never mentioned
irrational numbers at all, leaving that present editor to hear about those only by rumor
in a calculus class three years later.107 Perhaps by chance, he took the resulting discomfort
as a challenge to find further compensating triumphs in mathematics, rather than as a
disincentive for continued exploration! The authors of Geometrja did not leave that to
chance: they took painstaking care to be correct and clear.

105
Poland 1934, 160–164.
106
The curriculum for year V was to be phased into the new liceum schools in 1937. It was based on the
penultimate year of the pre-reform curriculum: Poland 1931, 66–67.
107
Stone and Mallory 1937, 343, 284.
9.9 Tarski’s Coauthored Textbook Geometrja 223

No detailed study has been made of the genesis of the curriculum standard Poland 1934
and the extent to which it was enforced in textbooks like Geometrja. No record survives
of the contributions to the content and style of that book by its individual coauthors.
Tarski’s influence seems most noticeable in three aspects. First is the frequent use and
discussion of logical principles in geometrical context. Sometimes that is explicit, as in
the discussion of closed systems of theorems in §3. Sometimes it is informal, as in the
words let us suppose that introduce area axioms in §34. These features reflect Tarski’s
concurrent authorship of his well-known [1935] 1946 logic textbook. Second is the
unusually deep and precise attention to the properties of the real number system in
§25–§27, couched in simple language appropriate for its informal introduction to serve
the need to measure all segments. Several years earlier, Tarski had sketched a compatible
approach adequate for dealing with the circumference of a circle at a somewhat more
advanced level. Finally, in Geometrja there is a flavor of persistent attention to detail,
precision, and completeness that reflects the same qualities required for solution of the
exercises for talented students that Tarski had proposed earlier in the journals Parametr
and Mody matematyk.108 Tarski did not leave the development of these talents, and
understanding mathematics, to chance.

For translation in chapter 13 of the present book, the editors selected ten sections of
Geometrja that are representative of its content and support these assessments:
§1 –§4 introduce the flavor of the text;
§25–§27 present its treatment of incommensurable segments and irrational
numbers;
§33–§35 develop the theory of polygonal areas.

The material gathered in the present book increases the accessibility of Tarski’s early
work and explains some of its relationships to the intellectual, political, and social milieu
of Poland between the world wars. The editors hope that it will spur broader investiga-
tion into the relationship of mathematics and its cultural setting during that era. That
would be particularly welcome for the connection between mathematical research and
mathematics education, as displayed by Tarski’s research on geometry and his practice
both in teaching secondary-school students and training secondary-school teachers. This
hope is a major reason for including in a single volume of selected translations works of
such contrasting mathematical sophistication as Tarski’s research on geometric decompo-
sition in Part Two and the secondary-school geometry lessons in chapter 13.

108
See Tarski [1932] 2014e, translated in section 3.3 of the present book, and the exercises translated and
discussed in 3.4.
224 9 Career and Family

Later editions of Geometrja. A second edition of the 1935 text Geometrja by Zygmunt
Chwiakowski, Wacaw Schayer, and Alfred Tarski was published in 1944 in Jerusalem by the
Ministerstwo Wyzna Religijnych i Owiaty Publicznej, Sekcji Wydawniczej Armii Polskiej na
Wschodzie (Ministry of Religious Denominations and Public Education, Publishing Section for the
Polish Army in the East). This branch of the Polish government-in-exile in Great Britain was set
up to provide services for the army recruited during 1941–1942 by General Wadysaw Anders from
Polish prisoners in the Soviet Union, and for the much larger group of civilians accompanying that
army.
The Soviets invaded eastern Poland two weeks after the German invasion from the west on
1 September 1939. They deported about 1.5 million Poles from the conquered territory to prisons
and labor camps deep in Siberia. They especially targeted middle-class Poles likely to oppose Soviet
rule. During the following two years, about half died. When the Germans invaded the Soviet Union
in June 1941, the Polish government-in-exile persuaded the British authorities to coerce the Soviets
into releasing their remaining Polish prisoners to form an army against the Germans. Polish agents
recruited about forty thousand men. The Soviets allowed them and about seventy-five thousand
civilians to leave, but provided virtually no means for them to do so. After harrowing journeys
through Central Asia to Iran, the Poles made their way to camps in Palestine and other Middle-
Eastern lands. The Soviets discontinued even this cooperation with the Poles in April 1943. The Pol-
ish Army units formed by then played significant roles in the defeat of the Germans, which was
concluded in May 1945. During this period a complex network of elementary, secondary, and trade
schools was established for the refugee children at many sites worldwide. According to Polish-
American sociologist Tadeusz Piotrowski (2004, 98), “The greatest impediment to education was the
lack of textbooks. The problem was somewhat mitigated by having them printed in Palestine and
later, in Iran.”
After the surrender, about six million Poles were stranded outside the new frontiers of Poland—
nearly 1.2 million in western Germany, most of whom had been enslaved there during the war. Most
could not travel, could not return to areas incorporated into the Soviet Union, or would not return
as Soviet rule descended over the rest of their country. To provide for displaced persons, more than
seven hundred camps were established worldwide. Many of these refugees were able to repatriate
or emigrate to another country within two years, but some remained in the camps for as long as nine
years. Polish society was reconstituted in them, and by fall 1945, about thirty-five thousand children
were attending Polish middle and secondary schools in the camps.*
The second edition of Geometrja was reprinted in 1946 in Hanover by Polski Zwizek Wychodctwa
Przymusowego (Polish Association of Forced Emigration) to serve those students.

*For further information about this history, consult N. Davies 1982, volume 2, 271–272 and chapter 20; Snyder
2010, chapter 4; T. Piotrowski 2004; and Jaroszyska-Kirchmann 2002, and 2004, chapters 1 and 2.
 10
Congress of
Mathematicians
of Slavic Countries (1929)

This chapter contains an English translation of Alfred Tarski’s [1929] 2014a article Zjazd
matematyków, which appeared in the journal Ogniwo. The journal identified the author
only as A.T. It is a brief account of the First Congress of Mathematicians of Slavic
Countries, held in Warsaw and Pozna, 23–27 September 1929. The journal’s full title
means The Link: Organ of the Trade Union of Polish Secondary-School Teachers and
Newsletter of Its Central Administration.1 For further information about the Congress,
the journal, and its milieu, consult section 9.5.

The translation is meant to be as faithful as possible to the original. Its only inten-
tional modernization is punctuation. Personal names have been adjusted to conform with
conventions of the present book. All [square] brackets in the translation enclose editorial
comments inserted for clarification, sometimes as footnotes.

1
The complete proceedings of the Congress are included in Leja 1930. The full title of the journal is Ogniwo:
organ Zwizku Zawodowego Nauczycielstwa Polskich Szkó rednich i biuletyn zarzdu gównego Z. Z.
N. P. S. r. The title of the article in the journal’s table of contents was Zjazd Matematyków Krajów
Sowiaskich; the added phrase means of Slavic Countries. Tarski claimed authorship in his 1965a
bibliography.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 225
DOI 10.1007/978-1-4939-1474-6_10, © Springer Science+Business Media New York 2014
A. T.

AN ASSEMBLY OF MATHEMATICIANS

During the days 23–27 of the previous month [September 1929], the first Congress of
Mathematicians of Slavic Countries took place inWarsaw. In the welcoming speech, deliv-
ered at the opening session of the Congress in the presence of the Minister of Religious
Denominations and Public Education2 and diplomatic representatives from Bulgaria,
Czechoslovakia, and Yugoslavia, the chairman of the executive committee, Prof. Wacaw
Sierpiski, emphasized among other things that the Congress did not in any way have
a political basis. Its goal was to establish closer intellectual relationships among research-
ers who work in the same branch of knowledge, who live in nearby territories and for
whom collaboration is made somewhat easier by the kinship of their native languages.
As a confirmation of his words Prof. Sierpiski pointed to the presence of several mathe-
maticians who came from non-Slavic countries with the goal of participating in the work
of the Congress.

Nearly one hundred participants took part in the proceedings of the Congress.
Naturally, Polish researchers predominated—from Warsaw as well as from other univer-
sity circles: from Lwów, Cracow, and Vilnius, and even from abroad, like Prof. Leon
Lichtenstein from Leipzig. In addition, a number of mathematicians came from Czecho-
slovakia, Bulgaria, and Yugoslavia, among them the eminent researcher Prof. Kyrille
Popoff from Sofia. Among the guests from non-Slavic countries should be named Prof.
W. H. Young from London, President of the International Mathematical Union; Prof.
Abraham A. Fraenkel from Kiel; Prof. Karl Menger from Vienna; Prof. Petre Sergescu
from Cluj in Romania; and even two Japanese mathematicians, Prof. Akitsugu Kawa-
guchi and Prof. Kinjiro Kunugui. Several eminent Western European mathematicians,
among them Prof. Jacques Hadamard from Paris and Prof. Leonida Tonelli from Bologna,
sent their communiqués to be read at the Congress sessions, with the goal of making
evident their affinity3 with the attendees. Striking, however, was the complete lack of
mathematicians from Soviet Russia, with whom Polish researchers remain in constant
research contact, and who participated in numbers in the first Polish Mathematical
Congress in Lwów in 1927 as well as in last year’s International Congress in Bologna, but
whom the Soviet government expressly prohibited from taking part in the present
Congress.

2
[Tarski wrote p. Ministra W. R. i O. P., which stands for pan Minister Wyzna Religijnych i Owiecienia
Publicznego. The Minister was Dr. Sawomir Czerwiski.]
3
[Tarski’s word was czno.]
10 Congress of Mathematicians of Slavic Countries (1929) 227

The proceedings of the Congress took place in general sessions and in five sections:
I. Foundations of mathematics, history, didactics of mathematics
II. Arithmetic, algebra, analysis
III. Set theory, topology, and their applications
IV. Geometry
V. Mechanics and applied mathematics.
Altogether, about eighty talks and communiques were delivered at the Congress. The
proceedings were carried out in Slavic as well as international languages, especially in
German and French. During the Congress there took place several official receptions—
for example, at the Presidium of the Council of Ministers—and social gatherings. During
the evening of 26 September a significant number of participants went on a special train
to Pozna, where the following day the Congress came to a close.4

Congress participants from abroad did not spare the organizers words of acknowledg-
ment for efficient organization as well as the high standard of the proceedings. The
Congress was one more proof of the outstanding position of Polish mathematics in the
modern scientific world.

In general, the community of teachers played a very small role in the Congress. Among
the presented talks only one, by Prof. Antoni omnicki from Lwów, carried a strict
didactic quality. In general, the experience of the most recent mathematical assemblies
reveals that including in the program of one and the same assembly issues of the exact
sciences and of didactics puts [didactics] at a disadvantage. Incidentally, that can be
foreseen a priori: for the most part, organizers of assemblies happen to be researchers—
professors in institutions of higher learning, little interested in the affairs of the educa-
tional system of the primary and secondary schools. And in addition, even those mathe-
maticians who have understanding of and interest in these affairs are inclined to take
them up less at any time in the duration of the assembly, wishing to take advantage of
a rare occasion to establish and deepen contacts with out-of-town professional colleagues.
On the other hand, every teacher of mathematics undoubtedly feels a need to discuss in
a broad group many of the problems from the realm of elementary mathematics and
mathematical didactics. Specific conditions of our educational system, related to the
frequent changes in the curriculum and the difficulties of realizing its individual sections,
strengthen and intensify this need. For all these reasons, it would be greatly desirable
to organize soon a special assembly, devoted exclusively to the problems of mathematics
in the framework of the educational system, primary and secondary, in which theoreti-
cians [and] researchers interested in these problems as well as practitioners—teachers
active in the primary and secondary schools—could participate significantly.

4
[In the original, the date of the trip to Pozna was incorrectly printed as 2 September.]
 11
Circumference
of a Circle (1932)

This chapter contains an English translation of Alfred Tarski’s [1932] 2014e article Teorja
dugoci okrgu w szkole redniej, which appeared in the October–December 1932 issue
of the journal Parametr.1 The journal’s target audience consisted of gimnazjum teachers
and their most serious students. For further information about that publication, consult
section 9.7 of the present book.

The translation is meant to be as faithful as possible to the original. Its only inten-
tional modernizations are punctuation and some changes in symbols where Tarski’s
conflict with today’s mathematical practice in English. In this paper, Tarski used the
same symbol, for example un, for a sequence u1 , u2 , u3 , ... of terms and for a typical term
of that sequence; and the same symbol, for example u( x), for a function u and a typical
value of that function. He also omitted notation such as n   that conventionally
appears under the limit symbol lim. The translation maintains his practices. Biblio-
graphic references and personal names have been adjusted to conform with conventions
of the present book. Some uses of alternative type styles for emphasis, enunciations, and
personal names have been modified. The present editors used white space to enhance
visual organization of the paper. All [square] brackets in the translation enclose editorial
comments inserted for clarification, often as footnotes.

1
This paper was reprinted in Tarski’s 1986a Collected Papers; several typographical errors were introduced
there. The Tarski 1965a and Givant 1986 bibliographies listed the paper’s title as The Theory of the Meas-
ure of the Circumference of a Circle for High School Teaching.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 229
DOI 10.1007/978-1-4939-1474-6_11, © Springer Science+Business Media New York 2014
DR. ALFRED TARSKI (Warsaw)

The Theory of the Circumference

of a Circle in the Secondary School

According to the curriculum requirements for secondary school,2 calculating the circumfer-
ence of a circle, the area of a disk, and the surface area and volume of basic solids of revo-
lution should be covered in close connection with the theory of limits, as examples of
applications of that theory. This recommendation is usually interpreted in such a way
that the circumference of a circle and the area of a disk, and so on, are defined as limits
of certain sequences; and from these definitions are derived the formulas used in the
calculation of the quantities under consideration. This way of proceeding I will call the
method of limits.

It is possible, however, to choose as a starting point such definitions that do not contain
the concept of a limit at all, but on the other hand imply a close relationship between the
quantities mentioned and the notion of continuity, characterizing them as certain cuts.3
Nevertheless, in this case, in deriving from the given definitions the formulas for calcula-
tion, it is convenient to use facts from the theory of limits. This method, which I will call
the method of cuts for short—although I will not use the term cut explicitly 4 —seems more
suitable in many respects.

The arrangement of material in the required curriculum allows facts from trigonometry
to be used while working out problems that interest us—to a wider or narrower extent
depending on the type of school. I have a suspicion that in this way one can present a
clearer and more easily accessible lecture.

In the present article I shall focus attention mainly on the theory of the circumference
of a circle. Making use of trigonometry, I shall outline a lecture about this theory
following [each of ] the two aforementioned methods; and then I shall compare both
methods from the point of view of their didactic value.

2
[The reprint of this paper in Tarski 1986a referred here to the curriculum of the seventh class of secondary
school. Before the 1932 reform, that was the penultimate year of Polish secondary schools, when students
were about seventeen years old. For more information about the Polish school system, see section 9.1.]
3
[Tarski used cut in the sense of Dedekind cut. His word was przekrój.]
4
[Tarski used here the Latin adverb explicite.]
11.1 §1 Method of Limits 231

§ 1. The Method of Limits

I assume that the class knows about the theory of limits to the extent provided in the
curriculum. The student has thus already grasped the notions of sequences, increasing
and decreasing sequences, sequences bounded above and below, and finally, convergent
sequences and limits. I also assume a familiarity with the following theorems:
1. Every convergent sequence has exactly one limit.
2. Every increasing sequence bounded from above is convergent, and its limit
is greater than all its terms; every decreasing sequence bounded from
below is convergent, and its limit is less than all its terms.
3. If all terms of a convergent sequence un are  (respectively, ) a given
number c, then lim un  c (respectively, lim un  c); if all terms of a
sequence un are = c, then lim un = c.
4. If c is any number and vn is a convergent sequence, then lim(c ± vn ) =
c ± lim vn and lim(c # vn ) = c # lim vn .
5. If un and vn are convergent sequences, then lim(un ± vn ) =
lim un ± lim vn and lim(un # vn ) = lim un # lim vn .

Later in the lecture it will be convenient to use the notion of mutually convergent
sequences:
6. We call sequences un and vn mutually convergent if
(1) un is an increasing sequence and vn , decreasing;
(2) the inequality un < vn always holds; and finally,
(3) lim(vn – un ) = 0.5

From this definition and from the theorems given above, the following conclusions,
among others, can be derived:
7. If the sequences un and vn are mutually convergent, then the sequences
are convergent [and] have a common limit, and this limit is the only
number greater than all terms of the first sequence but smaller than all
terms of the second.
8. If c > 0 and sequences un and vn are mutually convergent, then the
sequences c # un and c # vn are mutually convergent.

Last, the following lemma will be necessary for us:

5
Instead of mutually convergent sequences, one usually speaks of convergent sequences. However, because
of the double meaning of the word convergent, such terminology can be a source of misunderstanding. This
is also why in the present article I use the term convergent only to describe individual sequences having
a limit, and not as an expression of dependency between two sequences. [Tarski’s terms for convergent
and mutually convergent were zbieny and wspózbieny.]
232 11 Circumference of a Circle (1932)

9. If k < l and un is an increasing (respectively, decreasing) sequence, then

the arithmetic mean of the first k terms of the sequence is less than
(respectively, greater than) the arithmetic mean of the first l terms.
Proof. We have
l # (u1 + u2 + ## # + uk ) =
k # (u1 + u2 + ## # + uk ) + (l – k) # (u1 + u2 + ## # + uk ),
(1)
k # (u1 + u2 + ## # + uk +uk+1 + ## # + ul ) =
k # (u1 + u2 + ## # + uk ) + k # (uk+1 + ## # + ul ).
The product (l – k) # (u1 + u2 + ## # + uk ) can be expressed as a sum of k # (l – k) parts,
each of which is one of the terms u1 , u2 , ... , uk :
(l – k) # (u1 + u2 + ## # + uk ) =
(u1 + u2 + ## # + uk ) + (u1 + u2 + ## # + uk ) + ## # (l – k times)
+ (u1 + u2 + ## # + uk ).
Similarly, the product k # (uk+1 + ## # + ul ) can be expressed as a sum of equally many
parts, each being one of the terms uk+1 , ... , ul :
k # (uk+1 + ## # + ul ) =
(uk+1 + ## # + ul ) + (uk+1 + ## # + ul ) + ## # (k times)
+ (uk+1 + ## # + ul ).
Since the sequence un is increasing, every part of the first sum is less than every part
of the second sum; and since the number of terms in both sums is the same,
(l – k) # (u1 + u2 + ## # + uk ) < k # (uk+1 + ## # + ul ). (2)
Adding to both sides of inequality (2) the number k # (u1 + u2 + ## # + uk ), we obtain
k # (u1 + u2 + ## # + uk ) + (l – k) # (u1 + u2 + ## # + uk ) <
k # (u1 + u2 + ## # + uk ) + k # (uk+1 + ## # + ul );
and hence, because of (1),
l # (u1 + u2 + ## # + uk ) < k # (u1 + u2 + ## # + ul ). (3)
Finally, dividing both sides of inequality (3) by k # l then simplifying, we obtain the
inequality
u1  u2  "  uk u1  u2  "  ul
< , Q. E. D.
k l

Moving on to trigonometry, I assume that the class already knows the theory of angle
measurement (in degrees, but not in radians), the definitions and variation of the main
trigonometric functions sin D, cos D, and tan D —at least to the extent of the first
quadrant—and the fundamental relationships between these functions. Moreover,
11.1 §1 Method of Limits 233

familiarity with the formulas for the functions of sums and differences of angles is desired,
as well as with the transformations based on those formulas.

Based on the definitions of limit and of the function cos D, we show without any
difficulty that
10. lim cos D / n = 1.

Further, we establish the following lemma, which incidentally merits attention by

itself:
sin E E tan E
11. If 0( < D < E < 90(, then < < .
sin D D tan D

It could be expressed in this way: in the first quadrant, the sine increases more slowly
than the angle, and the tangent, faster.

Proof. We shall prove only a particular case of the theorem, which, incidentally, com-
pletely suffices for our purposes: specifically, [the case] in which angles D and E are
commensurable, so that there should exist natural numbers k and l and an angle J
such that
k < l, (1)
D= k # J, and E = l# J. (2)
For n = 1, 2 , ... , l we set
un = sin n # J – sin(n – 1) # J . (3)
Thus, in particular, u1 = sin J, u2 = sin 2 J – sin J, and so on. We show that
The sequence un is decreasing. (4)
In fact, in accordance with (3), when n < l we have
un = sin n # J – sin(n – 1) # J = 2 sin 1/ 2 J # cos(n – 1/ 2 ) # J,
un+1 = sin(n + 1) # J – sin n # J = 2 sin 1/ 2 J # cos(n + 1/ 2 ) # J .
Since the cosine decreases in the first quadrant,
cos(n – 1/ 2 ) # J > cos(n + 1/ 2 ) # J,
which yields
2 sin 1/ 2 J # cos(n – 1/ 2 ) # J > 2 sin 1/ 2 J # cos(n + 1/ 2 ) # J ;
that is, un > un+1 .
234 11 Circumference of a Circle (1932)

Should the class not yet know the trigonometric transformations used here, we can
establish the necessary inequalities by following a path of straightforward geometric
reasoning from the definition of sine.6

According to lemma 9, from (1) and (4) it follows that

u1  u2  "  uk u1  u2  "  ul
> . (5)
k l

On the other hand, from (3) we conclude with ease that

u1 + u2 + ## # + uk = sin J + ( sin 2 J – sin J) + ## #
( sin k # J – sin(k – 1) # J ) = sin k # J
and similarly, u1 + u2 + ## # + ul = sin l # J Therefore, in view of (5),

sin k º J sin l º J
>
k l
and hence
sin l º J l
< . (6)
sin k º J k

Statements (2) and (6) immediately yield

sin E E
< .
sin D D

We similarly justify the second required inequality,

E l
< .
D k
Specifically, for n = 1, 2 , ... , l we set vn = tan n # J – tan(n – 1) # J, and we show that this
sequence is increasing—either with the help of simple trigonometric transformations,
or following a path of straightforward geometric reasoning,7 then argue further as before.
Thus, finally,
sin E E tan E
< < , Q. E. D.
sin D D tan D

With the help of the lemma above we establish theorem

6
Such reasoning, along with a suitable drawing, is contained implicitly in the article Mihuowicz 1930,
220–221.
7
Compare Mihuowicz 1930, 221–222.
11.1 §1 Method of Limits 235

12. The sequences un = n # sin 180(/ n and vn = n # tan 180(/ n , for n  3, are
mutually convergent.
Proof. Beginning with n = 3, angles
180D 180D
D= and E=
n1 n
are acute and commensurate; moreover,
E n1
D<E [and] = .
D n
Therefore, according to lemma 11,

180D 180D
sin tan
n n 1 n
D < < .
180 n 180D
sin tan
n 1 n 1
180D
Multiplying both sides of the first part of this statement by n # sin , we obtain
n1
180D 180D
n # sin < (n + 1) # sin ,
n n1
so that un < un+1 . Similarly, the second part of the statement gives vn > vn+1 . In this way,
un is an increasing sequence, whereas vn [is] decreasing. (1)

Further, we have
un = n # sin 180(/ n = n # tan 180(/ n # cos 180(/ n = vn # cos 180(/ n .
Since cos 180(/ n < 1, therefore, we always have
un < v n . (2)

According to theorem 2, the sequence vn must be convergent, since [it is] decreasing
and bounded below (for example, by the number zero). It also follows from theorems 10
and 5 that the sequence 1 – cos 180(/ n is convergent; moreover,
lim(1 – cos 180(/ n ) = 1 – lim cos 180(/ n = 1 – 1 = 0.
Thus, in accordance with theorem 4, the sequence vn # (1 – cos 180(/ n ) is also convergent.
Moreover,
lim ( vn # (1 – cos 180(/ n ) ) = lim vn # lim(1 – cos 180(/ n ) = lim vn # 0 = 0.
On the other hand
vn # (1 – cos 180(/n ) = vn – vn # cos 180(/n =
vn – n # tan 180(/ n # cos 180(/ n = vn – n # sin 180(/ n = vn – un .
236 11 Circumference of a Circle (1932)

Thus, finally,
lim(vn – un ) = 0. (3)
In view of (1)–(3) and according to definition 6, sequences un and vn are mutually
convergent, Q. E. D.

It follows from theorems 7 and 12 that the sequences un = n # sin 180(/ n and vn =
n # tan 180(/ n have a common limit. To denote it we introduce a special symbol:
13. The common limit of the mutually convergent sequences
un = n # sin 180(/ n and vn = n # tan 180(/ n,
where n  3, we call the number S.
As a direct conclusion of theorem 8 and definition 13 we obtain
14. n # sin 180(/ n < S < n # tan 180(/ n for every natural number n  3.

This last statement gives a student the means to determine the number S with fairly
great precision with the aid of ordinary four-digit tables. (The question whether knowing
approximate values for this number was necessary in the construction of the tables does
not come into play here.) 8 For example, setting n = 180, we calculate
n # sin 180(/ n > 180 # 0.01745 = 3.141 ;
for n = 90 we obtain
n # tan 180(/ n < 90 # 0.03495 < 3.146 .9
Thus,
3.141 < S < 3.146 .

Already now, we can formulate a definition for the circumference of a circle and derive
from it a formula to calculate this quantity.
15. The circumference of a circle is the common limit of the sequence of perim-
eters of all regular polygons inscribed in the circle and the sequence of
perimeters of all regular polygons circumscribed about the circle.10

The circumference of a circle is often defined as the common limit of all sequences of
perimeters of regular polygons inscribed in the circle or [of those] circumscribed about

8
[Tarski’s verb phrase was nie wchodzi tu w gr. That is, the question is not covered in this lecture.]
9
[In the four-place table in omnicki 1930, 8, for example, the approximation for sin 1( is 0.0175, which
indicates that 0.01745  sin 1(  0.01755. Were sin 1( = 0.01745, this would have been rounded down
to 0.0174 because 4 is even. Thus, one can conclude from the table that 0.01745 < sin 1(. For tan 2(
the approximation is 0.0349, which indicates that 0.03485  tan 2(  0.03495. Were tan 2( = 0.03495,
this would have been rounded up to 0.0350 because 9 is odd. Thus, according to the table, tan 2( <
0.03495 . Analogous reasoning with tan 1( would not have yielded a strict inequality.]
10
[ The proof of theorem 16 shows that this common limit exists.]
11.1 §1 Method of Limits 237

the circle, [with the additional provisos that] the first polygon in the sequence have an
arbitrary number of sides, and each subsequent one have twice as many sides as the
previous one. This definition offends with its unnaturalness, and moreover does not
display any practical or theoretical advantages in comparison with definition 15.11

The fundamental theorem in the theory of the circumference of the circle reads,
16. Every circle has a precisely defined circumference l ; if r is the length
of the radius of the circle, then l is expressed by the formula l = 2 S # r.
Proof. Let un (respectively, vn ) be the perimeter of a regular n-gon12 inscribed in the
circle (respectively, circumscribed about the circle). With the help of a picture, it is easy
to show that un = 2 r # n # sin 180(/ n, while vn = 2 r # n # tan 180(/ n. Applying theorems 8 and
12, therefore, we conclude that sequences un and vn , [being] mutually convergent, have
a common limit; moreover, in accordance with theorem 4 and definition 13,
lim un = lim vn = 2 r # lim(n # sin 180(/ n ) = 2 r # lim(n # tan 180(/ n ) = 2 S # r.
According to definition 15, exactly that common limit is the circumference of the circle:
l = 2 S# r, Q. E. D.

As a direct consequence of the above theorem we obtain

17. The circumferences of circles are directly proportional to the lengths of
their radii.

The theory of the area of a circle does not show any essential differences. We define
the area of a circle as the limit of the sequence of areas of all regular polygons inscribed
in the circle or [of those] circumscribed about the circle. Denoting by the symbols r, s,
un , and vn the length of the radius of the circle, the area of the circle, the area of the
regular n-gon inscribed in the circle, and the area of the regular n-gon circumscribed
about the circle, respectively, we demonstrate that the sequences
un = r 2 # n # sin 180(/ n # cos 180(/ n and vn = r 2 # n # tan 180(/ n
have a common limit: lim un = lim vn = S # r 2. Therefore, according to the definition,
s = S # r 2.

In a completely analogous manner we develop the theory of the surface area and
volume for basic solids of revolution: the cylinder, cone, and sphere. Concerning the
latter in particular, let us consider the solids that arise from rotating regular polygons
with 2 n sides, inscribed in a circle and circumscribed about the circle, about an axis
passing through two opposite vertices of the polygon. We denote the length of the radius
of the circle by r, the surface area of the solid resulting from rotating the inscribed
(respectively, circumscribed) polygon by un (respectively, vn ), and the volume of the

11
Jerzy Mihuowicz (1930) has already mentioned this.
12
[Tarski’s term for closed polygon with n vertices was n-kt.]
238 11 Circumference of a Circle (1932)

solid by unU (respectively, vnU ). Using the formulas for surface area and volume of the
cylinder and cone, we show that
4S º r2
un = 4 S # r 2 # cos 90(/ n , vn = D
,
cos 90 / n

4S º r3
unU = 4/ 3 S # r 3 # cos2 90(/ n , vnU = D .
3 cos 90 / n
From this, by passing to the limit, we obtain the known formulas for surface area and
volume of a sphere.

§ 2. The Method of Cuts

Facts from the theory of limits 13 are necessary here to the same extent as in the pre-
vious method; only lemma 9 drops out.

On the other hand, a reminder to the class about the axiom of continuity (due to
Richard Dedekind) is imperative. This axiom played an essential role in the previous
discussions, too: without its help we would not have been able to develop the theory of
measuring segments and angles, to prove the existence of limits of mutually convergent
sequences, nor even to inscribe in a circle regular polygons with an arbitrary number of
sides. Now, however, we will apply it in a direct manner.

For a suitable formulation of the axiom, we shall first of all make the following two
conventions:
18. We say that a set A of numbers precedes a set B of numbers if every
number in set A is less than every number in set B.
19. We say that a number c separates a set A of numbers from a set B of
numbers if every number in set A is  c and every number in set B
is  c.

The axiom takes the following form:

20. If a set A of numbers precedes a set B of numbers, then there exists at
least one number c that separates the two sets.
This statement is one of the axioms of algebra or arithmetic;14 and similarly, we accept
a completely analogous axiom in geometry, formulating it not for numbers, but for points
and segments.

13
[Tarski wrote sequences, not limits, here.]
14
If the real numbers are defined as cuts in the set of rational numbers (and are not introduced axiomati-
cally), then statement 20 forfeits its role as an axiom and becomes a theorem.
11.2 §2 Method of Cuts 239

Reasoning based directly on the axiom of continuity (and on the notion of cut, [which
is] implicitly inherent in it) is generally harder than the considerations in which results
from the theory of limits are applied. To avoid these difficulties we establish the following
theorem, which in some sense extends a bridge between both of the methods considered:
21. If a set A of numbers precedes a set B of numbers, if moreover un and
vn are two mutually convergent sequences and all terms of the first
sequence belong to set A and [those of ] the second to set B, then there
exists exactly one number separating set A from set B. This unique
number, which in addition does not belong to either of the two sets, is the
common limit of both sequences un and vn .
Proof. According to axiom 20, there exists at least one number c that separates set A
from set B. Considering definition 19, every such number must be greater than all terms
of the sequence un and less than all terms of the sequence vn . Applying theorem 7, we
come to the conclusion that the unique such number is the common limit of the sequences
un and vn , which is exactly what we were supposed to prove.

As to trigonometry, the preparatory considerations within its scope admit a certain

simplification for the method of cuts.

Instead of equation 10, we establish the following equation, whose proof does not
present greater difficulty:
10U. lim cos D/ 2 n+1 = 1.

Lemma 11, whose justification was quite troublesome, basically drops out alto-
gether; we shall need one very special case of this lemma in which E = 2 D:
If 0 < D < 45( then sin 2 D < 2 sin D and tan 2 D > 2 tan D.
The proof is based either on the identities
2 tan D
sin 2 D = 2 sin D cos D , tan 2 D = ,
1  tan 2 D
or on straightforward geometric reasoning, which is very simple.

We modify theorem 12 in the following way:

12U. The sequences un = 2 n+1 # sin 180(/ 2 n+1 and vn = 2 n+1 # tan 180(/ 2 n+1 are
mutually convergent.

The proof of this differs from the proof of theorem 12 only in that instead of lemma 11
we use 11U.
240 11 Circumference of a Circle (1932)

Definition 13 changes analogously:

13U. The common limit of the mutually convergent sequences
un = 2 n+1 # sin 180(/ 2 n+1 and vn = 2 n+1 # tan 180(/ 2 n+1
we call the number S.

Coming to the theory proper of the measure of the circumference of a circle, we accept
the following definition—already applied more than once, incidentally, in the elementary
lecture:15
15U. The circumference of a circle is the number greater than the length of
every closed simple polygon inscribed in the circle, and less than the length
of every closed polygon circumscribed about the circle.16
The question of equivalence of definitions 15 and 15U, and of 13 and 13U, does not present
any great difficulties, nor will it interest us further here.

Taking definition 15U as a starting point, we prove theorem 16—that is, the main theo-
rem of the whole theory—with the help of the following reasoning.

Let A (respectively, B) be the set of all those numbers that are lengths of closed
polygons inscribed in a circle (respectively, circumscribed about the circle). Set A
precedes set B because, as is familiar from geometry, a convex polygon contained in the
interior of another closed polygon is shorter than that polygon.17

We form two mutually convergent sequences such that all terms of the first sequence
belong to set A, and of the second, to set B. We could use here those sequences that
were mentioned in definition 15; then, however, we would have to base the proof on
theorem 12, and thus indirectly on lemmas 9 and 11. To avoid this, we construct two
other sequences, skipping infinitely many of the terms of each of the sequences mentioned
just a moment ago. We take into consideration specifically the sequences of regular
polygons inscribed in a circle (respectively, circumscribed about the circle) in which the
first is a square and each subsequent one is obtained by doubling the number of edges
of the previous one. It is easy to prove that the nth polygon in each of these sequences
has 2 n +1 edges.

15
Compare Enriques and Amaldi [1903] 1916, 207–214 [chapter VII, §520].
16
[Tarski’s words were
Dugo okrgu jest to liczba, wiksza od dugoci kadej amanej zamknitej (niezwizanej),
wpisanej w okrg, a mniejsza od dugoci kadej amanej zamknitej, opisanej na okrgu.
He followed Enriques and Amaldi [1903] 1916, chapter VII, but modified its version of that definition.
In that version, all polygons under consideration were closed and convex, and thus simple. Tarski lifted
that restriction on the circumscribed polygons.]
17
[The polygons in A must be convex: see the preceding footnote. Tarski must have intended those in
B to be simple, else the notion of interior would be problematic. For the familiar result, consult Enriques
and Amaldi [1903] 1916, 208. Tarski did not indicate how to extend that presentation to allow the
polygons in B to be nonconvex or nonsimple.]
11.2 §2 Method of Cuts 241

Denoting by un (respectively, vn ) the perimeter of the nth inscribed (respectively,

circumscribed) polygon, we obtain

un = 2 r # 2 n +1 # sin 180(/ 2 n+1 , vn = 2 r # 2 n+1 # tan 180(/ 2 n+1 .

With the help of theorems 8 and 12U, we conclude from here that sequences un and vn
are mutually convergent. In accordance with theorem 4 and definition 13U we have in
addition lim un = lim vn = 2 S # r.

Essentially, since all terms of the sequence un belong to set A, and [those of]
sequence vn to set B, it follows from theorem 21 that the common limit of both of these
sequences is the unique number separating these two sets, and hence is the circumference
l of the circle,18 in accordance with definition 15U. Therefore, l = 2 S # r. Q. E. D.

From the point of view of pure research, the difference between the two methods
considered here is minimal. The situation is different from a didactic point of view. The
following ideas ought then to be emphasized.

1. The supporting body of knowledge indispensable in the development of the theory

is simpler for the second method than for the first, since two lemmas with somewhat
complicated proofs drop out.

2. The basic definition of the circumference of a circle is logically simpler in the second
case than in the first, since it does not depend on the notion of a limit. It is more natural,
since the regular polygons do not play “privileged” roles in it. Finally, it is more instruc-
tive, since it extends without any changes to arbitrary convex curves.

3. The main theorem in the theory, which contains a formula for calculating the
circumference of a circle, is easier to derive from the first definition than from the second;
however, I do not suppose that it would be possible to speak here of a fundamental
difference in the degree of difficulty.

In all these respects, the method of cuts possesses, in my opinion, greater didactic value
in the secondary-school setting than the method of limits.

18
[The previous l was missing from the original.]
 12
Exercises Posed by Tarski

This chapter contains translations and discussions of the fourteen exercises that Alfred
Tarski published for enhancing instruction in Polish gimnazjums. The translations
appear in boxes in each of the numbered sections that follow. They are meant to be as
faithful as possible to the originals, and follow the conventions used for the longer trans-
lations in other chapters. In particular, all [square] brackets in the translations enclose
editorial comments inserted for clarification. Each individual exercise is accompanied
by a discussion by the present editors: the material outside the box. These discussions
include contextual information, descriptions of responses from students, teachers, and
other mathematicians, and full or partial solutions by the present editors. The para-
graphs immediately following this one describe general aspects of all of the exercises.

The journal Parametr was founded in 1930 to remedy some perceived inadequacies in
Polish mathematics instruction at the gimnazjum level—at the schools that prepared
students to enter university. It was edited by Antoni M. Rusiecki and Stefan Straszewicz.
Rusiecki, a government official and trainer of teachers, was evidently the executive
editor; Straszewicz was a professor at the Technical University of Warsaw. Parametr
was intended to address both gimnazjum teachers of mathematics and their best students.
That breadth proved difficult to manage; after a year, the student-oriented content was
diverted to a second new journal, Mody matematyk (Young Mathematician). These two
journals featured coordinated problem sections that contained exercises presented for
readers’ consideration, and solutions submitted by readers or crafted by the journal
editors.1

In the present book, the word problem often indicates a significant mathematical
question with no known solution, offered to challenge research mathematicians. The
word exercise indicates a question offered to help instruct a class of students or to guide
individual students to a higher level of achievement.2 The phrase problem section is con-
ventionally used to designate a collection of either sort of questions, often with discussion

1
For a broader discussion of these journals see section 9.7. For more about Rusiecki and Straszewicz, con-
sult boxes in section 9.7.
2
Tarski used the word wiczenia for routine exercises commonly included in textbooks, such as those
translated in chapter 13. He used zadania for both nonroutine exercises such as those translated in the
present chapter, and for research problems.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 243
DOI 10.1007/978-1-4939-1474-6_12, © Springer Science+Business Media New York 2014
244 12 Exercises Posed by Tarski

or solutions. In his authoritative 1986 Tarski bibliography, Steven R. Givant included

a list of problems and exercises posed by Tarski. Givant did not distinguish individually
between the nine open research problems and the fourteen exercises intended for instruc-
tion. All of the latter were originally published in Parametr or Mody matematyk, and
receive their first translations in this chapter.3 For seven of Tarski’s exercises the author
was identified in the original only by the initials A. T. The journals did not include any
table of such identifiers, but authorship was confirmed in Tarski’s 1965a bibliography.
All of Tarski’s exercises were reprinted in Polish in his Collected Papers; some typograph-
ical errors were introduced there.4

Parametr and Mody matematyk included little explicit discussion to justify their
problem sections. They simply extended a tradition already well established in other
countries. The journal editors may have considered some or all of the following goals for
the problem sections and their potential contributors:
(1) to exhibit mathematical bravado
(2) to display mathematical beauty
(3) to identify and stimulate the elite mathematics students
(4) to enhance mathematical instruction of all gimnazjum students.
Those are listed roughly in order of increasing importance, in the present editors’ view.
Goal (1) was of course not stressed, but is everpresent in mathematical publication. The
problem sections did not specifically emphasize beauty—goal (2)—but these journals were
filled with beautiful mathematics! The journal editors were serious about using substan-
tial exercises to further instruction of good students—goal (3):
... sometimes there will be harder problems ... for those undertaking a more serious study of
mathematics. May the younger readers not grumble about such problems also being printed:
many of the mathematics lovers will try to tackle [such] a problem, and many will probably
be happy to have studied solutions of these problems.5

The editors’ attention to goal (4) is revealed by their commenting repeatedly about using
exercises to further instruction, without singling out the elite group. Their frustration
with their lack of success with goals (3) and (4) is described in the following paragraphs.
This may have been partially due to their own confusion about those goals, obscured by
Rusiecki’s overcommitment. They wrote,
We draw attention to the “Problems from Advanced Teachers’ Courses,” which we regard as
instructional exercises; we intend to give their solutions only in cases where readers encourage
us to do so by sending solutions (taking into account the comments in the article Straszewicz
1930). We also present more difficult problems for the advanced readers.

3
One of the research problems, Tarski 1925a, is translated in section 4.3, page 62. The others are easily
accessible elsewhere in French or English.
4
Tarski 1965a, 3; and 1986a, volume 4, 688–693.
5
Rusiecki and Straszewicz 1931–1932, 2.
12 Exercises Posed by Tarski 245

Thus, they intended to use some of their exercises explicitly as instruction, but only if
readers were already willing to air their own solutions! 6

During 1930–1932, the journals published 218 numbered exercises and a few others.
Rusiecki, evidently the editor of the problem sections, took personal credit for 76 of them,
about 35%. Tarski contributed the second largest number: fourteen, about 6%. Three
others—two teachers and one professor—each contributed more than four exercises.
Rusiecki collected most of the rest from various examinations.7

Nearly half of these 218 exercises received point ratings according to their difficulty.
No criterion for the ratings was published, and they seem somewhat haphazard. Some
ratings were assigned, and some revised, in later issues. Of the rated exercises,
• 38% received ratings below five points
• 52% —five to fourteen
• 10% —fifteen or more.
The journal editors would award generous prizes to readers who submitted complete
solutions for specific contest exercises or for various others whose ratings should total at
least one hundred points. The prizes were vouchers for 25 zlotys or more, to be spent on
books that the recipients could select. The large prize for Tarski’s very elementary
contest exercise (translated in section 12.1) and the journal editors’ remarks about the
desired completeness of solutions suggests that they were stressing the organization of
solutions and the quality of writing.8

The journal editors repeatedly expressed their dissatisfaction with the response to the
problem sections. For example,
The editors are receiving too few solutions. This is a bad sign. In order to improve the prob-
lem section, readers are asked to send their criticism and desiderata regarding the section.
Whoever is unhappy with the problem section may send in friendly criticism.9

Nevertheless, solutions were published for 84—about 39%—of the 218 numbered exer-
cises posed during 1930–1932. Most solutions were detailed; some, even for elementary
exercises, extended over several illustrated pages. In quite a few instances all or portions
of the published solutions were attributed to individual solvers; in some, only the pro-
poser was credited. The many remaining solutions were apparently constructed by the
editor or were composites of contributed solutions. Solutions were published for only two

6
Rusiecki and Straszewicz1930. The article Straszewicz 1930 emphasized some specific criteria for com-
pleteness of a solution. For Rusiecki’s workload, consult section 9.7.
7
Those three contributors were Kazimierz Gilewicz (from Pozna), Samuel Steckel (Biaystok), and
Wadysaw Wojtowicz (Warsaw). For more about Wojtowicz, see a box in section 9.3. Exercises published
after 1932 (that is, in 1939) were not analyzed for the present discussion, because there was no time for
readers to respond to them.
8
Rusiecki and Straszewicz 1930, 229, and 1931; Rusiecki 1931a. Contemporary issues of Parametr con-
tained advertisements for recently translated novels by James Oliver Curwood, which cost about five zlotys,
and for many Polish scientific books priced well below that.
9
Rusiecki and Straszewicz1930.
246 12 Exercises Posed by Tarski

exercises rated ten or higher. During 1930 and 1931, fifty different readers submitted
correct solutions:
• 20 gimnazjum students
• 13 gimnazjum faculty
• 17 others.10

The problem sections of Parametr and Mody matematyk were mainly the work of their
editor, Rusiecki. He provided 35% of the exercises, collected many others from various
examinations, contributed very many solutions, and probably assembled many others
from readers’ contributions. Even with a substantial backlog of material, his production
of the 1930 and 1931–1932 volumes must have taken a stupendous effort. No one could
have sustained that for long, on top of a full-time government position devoted to fostering
and administering secondary education in mathematics countrywide. The journals ceased
publication after 1932. Parametr was resurrected during 1939 until the German invasion.

During 1930–1932, Tarski submitted articles and exercises for this serious, exciting,
and potentially influential journal, connected with both government and academia. He
was a major contributor and his intent was serious. It is worthwhile to investigate his
contributions to determine what that was. To what extent did he seem to be motivated
by the goals described earlier? Tarski eventually developed a notably graceful style of
presenting mathematics; in fact he achieved it at great effort.11 But he was not known
for writing or speaking about that in public. Some of his exercises stand as examples of
elegance, but that seems not to have been a major goal for the whole set of exercises.

Tarski’s intent could have been to foster the education of talented students, addressing
both them and their teachers. He would have realized from experience that these groups’
capabilities overlapped. Moreover, improvement in teaching could be stimulated by these
students’ perception and increased interest. Or, he could have intended to help ferret out
the very topmost mathematics students in the gimnazjums, to recruit them to university
studies and mathematical careers. Many of Tarski’s contributions were advanced exer-
cises, which received no further attention in the journals. That suggests that he and the
journal editors did not intend them principally to foster the development of talented
students in general, but as stimuli for the very topmost and a means of identifying them.

The present editors analyzed the difficulty of Tarski’s exercises. They devised criteria
as follows, according to their own experience as students, teachers, and mathematicians.
• Half of Tarski’s exercises seem to involve only mathematics at gimnazjum
level. The others seem slightly or well above that.

10
Rusiecki and Straszewicz 1932a.
11
See Givant 1991, 18, for an account of Tarski’s private conversations about writing, decades later.
Another colleague reported, “... the more one worked with Tarski, the result tended to look less and less
laborious. In fact Tarski would work over a mathematical presentation until it achieved an elegance and
simplicity which disguised the difficulties hidden beneath the surface” (Szczerba 1986, 910).
12 Exercises Posed by Tarski 247

• Only two apply this mathematics in a way that seems to fit clearly within
the curriculum. The subjects of the others would have been unfamiliar to
some or all students.
• Most require reasoning beyond the sophistication of even the best students.
Four of Tarski’s exercises scored low on all three grounds: the present editors designated
them elementary. Six others scored highest on at least two of these criteria: they are
advanced. The other four are intermediate. The journals themselves assigned point
ratings to eleven of the exercises; most are consistent with the present editors’ designa-
tions.12 All these ratings are given with the discussions of the exercises in the following
sections.

In Tarski’s exercises there is a major emphasis on geometry: only three of the fourteen
seem not to require any geometric reasoning. All of the geometry exercises involve sym-
metry in some way; in three, that could be called a main concern. There is an emphasis,
too, on reasoning with inequalities: eight exercises involve that in some way, and in three
of those it is the main concern. Three exercises—two elementary and one advanced—
emphasize core areas of the curriculum: quadratics and logarithms. Three more involve
subjects—integer arithmetic and the floor and ceiling functions—that may not have
appeared significantly in the curriculum; one of those was elementary.

Organization of case-ridden arguments was emphasized in nine of Tarski’s fourteen

exercises. Some of those involve almost unmanageable proliferations of cases, demanding
either bookkeeping bordering on the obsessive or intense concentration beyond the ability
of most. Three more display that feature to a lesser extent. As scholar and university
teacher, Tarski was advancing the application of formal logic, and this type of argument
provided an excellent opportunity for that. But he rarely stated that goal explicitly.
The techniques are familiar now, probably because of their widespread use in computer
science.

Five of Tarski’s exercises asked, under what conditions on some parameters b, c, ...
does a problem involving them and a variable x have a solution x? For example, his first
exercise, in section 12.1, involved the discriminant condition b2 – 4 c  0 for solvability
of a quadratic equation x2 + b x + c = 0. In general, such a problem amounts to finding
a condition P(b, c, ...) equivalent to } x Q( x, b, c, ...) for a given condition Q( x, b, c, ...)
—that is, elimination of the quantifier } x. That was also a major thread in Tarski’s
logical research, stemming from his 1927–1928 University of Warsaw research seminar.
But its general discussion even now remains limited to advanced logic courses.

Only five of Tarski’s exercises were stated with complete clarity. For seven others, the
desired form of the solution was incompletely specified. While this is a way of assessing
students’ perspicacity and assertiveness, it is also a way of forcing solvers to learn
elsewhere, from some in-group, what form a solution should take.

12
The rectangle-cutting exercise in section 12.9 seems rated somewhat too high by the journal, and the
section 12.11 exercise on the parallel postulate, too low.
248 12 Exercises Posed by Tarski

Tarski’s exercises received little attention after their 1930–1932 publication in Param-
etr and Mody matematyk. Although the journals published solutions for about 39% of
the exercises posed during 1930–1932, only three solutions appeared for exercises by
Tarski, about 14%. Two were elementary; solution of the third—see section 12.10—was
provided by a general theorem in a research paper by Henryk Moese, also published in
Parametr. The journal Matematyka: Czasopismo dla nauczycieli (Mathematics: A Jour-
nal for Teachers), began publication soon after World War II. Its editors included Rusiecki
and Straszewicz, and its problem section clearly descended from those of the earlier
journals. One of Tarski’s advanced problems, from section 12.3, was reprinted and solved
there. Two others, from 12.9 and 12.10, were reprinted in 1975 as exemplars of the
mathematics in Mody matematyk.13

What accounts for this apparent lack of impact? Tarski may have intended most of
his exercises not to enhance instruction but to identify the very top students and to
stimulate them and some of their teachers. As such, they would not have required much
follow-up. But even had the exercises succeeded according to that goal, the journals
offered no feedback. On the other hand, Tarski and the journal editors may simply have
underestimated the difficulty of his exercises. A common fault of mathematicians
successful in research is to assume that good students in general learn the same way they
did. Moreover, many of the exercises require organizational persistence beyond the
patience of young students. That might have deterred readers from considering them,
especially when they were posed alongside other exercises that were simpler and offered
better opportunity for star students to shine.

Tarski’s 1931 exercise in section 12.11 was the first publication of his version of
Euclid’s parallel postulate. Although Tarski used that form in his own research and
teaching, he did not discuss it in print again until 1957. Its role in subsequent research
on foundations of geometry is described in section 12.11.

Tarski’s exercises confirm the connection between his research in mathematics and
logic and his work as a teacher. They show that he was keenly aware of that connection,
and made serious effort to ensure that each of these aspects of his activity influenced
the other.

13
Moese [1932] 2014, translated in section 7.3; for its impact, see sections 7.5 and 8.5. Iwaszkiewicz 1949b.
Dubiel 1975.
12.1 Exercise on Diluting Wine 249

12.1 [1930] 2014a , Exercise on Diluting Wine

Tarski’s contribution Zadanie o rozcieczaniu wina to the July–August 1930 issue of

Parametr (volume 1, page 229) is translated and discussed here. Unlike the other
exercises in this chapter, this one had no number, nor was it assigned a point value.

CONTEST PROBLEM II
Exercise on Diluting Wine

A winemaker had two barrels; one had a volume of a liters,

and the second, b liters. The first barrel was filled with pure
wine but the second was empty. The winemaker poured a
certain amount of wine— x liters—from the first barrel to
the second, and filled up the second barrel with water. After
mixing the wine and water, the winemaker poured from the
second barrel to the first just enough so that the first barrel
became full. It turned out that in the first barrel was the kind
of wine that we would obtain if, into the barrel containing a
liters, we poured c liters of pure wine and filled up the re-
maining [volume] with water.
Find x and report for what values of the givens a, b, c we
have (1) two solutions, (2) one solution, (3) a problem with no
solution.
Dr. Alfred Tarski (Warsaw)

Solution.14 After mixing, the concentration of pure wine in the second barrel was
x / b . After the first barrel was refilled with the mixture, it contained

(a – x) + ( x / b) x = c
liters of pure wine. After rearranging this equation into standard quadratic form and
considering the discriminant d, we see that the three conditions correspond to the cases
(1) d > 0, (2) d = 0, and (3) d < 0. The quadratic formula yields the x values.

14
Readers are reminded that the solutions in this chapter are the present editors’ unless otherwise noted.
250 12 Exercises Posed by Tarski

Discussion. This exercise is elementary and devoted to a core subject. Its solution
involves the quadratic formula, a very familiar example of quantifier elimination. The
journal editors characterized it as thought-provoking.15

There were three other contest exercises that year.16 The first challenged readers to
plan the longest airplane flight from sunrise to sunset, entirely within Poland; in the
present editors’ view, it was too vague. The other two were stated as precisely as Tarski’s:
to create a certain type of geometric puzzle based on equidecomposable polygons, and to
find a parameter m such that exactly six pairs of positive integers x, y satisfy the equa-
tion 5 x + 3 y = m.

For the best two solutions of Tarski’s contest problem, the journal editors offered
substantial prizes. The size of the reward for solving this elementary problem indicates
that the editors were stressing the organization of solutions and the quality of writing.17
There seems to have been no provision in the journal for recognizing contributed solutions
for this exercise.

12.2 [1930] 2014b, Exercise 83: Iteration of the Absolute Value Symbol

Tarski’s exercise Nr. 83: Równanie ze znakami bezwzgldnej wartoci is translated and
discussed in this section. It was published in the July–August 1930 issue of Parametr
(volume 1, page 231).

Exercise 83. Iteration of the absolute value symbol. Prove that

a – b = a+b + a–b – a – b .

[Three points for solution.] A. T. (Warsaw)

Discussion. This exercise is elementary and emphasizes inequalities and symmetry.

The equation is equivalent to

a + b – a+b = a–b – a – b .

15
“Arcydyskusyjny”—see Rusiecki and Straszewicz 1930, 229.
16
Rusiecki 1930c, 1930d, and Kaptur 1930.
17
See the discussion on page 245.
12.3 Exercise 88: Decomposition into Factors 251

That is, the amount by which one member of the triangle inequality a + b  a + b
exceeds the other is the same as that for the related inequality a – b  a – b ,
which is often employed in real-analysis texts.

The journal recognized contributed solutions from two students of a gimnazjum in

acut, a town in southeastern Poland. One received full credit. The next year, the
journal published an unsigned solution, which treated the cases separately.18

Solution. The coordinate axes and diagonals divide the a, b-plane into the origin and
eight closed unbounded wedge-shaped regions that overlap just along those lines. The
equation obviously holds in the region where a  b  0, and is invariant under the
reflections across the boundary lines; therefore it is valid in general. While that concise
argument is convincing for mathematicians, it might not be appropriate for a gimnazjum.

12.3 [1930] 2014c, Exercise 88: Decomposition into Factors

Tarski’s exercise Nr. 88: Rozkad na czynniki is translated and discussed in this section.
It was published in the September 1930 issue of Parametr (volume 1, page 277).

Exercise 88. Decomposition into factors. What conditions

must integers a and b —both different from zero—satisfy
so that these conditions are necessary and sufficient condi-
tions to be able to express the binomial a x 4 + b in the form
of a product of polynomials of at least first degree with integer
coefficients?
[Ten points for solution.] Dr. Alfred Tarski (Warsaw)

Discussion. This exercise is advanced. It emphasizes a core subject—factoring— and

another that may not have been encountered often in schools—integer arithmetic. The
exercise was stated awkwardly and gave no clue to the form of the desired solution.19 Two
teachers each received full credit for solving this exercise: Micha Hornowski from
Warsaw20 and Wadysaw Stojda from Piotrków Trybunalski, a city in central Poland.

18
The students were Józef Burda and Bronisaw Szul. For the solution, see Rusiecki 1931b.
19
The condition “to be able to express ...” in the exercise itself satisfies the stated requirement, but that is
not what Tarski intended.
20
M icha H ornowski was born in 1893 in Russia and educated in Warsaw. From about 1920 to 1958 he
taught mathematics in Polish institutes that trained teachers. He edited, coauthored, and reviewed many
textbooks and books about mathematics. Hornowski died in Warsaw in 1966. (Piotrowski 2003, 76.)
252 12 Exercises Posed by Tarski

Solution. Soon after World War II, the exercise was reprinted in the journal Matema-
tyka: Czasopismo dla nauczycieli. It received two additional correct solutions and pub-
lished one a year later: the condition that there should exist nonzero integers k, O, and
P such that either
k O2 = a and k P2 = – b, (1)
or
k O2 = a and k P2 = b (2)
and 2OP = Q 2
for some integer Q. (3)
The following proof sketch is adapted from that source.21

First use elementary algebra to verify that condition (1) implies that the given quartic
is the product of two quadratics with integer coefficients, and that the conjunction (2) &
(3) also implies this. Next, show that (1) is equivalent to the condition that
– a/b should be the square of a rational number.22 (1U)

Now suppose that the given quartic is the product of two quadratics with integer
coefficients D, ... , K :
a x 4 + b = ( Dx 2  Ex  J)( Gx 2  Hx  K). (4)
Let
O = gcd( D , G), k = DG/ O2, P = gcd( J , K).
By comparing coefficients of the members of equation (4), one sees that E = 0 if and
only if H = 0: that is, either
E = 0 = H or E, H =
/ 0. (5)
By a brief argument, the first alternative implies condition (1), and thus also (1U).
Similarly, the second alternative of (5) implies condition (2). Moreover, comparing
coefficients also yields 0 = EG  DH . The integers D = D / O and G = G / O are relatively
prime; since ODH =  OEG and O is nonzero, D divides E. That is, there exists an
integer Q such that E = Q D. Another brief argument yields condition (3).23

21
See Iwaszkiewicz 1949b. Stojda was given priority for the solution. The postwar solutions were by
Tadeusz Czarliski and Mieczysaw Warmus. Born in Breslau in 1918, M ieczysaw W armus became a
noted professor of mathematics and computer science; he died in Australia in 2007.
22
Moreover, the conjunction (2) & (3) is equivalent to the simpler condition that 4 a/b should be the fourth
power of some rational number; but that result is not used in this argument.
23
Readers are invited to complete the “brief” arguments mentioned in this paragraph. The first two rest
on showing that D and the ratio J = J / P divide each other, so that J = ± D. The third uses elementary
algebra to derive ± 2 O P = Q 2. The two ± signs must therefore be +.
12.4 Exercise 102: Interesting Identity 253

Finally, suppose that the given quartic Q( x) has a linear factor c x + d with integer
coefficients. Since Q( x) is even, it has roots ± d/c, and thus Q( x) = q( x) (c2 x 2 – d 2 ) for
some quadratic q( x) with rational coefficients. Let m be the least common multiple
of their denominators, so that the quartic mQ( x) is the product of two quadratics with
integer coefficients, which must fall under the first alternative of (5). Therefore,
mQ( x) must satisfy condition (1U): – (ma)/(mb) should be the square of a rational
number. But that ratio is the same as – a/b, so (1U), and hence condition (1), holds for
the given quartic Q( x).

12.4 [1930] 2014d, Exercise 102: Interesting Identity

Tarski’s exercise Nr. 102: Ciekawa tosamo is translated and discussed in this section.
It was published in the September 1930 issue of Parametr (volume 1, page 278).

Exercise 102. Interesting Identity. Show that for every natu-

ral number n  2 this identity holds:
n n

¦ E log
g 2
g n ¦E
k 2
k
n.

Remark. For a real number x, we denote by the symbol

E x the whole number p satisfying the conditions p  x <
p + 1. The symbol E is French shorthand for the word
entier; we read the symbol E x as the entirety of x.
Dr. Alfred Tarski (Warsaw)

Discussion. This exercise is intermediate in level. While it emphasizes core sub-

jects—logarithms, roots, and notation—it provides a taste of symmetry, inequalities, and
counting, and serves to introduce the floor function.24 Its elementary nature is disguised
by the bravado of its notation. But Tarski slyly gave a subtle hint by using different
indices with the sums. Two readers were recognized for contributing solutions: Perez
Halamon, from Kalisz, a city in central Poland, and Wadysaw Stojda. No solution was
published.

24
In this exercise, E x stands for a value O xP of the floor function.
254 12 Exercises Posed by Tarski

Solution. The decreasing function g ” log g n, graphed below in figure 1, maps the
interval [ n , n] to [1, n]. Its inverse is the decreasing function k ” k n . The first
n

and second sums in the exercise are the numbers of lattice points in the regions G and
K, hatched and in the figure, respectively, not including the points on the axes.
There are exactly n – 1 lattice points in each of the regions G – K and K – G, all on
the lines with equations k = 1 and g = 1, respectively. Thus, G and K contain the
same number of lattice points, and the sums are equal.

k = log g n
k
g= n

log 2 n

2
1
g
2 n n

Figure 1 (for Section 12.4)

12.5 [1930] 2014e: Exercise 118 255

12.5 [1930 ] 2014e , Exercise 118

Tarski’s exercise Nr. 118 is translated and discussed here. It was published in the
October 1930 issue of Parametr (volume 1, page 318). Unlike the other exercises in this
chapter, it had no descriptive title. It was included among “problems associated with the
content of articles”—probably selected as related to the paper Cwojdziski 1930 on
sequences, in the section of the same issue that targeted student readers.

Exercise 118. A sequence an was defined in the following

way:
2
if n is an odd number, then an = , and
n 1
2
if n is an even number, then an =  .
n
Construct a formula an = F(n) that would give all the terms
of the sequence.
A. T. (Warsaw)

Discussion. This exercise is elementary, but touches on a subject perhaps less

commonly encountered in schools—integer arithmetic. Teodor Hrycak, a teacher and
engineer from Stanisawów, a district in southeastern Poland, contributed a solution, but
none was published.

Solution. A glance at the previous exercise posed by Tarski, in section 12.4, would
suggest using the floor or ceiling operator. Here are two sample solutions, one with and
one without:

(1)n1 4
an = an = .
ǻ n / 2ȼ 1  (1)n (2n  1)
256 12 Exercises Posed by Tarski

12.6 [1930] 2014f, Exercise 136: Equidistant

Tarski’s exercise Nr. 136: Ekwidystanta is translated and discussed in this section. It was
published in the November–December 1930 issue of Parametr (volume 1, page 398).

25
Exercise 136. Equidistant. Find the set of all points whose
distance from the boundary of a given rectangle with edges
a and b is equal to a given segment of length r.

[Four points for solution.] A. T. (Warsaw)

Discussion. This exercise is elementary, with a taste of symmetry, inequalities, and

argument by cases. The word find (wyznaczy) in the exercise statement was unfortu-
nately vague. No solution by a reader was reported, but the journal published a solution
several years later.26 The journal editor asked a nice question: what is the analogous
result in three dimensions?

Solution. Illustrated below by figure 2 for the case a < b and r < a/ 2 , the desired
curve consists of an internal rectangle and an external one whose corners have been
replaced by quarter circles centered at the vertices of the original rectangle as indicated.
Tedious arguments will show that the points on this locus are exactly those desired.

When a < b and r = a/ 2 , the inner rectangle becomes a line segment; when a = b
and r = a/ 2 it becomes a point. When r > a/ 2 the locus has no interior component.

Figure 2 (for Section 12.6)

Equidistant Curve

25
In English the exercise title is an adjective used nominally to refer to the equidistant curve that the
exercise describes. The Polish title is a noun.
26
Rusiecki 1939, which was incorrectly labeled as a solution to a different exercise.
12.7 Exercise 167: System of Inequalities 257

12.7 [1931] 2014b, Exercise 167: System of Inequalities

Tarski’s exercise Nr. 167: Ukad nierównoci is translated and discussed in this section.
It was published in the February 1931 issue of Mody matematyk ( volume 1, page 46),
distributed with Parametr, volume 2.

Exercise 167. System of inequalities. Investigate what condi-

tions numbers k, l, m must satisfy so that there should exist
an angle M in the first quadrant (0  M  1/2 S) satisfying the
following two inequalities simultaneously:
k sin M + l cos M  m
k cos M + l sin M  m.
Ten points for solution. A. Tarski (Warsaw)

Discussion. This exercise and its companion, exercise 168 in the next section, were
identified as biproducts of Tarski’s research on a question that he posed in his [1931]
2014a paper on the degree of equivalence of polygons, translated in section 7.2. That
question was repeated in his exercise 170, translated in section 12.10. Shortly afterward,
Tarski’s original question was answered by an entirely different method in the paper
Moese [1931] 2014, translated in section 7.3.

Thus, this exercise’s interest lies mainly in its emphasis on types of reasoning that
Tarski considered important for the next generation of mathematicians. It is advanced,
and emphasizes symmetries and inequalities. Their relation to an asymmetric fea-
ture—specification of first quadrant—is particularly engaging. The exercise also makes
the investigator confront the need to reformulate arguments to reduce the great number
of cases involved. The word conditions in the statement of the exercise should probably
have been explained, to make it clear that this is an exercise in quantifier elimination.
No solutions by readers were reported, and no solution was published.

Partial solution. The following discussion assumes that k, l, k ± l, and m are

nonzero. Figure 3, on the next page, uses a Cartesian x, y coordinate system with origin
O to display lines with equations as follows:
x, y intercepts
g : kx + ly = m, m/ k , m/l ,
gU : l x + k y = m , m/ l , m/ k .
258 12 Exercises Posed by Tarski

The picture is symmetric with respect to the main diagonal and the lines intersect at point
m m m
P= , with distance OP = 2.
kl kl kl
The figure was drawn for the case in which P lies in the first quadrant: its coordinates
are positive.

m/k

m/ l
O
1

O
P

m/l
x
O
O O
1 m/k

Figure 3 (for Section 12.7)

The set of points  x, y that satisfy the simultaneous inequalities

kx + ly  m
 lx + ky  m
is one of the two hatched regions in the figure. It is the one containing O just in case
m > 0. The inequalities in Tarski’s exercise can be formulated as the two just shown, plus
the condition that  x, y lie on the first quadrant of the unit circle U , as shown in the
figure for a case in which OP < 1.

Should P fall in the third quadrant, Tarski’s inequalities will have a solution if and
only if m > 0. Should it lie in the first, there will be a solution just when m > 0 and
P or the intercept nearer O lies on or outside U , or when m < 0 and P lies on or
inside U . Therefore, under the listed assumptions, Tarski’s inequalities have a solution
if and only if
12.8 Exercise 168: Another System of Inequalities 259

m
• < 0 and m > 0; or
kl
m m m m
• > 0 and m > 0, and 2  1 or  1 or  1; or
kl kl k l
m m
• > 0 and m < 0, and 2  1.
kl kl

These conditions can be reformulated as Boolean combinations of linear inequalities

involving k, l, and m.

Readers of the present book may wish to complete this analysis to accommodate the
cases in which one or more of k, l, k ± l, and m is zero, and to simplify these
conditions.

12.8 [1931] 2014 c , Exercise 168: Another System of Inequalities

Tarski’s exercise Nr. 168: Ukad nierównoci is translated and discussed in this section.
It was published in the February 1931 issue of Mody matematyk (volume 1, page 46),
distributed with Parametr, volume 2.

Exercise 168. System of inequalities. Investigate what

conditions numbers k and l must satisfy so that the follow-
ing three inequalities should possess at least one common
solution:
(k + 1) x2 – 2 l x + (k – 1)  0
(l + 1) x2 – 2 k x + ( l – 1)  0
x2 – x  0 .
Indicate the relationship between this exercise and the
previous exercise.
Ten points for solution. A. Tarski (Warsaw)

Discussion. Background for this exercise was discussed with that of its companion,
exercise 167 in the previous section.
260 12 Exercises Posed by Tarski

This exercise, too, is advanced. It is ostensibly about inequalities and quadratics, and
quickly evolves into reasoning about many cases. Further investigation into its relation-
ship with the exercise in the preceding section might reveal a geometric connection. Like
that one, its statement could have been clarified to make it clear that this is an exercise
in quantifier elimination. No solutions by readers were reported, and no solution was
published.

Partial solution. Tarski’s system of three inequalities can be solved by straight-

forward but tedious algebraic manipulation, outlined as follows. It is assumed that
constants k, l =/ –1.

The Cartesian graphs of the first two quadratics are parabolas that open upward or
downward depending on whether the leading coefficients are positive or negative. Each
has two roots, one, or none, depending on whether its discriminant 4(l 2 – k2 +1) or
4(k2 – l 2 +1) is positive, zero, or negative. The sum of the discriminants is 8. One or
both of those quadratics must have two roots: otherwise, that sum would be negative or
zero. If (a) one or both parabolas should open downward, the set X of solutions of
Tarski’s first two inequalities will be empty, a single isolated point, two isolated points,
or a finite interval, depending on the linear order of the roots. If (b) both open upward,
X is the union of two or three disjoint intervals, two of which are infinite, depending on
the linear order of their roots. The isolated points and interval endpoints mentioned in
the preceding two sentences are roots of the quadratics.

Tarski’s third inequality merely describes the points x in the closed unit interval U.
Thus a solution of the system of all three inequalities exists just in case
a. one or both parabolas open downward, X is nonempty, and U intersects
one of the intervals described, or
b. both open upward and U intersects one of the intervals described.
The conditions describing the opening of the parabolas, whether X is empty in case (a),
and the order of the roots of the quadratics and endpoints of the unit interval, can all be
expressed as inequalities involving k and l. Thus, the existence of a solution of Tarski’s
system of three inequalities involving k, l, x is equivalent to a Boolean combination of
these inequalities involving k and l.

Readers of the present book may wish to construct these inequalities, simplify them
into manageable form, and then investigate how they can be reformulated, if necessary,
to accommodate the additional cases in which k or l equals –1. Moreover, the outline
just presented does not address the second part of Tarski’s exercise: how is this set of
inequalities related to those in the previous exercise 167 in section 12.7? That, too, is left
for future consideration.
12.9 Exercise 169: Cutting a Rectangle Out of a Square 261

12.9 [1931] 2014d, Exercise 169: Cutting a Rectangle Out of a Square

Tarski’s exercise Nr. 169: Wycicie prostokta z kwadratu is translated and discussed in
this section. It was published in the February 1931 issue of Mody matematyk (volume
1, page 46), distributed with Parametr, volume 2.

Exercise 169. Cutting a rectangle out of a square. With

regard to exercise 1 posed in issue 1 of Parametr from 1930,
and to the letter to the editor published in issue 5 of that year,27
prove the following assertion:
So that from a square with edge a it would be
possible to cut out a rectangle with edges b and
c, it is necessary and sufficient that either b  a
and c  a, or that b + c  a 2 .
Fifteen points for solution. A. Tarski (Warsaw)

Discussion. The assertion in this exercise was stated awkwardly in the original.
Exercise 1 in issue 1 asked for a general method to cut as many rectangular pieces as
possible, each with edges b and c, from a given rectangular sheet of paper (at least for
the case in which the edges of the rectangles are parallel to that of the sheet). In the cited
letter to the editor, Wodzimierz Krysicki wrote,
[Exercise 1] is a very interesting question, which initially seems very simple, but in reality is
very complicated. It seems as if the problem escapes mathematical methods.

Although exercise 1 was rated fifteen points, the editors awarded Krysicki twenty points
for his partial solution of a restricted case. A solution of Tarski’s exercise 169 would
provide a tool for solving the general case. This result and the method used for the solu-
tion presented in this section would provide useful tools for computer-assisted design.
However, the present editors have not found any reference to it in the literature of that
subject.28 In his 1975 historical description of Mody matematyk described in section 9.7,
Wadysaw Dubiel included this exercise among the six exemplars that he chose to reprint.

27
Rusiecki 1930a; Krysicki 1930. W odzimierz K rysicki was born in Warsaw in 1905 and educated there.
After earning a master’s degree in mathematics in 1928, he taught in a secondary school. Krysicki earned
the doctorate from the University of ódz in 1950 and taught there until his retirement in 1975. A noted
statistician and writer on mathematics, he died in 2001 (Pawlikowska-Broek 2003, 119).
28
Slightly more general versions of Tarski’s exercise were posed and solved in Garnett and Carver 1925 and
in Ford and Carver 1957.
262 12 Exercises Posed by Tarski

This exercise is intermediate in level. It emphasizes inequalities, but involves symme-

try as well, and is also an example of quantifier elimination.29 The assigned point value
seems too high. No solutions by readers were reported, and no solution was published.

Solution. If b  a and c  a, the rectangle can be placed in a corner of the square.

Assuming b + c  a 2 on the other hand, consider the subsegment AB of a diagonal
of the square, centered, with length b, as shown in figure 4, below. The remainder of
the diagonal consists of two segments, each of length d = 1/2 (a 2 – b)  c / 2 . The
perpendiculars to the diagonal at A and B meet the edges of the square at four points
that form a rectangle with edges of lengths b and 2 d  c. The desired rectangle can be
cut from this. Thus, Tarski’s condition is sufficient.

X x

B w
½c yes
W
V
b
no
W C
V
A

Figure 4 (for Section 12.9) Figure 5 (for Section 12.9)

To prove that the condition is necessary, suppose the desired rectangle can be cut from
the square in figure 5 but a < b or a < c. If b denotes the length of the longer edge, then
a < b in either case. Translate the rectangle horizontally then vertically within the
square, so that its left and right vertices become equidistant from the left and right edges
of the square, and its top and bottom vertices, equidistant from the top and bottom edges.
The translated rectangle is congruent to the original one, and has the same center C as
the square. Attempt to rotate the translated rectangle about C, within the square, to
make its major axis fall on one of the diagonals of the square. Should that fail, two
opposite vertices V, W of the rotated rectangle would have collided with opposite edges
of the square, as shown by the dashed rotation in figure 5. If only one pair of opposite
vertices collided with edges of the square, choose the other diagonal, which will intersect
the quadrants containing those vertices; after rotation, the rectangle would be situated

29
The word possible stands for an existential quantifier.
12.10 Exercise 170: On the Degree of Equivalence of Polygons 263

like the inner one in figure 4, and the previous reasoning will yield the desired inequality.
If both pairs of opposite vertices collided with edges, the rectangle would be situated like
V W X Y in figure 4, and the previous reasoning would imply b + c = a 2 .

To see that the axes of the rotated rectangle do coincide with the diagonals of the
square when its vertices all collide with edges, relax that requirement in figure 4. The
diagonals of V W X Y must nevertheless be equal, and they have simple formulas in terms
of a and the indicated lengths w and x. The resulting equation implies w = x.

12.10 [1931] 2014e, Exercise 170: On the Degree of Equivalence of Polygons

Tarski’s exercise Nr. 170: Twierdzenie o stopniu równowanoci wieloktów is translated

and discussed in this section. It was published in the February 1931 issue of Mody
matematyk (volume 1, page 46), distributed with Parametr, volume 2.

Exercise 170. Theorem on the degree of equivalence of poly-

gons. Use the assertion [in section 12.9, Exercise 169] to
prove the equation
W(4) = 4.
In the proof of this statement, it is permissible to refer to the
results of the exercises titled System of inequalities [in sec-
tions 12.7 and 12.8].
Ten points for solution. A. Tarski (Warsaw)

Discussion. The degree of equivalence and the function W were described and the
exercise posed as an unsolved problem in Tarski’s paper [1931] 2014a, translated in
section 7.2. It was reprinted in the present form later in the same issue of Mody
matematyk. Its level is advanced.

Tarski’s problem was solved the same year by Henryk Moese, as a corollary of a more
general result: W(n) = n for every positive integer n. Moese published this result in the
next volume of Parametr in the paper Moese [1931] 2014, which is translated in section
7.3. Moese’s methods had nothing to do with the suggestions that Tarski included in the
exercise statement.30 Director of a gimnazjum in Kpno, a town in southern Poland,
Moese was a regular contributor to these journals.

30
The present editors have not attempted to apply Tarski’s suggestions. For more information about Moese,
see a box in section 7.1.
264 12 Exercises Posed by Tarski

12.11 [1931] 2014f, Exercise 183: Postulate about Parallels

Tarski’s exercise Nr. 183: Postulat o równolegych is translated and discussed in this
section. It was published in the February–March 1931 issue of Parametr ( volume 2,
page 78).

Exercise 183. Postulate about parallels. Accepting the system

of axioms for Euclidean geometry given by David Hilbert, or
any system established in one of the Polish school texts on
elementary geometry,31 show that the axiom of parallelism
can be replaced by the statement,
For every point interior to a convex angle there
exists at least one segment that passes through this
point and has endpoints on the edges of the given
angle.
Five points for solution. A. T. (Warsaw)

Discussion. This exercise is advanced, not because of its mathematical content

—part of a core subject—but because of the sophistication required to consider alterna-
tives in the complicated axiom system of elementary geometry. It is somewhat delicate
because its interpretation and solution depend on how the notions of convex angle and
interior point are defined and how the parallel postulate is formulated.

Before introducing his version of the parallel postulate, Hilbert 32 presented postulates
that, for a given line g in a given plane H, allow classification of all points in H as lying
on g or on exactly one of two half-planes called the sides of g in H. Two distinct lines
g and h in H that intersect at a point O thus partition H into nine parts: O, four
open-ended rays emanating from O, and four regions, each of which is the intersection
of two sides of different lines. See figure 6 on page 266 for an example. The union of O
and a pair r and s of such rays, where r I g and s I h, is called a convex angle; the
rays are called its arms, and their intersection O, its vertex. Each arm lies entirely on

31
Tarski was familiar with Hilbert [1899] 1922. Section 9.2 contains a list of Polish geometry texts in use
at the time.
32
Hilbert [1899] 1922 or [1899] 1971, §§1–4; this present discussion applies to the 1903 second edition and
all later ones, but Hilbert’s treatment of the parallel postulate in preceding editions was different. For
further information see Hilbert 2003, 419.
12.11 Exercise 183: Postulate about Parallels 265

the same side of the line containing the other. The intersection of the side of g contain-
ing s and the side of h containing r is called the interior of the angle.33

Although Euclid’s original statement of the parallel postulate involved the notions of
side and angle, axiomatic presentations of Euclidean geometry often employ a simpler
version called Playfair’s postulate: in a given plane H, through any point O not on a
given line l there passes at most one line that does not intersect l. Such a line g is called
a parallel to l —in symbols, g // l.34 The other postulates in a presentation of Euclidean
geometry usually entail existence of at least one parallel to l in H.35

The present editors examined various editions of Polish texts available in the 1920s:
Enriques and Amaldi [1903] 1916, Kiselev 1917, omnicki 1923, Wojtowicz [1919] 1926,
and Zydler 1916. Each derived the existence of a parallel from postulates stated earlier,
and derived its uniqueness from Playfair’s postulate. Some discussed Euclid’s original
parallel postulate.

The form of the parallel postulate that Tarski presented in this exercise was apparently
first mentioned as such in Roberto Bonola’s [1906] 1955 history of non-Euclidean geome-
try. Bonola and others have suggested that the principle had been used earlier in Lorenz
1791–1792, but that seems doubtful.36

Solution. To show that Playfair’s postulate impliesY Tarski’s, assume

Y the former and
consider the convex angle S ROS formed by rays OR Z and
Y OS in lines g and h
respectively: see figure 6 on page 266. Let k be the line RS. Given any point P in the
interior of the angle formed by those rays, use the other postulates to find l // k in H
through P. By Playfair’s postulate, g must intersect l because l // k and g = / k;
similarly, h must intersect l, and thus the conclusion of Tarski’s postulate holds.

33
The word convex was used here because some geometry texts consider things called nonconvex angles,
with radian measure  0 or  S. They are not used in the presentation that Tarski was following.
34
Proposition 29 in Euclid’s book I ([1908] 1956, volume 1, 312) directly implies Playfair’s postulate. For a
proof of Euclid’s original postulate from Playfair’s, see the discussion by Thomas L. Heath (ibid., 313–314).
35
For example, drop a line k ] l from O, then erect in H a line g ] k at O.
36
Bonola (§28, page 58) reported that in an attempt to justify the parallel postulate, A. M. Legendre had
tacitly used this principle a century earlier in constructing from a given triangle a larger one whose defect
(180( – its angle sum) would be twice that of the original triangle. Bonola suggested that Lorenz had done
likewise. But in Euclid [1908] 1956, 220, the editor Heath reported that Johann Friedrich Lorenz had
in fact used a different principle. Neither Bonola nor Heath referred to a specific page of Lorenz
1791–1792. In his Berkeley doctoral dissertation supervised by Tarski, Haragauri N. Gupta (1964, page
11a) referred to Lorenz, volume 1, 101–102. The present editors have been unable to examine the original
edition of Lorenz. The 1798 second edition is inconsistent with all three citations. The 1804 third edition
(§§27, 28, 73) is consistent only with Heath’s citation. The principle that Heath mentioned occurred also
in the 1851 fourth edition (§28, which evidently stemmed from 1820) but not as a justification of the
parallel postulate.
 g H
h
O
S R Figure 6
k (for Section 12.11)
s r

l P

H
g
h
O Figure 7
D
P (for Section 12.11)

S Q R

O
SU P RU

S Q R

Figure 8 (for Section 12.11)

Tarski’s Parallel Postulate, in Terms of Betweenness E Alone:

EOPQ & ERUPSU & O =
/ P _ }R,S( EORUR & EOSUS & ESQR)
12.11 Exercise 183: Postulate about Parallels 267

To show the converse, assume Tarski’s postulate and (expecting a contradiction to

ensue) suppose a point O lay on distinct parallels g and h to a line l that does not pass
through O, as in figure 7 on page 266. Let P be any point on l, so that l would lie
entirely in theZYinterior of the angle D formed by the sides of g and h that contain P.
On the line OP select any point Q on the side of l opposite O and all other points on
g or h; then Q would be interior to angle D. By Tarski’s postulate, there would exist
a line k through Q that would intersect both g and h at some points R and
S, respectively, which would both have to lie on the side of l opposite Q. Thus, seg-
ments QR and QS would both intersect l. Those intersections would have to be
distinct, so that k = l, contrary to the hypotheses that g, h // l. This contradiction shows
that O cannot lie on distinct lines g, h // l, which is Playfair’s postulate.

Impact. No reader submitted any solution for this exercise, and none was published.
Tarski had begun using this form of the parallel postulate in his lectures on the axiom-
atics of Euclidean geometry at the University of Warsaw during 1926–1927.37 The
argument in the previous paragraph suggests that this form is closely entwined with the
properties of the betweenness relation E that are used to formulate reasoning about the
sides of a line in a plane. Indeed, this form of the postulate can be expressed in terms of
E alone38:
/ P _ }R,S( EORUR & EOSUS & ESQR).
EOPQ & ERUPSU & O =
This relationship is illustrated by figure 8 on page 266.

Tarski published his full system of postulates only after long delay, in the notable
[1957] 1959 paper What Is Elementary Geometry? There, he discussed his system’s organ-
ization in detail, but included no proofs. Nevertheless, that paper stimulated a consider-
able body of research by Tarski and others. For example, a version of the famous Pasch
postulate, used to describe the betweenness relation in both Euclidean and non-Euclidean
geometry, can be formulated in a form strikingly similar to Tarski’s parallel postulate:
EOQP & ERUPSU _ }R,S( EORRU & EOSSU & ESQR).
Readers of the present book may wish to consider the minor alteration of figure 8 required
to illustrate this so-called weak Pasch postulate. This analogy has suggested some intrigu-
ing, still unsolved problems in foundations of geometry.39

For more detailed analyses of Tarski’s system, see the papers Tarski and Geometry by
Lesaw W. Szczerba, and Tarski’s System of Geometry by Tarski and Steven R. Givant.
Givant constructed the latter publication from a long letter that he and Tarski had
written to Wolfram Schwabhäuser in 1978 to assist in the preparation of the 1983 book
Metamathematische Methoden in der Geometrie by Schwabhäuser, Wanda Szmielew, and
Tarski. That work finally included the proofs that demonstrated the adequacy of Tarski’s
postulate system for elementary Euclidean geometry. His axiomatization is now often
regarded as a standard for comparison of results in foundations of geometry.

37
Tarski and Givant 1999, 175.
38
Tarski and Givant 1999, 183. The symbol EOPQ means that points O and P are equal, or P lies
between O and point Q, or P = Q.
39
Tarski and Givant 1999, 197–198.
268 12 Exercises Posed by Tarski

12.12 [1931] 2014g, Exercise 186: Analytic Geometry of Space

Tarski’s exercise Nr. 186: Zadanie z geometrji analitycznej w przestrzeni is translated and
discussed in this section. It was published in the February–March 1931 issue of Parametr
(volume 2, page 78).

Exercise 183. Analytic geometry of space. Find the locus of

points equidistant from three given lines. Give an exhaustive
discussion of the possible cases.
Thirty points for solution. A. T. (Warsaw)

Discussion. This exercise and the next seem out of order: see the following section
12.13. This exercise is well known, not usually attributed to Tarski, and probably not
original to him. It is advanced, due to its requirement to consider very many cases, and
to its use of solid analytic geometry, including the intersection of two hyperbolic parabo-
loids. It still plays a role in geometric research.40 The word find (znale  ) in its statement
is unfortunately vague. No solutions by readers were reported, and no solution was
published.

Discussion of this exercise is continued at the end of the next section, 12.13, because
solution of the next exercise seems prerequisite to solution of this one.

12.13 [1932] 2014b, Exercise 213: Stereometry

Tarski’s exercise Nr. 213: Zadanie ze stereometrji is translated and discussed in this
section. It was published in the January–February 1932 issue of Parametr (volume 2,
page 207).

40
See Everett, Gillot, et al. 2009 and Everett, Lazard, et al. 2009.
12.13 Exercise 213: Stereometry 269

Exercise 213. Stereometry. Find the locus of points in space

equidistant from three given lines lying in one plane. Give an
exhaustive discussion of all possible cases.
Five points for solution. A. T. (Warsaw)

Discussion. This exercise and the previous one, in section 12.12, seem out of order.
Solution of this second one seems prerequisite to solution of the first. Following the solu-
tion of the present exercise is a partial solution of the earlier one. The level of the present
exercise is intermediate, due to its requirement to consider very many cases; but it is
devoted to a core subject in geometry. The word find ( poda) in its statement is unfortu-
nately vague. Aleksander Grzymaa, a reader from Warsaw, contributed a solution and
received full credit, but no solution was published.

Solution. First, let H be the given plane. Determining the locus OH of points in H
equidistant from the given lines is almost a routine plane-geometry problem. If (1) all
three coincide, then OH = H. If (2) just two of them are distinct and those intersect, then
OH is the union of the two lines that bisect the resulting vertical angles; the bisectors are
perpendicular to each other. If (3) just two of the given lines are distinct and those are
parallel, then OH is the line in H parallel to and midway between them. If (4) the lines
form a triangle, then OH consists of four distinct points: its incenter and excenters. If
(5) the lines are distinct and concurrent, then OH is their common point. If (6) the lines
are all distinct and parallel, then OH is empty.

The previous paragraph dealt with the routine planar cases. One planar case remains:
(7) two of the lines, g and h, are distinct and parallel and the third is a transversal
intersecting them at points V and W, respectively. The two pairs of bisectors of the
vertical angles formed at V and W form a rectangle V Y WZ with center X, depicted
with dashed lines in figure 9 on page 270. Its diagonals V W and YZ are congruent and
bisect each other and form two pairs of congruent isosceles triangles. By the exterior-
angle and isosceles-triangle theorems, SW X Z is congruent to the angle between the
transversal and g, on the side opposite Z. By the theorem about alternate interior
angles, Y and Z fall on the line m in H parallel to g and h and midway between
them: thus Z, and Y as well, belongs to the locus OH of points on H equidistant from
all three given lines. Conversely, any point of OH must lie on one of the bisectors and on
m, and hence must coincide with Z or Y. In this nonroutine planar case, the locus OH
consists of two points.
270 12 Exercises Posed by Tarski

g W

X
Y Z
m

h
V

Figure 9 (for Section 12.13)

Now it is simple to determine the desired locus O of all points in space that are
equidistant from the three given lines in H: it is the union of all perpendiculars to H
through points in OH . In the cases considered above, O consists of (1) all points in space,
(2) the union of two intersecting planes perpendicular to H, (3) a single plane perpendicu-
lar to H, (4) four distinct lines perpendicular to H, (5) a single line perpendicular to H,
(6) no points at all, or (7) the union of two distinct lines perpendicular to H.

Partial solution of exercise 183. The exercise described in the previous section asked
for the locus / of all points in space that are equidistant from three given lines config-
ured arbitrarily. Cases (1–3), discussed in the previous paragraph, cover the situations
in which the three lines all coincide or just two are distinct but coplanar. Cases (4)–(7)
cover the situations in which the three lines are all distinct but coplanar.

Several more cases can be analyzed relatively easily, in which the given lines are all
distinct but two pairs are coplanar. For case (8), in which the three lines are all parallel,
/ is the line parallel to them through the circumcenter of a triangle formed by their
intersections with a common perpendicular plane. If (9) two of the given lines intersect
and the third is parallel to one of them, then / = H  ( ] F K), where H is the plane of
points equidistant from the parallels, and ] and K are the two perpendicular planes of
points equidistant from the intersecting lines; in this case, / will consist of two inter-
secting lines. If (10) two pairs of given lines each intersect, then / is the intersection
of two unions of pairs of perpendicular planes; in general, / will consist of four lines
—two pairs of perpendiculars. In the concurrent case, these four lines will be concurrent
as well. Readers may wish to puzzle about whether special configurations of the given
lines in case (10) will yield a locus / that is simpler than the general one just described.

The remaining cases involve nonlinear analytic geometry—the reason for the title of
exercise 183. Of these, the simplest case (11) is the most important: just two of the given
lines, g and h, are distinct, but they are noncoplanar. The discussion here is limited
to an example. Use a Cartesian coordinate system with origin O and unit points U, V,
and W. Let g be the line through U parallel to the axis through V, and let h be the
12.14 Exercise 214: Arranging Two Segments in a Plane 271

axis through W. The squares of the distances to g and h from a point X =

 x 1 , x 2 , x 3  are ( x1 – 1) 2 + x 32 and x12 + x 22, respectively, and thus the locus / has
equation ( x1 – 1) 2 + x 32 = x12 + x 22. This is equivalent to the equation
x1 = 1/2 ( x 32 – x 22 ) + 1/2 ,
which identifies the locus as a hyperbolic paraboloid. The general case is similar to this
example in that two noncoplanar lines g and h always have a common perpendicu-
lar 41 analogous to the axis through U. But in general, h will not be perpendicular to
the plane of g and that axis. More analytic geometry is required to find the distance
from X to h, and the algebra leading to the equation of the locus / becomes formida-
ble. But it can nevertheless still be identified as a hyperbolic paraboloid.42

In the final cases, at most one pair of the given lines can be coplanar. If (12) those are
parallel, then the locus / is a conic section: the intersection of the plane of points
equidistant from the parallels with the hyperbolic paraboloid of points equidistant from
one of the other pairs. If (13) that given pair intersects, then / is the intersection of a
hyperbolic paraboloid with the union of two perpendicular planes. In general, such an
intersection will be the union of two conics; readers may wish to investigate whether
any special instance of this case might yield a simpler locus. Finally, if (14) no pair of the
given lines is coplanar, the locus / is the intersection of two hyperbolic paraboloids.
Readers may wish to investigate the nature of that curve.43

12.14 [1932] 2014c, Exercise 214: Arranging Two Segments in a Plane

Tarski’s contribution Nr. 214: Ukad dwóch odcinków na paszczy nie is translated and
discussed in this section. It was published in the January–February 1932 issue of Param-
etr (volume 2, page 207).

44
Exercise 214. Arranging two segments in a plane. A plane
figure consisting of two segments is given. Indicate all axes
and centers [of symmetry] of this figure, lying in its plane.
Give an exhaustive discussion of the possible cases.
Five points for solution. A. T. (Warsaw)

41
The common perpendicular to two noncoplanar lines is discussed by Thomas L. Heath in Euclid [1908]
1956, volume 3, 306–307, in his commentary on book XI, proposition 19. See also Enriques and Amaldi
[1903] 1916, chapter 12, §780, 322.
42
Readers of the present book may wish to ferret out interesting properties of this locus; see Forsythe 1969.
43
See Everett, Gillot et al. 2009 and Everett, Lazard et al. 2009.
44
Tarski’s term for arranging was ukad: arrangement or system. Givant (1986, 936) used system.
272 12 Exercises Posed by Tarski

Discussion. The level of this exercise is intermediate although it requires no deep

expertise in geometry. It forces students to consolidate familiarity with a broad scope of
elementary geometric objects and their properties. Organization and exposition of a
solution is its greatest challenge. Should a solution classify the methods of arranging the
components of the figure, just the point set that results, or the possible sets of symme-
tries? The solution presented below takes the last approach, which seems to lead to the
least redundancy. It also attempts to give enough detail that readers can reconstruct the
sketches that guided the present editors. No solutions by readers were reported, and no
solution was published.

Solution. Tarski did not specify whether his segments should contain their endpoints.
This solution assumes that they do, and that points are segments with length zero. The
degenerate case in which both given segments are the same single point O exhibits the
most symmetry: all rotations about O and reflections across all axes through it.

Otherwise, the largest possible symmetry group is that of a square, which contains two
quarter-turns and a half-turn about its center O and reflections across the four axes that
join O with the vertices and edge midpoints. This symmetry occurs only when the two
given segments are the diagonals of a square.

The next largest group is the symmetry group of a nonsquare rectangle: the Vierer-
gruppe, which contains a half-turn about the center O of the rectangle and reflections
about the two axes that join opposite edge midpoints. This symmetry occurs when the
two given segments are
• collinear and overlap, or
• collinear, disjoint, and congruent, or
• the diagonals or a pair of opposite edges of a nonsquare rectangle, or
• the diagonals of a nonsquare rhombus.

Some configurations are symmetric only by reflection about a single axis. This happens
when the two given segments are
• collinear and disjoint but incongruent, or
• a pair of congruent edges of an isosceles triangle, or
• the parallel edges or the congruent edges or the diagonals of an isosceles
trapezoid that is not a parallelogram, or
• the base of an isosceles triangle and a subsegment of its altitude, or
• the diagonals of a kite that is not a rhombus.

Two configurations are symmetric only by a single half-turn: the two given segments
must be a pair of opposite edges or the diagonals of a parallelogram that is neither a
rhombus nor a rectangle.

Other configurations of the given segments exhibit no symmetry.

 13
Geometry for the Third
Gimnazjum Class (1935)

This chapter is devoted to a translation of parts of the school-geometry text Geometrja

dla trzeciej klasy gimnazjalnej by Zygmunt Chwiakowski, Wacaw Schayer, and Alfred
Tarski. The text was published in 1935 by Pastwowe Wydawnictwo Ksiek Szkolnych,
the National Schoolbooks Publisher.1 Its title page is shown on page 276 of the present
book. The note there under the publisher’s logo says that the price of the book, including
the cost of the stamp of the Society Supporting Construction of Common Public Schools,
is 1.20 zlotys. That was comparable to the retail price of three kilograms of wheat flour,
one kilogram of pork, five liters of milk, or thirteen eggs.2

This translation is even closer to the original than the other translations in the present
book. Their intent is to convey content most accurately with changes in style only as
necessary for that. However, a main reason for translating parts of the Geometrja book
is to display its style,3 to provide a better glimpse of the atmosphere of mathematics
instruction in Poland during its time. All [square] brackets in the translation enclose
editorial comments inserted for clarification; this includes all the footnotes.

At the beginning of Geometrja, lengths are regarded as segments. For example,

consider the first sentence of §3 (page 281):
We know that the distance from a point M to a line PQ ... is the segment MN constructed from
point M perpendicular to PQ.

In the middle of Geometrja is a transition. The authors introduced numbers into geome-
try gradually. In §24, they used rational numbers as ratios of pairs of commensurable

1
Chwiakowski, Schayer, and Tarski [1935] 1946. For information about Chwiakowski and Schayer and
about later editions, see boxes in section 9.9.
2
These data, from Poland 1936, 168, are for 1935 in Warsaw, where the cost of living was higher than in
many other places in Poland.
3
The original use of boldface type for emphasis has been maintained. In the original, uppercase Latin
variables were not italicized; in the translation, they are. Minor changes in punctuation were sometimes
made to adhere more closely to conventions of English, and sometimes to enhance clarity. Most paragraphs
in the original, and in the translation, consist of single sentences.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 273
DOI 10.1007/978-1-4939-1474-6_13, © Springer Science+Business Media New York 2014
274 13 Geometry for the Gimnazjum

segments. In §25 (page 292) they showed that the diagonal BC of the unit square is
incommensurable with its edge AC. Then (page 294) they gave numbers a greater role:
We assume that the length BC with respect to the unit AC is a number. But it cannot be
a whole or fractional number ... . Therefore, this is a new kind of number. Previously we did
not know about such numbers, and only now do we become familiar with them.

This changeover presented a problem in translation. At first, concepts referred to as

segments, lengths, or distances are treated somewhat interchangeably. Congruent seg-
ments are called equal; so are certain segments and distances. Later, after introduction
of irrational numbers, the distinction between geometric segments and numerical lengths
or distances is clearer, but the authors’ terminology is not fully consistent.

The figures were redrawn using Mathematica software: their relative dimensions are
close approximations to those of the originals. The only systematic change was to use
solid rather than hollow dots to indicate points. Placement of figures in relation to
surrounding text is close to that of the original.

In the present book are translated the very beginning and very end of the Geometrja
text, plus some material from the middle: its sections 1–4, 25–27, 34–35, and table of
contents.4 The table is shown on the facing page. Sections 34–35 are devoted to the
theory of areas of polygonal regions; thus, they are closely related to Tarski’s research
reported in Part Two of the present book. Sections 25–27 of Geometrja introduce readers
to irrational numbers. They contain some material necessary for the later sections, and
are related to the approach to the real number system described in Tarski’s [1932] 2014
article on the circumference of a circle, translated in chapter 11 of the present book. Brief
editorial comments are inserted in boxes between the translations of sections 4 and 25,
and between those of sections 27 and 34, to describe the sections of Geometrja that are
not translated here.

For more detailed information about Geometrja and its background, see section 9.9 of
the present book. As noted there, the editors hope that including in this single volume
of selected translations works of such contrasting mathematical sophistication as Tarski’s
geometrical research in Part Two and the secondary-school geometry lessons in the
present chapter will spur broader investigation into the relation between mathematical
research and mathematics education, not just in Poland between the world wars, but in
more general contexts.

4
The book contained no significant front matter after the title page, and no back matter except the table
of contents. The §28 title in the original table was wrong; it was corrected in later editions and here.
 TABLE OF CONTENTS

PART ONE

Geometric Loci

page
§1. Circle and Disk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 277
§2. Construction Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
§3. Relative Positions of a Circle and a Line Lying in a Plane . . . . . . . . . . . . . . . . . . . . . . 281
§4. Construction Problems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285
§5. Tangents to a Circle –
§6. Axis of Symmetry of a Segment
§7. Construction Problems—Method of Geometric Loci
§8. Circumcircle of a Triangle
§9. Bisectors of Angles Formed from Two Intersecting Lines
§10. Incircle of a Triangle
§11. Parallel Lines
§12. Parallel Lines as Geometric Loci
§13. Geometric Locus of Points Equidistant from Two Parallel Lines

PART TWO not

–
translated
Angles in Relation to a Circle
§14. Central Angles
§15. Arcs and Chords
§16. Diameter of a Circle as Axis of Symmetry
§17. Chords and Their Distances from the Center of a Circle
§18. Angles between Chords and Tangents
§19. Inscribed Angles
§20. Arcs as Geometric Loci
§21. Construction Problems
§22. Quadrilaterals Inscribed in a Circle and Circumscribed about a Circle
§23. Regular Polygons –

PART THREE
Measuring Segments and Areas
§24. Commensurable Segments
§25. Incommensurable Segments—Irrational Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
§26. Measuring Incommensurable Segments with a Unit . . . . . . . . . . . . . . . . . . . . . . . . . . 296
§27. Rational Approximations of Irrational Numbers—
Operations on Irrational Numbers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
§28. A Theorem of Thales –
§29. On Similarity—Similar Triangles not
§30. Method of Similarity in Construction Problems –
translated
§31. Practical Applications
§32. Similar Polygons –
§33. Equivalent Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
§34. On Measuring Areas of Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
§35. Areas of Similar Polygons . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316
 PART ONE

Geometric Loci
§ 1. Circle and Disk

With point O as center, we describe a circle with radius equal to a given segment r
(figure 1).

Every point of the circle, such as A, G

B, C, is one radius distant from the
center of the circle and, conversely, B
each point of the plane whose distance A E
from the center O is equal to the
radius lies on the circle.
O F
If each point of a certain figure has
I
a given property and, conversely,
every point having this property be- D
longs to the figure, we say that this
figure is the geometric locus of points
having the given property. H
C

Thus we can provide the following Figure 1

definition of a circle.

Definition. A circle is a geometric locus of points in a plane whose distances

from a certain fixed point called the center are equal to each other.

The points whose distances from the center O are less than the radius, for example
D, E, F, we call interior to the given circle, and points whose distances from the center
O are greater than the radius, for example G, H, I, we call exterior.

The points on a circle and points interior to the circle constitute a disk.

Definition. A disk is a geometric locus of points in a plane whose distances

from the center are less than or equal to the radius.5

Recall from last year the definitions of chord, diameter, central angle, and arc, and
write them out.

Now we recall the following axiom:

Axiom. A segment or arc of a circle connecting a point interior to a circle

with an exterior point intersects that circle at one and only one point.

5
[In the original the terms for circle and disk were okrg and koo, respectively. The authors should have
defined the term radius before interior.]
278 13 Geometry for the Gimnazjum

Based on on this axiom,6 it is possible to prove the following theorem:

Theorem. A line7 passing through a point interior to the circle intersects the
circle at two points.

Let us also accept without proof this theorem:

Theorem. Any two points interior to a circle can be connected with a seg-
ment or polygonal path having no point in common with the circle.8

Consider whether the same can be said for any two points exterior to the circle.

Exercises. 1. Choose any point A [and] draw the geometric locus of points distant
from point A by a segment [of length] (a) 2 cm, (b) 3 cm, (c) 5 cm.
2. Choose any point A and shade the geometric locus of points distant from point A
by a segment [of length] (a) less than or equal to 2 cm, (b) greater than or equal to 3 cm.
3. Using any radius draw a circle and sketch the geometric locus of points distant from
the center of the circle by a segment equal to (a) half of the radius of the given circle, (b)
twice the radius of the given circle.
4. Draw any angle, not equal to a straight or full angle; choose point A (a) interior
to, (b) exterior to the angle.9
Find the points on the arms of the angle distant from point A by a given segment.
5. With point A as center, draw two circles, one with radius equal to 2 cm, the second
with radius equal to 5 cm.
Indicate on the drawing the geometric locus of points whose distance from the center
is (a) less than 2 cm, (b) equal to 2 cm, (c) greater than 2 cm but less than 5 cm, (d) equal
to 5 cm, (e) greater than 5 cm.

6
[Euclid’s very first proposition ([1908] 1956, 241–242) rested on this principle but he failed to state it
among his postulates. Now called the circle principle, it is often derived through analytic geometry from
a much stronger geometric continuity axiom. Geometrja alludes to a continuity axiom informally in §26
(page 296). The circle principle was stated formally as an axiom in the Polish texts Enriques and Amaldi
[1903] 1916 (§§191, 211, pages 56, 63, for a segment and an arc), omnicki 1934 (§§27, 28, pages 44, 47),
and Wojtowicz [1919] 1926, 21. (omnicki 1934, a textbook for the preceding gimnazjum year, adhered
to the same curriculum standards as Geometrja.) The text Zydler 1925 (22–23) considered the circle
principle worth mentioning, but did not indicate that it must be introduced as an axiom or derived from
a stronger one. See the discussion in Moise [1963] 1990, §§16.2 and 16.5.]
7
[The word prosta in the original is translated here as line. It connotes straight line.]
8
[In the original, the words for segment and polygonal path were odcinek and amana, respectively. It is
easy to prove this theorem using the exterior angle theorem and Euclid’s propositions 18 and 19 ([1908]
1956, 279–284).]
9
[What amounts to a definition of the phrase “interior of an angle with measure m such that 0(  m 
360(” is to be found in omnicki 1934, 22–23. The terms peny and pólpeny for full and straight angle
(360( and 180(—pó means half ) are explained there and used in this exercise in Geometrja.]
13.2 §2 Construction Problems 279

6. Draw a segment AB [with length] equal to 7 cm; with point A as center draw a
circle with radius 5 cm; with B [as center], a circle with radius 4 cm. Indicate the
geometric locus of points whose
(a) distance from A is less than 5 cm [and] from B, less than 4 cm;
(b) distance from A is less than or equal to 5 cm [and] from B, greater than 4 cm;
(c) distance from A is greater than 5 cm [and] from B, less than or equal to 4 cm;
(d) distance from A is greater than 5 cm [and] from B, greater than 4 cm.
7. Choose any two segments a and b ( a > b), and with an arbitrarily chosen point
A as center, draw concentric circles: one with radius a; a second with radius equal to
the sum a + b; [and] a third with [radius equal to] the difference a – b. Where are the
centers of the circles located, drawn with radius b, that with the first of these circles (a)
do not have even one point in common, (b) have one common point, (c) have two common
points?
8. With an arbitrarily chosen point as center, draw a circle with an arbitrary radius.
Choose a point A on this circle. Find points on the circle distant from point A by a
segment (a) equal to the radius of this circle, (b) less than the radius, (c) greater than the
radius but less than the diameter, (d) equal to the diameter.
Are there any points on this circle distant from point A by a segment greater than
the diameter?

§ 2. Construction Problems

Problem. Construct a triangle, given two edges a and b and the median m to edge a.

To find a way to solve this problem, we a

draw an arbitrary triangle ABC (figure b
2) and its median AD and assume that
m
it is precisely the triangle that we are
going to construct and that BC = a, A
AC = b, AD = m.

Now we consider how to draw such a b

m
triangle.
D
We can immediately draw edge CB, C B
equal to a: thus, we have two vertices of a
the triangle, B and C —[this] is only
about finding the third vertex A. Figure 2

The vertex A lies at a distance m from point D, which is the midpoint of edge BC.
Therefore A lies on the geometric locus of points at distance m from point D: that is,
on the circle with center D and radius m.
280 13 Geometry for the Gimnazjum

We can also say about the vertex A that it is located at a distance b from point C
[and] therefore lies on the geometric locus of points at distance b from point C —that
is, on the circle drawn with radius equal to b about point C as center.

Therefore, point A is a common point of these two circles.

We now know how to perform the construction of the triangle.

On a line (figure 3) we place the seg-

a
ment BC = a, [and] we find the point D
b
c that is the midpoint of segment BC
(CD = a/ 2 ). About point D we draw a
circle with radius m; and about point C
A we draw a circle with radius b. The
points A and AU of intersection of these
b circles we connect with points C and B.
m We have obtained two triangles: ABC
D and AUBC.
C B
a
Prove that each of these triangles has
two edges a and b and median m.

Prove that ' ABC and ' AUBC are

AU
congruent.

But can you always perform this

Figure 3 construction?

The following cases can arise: the circles drawn about points C and D
(1) have no points in common,
(2) have only one point in common (are tangent),
(3) have two points in common (they intersect).

If these circles did not intersect or were tangent, we could not construct the triangle.

Since we know what condition must be satisfied so that the two circles intersect, we
can conclude that the desired triangle can be constructed if and only if the segment CD =
a/ 2 is less than the sum of segments b and m and greater than their difference.

Many constructions amount to determining a point of intersection of two circles, then

proceeding as in the previous problem.

Exercises. 1. Choose any two points A and B and find a point whose distance from
point A equals the segment AB, while the distance from point B is equal to an
arbitrarily chosen segment a.
13.3 §3 Relative Positions of a Circle and a Line 281

Consider the cases in which (1) a < 2 AB, (2) a = 2 AB, (3) a > 2 AB.
2. Choose in the plane any three points A, B, and C. Find a point whose distance
from point A is equal to the segment AC and [whose] distance from point B is equal
to segment BC.
3. Construct a parallelogram, given an edge and the two diagonals.
4. Construct a rhombus, given an edge and a diagonal.

§ 3. Relative Positions of a Circle and a Line Lying in a Plane

We know that the distance from a point M to a line PQ (figure 4) is the segment MN
constructed from point M perpendicular to PQ.

If we are given a circle and a line, one and only one of three cases must then occur
(figure 5): the distance from the line to the center of the circle is (1) greater than the
radius, (2) equal to the radius, (3) less than the radius.

A B
E
C D
F G

P N Q

Figure 4

Figure 5

Now we prove the following three theorems. [Proofs follow the three statements.]

Theorem 1. If the distance from a line to the center of a circle is greater than
a radius, the line does not have any points in common with the circle.

Theorem 2. If the distance from a line to the center of a circle is equal to a

radius, the line has one point in common with the circle.

Theorem 3. If the distance from a line to the center of a circle is less than
the radius, the line has two points in common with the circle.
282 13 Geometry for the Gimnazjum

(1) Assumption: the distance from point

H E A O to the line AB is greater than the radius.
We shall prove that every point of the line
M
AB lies exterior to the circle (figure 6).

Point C lies exterior to the circle, since

OC > OD. Any other point M of the line
AB is farther from O than C, because
O D OM > OC (the hypotenuse [of a right trian-
C
G gle] is greater than a leg); thus, all points of
the line AB are exterior to the circle, and
therefore the line has no point in common
N
with the circle.

(2) Assumption: the distance from point

O to the line EF is equal to the radius r.
I F B
Therefore, when the perpendicular is
Figure 6 dropped from point O to EF, the segment
OD of this perpendicular is equal to r, and
thus the point D lies on the circle.

We shall prove that other than point D the line EF has no point in common with
the circle. Let us take on line EF any point N different from D, [so that] ON > OD,
[and] therefore N lies outside the circle. Thus all points of the line EF except D lie
in the exterior of the circle, and the line has only one point in common with the circle.
In this case we say that the line EF is tangent to the circle.

(3) Assumption: the distance from the line HI to O is less than the radius r. We
construct OG ] HI, so that OG < r; therefore G lies in the interior of the circle.

We know from the preceding that a line passing through an interior point of a circle
intersects the circle at two points; thus, HI intersects the circle at two points.

We call a line that intersects a circle at two points a secant.

The question arises whether the theorems converse to the previous [ones] are true.
We know that an assertion converse to a true one is not always also true. For example,
the theorem “if two angles are vertical, then they are equal” is true, but the converse of
this theorem, “if two angles are equal, then they are vertical,” is false.

Let us examine whether the following statement is true:

Assertion converse to theorem 1. If a line does not have any point in common
with a circle, then the distance from the line to the center of the circle is
greater than the radius.
13.3 §3 Relative Positions of a Circle and a Line 283

Assumption: the line does not have any points in common with the circle. [Desired]
conclusion: the distance from the line to the center of the circle is greater than the radius.

There are only three possible cases concerning the distance from the line to the center
of the circle: specifically, that this distance be (1) greater than the radius, (2) equal to
the radius, (3) less than the radius. One and only one of these cases must occur.

Let us suppose that the distance from the line to the center of the circle were equal to
the radius; then on the basis of theorem 2 the line would have one point in common with
the circle, which is contrary to the assumption of our theorem. Similarly, if the distance
from the line to the center of the circle were less than the radius, then on the basis of
theorem 3 the circle would have two points in common with the line, which again contra-
dicts the assumption of our theorem.

We have proved that the distance from the line to the center of the circle cannot be
equal to the radius or less than the radius; therefore, it is greater than the radius. In
this way the theorem has been proved.

Form the converses of the remaining two theorems, 2 and 3, and prove them by the
method of reduction to absurdity.10

We notice that for the proof of each of the theorems converse to theorems 1, 2, and 3,
it is sufficient to prove that the assumptions of theorems 1, 2, and 3 exhaust all possible
cases—that is, one and only one of them must occur—and moreover the conclusions of
theorems 1, 2, and 3 mutually exclude each other.

Thus, if we have several theorems whose assumptions exhaust all possible cases, and
the conclusions following from them mutually exclude each other, then the assertions
converse to the previous [ones] are true and can be proved by the method of reduction
to absurdity.11

We call a system of theorems satisfying the above conditions a closed system of

theorems.

The three previous theorems, which form a closed system of theorems, we can write
in abbreviated form (we denote the distance from the line to the center of the circle by
d, and the radius by r):
­(1) d > r, ­(1) have no points in common,
° °
If ®(2) d = r, then the line and the circle ®(2) have one point in common,
°̄(3) d < r, °̄(3) have two points in common.

10
[The authors used Polish words meaning reductio ad absurdum.]
11
[Tarski incorporated text similar to this in §50 of his well-known 1936 logic text. There, he called the
principle stated in this paragraph Hauber’s law. The assumptions of the several theorems need not be
mutually exclusive as required here, but merely exhaustive.]
284 13 Geometry for the Gimnazjum

In this way, write down the converse theorems.

Last year12 we encountered a closed system consisting of five theorems. Denoting by

d the distance between the centers of two circles, [and] by r and rU their radii (r > rU),
we write these five theorems as follows:

­ (1) d > r + rU, ­ (1) have no points in common—

° ° moreover, one lies in the
° ° exterior of the other;
°(2) d = r + rU, °(2) are externally tangent;
° °
If ®(3) r – rU < d < r + rU, then the circles ®(3) have two points in common;
°(4) d = r – rU, °(4) are internally tangent;
°(5) d < r – rU, °(5) have no points in common—
° ° moreover, one lies in the
° ° interior of the other.
¯ ¯
Formulate and write the converse theorems in this way.

In the future, when we prove that some theorems form a closed system, we will not
prove the converse theorems separately.

From the three theorems previously proved follow [some] corollaries.

Corollary 1. The line AB (figure 7) perpendicular to a radius OC and passing through

its endpoint C is tangent to the circle. (The distance to AB from O is equal to the
radius.)

This property gives us the ability to construct the tangent to a circle passing through
a point on the circle.

B B
M

C C

A O A O

Figure 7 Figure 8

12
[See Konarski 1934, §28, 44–49.]
13.4 §4 Construction Problems 285

Corollary 2. If through a point of contact C (figure 8) of a line AB with a circle we

pass a radius OC, then that is perpendicular to AB.

Proof. If OC were not perpendicular to AB, then we could drop OM ] AB; since
AB is tangent, its distance from O is equal to the radius r ; therefore, OM = r. Point
M lies thus on the circle. Since AB and the circle have only one point in common, point
M coincides with C, OM coincides with OC, [and] segment OC is perpendicular
to AB.

Exercises. 1. Draw an arbitrary line and a point A not lying on it. Choose any
segment a and locate a point on the line distant from point A by the segment a. Does
there exist such a point for all segments a, of any length?
What should the length of segment a be so that (1) there does not exist such a point,
(2) there exists only one such point, (3) there exist two such points?
2. Draw any line and choose a point O not lying on it. Trace the circle13 with center
O [and] tangent to that line.
3. On a line, choose any point A and with radius equal to a given segment trace circles
tangent to that line at point A.
4. Draw any line and a circle having no point in common with it. Pass a tangent to
the circle (a) parallel, (b) perpendicular, to this line.
5. Draw a circle and a line, choose any angle, and pass a tangent to the given circle,
forming with the given line an angle equal to the given [angle].
6. With a point A as center, draw a circle tangent to a given circle (a) externally, (b)
internally.
7. Trace a circle with any radius and a line having no point in common with it. On
the circle choose any point A. Trace a circle with center on this line and tangent to the
given circle at point A. For which locations of point A is the problem impossible?
8. On any line, choose any point A. With point A as center, draw a circle with radius
3 cm. Draw circles with 2-cm radii whose centers are on the line and that have one point
in common with this circle. How many such circles can you draw?

§ 4. Construction Problems 14

We have already solved several construction problems, using only compass and
straightedge. In the future we will be performing geometric constructions using only
these two tools. (For drawing parallels and perpendiculars, we can use set squares, since
both of these constructions can also be performed using compass and straightedge.)

13
[Here the original used the word koo, which usually meant disk.]
14
[The original titles of sections 2 and 4 were the same.]
286 13 Geometry for the Gimnazjum

If besides compass and straightedge we are going to use still other tools, that will be
specially discussed in the terms of the problems.

Problem 1. Construct a triangle, given two edges a and b and the angle E lying
opposite b, the smaller of them (figure 9). For what lengths of edge b is the construction
possible?

Let ABC be the desired triangle (figure 9). We can draw the edge BC and angle E
at once,15 still seeking vertex A, which is located on line BM (figure 10) and which is
distant from point C by segment b [and] thus is a point common to line BM and the
circle with center C and radius b.

From point C we drop the line CD perpendicular to BM.

a
b

E
a b
C

B AU D A M

E
B A

Figure 9 Figure 10

The following cases may occur.

(1) b > CD: then a circle with radius b and center C intersects BM in
points A and AU. Therefore we have two triangles ABC and AUBC
satisfying the conditions of the problem, and there are two solutions.
(2) b = CD: then the circle is tangent to BM, and we have only one triangle
BCD.
(3) b < CD: the circle and line BM have no point in common—the problem
has no solution.

15
[The original, in error, specified angle B in this sentence.]
13.4 §4 Construction Problems 287

Solving a construction problem often comes down to determining the point of intersec-
tion of a given line AB with a circle drawn about a given point M as center.

Then there can occur three cases:

(1) the radius of the circle is less than the distance from point M to AB,
so that there are no points of intersection of the circle with AB;
(2) the radius of the circle is equal to the distance from M to AB, so that
the circle is tangent and there exists one point common to the circle and
line AB;
(3) the radius of the circle is greater than the distance from M to AB, so
that there exist two points of intersection of the circle with AB.

If a construction problem is more complicated, then we proceed in the following way.

Assuming that the problem is solved, we draw by hand a figure similar to the one
sought and suppose that it satisfies the requirements of the problem. Next, we try to find
a relationship between the givens and the elements sought that allows us to reduce the
solution of the given problem to [that of] another problem that can we can already solve.
In addition, it is often useful to construct auxiliary circles or lines.

Such a procedure we call the analysis of the construction problem.

For example, we will solve the following a

problem. b
m
Problem 2. Construct a triangle, given
C
two edges a and b and the median m with
respect to the third edge.
a b
Let ' ABC (figure 11) be the desired tri-
m
angle and CD, the median to edge AB, [so
D A
that] BD = DA. We find that [by] extending B
CD, measuring off DE = CD, and joining
successively the points A, C, B, E, A, we
obtain the quadrilateral ACBE, which is a
parallelogram because [its] diagonals divide
each other into halves. Triangle EBC, E
which is half of the parallelogram, we can
construct from three edges: CB = a, BE = Figure 11
b, CE = 2 m. We can therefore construct the
desired triangle BCA.
288 13 Geometry for the Gimnazjum

Now we perform the construction and

B describe it. We construct triangle EBC
with edges a, b, 2 m. With this as goal, we
measure off (figure 12) ED = DC = m.
About points E and C we draw circles with
radii equal to b and a, respectively.

E
D
C
These circles intersect if and only if
16
a – b < 2 m < a + b.

We have two intersection points B and

BU that satisfy this condition, and construct
BU two triangles EBC and EBUC, “oppositely”
A
equal—that is, congruent after reversing one
of them. We draw only one of these trian-
gles, EBC. We construct BD, extend BD,
Figure 12 and on the extension we measure off DA =
BD; joining A with C, we obtain ' ABC,
as desired.

Now it is necessary to prove that the figure as drawn meets the conditions of the
problem. This part of the reasoning is called the proof. From the construction it follows
that BC = a, BD = DA, DC = m.

Since 'DBE = ' DAC (DB = DA, DC = DE, and angles ADC and BDE are equal),
[it follows that]
AC = BE = b.

We see that ' ABC satisfies the conditions of the problem. We still have to investigate
whether the problem is always solvable; if not always, it is necessary to investigate the
conditions under which solution is possible, and the number of solutions. This part of
the reasoning we call the determination of the conditions for the possibility of solutions,
and their number.

In our problem the construction of triangle ABC is possible if and only if it is possible
to construct the auxillary triangle EBC, and thus when the conditions (with a > b)
a – b < 2m < a + b
are satisfied.

16
[According to figure 12 it is assumed that a > b. This result was part of the previous year’s course: see
omnicki 1934, §28, page 47.]
13.4 §4 Construction Problems 289

When these conditions are satisfied it is possible to construct two congruent triangles
that satisfy the conditions of the problem.17

As we have explained in this example, a construction problem consists of the following

parts: (1) analysis, (2) construction, (3) proof, (4) determination of conditions for possibil-
ity and of the number of solutions.

Often the analysis is so short that we omit it and proceed at once to construction.

Exercises. 1. Construct a right triangle given a leg a and the hypotenuse b. Can
the problem be solved for all a and b?
2. Construct a triangle given two edges and the projection of the shorter of these onto
the longer. Investigate the possibility of the conditions of the problem.
3. Construct a triangle given an edge, an angle, and the median to the one remaining
edge that does not pass through [the vertex of ] this angle.
Choose the given segments so that
there exist (a) two solutions, (b) one,
[and] (c) so that there should be no
solution.
4. Construct a right trapezoid given
one of the bases a, edge b perpendicu-
lar to both bases, and edge c. Choose the c b
second edge so that (a) there are two
a
solutions, (b) there is one solution,
(c) there is no solution at all.
Hint. The construction and the inves-
Figure 13
tigation of the existence and number of
solutions is explained by figure 13.
5. Construct a triangle given a median a to an edge, and each of the angles that the
median forms with the remaining edges.
Hint. Perform the analysis of the problem by extending the median as in a problem
solved earlier.
6. In the interior of an arbitrary concave angle18 choose any point A, and pass through
A the line whose segment contained between the arms of the angle is bisected by the
point A.
Hint. The problem comes down to the construction of a parallelogram, two of whose
adjacent edges lie on the arms of the given angle and whose diagonals intersect at
point A.

17
[This sentence is puzzling: infinitely many triangles, all congruent, satisfy the requirements. The auth-
ors’ term for congruent was przystajcy. The meaning is unclear even if by that they meant indirectly
congruent, referring to a solution obtained by choosing BU in place of B, for it is possible to construct
from any solution an indirectly congruent one.]
18
[A concave (wklsy) angle must be one whose measure m satisfies 0( < m < 180(.]
290 13 Geometry for the Gimnazjum

7. Choose any three segments a, b, and c. Draw three circles, one with radius a,
the second, b, and the third, c, so that each of them should be externally tangent to the
other two.
Is the construction possible for all a, b, and c?
8. Construct a right triangle given the hypotenuse and the sum of the legs.
Hint: extend a leg beyond the vertex of the right angle and measure off on the exten-
sion a segment equal to the second leg.
9. Draw a right triangle given an acute angle and (a) the sum of the legs, (b) the
difference of the legs.
Hint. Perform an analysis and extend one of the legs beyond the vertex of the right
angle by a distance equal to the other leg, [and] join the endpoint of this extension with
the vertex of the right angle.19 What angle results?
10. Construct a trapezoid given its two bases and the two diagonals.
Hint. Construct a triangle with a base equal to the sum of the trapezoid’s bases and
edges equal to the diagonals.
11. Construct a square given the sum of a diagonal and an edge.
Hint. Draw a square ABCD, construct a diagonal, for example AC, extend edge
AB beyond20 point A by the distance AE = AC, construct segment EC, [and] prove
that ­ ACE = ­ AEC = 22(30U.
12. Draw two parallel lines AB and CD, choose a point E, and on the line CD mark
a point equidistant from point E and the line AB.21
Choose the point E so that (a) there exist two such points, (b) one such point, (c) no
such point.
13. Construct a parallelogram given the two diagonals and the angle included between
them.
14. Construct a trapezoid given its four edges a, b, c, d (a and c are the bases).
Hint. Construct a triangle one edge of which equals the difference between the bases
of the trapezoid, the others being equal to b and d.

19
[The hint must be mistaken: try joining the endpoint with another vertex instead.]
20
[Segment AB was specified incorrectly as AC in the original.]
21
[This occurrence of AB was printed incorrectly as CD in the original.]
13.5 Summary of §5–§24 291

13.5 Summary of §5–§24

This box describes the sections of the Geometrja text that lie between the previous
and next translated sections.

Some of these sections continue the review and presentation of very elementary
topics begun in the previous geometry course and in §1–§4:
§5 tangents to a circle §11–§13 parallel lines
§6 perpendicular bisector §24 rational arithmetic
§7 construction problems
In another group of sections these concepts are applied in somewhat more involved
situations:
§8 –§10 circumcircle, angle bisectors, incircle
§14–§19 circular arcs and chords and related angles
§20–§21 construction problems involving these
Two sections can be regarded as intermediate goals for that development:
§22 inscribed and circumscribed quadrilaterals
§23 regular polygons

These intervening sections present the standard theorems on their subjects. The
abundance of exercises is impressive. The simplicity of language is startling,
especially when contrasted with that of Enriques and Amaldi [1903] 1916. In these
sections the authors of Geometrja continued their practice of discussing logical
principles in the context of geometry. For example, the use of necessary and
sufficient conditions is discussed in detail in §6 in connection with equidistance from
two points and incidence with their perpendicular bisector, the notions of converse
and contrapositive are applied to organize §22, and conditions for existence and
uniqueness of solutions to construction problems are a recurring theme.

Section §24 introduces the notion of ratio of commensurable segments, and begins
to relate operations on segments and operations on ratios.
 PART THREE

Measuring Segments and Areas

§ 25. Incommensurable Segments—Irrational Numbers

So far, we have been considering commensurable segments. We shall now prove that
there are segments that are not commensurable.

We shall prove that [these] segments are not commensurable: the edge AC = b of a
square, and its diagonal BC = a (figure 75).

C
D

E F

B A

Figure 75

We suppose that a and b are commensurable, then let

l
a= b.
m

We divide BC into l equal parts. Segment CE, containing m such parts, is equal
to b:
EC = b.
13.6 §25 Irrational Numbers 293

Through the subdivision points of BC we draw parallels to AB. These lines divide
CA into l equal parts (recall the method for dividing a segment into equal parts). [Let]
CF include m of those parts: thus,
m
CF = AC. (1)
l

From point A we drop AG ] BC; therefore

BG = GC = 1 2 a.

Right triangles CFE and CGA are congruent, because AC = EC = b [and] ­ ECA
is common [to them]; therefore, CF = CG = 1 2 a. By virtue of (1) we would have
m
1 2 a= b;
l
l
substituting a = b, we would obtain
m
l m
b= b, or
2m l
2
§ l ·
¨ ¸ = 2. (2)
©m¹
Let us suppose that l/m is a fraction in irreducible form. We shall prove that equation
(2) is impossible.
2
From (2) it would follow that l / m2 = 2, [and] l 2 = 2 m2.

Since the right-hand side of this equation is divisible by 2, the left side would also be
divisible by 2, and hence l would be divisible by 2. Then l = 2 l1 for some whole
number l1 ,
l 2 = 4 l12, and thus
4 l12 = 2 m2 , 2 l12 = m2.
The left side of the last equation is divisible by 2; therefore, the right [side] would also
be divisible by 2, and thus
m = 2 m1 , for some whole number m1 .
We see that the fraction l/ m can be reduced by 2, which would contradict the assump-
tion that l/ m should be an irreducible fraction.

Thus, the supposition that BC and AC are commensurable led us to a contradiction.

Therefore, BC and AC are not commensurable.

We can provide many examples of pairs of segments that are not commensurable.
294 13 Geometry for the Gimnazjum

C You can prove, for example, that in

an equilateral triangle (figure 76) the
edge AB and the altitude CD are not
30( commensurable.22

Segments that are not commensurable

are called incommensurable.

60( Let us take segment AC for the unit

A D B length (figure 75 [page 292]). Because
segments BC and AC are incommen-
surable, there can be no relationship
Figure 76
between these segments [in the sense
of §24].

We assume that the length BC with respect to the unit AC is a number. But it
cannot be a whole or fractional number (that is, a ratio of whole numbers), because this
would mean that AC and BC would be commensurable. Therefore, this is a new kind
of number. Previously we did not know about such numbers, and only now do we become
familiar with them.

Whole and fractional numbers, which are measures of segments commensurable with
the unit, are called rational numbers; and those numbers that are measures of segments
that are incommensurable with the unit, we call irrational.

Taking AC = 1 decimeter (figure 75) as unit of length, construct a square as in figure

75, as accurately as you can, and measure its diagonal with the aid of a millimeter scale.

The length of the diagonal of this square, as we know, is an irrational number.

You will convince yourself that the desired length of the diagonal is
greater than 1 decimeter [dm] and less than 2 dm
" " 14 centimeters [cm] = 1.4 dm " " " 15 cm = 1.5 dm
" " 141 millimeters [mm] = 1.41 dm " " " 142 mm = 1.42 dm.

With [ just] the aid of a millimeter scale you cannot determine the length of the
diagonal any more accurately. With more accurate measuring devices, you could more
accurately identify bounds between which is included the length of the diagonal.

If we want the longer segment to have greater length,23 it is necessary to have accepted
that the number which is the length of the diagonal (we denote this number by D) is
larger than the numbers 1, 1.4, 1.41, and less than the numbers 2, 1.5, 1.42 —or

22
[In this sentence in the original, BC was misprinted for CD.]
23
[The present editors are puzzled by this phrase—Jeeli chcemy, aby wikszy odcinek mia wiksz dugo
in the original.]
13.6 §25 Irrational Numbers 295

1 < D< 2
1.4 < D < 1.5
1.41 < D < 1.42 .

Similarly, having constructed an equilateral triangle whose height is equal to 1 dm,

we would find that the number E (beta) that is the length of the edge of the triangle is
contained within the following bounds:
1 < E< 2
1.1 < E < 1.2
1.15 < E < 1.16 .

Carry out the indicated construction and convince yourself of it.

The numbers that are lengths of segments incommensurable with the unit are called
the positive irrational numbers. Alongside them we consider the negative irrational num-
bers: namely, we assume that to any positive irrational number, such as J (gamma),
corresponds the number – J (minus gamma). From now on, speaking of numbers, we
shall have in mind both positive and negative numbers, rational or irrational.

We assume that irrational numbers can be compared with each other [according to size]
and with rational numbers, and that the four arithmetic operations can be performed on
irrational numbers; moreover, these operations are governed by the same laws that
govern [corresponding] operations on the rational numbers.

Later we will learn how, in practice, to compare irrational numbers and perform
operations on them.

Exercises. 1. We have shown that there is no rational number l/ m such that

( / m) 2 = 2. Prove in a similar way that there is no rational number l/ m such that
l
( l/ m) 3 = 2 or ( l/ m) 3 = 4.
2. Let E be an arbitrary irrational number and J, any rational number. Prove that
/ 0, then E # J, E/ J , and J/ E
E + J and E – J are irrational numbers. Similarly, if J =
must be irrational. Hint: let E + J = G, so that E = G – J. What could be said about
E, if not only J but also G should be a rational number?
3. On the basis of the previous exercise, and knowing that the number D that was
introduced previously is irrational, prove that D + 1, D + 2, ... , 2 + D, 3 + D, 2 D, 2/3 D,
1/D , 2/D , ... are irrational numbers. How many different irrational numbers are there?
From among these indicated irrational numbers choose two whose sum, difference,
product, and quotient, respectively, are rational numbers.24

24
[The exercise was evidently garbled in the original: no two numbers in that list have a rational sum.]
296 13 Geometry for the Gimnazjum

§ 26. Measuring Incommensurable Segments with a Unit

We have established certain laws regarding segments whose lengths are commensurable
with the unit. Now we suppose that these laws remain in force, but in this case the seg-
ment that we measure is incommensurable with the unit; thus we assume what follows.

Having selected an arbitrary segment as the unit of length, [we may assign] to each
segment a number, called the length of the segment with respect to the given unit.
Moreover, the following laws are satisfied.

1. To every segment corresponds exactly one positive number (rational or irrational),

which is its length; we also call this number the ratio of that segment to the unit.

2. Every positive number corresponds to some segment, whose length is the given
number.

3. The length of the segment chosen as unit is the number 1.

4. Equal segments have the same length.

5. The length of the sum of two segments is equal to the sum of the lengths of these
segments.

6. The ratio of any two segments equals the ratio of their lengths.25

From the above assumptions follow [some] conclusions.

(1) The larger segment has the greater length. Rather than compare segments we can
compare their lengths.

(2) The length of the difference of two segments is equal to the difference of their
lengths.

Let us compare the two irrational numbers D and E [on page 295 ] with each other:
because E < 1.2 [and] 1.4 < D it follows that E < 1.2 < 1.4 < D —that is, E < D.

Exercises. 1. Prove that if we reduce the unit of length n-fold, the number express-
ing the length of the segment will increase n-fold.
Hint. Let a be [the length of] an arbitrary segment and b, the length using the
reduced unit; we apply condition 6.

25
[The ratio of two commensurable segments was already defined in §24. Condition 6 connects length with
that concept, and defines the ratios of incommensurable segments.]
13.8 §27 Rational Approximations of Irrational Numbers 297

2. We know that equal segments have equal length, and of two unequal segments the
longer has greater length and the shorter, less. Formulate the converse theorems: are
they true, and why?
Hint: a closed system of theorems. [See §3, page 283.]

§ 27. Rational Approximations of Irrational Numbers

Operations on Irrational Numbers

In §26 we found decimal numbers between which lie irrational numbers D and E. We
can find rational numbers arbitrarily close to any irrational number. These numbers are
called approximations of the irrational numbers.

We establish some properties of irrational numbers. The following properties stem

from theorems about incommensurable segments. Those results are presented at the end
of the present section.

Property 1. For each positive irrational number J, given a whole number

m we can find a whole number l such that J is contained between l/m and
(l + 1) / m .

The numbers l/m and (l + 1) / m are called the approximations of the number J with
precision 1/m — l/m from below, (l + 1) / m from above.

We have already learned about decimal numbers, which are decimal approximations
of the numbers D and E.

Property 2. If for each value of m two irrational numbers J and G have

the same approximations
l l +1
and
m m

with precision 1/m, then the numbers are equal to each other: J = G.

In practical calculations we can replace irrational numbers by their approximations

and perform appropriate operations on the approximations. The results of the operations
performed on approximations are approximate. If we add approximations of the numbers
D and E from below, we obtain an approximate value of the sum D + E, also from below;
and adding approximations of the numbers D and E from above, we obtain an overesti-
mate of the sum D + E. For example,
1.41 + 1.15 < D + E < 1.42 + 1.16
2.56 < D + E < 2.58
2 28/ 50 < D + E < 2 29/ 50 .
298 13 Geometry for the Gimnazjum

In this way, we found approximations of the sum D + E with precision 1/ 50 .

Subtracting from an approximation of the number D from below an approximation

to the number E from above, we obtain an approximation of the difference D – E from
below; subtracting from an approximation of the number D from above an approxima-
tion of the number E from below, we obtain an approximation of the difference D – E
from above:
1.41 – 1.16 < D – E < 1.42 – 1.15
0.25 < D – E < 0.27.

For the product D E and the quotient D : E we can write

1.41 # 1.15 < D E < 1.42 # 1.16

1.41 D 1.42
  .
1.16 E 1.15

Exercises. 1. As unit of length we take an edge of a square. Approximate the lengths

of the following segments from above and below: (1) half of the diagonal of the square,
(2) the sum of a diagonal and two edges of the square, (3) the difference of a diagonal and
an edge of the square.

2. As unit of length we take half of an edge of the equilateral triangle in figure 76 [on
page 294]. With precision one tenth, approximate the lengths of the radius of the
inscribed circle of this triangle and the radius of the circumscribed circle.

We can deduce property 1 of irrational numbers from the following theorem.

Theorem 1. For each segment AB incommensurable with the unit CD two

segments commensurate with CD can be found, one of which is longer and one
shorter than AB, and whose difference is equal to 1/ m CD, where m is any
whole number.

A EB F

C 1 D

Figure 77
13.8 §27 Rational Approximations of Irrational Numbers 299

We divide segment CD (figure 77) into m equal parts [and] place one part [thus]
obtained—segment a —in segment AB, [starting] from point A, as many times as
possible. We know from last year that by repeating this placement enough times, we will
cross point B. Suppose that we place segment a [there] l times without reaching point
B, but cross B on the (l + 1)st time, so that we can write
l l 1
# CD < AB < # CD . . . . . . . . . . . . . . . . . . . . . (1)
m m
l l 1
AE = # CD AF = # CD,
m m
[and] therefore
AE < AB < AF.

Thus, AE and AF are the desired segments.26

Let the ratio AB/CD = the irrational number J; from (1) follows

l l 1
< J< .
m m
From this, property 1 of irrational numbers follows directly.

The proof of property 2 is based on the following theorem.

Theorem 2. Given two unequal segments incommensurable with the unit

CD, we can find a segment, commensurable with the unit, that is smaller than
one of them and larger than the other.

A FB G E

Z e Y
C 1 D

Figure 78

We place these segments, AB and AE, as indicated in figure 78. We place segment
BE = e in CD, [starting] from point C, [ just] as many times as needed to cross point
D; suppose, to this end, it is necessary to place the segment m times, so that
CD < me.

26
[This and the preceding four lines were out of order in the original.]
300 13 Geometry for the Gimnazjum

Dividing CD and me by m, we get

(1/ m) CD < e;
thus, dividing CD into m equal parts, we obtain one such part a, smaller than e.

We place segment a in AB, [starting] from point A, [ just] as many times as needed
to cross point B. Let F be the end of the last segment before B, [and let] G be the end
of the next segment, containing B:
FG = a = FB + BG < e,
and thus, moreover,
BG < e.
Therefore, point G lies between B and E, and so segment AG is the required segment.

Letting the segments AB and AE have measures J and G (irrational numbers), we

can say this:

Given two unequal irrational numbers, we can find some rational number
that is larger than one of them and smaller than the other—that is, which is
contained between these numbers; we denote this rational number by w. Thus,
J < w < G.

We say that between two unequal irrational numbers we can can locate at least one
rational number.

Now property 2 can be easily proved. Indeed, let us suppose that J =

/ G —for instance,
J < G —then we can locate between J and G a rational number k / n :
J < k / n < G.
In this case the approximations of the numbers J and G with precision 1 / n would
be different, [since k / n would be larger than the approximation of the number J from
below with precision 1 / n and smaller than the approximation of the number G from
above with precision 1 / n .] This 27 is contrary to the assumption that the approximations
of numbers J and G are the same, so J < G cannot hold; similarly, J > G cannot hold,
and therefore J = G.

27
[The present editors provided the text in brackets to replace a confusing passage in the original.]
13.9 Summary of §28–§32 301

13.9 Summary of §28–§32

This box describes the sections of the Geometrja text that lie between the previous
and next translated sections.

The theorem of Thales presented in §28 is not the familiar proposition that an
angle inscribed in a semicircle is right, but rather the following one: if two pairs of
distinct parallel lines intersect two other lines at points A, B; C, D and E, F; G, H
respectively, then
AB EF
.
CD GH
For the case of congruent AB, CD this result was discussed earlier, in §11. In §28
the authors extended it easily to the commensurable case. For the incommensurable
case they showed that for any positive integers m and n,
m AB m1 m EF m1
  implies   .
n CD n n GH n
By property 2 in §27 this yields the equality of the ratios.

As consequences, §28 derives the converse of Thales’s theorem and the following
result:
Given a triangle ABC and points BU = / A and CU = / A on rays
AB and AC respectively, then BC BUCU if and only if AB/AUBU =
AC/AUCU; and in that case these ratios are also equal to BC/BUCU.

The next section, §29, defines similarity of polygons in general and derives from
this result the familiar AA, SAS, and SSS similarity theorems for triangles. Those
two sections together contain fifty-one exercises, which are supplemented with ten
more construction problems in §30. Two practical applications are discussed in §31:
they use similarity to measure indirectly some objects, portions of which are inacces-
sible. The next section, §32, discusses one more application, the pantograph,
continues with eighteen more excellent exercises, and contains one additional result
that is required for the area theory later in the book:
If through corresponding vertices in two similar convex polygons
we pass all the diagonals, then they will divide the polygons into the
same number of triangles, respectively similar and identically
arranged.
302 13 Geometry for the Gimnazjum

§ 33. Equivalent Polygons

If two polygons I and II (figure 106) have no interior points in common and polygon
III consists of all points of polygons I and II and contains no other points, we call polygon
III the sum of these polygons (components).

In figure 107 polygon IIIU [is] the sum of polygons IU and IIU. Adding polygon 3 to the
sum of polygons 1 and 2 (figure 108), we obtain the sum of three polygons. In a similar
way we can obtain the sum of any number of polygons.28

If polygon W is the sum of two polygons W1 and W2 then we call W1 a difference

of polygon W and polygon W2 .

For each of the polygons W1 and W2 , we say it is part of polygon W.

If a given polygon is the sum of two congruent polygons (figure 109) then each of these
polygons we call a half of that polygon; and if it is the sum of n congruent polygons [then
each of these polygons we call] an nth part of that polygon (figure 110).29

Polygons that are sums of polygons congruent in pairs (figures 106 and 107) we call
equivalent: for example, polygons III and IIIU.

Equivalent polygons, for example III and IIIU, might not be congruent.

Exercises. 1. Represent any triangle as the sum of (1) two triangles, (2) a triangle
and a quadrilateral, (3) two quadrilaterals, (4) a quadrilateral and a pentagon, (5) two
pentagons.
2. Represent any triangle as a difference of (1) two triangles, (2) a triangle and a
quadrilateral, (3) a quadrilateral and a pentagon.
3. Divide any rectangle (1) into two equal trapezoids, (2) into five equal rectangles,
(3) into six equal triangles.
4. Divide any isosceles triangle into two, into four, and into eight equal triangles.
5. A square with edge 6 cm and a rectangle with edges 4 cm and 9 cm are given. Prove
that these two quadrilaterals are equivalent, dividing each of them into three respectively
equal rectangles.

28
[The authors evidently regarded the associative and commutative laws for this addition operation as
inappropriate for discussion in the third gimnazjum class.]
29
[This sentence was garbled in the original; the authors may have intended to write “1/n czci” for
“nth part.” ]
 I

IIU
$$$
III
II $$$
IIIU

Figure 106 IU

2
1
Figure 107

Figure 108

Figure 110

Figure 109
304 13 Geometry for the Gimnazjum

6. A square with edge 6 cm is given, as well as a rectangle with edges 4 1 2 cm and 8 cm.
Prove that these two quadrilaterals are equivalent, dividing each of them into four
respectively equal rectangles.
7. A triangle ABC is given. Construct a parallelogram equivalent to the given tri-
angle, and having with it the common edge BC and common angle B. Hint: let D be
the midpoint of AB, and E, the midpoint of AC; if we extend DE by EF = DE, then
BCFD will be the desired parallelogram. Why?
8. A rectangle with edges 10 cm and 4 cm is given, as well as a parallelogram with edges
of 10 cm and 8 cm and acute angle 30(. Show that these two quadrilaterals are equivalent.
(Hint: place the quadrilaterals so that they have a common edge, and prove that the edges
of the quadrilaterals parallel to the common edge must lie on one line.)

§ 34. On Measuring Areas of Polygons

We have already learned to measure segments: that is, to assign each segment a number
called its measure or length. Not just segments can be measured, but also polygons: that
means, assign each polygon a number called its measure or polygonal area.

With some polygon selected as the unit of area, each polygon can be assigned a number
called the area of the given polygon with [respect to] the given unit.

Let us agree to adopt as unit the area of a square whose edge is equal to the unit length.
And thus, if we take as unit the length 1 cm, 1 m, or the like, then we adopt as the unit
area the area of the square with edge 1 cm, 1 m, or the like; we denote these units by
1 cm2, 1 m2, and the like.

It is necessary to remember this well while reading theorems that speak simultaneously
of lengths of segments and areas of polygons.

Area of a Rectangle

At the beginning, we shall deal with measuring the area of a rectangle. One of the edges
of a rectangle we call the base and [one] perpendicular to this edge, the altitude.

The area of the unit square is equal to 1.

Consider a rectangle having edges 3 and 4 units in length (figure 111). This rectangle
is the sum of 12 unit squares (3 # 4).
13.11 §34 Measuring Areas of Polygons 305

Let us suppose that

(1) congruent polygons have the same area;
(2) a polygon that is the sum of several polygons has area equal to the sum
of areas of those polygons.
Then the area of that rectangle equals 12 —that is, 3 # 4.

The area (figure 112) that comprises 1/ n of the unit square equals 1/ n, and the area
of the rectangle that comprises m/ n parts of the unit square equals m/ n.

Figure 111 Figure 112

Now we prove the following theorem:

Theorem. The area of a rectangle equals the product of the length of the base
by the length of the altitude.

Let us take a rectangle whose base has length 1 and altitude, length a. Consider two
cases:
(1) a is a rational number,
(2) a is an irrational number.

[In case] (1) a is a rational number—for example, a = l/ m , where l and m are

integers.

As shown in figure 113 [on the following page], we divided the altitude of the unit
square into m equal parts, the altitude of rectangle ABCD contains l such parts. We
denote the area of rectangle ABCD by P.

Through the points of division we passed parallels to the bases, which divided the
square and rectangle into equal rectangles. The area of one such rectangle = 1/ m ; there-
fore, P = l/ m —that is, P = a.
 B C

a P 1
___
l+1
m
_l
m
1
_l
m
A 1 D a P 1

Figure 113

Figure 114

b
D C F

a P
b
D C

A 1 B E
a
Figure 115

A 1 B _l ___
l+1
m m

Figure 116
13.11 §34 Measuring Areas of Polygons 307

[In case] (2) a is an irrational number (figure 114); let the approximations of this
number with precision 1/ m be l/ m and (l+1)/ m , so that
l l 1
a .
m m

The segment with length a is longer than the segment with length l/ m but shorter
than [that of length] (l+1)/ m . Then, as can be seen in the figure,
l l 1
P .
m m

We see that the numbers a and P have the same approximations [with precision]
and specifically l/ m and (l+1)/ m . Moreover, this will occur for every integer value
1/ m ,
m; therefore, according to property 2 of irrational numbers in §27 [page 297], numbers
a and P are equal:
P = a = 1 # a.

We see that if the base of the rectangle = 1, our theorem is true.

Now let us take a rectangle whose base has length b and altitude, length a —more-
over, a can be rational or irrational. Consider the two cases: (1) b rational (figure 115),
(2) b irrational (figure 116).

[In case] (1) b [is] rational: for example, b = l/ m (figure 115).

Denote by P the area of the rectangle with edges a and b, and by PU [that of]
rectangle ABCD with base 1 and altitude a . On the basis of the previous [discussion],
PU = a.

Lines parallel to AD (figure 115) divide AB into m equal parts; in b [the number
of] such parts is l. These lines divide the area of rectangle ABCD into m equal parts,
each of which is equal to ( 1/ m) a; the area of rectangle AEFD contains l such parts.30
Therefore,
l
P= PU,
m
and since PU = a,
l
P= a = b a = a b.
m

[In case] (2) b [is] an irrational number (figure 116); then, dividing AB into m equal
parts we obtain

30
[The authors neglected to specify points E, F in figure 115.]
308 13 Geometry for the Gimnazjum

l l 1
b .
m m
Also, for areas, as seen in figure 116, we have
l l 1
aP a.
m m

Dividing by a, we obtain
l P l 1
  .
m a m

We see that the numbers b and P/ a are contained between l/ m and (l+1)/ m ;
moreover, this occurs for each value of m. In view of this, these numbers are equal:
P/ a = b, or
P = a b.

Thus, we have proved the theorem in all cases.

Area of a parallelogram, triangle, trapezoid, and polygon

In a parallelogram, we call any two parallel edges the bases; then [we call] a segment
that measures the distance between the bases an altitude. For example (figure 117),
AB is a base of the parallelogram ABCD [and] CE is an altitude.

D C E F D C

A B E A B

Figure 117 Figure 118

Theorem. The area of a parallelogram is equal to the product of the length

of a base by the length of an altitude.
13.11 §34 Measuring Areas of Polygons 309

Parallelogram ABCD (figure 118) has base AB and altitude AE. Let us construct
the rectangle ABFE having base AB and altitude AE. We shall represent the area of
trapezoid ABCE as the sum of areas of polygons in two ways:
area ABCE = area ABCD + area ADE = area ABFE + area BCF.
Since ' ADE = ' BCF, it follows that area ABCD = area ABFE.

Therefore, the area of the parallelogram is equal to the area of the rectangle sharing
a common base and altitude [length]. Thus, it is equal to the product of the length of a
base by the length of an altitude.

Corollary 1. Parallelograms having equal bases and equal altitudes have equal areas.

Corollary 2. Dividing the area of a parallelogram by the length of a base, we obtain

the length of an altitude. Therefore, if two parallelograms have equal areas and a common
base and lie on the same side of the common base, then their other bases lie on one line
(since the altitudes of the parallelograms are equal).

Theorem. The area of a triangle is equal to half the product of the length of
a base by the length of the altitude.
B E
Given triangle ABC (figure 119) with
base AC and altitude BD, by construct-
ing [lines] parallel to its edges we can cre-
ate a parallelogram ABEC, half of which
is triangle ABC and which has with this
triangle a common base and altitude.

Let a, h, S denote the length of a A D C

base, length of the altitude, and the area
of the triangle: then Figure 119
S= 1 2 a h.
B

Corollary. Triangles having equal

bases and altitudes have equal areas.

Problem. Replace a triangle ABC by D G E

a rectangle with equal area (figure 120).

If we take a base AC of the triangle as

base of the desired rectangle, then half of
the altitude BF of the triangle—that is,
GF —as an altitude of the rectangle, then A F C
ADEC is the desired rectangle.
Figure 120
310 13 Geometry for the Gimnazjum

Theorem. The area of a trapezoid is equal to half the product of the sum of
the lengths of its bases by the length of an altitude.

In trapezoid ABCD (figure 121) parallel edges AB and CD are the bases; an altitude
is DE ] AB. We join D with the midpoint F of the edge CB, and extend DF to [its]
intersection with the line AB at point G.

We have ' FDC = ' FGB (FC = FB, ­1 = ­2, ­3 = ­4); 31 thus BG = CD [and]
area ADCB = area ADFB + area FDC = area ADFB + area FBG = area ADG.

We denote the length of AB by a, of DC or BG by b, [and] of DE by h.

The area of triangle ADG equals half the product of the length of AG (that is,
a + b) by h, [and thus]
area ADCB = 1 2 ( a + b) h.

D C

4
2 F
1
3

A E B G

Figure 121

To calculate the area of a polygon, we decompose it as usual into triangles (figure 122)
or trapezoids (figure 123) and calculate the sum of the areas of the triangles or trapezoids.

If the polygon is regular, the calculation of its area can be simplified. Let AB (figure
124) be the edge of a regular polygon with n edges; O, the center of the circumscribing
circle of the polygon; [and] OC, an apothem.

We know that rays OA, OB, and so on, divide the polygon into n congruent triangles.
Denoting the length of an edge of the polygon by a, of an apothem by r, [and] the area
by S, we have
S=n# 1 2 ar = na # 1 2 r, and therefore

The area of a regular polygon equals the product of the length of its
perimeter by half the length of an apothem.

31
[In the original, point G was incorrectly labeled as ­3.]
 B

Figure 122

G
H

F
I

Figure 123

A C B

Figure 124
312 13 Geometry for the Gimnazjum

Problem. Replace a polygon by a rectangle having area equal to the area of the
polygon.

In polygon ABCDE (figure 125) we C

draw a diagonal CE and a parallel DF
to it through vertex D, to its intersec-
tion with the extension of AE. Triangles B
CED and CEF have the same base CE
and the same altitude; therefore their D
areas are equal. If we remove triangle
CDE from polygon ABCDE and replace
it by triangle CEF, we will obtain the
quadrilateral ABCF, which has area
equal to the area of ABCDE. Replacing G A E F
in a similar way triangle ACB by trian-
gle ACG, we replace quadrilateral
ABCF by the triangle GCF, which has Figure 125
area equal to the area of ABCDE. Using
a familiar method we replace triangle
GCF by a rectangle with the same area.

Exercises. 1. In square ABCD, chosen as the unit area, O is the point of intersec-
tion of the diagonals; S, the midpoint of edge AB; and T, the midpoint of edge CD.
What areas should be assigned to (1) rectangle ASTD, (2) triangle ABC, (3) triangle
AOB, (4) triangle AOS, (5) trapezoid ASOD?
2. A rectangle whose one edge is the unit length and the second, three times larger,
is divided into four triangles by the diagonals. What are the areas of an undivided triangle
and each of the four triangles?
3. Given a square, construct a triangle with an area equal to one eighth of the square.
4. One edge of a parallelogram is equal to an edge of a square, and a second, to the
diagonal of the square; the acute angle of the parallelogram is equal to half of a right
angle. Show that the square and the parallelogram have equal areas.
5. Given a parallelogram, construct a triangle that should have an edge in common
with the parallelogram, as well as a common angle adjacent to this edge, and which
should, in addition, have area equal to the area of the parallelogram.
6. Construct a square that should have an area two times larger than the area of a
given square.
7. Through an arbitrary point E lying on the diagonal AC of a rectangle ABCD are
drawn two lines parallel to the edges of the rectangle. One of them intersects the edge
AB at point F and the edge CD at point H; tehe second [intersects] edge BC at point
G and edge AD at point I. Prove that rectangles FBGE and IEHD have equal areas.
13.11 §34 Measuring Areas of Polygons 313

Hint: represent the considered rectangles as the differences of [respectively] equal

triangles FBGE = ABC – AFE – EGC and similarly for IEHD.
8. Indicate on a drawing made for the previous exercise two more pairs of rectangles
with equal areas.
9. Can the theorem proved in exercise 7 be generalized, considering instead of the
rectangle an arbitrary parallelogram?
10. Prove the theorem: if a rectangle with consecutive edges a and b and a rectangle
with consecutive edges c and d have equal areas, then c : a = b : d. Formulate and
prove the converse theorem.
11. Given a rectangle with edges a and b; relying on the previous exercise, construct
a rectangle of equal area, one edge of which would be the segment c. Hint: we apply one
of the methods [in §28] for constructing the fourth proportional.
12. Explain geometrically the equation (a + b)2 = a2 + 2 a b + b2.
13. Prove that the area of a rhombus is equal to half of the product of the lengths of
its diagonals. Generalize this theorem, considering instead of a rhombus a kite (quadri-
lateral of which two consecutive edges are equal and the other two consecutive edges are
also equal).32
14. Prove that the area of a trapezoid with two nonparallel edges is equal to the length
of an altitude, multiplied by the length of the segment joining the midpoints of the
nonparallel edges.
15. Show that a median of a triangle divides this triangle into two triangles with equal
areas.
16. Divide a triangle into five triangles of equal areas with the help of segments passing
through one vertex.
17. In the convex quadrilateral ABCD, point O is the midpoint of the diagonal AC.
Show that quadrilaterals ABOD and BODC have equal areas.
18. Investigate whether the statement asserted in the previous exercise remains true
in the case in which quadrilateral ABCD is concave.
19. Two parallelograms ABCD and ABEF have equal areas and lie on the same side
of the line AB. Show that edges CD and EF must lie on the same line.
20. A parallelogram ABCD is given. Construct a parallelogram ABEF whose area
should be equal to the area of ABCD, and whose diagonal AE should be equal to a given
segment k. Investigate whether the problem is always possible and how many solutions
it has.
21. Two triangles ABC and ABD have equal areas and lie on the same side of the
line AB. Show that AB CD.

32
[The authors’ word for kite was deltoid.]
314 13 Geometry for the Gimnazjum

22. Triangle ABC is given, and an angle D. Construct a triangle ABD with area
equal to the area of the triangle ABC. Give the condition of solvability of the problem
in the case in which the angle D is right.
23. Triangle ABC is given. Construct a rectangle with equal area, whose base should
be a given segment k. Hint: first we construct a rectangle of equal area and base AB,
then apply exercise 11.
24. Solve the previous problem, taking instead of a triangle any polygon—for example,
a pentagon. Hint: divide the pentagon into triangles and for each of those construct a
corresponding rectangle; next, we add these rectangles to each other in order to obtain
as sum a rectangle with the same base.
25. Two arbitrary polygons are given; we want to convince ourselves which of them
has the greater area not with the help of measurements, but with geometric constructions.
Based on exercise 24, describe how we should proceed.
26. A parallelogram has acute angle 30( and edges [with lengths] 12 cm and 6 cm.
Calculate its area.
27. Two heights 33 h and l of a parallelogram are given, coming from a vertex A and
edge AB = c. Calculate the remaining edges of the parallelogram.
28. The area of a rhombus is 42 cm2 and one of the diagonals [measures] 7 cm.
Calculate the [length of the] second diagonal (see exercise 13).
29. A diagonal of a parallelogram measures 50 cm and forms a 30( angle with an edge
[having length] 36 cm. Calculate the area of a parallelogram.
30. In a triangle with base AB = 40 cm a rectangle with edges 10 cm and 8 cm has
been inscribed so that the longer edge lies on AB and the ends of the opposite edge of
the rectangle lie on the other edges of the triangle. Calculate the area of triangle ABC.
31. In an isosceles trapezoid the bases are equal to 56 cm and 42 cm. Calculate the
area of the trapezoid, knowing that the arms form 45( angles with a base.34
32. Calculate the area of an isosceles trapezoid with height 56 cm, knowing that its
diagonals are perpendicular to each other and their point of intersection divides them in
the ratio 2 : 5.
33. In an isosceles triangle one arm is 8 cm and the angle between the arms is equal
to 150(. Calculate the area of the triangle.
34. The edges of a parallelogram measure a = 5 cm and b = 8 cm. An altitude,
dropped to the longer edge, measures h = 3 cm. How long is the altitude dropped to the
shorter edge?

33
[Here, height means length of altitude.]
34
[Some text in this exercise was evidently scrambled in the original; the present editors have tried to
restore it.]
13.11 §34 Measuring Areas of Polygons 315

35. The bases of a trapezoid have lengths a = 7 cm and b = 11 cm; the height h =
4 cm. Calculate the area of the trapezoid, and the area of the triangle that is formed if
its edges are extended to their intersection.
36. Prove that the area of a trapezoid having two nonparallel edges may be expressed
as the product of one of the nonparallel edges by the distance from this edge to the
midpoint of the other edge.
37. An edge of a square measures a. Find the area of the square inscribed in it in such
a way that the vertices of the inscribed square lie on the midpoints of the edges of the
given square.
38. In a triangle ABC in which AB = 13 cm, BC = 14 cm, CA =15 cm, the height
was calculated [as] AD = 12 cm. Find the areas of the parts into which the bisector of
angle A divides the triangle.
39. In parallelogram ABCD the diagonals are given: AC = 18 cm and BD = 12 cm.
The distance from point A to diagonal BD is 5 cm. What is the distance from point
B to diagonal AC?
40. In a trapezoid ABCD in which the bases measure AD = 9 cm and BC = 13 cm
and the height h = 4 cm, a line is drawn parallel to the bases, dividing the area of the
trapezoid in half. How far from AD was this line drawn?
41. In a quadrilateral ABCD in which AB = BC and AD = DC, the diagonals are
given: AC = 8 cm, BD = 15 cm. Joining consecutive midpoints of the edges of this
quadrilateral, we obtain a new quadrilateral. Determine its shape and find the area.
42. In an isosceles trapezoid the bases AD = a = 24 cm and BC = b = 18 cm and
the height h = 14 cm are given. Find the area of the triangle between the edge AB and
the diagonals drawn through A and B.
43. When designing navigation chan-
nels it is assumed that the cross section of A E D
the channel should be 5.5 times as large
as the cross section of the submerged part
of a passing freight ship. Assuming the
ship’s beam h = 9 m and draft b =
1.75 m, as well as the channel depth l =
2.80 m, calculate the width of the channel
at the bottom and at the water level.35
The cross section of the channel is an B C
isosceles trapezoid ABCD (figure 126); in
addition,
36
EB : AE = 3 (EB ] AE). Figure 126

35
[A ship’s beam and draft are its width at its widest point and the depth of water needed to float it. The
authors’ words for them were szeroko and zanurzenie.]
36
[The trapezoid in the original was not isosceles, as required by the text.]
316 13 Geometry for the Gimnazjum

§ 35. Areas of Similar Polygons

Theorem. The areas of similar triangles stand in the same ratio as the squares
of the lengths of corresponding edges.

We have two similar triangles ABC and AUBUCU; moreover, corresponding vertices are
denoted by A and AU, B and BU, C and CU (figure 127). Let the lengths of the bases
be a and aU, and their corresponding heights, h and hU:
area ABC ½ ah a h
º .
area A B C ½a h a h

Since ' ADC  ' AUDUCU (why?),

AC h
.
AC h
From the assumptions, we have
AC AB a
.
AC AB a
Therefore
h a
,
h a
and thus
area ABC a a a2
º .
area A B C a a a2

Theorem. The areas of similar convex polygons stand in the same ratio as
the squares of the lengths of corresponding edges.

As we know, we can decompose similar polygons (figure 128) into similar triangles, with
the help of the diagonals passing through corresponding vertices. The areas of these
triangles stand in the same ratio as the squares of any two corresponding edges of these
polygons. Denoting these areas by P1 , P2, ... , p1 , p 2 , ... , and 37 corresponding edges of these
polygons by a and a1 ,
P1 a2 P2 a2
, , ... ,
p1 a12 p2 a12
and therefore
P1 P2
... .
p1 p2

37
[In the original, the letters P and p were interchanged in this paragraph and the next.]
13.12 §35 Areas of Similar Polygons 317

CU
C p2
p1
p3

AU DU BU P2
P1

P3
A D B

Figure 127 Figure 128

Thus, as we did similarly in §29 for perimeters of similar polygons, we can prove that

P1  P2  " a2
.
p1  p2  " a12

Not proving in general the analogous theorem about similar nonconvex polygons (figure
129), we explain a method for such a proof for similar concave quadrilaterals.

We have two similar concave quadrilaterals ABCD and AUBUCUDU (figure 130).

B BU
DU

EU CU
CU AU
D BU
C
E AU FU DU
B A
C

A F

Figure 129 Figure 130

318 13 Geometry for the Gimnazjum

Denoting the lengths of any corresponding edges by a and a1 ,

area ABCD area ABD  area BCD
, . . . . . . . . . . . . . . . (1)
area A B C D area A B D  area B C D

area ABD a2 area BCD a2

, ,
area A B D a12 area B C D a12
a2 a2
area ABD = # area AUBUDU , area BCD = # area BUCUDU.
a12 a12
Substituting these values in (1), we obtain38
area ABCD a2
.
area A B C D a12

Exercises. 1. Prove the theorem discussed here about areas of similar polygons, when
the polygons considered are nonconvex pentagons. Consider three cases: (1) pentagons
with one concave angle, (2) pentagons with two concave angles adjacent to the same edge,
(3) pentagons with two concave angles not adjacent to the same edge.
2. In two circles are inscribed two regular polygons with the same number of edges.
The radius of one circle is k times as large as the radius of the second circle. What is the
ratio of areas of these polygons?
3. Through the point of intersection of the medians of triangle ABC is drawn a line
parallel to BC, which intersects edge AB at point BU, and edge AC at CU. What part
of the area of triangle ABC is the area of triangle AUBUCU?
4. In trapezoid ABCD points E and F are midpoints of the nonparallel edges AD
and BC, respectively; the base AB is twice as long as the base CD. Show that the area
of the trapezoid ABFE is 1.4 times as large as area EFCD. Hint: we extend the edges
AD and BC to intersect at the point G, and calculate the ratios of the areas of all
triangles and trapezoids that we see in the figure 39 to the area of triangle ABG.
5. Calculate the ratio of the areas of the trapezoids ABFE and EFDC in the previous
problem, assuming that the base AB is four times as long as the base CD.
6. A plot of land has the shape of a pentagon ABCDE. The edges AB = 75 m, BC =
80 m, CD = 95 m, DE = 112 m, EA = 67 m, as well as the diagonals AC = 126 m and
AD = 149 m were measured. Calculate the value of the plot if 1 m2 costs 50 gr.40 Hint:
it is necessary to trace a pentagon with the given segments in a certain scale—for
example, 1 : 1000 —and replace it by a rectangle with equal area.

38
[In the original, equation (1) was typeset incorrectly, in an unfortunate way that seemed well-formed but
appeared to beg the question.]
39
[There was no figure in the original; moreover, point G was called E there.]
40
[1 gr is one grosz: one hundredth of a zloty.]
 14
Teaching and Logic
—To the New World

This book has pursued the following goals.

(1) Publish translations as necessary so that
• Alfred Tarski’s works will all be accessible in English, French, or
German; and
• his geometric works will all be accessible in English.
(2) Provide scientific and cultural background information about the works
translated here: their origin, context, structure, and impact.
(3) Update the Givant 1986 bibliography of Tarski’s publications, and
include an annotated list of major works about Tarski’s life and work.
The preceding portions of the book contain all of the translations and background infor-
mation needed to accomplish the first two goals, except for a few short works, not
concerned with geometry or teaching, that had been published only in Polish. Those are
translated in chapter 15, with some background information for specific items. One of
those short works has to do with politics; the rest, logic. The present chapter contains
some general background for chapter 15 and concludes the biographical material in this
book with an account of Tarski’s 1939 voyage to the New World. The update for Tarski’s
bibliography is in chapters 16–18. For more information on Tarski’s logical work in
Poland and on his career in the United States, consult the books and papers listed in
chapters 17 and 18—particularly the splendid biography, Feferman and Feferman 2004.

14.1 Teaching

During Alfred Tarski’s career in Poland, the major component of his workload was his
teaching at the eromski Gimnazjum. As noted in section 9.2, he probably spent as much
as twenty-four hours per week in classes there, in addition to five hours in lectures and
seminars at the university. After those commitments came class preparation and consul-
tation with students, presumably some administrative chores, and activity with family
and friends. Finally, there was his research and its presentation, evidently the intellectual
high point of his life, which will be discussed in the next section. Little direct record
remains of Tarski’s teaching: most records were destroyed, and his students’ impressions
overshadowed, by the terrors of the coming cataclysm.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 319
DOI 10.1007/978-1-4939-1474-6_14, © Springer Science+Business Media New York 2014
320 14 Teaching and Logic—To the New World

From the years 1937–1939 preceding his graduation from the eromski Gimnazjum,
Tarski’s student Jarosaw Rudniaski
... retained a memory of Tarski as a dignified and serene man (but somewhat nervous), with
a “clouded” face ... . As an instructor, he was patience itself as he set about explaining the
meanders of mathematics, but one saw clearly that he was interested much less in the students
than in the ... mathematics. His lessons usually proceeded this way: Tarski would himself solve
a problem at the board ... while commenting clearly and abundantly about the methods ... .
Then a student (more often than not, an average one) would have to solve a problem of the
same type, and Tarski would analyze the errors committed, according to the maxim, “Let us
learn from others’ mistakes.” In this connection he was loved—but by the gifted students.1

Leon Baszczyk, a eromski student during 1934–1939, reported that Tarski used text-
books by Wadysaw Nikliborc and by Stefan Banach, Wacaw Sierpiski, and Wodzimierz
Stoek. These two series entered publication after Tarski began teaching, in response to
the lack of Polish texts described in section 9.2. Baszczyk also commented that Tarski
was known for always recognizing—even though he never monitored students during examina-
tions—the pairs of works where one was copied to another; after correcting all of them,
he ... confined himself to asking who was the author of the “original.” 2

Witold Kozowski, Tarski’s Gimnazjum student during 1934–1938, has written,

[Tarski] was characterized by the great clarity of his discourse, and he knew how to excite us
in the discussions of various mathematical problems. He familiarized us with concepts such
as complex numbers and the theory of sets. He told us about the history of mathematics and
logic, going all the way back to Aristotle.3

According to Kozowski, Tarski preferred not to use textbooks directly, because they were
too distant from life. Students from his classes and seminars at the university would
attend his lessons at the gimnazjum. Kozowski confirmed Rudniaski’s observation
about the gifted gimnazjum students: the most talented (about one in ten) would meet
at Tarski’s home. To Kozowski, Tarski seemed cheerful and very witty; his language
was beautiful, tended toward irony, and he made fine distinctions in conversation.
Tarski’s favorite mathematical subject at that level was the theory of measurement.4
Tarski and his wife, Maria, both emphasized that athletics should be related to mathe-
matics: ball games and gymnastics, for example, provide many example applications.
Kozowski observed that Tarski was an excellent dancer.

Kozowski noted that Tarski was conscious of his own genius; but he told his students
that he would treat them personally, and did so. Kozowski believed that Tarski singled
him out because he showed the ability to view the school and his peers as a social system.

1
Jadacki 2003a, 145. J arosaw R udniaĔski was imprisoned by the Russians after their 1939 invasion of
Poland, escaped from Siberia with General Wadisaw Anders’s army, fought against the Germans in Italy,
then returned to Warsaw after the war to study with Tadeusz Kotarbiski. He became professor of
philosophy at the Polish Academy of Sciences (Gazeta wyborcza 2008).
2
Jadacki 2003a, 145. L eon T adeusz B aszczyk became a philologist.
3
Kozowski 2003, 2010.
4
Sections 13.6–13.8 and 13.10–13.13 of the present book are translations of the sections of Tarski’s coauth-
ored 1935 gimnazjum text Geometrja that cover the theory of measurement.
14.2 Logic 321

As a young lad, Kozowski had shown great sensitivity and individuality, and had
embraced atheism. He reported,
Graduation was approaching, and I was worrying—even if I did pass the examinations,
I wouldn’t be given the Diploma of Maturation. The prosaic obstacle—I had not been baptized,
and on my certificate my religious persuasion must be declared. I confessed my trouble to
Professor Tarski and his wife ... . I was baptized, in part thanks to Mrs. Tarski, who expressed
the desire to be my godmother.5

Kozowski seemed unaware that Tarski had confronted a similar situation at about the
same age (see chapter 3).

14.2 Logic

Alfred Tarski’s research in logic attracted worldwide attention during the 1930s. He
published most of those results in French or German, and as his fame increased, many
of them were translated into English. Nearly all of his works published after 1939
appeared originally in English; the others were in German or French. The translations
from Polish in chapter 15 are all brief, and concerned with subjects not closely related
to those covered earlier. The present section will describe their general context. Some
minimal explanations of their specific features are included in chapter 15 with the
translations.

Most of the works in chapter 15 originated in connection with colloquium presenta-

tions, sometimes by other scholars. In Warsaw, Tarski frequently reported his research
results at the regular meetings of three different organizations. Listed below are the full
names of the organizations and of the journals that published their proceedings:
• Warsaw Philosophical Institute, Logic Section
Przegld filozoficzny, Ruch filozoficzny6
• Polish Mathematical Society, Warsaw Section
Annales de la Société Polonaise de Mathématique7
• Warsaw Society of Sciences and Letters
Comptes rendus de la Société des Sciences et des Lettres de Varsovie,
classe III: sciences mathématiques et physiques8

5
Kozowski 2003 and 2010; Zalasiewicz 2009. W itold K ozowski was born in Oruro, Bolivia, in 1919. His
father, Roman, was a noted paleontologist and geoscientist, and later a professor at the University of War-
saw. Witold started university studies in Polish literature in 1939; during the war, he was active in under-
ground cultural life, particularly social services and journalism. He earned a master’s degree after the
war, in cultural education. He won many awards during his long career as journalist and author.
6
Warszawski Instytut Filozoficzny, Sekcja Logiczna; Philosophical Review, Philosophical Trends, the latter
published in Lwów.
7
Polskie Towarzystwo Matematyczne, Oddzia Warszawski; Rocznik Polskiego Towarzystwa Matematycznego.
8
Towarzystwo Naukowe Warszawskie; Sprawozdania z posiedze Towarszystwa Naukowego Warszaw-
skiego, wydzia III: nauk matematyczno-fizycznych.
322 14 Teaching and Logic—To the New World

Tarski made his first scientific presentations at the Warsaw Philosophical Institute in
1921. Several of his publications mentioned in section 1.1 and chapter 3 stemmed from
presentations at meetings of these three organizations. The discussions translated in
sections 15.1, 15.4, and 15.6 took place at sessions of the Institute in 1924, 1928, and 1930,
and section 15.7 contains a translation of the abstract of a presentation to the Warsaw
Society of Sciences and Letters in 1932. Tarski continued reporting to Warsaw societies
regularly throughout the 1930s. He served as vice-president of the Institute’s logic section
during the 1930s, while Jan ukasiewicz was its president.9

During the year after the September 1929 Congress of Mathematicians of Slavic
Countries Tarski continued to develop the fundamental concepts of the methodology of
deductive sciences that he had described there. The subject was probably receiving much
attention in his Warsaw research seminar. (For that background, see sections 8.1, 9.4,
and 9.5.) Tarski presented his work in several venues, culminating in six lectures during
the period from Wednesday through Tuesday, 10–16 December 1930, at meetings of the
Lwów sections of the Polish Philosophical and Mathematical Societies:
1930–
1931 Venue Subject
——— — ————————— ——————————————————————————— —
Feb 20 Vienna University Some basic concepts of metamathematics10
Mar 27 Warsaw Soc. of Sci. Some basic concepts of metamathematics
Oct 8 Warsaw Phil. Inst. Concept of a true sentence for formalized deductive systems11
Dec 10 Lwów Phil. Soc. 1. Scope of research, basic concepts, and assumptions
. . 11 ... .... 2. General properties of deductive systems
. . 12 .... .... 3. Bases of deductive systems
. . 13 .... .... 4. Consistency and completeness
. . 15 .... .... Concept of truth in formalized deductive sciences
. . 16 Lwów Math. Soc. Definable sets of real numbers
Mar 21 Warsaw Soc. of Sci. Concept of truth in the languages of deductive disciplines

The series of numbered lectures was called Fundamental Concepts of the Methodology of
Deductive Sciences. They were published in Vienna as Tarski [1930] 1983a; according
to the Philosophical Society’s minutes, a second part of that paper would follow in the
same journal, but that never occurred. Abstracts of the 15–16 December lectures were
published with their respective societies’ minutes. An abstract of the last lecture in this
list was published in 1932 in Vienna.12

9
Jadacki 2003a, 151, 156. For further information on the Warsaw logic milieu in this era, consult Woleski
1995a, section 4.
10
This title is equivalent to that of Tarski’s presentation to the 1929 congress. In Vienna, Tarski also gave
presentations on other subjects. See Menger 1994, chapter 12; Menger 1998, 78, 131; Dawson 1998,
33–36, and Feferman and Feferman 2004, 76–84.
11
Warszawski Instytut Filozoficzny 1931.
12
Polskie Towarzystwo Filozoficzne w Lwowie 1930–1931, 209–210; Tarski [1930–1931] 2014 and 1930b.
In April 1931 Tarski reported to the Warsaw Philosophical Institute about Kurt Gödel’s closely related
but independent [1931] 1967 work on the incompleteness of integer arithmetic: that was probably its
first exposure outside Vienna (Tarski [1931] 2014h, Feferman and Feferman 2004, 84).
14.2 Logic 323

Within a decade, the material in Tarski’s 1930 Lwów lectures became the framework
in which many fields of logic are investigated and presented. The abstract of his 15
December lecture was the first appearance in print of his celebrated theory of truth, a
particularly vital part of that framework: it supports much subsequent work in seman-
tics, including that in philosophy, linguistics, and computer science. It is translated in
section 15.6.

Chapter 15 also contains translations of Tarski’s remarks about six presentations by

other scholars at the Second and Third Polish Philosophical Congresses in 1927 and 1936.
The third congress occurred in the middle of Tarski’s major effort in the 1930s to present
throughout Europe his results in mathematics and logic, and disseminate his point of view
on logic. Together with reports of his Warsaw presentations, this portrays nearly frantic
activity, a major presence in European mathematics and philosophy, all grafted onto his
full-time occupation as a schoolteacher. Tarski’s presentations and publications in logic
and mathematics greatly overshadowed his work as a mathematics educator, and most
of them quickly became available in Western European languages. His renown in logic
explains why his work on geometry and teaching has been little known, even obscure.

The Second Polish Philosophical Congress took place during 23–28 September 1927
in Warsaw. Two hundred scholars attended. There were seventy lectures, apportioned
among seven sections:
• history of philosophy • logic
• metaphysics and theory of knowledge • semantics
• philosophy of natural sciences • aesthetics
• psychology 13
According to the conference report by Warsaw logician Janina Hosiassonówna, the most
notable sections were those on logic and psychology.14 Jan ukasiewicz gave the keynote
address, on the traditional and the deductive approaches to philosophy.15 He noted that
the discipline of philosophy had given other sciences the scientific and deductive methods,
but had never applied those to itself. As a consequence, traditional philosophy was dom-
inated by meaningless answers to meaningless questions. He urged application of the
deductive axiomatic method to all problems in philosophy. Although ukasiewicz men-
tioned no other modern works, his prescription echoed the doctrines of the Peano school
from a generation earlier, and reflected those then being formulated in Vienna by the
logical positivist school.16 Hosiassonówna reported that ukasiewicz’s address stirred up
much controversy, and that he remained the central figure of the congress. He gave two
talks in the logic section, comparing syllogistic to modern axiomatic logic, and outlining
the theory of deduction that he and Tarski were formulating. Tarski presented yet
another talk, continuing that discussion. (Two years later, Tarski would become uka-
siewicz’s official assistant.) These three presentations were not published in the volume

13
Until the 1930s, psychology was often regarded in academia as a branch of philosophy.
14
Hosiassonówna 1928. For information about Hosiassonówna, see the box on page 335.
15
ukasiewicz [1927] 1928.
16
See Marchisotto and Smith 2007, section 2.2, 126–128, and the introduction to Padoa [1900] 1901.
324 14 Teaching and Logic—To the New World

of congress proceedings. The latter two were probably subsumed by Tarski’s later paper
On Some Fundamental Concepts of Metamathematics.17

Sections 15.2 and 15.3 contain translations of Tarski’s very brief remarks on two other
presentations at the 1927 congress: by Henryk Greniewski on the theory of action, and
by Tadeusz Czeowski on causality. Both lay within the scope of ukasiewicz’s proposal.
Specific background for them is included in those sections.

The table below lists some of the congresses and other meetings outside Warsaw in
which Tarski participated, and indicates where some of those are discussed in the
present book.18 As noted, Tarski gave presentations at most of them. The list is surely
not complete, if only because there is no clear borderline between “official” presentations
at organized meetings and informal presentations in homes and offices of colleagues
throughout Europe. Tarski often visited other researchers en route to meetings or on
recreational excursions.

Tarski’s Colloquia Away from Warsaw

Date Location Meeting Discussed in

May 1923 Lwów First Polish Philosophical Congress chapter 3
Sep 1927 Lwów First Polish Mathematical Congress* section 9.4
Sep 1928 Bologna International Congress of Mathematicians* ..... 8.8
Feb 1930 Vienna Mathematisches Kolloquium* ..... 9.5
Dec 1930 Lwów Polish Philosophical Society, Lwów* ..... 14.2
Dec 1930 Lwów Polish Mathematical Society, Lwów* ..... 14.2
Sep 1931 Vilnius Second Polish Mathematical Congress
Jun 1932 Lwów Polish Philosophical Society, Lwów*,19
Aug 1934 Prague Vorkonferenz on Unity of Science* ..... 14.2
Sep 1934 Prague Eighth International Congress of Philosophy ..... 14.2
Jan– Jun 1935 Vienna University of Vienna* . . . . . 9.1, 14.2
Sep 1935 Paris First Congress for the Unity of Science* . . . . . 9.1, 14.2
Jun 1936 Cracow Third Polish Philosophical Congress . . 14.2, 15.8–11
Jul 1937 Paris Third Congress for the Unity of Science ..... 14.2
Jul 1937 Paris Ninth International Congress of Philosophy* ..... 14.2
Sep 1938 Amersfoort Modern Concepts of Reasoning ..... 14.2
Sep 1939 Harvard Fifth Congress for the Unity of Science* ..... 14.3
*Tarski presented a paper.

17
Polski Zjazd Filosoficzny [1927] 1928. Tarski [1930] 1983. See also Jadacki 2003b, 119. Stanisaw
Leniewski also presented at the congress an oral summary of three parts of the paper Leniewski
[1927–1930] 1992, introducing his theories of mereology, ontology, and logistic.
18
The information in the table was gleaned largely from Feferman and Feferman 2004, Dawson 1998, and
Jadacki 2003a, particularly the last. Jadacki claimed in error that Tarski had also presented a paper in
Vienna in 1936.
19
At the 1932 Lwów meeting Tarski spoke about categoricity of a theory and independence of its primitive
notions.
14.2 Logic 325

During summer 1934, Tarski participated in two meetings in Prague. The first, held
on 30 August and 1 September, was a Vorkonferenz devoted to planning the International
Congress on Unity of Science to be held the following summer. There, Tarski presented
his [1935–1936] 1983 paper “Some methodological investigations on the definability of
concepts.”20 The Vorkonferenz was followed immediately by the Eighth International
Congress of Philosophy in Prague, 2–7 September 1934. Many scholars associated with
the Vienna Circle and logical positivism attended both meetings.

The International Congress of Philosophy attracted some four hundred participants

from thirty-two countries, including all major countries in Europe except the Soviet
Union. Continuing a series of conferences begun in 1900, it attracted much attention:
nearly forty press representatives were listed as associates. The Czechoslovak foreign
minister, Edvard Beneš, delivered the opening address.21 Four of the twenty-eight con-
gress sessions were devoted to L’importance de l’analyse logique pour la connaissance.
Tarski participated in the discussions after the first and third of those, but presented no
paper himself. The first “connaissance” session was chaired by Rudolf Carnap, professor
at Prague and a leader in the Vienna Circle. The session featured papers on general
subjects by Tarski’s teacher and collaborator ukasiewicz; by Moritz Schlick, a leading
member of the Vienna Circle;22 and by Jørgen Jørgensen, professor of philosophy at
Copenhagen and the leading Danish exponent of logical positivism. The third “connais-
sance” session was chaired by Louis Rougier, from Besançon and Cairo, a conventionalist
and associate of the Vienna Circle. It included general presentations by Ernest Nagel,
of New York City, and by Eino Kaila, of Helsinki—both members of the logical positivist
movement—and by Kazimierz Ajdukiewicz, from Lwów. The remaining papers, by Janina
Hosiassonówna, Hans Reichenbach, and Zygmunt Zawirski, were devoted to the founda-
tions of probability theory and multivalued logic.

Tarski commented on Jørgensen’s paper and on the papers of Reichenbach and

Zawirski. He published the first of these commentaries—[1934] 2003b—and two almost
identical versions of the second—1935–1936 and [1934] 2003a—in the proceedings of the
two Prague meetings. These remarks, originally in German, were translated into Polish
and published in the collection Jadacki 2003b, 11–14. They are discussed in detail in
section 16.3, which is devoted to posthumously published contributions of Tarski.

Tarski spent January through June 1935 in Vienna, supported by a grant from the
Rockefeller Foundation.23 Afterward, he participated in the conference that had been
planned at the Prague Vorkonferenz the year before: the First Congress on the Unity of
Science in Paris, 15–21 September 1935. There, Tarski presented two papers, which have

20
See Frank 1935–1936 for an overview of the Vorkonferenz, and the individual papers in Erkenntnis 5(1)
for reports of its proceedings.
21
Before entering politics in 1918, Beneš had been a university professor. In 1936, he would become presi-
dent of Czechoslovakia.
22
Two years later, Schlick would become a famous victim of political assassination.
23
Feferman and Feferman 2004, 91. Meanwhile, Tarski’s wife, Maria, a new mother, had moved in with
his parents in Warsaw (see section 9.1).
326 14 Teaching and Logic—To the New World

since become famous: The Establishment of Scientific Semantics and On the Concept of
Following Logically.24 A photograph on page 327 shows him in Paris with the Lwów
philosopher Maria Kokoszyska. Back in Warsaw the following November, Tarski pre-
sented a report about the congress, jointly with Janina Hosiassonówna and her husband,
Adolph Lindenbaum, who had also presented papers there. Their text has not surfaced,
but Kokoszyska’s analogous 1937 report to the Polish Philosophical Society in Lwów
was published.25

On 24–26 September 1936, Tarski participated in the Third Polish Philosophy

Congress in Cracow. It was somewhat larger than the 1927 Second Congress; all listed
participants were Polish. Tarski presented no paper himself, but contributed to the
discussions of the plenary-session papers of Ajdukiewicz, Witold Wilkosz, Zawirski, and
Kokoszyska. These presentations are described in detail in sections 15.8–15.11, and
Tarski’s remarks are translated there. The specialized logic section of that congress
seems rather weaker. But there was a notable special section on Catholic Thinking and
Modern Logic on the last day of the congress.26

The Unity of Science conferences were held every summer from 1935 until World
War II. Tarski evidently missed the second one, in 1936 in Copenhagen. He attended
the 1937 conference in Paris, which immediately preceded the Ninth International
Congress of Philosophy. At the latter, he presented his 1937b expository paper, which
was “a general exposition for the nonspecialist of the nature of completely formalized
deductive systems in logic and mathematics.” 27 At the 1938 Unity of Science conference,
in Cambridge, England, Tarski was elected in absentia to the organizing committee for
the next one, to be held in September 1939 in Cambridge, Massachusetts.28

One of Tarski’s major goals throughout his career in Poland was to obtain a pro-
fessorship in a university there: a position appropriate for conducting his research in
mathematics, logic, and semantics; for guiding related research of other scholars; and
for fostering worldwide acceptance of his framework for logical investigations. The first
steps in that direction came easily. Tarski earned the venia legendi in 1925, which
qualified him for teaching university courses, and in 1929, he was appointed assistant to
ukasiewicz.29 Serving part-time in those capacities, Tarski offered some courses
especially for teachers, some general courses on the methodology of the deductive sciences,
and some courses on specific topics connected with his research, and he began supervising
research of advanced students in ukasiewicz’s seminar. Positions of this sort paid very
little. For his livelihood Tarski relied on his full-time employment as gimnazjum teacher,
on his work training teachers during vacation periods, and, later, on his wife’s employ-
ment as an elementary-school teacher.

24
Tarski [1935] 1936a in the congress section on language, and [1935] 1936b in the section on logic.
25
Warszawski Instytut Filozoficzny 1937; Kokoszyska 1937.
26
For further information about this meeting, see Polish Philosophy Congress 1936.
27
Description from Langford 1938.
28
Tarski [1956] 1983, viii; Jadacki 2003a, 158.
29
See section 9.3.
 Maria Kokoszyska,
Alfred Tarski
Paris, 1935

Maria KokoszyĔska was born in 1905 in Bóbrka, a town near Lwów, in the Austrian Empire (now
Bibrka, in Ukraine). At the University of Lwów, she studied with Kazimierz Ajdukiewicz and
Kazimierz Twardówski. Under the latter’s supervision, she earned the doctorate in philosophy in
1928. Kokoszyska became an assistant at that university in 1930. In 1934, she began participating
in meetings of the Vienna Circle. She met Alfred Tarski while there, and they became close personal
friends. Around this time she married the historian and political activist Roman Lutman.* During
the following years, Kokoszyska contributed significantly to the literature of philosophical logic.
In 1947, after World War II, she earned the venia legendi at the University of Pozna, and the next
year she became a professor at the University of Wrocaw. There she continued research in logic and
served the university, the philosophical community, and the city schools in many leadership roles.
Kokoszyska retired in 1976 and died in 1981.†

*In the literature her surname is often listed as Lutman- or Lutmanowa-Kokoszyska .

† Feferman and Feferman 2004, 88–92; Kokoszyska c1961; Rojszczak 2002, 36; Wójcicki 1981.
328 14 Teaching and Logic—To the New World

Tarski entered the competition for a university professorship in Poland just after
the establishment and expansion of academic institutions in Poland that followed its
attainment of independence. But Poland faced hard economic times and difficult political
questions during his entire career there. Early on, an opportunity appeared for a profes-
sorship in the philosophy faculty at the University of Lwów. Tarski was a strong con-
tender, favored by Kazimierz Twardowski, the head of the logic program there and
teacher of Tarski’s own teachers and senior colleagues. But Tarski had not yet gained
sufficient international recognition, and in fall 1930, after two years of deliberations, the
position was awarded to an eminent and more senior candidate, Leon Chwistek.30

Tarski pursued a professorship throughout the 1930s, with no success. In such a quest,
a scholar might expect to be supported by his own former research supervisor. But the
previously cited sources clearly document Stanisaw Leniewski’s antipathy toward his
illustrious—but only—doctoral student, Alfred Tarski. In that connection and probably
others, antisemitism played a role in frustrating Tarski’s aspirations, in spite of his official
conversion to Roman Catholicism and his change of surname. But the primary obstacle
was simply the scarcity of professorial positions in Tarski’s field in Poland, due to eco-
nomic and political pressures.31

On the other hand, Alfred Tarski and Maria Tarska held reasonably congenial positions
at the eromski gimnazjum, and his university research work and international reputa-
tion were soaring. In fall 1934 Tarski was named adjunkt of the ukasiewicz seminar;
this evidently provided more stature in the university hierarchy. Andrzej Mostowski
entered the university in 1931, and completed the doctorate in 1938 with a dissertation
on set theory supervised by Tarski. He was Tarski’s first doctoral student, although his
official sponsor was Kazimierz Kuratowski. Wanda Szmielew began studying logic in
Warsaw in 1935; by 1938, she had published her first research paper.32 Mostowski’s dis-
sertation and this research of Szmielew were very closely related to Tarski’s early work in
set theory. Moreover, Warsaw between the wars was a very exciting society, not just for
[The narrative continues on page 330.]

30
The competition for the Lwów professorship is reported in Jadacki 2003a, 147–150, referring to Twardow-
ski’s correspondence. See also Feferman and Feferman 2004, 66–68, and Woleski 1995a, section 2. For
more information on Chwistek, see page 329. (The self-portrait there is from Chwistek 1960.)
31
See chapter 3. A professorship opened up in Pozna in 1937, for which Tarski was clearly the top candi-
date; it was left unfilled, probably to avoid offering it to him. For further information on Tarski’s quest
for a professorship, see Feferman and Feferman 2004, 100–108; Jadacki 2003a, 154–158; and Woleski
1995a, the first section 4, section 5, and notes 13–15.
32
Mostowski 1938, reviewed in Tarski 1938c; Szmielew [1938] 1947—the initial date is from Kordos,
Moszyska, and Szczerba 1977, 241; Tarski 1924b, 1924e, and Lindenbaum and Tarski 1926—see section
8.3. For more information about Mostowski and Szmielew see section 9.4.
 Leon Chwistek Fencing
Self-portrait, 1929 by Leon Chwistek, circa 1920

Leon Chwistek was born in 1884 in Cracow, then part of the Austrian Empire. His father was a
doctor who specialized in hydrotherapy; his mother, a pianist and artist. In 1891 the family built
a house and clinic in Zakopane, a resort town in the Tatra mountains to the south. Chwistek was
schooled in Cracow, then entered an art school and the university there in 1902, beginning studies
in philosophy and mathematics. He earned the doctorate in 1906, with a dissertation On Axioms
supervised by Stanisaw Zaremba. Chwistek then taught in a gimnazjum. Leftist in politics, he was
a long-time close friend of the noted artist and writer Stanisaw Witkiewicz, also known as Witkacy.
Chwistek spent the years 1908–1910 in Göttingen and Vienna, and 1913–1914 in Paris, studying
mathematics, philosophy, and painting. There, he fought a famous duel, depicted above, with another
Zakopane artist, Wadysaw Dunin-Borkowski. They had disagreed over a slight to Olga Steinhaus-
ówna, the sister of the Lwów mathematician Hugo Steinhaus; Chwistek married her two years later.
After service in Józef Pisudski’s Polish Legion during World War I, Chwistek returned to teaching.
He established a reputation as artist, writer, and philosopher, publishing major works on art theory
and the theory of types. Chwistek was a founder of the Formist movement in Polish art. In 1928
he was awarded the venia legendi in philosophy at the University of Cracow, and the next year won
a competition with Alfred Tarski for a professorship at the University of Lwów. During the 1930s
Chwistek continued developing his philosophy of logic and science, stressing that mathematical logic
could not by itself form a basis for philosophy of the multiple realities of science. When the Germans
and Russians conquered Poland in 1939, he emigrated to the Soviet Union. He taught for two years
in Tiflis and a year in Moscow. Chwistek died there in 1944.*

*Estreicher 1971; Gromska 1948, 51–53; Skolimowski 1962. Witkacy created the portraits of Tarski and his
wife, Maria, on the covers of Feferman and Feferman 2004.
330 14 Teaching and Logic—To the New World

[continued from page 328]

research, but in all cultural matters. In spite of outbreaks of antisemitism in the univer-
sity, and the threats from Germany and the Soviet Union, which should have been all too
plain, Alfred Tarski and Maria Tarska did not seek to emigrate.33

14.3 To the New World

Returning to Warsaw from a meeting at Amersfoort, in the Netherlands near Amsterdam,

Alfred Tarski visited the philosopher Kurt Grelling in Berlin. There, on 1 October 1938,
on the Wilhelmplatz, they witnessed a violent speech by Adolf Hitler, who had just
returned to the Chancellery from his infamous Munich meeting with the leaders of Italy,
Great Britain, and France. They had agreed to let Germany annex the Sudeten part of
Czechoslovakia. The crowd in the adjacent Wilhelmplatz was ecstatic. Below are some
headlines from the next day’s Nazi party newspaper.34

The capital of the Reich pays homage

to the liberator of the Sudetenland
Berlin thanks Adolf Hitler Under protection
in the name of of the sharp arms
the 80-Million-Volk of the Reich

Wilhelmplatz: a unique hurricane

of jubilant exaltation

33
For more information on Warsaw between the wars, see Wynot 1983, chapter 8; Nowicki 1992; and
chapters 1, 3, and 9 of the present book. For information about the widespread movement during the
1930s to humiliate and restrict the education of Jewish students in Poland, see Natkowska 1999. Anti-
semitism penetrated deeply and peculiarly. For example, the following exercise (Rusiecki 1930b) was
posed in Parametr, the journal that included many of the exercises translated in chapter 12:
In a classroom the Jews constituted 22.58% of the total number of students. Determine how many
students were in the class!
The problem really was to determine positive integers m, n such that 0.22575  m/n  0.22585 and
n could be the size of a school class. Most present-day readers would cringe at the gratuitous reference
to Jews, but those in the intended audience could do so only in silence. The present editors found in that
journal no other exercises posed in loaded ethnic terms.
34
Völkischer Beobachter 1938. The Amersfoort congress took place during 19–25 September; the Munich
agreement and acceptance by Czechoslovakia, on 30 September. According to Domarus 1992, 1213–1214,
Hitler returned to Berlin on 1 October for two days only. See also Feferman and Feferman 2004, 103.
14.3 To the New World 331

After the demonstration, Tarski spent a long evening in discussions with other scholars
who had been there. Tarski’s biographers Anita B. and Solomon Feferman expressed
amazement that Tarski risked stopping in Berlin, but noted that he
was not alone thinking that somehow Poland would be protected by England and France.
This is not to say that he was unaware of ... what was occurring in Europe and in his own
country. Quite the opposite. He was well informed politically ... But his actual involvement
[in politics] was minimal; his own projects and those of his students consumed him, and he
had those two ... jobs.

Moreover, as late as 21 December 1938—after the terrible and widespread 10 November

Kristallnacht pogroms in Germany—Tarski attended a reception celebrating the honorary
doctorate that had been awarded to Jan ukasiewicz by the University of Münster.
Notable Germans were present, including the ambassador. Still, Tarski was evidently
not considering emigration, beyond changing his publication language to English.35

What were Tarski’s principal concerns during that academic year? The Tarskis’ daugh-
ter Ewa Krystyna (Ina) was born in Warsaw on 3 October 1938.36 Whether her mother
still taught school after giving birth to her daughter is not known.37 Alfred continued
teaching at the gimnazjum and university. In fall 1938 he offered his course on teaching
mathematics; in spring 1939, the one on methodology of deductive sciences.38 His super-
vision of research in connection with ukasiewicz’s seminar continued, as mentioned in
the preceding subsection. An outgrowth of that was his work with his former student
Andrzej Mostowski on the metatheory of well-ordering, which would be continued a
decade later (see section 8.2). Moreover, Tarski was continuing to collaborate with his
colleague Adolf Lindenbaum, both on questions of equidecomposibility and on a large
monograph about set theory, which they had begun several years before (see sections 8.5
and 15.12). Finally, Tarski was pursuing his own projects. A large paper in German was
still being composed for publication in Fundamenta Mathematicae; it belongs to the
filament of Tarski’s research on the measure problem,39 discussed in section 8.7. His
[1940] 1967 monograph on the completeness of elementary arithmetic and geometry
was under preparation for publication in English in Paris; he had also begun that work

35
Jadacki 2003a, 158. Tarski’s 1936a and 1936b papers and the original edition of his [1936] 1995 logic
text were has last major works in Polish; 1937b was his last in French. His first publication in English
(1937a) was a technical appendix to Joseph H. Woodger’s 1937 book The Axiomatic Method in Biology.
Tarski had met Woodger at Vienna Circle functions.
36
Feferman and Feferman 2004, 104; Ewa Krystyna’s birthday is stated incorrectly there. The date given
here was confirmed by Jan Tarski (2012).
37
Witold Kozowski (2010) recalled Maria Tarska as a secretary at the eromski Gimnazjum, not as a
teacher. He may have confused Maria with her sister Józefa Zahorska, who was indeed a secretary there.
38
The 1937 offering of the methodology course was described in section 9.3.
39
Tarski 1939–1945 was Tarski’s last paper in German. The editorial note Fundamenta Mathematicae 1945
explained its publication dates. The papers Tarski 1939a and 1939b and Mostowski and Tarski 1939,
on logic, set theory, and Boolean algebras, respectively, were probably completed during the previous
academic year.
Voyage to New York
11–22 August 1939

Menu, Cover Art

14.3 To the New World 333

a decade earlier: see section 9.4. Finally, Tarski was negotiating about publication of a
paper on the significance of the concept of truth, which he had drafted the previous year.40

The Fefermans reported that Tarski had received an official invitation to speak at the
upcoming September 1939 Fifth Congress on the Unity of Science in Cambridge, Massa-
chusetts, but had delayed his response. They attributed that to Stanisaw Leniewski’s
death from thyroid cancer on 13 May 1939, and Tarski’s desire to be present for any
deliberations about Leniewski’s successor, a position for which Tarski would obviously
be the leading candidate.41 According to the biographers, the Harvard professor and
logician Willard Van Orman Quine repeated the invitation and, in addition, urged Tarski
to consider emigration—Quine would set up some appropriate contacts.

Tarski accepted only at the last minute. He obtained an exit visa on 7 August and prob-
ably boarded a train for the port city, Gdynia, on 10 August. He sailed on 11 August on
the ocean liner M/S Pisudski. Tarski packed only a single suitcase, with light clothing
and some formal wear, intending to stay only about a month. On board, he was surprised
to meet the Lwów mathematician Stanisaw Ulam, en route to a position as lecturer at
Harvard. Ulam was accompanied by his young brother Adam, ready to start university
studies at Brown University, in Providence, Rhode Island.42 Some news bulletins were
aired on the ship, and passengers would have discussed the situation in Poland seriously
at meals. A photograph in the Feferman biography shows those three dining, in formal
dress. The ship itself and a menu are depicted on the facing page. The voyagers seemed
to deny the reality of the danger at home. On their pictured 22 August arrival in New
York City, they were met by very apprehensive colleagues, who did not: John von Neu-
mann for the Ulams, and Carl Hempel for Tarski. Decades later, Adam Ulam, by then
a distinguished Harvard political scientist, remarked,
In hindsight we realized we must have been blind ... we really didn’t think war was immediately
imminent. Sure, everyone talked about the possibility of war ... but there was also a lot of talk
and bravado about how France and England would defend Poland. In August ... when we left,
all the Poles were going off on their traditional August vacations, as if everything was normal.

40
According to Paolo Mancosu (2009, 132), Carl Hempel regarded the paper on truth as appropriate for the
journal Erkenntnis, and Tarski had even in 1936 described the monograph with Lindenbaum as nearly
ready for publication. No further trace of these drafts has surfaced. In the Tarski archive at the Bancroft
Library in Berkeley is a complete draft of an unpublished set-theory monograph by Richard Montague,
Dana S. Scott, and Tarski, circa 1970.
41
Feferman and Feferman 2004, 106. Tarski was himself a member of the organizing committee for the
congress.
42
S tanisaw and A dam U lam were born in Lwów in 1909 and 1922, respectively, to a prosperous Jewish
family. Stanisaw earned the doctorate there in 1933 under supervision of Kazimierz Kuratowski. His
early research interests were closely related to Tarski’s work in set theory. Stanisaw emigrated to the
United States in 1935, spent a year at Princeton to collaborate with John von Neumann, and then three
years at Harvard as a junior fellow. Stanisaw became a leading figure in applied mathematics in the
United States. Adam emigrated at the last possible moment, in August 1939, to start undergraduate stud-
ies at Brown University. He earned a doctorate in political science from Harvard in 1947 and spent a long
career as professor there, an authority on Russia and the Soviet Union. Stanisaw died in 1984 and Adam,
in 2000. See Ulam 1976 and Harvard University Gazette 2000.
334 14 Teaching and Logic—To the New World

On the day after Tarski arrived in New York City, the Germans and Soviets concluded
a secret agreement to destroy the Polish state.43

After sightseeing and socializing in New York City, Tarski traveled on 24 August by
train to Boston. He spent a week in nearby Cambridge to prepare for the congress. On
1 September, the Germans invaded Poland. Stanisaw Ulam reported that he and others
spent hours listening to the radio news. The congress began on schedule on 3 September.
The German philosopher Rudolph Carnap, who had emigrated to a professorship at the
University of Chicago, commented,
In spite of the exciting world events, we found it possible to devote ourselves to the theoretical
44
discussions of the Congress.

Tarski presented his paper New Investigations on the Completeness of Deductive

Theories on 9 September, the last day of the congress, at a concurrent meeting of the
Association for Symbolic Logic. Scheduled just after his talk were two by Kurt Grelling
and Janina Hosiasson-Lindenbaumowa; another, by the Lwów logician Leon Chwistek,
had been set on another day. Those speakers were ominously absent—read the boxes
that follow.45

Kurt Grelling was born in Berlin in 1886. His father, Richard, was a lawyer, active in politics; his
mother stemmed from a wealthy Jewish merchant family. Kurt finished his schooling in Gotha in
1904, then studied mathematics and philosophy at Göttingen. There, he discovered the antinomy
about heterological concepts that is now named for him. It was published jointly with the philosopher
Leonard Nelson. Under supervision of David Hilbert and Ernst Zermelo, Grelling earned the doctor-
ate in 1910 for research on set theory and the axiomatics of arithmetic. He translated several philo-
sophical treatises and wrote one on probability theory, but never habilitated, which was required for
a professorship. Instead, he taught in secondary schools and immersed himself in the labor movement
and politics. His argument against his father’s indictment of the German instigation of World War I
was publicized widely. Kurt Grelling became a leading exponent of the empiricist philosophy of the
Berlin School, in association with Hans Reichenbach. During the 1930s he pioneered work on the
logical foundations of gestalt psychology. He was fired from his teaching job in 1933, but by then was
financially independent. But with no means of support outside Germany, and trouble obtaining travel
documents, he and his wife left only in 1938, for Belgium. After the Germans conquered that country
in 1940, they deported the Grellings to Vichy France, and thence to Auschwitz for extermination in
1942. For further information, see Luchins and Luchins 2000.

43
Feferman and Feferman 2004, 124–127, which includes Adam Ulam’s remark. The voyage is also
recounted in Ulam 1976, 114–115, with the ship incorrectly identified. By 1939 Neumann and Hempel
had emigrated from Hungary and Germany and had found positions at Princeton and City College of New
York, respectively.
44
Feferman and Feferman 2004, 128, which includes Carnap’s remark; Ulam 1976, 119.
45
Tarski 1939a is an abstract of Tarski’s talk. For information about Chwistek see the box on page 335.
 Janina Hosiassonówna Adolf Lindenbaum

Janina Hosiassonówna was born in 1899 in Warsaw. She studied mathematics and philosophy there,
earned a doctorate in 1926 with a dissertation on inductive logic supervised by Tadeusz Kotarbiski,
and became a secondary-school teacher. Hosiassonówna continued research on inductive logic in
Warsaw, and spent some time around 1929–1930 in Cambridge, England, studying the work of several
scholars there on probability and induction. She played a significant role in establishing such studies
within the framework of formal logic. Hosiassonówna also translated into Polish several books by
Bertrand Russell. After her marriage to Alfred Tarski’s research collaborator Adolf Lindenbaum, she
used the surname Hosiasson-Lindenbaumowa. She was scheduled to present her research at the Unity
of Science Congress at Harvard in September 1939, but was unable to sail in time. After that month’s
invasions, she escaped the Nazi oppression temporarily by moving to Vilnius, which had been annexed
by the Soviet Union. The Germans invaded that region in 1941; they arrested and murdered Janina
Hosiasson-Lindenbaumowa in 1942.*

Adolf Lindenbaum was born in Warsaw in 1904, to a Jewish family that became involved in the
motion-picture business; he was independently wealthy. After graduating from gimnazjum in 1922,
Lindenbaum entered the University of Warsaw to study mathematics. He became active in student
organizations and left-wing politics, and earned the doctorate in 1927 under supervision of Wacaw
Sierpiski. Over the next decade, Lindenbaum contributed many results to the research seminar of
Jan ukasiewicz and Alfred Tarski, and published about twenty papers about general topology, set
theory, and mathematical logic, including especially significant ones authored jointly with Tarski and
Tarski’s student Andrzej Mostowski. Some now-fundamental concepts in logic are named for Linden-
baum. He married another Warsaw logician, Janina Hosiassonówna. During 1940–1941, under the
Soviet occupation, he taught at the pedagogical instutute in Biaystok. Adolf Lindenbaum was mur-
dered after the 1941 German invasion, near Vilnius.**

* Galavotti 2008; Gromska 1948, 57–59; Jedynak 2001.

** Skowska and Wglowska 2003; Woleski 1995a, 375–376.
336 14 Teaching and Logic—To the New World

14.4 Epilogue

After the Congress on the Unity of Science, Alfred Tarski was stranded in the United
States. After some difficulties with immigration status, he was able to obtain barely
adequate support while researching and teaching at various East Coast institutions.
Eventually, in September 1942, he obtained a position at the University of California
at Berkeley. The fighting in Europe ended on 30 April 1945; soon after that, Alfred
became a citizen of the United States. After eight months’ delay making arrangements,
Tarski was reunited in Berkeley with his wife, Maria; son, Janusz; and daughter,
Krystyna. After the German invasion in 1939, they had stayed in Warsaw with Maria’s
sister Józefa Zahorska. After the Warsaw Uprising, August–October 1944, they had
found shelter away from that city. Maria had succeeded in concealing her children’s
Jewish ancestry.

Alfred Tarski’s parents, brother, and his brother’s family were murdered by the
Germans. His colleagues Leon Chwistek, Janina Hosiasson-Lindenbaumowa, and Kurt
Grelling had not been able to leave Poland and Germany for the congress in time.
Chwistek escaped to Moscow, but sickened and died there. The others, and Tarski’s
colleague Adolf Lindenbaum, were murdered.

Tarski’s students Andrzej Mostowski and Wanda Szmielew survived the war. She
came to Berkeley in 1947, earned the doctorate there in 1950 as Tarski’s fifth doctoral
student, then returned to Warsaw. Both of those young scholars became leaders in
logical research in postwar Poland.

In Berkeley, Alfred Tarski founded a research program in logic and methodology of

science. Soon it became the world’s leading program in that subject, and Tarski
attained recognition as the world’s preeminent logician during the next three decades.46

46
Readers may obtain much further information about Tarski’s work in logic and his career in the United
States from the marvelous 2004 biography by Anita B. and Solomon Feferman. It will also be useful to
consult the numerous other works listed in chapters 17 and 18 of the present book.
 Part Four
Supplement
This book’s concluding Part Four consists of four chapters with different purposes. The
twelve sections of chapter 15 contain the remaining translations required to make all of
Alfred Tarski’s works accessible in at least one of the languages English, French, and
German. All of these items are short. In content they are generally unrelated to each
other and to the larger papers translated in previous parts of the book. Brief discussions
are provided to set them in context. Chapter 16 updates Steven R. Givant’s 1986 Tarski
bibliography by identifying all known items authored by Tarski and first published after
that date. Chapters 17 and 18 list and briefly annotate major published studies of
Tarski’s life and work. Their items were chosen because of unique relationships between
the authors and Tarski or because of the breadth of their coverage. The lists in chapters
17 and 18 are a biproduct of the research that led to the background discussions in earlier
parts of this book. The editors did not search exhaustively for material of the same sort
not mentioned in items discovered in that research.

Part Four employs boxed text somewhat differently from the earlier parts of this book.
As before, boxes with single borders contain informational essays that can be read inde-
pendently. Readers are not expected to visit them in sequence. But doubly outlined
boxes, as in chapter 12, enclose translations of Tarski’s writing (and one abstract by Jan
ukasiewicz) and separate them from background discussions by the present editors.
They are intended to be read in sequence. In several sections, doubly outlined translation
boxes are continued from one page to the next by omitting the bottom border of the one
and the top border of the next.
 15
Assorted
Contributions

This chapter is devoted to translations from Polish and discussions of twelve minor
publications of Alfred Tarski, in chronological order. They fulfill this book’s goal of
making all of Tarski’s publications accessible in English, French, or German. They are
included in this supplementary part of the book because they are not closely related to
any of the works featured in earlier parts. The associated discussions of their context and
content are sufficient to make them intelligible.

The items in sections 15.1–15.3 and 15.8–15.11 are Tarski’s remarks included in
published reports of other scholars’ presentations at meetings in Warsaw in 1925 and
1927, and in Cracow in 1936. They are about various aspects of philosophical logic.

Section 15.4 features a translation and discussion of a 1928 Warsaw presentation by

Jan ukasiewicz: On Definitions in the Theory of Deduction. ukasiewicz announced
some results of Tarski, who contributed to the subsequent discussion. The whole paper
is translated here because Tarski’s work is not easily separated from it.

Tarski’s 1929 paper on Polish pensions is translated in section 15.5. Published in a

journal on economics, it revealed a mistake in a pension law. Its nature is that of a report
on an involved applied exercise in elementary algebra.

The 1930–1931 paper On the Concept of Truth in Reference to Formalized Deductive

Sciences is probably the most interesting item in this chapter. Translated in section 15.6,
it is Tarski’s report of a presentation in Lwów, the first appearance in print of his cele-
brated theory of truth. His full account would not reach publication for several years.

Tarski’s 1932 Abstract On Geometric Properties of Banach’s Measure is translated in

section 15.7. The title of this record of his presentation to a Warsaw meeting was listed
in French in earlier surveys, but its text was in Polish.

The chapter concludes with translations of two letters that Tarski wrote to Wacaw
Sierpiski from Berkeley in 1946 and 1947. They are included here because they show
how a thread of Tarski’s geometric research in Poland extended into his career in
California.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 339
DOI 10.1007/978-1-4939-1474-6_15, © Springer Science+Business Media New York 2014
340 15 Assorted Contributions

15.1 Discussion of ukasiewicz 1925, on Understanding Deduction

On 8 December 1924, Jan ukasiewicz presented a paper to a meeting of the Logical

Section of the Warsaw Philosophical Institute. This event belonged to the same series
in which Alfred Tarski had presented two papers in 1921. The proceedings were edited
by Bolesaw Gawecki and published in volume 28 of the journal Przegld filozoficzny. The
published report of ukasiewicz’s presentation, entitled On a Certain Way of Understand-
ing the Theory of Deduction, seems to be an edited version of a listener’s notes.1 It
includes comments by Kazimierz Kuratowski, Stanisaw Leniewski, and Tarski, but no
response from the speaker.

ukasiewicz presented a graphical means of displaying formulas of Boolean logic with

colored tiles representing two connectives and propositional variables, and of manipulat-
ing them according to the substitution and modus-ponens inference rules to provide
proofs. He implemented this as a kind of single-player game. According to ukasiewicz,
the fact that the tiles convey no intuitive content shows that the process of deduction
involves only the form, not the content, of the formulas.

Kuratowski remarked that it would be useful to include additional types of tiles to

incorporate quantificational logic.

Leniewski echoed and extended Kuratowski’s comment. But he saw no advantage

of this game over ordinary presentations. He noted that it shows clearly that grouping-
symbols such as parentheses are unnecessary, and that ukasiewicz’s description of the
game included a recursive definition of formula in this system of logic.

Tarski commented,

The game suggests that for the future science of the theory
of deduction, undeveloped at this point, it might be possible
to find a geometric interpretation that would make it possible
to carry out a proof of the consistency of this science.

That remark is listed as Tarski [1925] 2014 in the present book’s bibliography.

Leniewski’s discussion reveals aspects of the state of development of logic in Poland

in 1924. The ideas of parenthesis-free notation and recursive definition of linguistic
constructs were new enough to warrant his attention. ukasiewicz first published his

1
ukasiewicz 1925. The Polish title is O pewnym sposobie pojmowania teorji dedukcji.
15.2 Discussion of Greniewski [1927] 1928, on Action 341

now-famous parenthesis-free notation four years later.2 On the other hand, in a 1936
letter to the Austrian philosopher Otto Neurath,3 Tarski recalled that while the view of
linguistic constructs as physical pictures had indeed been debated and regarded positively
by the Vienna Circle before 1930, it had been common in Warsaw since at least 1918, as
illustrated by ukasiewicz’s 1925 report.

ukasiewicz’s game was a precursor of a more elaborate and rather famous one
published decades later by Layman E. Allen: Wff ’N Proof. Allen used and explicitly
acknowledged ukasiewicz’s notation but was unaware of the earlier game.4

In the comment displayed on the preceding page, Tarski noted that geometric repre-
sentation of formulas might lead to a method for proving consistency. That might be
related to his later work on geometric dissections, in which he drew metamathematical
conclusions from other geometric considerations: see section 8.5.

15.2 Discussion of Greniewski [1927] 1928, on Action

At the Second Polish Philosophical Congress in September 1927 in Warsaw, Henryk

Greniewski presented the paper An Attempt to Formulate Precisely Some Notions of the
Theory of Action. The congress is described in section 14.2. The present section sum-
marizes Greniewski’s talk, displays Alfred Tarski’s [1927] 2014a comment about it
(translated, in the box on the next page), and describes the speaker’s response.5

Greniewski’s goal was to make precise the notions plan and method, which were
needed for a theory of action. He claimed that his work was inspired by Tadeusz Kotar-
biski or Jan ukasiewicz. According to Kotarbiski,
...if the plan is a certain description of a certain selection and sequence of actions, the method
is nothing more nor less than that very planned selection and sequence of actions integrated
by a common goal.6

Greniewski first noted that his system would consider not only individuals, but also
generalized individuals denoted by pseudonyms such as Zeus that stand for certain classes
of properties. Just as an event such as Aristotle philosophized should be a pair consisting
of an individual and a class of individuals (for example, philosophers), a generalized event

2
ukasiewicz remarked ([1931] 1970, 180): “I came upon the idea of a parenthesis-free notation in 1924.
I used that notation for the first time in [ukasiewicz 1929, 610 ff ]. See also [ukasiewicz [1929] 1963,
§§6 ff ] ... .”
3
Tarski [1930–1936] 1992, 26. These posthumously published letters are described in section 16.1.
4
Allen 1965, 2010. The title Wff ’N Proof is a double pun: Wff is a common English abbreviation for the
logical term well-formed formula; the Whiffenpoofs are a noted Yale University male a capella choral
ensemble. The game originated at Yale.
5
The Polish title of Greniewski [1927] 1928 was Próba precyzacji pewnych poje teorji czynu. For informa-
tion about its author, see the box on page 343. An English translation of part of Tarski’s remark was
included in Jan Woleski’s paper Tarski as a Philosopher (1993, 325–326).
6
Kotarbiski [1955] 1965, 56.
342 15 Assorted Contributions

should consist of a generalized individual and a class: Zeus thundered. Greniewski

considered time-dependent generalized events, and sets of them that might ramify to
represent possible future events. These would constitute plans, which are thus objects
of higher-order types. Plans could be compared and judged equivalent by a suitable
relation, and a method would be an equivalence class of plans.

The subsequent discussion included the following remark, which is listed as Tarski
[1927] 2014a in the present book’s bibliography.

Alfred Tarski drew attention to the origin of the ideas

contained in the paper, and pointed out that in interpret-
ing the most varied terms in pure logic, a certain moderation
is desirable, so that they retain only their formal properties.
An interpretation is desired in those cases in which the terms
satisfy two conditions: (1) these terms find an application
in any field of knowledge; (2) in experience (in the world
of concrete objects) no intuitive equivalents of these terms
are found.

In response, Greniewski granted that too great a reliance on higher-order types would
be harmful, but, as noted in a previous remark by Kazimierz Ajdukiewicz, abstraction
with equivalence classes simplifies the theory, as it does for arithmetic. Moreover,
Greniewski noted, there is no assurance that mental sensations should belong to the
lowest type of logical objects: that is a very dark issue.

Tarski’s remark certainly related to the subject at hand, particularly in view of

ukasiewicz [1927] 1928, the keynote address of the congress. But it does not seem
directly related to the published version of Greniewski’s presentation. Moreover, Gre-
niewski’s response does not seem to address Tarski’s remark directly. Probably, other
conversations were going on that were not recorded.

A recent survey7 of the theory of action reveals a rich, broadly applicable subject that
certainly contains traces of the ideas expressed in Kotarbiski’s and Greniewski’s early
work but does not mention that as a forerunner. An investigation of this possibility
would prove interesting. Both of those scholars continued working in this area and its
applications for decades.

7
Segerberg 2009.
15.3 Discussion of Czeowski [1927] 1928, on Causality 343

Henryk Greniewski was born in Warsaw, then a part of the Russian Empire, in 1903. He was
schooled at the same Zamoyski Gimnazjum on Smolna Street that Tadeusz Kotarbiski had attended
a generation earlier; somehow, they became lifelong friends. Greniewski studied briefly at the
University of Warsaw, then completed doctoral studies in 1926 at Cracow with a dissertation on
logical foundations of mechanics supervised (informally) by Leon Chwistek. A second concentration
was the mathematics of finance and insurance, which gained him secure employment in the financial
sector and at the Social Insurance Institution. During the 1930s, under the pseudonym Kozowski,
he was the author of many articles in the journal Gospodarka narodowa ( National Economy).
During World War II, Greniewski participated in underground socialist organizations. In 1948
he became director of the Group for Mathematical Apparatus of the Mathematical Institute of the
University of Warsaw, supervising construction of electronic computers. During 1951–1968 he served
the university as professor of education, then director of the Department of Econometrics. Greniew-
ski maintained activity in logic, serving as an editor of Studia Logica from 1953 on, and lecturing
on the foundations of computer science. He was known as an authority on the use of information
technology in business planning. Greniewski died in 1972.*

* Trzsicki 2010; Chodorowski et al. 2005, 49.

15.3 Discussion of CzeĪowski [1927] 1928, on Causality

In September 1927, Tadeusz Czeowski presented the paper On a Causal Relation at a

session on philosophy of natural sciences at the Second Polish Philosophical Congress in
Warsaw. The congress is described in section 14.2. Czeowski’s paper is summarized
here, followed by a translation of Alfred Tarski’s [1927] 2014b comment about it, in the
box on the next page. Two other comments and the speaker’s response are paraphrased
after that.8

Czeowski discussed various properties of phenomena and relations between them.

It is unclear what he meant by phenomena,9 but they could be temporal, spatial, or
repeating, and contiguous in time or space. He defined direct formal causal relationship
to stand between two classes A and B of phenomena just in case for each x  A there
exists y  B such that x and y are contiguous in time and space (or just in time if x
and y are not both spatial). Further, two phenomena a and b stand in direct material
causal relationship just in case there exist classes A and B of phenomena such that
a  A, b  B, and A and B stand in direct formal causal relationship. Indirect formal
and material causal relationship can be defined by generalizing the direct notions. Cze-
owski did not specify the generalization in detail.

8
The Polish title of Czeowski [1927] 1928 was O stosunku przyczynowym. For information about its author,
see the box on page 345. Tarski [1927] 2014b was listed as item 28ca in the bibliography Givant 1986, but
was omitted from the Tarski 1986a Collected Papers volumes.
9
Czeowski’s word was zjawiska.
344 15 Assorted Contributions

Czeowski claimed that seeking a formal causal relationship between classes of

phenomena is analogous to empirical inductive reasoning; no such relationship is logically
necessary. However, mere contiguity does not suffice to establish a material causal
relationship between two phenomena. One must verify that they constitute an example
of a formal causal relationship; that step may involve reasoning from logical necessity.
Czeowski suggested that there is an analogy between the notions of material and formal
causal relations and those of material and logical consequence.

The ensuing discussion included the following remark, which is listed as Tarski [1927]
2014b in the present book’s bibliography.

Alfred Tarski pointed out the following consequences of the

definitions given by the speaker:
(1) Every phenomenon is a direct material cause of every
other phenomenon contiguous with it in space and time.
(2) Every phenomenon is an indirect material cause of
every other phenomenon.
(3) Every class of phenomena is an indirect formal cause of
every other class of phenomena.

In the discussion, Kazimierz Ajdukiewicz noted that Tarski’s remark would apply even
if in the definition of material causality, classes A and B were required to have more
than one element.10 Henryk Mehlberg11 commented that the analogy between causality
and implication is incomplete because the fact that a falsehood materially implies any
proposition has no analogue for causality. Moreover, he doubted that there could be any
direct causal relationship according to this definition if time varied continuously.

Czeowski responded that the objection of Tarski and Ajdukiewicz applied only if
phenomenon was understood as the state of the entire universe at a given time, in which
case it said that the state at any moment is the cause of the state at the next. To Mehl-
berg he responded (roughly) that nevertheless there are similarities between the theory
of causality and the theory of formal deduction and logical truth, and that the notion of
contiguity is certainly related to that of continuity.

10
For more information on Ajdukiewicz, see a box in chapter 3.
11
Born in 1904 near Tarnopol, then in the Austrian Empire but now in Ukraine, H enryk M ehlberg studied
with Kazimierz Twardowski and Kazimierz Ajdukiewicz in Lwów in the 1920s and remained there through
World War II. In 1948 he emigrated to Toronto and soon moved to the University of Chicago, where he
continued a distinguished career as a philosopher of science. He specialized in theories of time and
causality. Mehlberg retired in 1970 and died in 1979. (Koterski 2001.)
15.3 Discussion of Czeowski [1927] 1928, on Causality 345

The lack of precise language and references makes it difficult to interpret Czeowski’s
paper, and must have contributed to the tone of Tarski’s devastating remark. All three
discussants’ objections were valid. The paper is an early, but unsuccessful, step in lifting
discussions in this area of traditional philosophy to the level of rigor that Jan ukasiewicz
demanded in his keynote address at the congress.12

Tadeusz
Czeowski
around 1925

Tadeusz CzeĪowski was born in Vienna in 1889; his father was a government official. Schooled in
Lwów, Tadeusz attended the university there and earned a teaching credential in 1912. He taught
in schools while he continued studies in logic with Kazimierz Twardowski. Czeowski completed the
doctorate in 1914 with a dissertation on the theory of classes. After military service, he worked with
Twardowski as a university administrator. Then Czeowski become director of the Department of
Science and Secondary Eduation in the Ministry of Religious Denominations and Public Education.
He was heavily involved in organizing the new Polish universities in Warsaw, Vilnius, and Pozna.
Czeowski habilitated in 1920, and in 1923 became professor at Vilnius. He was named full professor
there in 1936 and held administrative posts during 1933–1939. Czeowski contributed to many areas
of philosophical logic and became a major figure in the Polish logic community. During World War
II he taught philosophy clandestinely. Czeowski was honored by the State of Israel with the
designation “Righteous among the Nations” for undergoing great risk to save Vilnius Jews from the
Holocaust. After the war Czeowski transferred to Toru, again helped establish a new university,
and founded the Toru Philosophical Society. For many years he edited the journal Ruch filozoficzny.
Czeowski retired in 1960 and died in 1981.*

* Gumaski 1984

12
ukasiewicz [1927] 1928. See section 14.2.
346 15 Assorted Contributions

15.4 ukasiewicz 1928–1929, on Definitions

On 18 February 1928, Jan ukasiewicz presented a paper to a meeting of the Logical

Section of the Warsaw Philosophical Institute. This meeting belonged to the same series
as the one reported by ukasiewicz 1925, which is summarized in section 15.1. The
proceedings of the meeting were published in the journal Ruch filozoficzny, volume 11,
177–178. The published report of ukasiewicz’s presentation, entitled On Definitions in
the Theory of Deduction,13 seems to be a loosely edited version of a listener’s notes. Some
of the research that it mentions is described only vaguely. It includes comments by
Stanisaw Leniewski and Alfred Tarski, but no response from the speaker.14

In his presentation, ukasiewicz compared two methods for handling definitions in

propositional logic. One way, favored by Leniewski during the discussion, is to build a
notion of definitional equivalence into the system under study. Tarski provided some
technical steps toward that end. ukasiewicz favored regarding definitions as external
to the system, keeping it simpler and more focused. But he presented the advantages and
disadvantages of both methods. This paper provides an interesting preview of issues that
would reappear a generation later, in the design of programming languages.

ukasiewicz included discussion of some results of Tarski and of Mordchaj Wajsberg.

For that reason, this paper is listed in the appendix of the Tarski bibliography, Givant
1986, but Tarski’s discussion is not listed there. No part of the work is reproduced in
Tarski’s 1986a Collected Papers. Therefore, it is translated in its entirety in the following
boxes, as ukasiewicz [1928–1929] 2014. The translation adheres to the same standards
as earlier ones in this book. Items in [square] brackets and all footnotes, which fall
outside the boxes, are the work of the present editors.

On Definitions in the Theory of Deduction

Prof. Dr. Jan ukasiewicz

The speaker recognizes two types of definitions. [Definitions of type (1) are]
found in Principia Mathematica: for example, p H q .=: ~ p Z q Df.15 The authors
of Principia say that the “=” and “Df” form a whole, and what is on the left
side [of the “=”] means the same as what is on the right. [Method (2)] defines
primitive symbols in the system. It would seem that the only term suitable for
use in a definition [in the original system] is equivalence. However, we do not

13
The Polish title of this paper, ukasiewicz [1928–1929] 2014, is O definicyach w teoryi dedukcyi. In that
context, the term “theory of deduction” referred to propositional logic.
14
Additional comments by Henryk Greniewski and Bronisaw Knaster were mentioned but without detail.
15
Whitehead and Russell 1910–1913, vol. 1, 11.
15.4 ukasiewicz 1928–1929, on Definitions 347

know any theory of deduction with equivalence as a primitive term. In a system

whose primitive term is implication, Alfred Tarski first used definitions in the
form of two implications. That is the second method of definition.

The speaker gives an example that applies both of these methods of definition
in the proof of a theorem—specifically, definitions 16
(1) CNpq || A pq (2) CApqCNpq and CCNpqA pq.
[In this discussion he is] not treating the first definition as a hypothesis of the
system, but as a rule that allows us to write everywhere, in place of what is on
the left side of the definition, that which is on its right side, and vice versa. A
proof turns out to be shorter with the first method of definition.

The mere fact that the first method of definition is more consistent with what
we intuitively mean by a definition, and that the second method introduces defini-
tions as hypotheses, similar to axioms, leads to [the speaker’s] inclination to adopt
the first method. It would seem, however, that definitions of the first type are
logically stronger than [those of ] the second. In fact, having the identity law Cpp
in the system as well as a directive definition17 such as CNpq || A pq, you can
write the sentence CCNpqCNpq then replace the first occurrence of CNpq by
A pq and [do this a second time with] the second one. Two hypotheses are
obtained in this way, which constitute a suitable implicational definition.

Due to the following facts the speaker is definitely inclined to adopt the first
type of definition.

(1) Mordchaj Wajsberg found a sentence, [using just the connectives] implica-
tion and negation, sufficient to build a theory of deduction, but only by adopting
a definition of the second kind. He also showed that by using such a definition
(namely, definitional hypotheses CMqNq and CNqMq) one of the speaker’s
axiom systems can be simplified. The speaker presents a way to simplify [this],
which relies on [the possibility of ] deducing from two axioms a third, independent
of these,18 expressed in terms of the primitive [connectives] only, using the above
definition. According to the speaker, the purpose of these definitions lies in
replacing the more complex expressions by simpler ones that have more intuitive
value, rather than in introducing new theses into the system.19

16
These are definitions of the disjunction connective A in terms of implication C and negation N: that
is, of “p or q” as “not-p implies q.”
17
ukasiewicz’s term was definicy dyrektywaln.
18
ukasiewicz’s phrase was niezalenego od nich.
19
The present editors have been unable to locate any other report of this work of Wajsberg.
348 15 Assorted Contributions

Mr. Tarski stated that in writing a definition in the form of two implications
he [would] adjoin Cpp to the system. What is adjoined, moreover, [should also
be] adjoined with definitions of the first type, since, as has already been noted,
from such definitions suitable implicational definitions can be obtained using
Cpp. The speaker adds that directive definitions cannot imply any statements
about the primitive terms independent of the axioms; indeed, having obtained
a statement about primitive terms by such a definition, the same definition can
be used to replace the introduced term by the primitive [one] everywhere in the
proof. Therefore, considering the previously mentioned theorem of Mr. Tarski,
a new hypothesis, the identity law, [should be] given with an implicational
definition.20

(2) The speaker cites more facts confirming his conviction that we must use
definitions of the first type. He assumes David Hilbert’s axioms without [the law
of ] commutation.21 Adjoining to these axioms a definition of the expresion A pq
in the form of implicational hypotheses CA pqCNpq and CCNpqA pq does not
allow a statement to be obtained that differs from one of the axioms only in that
A pq stands in place of CNpq. (Of course, it can be obtained immediately by
introducing the definition CNpq || A pq.) On the other hand, there are in this
case as many as 2 12 different interpretation tables22 for A pq. Therefore, with
the [implicational] definition above, the expression is ambiguous. But directive
definition of the expression A pq allows here just one interpretation.

In the discussion, Stanisaw Leniewski, Henryk Greniewski, Mr.Tarski, and

Bronisaw Knaster spoke.

Prof. Leniewski favors adopting a definition of the second type. He mentions

four possible variations of the first type of definition, pointing to the difficulties
associated with each of them. [The one] most similar to the type first presented
by the speaker, according to Prof. Leniewski, is to accept a directive in advance,
according to which one is free to write as directed the replacements of some

20
It is unclear whether the first two sentences of this paragraph are the speaker’s or a comment from Tarski
in the audience. The original wording of the paragraph’s last clause was now tez , dan przez definicy
implikacyjn , jest prawo tosamoci.
21
ukasiewicz was probably referring to the axioms for the sentential calculus in Hilbert and Ackermann
[1928] 1950, page 29. ukasiewicz referred to that book in the introduction to his [1929] 1963 logic text,
and on page 47 identified the law of commutation as CCpCqrCqCpr : that is, ( p  (q  r))  ( q  ( p  r)).
22
ukasiewicz may have been alluding here to an investigation of the independence of the law of commuta-
tion, unknown to the present editors, which may have revealed very many ways to form multivalued
interpretation tables for the connectives C, N, and A, under which the remaining axioms of Hilbert’s
system and the hypotheses for the implicational definition of A might all be valid.
15.5 Tarski 1929, on Polish Pensions 349

expressions for others. There is a difficulty in formulating such a directive. Prof.

Leniewski further notes that it is possible to write a definition of the second type
using [connectives] other than implication as primitive: for example, Sheffer’s
stroke. In that case the definition is only a single sentence. In Prof. Leniewski’s
ontology, definitions have led to theses independent of the axioms. This is not
a defect—on the contrary: the definitions being introduced should be as creative
as possible. Finally, for a nonextensionalist, definitions of the first type are not
acceptable.23

Mr. Tarski notes that with the help of implication the [second type of ] defini-
tion can be written down in the form of only one hypothesis, specifically the
hypothesis CCDC Err, where D is the hypothetical implication in the one
direction and E, in the other.24 He continues, giving a way to state all definitions
in advance. Specifically, there is a countable number of possible definiens–
definiendum pairs; arrange them in sequences x1, x 2 , ... and y1, y2 ,... , and
accept that an appropriate x can be replaced by [the corresponding] y: that is,
that appropriate conditional [expressions] can be written down.

15.5 Tarski 1929, on Polish Pensions

This section is devoted to Alfred Tarski’s 1929 article Marginal Note on “Decree of the
President of the Republic Concerning Insurance of Nonmanual Workers as of 24 November
1927,” which appeared in the journal Ekonomista, kwartalnik powi cony nauce i
potrzebom  ycia, volume 29, pages 115–119.25

At that time Ekonomista was edited by Edward Lipiski, professor at the Warsaw
School of Economics, and published by the Association of Polish Economists and Statisti-
cians.26 Still published, it has traditionally addressed a broad audience. The issue that
contained Tarski’s paper also included articles and book reviews on economic history,
economics of municipal development, econometrics, aspects of the Polish economy and
politics, and reports and announcements of the association.

23
In the original, this sentence was Dla nieekstensyonalisty wreszcie definicye pierwszego rodzaju s nie do
przyjcia.
24
This seems to require that r be a contradictory sentence. Then, Cpr, or p  r in infix notation, is equiv-
alent to Np for any sentence p, and thus CCDCErr is equivalent to N( D  NE), which is in turn equiv-
alent to D & E. Now let D and E be the two implications that constitute the second type of definition.
25
The Polish title of the article was Na marginesie “Rozporzdzenia Prezydenta Rzeczypospolitej o ubezpie-
czeniu pracowników umysowych z dnia 24 listopada 1927 r.” The journal title means The Economist:
A Quarterly Devoted to Science and the Needs of Life.
26
Szkoa Gówna Handlowa w Warszawie; Towarzystwo Ekonomistów i Statystyków Polskich.
350 15 Assorted Contributions

Tarski’s article pointed out a probably unintended unfair consequence of the wording
of a government regulation that determined the amount of disability or old-age pension
for a Polish white-collar worker. The amount depended on a base figure established
elsewhere, on the number m of months of the worker’s contribution, and on the number
of his or her children. Tarski noted that fairness, almost certainly intended by the
legislature, would make the amount of a pension a nondecreasing function of m. He
demonstrated that in fact, this was not the case, and expressed hope that the anomaly
would be corrected.

Tarski was interested in other social issues too, possibly through his teacher colleagues
at the gimnazjum. In 1928 and 1931, he was elected docents’ representative to the faculty
council of the University of Warsaw. In 1937 he joined a socialist political club.27

This article about pensions is similar in approach to that of some curriculum materials
developed in the United States during the 1990s, which aimed at relating secondary and
lower-level university mathematics instruction closely to “real-world” applications.28
Tarski made his point through specific computational examples that used a precise
algebraic description of the algorithm for the amount of a pension. Tarski hinted that
elementary mathematics could yield a deeper and more general understanding of the
anomaly in the regulation, which would be necessary to correct it. The present editors
suggest this as an interesting project for a talented student at this level.

The present editors have not investigated the background of this regulation, deter-
mined the scope of its application, nor searched for any response to Tarski’s article. He
did not return to this subject in any later publication.

A full English translation of Tarski’s article, Tarski [1929] 2014b, appears in the boxes
that follow. This is its first translation. It is as close to the original as possible given the
inexact correspondence between 1927 Polish government terminology and that of con-
temporary English. All text in [square] brackets or outside the boxes was provided by the
present editors as explanation. Punctuation has been adapted to current English practice.
White space was inserted liberally to enhance clarity. For some text, indentation
has replaced emphasis by italics. Tarski’s numbers for his footnotes (inside the boxes)
have been replaced by distinctive symbols to avoid confusion with the present editors’
footnotes (outside).

27
Jadacki 2003a, 147, 152, 157. See sections 3.1–3.2.
28
For example, Hughes-Hallett et al. 2003.
15.5 Tarski 1929, on Polish Pensions 351

DR. ALFRED TARSKI

Docent, University of Warsaw

Marginal Note on
“Decree of the President of the Republic
Concerning Insurance of Nonmanual Workers
as of 24 November 1927”*

I would like to direct attention here to certain consequences that follow in an

undeniable way, in my opinion, from the words included in the title Decree, and
which remain, I presume, glaringly inconsistent with the intention of the
legislators.

Let us take into consideration an insured who satisfies the following

conditions:
(a) [he] has the right to a disablility or old-age pension (by virtue of
articles 22 and 24 of the Decree);
(b) the “basis 29 of pension benefits” (article 33 paragraph 3) yields for
him a specific sum, say P zlotys;
(c) his condition is of the sort that requires “constant care and
assistance by other persons” (compare article 40 paragraph 1);
and finally,
(d) the insured has a specified number, for example d, of children
younger than eighteen years of age, or even older than eighteen
years of age, as long as they satisfy the conditions provided by
article 28 paragraphs 3 and 4.

Under the above assumptions, according to Decree articles 38–40, the entire
pension together with all allowances that the insured draws from the Institution
for Nonmanual Workers’ Insurance 30 is dependent solely on the number of
“months of contribution” to the insurance. I will denote the number of “months
of contribution” by the symbol m and the pension together with allowances

* Poland President 1927.

29
The Polish term was podstawa wymiaru.
30
Zakad Ubezpiecze Pracowników Umysowych
352 15 Assorted Contributions

(calculated in zlotys), as a function of the variable m, by the symbol R(m).

Then, according to article 16 paragraph 5 and article 24 paragraph 4 the function
R(m) is defined when m assumes integer values in the interval [60, 480]:
60  m  480.†

I shall now find the value R(m) more precisely. In accordance with articles
38–40, this quantity is composed of three items:
(1) the proper disability or old-age pension;
(2) an allowance due to the fact that the insured “needs constant care
and assistance by other persons”;
(3) an allowance for children.
Since these three amounts, expressed in zlotys, depend more or less explicitly on
the number m of “months of contribution,” I shall denote them by the symbols
R1 (m), O(m), and D(m) respectively. Thus, a formula is obtained:
R(m) = R1 (m) + O(m) + D(m). (I)

Next (article 38 paragraph 1 and article 39) the proper pension R1 (m) consists
of a constant “principal amount” Z and the “amount of increase in pension”
W(m), dependent on the number m:
R1 (m) = Z + W(m). (II)

According to article 38 paragraph 2 we have

Z = 0.4 P. (III)
On the other hand, in accordance with paragraph 3 of that article —
The increase in pension begins after completion of one hundred twenty months of
contribution; the amount of increase in pension is 1/ 6 % of the basis for each addi-
tional month, and after four hundred eighty months of contribution reaches 60% of
the basis.—

† To simplify considerations, I do not take into account here the exceptional cases provided for in
article 16 paragraph 5 and article 33 paragraph 4.
15.5 Tarski 1929, on Polish Pensions 353

­ 0, when m < 120

°
W(m) = ® 1
when 120  m  480.‡
°̄ 600 P m  120 ,
(IV)

Next, according to article 40 paragraph 1—

A person receiving a disability or old-age pension, needing constant care and assis-
tance of other persons, receives an allowance equal to the difference between the
pension drawn and its basis (article 33 paragraphs 3 and 4).—

O(m) = P – R1 (m). (V)

Finally, in accordance with paragraph 2 [of article 40]—

A person receiving a disability or old-age pension receives for each child (articles 28
and 29) younger than eighteen years of age, or older than eighteen years of age under
the conditions provided in article 28 paragraphs 3 and 4, one tenth of the principal
amount (article 38 paragraph 2) if the pension together with the allowance for chil-
dren, but without the allowance provided for in the first paragraph, should not exceed
the base pension (article 33 paragraphs 3 and 4).—

we conclude that

D(m) = ^ 0.1 Zd , if R1(m) + 0.1 Z d  P

P  R1 ( m), if R1(m) + 0.1 Z d > P .
(VI)

The preceding formulas (I)–(VI) enable determination of each value R(m)

if P, d, and m are given.31

It seems certain that according to the intent of the legislators, the func-
tion R(m) should be a nondecreasing function. That is, expressed in ordinary
language:

‡ In order to satisfy article 38 paragraph 4, “Under no circumstance can the disability pension
be less than fifty zlotys a month,” some alterations should be introduced either in formula (II)
or in formulas (III) and (IV). For this purpose, for example, it is possible to retain formulas
(III) and (IV) but change (II) in the following way:

R1 (m) = ^ 50,
Z  W ( m),
if Z  W ( m) † 50
if Z  W ( m) ! 50.
(IIU)

Such an alteration would not make any impact on the further considerations.

31
Tarski’s variables are mnemonic:
Variable Polish English Variable Polish English
d, D . . . . dzieci . . . . . . . children R ..... renta . . . . . . . . . . . . . pension
m . . . . . miesi cy . . . . . months W ..... wzrost . . . . . . . . . . . . increase
O . . . . . . opieka . . . . . . . care Z ..... kwota zasadnicza . . . principal
P . . . . . . podstawa . . . . basis
Using the second alternative of (VI) in (I) would yield R(m) = O(m) + P.
354 15 Assorted Contributions

( A) It should not happen that of two persons, having the right to a dis-
ability or old-age pension, and finding themselves in the very same
situation with respect to all conditions that affect the size of pension
and allowances, the only difference being that the first has been
insured over a shorter time than the second,§ the first should collect
despite this a larger monthly sum from the Insurance Institution
than the second.

The validity of the above principle is so obvious that arguing its elementary
requirements of fairness or [discussing] examples of other laws and regulations
of a similar nature seems unnecessary.

In view of the preceding remarks [it seems that] the author or authors of the
Decree failed, no doubt, to notice the fact that the function R(m) determined by
formulas (I)–(VI) in accordance with the wording of the Decree, being generally
a nonincreasing function, actually decreases over some interval of variation of
m; the size of the interval increases with the increase in the number d, and
disappears only when d = 0.

One can convince oneself of the truth of the preceding sentence with the help
of completely elementary mathematical reasoning, which I will not include here.
I will illustrate on the other hand the perceptible situation with a certain espe-
cially clear example. Specifically, I suppose that the number d of children of the
insured is 15; for definiteness I assume, moreover, that the “basis of pension
benefits” [is] P = 300 zlotys. The course of values of the function R(m) —that
is, the total monthly pension together with allowances (expressed in zlotys),
depending on the number m of “months of contribution”—is then shown in the
following table calculated on the basis of formulas (I)–(VI):
m 60 120 180 240 300 360 420 480
R(m) 480 480 450 420 390 360 330 300

The following [results], among others, [can be inferred] from the preceding
table:

§ More precisely, it should be expressed, “the first person has been approved for insurance [zali-
czono do ubezpieczenia] over a shorter time than the second.”

Here, d assumes values in the interval [0, 15]: 0  d  15; starting from d = 15, the size of
the interval on which the function R(m) decreases does not change.
15.5 Tarski 1929, on Polish Pensions 355

(B) If two persons satisfy the following conditions—

(a) both are entitled to a disability or old-age pension;
(b) the first was insured over five years, and the second, over
forty years—
moreover, all other circumstances having influence on the amount of
the pension and allowances are in both cases completely identical,
namely—
(c) #
the “basis of pension benefits” for each person is 300 zlotys;
(d) both need “constant care and assistance of other persons”;
[and] finally
(e) both have the same number of children for whom they are
entitled to an allowance, namely 15 each—
then the first person receives in total from the Insurance Institution
480 zotys per month, but the second, only 300 zlotys.

I shall present yet another example, less crude than the previous one, and
likely to find significantly more applications in practical life. I shall suppose,
specifically, that the number d of children of the insured is 5, while the “basis
of pension benefits” P is 300 zlotys, as previously. The course of values of the
function R(m), depending on m, is then expressed in the following table:
m 60 120 180 240 300 360 420 480
R(m) 360 360 360 360 360 360 330 300

It follows that
(C) if two persons satisfy the following conditions—
(a) both are entitled to a disability or old-age pension;
(b) the first was insured over thirty years, and the second,
thirty-five years—
moreover, whereas—
(c) the “basis of pension benefits” is equal for both persons and
is 300 zlotys;
(d) both need “constant care and assistance of other persons”;
[and] finally
(e) each of them draws an allowance for 5 children—
then the first person receives in total from the Insurance Institution
360 zlotys per month, but the second, only 330 zlotys.

# Condition (c) is then satisfied, for example, when each of the persons considered earned 300
zlotys per month over the whole time and was approved for insurance (compare article 14 and
article 33 paragraph 3).
356 15 Assorted Contributions

The contradiction between principle (A) and the concrete consequences (B) and
(C) of the Decree is so obvious that it does not require comment.

One ought to expect that the shortcoming of the “Decree Concerning Insurance
of Nonmanual Workers” exposed here, having such great significance in general
for nonmanual workers in Poland, will be removed in the near future through
amendments to the law.

15.6 Tarski 1930 –1931, on the Concept of Truth

At a December 1930 series of meetings of the Lwów section of the Polish Philosophical
Society, Alfred Tarski presented the results of his investigations of fundamental concepts
of the methodology of deductive sciences. He had begun reporting on this topic in 1928
and had probably been perfecting the work in his Warsaw research seminar. Polished
versions of some of this material were being published already in 1930, in German.32 But
one of his lectures was aired only as an abstract in Polish, On the Concept of Truth in
Reference to Formalized Deductive Sciences.33 It appeared in the journal Ruch filozoficzny,
volume 12 (1930–1931), pages 210–211. Tarski’s celebrated theory of truth would not
be fully published for several years. The abstract was its first appearance in print. At
its conclusion it was labeled auto-referat: a report submitted by the speaker. A full
translation, Tarski [1930–1931] 2014, appears in the boxes that follow. This is the first
translation of the abstract.

Tarski presented another version of this talk to the Warsaw Society of Sciences and
Letters in March 1931. His brief 1932 paper, published in Vienna in German, reported
that talk; that paper is actually a description, not a repetition or a continuation, of the
work translated here. These abstracts were greatly revised, amplified, and published in
Polish as Tarski 1933. That monograph was edited and amplified further, translated, and
published in German as Tarski [1933] 1935. The German version was translated into
English and published in 1956; a revised English edition appeared as Tarski [1933] 1983.

The present translation, Tarski [1930–1931] 2014, is particularly faithful to the original
abstract. Text in [square] brackets has been inserted sparingly for clarification. The only
change in symbolic notation was italicizing variables. Tarski’s styles of quotation marks
have been maintained. Some other punctuation has been adapted to current English
practice. White space is sometimes used in that way, and often inserted to display lists,
a table, and the title. Even in those cases, Tarski’s words have been translated strictly.
Some particularly sensitive translation points have been explained in footnotes. Foot-
notes 38, 40, and 44 contain historical information, as well. All footnotes fall outside the
boxes, and are the work of the present editors.

32
Tarski [1930] 1983a,b; for further information about these presentations, see section 14.2 and Zygmunt
2010.
33
The Polish title of the article was O poj ciu prawdy w odniesieniu do sformalizowanych nauk deduk-
cyjnych. The journal title means Philosophical Trends.
15.6 Tarski 1930–1931, on the Concept of Truth 357

At the 307th plenary scientific meeting on 15 December 1930

Docent Dr. Alfred Tarski of the University of Warsaw gave a lecture

On the Concept of Truth in Reference to

Formalized Deductive Sciences

According to the intuitive understanding of the word „truth“ we say that the
sentence34 „snow is falling“ is true if and only if snow is falling. Generally, we
could say that „ p“ is true if and only if p, or still more generally, A is true if
and only if for some p, A = „ p“ and p. Such a definition of truth is incorrect,
however, since the occurrence of the expression »„ p“« in it must be considered
a function35 with a variable argument; and as Stanisaw Leniewski has demon-
strated, this leads to antinomies—specifically to the antinomy of the liar and [to
that] of heterological words.36 Moreover, in both of these formulations, a non-
extensional function occurs (that is, both »„ p“« as well as »A = „ p“«, since
both of these functions change their values even if for „p“ we substitute only
equivalent sentences).

The purpose of the lecture will not be to define the truth of any sentence
whatever in everyday language; this we are not able to do. We shall define the
truth of sentences occurring in the languages of certain deductive systems. We
could try saying it thus: a sentence of a certain system is true if and only if it
is a theorem37 in this system. Such a definition, however, does not conform to
intuitions relating to the word „truth“. In fact, it does not support the law of
excluded middle, which says that of two contradictory sentences one must be true:
it is in fact not necessary that one of two contradictory sentences be a theorem
of the system.

In order to give for the truth of the sentences of some deductive system a
definition correct and consistent with our intuitions, we distinguish first of all
the language of which we speak and the language in which we speak.38 We define

34
The word zdanie, in its various forms, is always translated here as sentence; this word occurs in the
translation only in that way.
35
In this paper Tarski used the word function ( funkcya) only to denote certain linguistic expressions.
36
Tarski’s phrase was do antynomii kamcy i wyrazów heterosemantycznych.
37
Tarski’s word was teza.
38
Tarski’s construction for this last clause was parallel and unemphasized: odrónimy przedewszystkiem
jzyk o którym mówimy i jzyk w którym mówimy. In the 1932 German abstract the language in which
we speak is called the Umgangssprache. In the 1933 and [1933] 1935 Polish and German monographs
it is the metajzyk or Metasprache, respectively, meaning metalanguage. That term does not occur in the
present paper.
358 15 Assorted Contributions

the truth of a sentence belonging to the language of which we speak. In contrast,

the definition that we give will now belong to the language in which we speak.
We give this definition only for a certain specific, very elementary, deductive sys-
tem. But, completely analogously, we could construct a definition of the truth
of a sentence for many other deductive systems (namely, for all those systems in
which the order of variables—in the sense of the so-called theory of types—does
not exceed some natural number given in advance).

The language of which we speak (which could be called the language of the
elementary algebra of classes) is presented as follows. It contains
three variables „a“, „b“, „c“;
a primitive term „G“ (we read this as „is part of “);
39
logical terms „“ (we read this as „it is not true that“),
„Z“ (we read this as „or“),
„3“ (we read this as „for all“);
and, moreover, parentheses.
The axioms of this system express certain properties of the relation „being
a proper part of “. Here are some examples:
(1) 3a(a G a),
(2) 3a 3b ((a G b) Z  (b G a)),
(3) 3a3b(a G b).40
Among the expressions of the language we distinguish the so-called sentential
functions. We call the inclusions „a G b“, „b G c“, and so on, functions of the
first order, meaning elementary; 41 there are nine kinds of them, there being
three variables. The functions of the second order appearing in this language
are functions formed from functions of the first order with the help of one of
the logical terms: for example, „(a G b) Z (b G a)“ is such a function.42 Functions
of the third order are functions formed from functions of the second order in
the same way as functions of the second order from functions of the first order.
Analogously, functions whose order is any natural [number] can be defined.
Sentential functions not containing any so-called free variables are called
sentences.

39
Tarski’s phrase was terminy logiczne.
40
From (2), Tarski omitted the outermost parentheses, which are necessary. At the corresponding point
in [1933] 1983, page 179, Tarski alluded to an axiom system of Edward V. Huntington (1904, 297) for the
inclusion relation I. The present example axioms can probably be related to a counterpart of that system
for the proper-inclusion relation G.
41
Tarski’s phrase was funkcyami pierwszego rzdu czyli elementarnemi. He used the same noun rzd for
order of a variable in the previous paragraph.
42
In error, Tarski wrote „(a G b) Z (b H a)“ here.
15.6 Tarski 1930–1931, on the Concept of Truth 359

The language in which we speak contains words and logical expressions to the
extent in which they occur, for example, in Principia Mathematica,43 thus in
particular, expressions equivalent to the terms of the language of which we speak:
for example, „is a part of “ (that is, „G“), „or“, and so on. Moreover, it contains
terms „z1“, „z2“, „z3“ which denote the variables
„a“, „b“, „c“, respectively;
a function „I(x, y)“ whose arguments are
of names 44 „z1“, „z2“, „z3“, while the
values [are] names of the
respective inclusions—that
is, names of expressions
„( a G b)“, „(b G a)“, and
so on;
the expressions „ x– “ denoting the negation of x,

„ x + y“ which denotes the logical

sum of x and y;
and the term „ B x y“ which denotes the general-
ization of the expression y
with respect to the variable
x —that is, an expression
of type „3 a p“.

By In, p we shall understand I(zn , z p ), where n, p = 1, 2, 3.

In constructing the definition of true sentence we shall use the notion of

satisfaction of a sentential function by a certain sequence of objects.45 Specifically,
in an inductive manner, we shall define when some sequence of objects satisfies
a sentential function of the nth order from the language of which we speak.
Sequence C satisfies a sentential function of first order—that is, an inclusion
In, p —if and only if
(1) the sequence C is defined for numbers 1, 2, 3;
(2) n, p = 1, 2, 3;
(3) the nth term of the sequence is a part of the pth term of the
sequence—that is, Cn G Cp .

43
Whitehead and Russell 1910–1913.
44
Tarski’s term was funkcya nazwowa.
45
Tarski’s word for objects was przedmioty.
360 15 Assorted Contributions

In order to say when some sequence satisfies a certain function of higher order,
it suffices—in view of the way in which the functions were formed—to say when
a negation, sum, or generalization of a function of a certain order is satisfied,
assuming that we already know when a sequence satisfies a function of that
order. Here, a sequence C satisfies x– if and only if
(1) (as above);
(2) the sequence C does not satisfy x.
The sequence C satisfies x + y if and only if
(1) (as above);
(2) the sequence C satisfies x or satisfies y.
Finally, the sequence C satisfies B zn x (that is, the generalization of x with
respect to the nth variable) if and only if
(1) (as above);
(2) n, p = 1, 2, 3;
(3) not only C, but also every sequence obtained from C through
a change just in the nth term, satisfies x.

We defined when a sequence of objects satisfies some sentential function in the

language of which we speak. Since sentences in this language are also certain
sentential functions, we have explained at the same time when a sequence of
objects satisfies a sentence in the language of which we speak.

With this in mind, we define,

A sentence in the language of which we speak is true
if and only if every sequence satisfies this sentence.

Such a definition is consistent with intuitions connected with the word „truth“.
As an argument for this, one can use the fact (among others) that by such a defi-
nition of the truth of a sentence the law of excluded middle is satisfied. This
holds because either every sequence satisfies sentence x, or none satisfies it. For
if even one sequence does not satisfy x, then no sequence satisfies this sentence,
and in that case every sequence of objects satisfies the sentence x– . Thus, if x
is not true, then x– is true. Similarly, the principle of contradiction is satisfied.
What is most important is that all the theorems constructed in the following way
15.6 Tarski 1930–1931, on the Concept of Truth 361

can be justified. We take under consideration the schema „ x is a true sentence

if and only p“ and any sentence y in the language of which we speak, then in
this schema we substitute for „ x“ the name of the sentence y (expressed in the
language in which we speak), and for „ p“ a sentence that is a translation of the
sentence y into the language in which we speak. Here is an example of such
theorems: B z1 B z2 I1, 2 is a true sentence if and only if for any a and b whatever,
a is part of b. Therefore, on the basis of the [ just] established definition of truth
we are able to explain, in a way appropriate from the intuitive point of view, the
meaning of the phrase „ x is a true sentence“ as applied to every sentence in the
language of which we speak.

A further consequence of the above definition is a proposition according to

which all theorems46 of the system are true sentences; from this it is easy to
conclude that the system of which we speak is consistent. In this way the con-
struction of a definition of truth also yields a method for a proof of the consistency
of the deductive system under consideration.

Tarski’s presentation was discussed by Kazimierz Ajdukiewicz, Roman Ingarden, Maria

Kokoszyska, Kazimierz Kuratowski, Tadeusz Witwicki, and Tarski.47 No further details
were recorded.

Why did Tarski undertake to define true sentence? Solomon Feferman has concluded
that Tarski and some other mathematicians had worked comfortably with informal ver-
sions of the notions explicated in this abstract at least since 1924. On the other hand,
another student and colleague of Tarski, Robert L. Vaught, claimed that48
Tarski appears to have been unhappy about various results obtained during the [1927–1929]
seminar because he felt that he did not have a precise way of stating them... no one had made
an analysis of truth [for example], not even of exactly what is involved in treating it [infor-
mally] ... . His major contribution was to show that the notion [of a true sentence] can simply
be defined inside of ordinary mathematics. ... Knowing that all of the semantical notions are
just ordinary mathematical notions, logicians and mathematicians doubtless felt able to behave
much more freely with them. ... In historical fact, this was only a part of Tarski’s motivation,
for he was also very much concerned with the positions and attitudes on the notion of truth
taken by various philosophers ... .

46
Tarski’s nouns twierdzenie and tezy have been translated here as proposition and theorems.
47
Born in Cracow in 1893, R oman I ngarden earned the doctorate at Freiburg in 1918 from Edmund Husserl.
He became professor of philosophy at Lwów and, after World War II, at Cracow. Ingarden was noted for
his work in ontology and aesthetics (Thomasson 2012). He died in 1970. T adeusz W itwicki was born in
Lwów in 1902; he earned the doctorate in psychology in 1927 from the university there. Witwicki
contributed to the literature of psychology and philosophy and taught at various institutions in Lwów,
Warsaw, and Toru until his death in 1970 (Hutnikiewicz 1965). For information about Ajdukiewicz,
Kokoszyska, and Kuratowski, see boxes in chapter 3 and sections 14.2 and 4.4, respectively.
48
S. Feferman 2008, 79; Vaught [1971] 1974, 160–161. For more information about the 1927–1929 seminar,
see section 9.4. The awkwardness of philosophical discussions involving the concept of truth, even by
those aware of some contemporary discussions in logic, is illustrated in section 15.3.
362 15 Assorted Contributions

Historical research reveals that considerations related to defining truth were much
in the air in the philosophical circles of the 1920s and most were controversial. Tarski’s
definition made specific use of several of these. First, he defined truth in terms of
satisfaction. Truth and satisfaction of what? For Tarski, truth-bearers are sentences,
not thoughts or judgments, as some had held. His sentences could be expressed in any
one of a rather broad class of languages. Tarski was not interested in colloquial lan-
guages, but in formal languages such as those employed in mathematics. Moreover, these
languages were regarded as interpreted: their nonlogical symbols must all be assigned
specific meanings not in question in his discussion. Since such languages were incapable
of describing states of affairs dependent on factors beyond those meanings, Tarski’s
conception of truth was nearly absolute: it was relative only to the choice of language.
This limited degree of relativity was sufficient for the scope of his study. Tarski empha-
sized that his discussion of truth would take place in a metalanguage distinct, in principle,
from the object language under analysis. The metalanguage could be the informal
language of ordinary mathematics or philosophy, or it could be formalized if that should
be necessary for application of a truth definition to a specific problem. Tarski showed
that the notion of satisfaction of a sentence in the formal object language could be defined
recursively according to its complexity. That technique fit neatly into the apparatus then
used in many elementary mathematical studies. Tarski noted in conclusion that his
definition of truth is consistent with the intuitive Aristotelian concept49 expressed in his
opening paragraph; but unlike that concept, it can be used with many languages of
interest without danger of contradiction.50

In the [1933] 1983 full monograph on truth, Tarski analyzed the concept of a truth
definition further. There he formulated his now famous condition T,51 which he later
explained in simpler terms as follows:
We stipulate that the use of the term “true” in its reference to sentences in [a given object
language] then and only then conforms with the classical conception of truth if it enables us
to ascertain every equivalence of the form
(3) “p” is true if and only if p
in which “p” is replaced on both sides by an arbitrary ... sentence [in that language]. If this
condition is satisfied, we shall say simply that the use of the term “true” is adequate.52

In that later discussion, Tarski first presented this condition for the English language,
and used the antinomy of the liar to show that no adequate truth definition is possible.

49
Tarski noted that this concept “can be found in Aristotle’s Metaphysics: ‘To say of what is that it is not,
or of 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 not, is true.’
Here ... the word ‘false’ means ... ‘not true’ ... . [This] is usually referred to as the classical, or semantic
conception of truth.” (Tarski 1969, 63; see also the reference to 1969 in [1933] 1983, 152, footnote †; and
Aristotle 1971, book *, chapter 7, page 23.)
50
The article Murawski and Woleski 2008 places Tarski’s work in context, with many more details. It
hardly mentions Stanisaw Leniewski’s thoughts in this area, but Arianna Betti has considered Tarski’s
relationship to him in detail (2004, 2008).
51
Tarski [1933] 1983, 187–188.
52
Tarski 1969, 64–65. In [1933] 1983, Tarski used the phrase materially adequate in this context. Two
years before, he had described a materially adequate definition as one “grasping the current meaning of
the notion as it is known intuitively” ([1931] 1983, 128–129).
15.7 Tarski 1932, on Banach’s Measure 363

Then he rephrased the condition for object languages in general, and discussed how, for
certain formalized object languages, an adequate truth definition could be achieved in the
metalanguage. In [1933] 1983 he showed that the feature of the object- and meta-
languages that is critical in this regard is their order: the maximum order of the logical
types of the variables that they can handle. He concluded,
A. For every formalized language a formally correct and materially adequate definition of true
sentence can be constructed in the metalanguage with the help only of general logical
expressions, of expressions of the language itself, and of terms from the morphology of
language—but under the condition that the metalanguage possess a higher order than the
language which is the object of investigation.
B. If the order of the metalanguage is at most equal to that of the language itself, such a defini-
tion cannot be constructed.53

Writing in retrospect, Tarski clarified the intent of his definition:

What will be offered can be treated in principle as a suggestion for a definite way of using the
term “true”, but the offering will be accompanied by the belief that it is in agreement with the
prevailing usage of this term in everyday language.54

This analysis of truth is a vital part of Tarski’s infrastructure for mathematical logic.
It has also stimulated much subsequent semantic research in philosophy, linguistics, and
computer science.

15.7 Tarski 1932, on Banach’s Measure

On 23 February 1932, Alfred Tarski contributed the paper On Geometric Properties of

Banach’s Measure to a meeting of the Warsaw Society of Sciences and Letters. The paper
was sponsored officially by Wacaw Sierpiski; whether Tarski presented it in person is
unclear. The published abstract,55 in volume 25 of the Comptes rendus des séances de la
Société des Sciences et des Lettres de Varsovie, Classe III, seems to be an edited version
of a listener’s notes. A full translation, its first, appears in the box on the next page.

In 1923 Stefan Banach had shown that there exist (finitely) additive extensions of
Lebesgue measure, invariant under congruence, to the algebras of all bounded sets of real
numbers and of all bounded sets of points in the Euclidean plane. That result played a
major role in the 1924 works of Tarski and Banach about equidecomposability of point

53
Tarski [1933] 1983, 273. For a nontechnical overview of that work, see Feferman and Feferman 2004,
Interlude III, 109–123. For further information on Tarski’s philosophical works, consult the sources listed
in chapter 18 of the present book.
54
Tarski 1969, 63.
55
The Polish title was O wasnociach geometrycznych miary Banacha. The journal also gave an equivalent
French title.
364 15 Assorted Contributions

sets.56 In his 1932 presentation, Tarski outlined a plan for recasting Banach’s work into
a geometric framework that would unfold more naturally from the familiar theory of
Peano– Jordan content. Tarski’s abstract includes no details at all. The following trans-
lation, listed as Tarski [1932] 2014 in this book’s bibliography, adheres to the same
standards as earlier ones in this book. The word in [square] brackets was inserted by the
present editors.

A l f r e d Ta r s k i

On Geometric Properties of Banach’s Measure

Presented by W. Sierpiski on 23 February 1932

Sur les propriétés géométriques de la mesure

de Banach
Mémoire présenté par M. W. Sierpiski dans la séance de 23 février 1932

Abstract

In his 1923 work Sur le problème de la mesure Banach solved

positively the broader problem of measure for linear and
planar sets. Banach’s construction method is not based on
concepts having clear geometric content and does not
resemble in any way the methods used in other constructions
in this subject (for example, in the Peano– Jordan or Lebesgue
theory of measure). In this paper the author outlines a plan
for a certain reconstruction of Banach’s results using con-
cepts with clear geometric content. As a consequence of this
reconstruction Banach’s theory of measure becomes a natural
generalization and development of the Peano– Jordan
measure theory. To this end the author defines the class of
absolutely measurable sets (that is, those that have the same
measure for each [compatible] definition of measure) and
he establishes some properties of this class.

Tarski returned to this project again in his 1938b paper on absolute measure; unfortu-
nately, even there the exposition was meager.

56
Tarski [1924] 2014b and Banach and Tarski [1924] 2014, translated in chapters 5 and 6. See chapter 4
for background information on these papers and Banach 1923.
15.8 Discussion of Ajdukiewicz 1936, on Idealism 365

15.8 Discussion of Ajdukiewicz 1936, on Idealism

On 25 September 1936 Kazimierz Ajdukiewicz presented the paper The Problem of Ideal-
ism in a Semantic Formulation at a plenary session of the Third Polish Philosophical
Congress in Cracow. The congress is described in section 14.2. Ajdukiewicz’s paper is
described here, followed by a translation of Alfred Tarski’s [1936] 2014a comment about
it in the box on the next page. Ajdukiewicz’s response is summarized after that.57

Ajdukiewicz’s 1936 paper is vague, probably in part because he was discussing a

subject very well known to his audience. In 1937 he published a longer, more informative,
paper on the same subject. Ajdukiewicz was attempting to explicate, in the framework
of contemporary logic, the thesis of transcendental idealism as formulated by the German
philosopher Heinrich Rickert. Ajdukiewicz regarded Rickert’s presentation of that
philosophy as the easiest to understand but nevertheless complained about its lack of
clarity. He cited no specific publication of Rickert.58 The Polish philosopher Leszek
Nowak has written,
According to Rickert, there exists the objective spirit conceived of as a set of trancenden-
tal norms and the reality is to be a correlate of it. The latter claim is interpreted by
Ajdukiewicz as saying that these norms show the set of all possible truths. These rules
are transcendental in that sense that we, people, can have an access to the set of all
possible truths only on the condition that we follow these rules. Therefore, they decide
about the truthfulness of our judgements. Ajdukiewicz paraphrases Rickert’s “transcen-
dental norms” as the rules of logical consequence relative to the scientific language.
Under this paraphrase, the thesis of transcendental idealism reads as the claim that in
the language of natural sciences only those sentences are true that are shown in the
rules of consequence characteristic of that language.59

Ajdukiewicz discerned certain features of Rickert’s philosophy that correspond to the

axioms (both logical and empirical) of some formal deductive system for natural science,
and other features that correspond to its inference rules. Thus, Ajdukiewicz claimed that
the judgments true in Rickert’s sense correspond exactly to the sentences provable in that
deductive system. Ajdukiewicz claimed that such a deductive system must include arith-
metic, and by Kurt Gödel’s famous incompleteness theorem,60 it must therefore be
incomplete: its language must include a sentence S such that neither S nor the nega-
tion of S is provable by the inference rules. Therefore, Ajdukiewicz claimed, Rickert’s
philosophy of transcendental idealism is inadequate for natural science because it fails
to satisfy the law of excluded middle: it cannot decide the truth or falsity of all sentences
of natural science.

57
The Polish title of Ajdukiewicz 1936 was Zagadnienie idealizmu w sformuowaniu semantycznym. It was
published in volume 39 of the journal Przegld filozoficzny. In the table of contents for the congress in
the journal, the paper’s title began instead with the word Problemat. For information about its author,
see a box in chapter 3.
58
Ajdukiewicz [1937] 1978, particularly 146 ff. Born in the Prussian city Danzig in 1863, H einrich R ickert
had been professor of philosophy at Heidelberg since 1916; he was regarded as a leader in the Baden
school of neo-Kantian philosophy. He died in 1936.
59
Nowak 1995, 213.
60
Gödel [1931] 1967.
366 15 Assorted Contributions

The ensuing discussion included the following remark, which is listed as Tarski [1936]
2014a in the present book’s bibliography. The translation adheres to the same standards
as earlier ones in this book. The word in [square] brackets was inserted by the present
editors.

Alfred Tarski emphasized:

(1) On the characteristics of deductive systems:
Methodologists of the deductive sciences would prefer
a characterization of deductive systems using a general
concept of consequence over [one] that used a concept
of direct consequence.
(2) On the relationship between the concepts of meaning
and direct consequence:
(a) It would seem that the concept of meaning deter-
mines the concept of consequence, but not of direct
consequence. (Let us assume that sentence b fol-
lows from a class A of sentences, but not directly;
can we, accepting the sentences of class A, reject
sentence b without changing the meaning of the
expressions occurring in these sentences?)
(b) Not only does the concept of meaning determine
the concept of direct consequence, but vice versa
(which the speaker did not mention in the talk, but
which follows from the speaker’s earlier work).
Only because of this can metalogic be considered
an adequate tool for carrying out epistemological
research.
(3) The view that the language of the natural sciences is
incomplete appears to be legitimate, but the given justi-
fication raises doubts. (For a certain definition of conse-
quence, arithmetic is a complete system.)

Ajdukiewicz had alluded to some deductive rules that would distinguish certain
consequences as indirect. Tarski’s comment (1) indicated that he would not consider such
rules in a characterization of deductive systems. His comment (2a) began with a verb in
15.9 Discussion of Wilkosz 1936, on the Significance of Logic 367

conditional mode: It would seem.61 Should Ajdukiewicz have really meant the clause that
followed it, he would have been contradicting Tarski’s comment (2b). Tarski’s parentheti-
cal remark in (2a) suggested a test that would judge that clause invalid. Together, (2a)
and (2b) explain why Tarski favored removing indirect consequences from consideration.62

Besides Tarski, ten others offered remarks about Ajdukiewicz’s presentation:

Jerzy Braun Henryk Mehlberg Adam Wiegner
Piotr Chojnacki Witold Steinberg Helena de Willman-Grabowska
Roman Ingarden Jarosaw Stpniewski Adam ótowski
Jan Leszczyski
Ajdukiewicz replied to some of them at length. Responding to Tarski, he discussed
alternative formulations using finitary and infinitary concepts of direct and general
consequence. He agreed that invoking the incompleteness of arithmetic depended on
the assumption of finitary inference rules in his proposed deductive system, and that
incorporation of an infinitary rule would lead to reinterpreting the basic thesis of tran-
scendental idealism. He concluded,
The subtle conceptual nuances that Dr. Tarski raised in the discussion illustrate the increased
level of philosophical investigations, when they are placed within the framework of modern
logic.

15.9 Discussion of Wilkosz 1936, on the Significance of Logic

On 26 September 1936, Witold Wilkosz presented the paper The Significance of Mathe-
matical Logic in Mathematics and Other Exact Sciences at a plenary session of the Third
Polish Philosophical Congress in Cracow. The congress is described in section 14.2. His
paper is summarized here, followed by a translation of Alfred Tarski’s [1936] 2014b
comment about it in the box on the next page. The speaker’s response is paraphrased
after that.63

In his presentation, Wilkosz first summarized the development of logic as a tool for the
justification of mathematics. He noted that mathematics always progressed in steps
ahead of the logic it needed: for example, with Euclid, Newton, Cauchy, and Cantor.
Researchers in logic had stressed consideration of axioms and primitive terms but
neglected the rules of deduction. However, recent work had permitted study of the
consistency of axiom systems and of logic itself. Logic should be, but was not yet, shaped
to allow no doubt of its consistency.

61
In Polish, wydawaoby si.
62
In his longer 1937 paper, Ajdukiewicz used the term consequence (wynikanie) repeatedly, but only with
the adjective direct (bezporedni). He did not seem to consider indirect consequences.
63
The Polish title of Wilkosz 1936 was Znaczenie logiki matematycznej dla matematyki i innych nauk
cisych. For information about its author, see the box on page 369.
368 15 Assorted Contributions

Wilkosz surveyed the state of the logical foundations of various disciplines. They were
most highly developed for mathematics, but physics had not yet reached that level of
precision and it was not clear whether that would even be beneficial. He mentioned the
possibility of developing logical foundations for social and even humanistic sciences, and
noted interesting work in foundations of theology, where the underlying assumptions
were not in question! Wilkosz pointed out that the development of logic has ignored
certain aspects of language—for example, shades of the grammar of conditional sen-
tences—and suggested that the time will come for that study. He alluded to work in
progress at Cracow on both broadening the scope of logic and simplifying its presentation.

After Wilkosz’s presentation, three scholars offered comments: J. M. Bocheski,

Benedykt Bornstein,64 and Alfred Tarski. Bocheski noted, and the speaker agreed, that
logic can and should also be developed as an independent science. Bornstein attempted
to relate Wilkosz’s paper to his own development of logic; the speaker pleaded
unfamiliarity with that alternative approach. The following translation of Tarski’s
comment is listed as Tarski [1936] 2014b in this book’s bibliography. It adheres to the
same standards as earlier translations in this book.

Alfred Tarski:
(1) Metalogic and metamathematics can give much to math-
ematics: even already they have given it a thing or two.
For example, theorems according to which all state-
ments of a certain form can be proved in some theory
can have an essential meaning for mathematics—and
also theorems that allow deduction of the properties of
mathematical notions from the content of a definition.
(2) Application of the axiomatic method and the framework
of logic is being tried in the various sciences, even the
nonmathematical ones. The research by Joseph H.
Woodger in the field of foundations of biology deserves
attention.

Responding, Wilkosz agreed with Tarski’s first comment, but suggested that it stem-
med from Tarski’s mishearing him. Wilkosz mentioned that he could give other examples
like those in Tarski’s second comment, but he regarded them as still fragmentary.

64
B enedykt B ornstein was born in 1880 in Warsaw, then part of the Russian Empire. He earned the
doctorate in philosophy at Lwów in 1907, and from 1918 taught at the Free University in Warsaw. After
World War II Bornstein taught at the University of ód. For information about Bocheski, see the box
on the next page.
 Witold Wilkosz
Broadcasting, 1929

Witold Wilkosz was born in 1891 in Cracow, in the Austrian Empire. His father taught Polish in a
gimnazjum there. A school classmate of Stefan Banach, Witold had extremely broad interests and
talents. While in gimnazjum, he won a study trip to Beirut with a paper on semitic linguistics. In
1910 he entered the university of Cracow to study philology. In 1912 he transferred to the University
of Turin, in Italy, and studied analysis with Giuseppe Peano. Wilkosz served during 1914–1915 in
Józef Pisudski’s Polish Legion, then returned to Cracow to study mathematics. After earning a
credential in 1917, Wilkosz taught for three years in private secondary schools in nearby cities, and
in Cracow he attended lectures on law. In 1918 he earned the doctorate there with a dissertation on
measure theory, supervised by Stanisaw Zaremba. He was awarded the venia legendi at Cracow in
1921 for further work in analysis, and in 1922 became a junior professor there. Wilkosz continued
research in analysis and also logic, and wrote several texts on elementary and advanced mathematics
and works popularizing mathematics. He taught courses for teachers for the Ministry of Religious
Denominations and Public Education and was heavily involved in the development of radio, both in
the technology and in production of educational broadcasts. In 1936 he became full professor. During
World War II his health deteriorated. He continued to teach at the School of Economics but died of
pneumonia in 1941.*
Innocentius Maria Boche Ĕski was born in 1902 in Czuszów, in south-central Poland, became a
Dominican monk, earned doctorates in philosophy and theology, then taught at the Pontifical College
in Rome until 1940. He served with the Polish Army during World War II, then became professor
and an academic leader at the University of Fribourg, in Switzerland. Fr. Bocheski wrote many
major studies in the history of philosophy, often using his birth name Józef. He died in 1995.†
Joseph Henry Woodger was born in 1894 in Great Yarmouth, England. During 1911–1922, except
for an interruption for World War I military service, he studied at University College, London. From
then until his retirement he taught at the university’s medical school. Woodger became a major
scholar of the foundations of biology. In particular, he carried to that field the ideas of the Vienna
Circle. Woodger met Tarski in 1935 at a function of that group. Tarski’s first publication in English
(1937a) was a technical appendix to Woodger’s 1937 book The Axiomatic Method in Biology. Woodger
edited the first published collection of Tarski’s works, the [1956] 1983 book Logic, Semantics, Meta-
mathematics. Woodger died in 1981.‡

†
* Pawlikowska-Broek 2011, Albiski 1976, Morscher, Neumaier, and Simons 2011
Rakoczy-Pindor 2003, redniawa 1961. ‡
Popper 1981.
370 15 Assorted Contributions

15.10 Discussion of Zawirski 1936, on Synthesis

On 26 September 1936, Zygmunt Zawirski presented the paper Concerning Scientific

Synthesis at a plenary session of the Third Polish Philosophical Congress in Cracow. The
congress is described in section 14.2. Zawirski’s paper is summarized here, followed by
a translation of Alfred Tarski’s [1936] 2014c remark about it. An explanation of that
remark is included, and the speaker’s response is paraphrased.65

Zawirski indicated that scientific synthesis and metaphysics cannot be strictly sepa-
rated and spoke about providing a basis for metaphysics that could fit within the Vienna
Circle guidelines. He suggested that although this subject is generally regarded as being
concerned with hypotheses that are essentially untestable, some of their consequences
may not be. He advocated a search for such statements, and underlined the delicacy and
precision with which that goal should be pursued. He mentioned two recently published
works that would help: Tarski’s now famous monograph The Concept of Truth in Formal-
ized Languages and Kazimierz Ajdukiewicz’s paper The Scientific World-View. Zawirski
also noted that care should be taken in selecting which of the logics now available should
underlie a scientific metaphysics, and said, “we have a legitimate right to assume that
only one of these systems corresponds best to reality.” 66

Besides Tarski, five others offered remarks about Zawirski’s presentation:

Jerzy Braun Bogumi Jasinowski Jarosaw Stpniewski
Piotr Chojnacki Maria Kokoszyska
The speaker responded to each of them. Maria Kokoszyska noted that the speaker could
be interpreted as suggesting that the Vienna Circle might have retreated from their
position of empiricism. That would not be fair, she said, for the Circle demanded then,
more than ever before, that scientific statements beyond those that are logically true
be verifiable experimentally. In the box on the next page is a translation of Tarski’s
remark. Listed as Tarski [1936] 2014c in this book’s bibliography, it adheres to the same
standards as earlier translations in this book. Tarski addressed at once Zawirski’s
presentation, an earlier one by Jan ukasiewicz, and a remark by Roman Ingarden about
ukasiewicz’s paper.67

65
The Polish title of Zawirski 1936 was W sprawie syntezy naukowej. For information about its author, see
the box on page 373.
66
Tarski [1933] 1983, Ajdukiewicz [1934] 1978; Zawirski 1936, 350.
67
ukasiewicz 1936. The journal identified Kokoszyska as Maria Lutman-Kokoszyska. For information
about these two philosophers and Ingarden, see boxes in sections 9.4 and 14.2 and a footnote in 15.6,
respectively.
15.10 Discussion of Zawirski 1936, on Synthesis 371

Alfred Tarski. Numerous talks and remarks in the current

Congress include polemical moments with respect to the
Vienna Circle, mainly due to the conflict this Circle has with
metaphysics; such moments occurred for example in the talks
of Professors ukasiewicz and Zawirski as well as in Professor
Ingarden’s remarks following the talks of Professor ukasie-
wicz. One of the reasons for this controversy seems to be a
terminological misunderstanding. The Vienna Circle denies
the scientific nature of any statements that cannot be checked
empirically or68 are not logical statements: in particular,
metaphysical statements. Moreover, simply speaking, all
statements that can be resolved—proved or refuted—by a
pure deductive method based on just the axioms and inference
rules accepted in the language are included as logical state-
ments; therefore, in the understanding of the Vienna Circle,
to the metaphysical statements belong only statements that
cannot be proved either by a deductive or by an empirical
method. On this point there is, as it seems, no essential
difference between the Vienna Circle and Professor ukasie-
wicz, who emphatically underlined that he knows of only two
scientific methods to prove statements—the deductive and
the empirical. Deductive metaphysics, whose possibility
Professor ukasiewicz raised, might be included by the
representatives of the Vienna Circle in parts of the logical
language; they would regard its propositions as logically true
statements. There is also no essentialdifference between the
Vienna Circle and the speaker, who recognizes the scientific
nature of metaphysical statements provided that they admit
empirical verification (or refutation); the difference is only
that the Vienna Circle might include statements metaphysical
in the speaker’s sense in empirical, scientific knowledge.
While between Professor Ingarden and the Vienna Circle
there seems to be no significant difference in understanding
of the term “metaphysics,” on the other hand deeper essential
differences are emphasized.

68
Tarski may have meant and instead of or: in Polish, i instead of lub. The original text of this sentence
is Ko o wiedeskie odmawia charakteru naukowego wszelkim zdaniom, które nie daj si sprawdzi na
drodze empirycznej lub nie s zdaniami logicznemi w szczególnoci za zdaniom metafizycznym.
372 15 Assorted Contributions

Contrary to Tarski’s remark, the present editors do not regard the published versions
of Zawirski’s and ukasiewicz’s congress presentations as notably polemic. Ingarden’s
published remark, with sufficient emphasis indicated, would indeed seem vehement.
These parts of the congress proceedings were probably redacted. ukasiewicz noted that
contemporary deductive logic, much more powerful than the old Aristotelian logic, can
be used with any sentences that are not concerned with size and number: such sentences
are logical truths.69 Moreover, if we use the new logic more heavily, we must expect it to
be applied to philosophical issues. Thus, many fundamental views should be reconsidered.
Finally, the discovery that bivalence is not a necessary feature of logic should cause great
upheaval. Ingarden’s remark advocated resisting the influence of the Vienna Circle,
which ukasiewicz had not mentioned, and emphasized the search for appropriate
methods of analyzing metaphysical questions, as had ukasiewicz.

It is perhaps significant that Tarski was careful here to describe the philosophical
position of the Vienna Circle without claiming it as his own.70

In response to Kokoszyska and Tarski, the speaker acknowledged that defending

metaphysics on the basis of experience does not affect the fundamental thesis of the
Vienna Circle. But he stated two reservations about the Circle. The first, he said,
coincided with that expressed in another recent article by ukasiewicz: 71 the Circle
favored the a priori sciences too much over the empirical and thus reduced too many
claims of empirical science to purely linguistic statements. Zawirski’s second reserva-
tion concerned the interdependence of mental and physical phenomena that underlie
empirical psychology.

69
ukasiewicz’s words were Wspóczesna logika matematyczna okazaa, e metod demonstratywn, czyli
matematyczn, mona stosowa i do zda, które nie dotycz wielkoci i liczby. Takimi zdaniami s prawa
logiczne, ... . They are remarkably evocative of the main point of Tarski’s posthumous 1986b paper What
are logical notions?
70
According to Jan Woleski (1993, 322), Tarski “was very self-restrained in expressing his philosophical
views.” See also the annotation of Tarski [1965] 2007 in section 16.1.
71
ukasiewicz [1936] 1970.
15.10 Discussion of Zawirski 1936, on Synthesis 373

Zygmunt Zawirski
around 1930

Z ygmunt Z awirski was born in 1882 near Tarnopol, then in the Austrian Empire (now Ternopil, in
Ukraine). His father, a Polish patriot, had taken refuge there after the 1863 uprising against the
Russian Empire. Zygmunt worked in schools during his 1901–1906 university studies in Lwów and
for some years afterward; he earned the full teaching credential in 1907. By 1910 he had completed
doctoral studies at Lwów under Kazimierz Twardowski, with a dissertation overlapping philosophy
and psychology. For a year afterward, Zawirski studied in Berlin and Paris. During 1922–1928 he
taught at the Lwów Polytechnic University. He earned the venia legendi in 1924 working with
Cracow philosopher Wadysaw Heinrich, then was appointed professor at the University of Pozna
in 1928. Zawirski continued research in fundamental areas of philosophy of science, emphasizing
the application of formal logic to physics problems, and was particularly concerned with the interpre-
tation of time. He continually and strongly emphasized an empirical approach, but stopped short
of the extreme views of the logical positivists. Zawirski was promoted to full professor in 1934, and
moved to the University of Cracow in 1937. He served as dean there both before and after World War
II. During that conflict, he taught classes clandestinely. From 1934 on, Zawirski edited the journal
Kwartalnik filozoficzny; from 1938, he was president of the Cracow section of the Polish Philosophical
Society. Zawirski was an especially effective communicator and statesman, both within the nation
and in its relations with foreign academics. He died in 1948.*

*Jadczak 1993a, Szumilewicz-Lachman 1994—these studies disagree on some details.

374 15 Assorted Contributions

15.11 Discussion of Kokoszy Ĕ ska 1936a, on Relative Truth

On 26 September 1936, Maria Kokoszyska presented the paper Concerning the Relativity
and Nonrelativity of Truth at a plenary session of the Third Polish Philosophical Congress
in Cracow.72 The congress is described in section14.2. Kokoszyska’s paper is summa-
rized here, and Alfred Tarski’s [1936] 2014d remark about it is translated here, an expla-
nation of that remark is included, and the speaker’s response is paraphrased.73

A former student of Kazimierz Twardowski, Kokoszyska had begun participating in

meetings of the Vienna Circle in 1934, and soon earned a reputation as a representative
of the Lwów–Warsaw School of Logic concerning such matters. In her presentation she
did not refer explicitly to Tarski’s new theory of truth.74 She referred to an ongoing
dispute about whether, in a description of a sentence S as true, truth should be regarded
as an absolute concept or as relative to other considerations. She indicated that the posi-
tivists tended to favor relativism. Truth, in her view, should be regarded as relative to
the meanings of the terms in S. Investigation of meanings was prerequisite to further
analysis of the language of S. Without that, the absolutist position would gain authority.

After Kokoszyska’s presentation, Walter Auerbach remarked that her concept of a

relative term seemed to differ from the everyday one.75 Tarski’s much longer remark is
translated in the box on the next page. Listed as Tarski [1936] 2014c in this book’s
bibliography, the translation adheres to the same standards as earlier ones in this book.
Tarski first suggested that it would be simpler to construe truth as relative to the lan-
guage in use. He did not say that this would suffice, though he may have implied tacitly
that linguistic analysis would constitute a tool in studying meaning. Tarski’s second
comment was addressed not so much to Kokoszyska’s paper, but to her, personally, and
to others doing similar studies. He did not want his claim of simplicity to appear
belittling, particularly of researchers for whom he had great respect, and of one with
whom he enjoyed very close acquaintance.76

72
The congress proceedings list Kokoszyska with this surname and under her married names Kokoszyska-
Lutmanowa and Lutman-Kokoszyska (Polish Philosophy Congress 1936, 351, 424, 544).
73
The Polish title of Kokoszyska 1936a was W sprawie wzgldnoci i bezwzgldnoci prawdy. For informa-
tion about its author, see a box in section 14.2. An English translation, by Jan Tarski, of part of Alfred
Tarski’s remark was included in Jan Woleski’s paper Tarski as a Philosopher (1993, 325). The present
translation is new; it seems consistent with that one.
74
Rojszczak 2002, 36; Mancosu 2008, 195; Tarski [1933] 1983.
75
W alter A uerbach , a former student of Kazimierz Twardowski and research collaborator of Tadeusz
Witwicki, was then living in Warsaw and publishing on the philosophy and psychology of memory. He
was murdered by the Nazis in Lwów in 1942. (Jadacki and Markiewicz 1993, 159–160.)
76
Woleski has discussed this interchange, including Kokoszyska’s response. His comments (2001b, 73)
on her paper and Tarski’s first remark have a slightly different thrust. He also discussed Tarski’s second
remark, without regard to its context (1993, 325). For the personal relationship between Kokoszyska
and Tarski, see Feferman and Feferman 2004, 88–92.
15.11 Discussion of Kokoszyska 1936a, on Relative Truth 375

Alfred Tarski. (1) The speaker’s words show, among other

things, that the notion of truth—in one of its interpreta-
tions—should be made relative to the notion of meaning.
Would it not be simpler to make the notion of truth relative
to the notion of language, which seems to be the clearer
notion and logically less complicated than the notion of
meaning?
(2) The second remark bears a more general character.
In recent times one could observe more than once the follow-
ing phenomenon: philosophers who are cultured in logic try
to apply the apparatus and methods of contemporary logic
to various classical problems of philosophy, and one finds
then that these problems, which in some cases were the
objects of philosophical investigations and speculations over
long centuries, acquire a more or less trivial character after
being subjected to logical analysis. By its nature, this
phenomenon gives rise to various doubts. In the first place,
the question arises whether by clarifying a problem and
liberating its formulation from vagueness and from various
imprecisions, one captures at the same time the “essential”
intentions of those who posed this problem or thought about
it in former times, even if they could not give to their investi-
gation appropriate formal dress. It is possible, and even
quite likely, that doubts of this kind are in many cases well
founded. Nevertheless, even in these cases the work of those
logicizing philosophers is surely not in vain: it forces oppo-
nents into making sufficiently precise the “essence” of the
problems at issue, so as to correspond to the demands of logic
and methodology, and so as to prevent future discussion of
these problems from having the character of a continuous
chain of never-ending misunderstandings.

Kokoszyska’s response has been described by the Catalan philosopher Manuel García-
Carpintero as conceding that Tarski may be correct
if languages are individuated not just by the well-formed combinations of sounds or inscriptions
belonging to them, but also by their correct translations into the language in which the
definition is given. If languages are only individuated “formally,” i.e. in the first way, then any
truth-definition for a language should be taken to be additionally relative to the way of
translating its sentences into sentences of the language in which the definition is carried out.77

77
García-Carpintero 1999, 143–144.
376 15 Assorted Contributions

Kokoszyska indicated that this second relativization is what she had called “relativiza-
tion to the meanings.” She said,
I am inclined both to propose and to consider my “formal” formulation of the problem of
the relativity of truth as an approximate approach to the traditional problem rather than a
final clarification.78

Responding to Tarski’s use of the word trivial, she claimed that granting the trivial nature
of this issue would result in characterization of both the notion of truth and that of
meaning as trivial, a position that would not garner much support.

15.12 Two Letters to Sierpi Ĕ ski, 1946 –1947

Two letters from Alfred Tarski to Wacaw Sierpiski are translated in the boxes in this
section. This is their first publication. The letters connect Tarski’s research in Poland
to his career in Berkeley. These introductory paragraphs and some later footnotes con-
tinue the account of his life and career in Poland that was included in chapters 8 and 14.
For further information about Tarski’s relocation to the New World, see Feferman and
Feferman 2004, particularly its chapters 6 and 7.

After the September 1939 conference at Harvard, Tarski spent three years in tempo-
rary research and teaching positions in the eastern United States. In September 1942
he took another temporary position as “utility” lecturer in mathematics at the University
of California, Berkeley. To a large extent, the university’s offerings were then a makeshift
contribution to the war effort. Nevertheless, Tarski quickly began creating a research
environment, to continue the work he had begun in Poland in earlier years. The univer-
sity recognized Tarski’s eminence and appointed him associate professor in 1945, then
full professor in 1946. World War II ended in Europe in April 1945. Two months later
Tarski earned United States citizenship. His wife and children finally arrived from
Poland in January 1946. With his salary and loans he bought a house in the hilly neigh-
borhood north of the Berkeley campus, at 1001 Cragmont Avenue. The Fefermans called
it “charming, if slightly pretentious.”79

Until the war’s end, Tarski had achieved only sporadic contact with his former Polish
teachers, colleagues, and students. When it became feasible, he resumed correspondence
with them. His 1946 and 1947 letters to Sierpiski show through their tone the sort of
relationships that Tarski would cultivate. They provide glimpses of

78
In her much longer paper on this subject, published three months later, Kokoszyska seemed to follow
Tarski’s suggestion (1936b, 155).
79
Feferman and Feferman 2004, 175.
15.12 Two Letters to Sierpiski, 1946–1947 377

• Tarski’s emerging research environment,

• his continuing research concerning cardinal arithmetic and the axiom of
choice and concerning equidecomposability in set theory,
• Tarski’s deep feeling for the murdered Adolf Lindenbaum and respect for
his work,
• obstacles to communication and publication, and
• the start of Tarski’s new research threads in universal algebra and the
theory of relations.80

The original letters were handwritten in Polish on stationery provided by the Univer-
sity of California.81 The translations, here designated Tarski [1946–1947] 2014, adhere
to the same standards as earlier translations in this book. In this case, moreover, their
layout is virtually identical with that of the originals. Personal names and bibliographic
references have been altered to conform with the conventions of the present book. The
present editors provided some bibliographical and historical information in footnotes that
fall outside the boxes.

1001 Cragmont Avenue

Berkeley 8, California
UNIVERSITY OF CALIFORNIA

DEPARTMENT OF MATHEMATICS 30 October 1946

BERKELEY 4, CALIFORNIA

Dear Professor!

I received your letter of September 14 ten days ago. I sent the manuscript to
the Annals of Mathematics while simultaneously expressing my willingness to
relieve you of editing it. I have not yet received a response; moreover, I have
asked to have the decision be announced to you and me at the same time.

I think your reasoning is really interesting for its simplicity and efficiency. I
would just like to make one remark (which I told Paul Erdos in due time). As I
recall, Adolf Lindenbaum had a more general result—a proof of the existence of
two sets not equivalent by countable decomposition in relation to an arbitrary

80
For biographical sketches of Sierpiski and Lindenbaum, see the boxes in section 4.1 and chapter 14,
respectively. Concerning Tarski’s continuing other research threads to Berkeley, particularly complete-
ness and decidability, see the long introduction to Tarski [1946] 2000 by its editor, Hourya Sinaceur.
81
Documents APAN sygn. III-194 in the Archives of the Polish Academy of Sciences.
378 15 Assorted Contributions

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 do not remember whether
the family of bijective transformations was subjected to some additional assump-
tions. 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. In this way, several appropriate topics
for master’s theses and perhaps even dissertations can be found. I spread some
propaganda to this end here. (Erdos’s article is to a certain extent the result of
this.) I will publish proofs—as a matter of fact I am doing this now—of Linden-
baum’s results that are closely associated with my own research or were
announced previously in our joint work.82

I really must congratulate you on the invitation to lecture at the Sorbonne.

I sincerely thank you for the card from Switzerland, the volume of Fundamenta
(it made a big impression here) and for sending copies of articles. I am promising
myself to reconsider some of your works in the near future—particularly those
that are related to Stefan Banach’s and my paradox. Please note that I have
always had the impression that the number 9 given by John von Neumann and
reduced by you can be easily obtained by directly analyzing the proof by Banach
and me.83

I sent a few of my articles printed in the United States along with general ship-
ments of books and articles for Polish mathematicians and logicians; I hope that
they have already arrived in Warsaw, or will arrive in the near future. On the
articles, I wrote in pencil the names of people for whom they are intended; I do
not need to mention that your name is on all the works which I believe may
interest you.

82
The first paragraph of this letter referred to Sierpiski’s 1947 paper Sur un théorème de A. Lindenbaum.
There, Sierpiski referred to Erdo  s’s 1943 paper Some Remarks on Set Theory, in the same journal. With
the aid

of the axiom of choice, Erdo  s had proved (page 144) that for every cardinal m there is a family
of 22 sets of real numbers, no two of which can be decomposed into m disjoint mutually congruent
0

subsets. Erdo s had credited Lindenbaum for announcing a less general version of that theorem, without
proof. Sierpiski proved the same result without using that axiom. Tarski thought that Lindenbaum
had achieved an even more general result.
83
Sorbonne refers to the University of Paris. Neumann’s number 9 is explained later, in footnote 89.
15.12 Two Letters to Sierpiski, 1946–1947 379

I am unsure whether I mentioned in my last letter to you that my book on

“cardinal algebras” should appear in print around 1 January 1947. It is a book on
the border between set theory and algebra, and contains reconstructions of all of
my and some of Lindenbaum’s results in the theory of bijective transformations, but
in completely new algebraic clothing. In a few weeks my student Bjarni Jónsson’s
and my seventy-page monograph on decompositions of simple finite algebraic sys-
tems will appear in print. I am working on a book on the theory of relations.84

I am curious whether Karol Borsuk has already left. In December, a mathematics

conference will be held at Princeton University in honor of its two-hundredth
anniversary. I was invited and hope that I can meet him there.85

I add sincere regards, and I ask [you] to pass along [my] regards to all [our]
colleagues.
[signed] Alfred Tarski
PS. For the information about the circumstances surrounding
my book about the concept of truth,86 thank you very much.

Tarski’s March 1947 letter to Sierpiski begins on the next page.

84
Tarski 1949a, Cardinal Algebras. Jónsson and Tarski 1947, Direct Decompositions of Finite Algebraic
Systems. Tarski had reported about his work on the theory of relations already in 1940, in an introductory
philosophy course at City College of New York: see Rosen 1985.
85
K arol B orsuk , born in 1905 in Warsaw, then part of the Russian Empire, earned the doctorate in 1930 from
the University of Warsaw with a dissertation on topology supervised by Stefan Mazurkiewicz. Borsuk
soon joined the faculty there and in 1946 became full professor. He contributed many major research
results and was instrumental in rebuilding Polish mathematics after World War II. He spent 1946–1947
at the Institute for Advanced Study in Princeton. Borsuk died in 1982. For further information, see
Keesling 1990. Tarski attended the Princeton Bicentennial Conference in summer 1946, and presented
the paper Tarski [1946] 2000.
86
Tarski 1933 or [1933] 1935.
380 15 Assorted Contributions

1001 Cragmont Avenue

Berkeley 8, California
UNIVERSITY OF CALIFORNIA

DEPARTMENT OF MATHEMATICS 18 March 1947

BERKELEY 4, CALIFORNIA

Dear Professor!

Thank you sincerely for your letter, papers, and books, as well as the invitation
to publish an article in Fundamenta Mathematicae about the equivalence of the
axiom of choice and the theorem quoted in your letter. I have already outlined
an article. I included certain results related to the older result, but obtained only
recently. Consequently, the article has grown slightly: I suppose it will take
about twenty pages to print. It will take some time, however, before someone
here corrects the article in terms of its style, and before I can rewrite it cleanly
and send it to you. I will of course be very grateful if anyone in Warsaw can
read the manuscript over carefully, check the accuracy of the reasoning and
later relieve me of doing the corrections; I would be extremely glad if Andrzej
Mostowski should take this up. Minor cosmetic changes may be made without
consulting me. In the article I use the symbolism of Fundamenta, varying slightly
from that which is used here (thus, symbols such as G, +, $, 6, 3, and not I,
, F, A, B ). Just one remark about cosmetics: I’ve heard here several times that
for British and American readers, use of the so-called expanded spacing is incom-
prehensible—the spaces are usually considered as errors made by the printing
machine. Instead of expanding the spacing, they use so-called “small caps”—that
is, the font is the size of the lowercase letters of the alphabet but has the capital-
ized shape. Because my article will be written in English, I will be glad if the
expanded spacing is replaced either by “small caps” or by the usual italic.87

I am truly happy that you are announcing other results by Lindenbaum and
me. You may be interested in knowing that some of those results (that is, those
that do not require the axiom of choice, and among others, a proof of the theorem
on the relationship between k # m  k # n and m  n) will appear simultaneously
elsewhere—in my book Cardinal Algebras (Oxford University Press, New York).*
They will appear there as an application of a certain general theory of an abstract
and algebraic character. It will not even be immediately apparent that the proof
does not require the axiom of choice—because in putting these results in abstract

* Unfortunately, the printing of this book has not yet begun. (It was to be published 1 January
1947). As you see, even here these things are not easy.

87
The article was Tarski 1948, Axiomatic and Algebraic Aspects of Two Theorems on Sums of Cardinals.
Tarski’s request that SMALL CAPS be used instead of e x p a n d e d l e t t e r s p a c i n g was not heeded.
15.12 Two Letters to Sierpiski, 1946–1947 381

algebraic terms, some limited use of the axiom of choice is necessary. But, follow-
ing the comments that I included in the book, the intelligent reader will likely
come up easily with purely set-theoretic proofs, not relying on the axiom of
choice.88 The material that I included in articles in Fundamenta will be helpful.

You probably heard from Borsuk [about] a result by a local colleague of mine,
Raphael M. Robinson (who is lecturing this year at Princeton). The result is
really beautiful. It concerns the equivalence between a ball and the union of two
balls with the same radius. Not being familiar with your earlier work in this field,
he obtained a stronger result—the number 5 instead of 9 or 8. Moreover, one
part consists of just a single point—the center of the ball—and this part can be
eliminated completely if, instead of balls, one considers their surfaces. It is easy
to see that these results are already the best possible. What is more important,
Robinson shows that the Cantor–Bernstein method, which Banach and I have
used in our work, is not needed in the treatment of this problem (it is probably
important when considering the general question of the equivalence of two
“solids”); [Robinson’s] method has a direct character, and depends on modifying
Felix Hausdorff’s methods.89 Another colleague of mine, Anthony P. Morse (at
Berkeley, already published in Fundamenta), has also obtained a certain result
in this field which, in my opinion, deserves to be published. As you know, the
assertion about the nonexistence of a “paradoxical decomposition” of a segment
can be proved without aid of the axiom of choice. With my advice, Morse took up
a similar problem for the square and obtained an identical result (although with
quite different methods). I urged the two— Robinson and Morse—to send their
results to Warsaw. In principle they are inclined to the idea, but are quite
overwhelmed with writing. Perhaps if you could write to the both of them (if you
consider it appropriate to do so), we may obtain a better outcome.90

88
To analyze the relationships between cardinal-arithmetic theorems and the axiom of choice, the then-
standard definition of cardinal number must be avoided because it depends on the axiom. Alternatives
are to introduce cardinals axiomatically, as in Tarski 1924e, or through the notion of rank, as in Lévy
1979, chapters II–III.
89
In his 1947 paper On the Decomposition of Spheres, Robinson showed that a ball can be decomposed into
five pieces, one consisting of a single point, that can be reassembled to form two balls of the same radius.
Neumann noted in his 1929 paper Zur allgemeinen Theorie des Masses, without proof, that the task could
be done with nine pieces. In his 1945 paper Sur le paradoxe de MM. Banach et Tarski, Sierpiski proved
that eight pieces were sufficient. Robinson also showed that four pieces were not enough. His methods
stemmed from those of Hausdorff 1914a.
90
The paper Morse and Randolph 1940, on measure theory, had been scheduled to appear in Fundamenta
Mathematicae in 1939, but was delayed until 1945; it was published elsewhere in 1940. In his 1949 paper
Squares are Normal, Morse showed, without the aid of the axiom of choice, that no decomposition of a
square is paradoxical in the sense of Banach and Tarski [1924] 2014. Morse and Randolph had derived
that result as a corollary of a more general one that required the axiom: see sections 5.2 and 6.2.
382 15 Assorted Contributions

From time to time I receive inquiries about current and past volumes of
Fundamenta. Among others, the University of Nevada, Reno, took my advice and
wrote directly to Warsaw a few months ago, expressing their desire to acquire all
of the volumes of Fundamenta. I do not know whether their letter arrived,
because there has as yet been no response.

I am unsure whether I communicated to you the following result related to a

problem of Mikhail Suslin.91 The problem can be summarized as follows: let Z
be an ordered set satisfying condition
(W1) every transfinite sequence of disjoint intervals is at most
countable;
is this set ordered like the set of real numbers? Instead of (W1) we may consider
an equivalent condition (W2) by replacing the word “disjoint” with the word
“decreasing.” Of course, both conditions (W1) and (W2) are necessary for Z to
be like a certain set of real numbers. Now consider condition (W3) that is
stronger than (W1) and (W2): namely,
(W3) if P0 , P1 , ... , P[ , ... ( [ < W) is a sequence of intervals in Z such
that for each pair of intervals P[ and PK where [ < K we have
P[ $ PK = I or P[ ƒ PK , then the sequence is at most countable.
It can be shown that this condition (W3 ) is necessary and sufficient for Z to be
similar to a certain set of real numbers. The proof is not too difficult, but it seems
to me it is not trivial (in either direction). From here it is easy to obtain a set of
conditions characterizing the order of the set of all real numbers.

My wife and I send warm greetings to both of you. Please pass on my regards
to all your colleagues; I am writing to Kazimierz Kuratowski at the same time.

[signed] A. Tarski

91
The famous Suslin’s Problem has been the subject of many investigations in set theory.
Alfred Tarski in 1968
 16
Posthumous
Publications

Chapter 16 is an annotated bibliography of works by Alfred Tarski that were not listed
or were incompletely described in the definitive 1986 bibliography by Steven R. Givant.1
None were reprinted in Tarski’s 1986a Collected Papers. Each listed item begins with
the author–date designation used in this book, continues with a description adapted from
this book’s bibliography,2 and is followed by a brief annotation. In some cases, the anno-
tation includes Tarski’s own words, enclosed in a box with a double border.

Many new editions and translations are included.3 However, the numerous translations
in this book are not listed here, though they logically belong with the items that are.
Readers may use this book’s bibliography to construct items for them analogous to those
in this chapter. For example, the following item would represent the first translation in
this book:
Tarski [1921] 2014. A contribution to the axiomatics of well-ordered sets. In Alfred
Tarski: Early Work in Poland—Geometry and Teaching, translated and edited by
Andrew McFarland, Joanna McFarland, and James T. Smith, chapter 2. New York:
Birkhäuser. Translation of “Przyczynek do aksjomatyki zbioru dobrze uporzdkowa-
nego,” Przegld filozoficzny 24, 85–94, Givant 1986 item 21. This is Tarski’s first
published paper.

16.1 Papers

For additional translations of papers by Tarski into Polish, see the annotation for Tarski
1995–2001 in section 16.2.

1
Four works listed here were published during Tarski’s lifetime, but not listed in Givant 1986: Tarski
[1930] 2014h, 1965a, Tarski 1972, and Mates, Henkin, and Tarski 1959. Tarski 1995–2001 contains a
less extensively updated bibliography.
2
This book’s bibliography includes more details of publication.
3
Some translations into Polish are mentioned only in the annotations. Reprints are mentioned in this
chapter only when particularly germane to the discussions in this book. Chapters 16–18 are a by-product
of the research that underlies earlier chapters. The editors did not search exhaustively for other works
of Tarski not mentioned in items discovered during that work.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 385
DOI 10.1007/978-1-4939-1474-6_16, © Springer Science+Business Media New York 2014
386 16 Posthumous Publications

Tarski [1923] 1998. On the primitive term of logistic. In Leniewski’s Systems:

Protothetic, edited by Jan T. J. Srzednicki and Zbigniew Stachniak, 43–68. Dordrecht:
Kluwer Academic Publishers.
Stachniak translated this from Tarski 1923a, “O wyrazie pierwotnym logistyki,” Przegld filozo-
ficzny 26: 68–89, Givant 1986 item 23. That was the original published version of Tarski’s
Warsaw doctoral dissertation, which was supervised by Stanisaw Leniewski. The original
was not reprinted in Tarski’s 1986a Collected Papers. The French version Tarski 1923b and
its English adaptation in [1956] 1983, 1–23, do not include the entire dissertation.

Tarski, Alfred. [1928] 2010. Remarks on fundamental concepts of the methodology

of mathematics. Translated by Robert Purdy and Jan Zygmunt. In Universal Logic:
An Anthology, from Paul Hertz to Dov Gabbay, edited by Jean-Yves Béziau, 67–68.
Basel: Birkhäuser.
Purdy and Zygmunt translated this from Tarski’s 1928 abstract: his first published contribu-
tion to general metamathematics. For information on the sequence of lectures that this
presentation inaugurated, see section 9.4 and Zygmunt 2010.

Tarski, Alfred. [1931] 2014h. O pewnym systemie logiki matematycznej i wynikajcych

z niego zagadnieniach metodologicznych i semantycznych. Ruch filozoficzny 12
(1930–1931): 232. Abstract of a 15 April 1931 presentation to the Logic Section of the
Warsaw Philosophical Institute. Omitted from the Givant 1986 bibliography and
Tarski’s 1986a Collected Papers, this abstract is translated in the box below. This is
probably the first exposure of Gödel’s famous incompleteness theorem outside Vienna
(Feferman and Feferman 2004, 84).

Dr. A. Tarski: On a certain system of mathematical logic and the resulting methodologi-
cal and semantic issues. The Speaker had previously shown how to build a very simple
fragment of logic that is sufficient for building all known mathematics. Such a system,
although perhaps inconvenient in practice, is very convenient for examining mathemat-
ics. After outlining such a system, the Speaker acquainted the audience with the latest
result of Gödel in Vienna, based on such a system. Namely, Gödel has proved that the
system of integer arithmetic is undecidable,4 even the sentences just about addition and
multiplication. After sketching a proof in the framework of the system, the Speaker
discussed further momentous consequences that follow from Gödel’s result.

Tarski, Alfred. [1934] 2003a. Discussion of two papers:

Reichenbach [1934] 1936, “Die Bedeutung des Wahrscheinlichkeitsbegriffs für die
Erkenntnis”;
Zawirski [1934] 1936, “Bedeutung der mehrwehrtigen Logik für die Erkenntnis
und ihr Zusammenhang mit der Wahrscheinlichkeits-
rechnung.”

4
Tarski’s word was nierozstrzygalny.
16.1 Papers 387

With a Polish translation by Mariusz Grygianiec. In Alfred Tarski: dedukcja i seman-

tyka (déduction et sémantique), edited by Juliusz Jacek Jadacki, 12–14. Warsaw:
Wydawnictwo Naukowe Semper. Reprint and translation of Tarski [1934] 1936a.5
This is Tarski’s contribution to a discussion that took place after presentations by the German
and Polish philosophers Hans Reichenbach6 and Zygmunt Zawirski at the Eighth International
Congress of Philosophy in Prague, 2–7 September 1934.7 As a way of surmounting certain
difficulties with the foundations of probability theory, Reichenbach and Zawirski each had
suggested formulating the theory within some version of multivalued logic. Tarski claimed
that those measures would not suffice. Moreover, the difficulties would disappear if the theory
were based instead on a function that assigns to a sentence s and a set Y of sentences the
probability of s under the conditions Y; ordinary logic would then suffice. Tarski referred
to Keynes 1921, which followed that approach, and the more recent studies Ajdukiewicz 1928
and Mazurkiewicz 1932, which extended it. Thus, Tarski claimed, reformulating logic to meet
the needs of probability theory was neither necessary nor desirable.

Tarski [1934] 2003b. Discussion of Jørgensen [1934] 1936, “Die logischen Grundlagen
der Wissenschaft.” With a Polish translation by Mariusz Grygianiec. Ibid., 11. Reprint
and translation of Tarski [1934] 1936b.
This is Tarski’s contribution to a discussion that took place after a presentation by the Danish
philosopher Jørgen Jørgensen8 at the congress mentioned in the previous item. Jørgensen had
considered in turn the meanings of the terms that made up the title of his presentation:
science, foundations of science, logical foundations of science. He regarded a science as a set
of sentences in a given language L that are held to be true and can be ordered as a kind of
deductive system. Tarski objected that the description was too general: any set of sentences
of L could be so ordered. Jørgensen went to great length to describe the role logic should play
in that ordering, and in the ensuing discussion emphasized that held to be true is different from
true. Tarski noted that most of Jørgensen’s claims depended on the notion of “tautologous”
sentence, which for most such languages had not yet been adequately defined. Tarski had
already published papers defining the notion of true sentence, and would soon publish another
that would define the notion of tautologous (logically true) sentence and distinguish it from the
former notion: see section 15.6 and the following two items.

5
The text of Tarski [1934] 1936a is nearly identical to that of Tarski 1935–1936. Tarski [1934] 1936a was
described incorrectly in Givant 1986 as a discussion of a paper by Janina Hosiasson-Lindenbaumowa. It
was reprinted in Tarski 1986a, but misidentified in the same way.
6
H ans R eichenbach was born in Hamburg in 1891 and schooled there. His father was a merchant. After
studying at several universities, Hans earned the doctorate at Erlangen in 1916 for studies in probability
theory. By 1926 he was wielding major influence in the philosophy of science and was appointed professor
at the University of Berlin. Dismissed in 1933 because of his Jewish ancestry, he emigrated to Istanbul,
and five years later to Los Angeles, where he helped establish a major program of philosophical studies.
Reichenbach died in 1953. (Schuster 1975.)
7
For biographical information about Zawirski, see a box in section 14.10. The congress is described in
section 14.2.
8
J ørgen J ørgensen was born in Haderup, in northeastern Denmark, in 1894. His father was a vicar. Jørgen
was graduated from the Soro Academy in 1912, earned a master’s degree in philosophy in 1918, worked
as a secretary for eight years, and in 1926 was named professor of philosophy at the University of Copen-
hagen. A leading exponent of logical positivism, he died in 1969. (See Witt-Hansen 1969.)
388 16 Posthumous Publications

Tarski [1935] 1984. Sobre el concepto de consecuencia logica. In Lecturas de lógica II:
Selección, introducción y traducción, edited by Pilar Castrillo and Luis Vega Reñon,
section A.2, 178–192. Madrid: Universidad Nacional de Educación a Distancia.
Vega translated this from Tarski [1935] 1936b, “Über den Begriff der logischen Folgerung,”
Actes du Congrès International de Philosophie Scientifique, Sorbonne, Paris, 1935, volume 7,
1–11, Paris: Hermann & Cie., Givant 1986 item 36g. Vega’s translation is preceded by a
significant introduction, pages 155–177.

Tarski [1935] 2002. On the concept of following logically. Translated from the German
and Polish, with commentaries, by Magda Stroiska and David Hitchcock. History and
Philosophy of Logic 23: 155–196.
Stroiska and Hitchcock translated this from Tarski’s German version, cited in the
previous item, and from Tarski 1936b, “O pojciu wynikania logicznego,” Przegld
filozoficzny 39: 58– 68, Givant 1986 item 36a. In addition to analyzing the content of
this paper, they systematically noted how the two originals differ from each other and
from the English version in Tarski [1956] 1983, 409–430. The translators’ account
of their methodology is particularly useful: it guided the translations from Polish in
the present book.

Tarski [1939–1940] 1995. Some current problems in metamathematics. Edited by Jan

Tarski and Jan Woleski. History and Philosophy of Logic 16: 159–168. The editors
constructed this article from Tarski’s almost-complete notes for a presentation, prob-
ably given at Harvard University. Here is their summary (page 159):
“In this article the author first described the developments which brought to focus the
importance of consistency proofs for mathematics, and which led Hilbert to promote the science
of metamathematics. Further comments and remarks concern the (partly analogous)
beginnings of the work on the decision problem, Gödel’s theorems and related matters, and
general metamathematics. An appendix summarizes a text by the author on completeness
and categoricity.”

Tarski [1944] 1985. Definition. American Mathematical Monthly 92: 358. This consists
merely of the following quotation from Tarski 1944, 359.

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.

Tarski [1946] 2000. Address at the Princeton University Bicentennial Conference on

Problems of Mathematics (December 17–19, 1946). Edited and with an introduction
by Hourya Sinaceur. Bulletin of Symbolic Logic 6: 1–44.
According to the editor (page 1), “This article presents Tarski’s address ... together with a sep-
arate summary. Two accounts of the discussion which followed are also included. The central
topic ... is decision problems. The introductory note gives information about the Conference,
16.1 Papers 389

about the ... subjects discussed ... and about subsequent developments ... .” The editor’s commen-
tary details the continuity of Tarski’s research activity as he moved from Poland to California.

Tarski [1955] 1993. Sur la théorie des modèles. Translated and edited by Anne Preller.
In Philosophie de la logique et philosophie du langage, edited by Jacques Bouveresse,
volume 2, 137–158. Paris: Éditions Odile Jacob.
This survey of the basic concepts of model theory emphasizes algebraic more than logical
aspects. According to the editor’s note, it “is the translation of a communication presented
by Alfred Tarski in Paris between 26 September and 1 October 1955 at the Seventieth Inter-
national Colloquium, Le raisonnement en mathématiques et en sciences expérimentales, of
the Centre National de la Recherche Scientifique. The typescript with some annotations and
corrections in the hand of Alfred Tarski was sent to me by Jan Tarski.” The typescript is
in the Bancroft Library of the University of California at Berkeley, Tarski Archive 84/69,
Container 12.

Mates, Henkin, and Tarski 1959. Jan Kalicki, philosophy, mathematics: Berkeley and
Davis. Berkeley: University of California Digital Archives.
J an K alicki was born in Warsaw in 1922. He was educated there and at the University of
London, where he earned a doctorate in mathematics in 1948. A prolific scholar in logic, he
soon emigrated to the United States. In 1953 Kalicki became assistant professor of philosophy
at the University of California, Davis. Three months later, he died in an automobile accident.
Tarski and his colleagues Benson Mates and Leon Henkin constructed this obituary for the
Berkeley community.

Tarski [1965] 2007. Two unpublished contributions by Alfred Tarski. Edited by Fran-
cisco Rodríguez-Consuegra. History and Philosophy of Logic 28: 257–264.
These are edited transcriptions of remarks delivered in April 1965 in Chicago, and in July 1965
in London, at symposia devoted to implications of Kurt Gödel’s incompleteness theorems and
to recent results in set theory, respectively. Tarski contributed to the discussions after the
invited talks. In the first discussion, according to the editor (page 257), Tarski “takes advan-
tage of the opportunity to announce his general philosophical position, very likely for the first
time in public: Tarski claims to be an extreme anti-Platonist.” The bulk of Tarski’s comments
responded to previous talks by Hilary Putnam, Andrzej Mostowski, and Georg Kreisel.

Tarski 1986b. What are logical notions? Edited by John Corcoran. History and Philoso-
phy of Logic 7: 143–154. Givant 1986 item 87?. According to the editor’s introduction
(page 143),
“In this article Tarski proposes an explication of the concept of logical notion. His earlier well-
known explication of the concept of logical consequence presupposes the distinction between
logical and extra-logical constants (which he regarded as problematic at the time). Thus, the
article may be regarded as a continuation of previous work. ... [Felix] Klein’s Erlanger Pro-
gramm for classifying geometrical notions is sketched ... . Then, generalizing beyond geometry,
a notion (individual, set, function, etc.) based on a fundamental universe of discourse is said
to be logical if and only if it is carried onto itself by each one-one function whose domain and
range both coincide with the entire universe of discourse.”
390 16 Posthumous Publications

Henkin, Monk, and Tarski 1986. Representable cylindric algebras. Annals of Pure and
Applied Logic 31: 23–60.
With Leon Henkin and J. Donald Monk, Tarski had written the 1971–1985 monograph
Cylindric Algebras. This article investigates which cylindric algebras are isomorphic to
algebras of relations.

Tarski and Givant 1999. Tarski’s system of geometry. Bulletin of Symbolic Logic 5:
175–214.
From the abstract (page 175, written by Steven R. Givant): “This paper is an edited form of
a letter written by [Tarski and Steven R. Givant] to Wolfram Schwabhäuser around 1978. It
contains extended remarks about Tarski’s system of foundations for Euclidean geometry, in
particular its distinctive features, its historical evolution, the history of specific axioms, the
questions of independence of axioms and primitive notions, and versions of the system suitable
for the development of 1-dimensional geometry.” The definitive presentation of Tarski’s system
is the 1983 monograph Metamathematische Methoden in der Geometrie by Schwabhäuser,
Wanda Szmielew, and Tarski.

16.2 Monographs

Tarski 1965a. Bibliography of Alfred Tarski. Berkeley: Department of Mathematics,

University of California, Berkeley. Superseded by the Givant 1986 bibliography.

Tarski 1986a. Collected Papers. Four volumes. Edited by Steven R. Givant and Ralph
N. McKenzie. Basel: Birkhäuser.
Givant 1986 item 81m a was a preliminary edition. Maddux 1993 is a thorough review, with
corrections, of Tarski 1986a and the 1986 bibliography by Steven R. Givant.

Tarski and Givant 1987. A Formalization of Set Theory without Variables. American
Mathematical Society colloquium publications, 41. Providence: American Mathemati-
cal Society. Item 8_m in Givant 1986, with incomplete publication data.
According to the 1989 review by J. Donald Monk, the main ideas and results in this book stem
from Tarski’s work during 1940–1945. Givant made many independent contributions while
writing the book with Tarski. Monk explained,
This last book written (in part) by Tarski ... contributes both to our understanding of what mathe-
matics is, as well as to technical foundational research. ... A simple equational language L × is
introduced, and it is shown that set theory (for example, ZFC) can be translated into equations
of L × which have no variables, so that sentences derivable in set theory are translated into equa-
tions derivable by equational rules of inference from the translates of the set-theoretical axioms.
... It is shown that, in principle, mathematics can be developed in the very simple framework of
equations and substitution of equals for equals, rather than the customary basis in set theory
formalized in first-order logic.

The 1990 review by István Németi continues,

All quantifiers, individual variables, and so forth disappear from the language of mathematics, and
all mathematical statements appear in the form of equations (and the only proofs are ... derivations
16.2 Monographs 391

of equations from equations following ... age-old rules). Surprisingly many mathematical state-
ments become shorter, more intuitive, and [more] suggestive.

Further, according to Németi,

The book provides a synthesis of the seemingly diverse schools of thought and research directions
[that] Tarski ... created or pursued seriously. ... The importance of finitistic methods in computer
science is well known, and ... the book provides a wealth of tools and results in this area. Another
connection is the theories and calculi of relations, the modern form of which Tarski originated.
And indeed, the book contains such a wealth of profound results and methods of this area not
available before (or anywhere else yet) that after reading the book one’s perspective of the whole
area is fundamentally transformed.

Tarski 1994. Introduction to logic and to the methodology of the deductive sciences. Fourth
edition. Edited and with a preface and biographical sketch by Jan Tarski. Oxford
Logic Guides, 24. New York: Oxford University Press. Item 41m in Givant 1986 lists
previous editions.
The third edition, Givant 1986 item 41m(10), was reprinted as an inexpensive paperback by
Dover Publications in 1995; it is still in print in 2014. According to the 1995 review by Elliott
Mendelson, this new edition, Tarski 1994,
... differs considerably from the third. ...The editor, Tarski’s son, has added a four-page “Short
biographical sketch” and a large number of new footnotes. Stylistic changes have been made and
the terminology and notation have been brought up to date. ... Some of the original exercises ...have
been replaced by somewhat more elementary problems.

Tarski 1996 (see a following item) is an augmented Polish translation of this work.

Tarski 1995–2001. Pisma logiczno-filozoficzne. Volume 1: Prawda. Volume 2:

Metalogika. Translated and annotated, with introduction, by Jan Zygmunt. With
a bibliography by Steven R. Givant, supplemented by Zygmunt. Warsaw: Wydaw-
nictwo Naukowe PWN. The title means Logical-Philosophical Writings; Prawda
means Truth.
These volumes (1, 2) contain reprints (r) or new Polish translations (t) of the following works
of Tarski—two are excerpts (x):
[1930] 1983a . . . . . . . . 2 t [1938] 1939 . . . . . . . . . 2 t 1953 . . . . . . . . . . . . . . . 2t
[1930–1931] 2014 . . . . 1r [1938] 1983 . . . . . . . . . 2 t [1955] 1993 . . . . . . . . . 2t
[1930–1936] 1992 . . . 1tx 1939b . . . . . . . . . . . . . . 1t 1965b . . . . . . . . . . . . . . 2t
1932 . . . . . . . . . . . . . . . 1t [1939–1940] 1995 . . . . 2 t 1969 . . . . . . . . . . . . . . . 1t
1933 . . . . . . . . . . . . . . . 1r [1940] 1967 . . . . . . . . . 2 t 1986b . . . . . . . . . . . . . . 2t
1934 . . . . . . . . . . . . . . . 2r [1942–1947] 1999 . . . 2 tx Lindenbaum/Tarski
[1934] 1983 . . . . . . . . . 2 t 1944 . . . . . . . . . . . . . . . 1t [1935] 1983 . . . . . . . 2t
1936a . . . . . . . . . . . . . . 1r [1944] 1987 . . . . . . . . . 1t ukasiewicz/Tarski
1936b . . . . . . . . . . . . . . 1r [1946] 2000 . . . . . . . . . 2 t [1930] 1983 . . . . . . . 2t
[1936] 2014d . . . . . . . . 1r [1948] 1957 . . . . . . . . . 2 t
Reviewed in Cesarski1995, this work also incorporates the biographical sketch Zygmunt 1995
and informative notes on all included papers. It contains Givant 1995, a version of the Givant
1986 Tarski bibliography, updated and tailored to fit a more limited scope.
392 16 Posthumous Publications

Tarski 1996. Wprowadzenie do logiki i do metodologii nauk dedukcyjnych. Second

edition. Translated by Monika Sujczyska from the fourth English edition, Tarski
1994 (see an earlier item). Edited and with a preface by Jan Tarski, with a supplement
on elements of predicate logic by Witold Marciszewski. Warsaw: Filia Uniwersytetu
Warszawskiego, Philomath, Aleph.
This was reviewed extensively in A. Wójtowicz 1996. The first edition was issued in 1995 by
the same publisher with a different supplement by Marciszewski: “Logika a dzielno
umysu”—“Logic and the boldness of mind.” In his letter 1996a, Jan Tarski complained about
the inappropriateness of the supplement. That may have occasioned the second edition.

16.3 Letters

Tarski [1918] 2014. Curriculum Vitae. Translation of handwritten document “ yciorys”

in the Archive of the University of Warsaw, sygnatura Teitelbaum-Tarski Alfred: RP
2909. Displayed in the box below and discussed in section 1.1. This is its first publica-
tion. The original has no other marks. Tarski’s signature is reproduced in chapter
3, page 38.

Capital City of Warsaw, 30 September 1918

Curriculum Vitae

I was born on 14 January 1901 in Warsaw, to my father Ignacy (Izaak) and mother Róa
(Rachel) Prussak.
I received my first instruction at home. In September 1910, I was sent to first grade
at State Gimnazjum IV in Warsaw. There I passed five classes. In September 1915,
having been promoted, I was accepted into grade VI of the eight-grade School of the
Mazovian Land.9 I attended this school for three years. In the spring of the current year
I passed the final exams successfully and obtained a certificate of maturation.
Now I plan to study with the Philosophical Faculty of the University of Warsaw,
wishing to devote myself to scientific work in the field of biology.
[signed] Alfred Teitelbaum

Tarski [1924] 2004. Letter to a Warsaw city government office, 18 March. In Feferman
and Feferman 2004, 40. The original handwritten document “Komisariat Rzdu na
m. st. Warszawa” is in the Archive of the University of Warsaw, sygnatura Teitelbaum-
Tarski Alfred: RP 2909. This is Alfred Tajtelbaum’s application for changing his
surname to Tarski.

9
In Polish, Szkoa Ziemi Mazowieckiej.
16.3 Letters 393

Tarski [1924] 2014a. Letter to the dean of the Philosophical Faculty, 2 June 1924. In
section 16.1 of the present book, Tarski 2014. Translation of handwritten document
“Do Pana Dziekana Wydziau Filozoficznego Uniwersytetu Warszawskiego” in the
Archive of the University of Warsaw, Signatura Teitelbaum-Tarski Alfred: RP 2909.
This is displayed in the box below and discussed in chapter 3. Alfred’s signature is
reproduced there. This is its first publication. The dean was Wiktor Lampe, a chemist.

To
Dean of the Philosophical Faculty of the University of Warsaw
Application
After obtaining the doctoral degree, wishing to complete my education in certain
branches of mathematics, especially in the field of theoretical physics, I respectfully ask
the Dean for permission to continue my studies in the mathematics and natural (physical)
sciences division of the Philosophical Faculty.
[signed] Dr. Alfred Tarski
Warsaw, 2 June 1924

I agree to the continuation of studies [Stamp of approval]

in the mathematics division.
3 June 1924 [signed] W. Lampe

Tarski [1930–1936] 1992. Drei Briefe an Otto Neurath. Edited,with a foreword, by Rudolf
Haller. Translated by Jan Tarski. Grazer philosophische Studien 43:1–32. The letters
are transcribed, then translated into English. According to the editor (pages 1–2),
“There are in the main two themes treated in Tarski’s letters to Neurath.10 The first presents
information about scientific philosophy in Poland and its possible links to the work of the
Vienna Circle. ... The second theme touches on syntax and semantics and Tarski’s influence
on research and discussion in the Vienna Circle.”

Tarski [1933] 2003. Letter to Kazimierz Twardowski, 12 July. In Alfred Tarski: dedukcja
i semantyka (déduction et sémantique), edited by Juliusz Jacek Jadacki, 19. Warsaw:
Wydawnictwo Naukowe Semper. This is about preparing Tarski 1933 for publication.
(For information on Twardowski, see a box in section 1.2.)

10
O tto N eurath , born in Vienna in 1882, was the son of a noted political economist. He earned the doctorate
in that same subject from the University of Berlin in 1906. Throughout his life, he held a variety of posi-
tions in the intellectual and socialist-political communities. He habilitated in 1917 but never became a
professor. Neurath was particularly influential in the study of symbolic languages and their use in educa-
tion, and in the philosophy of logical positivism. During the 1920s, with Hans Hahn and others, he helped
found the Vienna Circle. After the Nazi takeover in 1934, he fled to the Netherlands with his wife, Hahn’s
sister. She died there, and he then married another philosopher, the sister of the German mathematician
Kurt Reidemeister. In exile, Neurath founded the Encyclopedia of Unified Science, a major organ of logical
positivism. He fled again to England in 1940, and died there suddenly in 1945. (See Cat 2010.)
394 16 Posthumous Publications

Tarski [1934] 2014. Letter to the Ministry of Religious Denominations and Public
Education, 29 November. Translation of handwritten document N921/R15345, page
8, preserved in the archival fund No. 14, Ministerstwo Wyzna Religijnych i Owiecenia
Publicznego w Warszawie 1917–1939, sygnatura 6216, of the Archiwum Akt Nowych
(Central Archives of Modern Records) in Warsaw. Displayed in the box below, and dis-
cussed in section 9.1. At the bottom is the signature of the rector, the noted physicist
Stefan Piekowski, and a red stamp indicating that he read the letter on the same day.
The original document and its attached letter by Jan ukasiewicz, page 11 of the same
archive file, was transcribed in Jadacki , op. cit., 25–26.

To the Ministry of Religious Denominations and Public Education

(through the Dean of the Division of Mathematical and Natural Sciences
of the University of Warsaw)
I wish to go abroad, to Austria, France and Czechoslovakia, for scientific studies.
Please be so kind as to grant me permission to obtain a discounted passport for the period
between 15 December 1934 until 15 December 1935.
I have a serious chance of receiving a Rockefeller grant for a period of eleven to twelve
months, starting 1 January 1935. The formal decision of the committee is to take place
no earlier than December, and news of the decision on that date would be impossible to
use to obtain a passport on 1 January. Regardless of the grant, I want to leave in
mid-December for Prague, Czechoslovakia, and Vienna, to conduct with professors Rudolf
Carnap and Karl Menger a series of scientific conversations that are important to me.
[signed] Dr. Alfred Tarski
29 November 1934 docent, University of Warsaw
Two enclosures Warsaw, Koszykowa 51, apartment 14

I strongly endorse this application, referring to the attached

letter from Prof. Jan ukasiewicz.
[signed] Stefan Mazurkiewicz

Tarski [1935] 2014. Two letters to the Ministry of Religious Denominations and Public
Education, 25 November and 7 December. Translations of handwritten documents,
N840/R12478, pages 21–22, and N873/R13421, page 24, preserved in the archival fund
No. 14, Ministerstwo Wyzna Religijnych i Owiecenia Publicznego w Warszawie 1917–
1939, sygnatura 6216, Archiwum Akt Nowych (Central Archives of Modern Records)
in Warsaw. Displayed in the box on the facing page and discussed in section 9.1.
Transcriptions of the original letters were published in Jadacki 2003b, 26–27. Jadacki mistook
the date of the first letter. Below the body of second letter, in a position analogous to Mazurkie-
wicz’s endorsement of the first, someone added the word Czytaem (I have read) and an illegible
signature. Both letters carry a red stamp with that word, a date, and the signature of the Rec-
tor, Stefan Piekowski. Both letters carry various ministry stamps and endorsements.
16.3 Letters 395

To the Ministry of Religious Denominations and Public Education

(through the administration)
In January of this year, I went abroad for several months to study, using the leave
granted me by the Ministry. As a consequence of this trip, I was forced to give up the
apartment which I previously occupied. Incidentally, the apartment was too tight for me
as my family increased in size with the birth of a son. After returning to the country, I
encountered great difficulties in attempts to rent a new apartment. For housing that
remains under the protection of the Tenant Protection Act, a large sum was required of me
as a so-called consideration11 (the amount of which in recent years has significantly
increased). For housing in new homes, exorbitant rent was required. Therefore I was forced
to live temporarily with my wife and child with a larger family in one sublet room.12 Of
course, under these housing conditions, I can neither prepare properly to fulfill my profes-
sional duties—university lectures—nor continue my research.
Recently, an opportunity for obtaining housing opened up to me: the Social Insurance
Institution granted me the right to purchase a property, an apartment of two and one-half
rooms in one of its newly built homes. An essential requirement for purchasing the
apartment is that I pay the sum of 3500 zlotys. (The rest I can pay off in long-term install-
ments.) Not having any savings, I cannot make this payment without financial aid from
the Ministry.
In this situation, I most politely appeal to the Ministry, asking for the advance of six
months salary, which I wish to pay off in thirty-six monthly installments. I dare to say that
this matter is extremely urgent for me: if I do not make the payment within the shortest
possible time, the apartment will be given to someone else.
With this application I enclose a certificate13 issued by the Social Insurance Institution.
Dr. Alfred Tarski,
Warsaw, November 25, 1935 Docent, Józef Pisudski University,
Adjunct, Philosophical Seminar,
I support this application the same University.
[signed] Stefan Mazurkiewicz
.........................................................................
To the Ministry of Religious Denominations and Public Education
(through the administration)
In connection with the application submitted by me on November 26 of this year and
containing a request for a salary advance for the purchase of an apartment, I am pleased
to announce that I managed to obtain the necessary amount through a private loan, and
therefore I withdraw my request.
Docent Dr. Alfred Tarski
Warsaw, December 7, 1935 Adjunct, Philosophical Seminar
Józef Pisudski University

11
Tarski’s word was odstpne.
12
Tarski’s words were podnajtym przy wikszej rodzinie. It is not clear whether he was emphasizing or
downplaying his reliance on his parents.
13
See Zakad Ubezpiecze Spoecznych 1935.
396 16 Posthumous Publications

Tarski [1937] 2003. Two letters to Kazimierz Twardowski, 1 and 24 May. In Jadacki,
op. cit., 19–20. These notes are about some publications of the Polish Logic Society.

Tarski [1939] 2003. Letter to the dean of the Division of Mathematics and Natural
Sciences of the University of Warsaw, 16 January. Ibid., 27. Tarski requested that
a record of his university studies be sent to the state pension system.

Tarski [1940] 1996–1999. Letter to Józef M. Bocheski, 26 April. Filozofia nauki 4:

123–125, 7: 197–199. Polish transcription, with an English translation by Jan Tarski.
This letter, written from New York, was described and quoted in the Fefermans’ biography
of Tarski:
In understandably obsessive detail he told of his position in the United States ... and his so-far
unsuccessful attempts to get his family out of Poland ... and he concluded by saying he will be
grateful for anything Bocheski can do even if it is only to convey the information in the letter
to his wife.14

Tarski [1942–1947] 1999. Letters to Kurt Gödel, 1942–47. Edited by Jan Tarski. In
Alfred Tarski and the Vienna Circle: Austro–Polish Connections in Logical Empiri-
cism, edited by Jan Woleski and Eckehart Köhler, 261–273. Dordrecht: Kluwer
Academic Publishers. According to the editor (page 261), these
“... sixteen letters, which include notes on postcards, form (apparently) the only surviving series
of Tarski’s correspondence from the difficult transitional period, when he was in the course
of arranging his life in his new country. ... The letters split naturally into two groups. Those
of the first group, consisting of ten letters from June 1942 to April 1944, were written in Ger-
man; this language was Tarski’s principal international language in his former years. ...
The six letters of the second group, March 1945–January 1947, were written in English. All
of the letters were handwritten, and they deal with a mixture of professional and personal
matters. ... However, the reader will note a definite shift of point of view, from the primarily
personal in the first group of letters to the primarily professional in the second.”

Tarski [1944] 1987. A philosophical letter of Alfred Tarski. Edited by Morton White.
Journal of Philosophy 84: 28–32. Item 87 l in Givant 1986, with incomplete publication
data. According to the 1988 review by William S. Hatcher, this letter to White15
“... discusses certain philosophical questions, mainly related to semantical and logical issues.
Adopting a generally common-sensible and practical-minded approach throughout, Tarski opts
unequivocally for an empirical-linguistic rather than a purely linguistic origin for logical truth.

14
Feferman and Feferman 2004, 136–137. For information about Bocheski, see a box in section 15.9.
15
M orton W hite was born in New York City in 1917. He was educated there and earned a doctorate in
philosophy at Columbia University in 1942. During 1940 he was assistant to Tarski at City College of
New York. (See Rosen 1985, annotated in chapter 17, for a student’s view of Tarski’s course.) White
achieved note in several areas of philosophy and its history, particularly its relation to cultural and social
matters. He served at Harvard during 1953–1970, and after that at the Institute of Advanced Study in
Princeton. (White 1999.)
16.3 Letters 397

In particular, Tarski sees no ‘difference in principle’ between the status of logical and non-
logical premisses of science or of a given scientific discipline.” 16

Tarski [1946–1947] 2014. Two letters to Wacaw Sierpiski, 30 October 1946 and 18
March 1947. In section 15.12 of the present book, Tarski 2014. Translations of
handwritten documents, APAN signature III-194, in the Archiwum Polskiej Akademii
Nauk (Archive of the Polish Academy of Sciences) in Warsaw. This is their first
publication. They are discussed at length in section 15.12.

Tarski [1947–1955] 2003. Four letters to Tadeusz Kotarbiski, 6 December 1947 and 16
October 1953 from Berkeley, and 7 September 1954 and 10 December 1955 from
Amsterdam. In Jadacki, op. cit., 20–24.
These are about daily life, philosophical issues, conferences, colleagues, and publication of
Tarski [1956] 1983. For information on Kotarbiski, see a box in section 1.1.

Tarski [1963] 2003. Letter to Tadeusz Kotarbiski, 7 June. Ibid., 24.

Tarski signed this letter for the Group in Logic and Methodology at the University of California,
Berkeley. In English, it asked Kotarbiski to send any available reprints of his work and to
add the group to his mailing list.

Tarski et al. [1971] 2003. Letter to Tadeusz Kotarbiski, 24 April. Ibid., 25. This
postcard conveys warm wishes and greetings from eleven participants in a congress
at a Polish research institute in New York. One of the authors was Feliks Gross.

Tarski 1972. O szkicu Feliksa Grossa w Wiadomociach. Wiadomoci 27(1347, 23 Jan-

uary): 6. In English, with Polish commentary by the journal editor.
Tarski had written this letter to Feliks Gross urging republication of Gross’s article “Trip to
the moon: Why?” That article had appeared in the same journal in 1969 just before men first
stepped on the moon. Tarski wrote,
You concern yourself there with the impact of fantasy and creative myth on the cultural
development of mankind ... . This was the time when much shallow and demagogical criticism
of the present efforts to conquer space could be heard. ... . This superficial polemic ... has caused
a great deal of harm by weakening the ... support of the human community for the monumental
efforts needed to realize the dream. ... Your article ... can serve as a potent antidote for narrow-
minded criticism.
Gross was a noted sociologist who had escaped Poland during World War II. The journal,
published in London, addressed the Polish emigré community. Gross was one of the authors
of Tarski et al. [1963] 2003.

16
This view is often associated with W. V. O. Quine. About his relationship to Tarski, see Mancosu 2005.
398 16 Posthumous Publications

Tarski [1972–1973] 1995. Two letters to Svätoslav Mathé, 21 July 1972 and14 February
1973. Organon F 2: 56–58. The first is in Slovak, translated from Polish by Pavel
Cmorej. The second is handwritten in Polish. The journal is published in Bratislava.
Mathé is a Slovak political scientist and journalist. At the time he was an editor for Slovak
Television in Bratislava. Preparing a television film about Tarski, he had sent a script to Tarski
for comment. In the first letter Tarski expressed delight, no objections, and related some
pleasant experiences in Slovakia before World War II and in Prague the previous summer.
Months later, Mathé sent Tarski the film and an article about him. In the second letter, Tarski
conveyed his appreciation and assured Mathé that he should not be troubled by some words
of a certain Dr. Hajek in the film.17

17
Jan Zygmunt has suggested that this was the Czech logician Petr Hajek.
 17
Biographical
Studies

This chapter lists some personal biographies and sketches of Alfred Tarski. It is a by-
product of the research that underlies earlier chapters. The editors did not search
exhaustively for other sketches of Tarski. The list includes all longer works that they
encountered, but shorter ones only if their authors appear to have possessed unique
information or distinction. Each listed item begins with the author-date designation used
in this book and continues with a description adapted from this book’s bibliography;1 it
is followed by a brief annotation.

•Addison, John W. 1983. Eloge: Alfred Tarski, 1901–1983. California Monthly 94 ( 2,

December): 28. Reprinted in Annals of the History of Computing 6 (1984): 335–336.
Translated by Irena Czubiska as “Jeden z najwikszych logików,” Przegld techniczny
20 (1985): 39. That title means “One of the greatest logicians.”
Addison was Tarski’s colleague for twenty years at the University of California at Berkeley.
The journal was the university’s news magazine for alumni. Addison suggested that Tarski’s
success as a teacher was due to Tarski’s
dogged insistence on precision and clarity in the way his students expressed their ideas.
In a seminar he would never be satisfied with an “almost” clear account accompanied by
a wave of the hand to indicate “you see what I mean.” “No,” he would rejoin, “you must
say what you mean.”
Addison discussed Tarski’s truth definition:
The seeming simplicity of his famous example that the sentence “Snow is white” is true just
in case snow is white belies the depth and complexity of the consequences ... .2
Finally, Addison suggested an epitaph for Tarski: He Sought Truth and Found It.

•Chuaqui Kettlun, Rolando. 1984. Alfred Tarski, matemático de la verdad. Revista

universitaria 11: 30–34. Translated into Portuguese by S. P. Monoide as “Alfred
Tarski, matemático da verdade,” Boletim da Sociedade Paranaense de Matematica 6
(1985): 1–10.

1
This book’s bibliography includes more details of publication.
2
See section 15.6. Addison’s 55-word sentence was soon quoted in wonder by San Francisco’s celebrated
humor columnist Herb Caen, who quipped, “A snow job if I ever read one” (San Francisco Chronicle, 15
January 1984).

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 399
DOI 10.1007/978-1-4939-1474-6_17, © Springer Science+Business Media New York 2014
400 17 Biographical Studies

After training as a surgeon in Chile, Chuaqui earned a doctorate from Berkeley in 1965 in
foundations of probability theory. He became a professor in Chile, and exerted major influence
on the growth of interest in logic in South America. The Revista is published by the Universi-
dad Católica de Chile. This elegant resumé of Tarski’s research career emphasized those
aspects readily graspable by nonspecialists. Chuaqui particularly mentioned Tarski’s 1974–
1975 visit to Chile and Brazil and Tarski’s continued interest in their scientific communities.
Chuaqui concluded (page 34): “that is what I, along with all who knew him, feel for him:
admiration, gratitude, and friendship.”

•Feferman, Anita B. [1997] 1999. How the unity of science saved Alfred Tarski. In
Alfred Tarski and the Vienna Circle: Austro–Polish Connections in Logical Empiricism,
edited by Jan Woleski and Eckehart Köhler, 43–52. Dordrecht: Kluwer Academic
Publishers.
•———. 1999. Alfred Tarski. American National Biography 21: 330–332.
This lively article and professional biographical sketch are by-products of the research that
supported the full biography described in the next item. Feferman used her article title for
chapter 5 of that book.

•Feferman, Anita B., and Solomon Feferman. 2004. Alfred Tarski: Life and Logic. Cam-
bridge, England: Cambridge University Press. Feferman and Feferman 2009, described
in the next item, is a Polish translation. The following annotation is constructed
largely from five reviews.
This is the only full-length biography of Alfred Tarski. Anita B. Feferman is a historical
researcher and writer, with a previous biography of the logician Jean Van Heijenoort. Her
husband Solomon was Tarski’s tenth doctoral student, from 1948 to 1957 at Berkeley. Since
then Solomon has been a professor of mathematics and philosophy at nearby Stanford Univer-
sity, and a noted researcher in foundations of mathematics and computer science. The
American mathematician Anil Nerode wrote,
No one could be better qualified than Anita to write his personal biography. No one could
be better qualified than Sol to write his scientific biography.3
The book’s dust cover displays remarkable psychedelic portraits of Alfred Tarski and his wife,
Maria, which were created in 1934 and 1938, respectively, by the Polish artist and writer
Stanisaw Ignacy Witkiewicz. Famous under the name Witkacy, he was their close friend.4
According to Roger D. Maddux, Tarski’s twenty-second doctoral student,
The story of Tarski’s life and loves is told in 15 chapters, while his work in philosophy, logic,
and mathematics is described in 6 interludes. All mathematical and logical symbolism is
confined to the interludes. This structure makes this book accessible to the widest possible
audience.5
The Moroccan mathematician and historian Hourya Sinaceur described the book as
an enthralling success story of a self-confident, enterprising, untiring, and entrepreneurial
scientist, and a rich and scrupulous account of [his] numerous achievements ... . It is partic-
ularly remarkable that the Fefermans were able to reconstruct so vividly Tarski’s childhood

3
Nerode 2010, 286–288, published in the American Mathematical Monthly.
4
The original portraits are in the possession of Tarski’s son, Jan, and daughter, Ewa Krystyna.
5
Maddux 2005, 536, 540, published in the Journal of Symbolic Logic. Maddux earned the PhD in 1978.
The Fefermans included a list of doctorates completed under Tarski’s supervision (pages 385–386).
17 Biographical Studies 401

in Warsaw; the brilliant gymnasium and university years ... in short, the scientific, political,
cultural, and artistic atmosphere ... The reader follows month by month or year by year the
irresistible, though treacherous, ascension of a strong and passionate character.6
Nerode, a long-standing acquaintance of Tarski, reported,
I am amazed that Anita was able to uncover his personal life in such detail. People were
more discreet in previous generations.3
Sinaceur described the Fefermans’ interludes as
an introduction to the main problems and results that Tarski pushed to the forefront of
logical research. Written with a minimum of technicalities, [they] meet the demanding
standards of the exceptional teacher that Tarski was and serve perfectly the purpose of
pedagogical presentation. Simple but self-contained, explaining step-by-step everything
that is needed and leaving nothing fuzzy nor obscure, they are easily understandable even
by those not previously acquainted with the subject matters. They furnish a beautiful,
though partial, survey of a whole century of logic in Europe and America.6

The present editors note that the Fefermans did not uncover much about the critical years
1918–1920 in the lives of Alfred and his wife, Maria, nor about Alfred’s work as a full-time gim-
nazjum teacher and teacher trainer. The present book partially fills those gaps.7 Reviewers
in general have found few faults with the Fefermans’ biography; none mentioned those. The
American historian Irving Anellis would have appreciated more information about certain
publication and priority disputes and about influence on Polish thinkers by the American
philosopher Charles Sanders Peirce.8 According to the Brazilian philosopher Walter Carnielli,
The book probably overstresses the role of anti-semitism as a universal explanation for
everything that happened to Tarski in Poland.9
Moreover, he regretted the book’s “somewhat excessive insistence on Tarski’s love affairs, on
what he smoked and drank.” 10
Sinaceur concluded her review,
The complex and varied portrait the Fefermans have painted is a rigorous attempt at
capturing with greatest objectivity the complex socio-psychological facts that can help us
understand Tarski’s personal leanings, his high professional conscientiousness, his “unend-
ing concern for clarity, precision, and rigor,” and, last but not least, his strong will to put
his mark on his time.11
Tarski’s former student Maddux finished,
Many times while reading this book I thought, “Yes! That’s what he was like!” 5

•———. 2009. Alfred Tarski: ycie i logika. Translated by Joanna Goliska-Pilarek

and Marian Srebrny. With a preface by Jan Woleski. Warsaw: Wydawnictwa Aka-
demickie i Profesjonalne. Polish translation of Feferman and Feferman 2004.

6
Sinaceur 2007, 986–988, published in the Notices of the American Mathematical Society.
7
Sections 1.1 and 9.1 of the present book describe the 1919–1920 Polish-Soviet War from the viewpoints
of Alfred and Maria, respectively; Part Three is mostly devoted to Alfred’s teaching and teacher training.
8
Anellis 2005, 121–122, published in the Review of Modern Logic.
9
Carnielli 2006, 95, published in the journal Logic and Logical Philosophy. Carnielli was echoing the opin-
ion expressed in Woleski 1995a, notes 13–15. See also Jan Woleski’s preface to the Polish translation
described in the next item.
10
Ibid., 92.
11
Sinaceur 2007, 989, quoting Feferman and Feferman 2004, 41.
402 17 Biographical Studies

Goliska-Pilarek is a logician at the University of Warsaw. Srebrny is a theoretical computer

scientist at the Polish Academy of Sciences. The expansive preface by the noted Polish philo-
sopher Jan Woleski is remarkable for these features:
• an account of some of the early steps in the United States and Poland in the
creation of the Fefermans’ 2004 biography of Tarski;
• a discussion from an alternative viewpoint of the effects of antisemitism on Tarski’s
life in Poland;
• a discussion of nationalism in academic culture.

•Feferman, Solomon. [1997] 1999. Tarski and Gödel: between the lines. In Woleski
and Köhler, op. cit., 53–63.
This conference report could be regarded as a by-product of the author’s research for the Tarski
biography described in the previous two items. But it has another source, as well: Solomon
Feferman was the editor of the 1986–2003 Collected Works of Kurt Gödel. Feferman’s paper
presents bird’s-eye views of several personal encounters of Tarski and Gödel. It contains a
transcription of Gödel’s January 1931 letter to Tarski, in which Gödel first mentioned his
incompleteness result. For each thread of Tarski’s research that intertwined with one of
Gödel’s, Feferman explained the interaction.12 He concluded by comparing and contrasting the
public and private philosophies that underlay Tarski’s and Gödel’s mathematical works. Each
mathematician’s public practice was apparently opposite to his private philosophy, and the
private philosophies of the two were opposite, as were their public practices.

•———. 2003. Alfred Tarski and a watershed meeting in logic: Cornell, 1957. In
Philosophy and Logic in Search of the Polish Tradition: Essays in Honour of Jan
Woleski on the Occasion of his 60th Birthday, edited by Jaakko Hintikka et al.,
151–162. Dordrecht: Kluwer Academic Publishers.
This report, too, was a by-product of Solomon Feferman’s work on the Tarski biography
described in previous items. It is an account of the five-week 1957 Summer Institute for
Symbolic Logic held at Cornell University in New York. Feferman emphasized the institutional
background for the session, the political process of organizing it, and the vibrant social inter-
actions among the eighty-five participants. About ten of those, including him, were Tarski’s
research collaborators or students. Twenty others came from the computer industry: this was
the first organized interaction between pure and applied logical researchers. Feferman stressed
Tarski’s role in organizing the session and ensuring the participation of his associates. Tarski
stayed for three weeks, gave three presentations and participated significantly in the discus-
sions. However, Feferman noted, Tarski seemed to take little interest in the papers about
practical applications. It is remarkable how many concepts and results, soon to become core
elements of mathematical logic or theoretical computer science, received early exposure in the
eighty papers presented at Cornell. Feferman concluded by listing several similar conferences
that occurred during 1957–1960.

•Givant, Steven R. 1991. A portrait of Alfred Tarski. Mathematical Intelligencer 13 (3):

16–32. Translated into Czech by Helena Nešet ilová as “Portrét Alfréda Tarského,”

12
For a corresponding discussion from Gödel’s viewpoint, see Krajewski, S. 2004.
17 Biographical Studies 403

Pokroky matematiky, fysiky a astronomie 37 (1992): 185–205. Givant and Huber-Dyson

1996, described in the next item, is an augmented Polish translation.
Givant’s doctoral research at Berkeley during 1967–1975 was significantly influenced though
not formally supervised by Tarski. After that, Givant became a professor of mathematics at
nearby Mills College. He maintained an intense collaboration and friendship with Tarski until
the latter’s death. According to the American philosopher John Corcoran,
Givant knew Tarski and Tarski’s later mathematical work perhaps better than anyone else;
he is well qualified to present to the mathematical community this brief portrait of Tarski.13
The article contains many good photographs of Tarski, his family, and his associates. It is
an emotional and powerful presentation of major aspects of Tarski’s life, from a markedly
personal viewpoint. While many of its details have been subsumed by the Fefermans’ full
biography of Tarski, described in previous items, the tone of Givant’s study remains striking.
For example, here is Givant’s concluding sentence:
I don’t know if anyone who interacted with [Tarski] on a deeper level can completely sort
out the mixture of admiration, exasperation, loyalty, affection, frustration, anger, and
gratitude that he evoked. I know I can’t.
Corcoran, who also knew Tarski well, suggested that limitations of Givant’s relationship
with Tarski prevented Givant from presenting Tarski’s philosophical positions effectively; nor
did Givant sufficiently emphasize Tarski’s
enthusiasm for elementary algebra, elementary geometry, elementary analysis and
elementary set theory ... [That] informed many of his most sophisticated mathematical
achievements and it was important, sometimes decisive, in his philosophy.
Corcoran concluded,
Appreciative and intimate but not uncritical, the portrait reveals various occasionally
unflattering features of Tarski’s habits and character. The author resisted any temptation
to be hagiographic; his well-written, richly informative, and tasteful treatment of one of
the world’s greatest mathematicians is a pleasure to read.

•Givant, Steven R., and Verena Huber-Dyson. 1996. Alfred Tarski w kalejdoskopie
impresji osobistych. Translated from the English by Adelina Morawiec. Wiadomoci
matematyczne 32 (1996), 95–127. The title means “Alfred Tarski in a kaleidoscope of
personal impressions.”
The Swiss mathematician Huber-Dyson met Tarski at the 1957 Cornell meeting described by
Solomon Feferman in a previous item. Later she enjoyed a long career in Canada as professor
of philosophy at the University of Calgary. For more about her relationship withTarski, see
Feferman and Feferman 2004, chapters 9 and 11. She wrote a new introduction, coda, and
twenty-two long footnotes for the article Givant 1991 described in the previous item. The whole
was translated by the Polish logician Morawiec and published in the annual journal of the
Polish Mathematical Society. Unfortunately, the translated article does not include any of the
illustrations from the original.

•Goliska-Pilarek, Joanna, Joanna Porbska-Srebrna, and Marian Srebrny. 2009a. Król

logiki. Rzeczpospolita (4–5 April): A20–A21. The title means “King of logic.” The jour-
nal is a daily newspaper with nationwide circulation; its name means The Republic.

13
Corcoran 1993, published in Mathematical Reviews.
404 17 Biographical Studies

Two of these authors are the translators of the Tarski biography Feferman and Feferman 2009
described in an earlier item. The third is an architect and city planner. This article was pub-
lished on the same day as another newspaper article, Woleski 2009, described in a later item.
(Woleski had written the preface for that translated biography.) Readers can see how Tarski
is regarded by the educated public in Poland, and glimpse from a Polish viewpoint the prewar
society of Warsaw described by the Fefermans.
This article also details how its authors corrected an error in the Tarski biography. Tarski’s
home for nearly thirty years—his parents’ apartment building, Koszykowa 51—was not
destroyed in World War II, as the Fefermans had claimed, but stands today as Koszykowa 51a
in an attractive little residential court in downtown Warsaw. The authors provide a history
of that neighborhood. Many notable figures of an astonishingly vibrant society trod there in
Tarski’s time.14

•Hi, Henryk. 1971. Jubileusz Alfreda Tarskiego. Kultura 288: 134–140. The title
means “Jubilee of Alfred Tarski.” Kultura was a literary and political magazine
published monthly in Paris for the worldwide Polish emigré community.
As a Warsaw undergraduate in 1937, Hi attended Tarski’s course on the methodology of the
deductive sciences.15 After the war he emigrated to the United States, earned a doctorate in
logic, and became a noted researcher in linguistics. This article was prompted by the interna-
tional symposium held in Berkeley during 23–30 June 1971 to honor Tarski on the occasion
of his seventieth birthday. Its proceedings, edited by Leon Henkin and others, were published
in 1974.16 Describing the conference for his Polish emigré readers, Hi quipped that viewing
the crowd of participants gave him the impression that he was actually visiting a Berkeley
campus of the University of Warsaw! Hi’s account overlaps those proceedings only to a small
extent, mainly in the philosophical aspects of Tarski’s work. Hi began with a brief reminis-
cence about Tarski’s 1937 lectures, then continued with a discussion of the notion of truth and
a general survey of Tarski’s publications of interest to philosophers. Hi concentrated on rela-
tionships among the views of Tarski and Polish and Viennese philosophers contemporary with
him. Finally, Hi indicated those works of Tarski best suited for introducing a wide audience
to philosophical logic.

•Hodges, Wilfrid. 1986. Alfred Tarski. Journal of Symbolic Logic 51: 866–868.
This is the first of a series of essays by distinguished logicians about Tarski and his work,
published in volumes 51 (1986) and 53 (1988) of the journal. The others are described in
section 18.1 of the present book. Hodges also coauthored the one on decidable theories. The
content of this introduction is subsumed by the Fefermans’ 2004 biography of Tarski, except
for a list of about a dozen researchers who collaborated with Tarski on projects at Berkeley,
and for an invitation to contribute to a fund in honor of Tarski that would be used to support
further such projects. This article is also the source of the Fefermans’ list (2004, 385–386) of
Tarski’s doctoral students.

14
This article includes a contemporary photograph of Koszykowa 51a. In section 9.1 of the present book
is a 1946 photograph of the building, only slightly altered from its form in Tarski’s time. The same three
authors constructed in 2009 a presentation, Szlakiem Alfreda Tarskiego po Warszawie (Along the Route
of Alfred Tarski in Warsaw) with striking illustrations of Tarski’s residences and neighborhoods. Unfor-
tunately, it has not been published.
15
He is quoted in section 9.3, page 191.
16
This symposium was also reported by the article Melnick 1971, described in a later item.
17 Biographical Studies 405

•Jadacki, Jacek Juliusz. 2003a. Alfred Tarski à Varsovie. In Alfred Tarski: dedukcja
i semantyka (déduction et sémantique), edited by Jacek Juliusz Jadacki, 139–180.
Warsaw: Wydawnictwo Naukowe Semper. Translated by Wanda Jadacka from the
article “Alfred Tarski w Warszawie,” ibid., 112–137.
The author is a professor of philosophy at the University of Warsaw, specializing in logical
semiotics and the history of Polish philosophy. The article is in the proceedings of a 15 January
2001 symposium in Warsaw in honor of the centenary of Tarski’s birth. It is a wonderfully
detailed and richly documented chronology of Alfred Tarski’s life in Warsaw, plus snippets from
his American years. Without its guidance, assembling the background material for the present
book would have been impossible. The present editors owe a great debt to Jadacki.
The symposium proceedings also include a large number of fascinating illustrations and
transciptions or translations of letters and small works of Tarski. Many of those are cited
elsewhere in the present book.
Jadacki’s article is a lightly edited transcription of historical research notes, in day-by-day
sequence, with a multitude of references. He graciously acknowledged the assistance of Alfred
Tarski’s son, Jan, and daughter, Ewa Krystyna, a number of former students of Tarski’s
university and gimnazjum classes, and a number of relatives of Tarski’s wife, Maria. With their
help he also dug deeply into university and government archives and genealogical records.
Jadacki is to be congratulated and thanked for publishing his data in this form, for others to
use to add to the stories of Alfred and Maria. Jadacki indicated where his account differs from
others, and where he filled gaps. Particularly notable are a report of Maria’s involvement in
the 1919–1920 Polish–Soviet War, an alternative interpretation of Alfred’s conversion to
Catholicism, records of Alfred’s university teaching, and information that led the present
editors to a greater understanding of Alfred’s gimnazjum teaching.

•James, Ioan Mackenzie. 2009b. Napoleon of logic: Alfred Tarski (1901–1983). In

Driven to Innovate: A Century of Jewish Mathematicians and Physicists, 284–
288. Witney, Great Britain: Peter Lang.
Apparently derived from Feferman and Feferman 2004, this essay concentrates on Tarski’s
experiences as a Jew. It is virtually undocumented, and contains numerous small errors. The
author is a noted Oxford topologist. His historical introduction to the containing publication
is informative. But readers should also consider the highly critical review Siegmund-
Schultze 2010.

•Kozowski, Witold. 2003. Wspomnienie (14 1 1901 – 27 X 1983) Alfred Tarski. Gazeta
wyborcza 264 (November 13): 11. The first word in the title means “A memory of.”
The journal is a daily newspaper with nationwide circulation; its name means Electoral
Gazette.
An award-winning journalist, Kozowski was Tarski’s gimnazjum student during 1934–1938.
Section 14.1 contains a discussion of Tarski as a teacher based on this article and a 2010
personal interview with Kozowski. That section includes a brief biographical sketch of him.

•Melnick, Norman. 1971. The ‘Einstein of Mathematics’ at UC. San Francisco Chronicle
(27 June).
This half-page summary of Tarski’s career publicized the symposium then underway in
Berkeley to honor Tarski’s seventieth birthday. Melnick was a science writer for the Chronicle,
406 17 Biographical Studies

the leading daily newspaper in the Bay Area, where Berkeley is located. He had evidently
interviewed the Hungarian mathematician Paul Erdo  s, who described a dinner he had enjoyed
in Princeton in 1942 with Tarski and Albert Einstein. An Einstein headline was certain to
attract attention.17

•Pawlikowska-Broek, Zofia. 2003. Tarski (Tajtelbaum) Alfred (1901–1983): logik,

filozof, matematyk. In Sownik biograficzny matematyków polskich, edited by
Stanisaw Domoradzki, Zofia Pawlikowska-Broek, and Danuta Wglowska, 244–245.
Tarnobrzeg, Poland: Pastwowa Wysza Szkola Zawodowa.
This is a brief account of Tarski’s research career. The author is a historian of mathematics,
then on the staff of the State Advanced School of Professional Education, which published this
Biographical Dictionary of Polish Mathematicians.

•Polska Akademia Umiejtnoci. 1948. Alfred Tarski. In “Skad Polskiej Akademii

Umiejtnoci (stan w marcu 1948),” Rocznik Polskiej Akademii Umiejtnoci 1946–
1947: xli–xlii.
In 1946, Tarski was selected for the Polish Academy of Arts and Sciences in Cracow. This
biographical sketch is included in its March 1948 membership roster. It appears to be the first
biographical sketch of Tarski published in Poland; its details seem correct except for Tarski’s
year of birth. In 1949 Tarski’s membership was terminated because of inactivity.18

•Rosen, Saul. 1985. Alfred Tarski in 1940. Annals of the History of Computing 7:
364–365.
Rosen described one of the first classes that Tarski taught in the United States, at City College
of New York. Led to expect an introduction to the philosophy of mathematics, students instead
experienced Tarski’s carefully prepared English exposition of his current research in the theory
of relations. Conversation with most students in the class was effective only through intermedi-
aries who could translate between English and German. Rosen earned a bachelor’s degree
there in 1941. After service in the Army Signal Corps, he completed a doctorate and became
a pioneer in computer science, working particularly at Purdue University in Indiana.

•San Francisco Chronicle. 1983 (28 October). Alfred Tarski: Expert logician,
UC Professor.
•San Francisco Examiner. 1983 (28 October). Alfred Tarski.
These two sober and factual obituaries appeared in the Bay Area’s leading daily newspapers.19

•Sinaceur, Hourya. 2000. Introduction. In “Address at the Princeton University Bicenten-

nial Conference on Problems of Mathematics (December 17–19, 1946),” by Alfred Tarski, edited
by Hourya Sinaceur, Bulletin of Symbolic Logic 6: 1–20.
This introduction by a noted Moroccan mathematician and historian provides much informa-
tion about the effect on Tarski’s research career of his transition to the New World.

17
This symposium was also reported by the article Hi 1971, described in an earlier item.
18
Jadacki 2003a, 160.
19
In contrast, see footnote 2.
17 Biographical Studies 407

•———. 2009. Tarski’s practice and philosophy: between formalism and pragmatism.
In Logicism, Intuitionism, and Formalism: What Has Become of Them? edited by Sten
Lindström et al., 357–396. New York: Springer.
According to the American philosopher John Corcoran,20 who knew Tarski well, this paper is
a rather rambling ... report on, among other things, the author’s efforts to glean insights
into Tarski’s philosophy by combing his mathematical and logical works, his correspon-
dence, his spontaneous remarks at public conferences and meetings, and even things
Tarski read.
Corcoran explained that in this paper, pragmatism
refers vaguely to practice-oriented opportunism (see page 390) and not to the American
philosophical movement initiated by [Charles Sanders] Peirce and [William] James ... the
conclusion is that Tarski’s basically mathematical agenda required strong Platonist prem-
ises temporarily and pragmatically assumed for the sake of argument as long as they con-
tributed to the development of his program. This conclusion rings true.
Corcoran reported that Tarski once told him directly that “the closest thing to a statement of
his philosophy” might be the article Kotarbiski [1935] 1956 on pansomatism, which Tarski
helped translate. Corcoran criticized Sinaceur for not considering it. He also complained about
the inadequate editing of her paper.

•Suppes, Patrick, Jon Barwise, and Solomon Feferman. 1989. Commemorative meeting
for Alfred Tarski: Stanford University, November 7, 1983. In A Century of Mathemat-
ics in America, edited by Peter Duren et al., volume 3, 393–403. Providence: American
Mathematical Society.
This article is the proceedings of a meeting that consisted of talks by the three authors, all
Stanford professors of philosophy. Suppes described his experience as a professor sitting in
on Tarski’s Berkeley research seminars in the 1950s; only a summary of his talk is included.
Feferman described his own experience as a student in those seminars. Both of those accounts
of Tarski emphasize his personal characteristics and scientific and teaching style, from the
authors’ specific viewpoints. Feferman also discussed with some speculation Tarski’s private
philosophy. Barwise, who earned the doctorate in 1967 under Feferman’s supervision, spoke
about Tarski’s contributions to semantics and the theory of models.

•Tarski, Jan. 1994a. A short biographical sketch of Alfred Tarski. In Introduction to

Logic and to the Methodology of the Deductive Sciences, fourth edition, edited by Jan
Tarski, xviii-xxii. New York: Oxford University Press.
This edition of Alfred Tarski’s classic text is described in section 16.2. Alfred’s son, Jan, edited
it and wrote this very personal account. A Polish translation appears in the monograph Tarski
1996, which is also described in 16.2.

•———. 1996b. Philosophy in the creativity of Alfred Tarski. In Truth after Tarski,
edited by Micha Hempoliski, Dialogue and Universalism 6 (1–2), 157–159.
Published by the Polish Academy of Sciences, this is the journal of the International Society
for Universal Dialogue. The issue surveyed developments in the theory of truth since Tarski’s
1933 work (see section 15.6). The author, Alfred Tarski’s son, Jan, speculated about the

20
Corcoran 2011.
408 17 Biographical Studies

reasons underlying Alfred’s transition away from conceptual and philosophical investigations,
which he had emphasized during the 1930s, toward his more strictly mathematical and
technical research of the 1940s. This coincided with Alfred’s moving to the New World but,
Jan claimed, may have been only partly caused by that displacement.

•Wells, Benjamin. 2007. Alfred Tarski, friend and daemon. Notices of the American
Mathematical Society 54: 982–984.
Tarski’s last doctoral student, from 1963 to 1982, Wells became professor of mathematics and
computer science at the University of San Francisco. He explained his use of daemon: “a
leonine externalized conscience.” His intensely personal reminiscence describes the daemon
as mathematical, political, cultural, philosophical, moral, and spiritual.21

•Wojciechowska, Agnieszka. 2001, 2011. Alfred Tarski (1901–1983). Matematyka:

Czasopismo dla Nauczycieli (2001) 54 (3): 132–135; (2011) 64 (9): 3-9. In Polish.
These two articles are different. The journal’s title means Mathematics: Journal for Teachers.
The journal Parametr, to which Tarski had contributed in 1931–1932, ceased publication with
the World War II: see sections 9.7–9.8. The exciting journal Matematyka was founded in 1948
to fill that role. Dr. Wojciechowska is its current editor. The cover of its 2011 issue 9 carries
a fine color portrait of Tarski.

•Wójcik, Andrzej. 1985. Alfred Tarski: ycie i dzieo. Pismo literacko-artystyczne 4 (1):
174–178. The title phrase means “Life and work.” Pismo means Writings.
This account of Tarski’s scientific career was written when the author was a doctoral student
at the University of Silesia in Katowice. Later, he became a professor of philosophy there.

•Woleski, Jan. 1984. Alfred Tarski (1901–1983). Studia filozoficzne (2): 3–8. In Polish.
The author of this professional biographical sketch is a noted philosopher at the University
of Cracow. The journal was published monthly by the Polish Academy of Sciences.

•———. 1995b. On Tarski’s background. In From Dedekind to Gödel: Essays on the

Development of the Foundations of Mathematics, Synthese library 251, edited by Jaakko
Hintikka, 331–341. Dordrecht: Kluwer Academic Publishers Group.
This paper is partly about Tarski’s life and partly about his philosophical research. Woleski
aimed
to throw light on [the] cognitive conflict or dissonance of Tarski between his nominalistic
and empiricistic sympathies and his “Platonic” mathematical practice as well as why he was
so parsimonious in expressing his philosophical views.

•———. 2001a. Alfred Tarski–ksi logików: wzór na prawd. Polityka 2281 (20 Janu-
ary): 76–77. The title phrases mean “Prince of logicians: pattern for truth.” The
journal is a weekly newsmagazine.

21
Readers may want to explore the use of daemon in mythology, religion, and software engineering.
17 Biographical Studies 409

•———. 2009. Alfred Tarski: Logik z fantazj. Gazeta wyborcza (4–5 April): 24–25.
The title phrase means “Logician with flair.”
This essay appeared in one Polish national daily newspaper on the same day that Goliska-
Pilarek et al. 2009a appeared in another. The occasion was publication of the 2006 Tarski
biography by the Fefermans. Goliska-Pilarek and her colleague Marian Srebrny had trans-
lated the biography into Polish; their article and that book were discussed in earlier items in
this section. Woleski had written the book’s new preface. Under the title of this article,
alongside a famous dark portrait of Tarski as a thoughtful sage, Woleski quoted in bold letters,
Once someone asked him, “How do I become a great logician like you?” Tarski replied,
“That’s simple. You must be either a Jew or a Pole, and preferably both.”
From his section headings, readers may see what Woleski aimed to convey to the public:
Logik i sybaryta (Logician and Sybarite)
yd i Polak (Jew and Pole)
Geniusz bez profesury (Genius without Professorship)
Pedagog i bohater (Teacher and Hero)

•Woytak, Lidia. 1987. Alfred Tarski—matematyk o jzyku. Przegld polonijny 1987 (4):
67–75. The title phrase means “mathematician of language.” The journal title means
Polonia Review.
The author, a writer and editor at the Defense Language Institute in Monterey, California, has a doctor-
ate from the University of Pozna. The journal is published quarterly by the Polish Academy of Sciences;
its name refers to the Polish emigré community. Tarski’s thorough critique of an early draft of this
article is in the Tarski archive in the Bancroft Library in Berkeley: carton 13, folder 6.

•Zygmunt, Jan. 1995. Szkic biograficzny. In Pisma logiczno-filozoficzne, by Alfred

Tarski, edited by Jan Zygmunt, volume 1, vii–xx. Warsaw: Wydawnictwo Naukowe
PWN. The first words of the titles of the article and the containing publication mean
“sketch” and Writings.
•———. 2000. Alfred Tarski. Edukacja filozoficzna 29: 274–307. In Polish. Reprinted,
slightly updated, in 2001 in Polska filozofia powojenna, edited by Witold Mackiewicz,
volume 1, 342– 375. Warsaw: Agencja Wydawnicza Widmark.
The author is a noted Polish logician and historian of logic at the University of Wrocaw. The
first book, Pisma logiczno-filozoficzny, is described in section 16.2. The second, Polska filozofia
powojenna (Postwar Polish Philosophy), is a two-volume set of similar articles about other
Polish logicians and philosophers. According to the author’s summary of the five parts of the
latter article,
• I lists the main body of Tarski’s published works ...
• II gives a selection of papers on Tarski’s life and work ...
• III is a short [scientific] biography of Alfred Tarski ...
• IV surveys five areas of Tarski’s work ...
• V traces Tarski’s influence ... .
Part III is a dense stream of facts. The author’s 1995 essay is smoother, more polished.
 18
Research
Surveys

This final chapter lists some general surveys of Alfred Tarski’s research and legacy. It
is a by-product of the present editors’investigations that underlie earlier chapters. They
did not search exhaustively for other such surveys. Those that they encountered are listed
here if they cover substantial areas of Tarski’s work, not just single results. Each listed
item begins with the author-date designation used in this book and continues with a
description adapted from this book’s bibliography;1 it is followed by a brief annotation.

Of Tarski’s twenty-four doctoral students, eleven are authors of works about Tarski
mentioned in this or the previous chapter. Seven more of these authors were Tarski’s
“grand-students.” Those numbers attest to the persistence, respect, and loyalty that
Tarski inspired among his students and colleagues.

18.1 1986,1988 JSL Surveys

Major surveys of Alfred Tarski’s work in several areas of mathematics and logic were
published in volumes 51 (1986) and 53 (1988) of the Journal of Symbolic Logic (JSL).

•Blok, Willem J., and Don Pigozzi. 1988. Alfred Tarski’s work on general metamath-
ematics. 53: 36–50.
Blok earned the doctorate in 1976 at the University of Amsterdam for research in general
algebra. Pigozzi was Tarski’s nineteenth doctoral student, with a 1970 dissertation on cylindric
algebras.

•Doner, John, and Wilfrid A. Hodges. 1988. Alfred Tarski and decidable theories. 53:
20–35.
Tarski’s eighteenth doctoral student, Doner earned his degree in 1968 and collaborated with
Tarski on the theory of ordinal algebras. Hodges earned the doctorate in 1970 from Oxford
University, in model theory.

1
The entries in the present book’s bibliography include more publication details.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 411
DOI 10.1007/978-1-4939-1474-6_18, © Springer Science+Business Media New York 2014
412 18 Research Surveys

•Etchemendy, John. 1988. Tarski on truth and logical consequence. 53: 51–79.
Etchemendy earned the doctorate in1982 from Stanford University, working in model theory
and the theory of truth.

•Jónsson, Bjarni. 1986. The contributions of Alfred Tarski to general algebra. 51:
883–889.
Tarski’s second doctoral student, Jónsson earned his degree in 1946. He collaborated with
Tarski on the theory of cardinal algebras and on general algebra, and spent much of his career
at Vanderbilt University in Nashville,Tennessee.

•Lévy, Azriel. 1988. Alfred Tarski’s work in set theory. 53: 2–6.
Lévy earned the doctorate in 1958 from the Hebrew University of Jerusalem, in metamath-
ematics of set theory, and spent his career there.

•McNulty, George F. 1986. Alfred Tarski and undecidable theories. 51: 890–898.
McNulty was Tarski’s twentieth doctoral student, with a 1972 dissertation on general algebra.

•Monk, J. Donald. 1986. The contributions of Alfred Tarski to algebraic logic. 51:
899–906.
Monk was Tarski’s thirteenth doctoral student, with a 1961 dissertation on cylindrical algebras.
He continued collaborating with Tarski in that area. Monk is a professor at the University of
Colorado at Boulder.

• Suppes, Patrick. 1988. Philosophical implications of Tarski’s work. 53: 80–91.

Suppes earned the doctorate in 1950 from Columbia University in New York City, in the
philosophy of physics. He became noted in several areas of philosophy and mathematics, and
was the director of the Institute for Mathematical Studies in the Social Sciences at Stanford
University during 1959–1992.

• Szczerba, Lesaw W. 1986. Tarski and geometry,. 51: 907–912.

Szczerba collaborated with Tarski on foundations of geometry. He had studied with Wanda
Szmielew, who had been Tarski’s fifth doctoral student. Szczerba had a distinguished career
in Poland as a researcher and teacher in logic and as a university administrator.

•Van den Dries, Lou. 1988. Alfred Tarski’s elimination theory for real closed fields. 53:
7–19.
Van den Dries earned the doctorate in 1978 at the University of Utrecht, with a dissertation
on model theory and field theory.

•Vaught, Robert L. 1986–1987. Alfred Tarski’s work in model theory. 51: 869–882;
52(4): vii.
Vaught was Tarski’s eighth doctoral student, with a 1954 dissertation on model theory. From
1958 on he was Tarski’s colleague and collaborator at Berkeley.
18.2 Other Surveys 413

18.2 Other Surveys

This chapter concludes with a list of other surveys whose scope is comparable to those
listed in the previous section.

•Adamowicz, Zofia, et al., editors. 2004. Provinces of Logic Determined: Essays in the
memory of Alfred Tarski. Annals of Pure and Applied Logic, volumes 126 and 127.
Adamowicz earned the doctorate in 1975 under supervision of Andrzej Mostowski, Tarski’s first
doctoral student. She is now a researcher at the Polish Academy of Sciences. These volumes
consist of papers presented at the Alfred Tarski Centennial Conference held in Warsaw, 27
May–1 June 2001. Eighteen papers are on topics relevant to this section; most of them are
too specialized to be featured individually. However, taken as a group they constitute a major
survey of issues surrounding some aspects of Tarski’s research and its impact. They are listed
below.
Vol. 126
Solomon Feferman Tarski’s conception of logic 5–13
Jens Erik Fenstad Tarski, truth, and natural languages 15–26
Mario Gómez-Torrente The indefinability of truth in the Wahrheitsbegriff 27–37
Henryk Hi Reexamination of Tarski’s semantics 39–48
Ilkka Niiniluoto Tarski’s definition and truth-makers 57–76
John W. Addison Tarski’s theory of definability: common themes in 77–92
descriptive set theory, recursive function theory,
classical pure logic, and finite-universe logic
Wilfrid Hodges What languages have Tarski truth definitions? 93–113
Roman Kossak Undefinability of truth and nonstandard models 115–123
Jan Mycielski On the tension between Tarski’s nominalism and 215–224
his model theory (definitions for a mathematical
theory of knowledge)
Benjamin Wells Applying, extending, and specializing pseudo- 225–254
recursiveness
Pietro Benvenuti and On Tarski’s contribution to the additive measure 281–286
Radko Mesiar theory and its consequences
Andrzej Grzegorczyk Decidability without mathematics 309–312

Vol. 127
Ivo Düntsch and Boolean algebras arising from information systems 77–98
Ewa Orlowska
Roger D. Maddux Finite, integral, and finite-dimensional relation 117–130
algebras: a brief history
George F. McNulty Minimum bases for equational theories of groups 131–153
and rings: the work of Alfred Tarski and Thomas
Green.
Leo Esakia Intuitionistic logic and modality via topology 155–170
Urszula Wybraniec- Foundations for the formalization of metamathe- 243–266
Skardowska matics and axiomatizations of consequence theories
Arianna Betti Leniewski’s earlier liar, Tarski, and natural 267–287
language
Stanisaw Krajewski Gödel on Tarski 303–323
414 18 Research Surveys

•Chang, Chen-Chung. [1971] 1974. Model theory 1945–1971. In Proceedings of the

Tarski Symposium: An International Symposium Held to Honor Alfred Tarski on the
Occasion of His Seventieth Birthday, edited by Leon Henkin et al., 173–186. Provi-
dence: American Mathematical Society.
Chang was Tarski’s ninth doctoral student, with a 1955 dissertation on model theory. They
continued to collaborate. Chang became professor at the University of California, Los Angeles.
The major features of this survey are an extensive diagrammatic view of the influences
among different aspects of model theory in the last 20 years (grouped under the four
broad headings of first-order logic, algebra and syntax, compactness theorems, and
Löwenheim–Skolem–Tarski theorems) and some comments on the possible directions of
future research. There is an extensive bibliography of 119 items.2
See Vaught [1971] 1974, described in a later item, for an account of Tarski’s earlier work on
model theory.

•Czelakowski, Janusz, and Grzegorz Malinowski. 1985. Key notions of Tarski’s method-
ology of deductive systems. Studia Logica 44: 321–351.
The authors are professors of mathematics and logic in Poland. According to the American
mathematician John W. Dawson, Jr.,
The aim of this article, as the authors state in their abstract, is “to outline the historical
background and the present state of the methodology of deductive systems invented by
Alfred Tarski.” As such, it provides a survey that is at once broad but (necessarily, in view
of its brevity) shallow. Basic notions are defined, major directions of research indicated, and
subsequent developments traced through citations of important results.
Dawson criticized the authors’ unsubstantiated claims that various later developments were
“implicit” in Tarski’s work.3

•Feferman, Solomon. 2006. Tarski’s influence on computer science. Logical Methods

in Computer Science 2(3): 1-1–1-13.
Feferman was Tarski’s tenth doctoral student, with a 1957 dissertation on model theory and
proof theory. Since then he has been a professor at Stanford University and a noted researcher
in foundations of mathematics and computer science. From his first page:
Here we survey 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.

•Givant, Steven R. [1988] 1991. Tarski’s development of logic and mathematics based
on the calculus of relations. In Algebraic Logic: Colloquia Mathematica Societatis
János Bolyai, 54 (Budapest, 8–14 August 1988), edited by Hajnal Andréka et al., 189–
215. Amsterdam: North–Holland Publishing Company.
By including the operation of relation composition, the calculus of binary relations transcends
the power of Boolean algebra and incorporates some of that of first-order logic. The author

2
Mathematical Reviews 1979.
3
Dawson 1987.
18.2 Other Surveys 415

presents a sketch of the history of works dedicated, by Tarski or at his instigation, to the
delineation of this part, as well as the results obtained (the strongest is that in this fragment
of first-order logic it is still possible to formalize set theory and arithmetic, and, ultimately,
all that can be formalized in mathematics by means of the latter) ... .4

•———. 1999. Unifying threads in Alfred Tarski’s work. Mathematical Intelligencer

21 (1): 47–58.
The chapter 17 annotation of Givant 1991, a personal biographical sketch of Tarski, describes
the close relationship of this author and his subject. The present article is a deeper study of
Tarski’s vast research corpus, an attempt “to find underlying unity in Tarski’s work, to trace
steps that may have led to some of his discoveries.” Givant tackled the following questions:
How did a logician end up working in so many different areas? Were there interconnections
in his work that led him from one field to another? What drew him to set theory in the first
place, and what drew him away from it a few years later? What sparked his interest in
algebra and geometry? How did he become involved in the problem of defining truth? Why
did he work so intensively in algebraic logic ... and what did this work have to do with his
other research? Just why did he go into logic, anyway?

•Gómez-Torrente, Mario. 2006. Alfred Tarski. Stanford Encyclopedia of Philosophy.

On the Internet at http://plato.stanford.edu/entries/tarski/.
The author earned the doctorate at Princeton University in 1996; he now serves at the Uni-
versidad Nacional Autónoma de México. This article surveys only Tarski’s research in logic.

•Jadacki, Jacek Juliusz, editor. 2003b. Alfred Tarski: dedukcja i semantyka (déduction
et sémantique). Warsaw: Wydawictwo Naukowe Semper.
This book consists of materials for a January 2001 symposium sponsored by the University of
Warsaw, the Polish Philosophical Society, and the Warsaw Society of Sciences and Letters to
commemorate the centennial of Tarski’s birth. The author is professor of philosophy at the
university. The book contains many historical materials on Tarski: photographs, transcrip-
tions, Polish translations of short Tarski papers, and the indispensible chronological record
“Alfred Tarski in Warsaw.” Many of these are cited or described in preceding chapters. The
materials also include nine papers presented at the symposium, all related to Tarski’s research:
Subject Authors
Tarski’s foundation of the theory of deduction . . Urszula Wybraniec-Skardowska5
Tarski’s definition of true sentence . . . . . . . . . . . . Adam Nowaczyk
Joanna Odrow-Sypniewska
Jan Woleski
Tarski’s definition of logical constant . . . . . . . . . . Janusz Maciaszek
Other aspects of Tarski’s semantics . . . . . . . . . . . Adam Grobler
Artur Rojszczak
Marcin Selinger
Tarski and the definition of computability . . . . . . Andrzej Grzegorczyk

All these materials are in Polish, except for “Alfred Tarski in Warsaw,” which is provided in
both Polish and French.

4
Guillaume 1993.
5
This paper is similar to her English paper mentioned in the annotation of Adamowicz 2004 earlier in this
section.
416 18 Research Surveys

•os, Jerzy. 1986. O Alfredzie Tarskim. Ruch filozoficzny 43: 3–10.

The author, a noted Polish logician, was not a direct academic descendent of Tarski; neverthe-
less, he stressed (page 4): “I feel like his disciple.” The journal is published quarterly in Toru
by the Polish Philosophical Society. In this biographical sketch, o emphasized Tarski’s
research on the borderline of mathematics and philosophy.

•Mancosu, Paolo. 2009. Tarski’s engagement with philosophy. In The Golden Age of
Polish Philosophy: Kazimierz Twardowski’s Philosophical Legacy, edited by Sandra
Lapointe et al., chapter 9, 131–153. Dordrecht: Springer.
The author earned the doctorate in 1989 at Stanford University, supervised by Tarski’s former
student Solomon Feferman. Mancosu is now professor of philosophy at Berkeley. In this study
he started from Jan Woleski’s significant analysis of Tarski’s work in philosophy (Woleski
1993, discussed later in this section, and 1995b, discussed in chapter 17). Mancosu noted that
Woleski’s discussion was based almost entirely on published works of Tarski. In this article,
Mancosu reported his own further pursuit of these matters using various archival materials.
He indicated that further investigations of that nature might prove productive.

•Moore, Gregory H. 1990. Alfred Tarski. In Dictionary of Scientific Biography 1990,

volume 18: supplement II, 893–896.
The author is a noted historian of logic and mathematics at McMaster University in Hamilton,
Ontario. The article’s synopses of the impact of Tarski’s research are very effective: for
example,
He brought clarity and precision to the semantics of mathematical logic, and in so doing
he legitimized semantic concepts, such as truth and definability, that had been stigmatized
by the logical paradoxes.
Readers should be aware that there are a number of errors in historical detail.

•Mostowski, Andrzej. 1967. Alfred Tarski. In The Encyclopedia of Philosophy, edited

by Paul Edwards, volume 8, 77–81. New York: Macmillan Publishing Company.
Mostowski was Tarski’s first doctoral student, during 1931–1938, in Warsaw. After World War
II he continued collaboration with Tarski and led the redevelopment of logic research in Poland.
This article surveys only Tarski’s research in logic.

•Murawski, Roman. 2011. Alfred Tarski. In Filozofia matematyki i logiki w Polsce mi-
dzywojennej, section 3.6, 134–155. Toru: Wydawnictwo Naukowe Uniwersitetu
Mikoaja Kopernika. The book’s title means Philosophy of Mathematics and Logic in
Interwar Poland.
The author is a noted logician at the University of Pozna. The article is one of many biograph-
ical sketches in his comprehensive book.

•Patterson, Douglas, editor. 2008. New Essays on Tarski and Philosophy. Oxford, U.K.:
Oxford University Press.
Most of the essays in this high-quality collection are too specialized to merit individual inclu-
sion in this section. However, taken as a group they consistitute a major survey of issues
18.2 Other Surveys 417

surrounding Tarski’s philosophical work and its impact. Several of them have served as
resources for earlier sections of the present book. Here is the table of contents:
Roman Murawski Tarski and His Polish Predecessors on Truth 21–43
and Jan Woleski
Arianna Betti Polish Axiomatics and Its Truth: On Tarski’s 44–71
Leniewskian Background and the Ajdukiewicz
Connection.
Solomon Feferman Tarski’s Conceptual Analysis of Semantical Notions 72–93
Wilfrid Hodges Tarski’s Theory of Definition 94–132
Marian David Tarski’s Convention T and the Concept of Truth 133–156
Douglas Patterson Tarski’s Conception of Meaning 157–191
Paolo Mancosu Tarski, Neurath, and Kokoszyska on the Semantic 192–224
Conception of Truth
Greg Frost-Arnold Tarski’s Nominalism 225–246
Panu Raatikainen Truth, Meaning, and Translation 247–262
John Etchemendy Reflections on Consequence 263–299
Gila Sher Tarski’s Thesis 300–339
Mario Gómez-Torrente Are There Model-Theoretic Logical Truths That 340–368
Are Not Logically True?
Peter Simons Truth on a Tight Budget: Tarski and Nominalism 369–389
Jody Azzouni Alternative Logics and the Role of Truth in the 390–429
Interpretation of Languages

•Patterson, Douglas. 2012. Alfred Tarski: Philosophy of Language and Logic. New York:
Palgrave Macmillan.
According to the book’s cover, Tarski conceived of his logical work in the late 1920s as
a contribution to a view that he called “Intuitionistic Formalism.” The book explains this
view in terms of the views of Tarski’s teachers, in particular as found in the work of
Stanisaw Leniewski and Tadeusz Kotarbiski. These figures conceived of meaning not
in terms of the semantic relation of words to the world, but in terms of the expression of
thoughts. Questions about Tarski’s work are then addressed by applying this reading of
Tarski. Throughout the book, close attention is paid both to the development of Tarski’s
work through [the mid-1930s] and to the details of the expression of his ideas ... .

•Pla i Carrera, Josep. 1984. Alfred Tarski i la lògica contemporània. Primera part.
Butlletí de la Societat Catalana de Matemàtiques 17: 26–46. In Catalan.
•———. 1989. Alfred Tarski i la teoria de conjunts. Theoria (Spain) (series 2) 4: 343–417.
The title means “Alfred Tarski and set theory.” In Catalan.
The author, professor of logic, history, and philosophy of science at the University of Barcelona,
published several articles about Tarski in Catalonian and Spanish journals. The first article
discusses Tarski’s work on propositional and first-order logic, and on consequence operations
and cylindric algebras. There appears to be no second part of that article. The second article
emphasizes Tarski’s work on the axiom of choice and related topics.
418 18 Research Surveys

•Rodríguez-Consuegra, Francisco. 2005. Tarski’s intuitive notion of set. In Essays on

the Foundations of Mathematics and Logic, edited by Giandomenico Sica, 227–266.
Monza, Italy: Polimeric International Scientific Publisher.
The author, professor of philosophy at the University of Valencia, has published major studies
of Bertrand Russell’s mathematical philosophy. He noted (pages 227–228) that
Tarski made important contributions to set theory ... . Also ... set theory was the main
instrument used by Tarski in his most significant contributions which had philosophical
implications and presuppositions. ... [For] Tarski set theory was reliable as a working
instrument, then presumably as a conceptual ground. ... Like Russell and [Kurt] Gödel,
Tarski was much more sincere about his philosophical tendencies when no publication was
involved ... . Thus, in this case the archival work has been once again the only effective way
to try to understand Tarski’s actual ideas. The resulting picture is a fascinating struggle
between his nominalistic tendencies and his professional need to behave as if the mathe-
matical entities with which he was working actually existed.

•Scanlan, Michael. 2003. American postulate theorists and Alfred Tarski. History and
Philosophy of Logic 24: 307–325.
The author spent a long career at Oregon State University. He had earned the doctorate in
1982 under supervision of John Corcoran at the State University of New York at Buffalo, with
a dissertation on the American postulate theorists. This article outlines their results and
their influence on Tarski’s work in the 1930s that was to be foundational for model theory.
The American Postulate Theorists were influenced by the European foundational work of
the period around 1900. ... Their work served as paradigm examples of the theories and
concepts investigated in model theory. The article also examines the possibility of a more
specific impetus to Tarski’s model theoretic investigation, arising from his having studied
in 1927–1929 a paper by C. H. Langford proving completeness for various axiom sets for
linear orders. This used the method of elimination of quantifiers.6

•Simmons, Keith. 2009. Tarski’s logic. In Handbook of the History of Logic, edited by
Dov M. Gabbay and John Woods, volume 5: Logic from Russell to Church, 511–616.
Amsterdam: Elsevier.
The author is professor of philosophy at the University of North Carolina. This detailed outline
covers of all of Tarski’s research except some works of strictly technical mathematics. More-
over, the author assesses its impact and describes its relationship to some later works of others.

•Sinaceur, Hourya. 1996. Mathématiques et métamathématique du congrès de Paris

(1900) au congrès de Nice (1970): Nombres réels et théorie des modèles dans les
travaux de Tarski. Rendiconti del Circolo Matematico di Palermo (serie 2 supplemento)
44: 113–132.
This study by a noted Moroccan mathematician and historian is devoted to Tarski’s contribu-
tions to the metamathematics of the real-number system and to model theory: in particular,
the decidability and completeness of the elementary theory of real numbers.

6
From the author’s abstract. Langford 1926, 1927. For information on Tarski’s 1927–1929 seminar see
section 9.4.
18.2 Other Surveys 419

•———. 2001. Alfred Tarski: semantic shift, heuristic shift in metamathematics.

Synthese 126: 49–65.
In his review of this paper, the American–French logician Yehuda Rav wrote,
The author traces, with impeccable scholarship, the remarkable shift in the place of meta-
mathematics within the mathematical sciences due to the work of Tarski. The emphasis
by [David] Hilbert and his school was on syntactic methods, and the early [Rudolf] Carnap
followed this trend. ... Tarski ... brought about a major shift. ..through the introduction of
semantic methods, culminating in the development of model theory, with its interplay of
semantic and syntactic relations. Thus, metamathematics ... became part and parcel of math-
ematics, on a par with any other branch of mathematics.7

•Suppes, Patrick, Jon Barwise, and Solomon Feferman. 1989. Commemorative meeting
for Alfred Tarski: Stanford University, November 7, 1983. In A Century of Mathemat-
ics in America, edited by Peter Duren et al., volume 3, 393–403. Providence: American
Mathematical Society.
This article is the proceedings of a meeting that consisted of talks by the three authors, all
Stanford professors of philosophy. The biographical talks by Suppes and Feferman were
described in chapter 17. Barwise, who earned the doctorate in 1967 under Feferman’s supervi-
sion, spoke in more technical terms about Tarski’s contributions to semantics and the theory
of models.

•Vaught, Robert L. [1971] 1974. Model theory before 1945. In Henkin et al., op. cit.,
153–172.
The section 18.1 annotation of the later study Vaught 1986–1987 describes the author’s
relationship to Tarski. The present article emphasizes the Löwenheim–Skolem theorem, its
elaboration in Tarski’s 1926–1928 Warsaw seminar (see section 9.4), Tarski’s theory of truth,
and related developments. See Chang [1971] 1974 for an account of later work on model theory
by Tarski and his associates.

•Woleski, Jan. 1987. Alfred Tarski jako filozof. Wiadomoci matematyczne 27: 247–259.
•———. 1993. Tarski as a philosopher. In Polish Scientific Philosophy: The Lvov–
Warsaw School, edited by Francesco Coniglione et al., 319–338. Amsterdam: Rodopi.
The second paper is a translated and greatly expanded version of the first. The author, a
leading Polish philosopher,
describes the intellectual climate in which Tarski worked (the Vienna Circle, the Lwów–
Warsaw logic school), studies his philosophical opinions concerning mathematics, and dis-
cusses the philosophical components of Tarski’s works in logic (primarily of his semantics).8

7
Rav 2001. The issue Synthese 126 (1/2) is devoted to recent considerations of the concept of truth.
8
Murawski 1988. Tarski’s son, Jan, also published comments about this paper (Tarski, J. 1994b).
420 18 Research Surveys

•Zygmunt, Jan. 2009. Alfred Tarski—logik i metamatematyk. In O przyrodzie i

kulturze, edited by Ewa Dobierzewska-Mozrzymas and Adam Jezierski, 305–327.
Wrocaw: Wydawnictwo Uniwersitytetu Wrocawskiego. The article’s title phrase and
book’s title mean “logician and metamathematician” and On Nature and Culture.
The author is a noted Polish logician and historian of logic at the University of Wrocaw.
Contents:
1. Scientific biography
2. Set theory
3. Geometry and measure theory
4. Decidable and undecidable theories
5. Formalizing foundations of mathematics based on the relational calculus
 Bibliography
This bibliography includes all, and only, works referred to in the present volume. Since
most of the book is about the first third of Alfred Tarski’s career, in Poland, very many
works that Tarski completed later, in the United States, are not mentioned here. In 1986,
Steven R. Givant prepared an authoritative, complete bibliography of Tarski’s works.
That is updated in this book by lists of additional Tarski publications in chapter 16 and
supplemented with lists of major works about Tarski in chapters 17 and 18. The entries
of those lists are related to corresponding entries in the following bibliography, but were
tailored to suit their special purposes.

In this book, bibliographic references are cited by giving a short version of the author’s
name and a date, nearby and usually, but not always, in that order. The book mentions
more than one person named Tarski; citations that include this surname only are refer-
ences to Alfred Tarski. Joint authors, up to three, are listed in title-page order. Alpha-
betization of the bibliography ignored diacritical marks, punctuation, and spaces within
names. Warning: this resulted in alphabetization different from the Polish standard!

Some cited works have appeared in several versions. Information about the first often
has historical interest, even when reference to a later one is more appropriate. In such
cases, both dates are given, as in
Ajdukiewicz, Kazimierz. [1921] 1966. ...
The remaining data in such entries refer to the later version unless otherwise specified.
Some journals cited here identify their volumes by year ranges rather than single years.
An article in such a journal is identified by its year of publication if possible; the range
is then given in parentheses following the volume number. Often, when a cited item is
contained in a larger publication, both are cited, with cross-references. The entries for
five large collections of Tarski’s works are displayed in boxes.

After author, date, and title, each entry of the bibliography includes information about
the item’s publication, or its location in case it has not been published. This is followed
by an annotation that explains its origin, import, and relationship to other items. When
a title does not clearly suggest the language of a non-English item, the annotation does
so. For an item listed in the Givant 1986 Tarski bibliography, the first element in the
annotation is the identification code that Givant assigned. An annotation may include
a code such as “JFM: Name” or “MR: Name” to indicate that the cited work was reviewed
in the Jahrbuch über die Fortschritte der Mathematik or in Mathematical Reviews, by the
named reviewer. Many annotations include English translations of Slavic titles or institu-
tional names. Each annotation identifies sections of this book or other bibliography items
that refer to it.1

1
The terms przegld and ruch—review and trends—recur in Polish journal titles. A section number of the
form n.0 refers to text in chapter n before the first numbered section.

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 421
DOI 10.1007/978-1-4939-1474-6, © Springer Science+Business Media New York 2014
422 Bibliography

As much as possible, titles appear here as they did on their published title pages. This
practice, advocated by current documentation authorities,2 is consistent with major
catalogs now searchable electronically. But readers may encounter difficulty matching
titles, particularly of journals, with nomenclature in past literature, or in catalogs based
on it, which often differs considerably from that of title pages. For example, the Giornale
di matematiche was often referred to as the Giornale di Battaglini, after the original
editor; and the journal published by the organization known as the Dorpater Natur-
forschergesellschaft was sometimes listed by an institutional name—Universität
Tartu—or in the language—Estonian—more politically correct at the time the catalog
was constructed.3

Adamowicz, Zofia, et al., editors. 2004. Provinces of Logic Determined: Essays in the Memory
of Alfred Tarski. Same as volumes 126–127 of Annals of Pure and Applied Logic. Cited in 18.2.
Addison, John W. 1983. Eloge: Alfred Tarski, 1901–1983. California Monthly 94 (2, December):
28. Reprinted in Annals of the History of Computing 6 (1984): 335–336. Translated by Irena
Czubiska as “Jeden z najwikszych logików,” Przegld techniczny 20 (1985): 39. That title
means “One of the greatest logicians.” Cited in chapter 17.
Ajdukiewicz, Kazimierz. [1921] 1966. From the methodology of the deductive sciences. Translated
by Jerzy Giedymin. Studia Logica 19: 9–46. Originally published as Z metodologii nauk
dedukcyjnych, publication 10, by the Polish Philosophical Society in Lwów (Polskie Towar-
zystwo Filozoficzne we Lwowie). Cited in chapter 3.
———. 1928. Gówne zasady metodologji nauk i logiki formalnej. Lectures delivered during
academic year 1927–1928. Authorized transcript by Mojz esz Presburger. Publication 16.
Warsaw: Faculty of Mathematics and Physics of the University of Warsaw. The title means
Main Principles of the Methodology of Science and Formal Logic. Cited in 9.4 and 16.1.
———. [1934] 1978. The scientific world-perspective. Translated by Wilfrid Sellars. In Ajdukie-
wicz 1978, 111–117. Originally published as “Naukowa perspektywa wiata,” Przegld filo-
zoficzny 37 (1934): 409–416. Cited in 15.10.
———. 1936. Zagadnienie idealizmu w sformuowaniu semantycznym. In Polski Zjazd Filozo-
ficzny 1936, 334–340. Contains the original version of the comment Tarski [1936] 2014a. The
title means “The problem of idealism in a semantic formulation.” This was published in
Przegld filozoficzny 39. In the table of contents for the congress in the journal, the paper’s
title began instead with the word “Problemat.” Cited in 15.8.
———. [1937] 1978. A semantical version of the problem of transcendental idealism. Translated
by Jerzy Giedymin. In Ajdukiewicz 1978, 140–154. Originally published as “Problemat
transcendentalnego idealizmu w sformuowaniu semantycznym,” Przegld filozoficzny 40:
271–287. Cited in 15.8.
———. 1978. The Scientific World-Perspective and Other Essays, 1931–1963. Synthese library
108. Edited by Jerzy Giedymin. Dordrecht: D. Reidel. Contains Ajdukiewicz [1934] 1978,
[1937] 1978, and Giedymin 1978.
Albiski, Marian. 1976. Wspomnienia o Banachu i Wilkosu. Wiadomoci matematyczne 19:
133–135. The title means “Memories of Banach and Wilkosz.” Cited in 15.9.
Allen, Layman E. 1962. Wff ’N Proof: The Game of Modern Logic. New Haven, Connecticut:
Wff ’n Proof. Cited in 15.1.
———. 2010. Personal communication to James T. Smith, 25 June 2010.

2
Chicago Manual 1993, 15.208–209; Mann 1998.
3
The serials list in May 1973, Appendix 2, may help with this problem.
Bibliography 423

Amaldi, Ugo. [1900] 1912. Sulla teoria della equivalenza. In Enriques [1900] 1912–1914, volume
1, 145–198. Revision of the original version in Enriques 1900. The 1911 first volume of
Enriques [1900] 1907–1911 contains a German translation. Amaldi [1900] 1914 is a Polish
translation.
———. [1900] 1914. O teorji równowanoci. Translated by Stefan Kwietniewski and Wadysaw
Wojtowicz. In Enriques [1900] 1914–1917, article 6, 132–181. Translation of Amaldi [1900]
1912. Cited in 4.1, 4.3, and chapter 5.
Amaldi, Ugo, and Tullio Levi-Cività. 1920. Lezioni di meccanica razionale. Padova: “La litotipo”
Editrice Universitaria. Cited in 9.2.
Andréka, Hajnal, J. Donald Monk, and István Németi, editors. 1991. Algebraic Logic: Colloquia
Mathematica Societatis János Bolyai, 54 (Budapest, 8–14 August 1988). Amsterdam: North–
Holland Publishing Company. Contains Givant [1988] 1991. Cited in 18.2.
Anellis, Irving H. 2005. Review of Feferman and Feferman 2004. Review of Modern Logic 10:
117–129. Cited in chapter 17.
Apostol, Tom M., and Mamikon A. Mnatsakanian. 2011. Complete dissections: converting
regions and their boundaries. American Mathematical Monthly 118: 789–798. MR: Richter.
Cited in 8.5.
Apt, Krzystof R. 2007. Review of Feferman and Feferman 2004. Mathematical Intelligencer
29(2): 79–80. Cited in chapter 17.
Arboleda, Luis Carlos. 1990. Kuratowski, Kazimierz. In Dictionary of Scientific Biography
1970–, volume 17, supplement 2: 519–521. Cited in 4.4.
Archimedes. [1897] 2002a. Measurement of a circle. In Archimedes [1897] 2002b, 91–98. Cited
in 9.8.
———. [1897] 2002b. The Works of Archimedes, edited by Thomas L. Heath. Mineola, New York:
Dover Publications. JFM: Gibson. MR: Schneider. Originally published in 1897 in Cambridge,
England, by Cambridge University Press, with a supplement in 1912. Contains Archimedes
[1897] 2002a.
Aristotle. 1971. Metaphysics, Books *, ' , and ( . Translated with notes by Christopher Kirwan.
Oxford: Clarendon Press. Cited in 15.6.

Bafia, Stanisaw, and Andrzej Maryniarczyk. 2001. Powszechna encyklopedia filozofii. Nine
volumes. Lublin: Polskie Towarzystwo Tomasza z Akwinu. The first word of the title means
Universal. The publisher is the Polish Society of Thomas Aquinas. Contains Majdaski and
Lekka-Kowalik 2001.
Banach, Stefan. 1923. Sur le problème de la mesure. Fundamenta Mathematicae 4: 7–33. MR:
Rosenthal. Reprinted in Banach 1967, 66–89. Cited in 4.2, ... , 4.4, 5.0, 6.0, 6.2, 8.7, and 15.7.
———. 1924. Un théorème sur les transformations biunivoques. Fundamenta Mathematicae
6: 236–239. Reprinted in Banach 1967, 114–117. Cited in 4.4, 6.1, and 8.8.
———. [1931] 1987. Theory of Linear Operations. Translated by Francis Jellett. Contains “Some
aspects of the present theory of Banach spaces,” by Aleksander Pelczyski and Czesaw Bessaga,
161–237. Amsterdam: North–Holland Publishing Company. Translated from the 1932 French
edition. Originally published in Warsaw by the Mianowski Fund as Teorja operacyj, volume
1: Operacje linjowe. JFM: Aumann. Cited in 4.2.
Banach, Stefan. 1967. Oeuvres. Volume 1: Travaux sur les fonctions réeles et sur les séries ortho-
gonales. Edited by Stanisaw Hartman and Edward Marczewski. Warsaw: PWN—Éditions
scientifiques de Pologne. MR: Zygmund. Contains reprints of Banach 1923 and 1924 and of
the original version of Banach and Tarski [1924] 2014.
Banach, Stefan, Wacaw Sierpiski, and Wodzimierz Stoek. 1933. Arytmetyka dla I klasy
gimnazjalnej. Lwów: Ksinica-Atlas. The title means Arithmetic for the First Class in
Gimnazjum. Cited in 4.2.
424 Bibliography

Banach, Stefan, and Alfred Tarski. [1924] 2014. On decomposition of point sets into respectively
congruent parts. Chapter 6 of the present book, Tarski 2014. Translation of “Sur la décomposi-
tion des ensembles de points en parties respectivement congruents,” Fundamenta Mathematicae
6: 244–277, Givant 1986 item 24d, reprinted in Banach 1967, 118–148, and in Tarski 1986a,
volume 1, 119–154. JFM: Rosenthal. Cited in 4.0, ... , 4.4, 5.3, 8.4, 8.6, 15.7, and 15.12.
Bartocci, Claudio. 2012. Una piramide di problemi: Storie di geometrie da Gauss a Hilbert.
Milan: Raffaello Cortina Editore. Cited in 4.1.
Barzin, Marcel. [1938] 1939. Langage et réalité; discussion. In Institut International de Col-
laboration Philosophique 1939, 20–27, 35–65. Contains Tarski [1938] 1939.
Baumslag, Naomi. 2005. Murderous Medicine: Nazi Doctors, Human Experimentation, and
Typhus. Westport, Connecticut: Praeger Publishers. Cited in 4.2.
Benko, David. 2007. A new approach to Hilbert’s third problem. American Mathematical
Monthly 114: 665–676, 770. MR: Linhart. Cited in 4.1.
Bettazzi, Rodolfo. 1890. Teoria delle grandezze. Opera premiata dalla Reale Accademia dei
Lincei. Pisa: Enrico Spoerri. JFM: Vivanti. Cited in 4.2.
Betti, Arianna. 2004. Leniewski’s early liar, Tarski, and natural language. Annals of Pure and
Applied Logic 127: 267–287. MR: Murawski. Cited in 1.1, chapter 3, 15.6, and 18.2.
———. 2008. Polish axiomatics and its truth: On Tarski’s Leniewskian background and the
Ajdukiewicz connection. In Patterson 2008, chapter 3, 44–71. Cited in 15.6 and 18.2.
———. 2011. Kazimierz Twardowski. In Zalta 2011 at http://plato.stanford.edu/
archives/sum2011/entries/twardowski/.
Béziau, Jean-Yves, editor. 2010. Universal Logic: An Anthology, from Paul Hertz to Dov Gabbay.
Basel: Birkhäuser. Contains Tarski [1928] 2010.
Birkhoff, Garrett. 1948. Lattice Theory. Revised edition. Colloquium publication 25. Providence,
Rhode Island: American Mathematical Society. MR: Frink. Cited in 4.4.
Baejewski, Stanisaw, Janusz Kutta, and Marek Romaniuk. 1994–2006. Bydgoski sownik
biograficzny. Seven volumes. Bydgoszcz: Kujawsko-Pomorskie Towarzystwo Kulturalne
Wojewódzki Orodek Kultury. The title and publisher’s name mean Bydgoszcz Biographical
Dictionary and Cultural Center of the Kujawsko-Pomorskie Regional Cultural Association.
Contains Kutta 1997.
Blok, Willem J., and Don Pigozzi. 1988. Alfred Tarski’s work on general metamathematics. Jour-
nal of Symbolic Logic 53: 36–50. MR: Mendelson. Cited in 18.1.
Bocheski, J. M. 1994. Morals of thought and speech—reminiscences. In Woleski 1994, 1–8.
MR: Veldman. Cited in 1.1.
Bojarski, Bogdan, Julian awrynowicz, and Yaroslav G. Prytula, editors. 2009. Lvov Mathemati-
cal School in the Period 1915–1945 as Seen Today. Warsaw: Polish Academy of Sciences,
Institute of Mathematics. Contains Schinzel 2009.
Boltyanski, Vladimir Grigorevich. 1978. Hilbert’s Third Problem. Translated by Richard A.
Silverman. Introduced by Albert B. J. Novikoff. Washington: V. H. Winston and Sons. MR:
Firey. Cited in 4.4.
Bolyai Farkas. [1832] 1904. Tentamen Iuventutem Studiosam in Elementa Matheseos Purae
Elementaris ac Sublimioris Methodo Intuitiva Evidentiaque Huic Propria Introducendi, cum
Appendice Triplici. Volume two: Elementa Geometriae et Appendices. Second edition. Edited
by József Kürschák, Móricz Réthy, and Béla T tössy de Zepethnek. Budapest: Hungarian
Academy of Sciences. The main title means Essay for Studious Youth on the Elements of
Elementary and Higher Pure Mathematics, Properly Introduced Here by an Intuitive and Clear
Method, with Three Appendices. Cited in 4.1.
Bolyai János. [1832] 1891. The science absolute of space independent of the truth or falsity of
Euclid’s axiom XI (which can never be decided a priori). Translated by George Bruce Halsted.
Scientiae Baccalaureus 1: 203–260. Reprinted in Bonola [1906] 1955. Originally published
Bibliography 425

as “Scientiam spatii absolute veram exhibens: a veritate aut falsitate axiomatis XI Euclidiei
(a priori haud unquam decidenda) independentem,” appended to the [1832] 1904 book by
Bolyai’s father Farkas.
Bonola, Roberto. [1906] 1955. Non-Euclidean Geometry: A Critical and Historical Study of Its
Developments. Translated, with additional appendices by H. S. Carslaw. With an introduction
by Federigo Enriques. With a supplement containing the translations J. Bolyai [1832] 1891
and Lobachevsky [1840] 1914 by George Bruce Halsted. New York: Dover Publications.
Bonola’s book was first published by Nicola Zanichelli in Bologna as La geometria non-euclidea:
Esposizione storico-critico de suo sviluppo, JFM: Lampe. Cited in 12.11.
Borel, Émile. 1914. Leçons sur la théorie des fonctions. Second edition. Paris: Gauthier-Villars.
The first edition was published in 1898. JFM: Lampe. Cited in 4.4.
Borowski, Marjan. 1922. Letter to Kazimierz Twardowski, 24 February 1922. In the “Documen-
tation on Leniewski” section of Coniglione and Betti, editors 2001–. Translated by Jennifer
Smith. The original is in the Twardowski Archives in Warsaw. Cited in 1.1.
Bouveresse, Jacques, editor. 1991–1993. Philosophie de la logique et philosophie du langage.
Two volumes. L’âge de la science 4–5. Paris: Éditions Odile Jacob. Contains Tarski
[1955] 1993.
Brandes, Hans. 1907. Über die axiomatische Einfachheit mit besonderer Berücksichtigung der
auf Addition beruhenden Zerlegungsbeweise des Pythagoräischen Lehrsatzes. Inaugural disser-
tation presented to the Philosophical Faculty of the University of Halle–Wittenberg. Bruns-
wick: Friedrich Vieweg und Sohn. JFM: Zacharias. Cited in 7.1.
Braun, Stephania, Edward Szpilrajn, and Kazimierz Kuratowski. 1935. Annexe. Fundamenta
Mathematicae 1 (second edition of 1920 volume): 225–254. Cited in 4.4.
Brieskorn, Egbert, editor. 1996. Felix Hausdorff zum Gedächtnis, volume 1: Aspekte seines
Werkes. Braunschweig: Friedrich Vieweg & Sohn. MR: Scriba. Contains Schreiber 1996.
Brzk, Gabriel. 1983. Henryk Raabe: 1882–1951. Lublin: Uniwersitet Marii Curie-Skodowskiej.
Cited in 9.2.
Bury, Jan. 2004. Polish codebreaking during the Russo-Polish war of 1919–1920. Cryptologia
28: 193–203. Cited in 1.1.

Cantor, Georg. 1883. Über unendliche lineare Punktmannigfaltigkeiten, 5: Grundlagen einer

allgemeinen Mannigfaltigkeitslehre. Mathematische Annalen 21: 51–58, 545–586. JFM:
Schlegel. Reprinted in Cantor 1962, 165–209. Cantor [1883] 1996 is an augmented translation.
Cited in 1.2.
———. [1883] 1996. Foundations of a general theory of manifolds: A mathematico-philosophical
investigation into the theory of the infinite. Translated by William B. Ewald. In Ewald 1996,
volume 2, section 19.C, 878–920. The original, published by B. G. Teubner, is a reprint, with
additional preface and footnotes, of Cantor 1883.
———. 1895–1897. Beiträge zur Begründung der transfiniten Mengenlehre, Mathematische
Annalen 46 (1895): 481–512; 49 (1897): 207–246. JFM: Vivanti. Reprinted in Cantor 1962,
282–356. Cantor [1895–1897] 1952 is a translation.
———. [1895–1897] 1952. Contributions to the Founding of the Theory of Transfinite Numbers.
Translated and with a preface, introduction, and notes by Philip E. B. Jourdain. New York:
Dover Publications. JFM: Vivanti. Translation of Cantor 1895–1897, originally published in
1915 by the Open Court Publishing Company. Cited in 1.2.
———. 1962. Gesammelte Abhandlungen mathematischen und philosophischen Inhalts, mit
erläuternden Anmerkungen sowie mit Ergänzungen aus dem Briefwechsel Cantor–Dedekind.
Edited by Ernst Zermelo. With a biography of Cantor by Abraham A. Fraenkel. Hildesheim:
Georg Olms Verlagsbuchhandlung. JFM: Pannwitz. Reprint of the original 1932 edition.
Contains reprints of Cantor 1883 and 1895–1897.
426 Bibliography

Carnielli, Walter. 2006. Review of Feferman and Feferman 2004. Logic and Logical Philosophy
15: 91–96. Cited in chapter 17.
Castelnuovo, Guido. 1947. Federigo Enriques. Accademia Nazionale dei Lincei, classe di scienze
fisiche, matematiche e naturali: Rendiconti (series 8) 2: ix–xix. In Italian. Cited in 9.2.
Castelnuovo, Guido, and Federigo Enriques. 1908. Grundeigenschaften der algebraischen
Flächen. In Meyer and Mohrmann 1907–1934, part 2, half 1, 635–673. Article III C, 6a. JFM:
Ostrowski. Cited in 9.2.
———. 1914. Die algebraischen Flächen vom Gesichtspunkte der birationalen Transformationen
aus. In Meyer and Mohrmann 1907–1934, part 2, half 1, 674–767. Article III C, 6b. JFM:
Ostrowski. Cited in 9.2.
Castrillo, Pilar, and Luis Vega Reñon, editors. 1984. Lecturas de lógica II: Selección,
introducción y traducción. Madrid: Universidad Nacional de Educación a Distancia. Contains
Tarski [1935] 1984. Cited in 16.1.
Cat, Jordi. 2010. Otto Neurath. In Zalta 2011– at http://plato.stanford.edu/entries/
neurath/. Cited in 16.3.
Cesarski, Wojciech. 1995. Review of Tarski 1995–2001, volume 1. Gazeta wyborcza 267 (17
November): 2. In Polish. The journal is a daily newspaper with national circulation. Cited
in 16.2.
Chang, Chen-Chung. [1971] 1974. Model theory 1945–1971. In Henkin et al. 1974, 173–186.
MR: Anonymous. Cited in 18.1 and 18.2.
Chatterji, Srishti D. 2001. Commentary on Hausdorff 1914b. In Hausdorff 2001–2002, volume
4, 11–18. In English. Cited in 4.4.
———. 2002. Hausdorff als Masstheoretiker. Mathematische Semesterberichte 49: 129–143.
Cited in 4.2.
Chicago Manual of Style, The. 1993. Fourteenth edition. Chicago: University of Chicago Press.
Cited in the preface and this bibliography’s introduction.
Chicago, University of, News Office. 2008. Izaak Wirszup, 1915–2008. Press release, 31 January.
Chicago: University of Chicago News Office. Cited in 7.1.
Chodorowski, Barbara, et al. 2005. 100 lat na Smolnej czyli dzieje gimnazjum i liceum im. Jana
Zamoyskiego w Warszawie. Warsaw: XVIII Liceum im. Jana Zamoyskiego. The title means
100 Years of History of the Jan Zamoyski Gimnazjum and Liceum on Smolna Street. Cited in
9.9 and 15.2.
Choquet, Gustave, et al., editors. 2004. Autour du centenaire Lebesgue. Panoramas et syn-
thèses 18. Paris: Société Mathématique de France. MR: Anonymous. Contains item De la
Harpe 2004.
Chuaqui Kettlun, Rolando. 1984. Alfred Tarski, matemático de la verdad. Revista universitaria
11: 30–34. Translated into Portuguese by S. P. Monoide as “Alfred Tarski, matemático da ver-
dade,” Boletim da Sociedade Paranaense de Matematica 6 (1985): 1–10. Cited in chapter 17.
Church, Alonzo. 1941–1947. Reviews of editions of Tarski [1936] 1995. Journal of Symbolic
Logic 6(1941): 30–32; 12(1947): 61.
———. 1956. Introduction to Mathematical Logic. Princeton: Princeton University Press. MR:
Gál. There is no volume two. Cited in chapter 3.
Chwiakowski, Zygmunt. 1914. " /-!3]#!?!=1; -' !!]E1; . Warsaw: *%# '/]E '7 -
) # ,5! # "'- . These are the original spellings of the title and publisher’s name; the
author’s name was spelled (. . 0 E# )]. The title and publisher’s name mean On Func-
tional Equations and Typographer of the Warsaw Schools. Cited in 3.1.
Chwiakowski, Zygmunt, Wacaw Schayer, and Alfred Tarski. [1935] 1946. Geometrja dla trzeciej
klasy gimnazjalnej. Second edition, reprinted. Hanover: Polski Zwizek Wychodctwa Przy-
musowego. Givant 1986 item 35m. The title and publisher’s name mean Geometry for the Third
Gimnazjum Class and Polish Association of Forced Emigration. The 1935 edition was
Bibliography 427

published in Lwów by the Pastwowe Wydawnictwo Ksiek Szkolnych (National School Books
Publisher). The 1946 edition was originally published in Jerusalem in 1944 by the Min-
isterstwo Wyzna Religijnych i Owiaty Publicznej, Sekcji Wydawniczej Armii Polskiej na
Wschodzie (Ministry of Religious Denominations and Public Education, Publishing Section for
the Polish Army in the East). For an advertisement, see Pastwowe Wydawnictwo Ksiek
Szkolnych 1937. Cited in 9.7, 9.9, 13.0, and 14.1.
Chwistek, Leon. 1960. Wielo rzeczywistoci w sztuce i inne szkice literacke. With preface by
Karol Estreicher. The title means The Multiplicity of Reality in Art, and Other Literary Essays.
Warsaw: Czytelnik. Cited in 14.2.
Congrès Polonais de Mathématiques. [1937] 1938. Comptes-rendus et analyses. Annales de la
Société Polonaise de Mathématique 16: 182–212. Contains Lindenbaum [1937] 1938.
Coniglione, Francesco, and Adrianna Betti. 2001. Kazimierz Ajdukiewicz. In the “Main Polish
Philosophers of the 20th Century” section of Coniglione and Betti, editors 2001–. Cited in
chapter 3.
Coniglione, Francesco, and Arianna Betti, editors. 2001–. Polish Philosophy Page. Catania, Italy:
University of Catania. Internet: http://segr-did2.fmag.unict.it/~polphil/PolHome.html.
Contains Borowski 1922 and Coniglione and Betti 2001.
Coniglione, Francesco, Roberto Poli, and Jan Woleski, editors. 1993. Polish Scientific Philoso-
phy: The Lvov–Warsaw School. Pozna studies in the philosophy of the sciences and the
humanities 28. Amsterdam: Rodopi. Contains Woleski 1993. Cited in 18.2.
Corcoran, John. 1991. Review of Tarski 1986a. Mathematical Reviews 91: MR1015501 (91h:
01101). Cited in the preface.
———. 1993. Review of Givant 1991. Mathematical Reviews 93: MR1117874 (93a: 01046).
Cited in chapter 17.
———. 2011a. Personal communication to James T. Smith, 20 February 2011. Cited in
chapter 3.
———. 2011b. Review of Sinaceur 2009. Mathematical Reviews 2011: MR2509665 (2011b:
03006). Cited in chapters 3 and 17.
Couturat, Louis. [1905] 1918. Algebra logiki. Translated by Bronisaw Knaster and Jan ukasie-
wicz. Warsaw: Wydawnictwo Koa Matematyczno-Fizycznego Suchaczów Universytetu
Warszawskiego. The publisher’s name means “Publisher of the Circle of Mathematics and
Physics Students of the University of Warsaw.” Translation supported by the Mianowski Fund.
Originally published in Paris by Gauthier-Villars as L’algèbre de la logique. Cited in chapter 3.
Cwojdziski, Kazimierz. 1930. Kilka sów o cigach. Parametr 1: 300–305. The title means
“Some words about sequences.” Cited in 12.5.
Czelakowski, Janusz, and Grzegorz Malinowski. 1985. Key notions of Tarski’s methodology of
deductive systems. Studia Logica 44: 321–351. MR: Dawson. Cited in 18.2.
Czeowski, Tadeusz. [1927] 1928. O stosunku przyczynowym. In Polski Zjazd Filozoficzny 1928,
165–168. The title means “On a causal relation.” Contains the original version of the comment
Tarski [1927] 2014b. Cited in 15.3.
Czy, Janusz. 1990. “Enigmatyczna” wojna 1920 r. Wiedza i ycie 44: 44–48. The paper and
journal titles mean “The ‘enigmatic’ war of 1920” and Knowledge and Life. Cited in 1.1.
———. 1994. Paradoxes of Measures and Dimensions Originating in Felix Hausdorff ’s Ideas.
Singapore: World Scientific Publishing Company. MR: Grattan-Guinness. Cited in 4.4 and 8.6.

Davies, Charles. 1890. Elements of Geometry and Trigonometry from the Works of A. M. Legendre,
Adapted to the Course of Mathematical Instruction in the United States. Edited by J. Howard
van Amringe. New York: American Book Company. Cited in 9.8.
Davies, Norman. 1972. White Eagle, Red Star: The Polish–Soviet War, 1919–20. With forward
by A. J. P. Taylor. New York: St. Martin’s Press. Cited in 1.1.
428 Bibliography

———. 1982. God’s Playground: A History of Poland. Volume 1: The Origins to 1795. Volume
2: 1795 to the Present. New York: Columbia University Press. Cited in 1.1, chapter 3, 9.1,
and 9.9.
Davis, Anne C. 1955. A characterization of complete lattices. Pacific Journal of Mathematics
5: 312–319. MR: Jónsson. Cited in 8.8.
Davis, Martin. 2005. The man who defined truth. American Scientist 93 (2, Scientist’s Bookshelf)
1. Review of Feferman and Feferman 2004. Cited in chapter 17.
Davis, Philip J. 2005. A life of logic and the illogic of life. Review of Feferman and Feferman
2004. SIAM News 38 (2, March). Cited in chapter 17.
Dawson, John W., Jr. 1987. Review of Czelakowski and Malinowski 1985. Mathematial Reviews
87: MR0832393 (87f:03083). Cited in 18.2.
———. 1998. Logical contributions to the Menger colloquium. In Menger et al. 1998, 33–42.
Cited in 14.2.
Dehn, Max. 1901–1902. Ueber den Rauminhalt. Mathematische Annalen 55: 465–478. Cited
in 4.1, 4.3, and 5.3.
De la Harpe, Pierre. 2004. Mesures finiment additives et paradoxes. In Choquet 2004, 39–61.
MR: Rosenblatt. Cited in 8.6.
De Zolt, Antonio. 1881. Principii della eguaglianza di poligoni preceduti da alcuni critici sulla
teoria della equivalenza geometrica. Milan: Briola. This booklet is rare. Cited in 4.1.
———. 1883. Principii della eguaglianza di poliedri e di poligoni sferici. Milan: Briola. This
booklet is rare. Cited in 4.1.
Derkowska, Alicja. 2001. Zenon Waraszkiewicz. Wiadomoci matematyczne 37: 135–137. In
Polish. The published title included an incorrect date for Waraszkiewicz’s death: see
Tatarkiewicz 2003. Cited in 7.1.
Dictionary of Scientific Biography. 1970–. Eighteen volumes. Charles C. Gillispie et al., editors.
New York: Charles Scribner’s Sons. Contains Eisele 1971, Keesling 1960, Knaster 1973, and
Moore 1990. Cited in chapter 17 and 18.2.
Dierkesmann, Magda, et al. 1967. Felix Hausdorff zum Gedächtnis. Jahresbericht der Deutschen
Mathematiker-Vereinigung 69: 51–76. Cited in 4.4.
Dobierzewska-Mozrzymas, Ewa, and Adam Jezierski. 2009. O przyrodzie i kulturze. Studium
Generale 13. Wrocaw: Wydawnictwo Uniwersytetu Wrocawskiego. The title means On
Nature and Culture. Contains Zygmunt 2009.
Dobrzycki, Stanisaw. 1971. Kwietniewski Wadysaw (1840–1902). Polski sownik biograficzny
16: 379–380. In Polish. Cited in 9.3.
Dobrzyska, Boena, and Olszewska, Irena. 1998. Chwiakowski Zygmunt. Bibliografia polska
1901–1939. Volume 4: Ch–Cy. In Polish. Warsaw: Biblioteka Narodowa. Cited in 9.9.
Domarus, Max. 1992. Hitler: Speeches and Proclamations, 1932–1945. Volume 2: The Years
1935 to 1938. Translated by Chris Wilcox and Mary Fran Gilbert. Wauconda, Illinois:
Bolchazy–Carducci Publishers. Cited in 14.3.
Domoradzki, Stanisaw. 2011. Personal communication to Andrew McFarland, May 2011. Cited
in 9.4.
Domoradzki, Stanisaw, Zofia Pawlikowska-Broek, and Danuta Wglowska. 2003. Sownik bio-
graficzny matematyków polskich. Tarnobrzeg, Poland: Pastwowa Wysza Szkola Zawodowa.
The title and publisher’s name mean Biographical Dictionary of Polish Mathematicians and
State Higher Vocational School. Contains Pawlikowska-Broek 2003a and 2003b, W. Piotrow-
ski 2003, Skowska and Wglowska 2003, and Pawlikowska-Broek 2011 (in a future edition).
Cited in chapter 17.
Doner, John E., and Wilfrid A. Hodges. 1988. Alfred Tarski and decidable theories. Journal of
Symbolic Logic 53: 20–35. MR: Plotkin. Cited in 18.1.
Bibliography 429

Doner, John E., Andrzej Mostowski, and Alfred Tarski. 1978. The elementary theory of well-
ordering—a metamathematical study. In Macintyre, Pacholski, and Paris 1978, 1–54. Givant
1986 item 78. MR: Slomson. Cited in 8.2.
Drozdowski, Marian, editor. 1968–1973. Warszawa II Rzeczypospolitej 1918–1939. Five volumes.
Warsaw: Pastwowe Widawnictwo Naukowe. The title means Warsaw under the Second
Republic. Contains Konarski 1971 and 1973.
Dubiel, Wadysaw. 1975. Mody matematyk—czasopismo dla modziey szkolnej. Delta 4:
2–6. The title phrase means “magazine for young scholars.” Contains a reprint of the original
version of Tarski [1931] 2014d. Cited in 9.7 and 12.0.
———. 1990. Polskie czasopisma powicone matematyce i jej nauczaniu (1911–1939). Lublin:
Wydawnictwo Universytetu Marii Curie-Skodowskiej. The title means Polish Periodicals
Dedicated to Mathematics and Teaching (1911–1939). Cited in 9.7.
Duda, Roman. 1996. Fundamenta Mathematicae and the Warsaw school of mathematics. In
Goldstein, Gray, and Ritter 1996, chapter 21, 479–498. Cited in 4.4.
———. 2009. Facts and myths about Stefan Banach. Translated by Joseph Pomianowski. News-
letter of the European Mathematical Society 71 (March): 29–34. Cited in 2.1.
Duren, Peter, Richard A. Askey, Harold M. Edwards, and Uta C. Merzbach, editors. 1989. A
Century of Mathematics in America. Three volumes. Providence: American Mathematical
Society. Contains Suppes, Barwise, and Feferman 1989.

Edwards, Paul. 1967. The Encyclopedia of Philosophy. Eight volumes. New York: Macmillan
Publishing Company. Contains Mostowski 1967. Cited in 18.2.
Ehrenfeucht, Andrzej, Wiktor W. Marek, and Marian Srebrny, editors. 2008. Andrzej Mostowski
and Foundational Studies. Amsterdam: IOS press. Contains Krajewski and Srebrny 2008 and
Woleski 2008.
Eisele, Carolyn. 1971. Federigo Enriques. In Dictionary of Scientific Biography 4: 373–375.
Cited in 9.2.
Enriques, Federigo. 1907. Prinzipien der Geometrie. In Meyer and Mohrmann 1907–1934, part
1, half 1, 1–129. Article III A, B 1. JFM: Lampe. Cited in 9.2.
Enriques, Federigo, editor. 1900. Questioni riguardanti la geometria elementare. Bologna: Nicola
Zanichelli. JFM: Loria. Also reviewed in Revue de métaphysique 1904. Enriques [1900]
1907–1911 is a German translation. Enriques [1900] 1912–1914 is a revised and augmented
edition. The [1900] 1914–1917 Polish translation incorporates those revisions. Contains the
original versions of Amaldi [1900] 1912 and [1900] 1914.
———. [1900] 1907–1911. Fragen der Elementargeometrie. Volume 1 (1911): Die Grundlagen
der Geometrie, translated and edited by Hermann Thieme, incorporating revisions drafted for
Enriques [1900] 1912–1914. Volume 2 (1907): Die geometrischen Aufgaben, ihre Lösung und
Lösbarkeit, translated and edited by Hermann Fleischer. Leipzig: B. G. Teubner. Translation
of Enriques 1900. Volume 1 contains a translation of Amaldi [1900] 1912. JFM: Lampe.
———. [1900] 1912–1914. Questioni riguardanti le matematiche elementari. Volume 1: Critica
dei principii. Volume 2: Problemi classici della geometria, numeri primi e analisi indetermi-
nata, massimi e minimi. Bologna: Nicola Zanichelli. JFM: Vivanti. Revised and augmented
edition of Enriques 1900. Partially translated in Enriques [1900] 1914–1917. Volume 1
contains Amaldi [1900] 1912.
———. [1900] 1914–1917. Zagadnienia dotyczce gieometrji elementarnej. Volume 1: Krytyka
podstaw. Volume 2: Konstrukcje gieometryczne i teorja izoperymetrów. Translated by Stefan
Kwietniewski and Wadysaw Wojtowicz. Warsaw: Skad gówny w Ksigarni Gebethnera i
Wolfa. Translation of those articles in Enriques [1900] 1912–1914 whose original versions were
in Enriques 1900. Volume 1 contains Amaldi [1900] 1914 and Kwietniewski and Wojtowicz
1914. Supported by the Mianowski Fund. Cited in 9.3.
430 Bibliography

Enriques, Federigo, and Ugo Amaldi. [1903] 1913. Elementi di geometria ad uso delle scuole
secondarie superiori. Sixth edition. Bologna: Nicola Zanichelli. This book appeared in many
editions, tailored to various audiences. Enriques and Amaldi [1903] 1916 is a Polish translation
of one of them. The first edition was reviewed in Revue de métaphysique 1904. Cited in 4.2,
9.2, and 9.3.
———. [1903] 1916. Zasady gieometrji elementarnej do uytku szkó rednich. Translated by
Wadysaw Wojtowicz. Warsaw: E. Wende i Spólka. Translation of the fifth edition of Enriques
and Amaldi [1903] 1913. Polish libraries list this as published in 1914 or 1916. The book the
present editors have studied carries no such information, but indicates that it was cleared by
the German censors in 1916. Cited in 4.1, 4.2, chapter 5, 9.2, 9.8, 11, 12.11, and 12.13.
Erdo s, Paul. 1943. Some remarks on set theory. Annals of Mathematics (2) 44: 643–646. MR:
Todd. Cited in 15.12.
Estreicher, Karol. 1971. Leon Chwistek: Biografia artysty (1884–1944). Cracow: Pastwowe
Wydawnictwo Naukowe. Cited in 14.2.
Etchemendy, John. 1988. Tarski on truth and logical consequence. Journal of Symbolic Logic
53: 51–79. MR: Luce. Cited in 18.1.
Euclid. [1908] 1956. The Thirteen Books of Euclid’s Elements. Translated from the text of
Heiberg, with introduction and commentary by Thomas L. Heath. Second edition. Three
volumes. New York: Dover Publications. Cited in 2.1, 12.11, 12.13, and 13.1.
Everett, Hazel, Christian Gillot, et al. 2009. The Voronoi diagram of three arbitrary lines in \3.
To appear in Proceedings of the 25th European Workshop on Computational Geometry (Euro-
CG’09). Brussels: Université Libre de Bruxelles. Cited in 12.12 and 12.13.
Everett, Hazel, Daniel Lazard, et al. 2009. The Voronoi diagram of three lines. Discrete and
Computational Geometry 42: 94–130. MR: Anton. Cited in 12.12 and 12.13.
Eves, Howard W. 1963–1965. A Survey of Geometry. Two volumes. Boston: Allyn and Bacon.
MR: Kelly. Cited in 8.5.
Ewald, William B., editor. 1996. From Kant to Hilbert: A Source Book in the Foundations of
Mathematics. Two volumes. Oxford: Clarendon Press. Contains Cantor [1883] 1996.

Feferman, Anita B. [1997] 1999. How the unity of science saved Alfred Tarski. In Woleski and
Köhler 1999, 43–52. Cited in chapter 17.
———. 1999. Alfred Tarski. American National Biography 21: 330–332. Cited in chapter 17.
Feferman, Anita B., and Solomon Feferman. 2004. Alfred Tarski: Life and Logic. Cambridge,
England: Cambridge University Press. MR: Dauben. Also reviewed in Anellis 2005, Apt 2007,
Carnielli 2006, M. Davis 2005, P. Davis 2005, Maddux 2005, Nerode 2010, and Sinaceur 2007.
Feferman and Feferman 2009 is a Polish translation. Contains Tarski [1924] 2004. Cited in
the preface, 1.1, chapter 3, 4.1, 4.3, 9.1, 14.0, 14.2, ... , 14.4, 15.6, 15.11, 15.12, 16.1, 16.3, and
chapter 17.
———. 2009. Alfred Tarski: ycie i logika. Translated by Joanna Goliska-Pilarek and Marian
Srebrny. With preface by Jan Woleski. Warsaw: Wydawnictwa Akademickie i Profesjonalne.
Polish translation of Feferman and Feferman 2004. Cited in the preface and chapter 17.
Feferman, Solomon. [1997] 1999. Tarski and Gödel: between the lines. In Woleski and Köhler
1999, 53–63. Cited in chapter 17.
———. 2003. Alfred Tarski and a watershed meeting in logic: Cornell, 1957. In Hintikka et al.
2003, 151–162. Cited in chapter 17.
———. 2006. Tarski’s influence on computer science. Logical Methods in Computer Science 2(3):
1-1–1-13. Originally published in Proceedings of the 20th Annual IEEE Symposium on Logic
in Computer Science, 2005, 342 ff. MR: S. Feferman. Cited in 8.8 and 18.2.
———. 2008. Tarski’s conceptual analysis of semantical notions. In Patterson 2008, chapter 4,
72–93. Cited in 15.6 and 18.2.
Bibliography 431

Ford, Lester R., and Carver, Walter B. 1957. Laying a rug. American Mathematical Monthly 64:
114–116. Cited in 12.9.
Forsythe, Alexandra. 1969. What points are equidistant from two skew lines? Mathematics
Teacher 62: 97–101. Cited in 12.13.
Frckowiak, Danuta. 1998. Aleksander Jaboski in the eyes of his daughter. In Szudy 1998,
19–40. Cited in chapter 3.
Frank, Phillip. 1935–1936. Prager Vorkonferenz der Internationalen Kongresse für Einheit der
Wissenschaft—1934: Vorbemerkungen. Erkenntnis 5: 1–2. Cited in 14.2.
Fryde, Matthew M. 1964. Wacaw Sierpiski—mathematician. Scripta Mathematica 27: 105–
111. Cited in chapter 3.
Fuchs, Eduard. 1921. Die Juden in der Karikatur: ein Beitrag zur Kulturgeschichte. Munich:
Albert Langen Verlag. Cited in 1.1.
Fundamenta Mathematicae. 1934 (volume 23, 161). Bemerkung der Redaktion. Cited in 9.4.
———. 1945 (volume 33, ix). Editorial note. This explains the gap in publication between the
two parts of Tarski 1939–1945. Cited in 14.3.

Gabbay, Dov M., and John Woods. 2009. Handbook of the History of Logic. Volume 5: Logic from
Russell to Church. Amsterdam: Elsevier. MR: Kahle. Contains Simmons 2009. Cited in 18.2.
Galavotti, Maria Carla. 2008. A tribute to Janina Hosiasson Lindenbaum: A philosopher victim
of the Holocaust. In Scazzieri and Simili 2008, 179–194. Cited in 14.3.
García-Carpintero, Manuel. 1999. The explanatory value of truth theories embodying the seman-
tic conception. In Peregrin 1999, 129–148. Cited in 15.11.
Gardner, Richard J., and Stan Wagon. 1989. At long last, the circle has been squared. Notices
of the American Mathematical Society 36: 1338–1343. MR: Sah. Cited in 4.3.
Garlicki, Andrzej, et al. 1982. Dzieje Uniwersytetu Warszawskiego 1915–1939. Warsaw: Past-
wowe Wydawnictwo Naukowe. The title means History of Warsaw University 1915–1939.
Cited in 1.1.
Garnett, F. M., and Walter B.Carver. 1925. Problem 3036 [1923, 337]. American Mathematical
Monthly 32: 47–49. Cited in 12.9.
Gazeta Warszawska. 1920 (20 July: 3). Profesorowie na front. The title means “Professors on
the front.” The journal is a Warsaw daily newspaper. Cited in 1.1.
Gazeta wyborcza. 2008 (25 April: 3). Profesor Jarosaw Rudniaski. In Polish. The journal is
a daily newspaper with national circulation. Cited in 14.1.
Gerwien, P. 1833a. Zerschneidung jeder beliebigen Anzahl von gleichen geradlinigen Figuren
in dieselben Stücke. Journal für die reine und angewandte Mathematik 10: 228–234, 301.
Cited in 4.1.
Gerwien, P. 1833b. Zerschneidung jeder beliebigen Menge von verschieden gestalteter Figuren
von gleichem Inhalt auf der Kugelfläche in dieselben Stücke. Journal für die reine und
angewandte Mathematik 10: 235–241, 303. Cited in 4.1 and 6.2.
Giedymin, Jerzy. 1978. Ajdukiewicz’s life and personality. In Ajdukiewicz 1978, xiii–xvi.
Reprinted in Sinisi and Woleski 1995, 10–12. Cited in chapter 3.
Givant, Steven R. 1986. Bibliography of Alfred Tarski. Journal of Symbolic Logic 51: 913–941.
MR: anonymous. Also reviewed, with corrections, in Maddux 1993. Supersedes Tarski 1965a.
Reprinted in Tarski 1986a, volume 4, 729–757. Updated and partially translated into Polish
in Givant 1995. Cited in the preface, chapter 11, 12.0, the introduction to part four, 15.3, 15.4,
16.0, ... , 16.2, and this bibliography’s introduction.
———. [1988] 1991. Tarski’s development of logic and mathematics based on the calculus of rela-
tions. In Andréka et al. 1991, 189–215. MR: Guillaume. Cited in 18.2.
———. 1991. A portrait of Alfred Tarski. Mathematical Intelligencer 13(3): 16–32. MR: Cor-
coran. Translated into Czech by Helena Nešet ilová as “Portrét Alfréda Tarského,” Pokroky
432 Bibliography

matematiky, fysiky a astronomie 37 (1992): 185–205. Givant and Huber-Dyson 1996 is an

augmented Polish translation. Cited in 12.0, chapter 17, and the permissions section.
———. 1995. Bibliografia prac Alfreda Tarskiego. In Tarski 1995–2001, volume 1, 333–372.
Translated and updated from Givant 1986, and trimmed to the narrower scope of Tarski
1995–2001. Cited in 16.2.
———. 1999. Unifying threads in Alfred Tarski’s work. Mathematical Intelligencer 21(1):
47–58. MR: Murawski. Cited in 1.1, chapter 3, 8.0, 9.1, 9.3, 18.2, and the permissions section.
Givant, Steven R., and Verena Huber-Dyson. 1996. Alfred Tarski w kalejdoskopie impresji osobis-
tych. Translated from English by Adelina Morawiec. Wiadomoci matematyczne 32 (1996),
95–127. The title means “Alfred Tarski, a kaleidoscope of personal impressions.” Givant 1991
stemmed from the English draft of this work. Cited in chapter 17.
Gödel, Kurt. [1931] 1967. On formally undecidable propositions of Principia Mathematica and
related systems I. Translated by Jean van Heijenoort. In Van Heijenoort [1967] 1970,
592–617. There is no part II. Cited in 14.2 and 15.8.
Goldstein, Catherine, Jeremy Gray, and Jim Ritter, editors. 1996. L’Europe mathématique:
Histoires, mythes, identités. Paris: Éditions de la Maison des Sciences de l’Homme. MR: Lewis.
Contains Duda 1996.
Goliska-Pilarek, Joanna, Joanna Porbska-Srebrna, and Marian Srebrny. 2009a. Król logiki.
Rzeczpospolita (4–5 April): A20–A21. The titles mean “King of logic” and The Republic. The
journal is a daily newspaper with national circulation. Cited in 1.1 and chapter 17.
———. 2009b. Szlakiem Alfreda Tarskiego po Warszawie. Illustrated presentation, 1 April. The
title means Along the Route of Alfred Tarski in Warsaw. Cited in 1.1 and chapter 17.
Gómez-Torrente, Mario. 2006. Alfred Tarski. In Zalta 2011– at http://plato.stanford.edu/
entries/tarski/. Cited in 18.2.
Goodall, E. W. 1920. Typhus fever in Poland, 1916 to 1919. Proceedings of the Royal Society of
Medicine, Section of Epidemiology and State Medicine 13: 277–289. Cited in 1.1.
Grattan-Guinness, Ivor. 1989. Review of Tarski [1956] 1983. Journal of Symbolic Logic 54:
281–282.
———. 2000. The Search for Mathematical Roots 1870–1940: Logics, Set Theories and the
Foundations of Mathematics from Cantor through Russell to Gödel. Princeton: Princeton
University Press. MR: Murawski. Cited in the foreword.
Greniewski, Henryk. [1927] 1928. Próba precyzacji pewnych poj teorji czynu. In Polski Zjazd
Filozoficzny 1928, 157–160. The title means “An attempt to formulate precisely some notions
of the theory of action.” Contains the original version of the comment Tarski [1927] 2014a.
Cited in 15.2.
Gromska, Daniela. 1948. Philosophes polonais morts entre 1938 et 1945. Studia Logica 3:
31–97. Cited in 14.2 and 14.3.
Gross, Feliks. 1969. Podróz na ksiic—dlaczego? Wiadomoci 24 (22 June, issue 1212): 3. The
title means “Trip to the moon—why?” Cited in 16.3.
Grosse Brockhaus. 1928–1935. Der Grosse Brockhaus: Handbuch des Wissens. Fifteenth,
fully revised edition. Twenty volumes. Leipzig: F. A. Brockhaus. Cited in item Grosse Brock-
haus 1931.
———. 1931. Felix Hausdorff. In Grosse Brockhaus 1928–1935, volume 8, 239. In German.
Cited in 4.4.
Gruszczyński, Rafa, and Andrzej Pietruszczak. 2008. Full development of Tarski’s geometry of
solids. Bulletin of Symbolic Logic 14: 481–540. MR: Pambuccian. Cited in 9.4.
Guillaume, Marcel. 1993. Review of Givant [1988] 1991. Mathematical Reviews 93: MR1153425
(93d: 03066). Cited in 18.2.
Gumaski, Leon. 1984. Tadeusz Czeowski: an advocate of scientific philosophy. Translated
by Sylwia Twardo. Dialectics and Humanism 11: 657–664. Cited in 15.3.
Bibliography 433

Gupta, Haragauri N. 1965a. Contributions to the Axiomatic Foundations of Geometry. PhD

dissertation. Berkeley: University of California. Cited in 12.11.

Haaparanta, Leila, editor. The Development of Modern Logic. Oxford: Oxford University Press.
Contains Mancosu, Zach, and Badesa 2009.
Hadwiger, Hugo. 1957. Vorlesungen über Inhalt, Oberfläche und Isoperimetrie. Grundlehren
der Mathematischen Wissenschaften 93. Berlin: Springer-Verlag. MR: Hewitt. Cited in 8.7.
Harvard University Gazette. 2000 (6 April). Adam Ulam, authority on Russia, dies at 77. Cited
in 14.3.
Hatcher, William S. 1988. Review of Tarski [1944] 1987. Mathematical Reviews 88: MR876004
(88d: 03008). Cited in 16.3.
Hauber, Karl Friedrich. 1829. Scholae Logico-Mathematicae, in Quibus Ars Cogitandi et Elo-
quendi, Inveniendi et Demonstrandi circa Unam Propositionem Quae Est Euclidis Elmentorum
Theorema Primum, Multis Modis et Magna Exemplorum Varietate Exercetur, Proponuntur et
Varia Generalia de Methodo, et Nova Quaedam Tum ad Logicam Theoreticam Pertinentia, Tum
de Porismatibus in Analysi Geometrica Antiquorum. Reutlingen, Württemberg: Officina
Literaria. Cited in 9.9.
Hausdorff, Felix. 1914a. Grundzüge der Mengenlehre. Leipzig: Veit & Comp. JFM: Lampe.
Reprinted in Hausdorff 2001–2002, volume 2, 90–476, with extensive notes and commentaries.
Cited in 4.1, ... , 4.4, 6.0, 6.2, 6.3, and 15.12.
———. 1914b. Bemerkung über den Inhalt von Punktmengen. Mathematische Annalen 75:
428–433. Reprinted, with a 2001 commentary by Srishti D. Chatterji, in Hausdorff 2002,
volume 3, 3–18. JFM: Lampe. Cited in 4.4.
———. 2001–2002. Gesammelte Werke, volume 2: Grundzüge der Mengenlehre, volume 4: Analy-
sis, Algebra, und Zahlentheorie. Edited by Egbert Brieskorn et al. Berlin: Springer-Verlag.
Contains Chatterji 2001, Purkert 2002, and reprints of Hausdorff 1914a and 1914b.
Heath, Thomas L. 1921. A History of Greek Mathematics. Two volumes. Oxford: Clarendon
Press. Cited in chapter 11.
Heflich, Aleksander, and Stanisaw Michalski, editors. 1915–1932. Poradnik dla samouków. Ten
volumes. Warsaw: Gebethner i Wolff, Mianowski Fund. The title means Handbook for Self-
Education. Supported by the Mianowski Fund, which took over publication after volume 2.
Contains Kwietniewski 1923. Cited in 9.3.
Heller, Celia S. 1994. On the Edge of Destruction: Jews of Poland between the Two World Wars.
Detroit: Wayne State University Press. Originally published in 1977. Cited in 1.1 and
chapter 3.
Hempoliski, Micha, editor. 1996. Truth after Tarski. Dialogue and Universalism 6 (1–2): 160
pages. Contains J. Tarski 1996a. Cited in chapter 17.
Henkin, Leon, et al., editors. 1974. Proceedings of the Tarski Symposium: An International
Symposium Held to Honor Alfred Tarski on the Occasion of His Seventieth Birthday. Proceed-
ings of symposia in pure mathematics 25. Providence: American Mathematical Society. MR:
Anonymous. Held at the University of California, Berkeley, 23–30 June 1971. Contains Chang
[1971] 1974 and Vaught [1971] 1974. Cited in chapter 17.
Henkin, Leon, J. Donald Monk, and Alfred Tarski. 1971–1985. Cylindric Algebras. Two parts:
Part I with an Introductory Chapter, General Theory of Algebras. Studies in logic and the
foundations of mathematics 64 and 115. Amsterdam: North–Holland Publishing Company.
Givant 1986 items 71m and 85m. MR: Preller, Comer. Cited in 16.1.
———. 1986. Representable cylindric algebras. Annals of Pure and Applied Logic 31: 23–60.
MR: Georgescu. Cited in 16.1.
Henkin, Leon, Patrick Suppes, and Alfred Tarski, editors. 1959. The Axiomatic Method with
Special Reference to Geometry and Physics: Proceedings of an International Symposium Held
434 Bibliography

at the University of California, Berkeley, December 26, 1957– January 4, 1958. Amsterdam:
North–Holland Publishing Company, 1959. Contains Tarski [1957] 1959.
Hessenberg, Gerhard. 1906. Grundbegriffe der Mengenlehre: zweite Bericht über das Unendliche
in der Mathematik. Göttingen: Vandenhoeck & Ruprecht. JFM: Meyer. Cited in chapter 3.
Hilbert, David. [1899] 1900. Les principes fondamentaux de la géométrie. Translated by Léonce
Laugel. Paris: Gauthier-Villars. JFM: Engel. Translation of the first edition, revised, of
Hilbert [1899] 1922. Cited in 9.3.
———. [1899] 1922. Grundlagen der Geometrie. Fifth, enlarged, edition. Wissenschaft und
Hypothese 7. Leipzig: B. G. Teubner. The original 1899 edition is reprinted in Hilbert 2003,
chapter 5. JFM: Engel, Dehn. Hilbert [1899] 1900 is a French translation of the original
edition. Hilbert [1899] 1971 is an English translation of the tenth edition. Cited in 4.1, chap-
ter 5, 9.3, and 12.11.
———. [1899] 1971. Foundations of Geometry (Grundlagen der Geometrie). Second edition.
LaSalle, Illinois: Open Court. Translated by Leo Unger from the 1968 tenth German edition
of Hilbert [1899] 1922, which was revised and enlarged by Paul Bernays and published in
Stuttgart by B. G. Teubner. Cited in 9.3, 9.4, and 12.11.
———. [1900] 2000. Mathematical problems. Translated by Mary Winston Newson. Bulletin
of the American Mathematical Society (new series) 37: 407– 436. Reprinted from Bulletin of
the American Mathematical Society, 8 (1902): 437– 479. Originally published as “Mathe-
matische Probleme,” Nachrichten von der Königlichen Gesellschaft der Wissenschaften zu
Göttingen, Mathematisch-physikalische Klasse 1900: 253–297, JFM: Wallenberg; reprinted
in Hilbert [1932–1935], volume 3, 290–329. A shorter French version is included in Interna-
tional Congress of Mathematicians 1902, 58–114. Cited in 4.1.
———. [1932–1935] 1970. Gesammelte Abhandlungen. Three volumes. Second edition. Berlin:
J. Springer. Contains a reprint of the original version of Hilbert [1900] 2000.
———. 2003. David Hilbert’s Lectures on the Foundations of Geometry, 1891–1902. Edited by
Michael Hallett and Ullrich Majer. Berlin: Springer-Verlag. Contains a reprint of the original
edition of Hilbert [1899] 1922. Cited in 12.11.
Hilbert, David, and Wilhelm Ackermann. [1928] 1950. Principles of Mathematical Logic. Trans-
lated and edited by Lewis M. Hammond, George G. Leckie, F. Steinhardt, and Robert E. Luce.
New York: Chelsea Publishing Company. JFM: Dürr. Originally published by Springer -Verlag
as Grundzüge der theoretischen Logic. Cited in 15.4.
Hinkis, Arie. 2013. Proofs of the Cantor–Bernstein Theorem: A Mathematical Excursion. Science
networks historical studies, 45. Basel: Birkhäuser. MR: Plotkin. Cited in 8.8.
Hintikka, Jaakko, editor. 1995. From Dedekind to Gödel: Essays on the Development of the Foun-
dations of Mathematics. Synthese library 251. Dordrecht: Kluwer Academic Publishers Group.
Contains Woleski 1995b. Cited in chapter 17.
Hintikka, Jaakko et al., editors. 2003. Philosophy and Logic in Search of the Polish Tradition:
Essays in Honour of Jan Woleski on the Occasion of his 60th Birthday. Synthese library 323.
Contains S. Feferman 2003. Dordrecht: Kluwer Academic Publishers. Cited in chapter 17.
Hi, Henryk. 1971. Jubileusz Alfreda Tarskiego. Kultura 288: 134–140. The title means “Jubilee
of Alfred Tarski.” Cited in 9.3 and chapter 17.
Hodges, Wilfrid A. 1986. Alfred Tarski. Journal of Symbolic Logic 51: 866–868. Cited in chap-
ter 17.
Holdener, Judy. 2009. Review of Sally and Sally 2007. American Mathematical Monthly 116:
754–758.
Hoormann, Cyril F. A., Jr. 1971. On Hauber’s statement of his theorem. Notre Dame Journal
of Formal Logic 12: 86–88. Cited in 9.9.
Horn, Alfred, and Alfred Tarski. 1948a. Measures in Boolean algebras. Bulletin of the American
Mathematical Society 54: 79. Givant 1986 item 48 a. Abstract of a presentation at an American
Bibliography 435

Mathematical Society meeting in Pasadena, California, 29 November 1947. Reprinted in Tarski

1986a, volume 4, 574. This does not cover the material in Horn and Tarski 1948b that is
relevant to the present book. Cited in 8.7.
———. 1948b. Measures in Boolean algebras. Transactions of the American Mathematical Soci-
ety 64: 467–497. MR: Loomis. Givant 1986 item 48c. Reprinted in Tarski 1986a, 199–232.
Horn and Tarski 1948a is an abstract for part of this. Cited in 8.7.
Hosiassonówna, Janina. 1928. Der II. polnische Philosophenkongress (Warschau, 23.–28. Septem-
ber 1927). Archiv für systematische Philosophie und Soziologie (new series) 31: 98–103. Polski
Zjazd Filozoficzny 1928 is the congress proceedings. Cited in 14.2.
Hübner, Piotr, Jan Piskurewicz and Leszek Zasztowt. 1992. Józef Mianowski Fund: a foundation
for the promotion of science, established 1881. History. Translated by Jacek Soszyski. On
the Internet at http://www.mianowski.waw.pl/. Originally published as “Zarys historii Kasy
im. Józefa Mianowskiego,” in Kasa imienia Józefa Mianowskiego Fundacja Popierania Nauki
1881–1991, Warsaw: Kasy im. J. Mianowskiego, 7–54. Cited in 9.3.
Hughes-Hallett, Deborah, et al. 2003. Applied Calculus. Second edition. Hoboken, New Jersey:
John Wiley & Sons. Cited in 15.5.
Huntington, Edward V. 1904. Sets of independent postulates for the algebra of logic. Transac-
tions of the American Mathematical Society 5: 288–309. JFM: Jahnke. Cited in 15.6.
Hutnikiewicz, Artur, editor. 1965. Witwicki Tadeusz. Sprawozdania Towarzystwa Naukowego
w Toruniu 17 (1963): 53–55. In Polish. The journal title means Reports of the Toru Scientific
Society. The article was anonymous. Cited in 15.6.

Infeld, Leopold. 1924. O dowodzie tw. Talesa w szkole redniej. Przegld matematyczno-fizyczny
2: 87–89. The title means “On the proof of Thales’s theorem in secondary school.” Cited in 4.3.
———. 1980. Quest: An Autobiography. Second edition. New York: Chelsea Publishing Com-
pany. The first edition was published in 1941. Cited in chapter 3 and 4.3.
Institut International de Collaboration Philosophique. 1939. Les conceptions modernes de la
raison: Entretiens d’été, Amersfoort 1938. Volume 1: Raison et monde sensible. Edited by
Raymond Bayer. Actualités scientifiques et industrielles 849. Paris: Hermann & Cie.
Contains Barzin [1938] 1939 and Tarski [1938] 1939.
International Congress for the Unity of Science. 1936. Actes du Congrès International de
Philosophie Scientifique, Sorbonne, Paris, 1935. Eight volumes. Actualités scientifiques et
industrielles 388–395. Paris: Hermann & Cie. Contains Tarski [1935] 1936a and [1935] 1936b.
International Congress of Mathematicians. 1902. Compte rendu du Deuxième Congrès
International des Mathématiciens tenu à Paris du 6 au 12 août 1900. Procès-verbaux et
communications. Edited by Ernest Duporcq. Paris: Gauthier-Villars. Contains a French
translation of part of the original version of Hilbert [1900] 2000.
———. 1930. Atti del Congresso Internazionale dei Matematici, Bologna, 3–10 settembre 1928,
six volumes. Bologna: Nicola Zanichelli Editore. Contains Tarski [1928] 1930.
International Congress of Philosophy. 1900–1903. Bibliothèque du Congrès International de
Philosophie. Four volumes. Paris: Librairie Armand Colin. Contains Padoa [1900] 1901.
———. 1936. Actes du Huitième Congrès International de Philosophie à Prague, 2–7 septembre
1934. Prague: Comité d’Organisation du Congrès. Reprinted in 1968 in Nendeln, Liechtenstein,
by Kraus Reprint Limited. Contains Jørgensen [1934] 1936, Reichenbach [1934] 1936, Tarski
[1934] 1936a and [1934] 1936b, and Zawirski [1934] 1936. Cited in 16.1.
———. 1937. Travaux du IX e Congrès International de Philosophie: Congrès Descartes. Twelve
fascicles. Edited by Raymond Bayer. Actualités scientifiques et industrielles 530–541. Paris:
Hermann & Cie. Contains Tarski 1937b.
436 Bibliography

Iwaszkiewicz, Bolesaw. 1935. Geometrja dla III klasy gimnazjalnej. Lwów: Wydawnictwo
Zakadu Narodowego Imienia Ossoliskich. The title means Geometry for the Third Year of
Gimnazjum. Cited in 9.9.
———. 1949a. Od Redakcji. Matematyka: Czasopismo dla nauczycieli 2: 6–7. The article and
journal titles mean “From the Editor” and Mathematics: A Journal for Teachers. The article
was unsigned; Iwaszkiewicz was the journal editor. Marxist commentary on Jakowski 1949.
Cited in 9.7.
———. 1949b. Rozwizania zada z zeszytu nr. I. Zadania z Parametra. Matematyka: Czaso-
pismo dla nauczycieli 2(4): 39–42. The article and journal titles mean “Solutions of exercises
in volume 1. Exercises from Parametr” and Mathematics: A Journal for Teachers. Contains
a solution of Tarski [1930] 2014c. This material was unsigned. Cited in 9.7, 12.0, and 12.3.

Jadacki, Jacek Juliusz. 2003a. Alfred Tarski à Varsovie. In Jadacki 2003b, 139–180. Translation
of “Alfred Tarski w Warszawie” by Wanda Jadacka, ibid., 112–137. Cited in 1.1, chapter 3, 9.1,
9.3, 9.4, 14.1, ... , 14.3, 15.5, and chapter 17.

Jadacki, Jacek Juliusz, editor. 2003b. Alfred Tarski: dedukcja i semantyka (déduction et
sémantique). Materiay Sympozjum Instytutu Filozofii Uniwersytetu Warszawskiego,
Polskiego Towarzystwa Filozoficznego i Towarzystwa Naukowego Warszawskiego
odbytego 15 stycznia 2001 roku w Sali Lustrzaney Paacu Staszica w Warszawie, Nowy
wiat 72 (I pitro) z okazji senej rocznicy urodzin ALFREDA TARSKIEGO (14 I 1901,
Warszawa—27 X 1983, Berkeley). Studia semiotica 25–26. Warsaw: Wydawnictwo
Naukowe Semper. This contains
• Jadacki 2003a; Tarski et al. [1971] 2003;
Tarski [1933] 2003, [1937] 2003, [1939] 2003, [1947–1955] 2003,
[1963] 2003;
• Polish translations Tarski [1934] 2003a, [1934] 2003b, with reprints of
their original versions;
• and reprints of the original versions of
ukasiewicz [1928–1929] 2014 (in part);
Tarski [1925] 2014, [1927] 2014a, [1927] 2014a, [1934] 2014,
[1935] 2014, [1936] 2014a, ... , [1936] 2014c.
Cited in 3.1, 14.2, 14.3, 16.1, 16.3, chapter 17, and 18.2.

Jadacki, Jacek Juliusz, and Barbara Markiewicz, editors. 1993. A mdroci zo nie przemoe.
Warsaw: Polskie Towarzystwo Filozoficzne. The title means Evil Never Overpowers Wisdom.
Cited in 15.11.
Jadczak, Ryszard. 1993a. Lwowski i Poznaski okres w dziaalnoci naukowej Zygmunta Zawir-
skiego. Edukacja filozoficzna 16: 197–205. The title means “The Lwów and Pozna period
in the research of Zygmunt Zawirski.” Cited in 15.10.
———. 1993b. Pozycja Stanisawa Leniewskiego w Szkole Lwowsko-Warszawskiej (Polskie
Towarzystwo Filozoficzne, oddzia w Toruniu, 9 grudnia 1992 r.). Ruch filozoficzny 50: 311–316.
The title means “The position of Stanisaw Leniewski in the Lwów-Warsaw school (Polish
Philosophical Society, Toru section, 9 December 1992). Cited in 1.1.
Jahnke, Hans Niels, editor. 2003. A History of Analysis. History of Mathematics 24. Providence
and London: American Mathematical Society and London Mathematical Society. MR: Barbeau.
Cited in the foreword.
Bibliography 437

Jakimowicz, Emilia, and Adam Miranowicz, editors. 2007. Stefan Banach: Remarkable Life,
Brilliant Mathematics. Gdask: Gdask University Press. Cited in 4.2.
Jakubowski, J. L. 1964. Docent Doktor Roman Hampel. In migielski 1964, 335–339. In Polish.
Cited in 9.1.
James, Ioan Mackenzie. 2009a. Driven to Innovate: A Century of Jewish Mathematicians and
Physicists. Witney, Great Britain: Peter Lang. MR: Lewis. Also reviewed in Siegmund-
Schultze 2010. Contains James 2009b. Cited in chapter 17.
———. 2009b. Napoleon of logic: Alfred Tarski (1901–1983). In James 2009a, 284–288. Cited
in chapter 17.
Janiszewski, Zygmunt. [1918] 1968. The needs of mathematics in Poland. Translated by Mary
Grace Kuzawa. In Kuzawa 1968, appendix, 112–118. The original paper, “O potrzebach
matematyki w Polsce,” was published in Nauka polska 1: 11–18. Cited in 1.1 and 4.4.
Jaroszyska-Kirchmann, Anna D. 2002. Patriotism, responsibility, and the Cold War: Polish
schools in DP camps in Germany, 1945–1951. Polish Review 47: 35–66. Cited in 9.9.
———. 2004. The Exile Mission: The Polish Political Diaspora and Polish Americans, 1939–
1956. Cited in 9.9.
Jakowski, Stanisaw. 1948. Une modification des définitions fondamentales de la géométrie des
corps de M. A. Tarski. Annales de la Société Polonaise de Mathématique 21: 298–301. MR:
Blumenthal. Cited in 9.4.
———. 1949. Geometria bry. Matematyka: Czasopismo dla nauczycieli 2: 1–6. The titles mean
“Geometry of solids” and Mathematics: A Journal for Teachers. Iwaszkiewicz 1949a is a Marx-
ist commentary on this. Cited in 9.4 and 9.7.
Jastrzbowski, Wojciech. 1929. Powszechna Wystawa Krajowa. Poster for the General Polish
Exposition, Pozna, May–September. Bydgoszcz, Poland: Zakady Graficzne “Bibljoteka
Polska”. Cited in 9.5.
Jedynak, Anna. 2001. Janina Hosiasson-Lindenbaum: The Logic of Induction. In Krajewski
2001, 97–101. Cited in 14.3.
Jerzmanowski, Andrzej. 2013. Personal communication to James T. Smith, 2 June. Cited in 9.9.
Jónsson, Bjarni. 1986. The contributions of Alfred Tarski to general algebra. Journal of Sym-
bolic Logic 51: 883–889. MR: Hawkins. Cited in 18.1.
Jónsson, Bjarni, and Alfred Tarski. 1947. Direct Decompositions of Finite Algebraic Systems.
Notre Dame mathematical lectures 5. Notre Dame: University of Notre Dame Press. Givant
1986 item 47m. MR: Birkhoff. Cited in 15.12.
Jørgensen, Jørgen. [1934] 1936. Die logischen Grundlagen der Wissenschaft. In International
Congress of Philosophy 1936, 100–116. Discussed in Tarski [1934] 1936b and [1934] 2003b.
Cited in 16.1.

Kac, Mark. 1931. O nowym sposobie rozwizywania równa stopnia trzeciego. Mody matematyk
1 (1931–1932): 69–71. Kac was then a gimnazjum student; the journal listed his name as
Marek Katz. The title means “A new method for solving third-degree equations.” Cited in 9.7.
———. 1985. Enigmas of Chance: An Autobiography. New York: Harper & Row, Publishers.
MR: Hobbs. Cited in chapter 3 and 9.7.
Kaua, Roman. 1996. Through a Reporter’s Eyes: The Life of Stefan Banach. Translated and
edited by Ann Kostant and Wojbor Woyczyski. Boston: Birkhäuser. MR: Cooke. Cited in 2.1.
Kalicun-Chodowicki, Bazyli. 1925. Geometrja wykrelna: dla wyszych klas szkó rednich (kurs
systematyczny, wedug programu MWRiOP. Lwów: Ksigarnia Naukow. The title means De-
scriptive Geometry for the Higher Classes of Secondary Schools (A Systematic Course Following
the Program of the Ministry of Religious Denominations and Public Education). Cited in 9.2.
Kaptur, Leon. 1930. IV zadanie konkursowe: Dobór parametra. Parametr 1: 397. The title
means, “Fourth contest problem: selection of a parameter.” Cited in 12.1.
438 Bibliography

Karski, Jan. 1944. Story of a Secret State. Boston: Houghton Mifflin Company 1944. Cited in
the preface and 9.9.
Kasperowiczowa, Helena. 1969. Warszawskie szkoy rednie ogólnoksztacce w latach 1915–
1944. Przegld historyczno-owiatowy 12(2): 187–202. The titles mean “Warsaw general
secondary schools in the years 1915–1944” and Historical-educational review. Cited in 9.1.
Keesling, James. 1990. Karol Borsuk. In Dictionary of Scientific Biography 1970–, volume 17,
supplement 2: 93–95. Cited in 15.12.
Kpno. Pastwowe Gimnazjum Koedukacyjne. 1931–1933. Sprawozdanie dyrekcji Pastwowego
Gimnazjum Koedukacyjnego w Kpnie. Volumes for academic years 1929/1930–1930/1931,
1931/1932, and 1932/1933. Edited by Henryk Moese. Kpno: Drukarnia Dziennika
Poznaskiego Spóka Akcyjna, Drukarnia Spókowa w Kpnie. The title means Report of the
directorate of the National Coeducational Gimnazjum in Kpno. Cited in 7.1.
Keynes, John Maynard. 1921. A treatise on probability. London: Macmillan and Company. Cited
in 16.1.
Kiselev, Andrej Petrovi. 1917. Gieometrja elementarna: do uytku szkó rednich. Two volumes.
Translated by Bolesaw Herman. ód: Ksigarnia Ludwika Fiszera. Originally published in
1892 as @!+'!E #+'E, E )' !1; -5!=1;   !], Moscow: Dumnov’s
Bookstore (. . -!# ). The title means Elementary Geometry for Use in Secondary Schools.
Kiselev 2006–2008 is an English translation. Cited in 9.2.
———. 2006–2008. Kiselev’s Geometry. Book 1: Planimetry. Book 2: Stereometry. Adapted from
a later edition of Kiselev 1917 by Alexander Givental. El Cerrito, California: Sumizdat. Cited
in 9.2 and 12.11.
Knaster, Bronisaw. 1927. Un théorème sur les fonctions d’ensembles. Annales de la Société
Polonaise des Mathématique 6: 133–134. Givant 1986 item 27. JFM: Hurewicz. Reprinted
in Tarski 1986a, 546–547. Abstract of a 9 December talk to the Warsaw section of the Polish
Mathematical Society presenting joint results of Knaster and Tarski. Cited in 8.8.
———. 1973. Janiszewski, Zygmunt. Dictionary of Scientific Biography 7: 71–73. Cited in 1.1.
Kokoszyska, Maria. 1936a. W sprawie wzgldnoci i bezwzgldnoci prawdy. In Polski Zjazd
Filozoficzny 1936, 424–426. The title means “Concerning the relativity and nonrelativity of
truth.” Contains the original version of the comment Tarski [1936] 2014d. Cited in 8.1 and
15.11.
———. 1936b. Über den absoluten Wahrheitsbegriff und einige andere semantische Begriffe.
Erkenntnis 6 (December): 143–165. Cited in 15.11.
———. 1937. Wraenia z Pierwszego Midzynarodowego Kongresu Filozofii Naukowej (Pary,
15.–23. wrzenia 1935). Ruch filozoficzny 13: 77–76. The title means “Impressions of the First
International Congress of Scientific Philosophy (Paris, 15–23 September 1935). Cited in 14.2.
———. c1961. Résumé. Item RC 088-57-07 in the Carnap Collection of the library of the
University of Pittsburgh. Cited in 14.2.
Komjáth, Péter. 2006. Review of Wapner 2005. Notices of the American Mathematical Society
53: 1035–1036.
Konarski, Stanisaw. 1971. Warszawskie szkolnictwo powszechne w latach 1918–1939. In Droz-
dowski 1968–1973, volume 3, 215–246. The title means “Warsaw elementary schools in the
years 1918–1939.” There is a preprint of this dated 1966. Cited in 9.1.
———. 1973. Warszawskie rednie szkolnictwo ogólnoksztacce w latach 1918–1939. In
Drozdowski 1968–1973, volume 5, 179–252. The title means “Warsaw general secondary
schools in the years 1918–1939.” Cited in 9.1.
——— 1994. Schayer Stanisaw. Polski sownik biograficzny 35: 411–412. In Polish. Cited
in 9.9.
Kordos, Marek, Maria Moszyska, and Lesaw W. Szczerba. 1977. Wanda Szmielew 1918–1976.
Translated by J. Smólska. Studia Logica 36: 241–244. Originally published in Polish in Wiado-
Bibliography 439

moci Matematyczne (2) 21 (1978): 25–28. Establishes the date of the work in Szmielew [1938]
1947. Cited in 9.4 and 14.2.
Kotarbiski, Tadeusz. [1935] 1955–1956. The fundamental ideas of pansomatism. Mind 64:
488–500, 65: 288. Translated by Alfred Tarski and David Rynin. Givant 1986 item 56.
Originally published as “Zasadnicze myli pansomatyzmu,” Przegld filozoficzny 38, 283–284.
Reprinted in Tarski 1986a, volume 3, 577–592. Cited in chapter 17.
Kotarbiski, Tadeusz. [1955] 1965. Praxiology: An Introduction to the Sciences of Efficient
Action. Translated by Olgierd Wojtasiewicz. Oxford: Pergamon Press. The original edition
was published in Wrocaw as Traktat o dobrej robocie by the Zakad Narodowy im. Ossoliskich
Wydawnictwo Polskiej Akademii Nauk. Cited in 15.2.
———. 1966. Garstka wspomnie o Stanisawie Leniewskim. Ruch filozoficzny 24: 155–163.
The title means “A handful of memories about Stanisaw Leniewski.” A typewritten English
translation has been circulated but evidently not published. Cited in 1.1.
Koterski, Artur. 2001. Henryk Mehlberg—The Reach of Science. In W. Krajewski 2001, 121–127.
Cited in 15.3.
————. 1977. Philosophical self-portrait. Dialectics and Humanism 4: 22–26. Cited in 1.1.
Kötter, Fritz. 1891. Review of Réthy 1891. Jahrbuch über die Fortschritte der Mathematik 23:
review 23.0532.01. Cited in 4.3.
Kozowski, Witold. 2003. Wspomnienie. Alfred Tarski (14 1 1901–27 X 1983). Gazeta wyborcza
264 (November 13): 11. The first word in the title means “A memory of.” The journal is a daily
newspaper with national circulation. Cited in 14.1 and chapter 17.
———. 2010. Personal interview with Andrew and Joanna McFarland, 6 February, in Warsaw.
Cited in 14.1 and 14.3.
Krajewski, Stanisaw. 2004. Gödel on Tarski. Annals of Pure and Applied Logic 127: 303–323.
Cited in chapter 17.
Krajewski, Stanisaw, and Marian Srebrny. 2008. On the life and work of Andrzej Mostowski
(1913–1975). In Ehrenfeucht, Marek, and Srebrny 2008, 3–14. MR: Surma. Cited in 9.4.
Krajewski, Wadysaw, editor. 2001. Polish Philosophers of Science and Nature in the 20th
Century. Amsterdam: Rodopi. Contains Jedynak 2001 and Koterski 2001.
Królikowski, Lech. 1991. Rusiecki Antoni Marian (1892–1956). Polski sownik biograficzny 33(1):
125–126. In Polish. Cited in 9.7.
Krysicki, Wodzimierz. 1930. List do Redakcji dotyczcy zadania Nr. 1. Parametr 1: 196–197.
The title means “Letter to the Editor about exercise 1.” That exercise is Rusiecki 1930a. Cited
in 12.9.
Ksigarnia w. Wojciecha. 1930. Sowo wstpne. Parametr 1(1930): 1. The author name and
title mean “St. Wojciech’s Bookstore” and “Foreword.” St. Wojciech is often referred to as St.
Adalbert. Cited in 9.7.
Kuratowski, Kazimierz. 1924. Une propriété des correspondances biunivoques. Fundamenta
Mathematicae 6: 240–243. JFM: Rosenthal. Cited in 4.4, 6.1, and 8.8.
———. 1958. Topologie. Volume 1. Fourth edition, corrected and augmented, with an appendix
and two notes by Andrzej Mostowski and Roman Sikorski. Monografie matematyczne 20.
Warsaw: Pastwowe Wydawnictwo Naukowe. MR: Roberts. Cited in 4.4.
———. 1974. Wacaw Sierpiski. In Sierpiski 1974–1976, volume 1, 9–14. In French. Cited
in chapter 3.
———. 1980. A Half Century of Polish Mathematics: Remembrances and Reflections. Translated
by Andrzej Kirkor. Warsaw: Polish Scientific Publishers. MR: Mendelson. The original Polish
edition was published in 1973. Cited in 1.1, chapter 3, and 4.4.
———. 1981. Notatki do autobiografii. Warsaw: Czytelnik. Cited in 1.1.
Kurzawa, Jan, and Stanisaw Nawrocki. 1978. Dzieje Kpna. Warsaw: Pastwowe Wydawnictwo
Naukowe, oddzia w Poznaniu. The title means “History of Kpno.” Cited in 7.1.
440 Bibliography

Kutta, Janusz. 1997. Pasenkiewicz Kazimierz (1897–1995). In Baejewski et al. 1997, volume
4, 81–83. In Polish. Cited in chapter 3.
Kuzawa, Mary Grace. 1968. Modern Mathematics: The Genesis of a School in Poland. New
Haven, Connecticut: College and University Press. Contains Janiszewski [1918] 1968. Cited
in 9.3.
Kwietniewski, Stefan. 1923. Review of W. Wojtowicz [1919] 1926. In Heflich and Michalski
1915–1932, volume 3, 19. Cited in 9.2.
Kwietniewski, Stefan, and Wadysaw Wojtowicz. 1914. Od tumaczów. In Enriques [1900]
1914–1917, volume 1, i. The title means “From the translators.” Cited in 9.3.

Laczkovich, Miklós. 1990. Equidecomposability and discrepancy: a solution of Tarski’s circle-

squaring problem. Journal für die reine und angewandte Mathematik 404: 77–117. MR:
Wagon. Cited in 4.3 and 8.6.
Langford, C. H. 1926. Some theorems on deducibility. Annals of Mathematics 28: 16–40. Cited
in 18.2.
———. 1927. Theorems on deducibility. Annals of Mathematics 28: 459–71. Cited in 18.2
———. 1938. Review of Tarski 1937b. Journal of Symbolic Logic 3: 56. Cited in 14.2.
Lapointe, Sandra, et al., editors. 2009. The Golden Age of Polish Philosophy: Kazimierz Twar-
dowski’s Philosophical Legacy. Dordrecht: Springer. Contains Mancosu 2009. Cited in 18.2.
Lebesgue, Henri. 1904. Leçons sur l’intégration et la recherche des fonctions primitives, professées
au Collège de France. Paris: Gauthier-Villars. JFM: Stäckel. Cited in 4.2.
———. 1922. Apropos d’une nouvelle revue mathématique: Fundamenta Mathematicae. Bulletin
des sciences mathématiques (2) 46: 35–48. JFM: Lichtenstein. Cited in the foreword.
Leja, Franciszek, editor. 1930. Comptes-rendus du I Congrès des Mathématiciens des Pays Slaves,
Warszawa, 1929. Warsaw: Ksinica Atlas. Contains Sierpiski [1929] 1930 and the original
version of Presburger [1929] 1991. Cited in 9.5 and chapter 10.
Leniewski, Stanisaw. [1911] 1992. A Contribution to the Analysis of Existential Propositions.
Translated by Stanisaw J. Surma and Jan W. Wójcik. In Leniewski 1992, volume 1, 1–19.
Originally published as “Przyczynek do analizy zda egzystencjalnych,” Przegl d filozoficzny
14 (1911), 329–345. Edited version of Leniewski’s doctoral dissertation. Cited in 1.1.
———. [1927–1930] 1992. On the foundations of mathematics. Translated by D. I. Barnett. In
Leniewski 1992, volume 1, 174–382. Originally published as “O podstawach matematyki,”
Przegld filozoficzny 30 (1927): 164–206; 31 (1928): 261–322; 32 (1929): 60–101; 33 (1930):
77–105. Cited in chapter 3 and 14.2.
———. [1929] 1992. Fundamentals of a new system of the foundations of mathematics. Trans-
lated by Michael P. O’Neil. In Leniewski 1992, volume 2, 410–605. Originally published as
“Grundzüge eines neuen Systems der Grundlagen der Mathematik,” Fundamenta Mathematicae
14: 1–81, JFM: Skolem. Cited in chapter 3.
———. 1992. Collected Works. Two volumes. Edited by Stanisaw J. Surma, Jan T. J.
Srzednicki, D. I. Barnett, and V. Frederick Rickey. Nijhoff international philosophy series 44.
Dordrecht: PWN-Polish Scientific/Kluwer Academic Publishers. Reviewed in Woleski
2000–2001. Contains Leniewski [1911] 1992, [1927–1930] 1992, and [1929] 1992. Cited in 1.1.
Levin, Aleksey E. 1988. Expedient catastrophe: a reconsideration of the 1929 crisis at the Soviet
Academy of Science. Slavic Review 47: 261–279. Cited in 9.5.
Lévy, Azriel. 1979. Basic Set Theory. Berlin: Springer-Verlag. MR: Kanamori. Cited in 15.12.
———. 1988. Alfred Tarski’s work in set theory. Journal of Symbolic Logic 53: 2–6. MR: Surma.
Cited in 18.1.
Lindenbaum, Adolf. [1937] 1938. Sur l’équivalence de deux figures par décomposition en nombre
fini de parties respectivement congruents. In Congrès Polonais de Mathématiques [1937] 1938,
197. Abstract. Cited in 7.1.
Bibliography 441

Lindenbaum, Adolf, and Alfred Tarski. 1926. Communication sur les recherches de la théorie
des ensembles. Comptes rendus de la Société des Sciences et des Lettres de Varsovie, classe III
19: 299–330. JFM: Ruziewicz. Givant 1986 item 26. Reprinted in Tarski 1986a, volume 1,
171–204. Cited in 8.3, 8.6, and 14.2.
———. [1935] 1983. On the limitations of the means of expression of deductive theories. In
Tarski [1956] 1983, 384–392. The original paper, “Über die Beschränktheit der Ausdrucks-
mittel deduktiver Theorien,” Ergebnisse eines mathematischen Kolloquiums 7: 15–22, Givant
1986 item 36b, JFM: Bachmann, was reprinted in Menger et al. 1998, 333–340, and Tarski
1986a, volume 2, 203–212, and translated into Polish in Tarski 1995–2001, volume 2, pages
147–157. Cited in 16.2.
Lindström, Sten, et al., editors. 2009. Logicism, Intuitionism, and Formalism: What Has Become
of Them? Synthese library 341. New York: Springer. Contains Sinaceur 2009. Cited in chap-
ter 17.
Lobachevsky, N. I. [1840] 1914. Geometrical Researches on the Theory of Parallels. Translated
by George Bruce Halsted. Chicago: Open Court Publishing Company. Originally published
by G. Fincke in Berlin as Geometrische Untersuchung zur Theorie der Parallellinien. Reprinted
in Bonola [1906] 1955.
ód. Rzd Miejski. 1946. wiadczenie o czasowym pobycie na terenie m. odzi: Szmielew
Wanda ur. Montlak. Document K. 7175, Zespó Archiwalny Akt Pracowniczych, Archiwum
Uniwersytetu Warszawskiego. The author name, title phrase, and source names mean
“Municipal administration of ód,” “Certification for temporary residence in ód,” and
“Archival Unit for Personnel Records of the University of Warsaw.” Cited in 9.4.
omnicki, Antoni. 1919. O ukadach zasad koniecznych i dostatecznych, sucych do definicyi
pojcia wielkoci. (Sur les systèmes de principes, nécessaires et suffisants servant à la défini-
tion de la notion de grandeur.) Wiadomoci matematyczne 23: 37–70. Cited in 4.2.
———. 1922. O zasadzie dysjunkcyi w logistyce i matematyce. Ruch filozoficzny 6 (1921–1922):
144–146. Presented before the Lwów section of the Polish Philosophical Society on 21 January
1922. The title means, “On the principle of disjunction in logic and mathematics.” Cited in 4.2.
———. 1923. Geometrja: Podrcznik dla szkó rednich. Volume 1: Planimetrja-stereometrja.
Lwów: Ksinica Polska Towarzystwa Nauczycieli Szkó Wyszych. The title phrase and
publisher’s names mean Textbook for Secondary Schools and “Library of the Polish Society of
Teachers in Higher Schools.” Revision of an edition published in Lwów in 1911. Cited in 4.2,
9.2, and 12.11.
———. 1930. Tablice matematyczno-fizyczne czterocyfrowe, dla uytku szkó rednich. Second
edition. Lwów: Ksinica-Atlas. The title means Four-digit matematical and physical tables
for use in secondary schools. The first edition was published in 1926. Cited in chapter 11.
———. 1934. Geometrja dla II klasy gimnazjalnej. Lwów: Ksiica-Atlas. The title means
Geometry for the Second Year of Gimnazjum. Cited in 9.9, 13.1, and 13.4.
———. 1935. Geometrja dla III klasy gimnazjalnej. Lwów: Ksiica-Atlas. Cited in 9.9.
Lorenz, Johann Friedrich. 1798–1807. Grundriss der reinen und angewandten Mathematik, oder
der erste Cursus der gesamten Mathematik. Two volumes. Volume I: Der erste cursus der
Arithmetik und Geometrie. Volume II: Der erste cursus der angewandten Mathematik.
Helmstädt: Carl Gottfried Gleckeisen. Cited in 12.11.
o, Jerzy. 1986. O Alfredzie Tarskim. Ruch filozoficzny 43: 3–10. Cited in 18.2.
Lowry, John, and William Wallace. 1814. Question 269 and solution. New Series of the Mathe-
matical Repository 3: 44–46. Cited in 4.1.
Luchins, Abraham S., and Edith H. Luchins. 2000. Kurt Grelling: steadfast scholar in a time
of madness. Gestalt Theory 22: 228–281. Cited in 14.3.
Luciano, Erika, and Clara Silvia Roero. 2010. La scuola di Peano. In Roero 2010, xi–xviii, 1–212.
Cited in the foreword.
442 Bibliography

ukasiewicz, Jan. 1910. O zasadzie sprzecznoci u Arystotelesa: Studyum krytyczne. Cracow:

Akademia Umiejtnoci. The title means The Principle of Contradiction in Aristotle: A Critical
Study. Cited in 9.4.
———. 1916. O pojciu wielkoci. Przegld filozoficzny 19: 1–70. The title means “On the
concept of magnitude.” Cited in 4.2.
———. 1925. O pewnym sposobie pojmowania teorji dedukcji. Przegld filozoficzny 28: 134–136.
Contains the original version of the comment Tarski [1925] 2014. The title means “On a
certain way of understanding the theory of deduction.” Cited in 8.1 and 15.1.
———. [1927] 1928. O metod w filozofji. In Polski Zjazd Filozoficzny 1928, 3–5. The title
means “On method in philosophy.” Cited in 14.2, 15.2, and 15.3.
———. [1928–1929] 2014. On definitions in the theory of deduction. In section 15.4 of the
present book, Tarski 2014. Translation of “O definicyach w teoryi dedukcyi,” Ruch filozoficzny
11: 177–178, Givant 1986 item 28/29. This abstract announced some results of Tarski, which
are reprinted in Jadacki 2003b, 10. Cited in 8.1, 15.0, and 15.4.
———. 1929. O znaczeniu i potrzebach logiki matematycznej. Nauka polska 10: 604–620. The
title means “On the significance and needs of mathematical logic.” Cited in 15.1.
———. [1929] 1963. Elements of Mathematical Logic, translated by Olgierd Wojtasiewicz,
International series of monographs on pure and applied mathematics 31. New York: Macmil-
lan Company. Originally published as Elementy logiki matematycznej, edited by Mojz esz
Presburger, publication 18, by the Publishing Commission of the Mathematics and Physics
Students of the University of Warsaw. Cited in 15.1, 15.4, and 9.4.
———. [1931] 1970. Comments on Nicod’s axiom and on “generalizing deduction.” Translated
by Olgierd A. Wojtasiewicz. In ukasiewicz 1970, 179–196. Originally published as “Uwagi
o aksjomacie Nicoda i ‘dedukcji uogólniajcej,’” in Polskie Towarzystwo Filozoficzne we Lwowie
1931, 366–382. The translator is not identified. Cited in 15.1.
———. 1936. Co daa filozofii logika matematyczna? In Polski Zjazd Filozoficzny 1936, 325–329.
The title means “What has mathematical logic given to philosophy?” Cited in 15.10.
———. 1936 [1970]. Logistic and philosophy. Translated by Olgierd A. Wojtasiewicz. In
ukasiewicz 1970, 218–235. Originally published as “Logistyka a filozofja,” Przegld
filozoficzny 39: 115–131. Cited in 15.10.
———. [1951] 1957. Aristotle’s Syllogistic from the Standpoint of Modern Formal Logic. Second,
enlarged edition. MR: Martin. Cited in 9.4.
———. [1953] 1994. Curriculum vitae of Jan ukasiewicz. Metalogicon 7: 133–137. MR: Anony-
mous. Incorrectly reprinted in B. Sobociski 1956. Cited in 1.1 and 9.4.
———. 1970. Selected Works. Edited by Ludwik Borkowski. Amsterdam: North–Holland Pub-
lishing Company. MR: anonymous. Contains ukasiewicz [1931] 1970, [1936] 1970, and a
reprint of ukasiewicz and Tarski [1930] 1983.
ukasiewicz, Jan, and Alfred Tarski. [1930] 1983. Investigations into the sentential calculus.
In Tarski [1956] 1983, chapter 4, 38–59. Givant 1986 item 30d. Translation of “Unter-
suchungen über den Aussagenkalkül,” Comptes rendus de la Société des Sciences et des Lettres
de Varsovie, classe III, 23: 30–50, reprinted in Tarski 1986a, volume 1, 321–344. This transla-
tion was reprinted in ukasiewicz 1970, 131–152. The original paper was translated into Polish
in Tarski 1995–2001, volume 2, pages 3–30. Cited in 9.4.

Macintyre, Angus, Leszek Pacholski, and Jeff Paris, editors. 1978. Logic Colloquium ’77:
Proceedings of the Colloquium Held in Wrocaw, August 1977. Amsterdam: North–Holland
Publishing Company. Contains Doner, Mostowski, and Tarski 1978.
Mackiewicz, Witold, editor. 2001. Polska filozofia powojenna. Two volumes. Warsaw: Agencja
Wydawnicza Widmark. The title means Postwar Polish Philosophy. Cited in chapter 17.
Mac Lane, Saunders. 1938. Review of Tarski [1936] 1995. Journal of Symbolic Logic 3: 51–52.
Bibliography 443

Maddux, Roger D. 1993. Review of Alfred Tarski, Collected Papers. Modern Logic 3: 174–188.
Review, with corrections, of Givant 1986 and Tarski 1986a. Cited in 16.2.
———. 2005. Review of Feferman and Feferman 2004. Bulletin of Symbolic Logic 11: 535–540.
Cited in chapter 17.
Mahlo, Paul. 1908. Topologische Untersuchungen über Zerlegung in ebene und sphaerische
Polygone. Inaugural dissertation presented to the Philosophical Faculty of the University of
Halle–Wittenberg. Halle: Hofbuchdruckerei von C. A. Kaemmerer & Co. JFM: Dehn. Cited
in 7.1.
Majdaski, Stanisaw, and Agnieszka Lekka-Kowalik. 2001. Drewnowski Jan Franciszek. In
Bafia and Maryniarczyk 2001, volume 2, 717–721. In Polish. Cited in chapter 3.
Maligranda, Lech. 2008. Antoni omnicki (1991–1941). Wiadomoci matematyczne 44: 61–112.
In Polish. Cited in 4.2.
Mancosu, Paolo. 2005. Harvard 1940–1941: Tarski, Carnap and Quine on a finitistic language
of mathematics for science. History and Philosophy of Logic 26: 327–357. Cited in 16.3.
———. 2008. Tarski, Neurath, and Kokoszyska on the semantic conception of truth. In
Patterson 2008, 192–224. Cited in 15.11 and 18.2.
———. 2009. Tarski’s engagement with philosophy. In Lapointe et al. 2009, chapter 9, 131–153.
Cited in 14.3 and 18.2.
Mancosu, Paolo, editor. 1998. From Brouwer to Hilbert. With the collaboration of Walter P. van
Stigt. New York: Oxford University Press. MR: Pambuccian. Papers translated from Dutch,
French, and German. Cited in the foreword.
Mancosu, Paolo, Richard Zach, and Calixto Badesa. 2009. The development of mathematical logic
from Russell to Tarski: 1900B1935. In Haaparanta 2009, 318B470. Cited in 9.4.
Mann, Thomas. 1998. The Oxford Guide to Library Research. New York: Oxford University
Press. Cited in this bibliography’s introduction.
Manteuffel, Tadeusz. 1936. Uniwersytet Warszawski w latach 1915/16—1934/35: Kronika.
Warsaw: Nakadem Uniwersytetu Józefa Pisudskiego. Latach means years. Cited in 1.1 and
chapter 3.
Maramorosch, Karl, and Farida Mahmood, editors. 1999. Maintenance of Human, Animal,
and Plant Pathogen Vectors. Enfield, New Hampshire: Science Publishers. Contains
Szybalski 1999.
Marchisotto, Elena A., and James T. Smith. 2007. The Legacy of Mario Pieri in Geometry and
Arithmetic. Boston: Birkhäuser. MR: Lewis. Contains Pieri [1908] 2007. Cited in 4.3, 9.3,
9.4, and 14.2.
Marcus, Joseph. 1983. Social and Political History of the Jews in Poland, 1919–1939. Berlin:
Mouton Publishers. Cited in 9.1.
Marczewski, Edward. 1971. Kwietniewski Stefan (1874–1940). Polski sownik biograficzny 16:
378–379. In Polish. Cited in 9.3.
Maryniarczyk, Andrzej, et al. 2011. Encyklopedia filozofii polskiej. Two volumes. Lublin: Polskie
Towarzystwo Tomasza z Akwinu. The publisher is the Polish Society of St. Thomas Aquinas.
Contains an abridgment of Trzsicki 2010.
Mates, Benson, Leon Henkin, and Alfred Tarski. 1959. Jan Kalicki, philosophy, mathematics:
Berkeley and Davis. Berkeley: University of California Digital Archives. Cited in 16.1.
Mathematical Reviews. 1979. Review of Chang [1971] 1974. MR0472502 (57 #12200). Cited
in 18.2.
May, Kenneth O. 1973. Bibliography and Research Manual of the History of Mathematics.
Toronto: University of Toronto Press. MR: anonymous. Cited in this bibliography’s
introduction.
444 Bibliography

Mazurkiewicz, Stefan. 1932. Zur Axiomatik der Wahrscheinlichkeitsrechnung. Comptes rendus

de la Société des Sciences et des Lettres de Varsovie, classe III 25: 1–4. JFM: Dörge. Cited in 16.1.
———. 1939. Stanisaw Leniewski (1886–1939). Przegld filozoficzny 42: 115. In Polish. Cited
in 1.1.
———. 1956. Podstawy rachunku prawdopodobiestwa. Edited by Jerzy o. Monografie mate-
matyczne 32. Warsaw: Pastwowe Wydawnictwo Naukowe. MR: Birnbaum. The title means
Foundations of the Calculus of Probability. Cited in 1.1.
Mazurkiewicz, Stefan, and Wacaw Sierpiski. 1914. Sur un ensemble superposable avec chacune
de ses deux parties. Comptes rendus hebdomadaires des séances de l’Académie des Sciences
158: 618–619. JFM: Korn. Reprinted in Sierpiski 1974–1976, volume 1, 87–88. Cited in 4.1.
McCall, Storrs, editor. 1967. Polish Logic 1920–1939: Papers by Ajdukiewicz, Chwistek,
Jakowski, Jordan, Leniewski, ukasiewicz, Supecki, Sobociski, and Wajsberg. With an
introduction by Tadeusz Kotarbiski. Translated by Bohdan Gruchman, Henryk Hi, Zbigniew
Jordan, Eugene C. Luschei, Storrs McCall, Wendy Teichmann, Horst Weber, and Peter
Woodruff. London: Oxford University Press. Cited in the foreword.
McNulty, George F. 1986. Alfred Tarski and undecidable theories. Journal of Symbolic Logic
51: 890–898. MR: Anshel. Cited in 18.1.
Mehrtens, Herbert. 1980. Felix Hausdorff: Ein Mathematiker seiner Zeit. With forewords by
Egbert Brieskorn and Eberhard Voigt. Bonn: Fachschaftsrat Mathematik und Mathematisches
Institut der Universität Bonn. MR: Kline. Cited in 4.4.
Melnick, Norman. 1971. The ‘Einstein of mathematics’ at UC. San Francisco Chronicle (27
June). Cited in chapter 17.
Mendelson, Elliott. 1995. Review of Tarski 1994. Mathematical Reviews 95: MR1269114 (95e:
03005). Cited in 16.1.
Menger, Karl. 1934. Ist die Quadratur des Kreises lösbar? In Alte Probleme—neue Lösungen
in den exacten Wissenschaften: fünf Wiener Vorträge, zweiter Zyklus, edited by Karl Menger,
1–28. Leipzig: Franz Deuticke. JFM: Pinl. Cited in 4.3 and 8.6.
———. 1994. Reminiscences of the Vienna Circle and the Mathematical Colloquium. Edited by
Louise Golland, Brian McGuinness, and Abe Sklar. Dordrecht: Kluwer Academic Publishers.
MR: Harkleroad. Cited in 14.2.
Menger, Karl, Egbert Dierker, and Karl Sigmund, editors. 1998. Ergebnisse eines mathematischen
Kolloquiums. With contributions, foreword, and afterword by John W. Dawson Jr., Richard
Engelking, W. Hildenbrand, Gérard Debreu, and Franz Alt. Vienna: Springer-Verlag. MR:
Duda. Contains Dawson 1998, and reprints of Tarski 1931 and the original version of
Lindenbaum and Tarski [1935] 1983. Cited in 14.2.
Meschkowski, Herbert. 1966. Unsolved and Unsolvable Problems in Geometry. Translated by
Jane A. C. Burlak. New York: Frederick Ungar Publishing Co. Originally published in 1960
as Ungelöste und unlösbare Probleme der Geometrie, in Brunswick, Germany, by Friedrich
Vieweg & Sohn. MR: Dyer-Bennet. Cited in 4.3.
Meyer, Franz, and Hans Mohrmann, editors. 1907–1934. Encyklopädie der mathematischen
Wissenschaften mit Einschluss ihrer Anwendungen. Volume 3: Geometrie. Leipzig: B. G.
Teubner. Three parts; parts 1 and 2 have two halves; part 2, half 2, has subdivisions A and
B. Other volumes and editions have different editors. Contains Castelnuovo and Enriques
1908 and 1914 and Enriques 1907.
Mihuowicz, Jerzy. 1930. O okreleniu obwodu kola. Parametr 1 (1930–1931): 219–223. The
title means, “On defining the circumference of a circle.” Cited in chapter 11.
Ministerstwo Wyzna Religijnych i Owiecenia Publicznego. 1930. Czasopismo Parametr.
Owiata i wychowanie 2: 983. The author name means Ministry of Religious Denominations
and Public Education; Czasopismo means journal. The article is unsigned. Owiata i
wychowanie was a journal published by the Ministry. Cited in 9.7.
Bibliography 445

Monik, Franišek. 1896. Geometrya pogldowa dla klas niszych szkó rednich. Adapted by
Grzegórz Maryniak from a German edition. Lwów: Ksigarnia Seyfartha i Czajkowskiego.
JFM: Maynz. The title means Intuitive Geometry for the Lower Classes of Secondary Schools.
The German version of the author’s name is Franz, Ritter von Monik. The original publica-
tion was probably the 24th edition of Geometrische Anschauungslehre für Untergymnasien,
published in Vienna by F. Tempsky. Cited in 9.2.
Moese, Henryk. 1929. Problem z Lîlâvatî: Problem siedmiu siódemek. In rem 1929, 9–24. The
title means “Problem from Lîlâvatî: The problem of seven sevens.” Cited in 7.1.
———. [1932] 2014. Contribution to A. Tarski’s problem “On the degree of equivalence of
polygons.” Section 7.3 of the present book, Tarski 2014. Translation of “Przyczynek do prob-
lemu A. Tarskiego: ‘O stopniu równowanoci wieloktów,’” Parametr 2 (1931–1932): 305–309.
A draft translation by Izaak Wirszup, with the same title, appeared in Tarski and Moese 1952,
9–14. Cited in 4.3, 7.0, ... , 7.4, 8.5, 9.7, 9.8, 12.0, and 12.7.
Moise, Edwin E. [1963] 1990. Elementary Geometry from an Advanced Standpoint. Third edition.
Reading, Massachusetts: Addison–Wesley Publishing Company. Cited in 13.1.
Monk, J. Donald. 1986. The contributions of Alfred Tarski to algebraic logic. Journal of Symbolic
Logic 51: 899–906. MR: Surma. Cited in 18.1.
———. 1989. Review of Tarski and Givant 1987. Bulletin of the American Mathematical Society
20: 236–239. Cited in 16.2.
Montague, Richard, Dana S. Scott, and Alfred Tarski. Circa 1970. An Axiomatic Approach to Set
Theory. Berkeley: Bancroft Library, Tarski Archive 84/69, Container 1. Cited in 14.3.
Moore, Gregory H. 1982. Zermelo’s Axiom of Choice: Its Origins, Development, and Influence.
New York: Springer-Verlag. MR: Smoryski. Cited in 4.4.
———. 1990. Alfred Tarski. Dictionary of Scientific Biography 18 (supplement 2): 893–896.
Cited in 18.2.
Morscher, Edgar, Otto Neumaier, and Peter M. Simons. 2011. Ein Philosoph mit “Bodenhaft-
ung”: zu Leben und Werk von Joseph M. Bocheski. Projekte zur Philosophie 9. Sankt
Augustin, Germany: Academia. Cited in 15.9.
Morse, Anthony P. 1949. Squares are normal. Fundamenta Mathematicae 36: 35–39. MR:
Randolph. Cited in 15.12.
Morse, Anthony P., and Randolph, John F. 1940. Gillespie measure. Duke Mathematical Journal
6: 408–419. JFM: Haupt. MR: Jeffrey. Also published in Fundamenta Mathematicae 33
(1945): 12–26. Cited in 15.12.
Mostowski, Andrzej. 1938. O niezalenoci definicji skoczonoci w systemie logiki. Dodatek do
Rocznika Polskiego Towarzystwa Matematycznego 11. Cracow: Drukarnia Uniwersytetu Jagiel-
loskiego. The title means “On the Independence of the Definitions of Finite in a System of
Logic: Supplement to Annales de la Société Polonaise de Mathématique.” Doctoral dissertation,
University of Warsaw. Reviewed in Tarski 1938c. Cited in 9.4, 14.2, and 14.3.
———. 1967. Alfred Tarski. In Edwards 1967, volume 8, 77–81. Cited in 18.2.
Mostowski, Andrzej, and Tarski, Alfred. 1939. Boolesche Ringe mit geordneter Basis. Funda-
menta Mathematicae 32: 69–86. Givant 1986 item 39a. JFM: Skolem. Reprinted in Tarski
1986a, volume 2, 529–548. Cited in 14.3.
———. 1949. Arithmetical Classes and Types of Well-Ordered Systems. Bulletin of the American
Mathematical Society 55: 65, 1192. Cited in 8.2.
Murawski, Roman. 1988. Review of Woleski 1987. Mathematical Reviews 88: MR0908888
(88k:01047). Cited in 18.2.
———. 2011. Alfred Tarski. In Filozofia matematyki i logiki w Polsce midzywojennej, section
3.6, 134–155. Toru: Wydawnictwo Naukowe Uniwersitetu Mikoaja Kopernika. In Polish.
MR: Anonymous. The book’s title means Philosophy of Mathematics and Logic in Interwar
Poland. Cited in 18.2.
446 Bibliography

Murawski, Roman, and Jan Woleski. 2008. Tarski and his Polish predecessors on truth. In
Patterson 2008, chapter 2, 21–43. Cited in 15.6 and 18.2.
Mycielski, Jan. 1979. Finitely additive measures, I. Colloquium Mathematicum 42: 309–318.
MR: Ramsey. Cited in 4.2.

2005–. Biografie di matematici Italiani. On the Internet at http://

Nastasi, Pietro.
matematica-old.unibocconi.it/presentazione.htm . This website augments Tricomi 1962.
Natkowska, Monika. 1999. Numerus clausus, getto awkowe, numerus nullus, “paragraf aryjski”:
Antysemityzm na Uniwersytecie Warszawskim 1931–1939. Warsaw: ydowski Instytut Histor-
yczny. The Polish phrases in the first clause of the title mean “ghetto benches” and “Aryan
paragraph”; the publisher’s name means “Institute of Jewish History.” Cited in 14.2.
Németi, István. 1990. Review of Tarski and Givant 1987. Journal of Symbolic Logic 55: 350–
352. Cited in 16.2.
Nerode, Anil. 2010. Review of Feferman and Feferman 2004. American Mathematical Monthly
117: 286–288. Cited in chapter 17.
Neumann, Johann von. 1929. Zur allgemeinen Theorie des Masses. Fundamenta Mathematicae
13: 73–116, 133. JFM: Pannwitz. Cited in 4.2, 8.6, and 15.12.
Nikodym, Otton. 1930–1937. Dydaktyka matematyki czystej. Volume 1: W zakresie gimnazjum
wyszego, liczby naturalne. Volume 2: W zakresie wyszych klas szkoy redniej, uamki oraz
ich algebra. Lwów, Warsaw: Ksinica–Atlas, Nasza Ksigarnia. The titles mean Teaching
pure mathematics; Natural Numbers in the Higher Gimnazjum Classes; and Fractions and
Their Algebra In the Higher Classes of Secondary School. Cited in 9.3.
Nowak, Leszek. 1995. Ajdukiewicz and the status of the logical theory of natural language. In
Sinisi and Woleski 1995, 205–219. Cited in 15.8.
Nowicki, Ron. 1992. Warsaw: The Cabaret Years. San Francisco: Mercury House. Cited in 9.1
and 14.2.
Nowik, Grzegorz. 2004–2010. Zanim zamano ENIGM ... Polski radiowywiad podczas wojny z
bolszewick Rosj 1918–1920. Two volumes. The title of volume 2 includes after the ellipsis
the clause, rozszyfrowano REWOLUCJ . Warsaw: Oficyna Wydawnicza RYTM. That title means
Before ENIGMA Was Broken ... REVOLUTION Was Decoded: Polish Radio-Intelligence during the
War Against Bolshevik Russia 1918–1920; the uppercase names refer to a German cypher
machine and a Russian code. Cited in 1.1.
Nye, Mary Jo, editor. 2003. The Cambridge History of Science. Volume 5: The Modern Physical
and Mathematical Sciences. Cambridge, United Kingdom: Cambridge University Press. MR:
Gray. Contains Rowe 2003.

Ogniwo: Organ Zwizku Zawodowego Nauczycielstwa Polskich Szkól rednich i biuletyn rzdu
gównego Z. Z. N. P. S. r. 1931 (volume 11, issue 3, 112–113). Drugie zawody strzeleckie
nauczycieli szkól rednich w Warszawie. The titles mean Organ of the Trade Union of Polish
Secondary School Teachers and Bulletin of the General Administration (of the same organiza-
tion) and “Second shooting competition for Warsaw secondary-school teachers.” Cited in 9.1.

Padoa, Alessandro. [1900] 1901. Essai d’une théorie algébrique des nombres entiers, précédé
d’une introduction logique à une théorie déductive quelconque. In International Congress of
Philosophy 1900–1903, volume 3 (1901), 309–365. The introduction, “Logical introduction
to any deductive theory,” translated by Jean van Heijenoort, is in Van Heijenoort [1967] 1970,
118–123. Cited in 14.2.
Pambuccian, Victor. 2004. Early examples of resource-consciousness. Studia Logica 77: 81–86.
MR: Anonymous. Cited in 7.1.
Bibliography 447

Pastwowe Wydawnictwo Ksiek Szkolnych we Lwowie. 1937. Advertisement. Przegld peda-

gogiczny 56: 221. The author name means “National Schoolbook Publisher in Lwów.” This
lists Chwiakowski, Schayer, and Tarski [1935] 1946. Cited in 9.9.
Parafia Wszystkich witych w Warszawie, Proboszcz. [1929] 1940. Akt Maestwa. Nr. 367.
Berkeley: Bancroft Library, Tarski Archive 84/69, Carton 13. The author and title mean
Parish of All Saints in Warsaw, Rector; Marriage Ceremony. Handwritten official copy of
parish records, accompanied by a typewritten German translation, both evidently made in 1940
for the German occupation authorities. Cited in 9.1.
Pasenkiewicz, Kazimierz. 1935. Wykaz wykadów i wicze, pod L. 7615. In the Archive of the
University of Warsaw, Sygnatura Pasenkiewicz Kazimierz: RP 7615. The title phrase means
List of Lectures and Recitations. Pasenkiewicz’s handwritten official record of studies,
1921–1935. Cited in chapter 3.
———. 1984. Recollections about Alfred Tarski from the years 1921–1925. Handwritten trans-
lation, eight pages, by Jan Tarski. In the Alfred Tarski Archive, Bancroft Library, University
of California at Berkeley, Carton 15. Cited in 1.1 and chapter 3.
Patterson, Douglas. 2012. Alfred Tarski: Philosophy of Language and Logic. New York:
Palgrave Macmillan. Cited in 18.2.
———, editor. 2008. New Essays on Tarski and Philosophy. Oxford: Oxford University Press.
Contains Betti 2008, S. Feferman 2008, Mancosu 2008, and Murawski and Woleski 2008.
Cited in 4.6 and 18.2.
Pawlikowska-Broek, Zofia. 1975. Mazurkiewicz Stefan (1888–1945). Polski sownik biograficzny
20: 313–314. In Polish. Cited in 1.1.
———. 2003. Krysicki Wodzimierz (1905–2001); Tarski (Tajtelbaum) Alfred (1901–1983): logik,
filozof, matematyk. In Domoradzki, Pawlikowska-Broek, and Wglowska 2003, 119, 244–245.
In Polish. Cited in 12.9 and chapter 17.
———. 2011. Wilkosz Witold. To appear in the next edition of Domoradzki, Pawlikowska-Broek,
and Wglowska 2003. In Polish. Cited in 15.9.
Peano, Giuseppe. 1903. De latino sine flexione: lingua auxiliare internationale. Revista de
mathematica 8: 74–83. JFM: Vivanti. Included in Peano 2002 as file 1903d.pdf . Reissued in
Roero 2003. This specification starts in Scholastic Latin; as it proposes syntax simplifications,
it implements them, so that it ends in Latino sine Flexione (often called Lingua Peano). Cited
in 9.7.
———. 1924. Prosta teorja logarytmów. Translated by Samuel Dickstein. Przegld
matematyczno-fizyczny 2: 150–152. Originally published in Latino sine Flexione as “Theoria
simplice de logarithmos” in Wiadomoci matematyczne 26 (1924): 53–55. Cited in 4.3.
———. 2002. L’opera omnia di Giuseppe Peano. Edited by Clara S. Roero. Turin: Dipartimento
di Matematica, Università di Torino. CD-ROM. Contains a reprint of Peano 1903.
Peposki, Andrzej. 1995. Polski wywiad wojskowy w okresie bitwy warszawskiej w sierpniu
1920r. Niepodlego i Pami 1995(2), 39–49. The titles means “Polish military intelligence
during the battle of Warsaw in August 1920” and Independence and Memory. Cited in 1.1.
Peregrin, Jaroslav, editor. 1999. Truth and Its Nature (If Any). Synthese library 284. Dordrecht:
Kluwer Academic Publishers. Contains García-Carpintero 1999.
Petrozolin-Skowroska, Barbara, et al., editors. 1994. Gimnazjum im. Stefana eromskiego.
Encyklopedia Warszawy: 210. Warsaw: Wydawnictwo Naukowe PWN. Cited in 9.1.
Pieri, Mario. [1908] 1915. Gieometrja elementarna oparta na pojciach “punktu” i “kuli.” Trans-
lated by Stefan Kwietniewski. Warsaw: Bibljoteka Wektora A3, Skad Gówny w Ksigarni
Gebethnera i Wolffa, Warsaw. Translation of the original version of Pieri [1908] 2007. This
item was incorrectly called Pieri 1914 in Marchisotto and Smith 2007. Supported by the
Mianowski Fund. Cited in 9.3.
448 Bibliography

———. [1908] 2007. Elementary geometry based on the notions of point and sphere. In Marchi-
sotto and Smith 2007, chapter 3. Translation of “La Geometria Elementare istituita sulle
nozioni di ‘punto’ e ‘sfera,’ ” Memorie di matematica e di fisica della Società Italiana delle
Scienze (series 3) 15: 345–450, presented to the Society by Guido Castelnuovo and approved
by Corrado Segre, reprinted in Pieri 1980, 455–560. Pieri [1908] 1915 is a Polish translation
of the original publication. Cited in 9.3 and 9.4.
———. 1980. Opere sui fondamenti della matematica. Edited by the Unione Matematica Italiana,
with contributions by the Consiglio Nazionale delle Ricerche. Contains a reprint of the original
version of Pieri [1908] 2007. Bologna: Edizioni Cremonese.
Piatowicz, Józef. 2006. Straszewicz Stefan (1889–1983). Polski sownik biograficzny 44(1):
233–236. In Polish. Cited in 9.7.
Piotrowski, Tadeusz, editor. 2004. The Polish Deportees of World War II: Recollections of
Removal to the Soviet Union and Dispersal Throughout the World. Jefferson, North Carolina:
McFarland & Company. Cited in 9.9.
Piotrowski, Walerian. 2003. Hornowski Micha (1893–1966), Rusiecki Antoni Marian (1892–
1956), Wojtowicz (Wójtowicz) Wadysaw Bazar (1874–1942). In Domoradzki, Pawlikowska-
Broek, and Wglowska 2003, 76, 206, 261. Cited in 9.3, 9.7, and 12.3.
Pla i Carrera, Josep. 1984. Alfred Tarski i la lògica contemporània. Primera part. Butlletí de
la Societat Catalana de Matemàtiques 17: 26–46. MR: Anonymous. In Catalan. Cited in 18.2.
———. 1989. Alfred Tarski i la teoria de conjunts. Theoria (Spain) (series 2) 4: 343–417. MR:
Anonymous. In Catalan. “Conjunts” means “sets.” Cited in 18.2.
Poland. Gówny Urzd Statystyczny Rzeczypospolitej Polskiej. 1936. May rocznik statystyczny.
Warsaw: Nakadem Gównego Urzdu Statystycznego. The author name and title mean Main
Statistical Office of the Polish Republic, Concise Statistical Yearbook. Cited in 13.0.
Poland. Ministerstwo Wyzna Religijnych i Owiecenia Publicznego. 1931. Program gimnazjum
pastwowego: wydzia matematyczno-przyrodniczy. Fifth edition. Lwów: Pastwowe Wydaw-
nictwo Ksiek Szkolnych. The author is the Mininstry of Religious Denominations and Public
Education and the title means Program for Government Gimnazjums: Mathematical-Scientific
Section. Cited in 9.2.
———. 1934. Program nauki w gimnazjach pastwowych z polskim jzykiem nauczania
(Tymczasowy). Biblioteka owiaty i wychowania, 2. Lwów: Pastwowe Wydawnictwo Ksiek
Szkolnych. The title means Program of Instruction in Government Gimnazjums with Polish
as the Language of Instruction (Provisional). The Program was instituted in 1932. This book
carries no date, but Polish library records date it 1934. Cited in 9.9.
Poland. President. 1927. Rozporzdzenia Prezydenta Rzeczypospolitej z dnia 24 listopada 1927r.
o ubezpieczeniu pracowników umysowych. Dziennik ustaw Rzeczypospolitej Polskiej 1927
(numer 106, pozycja 911): 1463–1488. On the Internet at http://isap.sejm.gov.pl/
DetailsServlet?id= WDU19271060911 in file D19270911.pdf. The titles mean “Decree of the
President of the Republic as of 24 November 1927 concerning insurance of nonmanual workers”
and Journal of Laws of the Republic of Poland. Analyzed in Tarski [1929] 2014b. Cited in 15.5.
Poland. Sejm. 1932. Ustawa z dnia 11 marca 1932 r. o ustroju szkolnictwa. Dziennik ustaw
Rzczypospolitej Polskiej 1932 (numer 38, pozycja 389): 639–645. On the Internet at
http://isap.sejm.gov.pl/DetailsServlet?id=WDU19320380389, in file D19320389.pdf. The
author is the parliament of Poland; the titles mean “Law of 11 March 1932 concerning the
educational system” and Journal of Laws of the Republic of Poland. Cited in 9.1.
Polonsky, Antony. 2012. The Jews in Poland and Russia. Volume 3: 1914 to 2008. Portland,
Oregon: Littman Library of Jewish Civilization. Cited in 1.1.
Polska Akademia Umiejtnoci. 1948. Alfred Tarski. In “Skad Polskiej Akademii Umiejtnoci
(Stan w Marcu 1948),” Rocznik Polskiej Akademii Umiejtnoci 1946–1947: xli–xlii. The author
is the Polish Academy of Art and Sciences in Cracow. Cited in chapter 17.
Bibliography 449

Polskie Towarzystwo Filozoficzne we Lwowie. 1930–1931. December 1930 minutes. Ruch

filozoficzny 12: 209–210. The author is the Polish Philosophical Society in Lwów. Cited in 14.2.
———. 1931. Ksiga pamitkowa Polskiego Towarzystwa Filozoficznego we Lwowie 12.II.1904–
12.II.1929. Lwów: Ksinica–Atlas. Contains the original version of ukasiewicz [1931] 1970.
The title means Commemorative Book of the Polish Philosophical Society in Lwów.
Polskie Towarzystwo Matematyczne. 1927. Roster. Annales de la Société Polonaise de Mathéma-
tique 6: 135–140. Cited in 8.7.
Polski Zjazd Filozoficzny. 1927. Ksiga pamitkowa Pierwszego Polskiego Zjazdu Filozoficznego
(Lwów, 1923). Przegld filozoficzny 30: 256–366. The title means, “Commemorative book
for the First Polish Philosophical Congress.” Cited in chapter 3.
———. 1928. Ksiga pamitkowa Drugiego Polskiego Zjazdu Filozoficznego, Warszawa, 1927.
Przegld filozoficzny 31(1–2): 1–225. The title means “Commemorative Book of the Second
Polish Philosophical Congress.” Contains Czeowski [1927] 1928, Greniewski [1927] 1928, and
the original versions of the comments Tarski [1927] 2014a and[1927] 2014b. Described in
Hosiassonówna 1928.
———. 1936. Ksiga pamitkowa III Polskiego Zjazdu Filozoficznego, Kraków, 1936. Przegld
filozoficzny 39(4): 317–548. The title means “Commemorative Book of the Third Polish Philo-
sophical Congress.” Contains Ajdukiewicz 1936, Kokoszyska 1936a, ukasiewicz 1936,
Wilkosz 1936, Zawirski 1936, and the original versions of the comments Tarski [1936]
2014a, ... , [1936] 2014d. Cited in 4.11.
Polskie Zjazd Matematycznego. 1929. Ksiga pamitkowa Pierwszego Polskiego Zjazdu Mate-
matycznego: Lwów, 7–10.IX.1927. Supplement to Annales de la Société Polonaise de Mathé-
matique 6. The title means Commemorative Book of the First Polish Mathematical Congress.
Contains Tarski [1927] 1929.
Popper, Karl. 1981. Joseph Henry Woodger. British Journal for the Philosophy of Science 32:
38–330. Cited in 15.9.
Powszechna Wystawa Krajowa. 1929. Powszechna Wystawa Krajowa: Pozna, maj–wrzesie
1929. Pozna: Polskie Towarzystwo Ksigarni Kolejowych “Ruch.” The title means The
National Exhibition: Pozna, May–September 1929. Cited in 9.5.
Presburger, Mojesz. [1929] 1991. On the completeness of a certain system of arithmetic of whole
numbers in which addition occurs as the only operation. Translated, with commentary, by Dale
Jacquette. History and Philosophy of Logic 12: 225–233. Translation of “Über die Vollständig-
keit eines gewissen Systems der Arithmetik ganzer Zahlen, in welchem die Addition als einzige
Operation hervortritt,” in Leja 1930, 92–101, 395, JFM: Fraenkel. Cited in 9.4.
Przegld pedagogiczny. 1929 (volume 48, February). Walne Zebranie Koa Warszawskiego. 187–
188. The title means “General assembly of Warsaw circles.” Cited in 9.9.
———. 1931 (volume 50). Odczyty o programach matematyki: 23. The title means “Lectures
in programs for mathematicians.” Cited in 9.9.
Purkert, Walter. 2002. Grundzüge der Mengenlehre: historische Einführung. In Hausdorff 2002,
1–89. Cited in 2.1 and 4.3.
———. 2008. The double life of Felix Hausdorff/Paul Mongré. Translated by Hilde Rowe and
David E. Rowe. Mathematical Intelligencer 30(4): 36–50. Cited in 4.4.

Raabe, Henryk. 1926. III Gimnazjum Mskie Zwizku Zawodowego Nauczycielstwa Polskich
Szkó rednich w Warszawie. Ogniwo 2(6): 30–33. The title is the name of the school where
Tarski taught. Cited in chapter 3 and 9.1.
Rakoczy-Pindor, Krystyna. 2003. Przygoda matematyka z radiem czyli profesor Witold Wilkosz
pionierem radiotechniki i radiofonii w Polsce. Wiadomoci Matematyczne 39: 151–156. The
beginning of the title means “Adventure in radio and mathematics.” Cited in 15.9.
450 Bibliography

Rav, Yehuda. 2001. Review of Sinaceur 2001. Mathematical Reviews 2001: MR1813395 (2001k:
03004). Cited in 18.2.
Redzik, Adam. 2004. Polish Universities During the Second World War. Logroño, Spain: Encuen-
tros de Historia Comparada Hispano-Polaca. Conference presentation. Cited in 9.9.
Reichenbach, Hans. [1934] 1936. Die Bedeutung des Wahrscheinlichkeitsbegriffs für die
Erkenntnis. In International Congress of Philosophy 1936, 163–169. Discussed in Tarski
[1934] 1936a and [1934] 2003a. Cited in 16.1.
Reid, Constance. 1982. Neyman—from Life. New York: Springer-Verlag. MR: Grimm. Cited
in chapter 3.
Réthy, Móricz. 1891. Endlich-gleiche Flächen. Mathematische Annalen 38: 405–428. JFM:
Kötter. Kötter found a flaw; see Amaldi [1900] 1914, 172. Cited in 4.3.
Revue de métaphysique et de morale. 1904 (volume 12, supplement for May, 11–12). Reviews of
the first edition of Enriques and Amaldi [1903] 1913 and of Enriques 1900. Cited in 9.2.
Rickey, V. Frederick. 2011. Polish logic from Warsaw to Dublin: the life and work of Jan
ukasiewicz. Manuscript. Cited in 9.4.
Robinson, Raphael M. 1947. On the decomposition of spheres. Fundamenta Mathematicae 34:
246–260. MR: Blumberg. Cited in 8.6 and 15.12.
Rodríguez-Consuegra, Francisco. 2005. Tarski’s intuitive notion of set. In Sica 2005, 227–266.
Cited in 18.2.
Roero, Clara Silvia, editor. 2003. Le riviste di Giuseppe Peano. Turin: Dipartimento di Matema-
tica, Università di Torino. CD-ROM. Contains a reprint of Peano 1903.
————. 2010. Peano e la sua scuola fra matematica, logica e interlingua: Atti del congreso
internazionale di studi (Torino, 6–7 Ottobre 2008). Turin: Deputazione Subalpina di Storia
Patria. Contains Luciano and Roero 2010.
Rojszczak, Artur. 2002. Philosophical background and philosophical content of the semantic
definition of truth. Erkenntnis 56: 29–62. Cited in 14.2 and 15.11.
Roliski, Adam. 1996. Bolesaw Zahorski–“Lubicz” (1887–1922). A gdy na wojenk szli Ojczy -
nie suy. Pieni i piosenki onierskie z lat 1914–1918. Antologia, 482. Cracow: University
of Cracow Press. The title means And When They Went to Battle to Serve the Motherland:
Soldier Songs and Lyrics ... . Cited in 9.1.
Rosen, Saul. 1985. Alfred Tarski in 1940. Annals of the History of Computing 7: 364–365. Cited
in 15.12, 16.3, and chapter 17.
Rowe, David E. 2003. Mathematical schools, communities, and networks. In Nye 2003, chapter
6, 113–132. Cited in the preface.
Rusiecki, Antoni M. 1923–1924. Obrazy liczbowe na pitce oparte. Przegld matematyczno-
fizyczny 1: 21–25 and 2: 117–134. The title means “Number pictures based on 5.” Cited in 4.3.
———. 1930a. Exercise 1. Parametr 1: 35. Krysicki 1930 is a solution. Reprinted in Mody
matematyk 1(1931, issue 3 supplement): I. Cited in 3.4 and 12.9.
———. 1930b. Osobliwe zadanie. Parametr 1: 75, 271–274. The title means “A peculiar prob-
lem.” Cited in 14.2.
———. 1930c. Zadanie konkursowe: Rekord nie do pobicia. Parametr 1, 193–194. The title
means, “Contest problem: an unbreakable record.” Cited in 12.1.
———. 1930d. III Zadanie konkursowe: amigówka geometryczna. Parametr 1: 316. The title
means, “Third contest problem: geometric puzzle.” Cited in 12.1.
———. 1931. Lingua Peano. Mody matematyk 1 (4–5): 49–53. Cited in 9.7.
Rusiecki, Antoni M., editor. 1931a. Dodatek. Mody matematyk 1 (3, March 1931): I–V (following
page 48). The title means “appendix.” This assigned point values to exercises Tarski [1930]
2014b, [1930] 2014c, and [1930] 2014f. Cited in 12.0.
Bibliography 451

———. 1931b. Solution to Tarski [1930] 2014b. Mody matematyk 1: 90. This issue of the jour-
nal was distributed with Parametr 2 (1931–1932, issue 6–7). The solution was unsigned. Cited
in 12.2.
———. 1939. Solution to Tarski [1930] 2014f. Parametr 3: 93–94. This was incorrectly labeled
as a solution to a different exercise. The solution was unsigned. Cited in 12.6.
Rusiecki, Antoni M., and Stefan Straszewicz, editors. 1930. Dzia zada; Apel; Od redakcji.
Parametr 1: 116, 229, 317, 323. The titles mean “Problem section,” “Appeal,” and “From the
editors.” The articles were unsigned. Cited in 9.7 and 12.0.
———. 1930–1931. Kursy wakacyjne. Parametr 1 (1930)(3, March), 111–112; (6, July–August),
224; (1931)(4–5, July), 138. The title means “Vacation courses.” The articles were unsigned.
Cited in 9.3.
———. 1931. Zadania. Mody matematyk 1: 13. The title means “Problems.” The article was
unsigned. Cited in 12.0.
———. 1931–1932. Od redakcji. Mody matematyk 1: 1–2, 97. The title means “From the
editors.” The article was unsigned. Cited in 9.7 and 12.0.
———. 1932a. Konkurs zadaniowy. Mody matematyk 1: 109–111, 154–155. The title means
“Problem contest.” The article was unsigned. Cited in 7.1 and 12.0.
———. 1932b. Od redakcji. Parametr 2: 33, 145. The title means “From the editors.” The article
was unsigned. Cited in 9.7.
Russell, Bertrand. 1903. The Principles of Mathematics. Cambridge, U. K.: Cambridge University
Press. JFM: Engel. There are several newer editions and reprints. Cited in 1.2.

Sadowska, Joanna. 2001. Ku szkole na miar Drugiej Rzeczypospolitej: geneza, zaoenia i

realizacja reformy Jdrzejewiczowskiej. Biaystok: Wydawnictwo Uniwersytetu w Biaymstoku.
The title means Toward Schools, for the Measure of the Second Republic: The Origin, Estab-
lishment, and Implementation of the Jdrzejewiczowski Reforms. Cited in 1.1.
Sally, Judith D., and Paul J. Sally, Jr. 2007. Roots to Research: A Vertical Development of Mathe-
matical Problems. Providence: American Mathematical Society. MR: Muldoon. Also reviewed
in Holdener 2009. Cited in 8.6.
San Francisco Chronicle. 1983 (28 October). Alfred Tarski: Expert Logician, UC Professor.
Cited in chapter 17.
San Francisco Examiner. 1983 (28 October). Alfred Tarski. Cited in chapter 17.
Scanlan, Michael. 2003. American postulate theorists and Alfred Tarski. History and Philosophy
of Logic 24: 307–325. Cited in 1.2 and 18.2.
Scazzieri, Roberto, and Raffaella Simili, editors. 2008. The Migration of Ideas. Sagamore Beach,
Massachusetts: Science History Publications. Contains Galavotti 2008.
Schayer, Wacaw. 1924. yciorys. Handwritten document in the Archive of the University of War-
saw, Sygnatura Schayer Wacaw: RP 17204. The title means “Curriculum Vitae.” Cited in 9.9.
———. 1932. Wykaz wykladów i wicze, pod L. 17204. In the Archive of the University of
Warsaw, Sygnatura Schayer Wacaw: RP 17204. The title phrase means List of Lectures and
Recitations. Schayer’s handwritten official record of studies, 1924–1932. Cited in 9.9.
Schinzel, Andrzej. 1974. Wacaw Sierpiski. Warsaw: Iskry. In Polish. Cited in chapter 3.
———. 2000. Pidziesit lat Olimpiady Matematycznej. Wiadomoci matematyczne 36: 155–161.
The title means “Fifty years of the Mathematical Olympiad.” Cited in 9.9.
———. 2009. The Lvov years of Wacaw Sierpiski. In Bojarski, awrynowicz, and Prytula 2009,
87–89. Cited in 4.1.
Schmidt am Busch, Hans-Christoph, and Kai F. Wehmeier. 2007. On the relations between Hein-
rich Scholz and Jan ukasiewicz. History and Philosophy of Logic 28: 67–81. Cited in 9.4.
Scholz, Heinrich. 1957. In memoriam Jan ukasiewicz. Archiv für mathematische Logik und
Grundlagenforschung 3: 3–18. Cited in 9.4.
452 Bibliography

Schönflies, Arthur. 1913. Entwicklung der Mengenlehre und ihrer Anwendung. Umarbeitung
des im VIII. Bande der Jahresbericht der Deutschen Mathematiker-Vereinigung erstatteten
Berichts. Erste Hälfte: Allgemeine Theorie der unendlichen Mengen und Theorie der Punkt-
mengen. Edited by Hans Hahn and Artur Schoenflies. Leipzig: B. G. Teubner. JFM: Vivanti.
Cited in 4.3.
Schreiber, Peter. 1996. Felix Hausdorffs paradoxe Kugelzerlegung im Kontext der Entwicklung
von Mengenlehre, Masstheorie und Grundlagen der Mathematik. In Brieskorn 1996, volume
1, 135–148. Cited in 8.6.
———. 1999. Über Beziehungen zwischen Heinrich Scholz und polnischen Logikern. History
and Philosophy of Logic 20: 97–109. MR: Glas. Cited in 9.4.
Schuster, Cynthia A. 1975. Hans Reichenbach. In Dictionary of Scientific Biography 1970–,
volume 11: 355–359. Cited in 16.1.
Schwabhäuser, Wolfram, Wanda Szmielew, and Alfred Tarski. 1983. Metamathematische Metho-
den in der Geometrie. Givant 1986 item 83 m. Berlin: Springer-Verlag. MR: Rautenberg. Cited
in 9.4 and 12.11.
Segerberg, Krister. 2009–. The logic of action. In Zalta 2011– at http://plato.stanford.edu/
entries/logic-action/. Cited in 15.2.
Skowska, Magdalena, and Wglowska, Danuta. 2003. Lindenbaum Adolf (1904–1941). In
Domoradzki, Pawlikowska-Broek, and Skowska 2003, 136. In Polish. Cited in 14.3.
Shilleto, James R. 2012. Personal communication to James T. Smith, June 2012. Cited in 7.1
and 8.5.
Shneiderman, S. L. 1995. Notes for an autobiography. Translated by Fannie Peczenik. College
Park, Maryland: University of Maryland Libraries, S. L. and Eileen Shneiderman Collection
of Yiddish Books. Cited in chapter 3.
Sica, Giandomenico, editor. 2005. Essays on the Foundations of Mathematics and Logic. Monza,
Italy: Polimeric International Scientific Publisher. Contains Rodríguez-Consuegra 2005. Cited
in 18.2.
Siegmund-Schultze, Reinhard. 2010. On a missed opportunity for collaboration between histori-
ans and mathematicians: a biographical avalanche triggered by Professor Ioan James, FRS.
Historia Mathematica 37: 693–707. Review of James 2009a and some other works. Cited in
chapter 17.
Sierpiski, Wacaw. 1918. Pewnik p. Zermelo oraz jego rola w teorji mnogoci i w analizie.—
L’axiome de M. Zermelo et son rôle dans la théorie des ensembles et l’analyse. Bulletin inter-
national de l’Académie des Sciences de Cracovie, classe des sciences mathématiques et
naturelles, série A: Sciences mathématiques: 97–152. JFM: Rosenblatt. In French. Reprinted
in Sierpiski 1974–1976, volume 2, 208–255. Cited in 4.1 and 6.3.
———. 1922. Sur l’égalité 2m = 2n pour les nombres cardinaux. Fundamenta Mathematicae
3: 1–6. JFM: Rosenthal. Reprinted in Sierpiski 1974–1976, volume 2, 417–421. Cited in 4.4.
———. 1923. Zarys teorji mnogoci. Volume 1: Liczby pozaskoczone. Second edition. Warsaw:
Mianowski Fund. JFM: Lichtenstein. Originally published in 1912. The titles mean Outline
of Set Theory and Transfinite Numbers. Sierpiski [1928] 1950 is a French version. Cited in
4.3 and chapter 5.
———. [1928] 1950. Leçons sur les nombres transfinis. Paris: Gauthier-Villars. New printing
of the original 1928 edition. French version of Sierpiski 1923. JFM: Feigl.
———. [1929] 1930. Przemówienie. In Leja 1930, 7–10. The title means “Address.” Cited in 9.5.
———. 1945. Sur le paradoxe de MM. Banach et Tarski. Fundamenta Mathematicae 33: 229–
234. MR: Blumberg. Reprinted in Sierpiski 1974–1976, volume 3, 439–444. Cited in 15.12.
———. 1947. Sur un théorème de A. Lindenbaum. Annals of Mathematics (series 2) 48: 641–642.
MR: Erdo  s. Reprinted in Sierpiski 1974–1976, volume 3, 469–471. Cited in 15.12.
Bibliography 453

———. 1954. On the Congruence of Sets and Their Equivalence by Finite Decomposition. Luck-
now University studies 20: Faculty of Science, session 1948–1949. Lucknow, India: Lucknow
University. MR: Halmos. Cited in 4.2 and 4.3.
———. 1974–1976. Oeuvres choisies. Three volumes. Edited by Stanisaw Hartman and Andrzej
Schinzel. Warsaw: PWN—Éditions Scientifiques de Pologne. Contains Kuratowski 1974 and
reprints of Sierpiski 1918 , 1922, 1945, and 1947, and Mazurkiewicz and Sierpiski 1914.
Simmons, Keith. 2009. Tarski’s logic. In Gabbay and Woods 2009, 511–616. Cited in 18.2.
Sinaceur, Hourya. 1996. Mathématiques et métamathématique du congrès de Paris (1900) au
congrès de Nice (1970): Nombres réels et théorie des modèles dans les travaux de Tarski.
Rendiconti del Circolo Matematico di Palermo (serie 2 supplemento) 44: 113–132. MR:
Murawski. Cited in 18.2.
———. 2001. Alfred Tarski: semantic shift, heuristic shift in metamathematics. Synthese 126:
49–65. MR: Rav. Cited in 18.2.
———. 2007. Review of Feferman and Feferman 2004. Notices of the American Mathematical
Society 54: 986–989. Cited in chapter 17.
———. 2009. Tarski’s practice and philosophy: between formalism and pragmatism. In Lind-
ström et al. 2009, 357–396. MR: Corcoran. Cited in chapter 17.
Singleterry, Ann M. 1966. Review of the third edition of Tarski [1936] 1995. Journal of Symbolic
Logic 31: 674–675.
Sinisi, Vito, and Jan Woleski, editors. 1995. The Heritage of Kazimierz Ajdukiewicz. Pozna
Studies in the philosophy of the sciences and the humanities 40. Amsterdam: Rodopi.
Contains Nowak 1995 and a reprint of Giedymin 1978.
Skolimowski, Henryk. 1962. Colourful philosopher. Polish Perspectives 5 (8–9): 122–126. Cited
in 14.2. About Leon Chwistek.
migielski, Henryk, editor. 1964. Matematyka i jej zastosowania. Zeszyty naukowe Politechniki
Warszawskiej 99. Warsaw: Wydawnictwa Politechniki Warszawskiej. The title means Mathe-
matics and Its Applications. Contains Jakubowski 1964.
Snyder, Timothy. 2010. Bloodlands: Europe between Hitler and Stalin. New York: Basic Books.
Cited in 9.9.
Sobociski, Bolesaw. 1956. In memoriam Jan ukasiewicz (1878–1956). Philosophical Studies:
Annual Journal of The Philosophical Society, St. Patrick’s College, Ireland 6: 3–49. Contains
an erroneous version of ukasiewicz [1953] 1994. Cited in 9.4.
Sobociski, Leon. 1935. W gimnazjum im. Mikoaja Kopernika w Toruniu. Gazeta Gdaska 14:
7–8. The title means “In the Nicolaus Copernicus Gimnazjum in Toru.” Cited in 7.1.
Sosnowski, Pawe. 1923. Pastwowy Instytut Pedagogiczny: sprawozdanie. Przegld pedago-
giczny 42(1): 74–83. The article title means “National Pedagogical Institute: report.” The
author was the director of the institute. Cited in chapter 3.
redniawa, Bronisaw. 1961. W dwudziest rocznic mierci prof. Witolda Wilkosza. Postpy
fizyki 13: 389–391. The title phrase and journal title mean “On the twentieth anniversary of
the death ...” and Advances in physics. Cited in 15.9.
rem. Pastwowe Gimnazjum im. Generaa Józefa Wybickiego. 1928, 1930, 1931. Sprawozdanie
dyrekcji Pastwowego Gimnazjum im. Generaa Józefa Wybickiego w remie. Volumes for aca-
demic years 1919/1920–1927/1928, 1929/1930, and 1930/1931. rem: Drukarnia „Gazety
Powszechnej“ w Poznaniu, Drukarnia Centralna w remie. The title means Report of the
directorate of the National Gimnazjum in rem named for General Józef Wybicki. The material
for 1928/1929 is contained in rem 1929. Cited in 7.1.
rem. Pastwowe Gimnazjum im. Generaa Józefa Wybickiego. 1929. Sprawozdanie dyrekcji
Pastwowego Gimnazjum im. Generaa Józefa Wybickiego w remie za 10-lecie 1918/19–
1928/29. rem: Drukarnia „Gazety Powszechnej“ w Poznaniu. The title means Report of the
454 Bibliography

directorate of the National Gimnazjum in rem named for General Józef Wybicki, for the ten
years 1918/1919–1928/1929. Overlaps rem 1928. Cited in 7.1.
Srzednicki, Jan T. J., and Zbigniew Stachniak, editors. 1998. Leniewski’s Systems: Protothetic.
Nijhoff international philosophy series 54. Dordrecht: Kluwer Academic Publishers. MR:
Murawski. Contains Tarski [1923] 1998.
Steinhaus, Hugo. 1924. O mierzeniu pól paskich. Przegld matematyczno-fizyczny 2: 24–29.
The title means “On measuring area in the plane.” Cited in 4.3.
Stone, John C., and Virgil S. Mallory. 1937. New Plane Geometry. Chicago: Benj. H. Sanborn
& Co. Cited in 9.9.
Stone, Marshall H. 1938. The representation of Boolean algebras. Bulletin of the American
Mathematical Society 44(1): 807–816. JFM: Rinow. Cited in 8.7.
Straszewicz, Stefan. 1930. Uwagi o t. zw. zadaniach dyskusyjnych w geometrji szkolnej. Para-
metr 1: 53–62. The title means “Remarks on the so-called geometrical discussion problems
in school.” Cited in 12.0.
Straszewicz, Stefan, and Stefan Kulczycki. [1935] 1948. Geometria dla III klasy gimnazjum.
Warsaw: Ksinica Atlas. The title means Geometry for the Third Gimnazjum Class. Cited
in 9.7 and 9.9.
Stromberg, Karl. 1979. The Banach–Tarski paradox. American Mathematical Monthly 86:
151–161. MR: Bressler. Cited in 4.4 and 8.6.
Suppes, Patrick. 1960. Axiomatic Set Theory. Princeton, New Jersey: D. Van Nostrand Company.
MR: Rieger. Cited in 1.2.
———. 1988. Philosophical implications of Tarski’s work. Journal of Symbolic Logic 53: 80–91.
MR: Kielkopf. Cited in 18.1.
Suppes, Patrick, Jon Barwise, and Solomon Feferman. 1989. Commemorative meeting for Alfred
Tarski: Stanford University, November 7, 1983. In Duren et al. 1989, 393–403. Cited in
chapter 17.
Surma, Stanisaw J. 1977. Mordchaj Wajsberg: life and work. In Wajsberg 1977, 7–11. Cited
in chapter 3.
Szczepaski, Janusz. 1995. Wojna 1920 roku na Mazowszu i Podlasiu. Warsaw: Putusk. The
title means The 1920 War in Mazovia and Podlachia. Cited in 1.1.
Szczerba, Lesaw W. 1986. Tarski and geometry. Journal of Symbolic Logic 51: 907–912. Cited
in 8.4, 9.3, 12.0, and 18.1.
Szkoa Ziemi Mazowieckiej. [1918] 1927. Document “8-mio Klasowa Szkoa Mazowiecka” dated
18 February 1927. In the Archive of the University of Warsaw, Sygnatura Teitelbaum-Tarski
Alfred: RP 2909. Certified handwritten copy of Alfred Tarski’s 8 June 1918 certificate of
graduation from the Mazowieckie gimnazjum, with a record of his spring 1918 grades. Cited
in 1.1.
Szmielew, Wanda. [1938] 1947. On choices from finite sets. Fundamenta Mathematicae 34:
75–80. MR: Martin. According to Kordos, Moszyska, and Szczerba 1977, these results were
obtained in 1938. Cited in 9.4 and 14.2.
———. 1955. Elementary properties of Abelian groups. Fundamenta Mathematicae 41: 203–271.
MR: Lyndon. Cited in 9.4.
Szudy, Józef, editor. 1998. Born 100 Years Ago: Aleksander Jaboski (1898–1980). Toru: Uni-
wersytet Mikoaja Kopernika. Contains Frckowiak 1998.
Szumilewicz-Lachman, Irena. 1994. Zygmunt Zawirski: a short bio-bibliography (28.IX.1882–
2.IV.1948). Translated by Feliks Lachman. In Szumilewicz-Lachman, Cohen, and Bergo 1994,
Part 1, chapter 5, 33–80. Cited in 15.10.
Szumilewicz-Lachman, Irena, Robert S. Cohen, and Bettina Bergo, editors. 1994. Zygmunt Zawir-
ski: His Life and Work, with Selected Writings on Time, Logic, and the Methodology of Science.
Bibliography 455

Boston studies in the philosophy of science 157. Dordrecht: Kluwer Academic Publishers.
Contains Szumilewicz-Lachman 1994.
Szybalski, Wacaw. 1999. Maintenance of human-fed live lice in the laboratory and production
of Weigl’s exanthematous typhus vaccine. In Maramorosch and Mahmood 1999, chapter 8,
161–180. Cited in 2.1.

Tarski, Alfred. See also joint entries after the following single-author entries, and under Banach,
Chwiakowski, Doner, Henkin, Horn, Jónsson, Lindenbaum, ukasiewicz, Mates, Montague,
and Mostowski.
———. [1918] 2014. Curriculum Vitae. In section 16.1 of the present book, Tarski 2014. Trans-
lation of the handwritten document “ yciorys” in the Archive of the University of Warsaw,
Sygnatura Teitelbaum-Tarski Alfred: RP 2909. Cited in 1.1, chapter 3, and 16.3.
———. 1919–1920. Two documents titled “Do jego Magnificencji Rektora Uniwersytetu
Warszawskiego,” dated trymestrze I 1919/20 and zimowym 1920/21. Printed forms filled out
by hand, in the Archive of the University of Warsaw, Sygnatura Teitelbaum-Tarski Alfred: RP
2909. The titles mean “To his Magnificence, the Rector of the University of Warsaw,” autumn
trimester 1919/1920 and autumn 1920/1921.” Tarski crossed out unused portions, signed them
Alfred Tajtelbaum, and submitted them. They are about returning to the university after
summer military leave. Cited in 1.1 and chapter 3.
———. [1921] 2014. A contribution to the axiomatics of well-ordered sets. Chapter 2 of the
present book, Tarski 2014. Translation of “Przyczynek do aksjomatyki zbioru dobrze uporzd-
kowanego,” Przegld filozoficzny 24, 85–94, Givant 1986 item 21, reprinted in Tarski 1986a,
volume 1, 1–12. JFM: Chwistek. The author’s surname on the original is Tajtelbaum; its title
acknowledges the seminar of Prof. Stanisaw Leniewski at the University of Warsaw. Cited
in 1.2, 8.2, 16.0, and 16.1.
———. 1921–1926. Do kwestury, letni 1921, zimowy 1921/2, zimowy 1923/4, 1 czerwiec 1924,
27 stycze 1925, 23 stycze 1926. Printed forms filled out by hand, in the Archive of the
University of Warsaw, Sygnatura Teitelbaum-Tarski Alfred: RP 2909. The title means “To the
bursar” with the dates. Receipts for Alfred Tarski’s tuition (sometimes waived). The docu-
ment dated zimowy 1923/4 (autumn 1923/1924) is reproduced in chapter 3 of the present book,
Tarski 2014. Cited in chapter 3.
———. 1923a. O wyrazie pierwotnym logistyki. Przegld filozoficzny 26: 68–89. Givant 1986
item 23. Tarski’s doctoral dissertation. The author’s surname is listed as Tajtelbaum-Tarski.
JFM: Rosenthal. Translated into English as Tarski [1923] 1998. Adapted in French as Tarski
1923b and 1924d. Adapted and translated into English in Tarski [1956] 1983, 1–23. Cited in
chapter 3 and 16.1.
———. 1923b. Sur le terme primitif de la logistique. Fundamenta Mathematicae 4: 196–200.
Givant 1986 item 23a. Adaptation of part of Tarski 1923a (the other part is adapted in Tarski
1924d). JFM: Rosenthal. Reprinted in Tarski 1986a, volume 1, 13–20. Cited in chapter 3
and 16.1.
———. [1923] 1998. On the primitive term of logistic. Translation of the Polish original, Tarski
1923a, by Zbigniew Stachniak. In Srzednicki and Stachniak 1998, 43–68. The author’s sur-
name is listed as Tajtelbaum-Tarski. Cited in 16.1.
———. 1924a. Une remarque concernant les principes d’arithmétique théorique. Annales de
la Société Polonaise de Mathématique 3: 150. Givant 1986 item 24a a. Reprinted in Tarski
1986a, volume 4, 535. Cited in 4.3.
———. 1924b. Sur les ensembles finis. Fundamenta Mathematicae 6: 45–95. Givant 1986 item
24c. JFM: Rosenthal. Reprinted in Tarski 1986a, volume 1, 65–118. Cited in chapter 3, 4.3,
4.4, 8.3, 9.3, and 14.2.
456 Bibliography

———. 1924c. Sur les principes de l’arithmétique des nombres ordinaux (transfinis). Annales
de la Société Polonaise des Mathématique 3: 148–149. Givant 1986 item 24a. Reprinted in
Tarski 1986a, volume 4, 533–534. Cited in 1.1, 4.3, and 8.2.
———. 1924d. Sur les truth-functions au sens de MM. Russell et Whitehead. Fundamenta Math-
ematicae 5: 59–74, Givant 1986 item 24. Adaptation of part of Tarski 1923a (the other part
is adapted in Tarski 1923b). JFM: Rosenthal. Reprinted in Tarski 1986a, volume 1, 21–38.
The author’s surname is listed as Tajtelbaum-Tarski. Cited in chapter 3 and 4.3.
———. 1924e. Sur quelques théorèmes qui équivalent à l’axiome du choix. Fundamenta Math-
ematicae 5: 147–154. Givant 1986 item 24a (which has incorrect publication data). JFM:
Rosenthal. Reprinted in Tarski 1986a, volume 1, 39–48. The author’s surname is listed as
Tajtelbaum-Tarski. Cited in chapter 3, 4.3, 4.4, 8.3, 9.4, 14.2, and 15.12.
———. 1924f. Wykaz wykadów i wicze, no. 2909. In the Archive of the University of Warsaw,
Sygnatura Teitelbaum-Tarski Alfred: RP 2909. The title phrase means List of Lectures and
Recitations. Tarski’s handwritten official record of studies, 1918–1924. Cited in 1.1, chapter
3, and 4.1.
———. [1924] 2004. Letter to a Warsaw city government office, 18 March. In Feferman and
Feferman 2004, 40. The original handwritten document “Komisariat Rzdu na m. st.
Warszawa” is in the Archive of the University of Warsaw, Sygnatura Teitelbaum-Tarski Alfred:
RP 2909. Cited in chapter 3 and 16.3.
———. [1924] 2014a. Letter to the dean of the Philosophical Faculty, 2 June 1924. In section
16.3 of the present book, Tarski 2014. Translation of handwritten document “Do Pana Dzie-
kana Wydziau Filozoficznego Uniwersytetu Warszawskiego” in the Archive of the University
of Warsaw, Signatura Teitelbaum-Tarski Alfred: RP 2909. Cited in chapter 3.
———. [1924] 2014b. On the equivalence of polygons. Chapter 5 of the present book, Tarski
2014. Translation of “O równowanoci wieloktów,” Przegld matematyczno-fizyczny 2:
47–60, Givant 1986 item 24b, reprinted in Tarski 1986a, volume 1, 49–64. Cited in 4.0, ... , 4.4,
6.0, 8.4, 8.6, 9.2, 9.3, 9.7, 9.8, and 15.7.
———. 1925a. Problème 38. Fundamenta Mathematicae 7: 381. Givant 1986 item 25p. The
wrong text was printed for this problem in Tarski 1986a; the original text is reproduced in
section 4.3 of the present book, Tarski 2014. Cited in 4.3 and 12.0.
———. 1925b. Quelques théorèmes sur les alephs. Fundamenta Mathematicae 7: 1–14. Givant
1986 item 25, reprinted in Tarski 1986a, volume 1, 155–170. JFM: Pannwitz. Cited in 8.3
and 9.4.
———. [1925] 2014. Discussion of ukasiewicz 1925. In section 15.1 of the present book, Tarski
2014. Translation of a discussion in ukasiewicz 1925, 136: Givant 1986 item 25c, reprinted
in Tarski 1986a, volume 4, 686, and in Jadacki 2003b, 19. Cited in 8.1.
———. 1927a. Sur quelques propriétés caractéristiques des images d’ensembles. Annales de la
Société Polonaise de Mathématique 6: 127–128. Givant 1986 item 27a. JFM: Hurewicz.
Reprinted in Tarski 1986a, volume 4, 541–543. Abstract of an 18 February talk presented to
the Warsaw section of the Polish Mathematical Society. Cited in 8.8.
———. 1927b. Quelques théorèmes généraux sur les images d’ensembles. Annales de la Société
Polonaise de Mathématique 6: 132–133. Givant 1986 item 27aa. JFM: Hurewicz. Reprinted
in Tarski 1986a, volume 4, 544–545. Abstract of a 9 December talk presented to the Warsaw
section of the Polish Mathematical Society. Cited in 8.8.
———. [1927] 1929. Les fondements de la géométrie des corps. In Polski Zjazd Matematyczny
1929, 54–61. Givant 1986 item 29. (Givant reported the pages incorrectly.) Reprinted in
Tarski 1986a, volume 1, 225–232. Tarski [1927] 1983 is an English translation and adaptation.
———. [1927] 1983. Foundations of the geometry of solids. In Tarski [1956] 1983, 24–29. Givant
1986 item 29(1). Translated and adapted from Tarski [1927] 1929. Cited in 2.1, 9.4, and 9.7.
Bibliography 457

———. [1927] 2014a. Discussion of Greniewski [1927] 1928. In section 15.2 of the present book,
Tarski 2014. Translation of a discussion in Greniewski [1927] 1928, 160: Givant 1986 item
28c, reprinted in Tarski 1986a, volume 4, 687, and in Jadacki 2003b, 19.
———. [1927] 2014b. Discussion of Czeowski [1927] 1928. In section 15.3 of the present book,
Tarski 2014. Translation of a discussion in Czeowski [1927] 1928, 167: Givant 1986 item 28ca,
reprinted in Jadacki 2003b, 19, and in Tarski 1986a, 549.
———. 1928. Remarques sur les notions fondamentales de la méthodologie des mathématiques.
Annales de la Société Polonaise de Mathématique 7: 270–273. Givant 1986 item 28a b. Report
of a 14 September 1928 presentation in Warsaw. The journal gave that date incorrectly as
1929. Reprinted in Tarski 1986a, volume 4, 552. Tarski [1928] 2010 is an English translation.
Cited in 9.5 and 16.1.
———. [1928] 1930. Über Äquivalenz der Mengen in Bezug auf eine beliebige Klasse von Abbild-
ungen. In International Congress of Mathematicians 1930, volume 2, 243–252. JFM: Fraenkel.
Givant 1986 item 30b, reprinted in Tarski 1986a, volume 1, 299–310. Cited in 8.8.
———. [1928] 2010. Remarks on fundamental concepts of the methodology of mathematics.
Translated by Robert Purdy and Jan Zygmunt. In Béziau 2010, 67–68. Translation of Tarski
1928. Cited in 16.1.
———. [1929] 2014a. An assembly of mathematicians. Chapter 10 of the present book, Tarski
2014. Translation of “Zjazd matematyków,” Ogniwo: organ Zwizku Zawodowego Nauczy-
cielstwa Polskich Szkó rednich i biuletyn zarzdu gównego Z. Z. N. P. S. r. 9: 401–402,
Givant 1986 item 29e, reprinted in Tarski 1986a, volume 1, 251–252. The journal title means
Organ of the Trade Union of Polish Secondary School Teachers and Bulletin of the General
Administration (of the same organization). Cited in 9.4 and 9.5.
———. [1929] 2014b. Marginal note on “Decree of the President of the Republic Concerning
Insurance of Nonmanual Workers as of 24 November 1927” In section 15.5 of the present book,
Tarski 2014. Translation of “Na marginesie ‘Rozporzdzenia Prezydenta Rzeczypospolitej o
ubezpieczeniu pracownikow umysowych z dnia 24 listopada 1927 r,’ ” Ekonomista, kwar-
talnik powicony nauce i potrzebom ycia 29: 115–119, Givant 1986 item 29d, reprinted
in Tarski 1986a, volume 1, 269–273. The decree is Poland President 1927. The journal title
means The Economist: A Quarterly Devoted to Science and the Needs of Life. Cited in 9.2, 15.0,
and 15.5.
———. 1929–1930. Sur les fonctions additives dans les classes abstraites et leur application au
problème de la mesure. Comptes rendus des séances de la Société des Sciences et des Lettres
de Varsovie, classe III 22: 114–117. JFM: Ruziewicz. Givant 1986 item 29b, reprinted in Tarski
1986a, volume 1, 243–248. Cited in 8.6.
———. 1930a. Une contribution à la théorie de la mesure. Fundamenta Mathematicae 15:
42–50. Givant 1986 item 30, reprinted in Tarski 1986a, volume 1, 275–286. JFM: Feigl. Cited
in 2.1.
———. 1930b. Über definierbare Zahlenmengen. Annales de la Société Polonaise de Mathéma-
tique 9: 206–207. Givant 1986 item 30 a. Reprinted in Tarski 1986a, volume 4, 560–561.
Preliminary version of Tarski [1931] 1983. Cited in 8.6 and 14.2.
———. [1930] 1983a. Fundamental concepts of the methodology of the deductive sciences, I.
In Tarski [1956] 1983, 60–109. The original paper, “Fundamentale Begriffe der Methodologie
der deduktiven Wissenschaften,” Monatshefte für Mathematik und Physik 37: 361–404, Givant
1986 item 30e, JFM: Skolem, is reprinted in Tarski 1986a, volume 1, 345–390, and translated
into Polish in Tarski 1995–2001, volume 2, 31–92. There is no part II. Cited in 14.2, 15.6,
and 16.1.
———. [1930] 1983b. On some fundamental concepts of metamathematics. In Tarski [1956]
1983, 30–37. The original paper, “Über einige fundamentale Begriffe der Metamathematik,”
Comptes rendus de la Société des Sciences et des Lettres de Varsovie, Classe III 23: 22–29,
458 Bibliography

Givant 1986 item 30c, is reprinted in Tarski 1986a, volume 1, 311–320. Cited in chapter 3, 8.1,
14.2, and 15.6.
———. [1930] 2014a. Exercise: On diluting wine. In section 12.1 of the present book, Tarski
2014. Translation of “Zadanie o rozcieczaniu wina,” Parametr 1: 229, Givant 1986 item 30p,
reprinted in Tarski 1986a, volume 4, 688.
———. [1930] 2014b. Exercise 83: Iteration of the absolute value symbol. In section 12.2 of the
present book, Tarski 2014. Translation of “Nr. 83: Iteracja znaku bezwzgldnej wartoci,”
Parametr 1: 231, Givant 1986 item 30pa, reprinted in Tarski 1986a, volume 4, 688. Rusiecki
1931a assigned it a point value and Rusiecki 1931b is a solution.
———. [1930] 2014c. Exercise 88: Decomposition into factors. In section 12.3 of the present
book, Tarski 2014. Translation of “Nr. 88: Rozkad na czynniki,” Parametr 1: 277, Givant 1986
item 30p b, reprinted in Iwaszkiewicz 1949b, with a solution, and in Tarski 1986a, volume 4,
689. Rusiecki 1931a assigned it a point value. Cited in 9.7.
———. [1930] 2014d. Exercise 102: Interesting identity. In section 12.4 of the present book,
Tarski 2014. Translation of “No. 102: Ciekawa tosamo ,” Parametr 1: 278, Givant 1986 item
30pc, reprinted in Tarski 1986a, volume 4, 689.
———. [1930] 2014e. Exercise 118. In section 12.5 of the present book, Tarski 2014. Translation
of “Nr. 118,” Parametr 1: 318, Givant 1986 item 30pd, reprinted in Tarski 1986a, volume 4,
689–690.
———. [1930] 2014f. Exercise 136: Equidistant. In section 12.6 of the present book, Tarski
2014. Translation of “Nr. 136: Ekwidystanta,” Parametr 1: 398, Givant 1986 item 30pe,
reprinted in Tarski 1986a, 690. Rusiecki 1931a assigned it a point value. Rusiecki 1939 is a
solution.
———. [1930–1931] 2014. On the concept of truth in reference to formalized deductive sciences.
In section 15.6 of the present book, Tarski 2014. Translation of “O pojciu prawdy w odnie-
sieniu do sformalizowanych nauk dedukcyjnych,” Ruch filozoficzny 12: 210–211, Givant 1986
item 30/31a, reprinted in Tarski 1986a, volume 4, 555–559, and Tarski 1995–2001, volume 1,
3–8. Abstract. Cited in 8.1, 15.0, and 16.2.
———. [1930–1936] 1992. Drei Briefe an Otto Neurath. Edited, with a foreword, by Rudolf
Haller. Translated by Jan Tarski. Grazer philosophische Studien 43:1–32. Partially translated
into Polish in Tarski 1995–2001, volume 1, 206–213. Cited in 15.1 and 16.3.
———. 1931. Neue Resultate und unentschiedene Probleme der Kardinalzahlarithmetik.
Monatshefte für Mathematik und Physik 38: 23. Givant 1986 item 31a. Reprinted in Ergebnisse
eines mathematischen Kolloquiums 2 (1932), 12–13; in Menger et al. 1998, 130–131; and in
Tarski 1986a, volume 4, 562–563. Report of a 19 February 1930 presentation in Vienna. Cited
in 9.5.
———. [1931] 1951. Al ma’alat shivyon-ha-peruk shel metsula (   r –  
). Riveon lematematika 5(1951): 32–38. Hebrew translation by Dov Jarden of the original
version of Tarski [1931] 2014a. Givant 1986 item 31b(2). MR: Jerison. Cited in 7.1 and 8.5.
———. [1931] 1952. The degree of equivalence of polygons. In Tarski and Moese 1952, 1–8.
Givant 1986 item 31b(1). Draft translation, by Isaak Wirszup, of the original version of Tarski
[1931] 2014a. Reprinted in Tarski 1986a, volume 1, 571–580.
———. [1931] 1983. On definable sets of real numbers. In Tarski [1956] 1983, chapter VI,
110–142. Translation of “Sur les ensembles définissables de nombres réels, I,” Fundamenta
Mathematicae 17: 210–239. Givant 1986 item 31, reprinted in Tarski 1986a, vol. 1, 517–548.
JFM: Fraenkel. Cited in 8.6 and 15.6.
———. [1931] 2014a. On the degree of equivalence of polygons. Section 7.2 of the present book,
Tarski 2014. Translation of “O stopniu równowanoci wieloktów,” Mody matematyk 1
(1931–1932): 37–44, Givant 1986 item 31b, reprinted in 1975 in Delta 5 (4): 3–6. Tarski [1931]
1952 is a draft translation of the original paper. The original and that draft were reprinted
Bibliography 459

in Tarski 1986a, volume 1, 561–580. Tarski [1931] 1951 is a Hebrew translation. Cited in 4.3,
7.0, ... , 7.4, 8.5, 9.2, 9.7, 9.8, 12.7, and 12.10.
———. [1931] 2014b. Exercise 167: System of inequalities. In section 12.7 of the present book,
Tarski 2014. Translation of “Nr. 167: Ukad nierównoci,” Mody matematyk 1 (1931–1932):
46, Givant 1986 item 31p, reprinted in Tarski 1986a, volume 4, 692.
———. [1931] 2014c. Exercise 168: [Another] system of inequalities. In section 12.8 of the pres-
ent book, Tarski 2014. Translation of “Nr. 168: Ukad nierównoci,” Mody matematyk 1
(1931–1932): 46, Givant 1986 item 31pa, reprinted in Tarski 1986a, volume 4, 692.
———. [1931] 2014d. Exercise 169: Cutting a rectangle out of a square. In section 12.9 of the
present book, Tarski 2014. Translation of “Nr. 169: Wycicie prostokta z kwadratu,” Mody
matematyk 1 (1931–1932): 46, Givant 1986 item 31p b, reprinted in Dubiel 1975 and Tarski
1986a, volume 4, 692.
———. [1931] 2014e. Exercise 170: A theorem on the degree of equivalence of polygons. In sec-
tion 12.10 of the present book, Tarski 2014. Translation of “Nr. 170: Twierdzenie o stopniu
równowanoci wieloktów,” Mody matematyk 1 (1931–1932): 46, Givant 1986 item 31pc,
reprinted in Tarski 1986a, volume 4, 693. Cited in 4.3.
———. [1931] 2014f. Exercise 183: Postulate about parallels. In section 12.11 of the present
book, Tarski 2014. Translation of “No. 183: Postulat o równolegych,” Parametr 2 (1931–1932):
78, Givant 1986 item 31/32 p, reprinted in Tarski 1986a, volume 4, 691. Cited in 9.3 and 9.4.
———. [1931] 2014g. Exercise 186: Analytic geometry of space. In section 12.12 of the present
book, Tarski 2014. Translation of “No. 186: Zadanie z geometrji analitycznej w przestrzeni,”
Parametr 2 (1931–1932): 78, Givant 1986 item 31/32 p a, reprinted in Tarski 1986a, vol-
ume 4, 691.
———. [1931] 2014h. On a certain system of mathematical logic and the resulting methodological
and semantic issues. In section 16.1 of the present book, Tarski 2014. Translation of “O
pewnym systemie logiki matematycznej i wynikajcych z niego zagadnieniach metodologicznych
i semantycznych,” Ruch filozoficzny 12 (1930–1931): 232. Abstract of a 15 April 1931 presenta-
tion to the Logic Section of the Warsaw Philosophical Institute. Cited in 14.2.
———. 1932. Der Wahrheitsbegriff in den Sprachen der deduktiven Disziplinen. Akademie der
Wissenschaften in Wien, mathematisch-naturwissenschaftliche Klasse, akademischer Anzeiger
69: 23–25. Givant 1986 item 32. JFM: Ackermann. Reprinted in Tarski 1986a, volume 1,
613–618. Presented to the Akademie by Hans Hahn. Translated into Polish in Tarski
1995–2001, volume 1, 9–12. Cited in 14.2 and 15.6.
———. [1932] 1952. Further remarks about the degree of equivalence of polygons. In Tarski
and Moese 1952, 14–20. Givant 1986 item 31/32a(1). Tarski [1932] 1952 is a draft translation
of the original paper. Reprinted in Tarski 1986a, volume 1, 603–612.
———. [1932] 2014a. On geometric properties of Banach’s measure. In section 15.7 of the
present book, Tarski 2014. Translation of “O wasnociach geometrycznych miary Banacha,”
Comptes rendus des séances de la Société des Sciences et des Lettres de Varsovie, classe III 25
(1932–1933): 12, Givant 1986 item 32a. Presented to the Society by Wacaw Sierpiski on 23
February 1932. Reprinted in Tarski 1986a, volume 4, 567. Cited in 8.7 and 15.0.
———. [1932] 2014b. Exercise 213: Stereometry. In section 12.13 of the present book, Tarski
2014. Translation of “No. 213: Zadanie ze stereometrji,” Parametr 2 (1931–1932): 207, Givant
1986 item 31/32 p b, reprinted in Tarski 1986a, volume 4, 691.
———. [1932] 2014c. Exercise 214: Arrangement of segments in a plane. In section 12.14 of
the present book, Tarski 2014. Translation of “No. 214: Ukad dwóch odcinków na pasz-
czynie,” Parametr 2 (1931–1932): 207, Givant 1986 item 31/32 pc, reprinted in Tarski 1986a,
volume 4, 691.
———. [1932] 2014d. Remarks on the degree of equivalence of polygons. Section 7.4 of the pres-
ent book, Tarski 2014. Translation of “Uwagi o stopniu równowanoci wieloktów,” Parametr
460 Bibliography

2 (1931–1932): 310–314, Givant 1986 item 31/32a. A draft translation by Izaak Wirszup
appeared in Tarski and Moese 1952, 15–20, Givant 1986 item 31/32a(1). The original and that
draft were reprinted in Tarski 1986a, volume 1, 595–612. Cited in 4.3, 7.0, ... , 7.4, 8.5, 9.7,
and 9.8.
———. [1932] 2014e. The theory of the circumference of a circle in the secondary school.
Chapter 11 of the present book, Tarski 2014. Translation of “Teorja dugoci okrgu w szkole
redniej,” Parametr 2 (1931–1932): 257–267, Givant 1986 item 31/32, reprinted in Tarski 1986a,
volume 1, 581–594. Cited in 9.7, ... , 9.9.
———. 1933. Pojcie prawdy w jzykach nauk dedukcyjnych. Prace Towarzystwa Naukowego
Warszawskiego, Wydzia III—Nauk Matematyczno-fizycznych 34. Givant 1986 item 33 m. The
title means The Concept of Truth in the Languages of Deductive Sciences. Reprinted in Tarski
1995–2001, volume 1, 13–172. Tarski [1933] 1983 is an English translation of the German
translation [1933] 1935 which had an added appendix. Tarski [1930–1931] 2014 is a translation
of a preliminary version. Cited in 9.3, 15.6, 15.12, 16.2, and 16.3.
———. [1933] 1935. Der Wahrheitsbegriff in den formalisierten Sprachen. Translated by
Leopold Blaustein. With a postscript by Alfred Tarski. Studia Philosophica 1: 261–405.
Translation of Tarski 1933. Givant 1986 item 35b. JFM: Ackermann. Reprinted in Tarski
1986a, volume 2, 51–198. Cited in 15.6 and 15.12.
———. [1933] 1983. The concept of truth in formalized languages. Translated by J. H. Woodger.
In Tarski [1956] 1983, chapter 8, 152–278. Givant 1986 item 35b(1). Translation of Tarski
[1933] 1935, which was a translation of Tarski 1933. Cited in 9.3, 15.6, 15.10, and 15.11.
———. [1933] 2003. Letter to Kazimierz Twardowski, 12 July. In Jadacki 2003b, 19. Cited
in 16.3.
———. 1934. Z bada metodologicznych nad definjowalnoci terminów. Przegld filozoficzny
37: 438–460. Givant 1986 item 34. Reprinted in Tarski 1995–2001, volume 2, 114–146.
Tarski [1935–1936] 1983 is an English translation of a German translation. Cited in 16.2.
———. [1934] 1936a. Discussion of Reichenbach [1934] 1936 and Zawirski [1934] 1936.
In International Congress of Philosophy 1936, 197–198. Givant 1986 item 36ca. (Givant
incorrectly described that item as a contribution to a discussion of a paper by Janina Hosiasson-
Lindenbaumowa.) Reprinted in Tarski 1986a, volume 4, 695. Reprinted and translated into
Polish in Tarski [1934] 2003a. The original text of this discussion is nearly identical to that
of Tarski 1935–1936. Cited in 16.1.
———. [1934] 1936b. Discussion of Jørgensen [1934] 1936. In International Congress of
Philosophy [1936] 1968, 119. Givant 1986 item 36c. Reprinted in Tarski 1986a, volume 4, 694.
Reprinted and translated into Polish in Tarski [1934] 2003b. Cited in 16.1.
———. [1934] 1983. Some observations on the concepts of Z-consistency and Z-completeness.
In Tarski [1956] 1983, 279–295. The original paper, “Einige Betrachtungen über die Begriffe
der Z-Widerspruchsfreiheit und Z-Vollständigkeit,” Monatshefte für Mathematik und Physik
40: 97–112, Givant 1986 item 33, JFM: Fraenkel, is reprinted in Tarski 1986a, volume 1,
619–636, and translated into Polish in Tarski 1995–2001, volume 2, pages 93–113. Cited
in 16.2.
———. [1934] 2003a. Discussion of Reichenbach [1934] 1936 and Zawirski [1934] 1936. With
Polish translation by Mariusz Grygianiec. In Jadacki 2003b, 12–14. Reprint and translation
of Tarski [1934] 1936a. Cited in 16.1.
———. [1934] 2003b. Discussion of Jørgensen [1934] 1936. With Polish translation by
Mariusz Grygianiec. In Jadacki 2003b, 11. Reprint and translation of Tarski [1934] 1936b.
Cited in 16.1.
———. [1934] 2014. Letter to the Minister of Religious Denominations and Public Education,
29 November. Translation of a handwritten document N921/R15345, page 8, preserved in the
archival fund No. 14, Ministerstwo Wyzna Religijnych i Owiecenia Publicznego w Warszawie
Bibliography 461

1917–1939, sygnatura 6216, of the Archiwum Akt Nowych (Central Archives of Modern
Records) in Warsaw. In section 16.3 of the present book, Tarski 2014. The original document
was transcribed in Jadacki 2003b, 25–26. Cited in 9.1 and 16.3.
———. [1935] 1936a. Grundlegung der wissenschaftlichen Semantik. In International Congress
for the Unity of Science 1936, volume 3, 1–8. Givant 1986 item 36f. German version of Tarski
1936a. Reprinted in Tarski 1986a, volume 2, 259–268. Tarski [1935] 1983 is an English
translation. Cited in 14.2.
———. [1935] 1936b. Über den Begriff der logischen Folgerung. In International Congress for
the Unity of Science 1936, volume 7, 1–11. Givant 1986 item 36g. German version of Tarski
1936b. Reprinted in Tarski 1986a, volume 2, 269–282. Tarski [1935] 2002 and [1956] 1983,
409–430, are English translations. Tarski [1935] 1984 is a Spanish translation. Cited in 14.2
and 16.1.
———. [1935] 1983. The establishment of scientific semantics. In Tarski [1956] 1983, chapter
15, 401–408. Givant 1986 item 36f(1). Translation of Tarski [1935] 1936a.
———. [1935] 1984. Sobre el concepto de consecuencia logica. Translation of Tarski [1935]
1936b by Luiz Vega Reñon. In Castrillo and Vega 1984, section A.2, 178–192. Cited in 16.1.
———. [1935] 2002. On the concept of following logically. Translated from the Polish and
German originals Tarski [1935] 1936b and 1936b by Magda Stroiska and David Hitchcock,
with their commentaries. History and Philosophy of Logic 23: 155–196. MR: Murawski. Cited
in the preface and 16.1.
———. [1935] 2014. Two letters to the Ministry of Religious Denominations and Public
Education, 25 November and 7 December. In section 16.3 of the present book, Tarski 2014.
Translations of handwritten documents, N840/R12478, pages 21–22, and N873/R13421, page
24, preserved in the archival fund No. 14, Ministerstwo Wyzna Religijnych i Owiecenia
Publicznego w Warszawie 1917–1939, sygnatura 6216, Archiwum Akt Nowych (Central
Archives of Modern Records) in Warsaw. Transcriptions of the original letters were published
in Jadacki 2003b, 26–27. Jadacki misstated the date of the first letter. Cited in 9.1 and 16.3.
———. 1935–1936. Wahrscheinlichkeitslehre und mehrwertige Logik. Erkenntnis 5: 174–175.
Givant 1986 item 35d. The text of this report is nearly identical to that of Tarski [1934] 1936a.
Reprinted in Tarski 1986a, volume 2, 199–202. Cited in 16.1.
———. [1935–1936] 1983. Some methodological investigations on the definability of concepts.
In Tarski [1956] 1983, 296–319. JFM: Hempel. Givant 1986 item 35c(1). The original paper,
“Einige methodologische Untersuchungen über die Definierbarkeit der Begriffe,” Erkenntnis
5: 80–100, Givant 1986 item 35c, is reprinted in Tarski 1986a, volume 1, 637–659. That, in
turn, was a translation of Tarski 1934.
———. 1936a. O ugruntowaniu naukowej semantyki. Przegld filozoficzny 39: 50–57. Givant
item 36. Reprinted in Tarski 1995–2001, volume 1, 173–185. Tarski [1935] 1983 is an English
translation of the German version Tarski [1935] 1936a. Cited in 14.3 and 16.2.
———. 1936b. O pojciu wynikania logicznego. Przegld filozoficzny 39: 58–68. Givant 1986
item 36a. Reprinted in Tarski 1995–2001, volume 1, 186–202. Tarski [1935] 2002 is an
English translation. In Tarski [1956] 1983, 409–420, is an English translation of the German
version Tarski [1935] 1936b. Cited in 14.3, 16.1, and 16.2.
———. [1936] 1995. Introduction to Logic and to the Methodology of the Deductive Sciences.
Second, revised edition. Translated by Olaf Helmer. New York: Dover Publications. Originally
published in 1941 by Oxford University Press, with second and third editions in 1946 and 1965.
Tarski 1994 is a fourth edition. Item 41m in Givant 1986. MR: Frink. Also reviewed in Church
1941–1947 and Singleterry 1966. Expansion and translation of O logice matematycznej i meto-
dzie dedukcyjnej, Bibljoteczka matematyczna 3-5, Lwów and Warsaw: Ksinica–Atlas, item
36m in Givant 1986, JFM: Ackermann, also reviewed in Mac Lane 1938. The original title means
On Mathematical Logic and the Deductive Method. Cited in 9.3, 9.9, 14.2, and 14.3.
462 Bibliography

———. [1936] 2014a. Discussion of Ajdukiewicz 1936. In section 15.8 of the present book, Tarski
2014. Translation of a discussion in Ajdukiewicz 1936, 337, Givant 1986 item 36c b, reprinted
in Tarski 1986a, volume 4, 698, and in Jadacki 2003b, 15.
———. [1936] 2014b. Discussion of Wilkosz 1936. In section 15.9 of the present book, Tarski
2014. Translation of a discussion in Wilkosz 1936, 346–347, Givant 1986 item 36c, reprinted
in Tarski 1986a, volume 4, 699, and in Jadacki 2003b, 15.
———. [1936] 2014c. Discussion of Zawirski 1936. In section 15.10 of the present book, Tarski
2014. Translation of a discussion in Zawirski 1936, 350–351, Givant 1986 item 36c d, reprinted
in Tarski 1986a, volume 4, 700, and in Jadacki 2003b, 15–16.
———. [1936] 2014d. Discussion of Kokoszyska 1936a. In section 15.11 of the present book,
Tarski 2014. Translation of a discussion in Kokoszyska 1936a, 425, Givant 1986 item 36c e,
reprinted in Tarski 1986a, volume 4, 701, and Tarski 1995–2001, volume 1, 203–205. Cited
in 16.2.
———. 1937a. Appendix E. In Woodger 1937, 161–172. Givant 1986 item 37b. Reprinted in
Tarski 1986a, volume 2, 335–350. Tarski’s first publication in English. Cited in 14.3.
———. 1937b. Sur la méthode déductive. In International Congress of Philosophy 1937, fascicle
6: Logique et mathématique, 95–103. Givant 1986 item 37a, reprinted in Tarski 1986a, volume
2, 325–333. Also reviewed in Langford 1938. Tarski’s last publication in French. Cited in 14.2
and 14.3.
———. 1937c. Über additive und multiplikative Mengenkörper und Mengenfunktionen. Comptes
rendus des séances de la Société des Sciences et des Lettres de Varsovie, classe III 30: 151–181.
Givant 1986 item 37, reprinted in Tarski 1986a, volume 2, 289–322. JFM: Aumann. Cited
in 8.7.
———. [1937] 2003. Two letters to Kazimierz Twardowski, 1 and 24 May. In Jadacki 2003b,
19–20. Cited in 16.3.
———. 1938a. Algebraische Fassung des Massproblems. Fundamenta Mathematicae 31: 47–66.
Givant 1986 item 38g, reprinted in Tarski 1986a, volume 2, 451–472. JFM: Feigl. Cited in 8.7.
———. 1938b. Über das absolute Mass linearer Punktmengen. Fundamenta Mathematicae 30:
218–234. Givant 1986 item 38e, reprinted in Tarski 1986a, volume 2, 425–444. JFM: Radon.
Cited in 8.7, 15.7, and items Givant 1986 and Tarski 1986a.
———. 1938c. Review of Mostowski 1938. Journal of Symbolic Logic 3: 115–116. Givant 1986
item 38r. Reprinted in Tarski 1986a, volume 4, 704–706. Cited in 14.2.
———. [1938] 1939. Contribution to the discussion. In Barzin [1938] 1939, 55. Givant 1986 item
39c. Reprinted in Tarski 1986a, volume 4, 707. Translated into Polish in Tarski 1995–2001,
volume 2, 472. Cited in 16.1.
———. [1938] 1983. Sentential calculus and topology. In Tarski [1956] 1983, 421–454. The orig-
inal paper, “Der Aussagenkalkül und die Topologie,” Fundamenta Mathematicae 31: 103–134,
Givant 1986 item 38h, JFM: Rinow, is reprinted in Tarski 1986a, volume 2, 473–506, and
translated into Polish in Tarski 1995–2001, volume 2, 158–200. Cited in 16.2.
———. 1939a. New investigations on the completeness of deductive theories. Journal of Sym-
bolic Logic 4: 176. Givant 1986 item 39a. Reprinted in Tarski 1986a, volume 4, 571. Cited
in 14.3.
———. 1939b. On undecidable statements in enlarged systems of logic and the concept of truth.
Journal of Symbolic Logic 4: 105–112. Givant 1986 item 39c. JFM: Skolem. MR: Frink.
Translated into Polish in Tarski 1995–2001, volume 1, 214–227. Reprinted in Tarski 1986a,
volume 2, 559–568. Cited in 14.3 and 16.2.
———. 1939c. On well-ordered subsets of any set. Fundamenta Mathematicae 32: 176–183.
Givant 1986 item 39b. JFM: Skolem. Reprinted in Tarski 1986a, volume 2, 549–558. Cited
in 14.3.
Bibliography 463

———. [1939] 2003. Letter to the dean of the Division of Mathematics and Natural Sciences of
the University of Warsaw, 16 January. In Jadacki 2003b, 27. Cited in 16.3.
———. [1939–1940] 1995. Some current problems in metamathematics. Edited by Jan Tarski
and Jan Woleski. History and Philosophy of Logic 16: 159–168. MR: Angelelli. Translated
into Polish in Tarski 1995–2001, volume 2, pages 380–395. Cited in 16.2.
———. 1939–1945. Ideale in vollständigen Mengenkörpern. Fundamenta Mathematicae 32:
69–86 and 33: 51–65. Givant 1986 items 39 and 45. JFM: Skolem. MR: Birkhoff. Reprinted
in Tarski 1986a, volume 2, 507–528, and volume 3, 1–18. The note Fundamenta Mathematicae
1945 explains the publication dates. Cited in 8.7 and 14.3.
———. [1940] 1967. The Completeness of Elementary Algebra and Geometry. Paris: Centre
National de la Recherche Scientifique, Institute Blaise Pascal. Givant 1986 item 67ma. Reprint
from page proofs of a booklet scheduled for 1940 publication in Paris by Hermann & Cie. but
destroyed by war. Reprinted again in Tarski 1986a, volume 4, 289–346, and translated into
Polish in Tarski 1995–2001, volume 2, pages 231–252. Cited in 9.4 and 16.2.
———. [1940] 1996–1999. Letter to J. M. Bocheski, 26 April. Filozofia nauki 4: 123–125, 7:
197–199. Polish transcription, with an English translation by Jan Tarski. Cited in 16.3.
———. [1942–1947] 1999. Letters to Kurt Gödel, 1942–47. Edited by Jan Tarski. In Woleski
and Köhler 1999, 261–273. Translated into Polish in Tarski 1995–2001, volume 2, pages
467–471. Cited in 16.2 and 16.3.
———. 1944. The semantic conception of truth and the foundations of semantics. Philosophy
and Phenomenological Research 4: 341–375. Givant 1986 item 44a. MR: Martin. Translated
into Polish in Tarski 1995–2001, volume 1, 228–282. Reprinted in Tarski 1986a, volume 2,
661–699. Tarski [1944] 1985 is an excerpt. Cited in 16.1 and 16.2.
———. [1944] 1985. Definition. American Mathematical Monthly 92: 358. Quotation from
Tarski 1944, 359. Cited in 16.1.
———. [1944] 1987. A philosophical letter of Alfred Tarski. Edited by Morton White. Journal
of Philosophy 84: 28–32. Givant 1986 item 87l, with incomplete publication data. MR: Hatcher.
Translated into Polish in Tarski 1995–2001, volume 1, 283–291. Cited in 16.2 and 16.3.
———. [1946] 2000. Address at the Princeton University Bicentennial Conference on Problems
of Mathematics (December 17–19, 1946). Edited, with an introduction, by Hourya Sinaceur.
Bulletin of Symbolic Logic 6: 1–44. MR: Dawson. Excerpts were translated into Polish in
Tarski 1995–2001, volume 2, 396–413. Cited in 15.12, 16.1, 16.2, and chapter 17.
———. [1946–1947] 2014. Two letters to Wacaw Sierpiski, 30 October 1946 and 18 March 1947.
In section 15.12 of the present book, Tarski 2014. Translations of handwritten documents,
APAN sygn. III-194, in the Archiwum Polskiej Akademii Nauk (Archive of the Polish Academy
of Science) in Warsaw. This is their first publication. Cited in 8.6,15.12, and 16.3.
———. [1947–1955] 2003. Four letters to Tadeusz Kotarbiski, 6 December 1947 and 16 October
1953 from Berkeley, and 7 September 1954 and 10 December 1955 from Amsterdam. In
Jadacki 2003b, 20–24. Cited in 16.3.
———. 1948. Axiomatic and algebraic aspects of two theorems on sums of cardinals. Funda-
menta Mathematicae 35: 79–104. Givant 1986 item 48a. MR: Jónsson. Reprinted in Tarski
1986a, volume 3, 171–198. Cited in 15.12.
———. [1948] 1957. A Decision Method for Elementary Algebra and Geometry. Report R-109.
Prepared for publication by J. C. C. McKinsey. Second edition. Santa Monica, California:
RAND Corporation. Givant item 48 m. MR: Heyting. Reprinted in Tarski 1986a, volume 3, 297–
368, and translated into Polish in Tarski 1995–2001, volume 2, 253–337. Cited in 9.4, 8.5,
and 16.2.
———. 1949a. Cardinal Algebras. With an appendix, “Cardinal products of isomorphism types,”
by Bjarni Jónsson and Alfred Tarski. New York: Oxford University Press. MR: Mac Lane.
Givant 1986 item 49 m. Cited in 8.7, 8.8, and 15.12.
464 Bibliography

———. 1949b. A fixpoint theorem for lattices and its applications: preliminary report. Bulletin
of the American Mathematical Society 55: 1051–1052, 1192. Givant 1986 item 49 a j, reprinted
in Tarski 1986a, volume 4, 592. Abstract of a talk presented at the summer meeting of the
American Mathematical Society at Boulder, Colorado, 30 August–2 September 1949. Cited
in 8.8.
———. 1953. A general method in proofs of undecidability. In Tarski, Mostowski, and Robinson
1953, chapter 1, 1–35. Translated into Polish in Tarski 1995–2001, volume 2, 338–379. Cited
in 16.2.
———. 1955. A lattice-theoretical fixpoint theorem and its applications. Pacific Journal of
Mathematics 5: 285–309. Givant 1986 item 55, reprinted in Tarski 1986a, volume 3, 549–
576. MR: Jónsson. Cited in 8.8.
———. [1955] 1993. Sur la théorie des modèles. Translated and edited by Anne Preller. In
Bouveresse 1991–1993, volume 2, 137–158. Translated into Polish in Tarski 1995–2001,
volume 2, 414–445. Cited in 16.2.

Tarski, Alfred. [1956] 1983. Logic, Semantics, Metamathematics: Papers from 1923 to
1938. Translated by J. H. Woodger. Second edition, edited with introduction and
analytical index by John Corcoran. Indianapolis: Hackett Publishing Company. Givant
1986 item 56 m. MR: Mendelson. Also reviewed in Grattan-Guinness 1989. This contains
• Lindenbaum and Tarski [1935] 1983; ukasiewicz and Tarski [1930] 1983;
• Tarski [1927] 1983, [1930] 1983a, [1930] 1983b, [1931] 1983, [1933] 1983,
[1934] 1983, [1935] 1983, [1935–1936] 1983, [1938] 1983;
• translations of [1935] 1936b and an adaptation of Tarski 1923a.
Cited in 8.1 and 16.1.

Tarski, Alfred. [1957] 1959. What is elementary geometry? In Henkin, Suppes, and Tarski 1959,
16–29. Givant 1986 item 59. Reprinted in Tarski 1986a, volume 4, 17–32. MR: Mendelson.
Cited in 9.4 and 12.11.
———. [1963] 2003. Letter to Tadeusz Kotarbiski, 7 June. In Jadacki 2003b, 24. Cited in 16.3.
———. 1965a. Bibliography of Alfred Tarski. Berkeley: Department of Mathematics, University
of California, Berkeley. Superseded by Givant 1986. Cited in chapters 10 and 11, 12.0,
and 16.2.
———. 1965b. A simplified formalization of predicate logic with identity. Archiv für mathe-
matische Logik und Grundlagenforschung 7: 61–79. Givant 1986 item 65, reprinted in Tarski
1986a, volume 4, 251–272, and translated into Polish in Tarski 1995–2001, volume 2, 201–230.
MR: Monk. Cited in 16.2.
———. [1965] 2007. Two unpublished contributions by Alfred Tarski. Edited by Francisco
Rodríguez-Consuegra. History and Philosophy of Logic 28: 257–264. Cited in 15.10 and 16.1.
———. 1969. Truth and proof. Scientific American 220: 63–77. Givant 1986 item 69. Reprinted
in Tarski 1986a, volume 4, 399–424. Translated into Polish in Tarski 1995–2001, volume 1,
292–332. Cited in 15.6, 16.1, and 16.2.
———. 1972. O szkicu Feliksa Grossa w Wiadomociach. Wiadomoci 27(1347, 23 January): 6.
In English, with Polish commentary by the journal editor. Tarski had written this letter to
Feliks Gross urging republication of Gross’s article “Trip to the moon: Why?”
———. [1972–1973] 1995. Two letters to Svätoslav Mathé, 21 July 1972 and 14 February 1973.
Organon F 2: 56–58. The first is in Slovak, translated from Polish by Pavel Cmorej. The
second is handwritten in Polish. Cited in 16.3.
Bibliography 465

Tarski, Alfred. 1986a. Collected Papers. Edited by Steven R. Givant and Ralph McKenzie.
Four volumes. Basel: Birkhäuser. Item 81ma in Givant 1986 is a preliminary version.
MR: Corcoran. Also reviewed, with corrections, in Maddux 1993. This contains reprints of
• Givant 1986; Knaster 1927;
• Horn and Tarski 1948a and 1948b; Kotarbiski [1935] 1955–1956;
• Lindenbaum and Tarski 1926; Mostowski and Tarski 1939;
• Tarski 1923b, 1924a ... e, 1925b, 1927a ... b, [1927] 1929, 1928,
[1928] 1930, 1929–1930, 1930a ... b, 1931, [1931] 1952, 1932,
[1932] 1952, [1933] 1935, [1935] 1936a ... b, 1935–1936,
1937a ... c, 1938a ... c, [1938] 1939, 1939a ... c, 1939–1945,
[1940] 1967, 1944, 1948, [1948] 1957, 1949b, 1955,
[1957] 1959, 1965b, 1969;
and reprints of the original versions of
• Banach and Tarski [1924] 2014; Lindenbaum and Tarski [1935] 1983;
• ukasiewicz and Tarski [1930] 1983;
• Tarski [1921] 2014, [1924] 2014b, [1925] 2014, [1927] 1983,
[1927] 2014a ... b, [1928] 2010, [1929] 2014a ... b, [1930] 1983a ... b,
[1930] 2014a ... f, [1930–1931] 2014, [1931] 1983, [1931] 2014a ... g,
[1932] 2014a ... e, [1934] 1936a ... b, [1934] 1983, [1935–1936] 1983,
[1936] 2014a ... d, [1938] 1983.
Cited in the preface, 1.2, chapter 11, 12.0, 15.3, and 16.0, ... , 16.2.

Tarski, Alfred. 1986b. What are logical notions? Edited by John Corcoran. History and Philoso-
phy of Logic 7:143–154. Givant 1986 item 87?. MR: Cocchiarella. Translated into Polish in
Tarski 1995–2001, volume 2, 446–466. Cited in 16.1.
———. 1994. Introduction to Logic and to the Methodology of the Deductive Sciences. Fourth
edition. Translated from the Polish by Olaf Helmer. Edited, with a preface and biographical
sketch, by Jan Tarski. Oxford logic guides 24. New York: Oxford University Press. MR:
Mendelson. Contains J. Tarski 1994a. Tarski 1996 is a Polish translation. For earlier editions,
see item Tarski [1936] 1995. Cited in 16.2.
———. 1995–2001. See a box on the next page.
———. 1996. Wprowadzenie do logiki i do metodologii nauk dedukcyjnych. Second edition.
Translated by Monika Sujczyska from the fourth English edition, edited and with a preface
by Jan Tarski, with a supplement on elements of predicate logic by Witold Marciszewski.
Warsaw: Filia Uniwersytetu Warszawskiego, Philomath, Aleph. Translation of Tarski 1994
and J. Tarski 1994a. Reviewed extensively in A. Wójtowicz 1996. The first edition had been
issued in 1995 by the same publisher with a different supplement, “Logika a dzielno
umysu”—“Logic and the boldness of mind”—by Marciszewski. See J. Tarski 1996a for a
complaint about that. Cited in 16.2.
Tarski, Alfred, and Steven R. Givant. 1987. A Formalization of Set Theory without Variables.
American Mathematical Society colloquium publication 41. Providence: American Mathemati-
cal Society. Givant 1986 item 8–m, with incomplete publication data. MR: Monk. Also reviewed
in Monk 1989 and Németi 1990. Cited in 16.2.
———. 1999. Tarski’s system of geometry. Bulletin of Symbolic Logic 5: 175–214. MR: Pambuc-
cian. Cited in 9.4, 12.11, and 16.1.
466 Bibliography

Tarski, Alfred. 1995–2001. Pisma logiczno-filozoficzne. Volume 1: Prawda. Volume 2:

Metalogika. Translated and annotated, with introduction, by Jan Zygmunt. With
bibliography by Steven Givant, supplemented by Jan Zygmunt. Warsaw: Wydawnictwo
Naukowe PWN. The title means Logical-Philosophical Writings. Reviewed in Cesarski
1995. This contains Givant 1995, Zygmunt 1995, and
• reprints of Tarski 1933, 1934, 1936a... b;
• and of the original versions of Tarski [1930–1931] 2014, [1936] 2014d.
In addition, it contains Polish translations of
• Tarski 1932, [1938] 1939, 1938–1939, 1939b, [1939–1940] 1995,
[1940] 1967, [1942–1947] 1999, 1944, [1944] 1987, [1948] 1957,
1953, [1955] 1993, 1965b, 1969, [1946] 2000, 1986b,
parts of Tarski [1930–1936] 1992, [1946] 2000;
and translations of the original versions of
• Lindenbaum and Tarski [1935] 1983; ukasiewicz and Tarski [1930] 1983;
• Tarski [1930] 1983a, [1934] 1983, [1938] 1983, [1938] 1983.
Cited in 16.0, ... , 16.2.

Tarski, Alfred. 2014. Early Work in Poland—Geometry and Teaching. Translated and
edited by Andrew McFarland, Joanna McFarland, and James T. Smith. New York: Birk-
häuser. The present book. This contains
• Banach and Tarski [1924] 2014; ukasiewicz [1928–1929] 2014;
• Moese [1932] 2014;
• Tarski [1921] 2014, [1924] 2014a ... b, [1925] 2014, [1927] 2014a ... b,
[1929] 2014a ... b, [1930] 2014a ... f, [1930–1931] 2014,
[1931] 2014a ... g, [1932] 2014a ... e, [1934] 2014, [1935] 2014,
[1936] 2014a ... d, [1946–1947] 2014,
and a reprint of Tarski 1925a. See the overviews in the preface and in the introductions
to Parts One to Four.

Tarski, Alfred, and Henryk Moese. 1952. Concerning the Degree of Equivalence of Polygons.
Translated by Izaak Wirszup. Chicago: The College, University of Chicago. A reproduced type-
script, this contains draft translations Tarski [1931] 1952 and [1932] 1952 of the original
versions of Tarski [1931] 2014a and [1932] 2014d and a draft translation of the original version
of Moese [1932] 2014. Cited in 7.0, ... , 7.4, 8.5, and 9.8.
Tarski, Alfred, Andrzej Mostowski, and Raphael M. Robinson. 1953. Undecidable Theories.
Amsterdam: North–Holland Publishing Company. Givant 1986 item 53 m. MR: Kreisel. Con-
tains Tarski 1953. Cited in 16.2.
Tarski, Alfred, et al. 1971. Letter to Tadeusz Kotarbiski, 24 April. In Jadacki 2003b, 25. There
were ten cosigners. Cited in 16.3.
Tarski, Jan. 1994a. A short biographical sketch of Alfred Tarski. In Tarski 1994, xviii–xxi.
Tarski 1996 contains a Polish translation. Cited in chapter 17.
———. 1994b. Uwagi o artykule Woleskiego “Alfred Tarski jako filozof ”. Wiadomoci mate-
matyczne 30: 265–269. The title means “Remarks on Wolenski 1987.” Cited in 18.2.
Bibliography 467

———. 1996a. W sprawach wydawniczych. Filozofia nauki 4(4): 153–154. The title means “On
matters of publishing.” Dated 23 March, this letter to the editor is about the first edition of
Tarski 1996.
———. 1996b. Philosophy in the creativity of Alfred Tarski. In Hempoliski 1996, 157–159.
Cited in chapter 17.
———. 2012. Personal communication to James T. Smith, 20 February. Cited in 14.3.
Tatarkiewicz, Krzysztof. 2003. W sprawie daty mierci Z. Waraszkiewicza. Wiadomoci matema-
tyczne 39: 221–222. The title means “On the date of death of Z. Waraszkiewicz.” This corrects
an error in Derkowska 2001. Cited in 7.1.
Terracini, Alessandro. 1958. Ugo Amaldi. Atti della Accademia delle Scienze di Torino, classe
di scienze fisiche, matematiche e naturali 92 (1957–1958): 687–695. In Italian. Cited in 9.2.
Thomasson, Amie. 2012. Roman Ingarden. In Zalta 2011– at http://plato.stanford.edu/
archives/fall2012/entries/ingarden. Cited in 15.6.
Tomczak, Roman. 1927. Kalendarz nauczycielski na rok 1927. Warsaw: Nakadem Zarzdu
Gównego Zwizku Polskiego Nauczycielstwa Szkó Powszechnych. The title means Teacher’s
Calendar for the Year 1927. Cited in 9.6.
Toru, University of. 1995. Pracownicy nauki i dydaktyki Uniwersytetu Mikoaja Kopernika,
1945–1994: Materiay do biografii. Edited by Sawomir Kalembki. Toru: Uniwersytet Mikoaja
Kopernika w Toruniu. The initial title phrase means Scientific and Teaching Staff of the
Nicolaus Copernicus University. Cited in 7.1.
Tricomi, Francesco G. 1962. Matematici Italiani del primo secolo dello stato unitario. Memorie
dell’Accademia delle Scienze di Torino, classe di scienze fisiche, matematiche e naturali (series
4) 1: 1–120. Augmented by PRISTEM 2005–. Cited in 4.1.
Trzsicki, Kazimierz. 2010. Greniewski Henryk (1903–1972). Warsaw: Polskie Towarzystwo
Logiki i Filozofii Nauki. On the Internet at http://logic.org.pl/doc/historia/hg.pdf,
the site of the Polish Society for Logic and Philosophy of Science. In Polish. An abridgement
is in Maryniarczyk et al. 2011, volume 1, 451. Cited in 15.2.
Turkowski, Romuald. 1994. Schayer Wacaw. Polski sownik biograficzny 35: 413–415. In
Polish. Cited in 9.9.

Ulam, Stanisaw Marcin. 1976. Adventures of a Mathematician. New York: Charles Scribner’s
Sons. MR: Oxtoby. Cited in 14.3.

Van den Dries, Lou. 1988. Alfred Tarski’s elimination theory for real closed fields. Journal of
Symbolic Logic 53: 7–19. MR: Coplakova. Cited in 18.1.
Van Heijenoort, Jean. [1967] 1970. From Frege to Gödel: A Source Book in Mathematical Logic,
1879–1931. Cambridge, Massachusetts: Harvard University Press. MR: Dawson. Contains
Gödel [1931] 1967 and Padoa [1900] 1901. Cited in 4.1.
Vaught, Robert L. [1971] 1974. Model theory before 1945. In Henkin et al. 1974, 153–172. MR:
Anonymous. Cited in 15.6, 18.1, and 18.2.
———. 1986–1987. Alfred Tarski’s work in model theory. Journal of Symbolic Logic 51:
869–882; 52(4): vii. MR: Baldwin. Cited in 18.1 and 18.2.
Veblen, Oswald. 1911. The foundations of geometry. In Young 1911, 3–54. Cited in 9.4.
Viola, Tullio. 1957. Ugo Amaldi. Bollettino della Unione Matematica Italiana (3) 12: 727–730.
In Italian. Cited in 9.2.
Vitali, Giuseppe. 1905. Sul problema della misura dei gruppi di punti di una retta: Nota.
Bologna: Gamberini e Parmeggiani. Five-page pamphlet. Cited in 4.2, 4.4, and 6.0.
Völkischer Beobachter. 1938 (North German edition, 2 October, 1–3). Die Reichshauptstadt
huldigt dem Befreier des Sudetenlandes. Berlins stürmischer Dank. Wilhelmplatz: ein
einziger Orkan jubelnder Begeisterung. Cited in 14.3.
468 Bibliography

Wagon, Stan. 1993. The Banach–Tarski Paradox. Paperback edition. Cambridge: Cambridge
University Press. MR: Bzyl. Originally published in 1985; the later edition has some addi-
tional information. Cited in 4.2, ... , 4.4, 8.6, and 8.7.
Wajsberg, Mordchaj. 1977. Logical Works, edited by Stanisaw Surma. Wrocaw: Polish Academy
of Sciences, Institute of Philosophy and Sociology. Contains Surma 1977.
Wapner, Leonard M. 2005. The Pea and the Sun: A Mathematical Paradox. Wellesley, Massa-
chusetts: A K Peters. MR: Plotkin. Also reviewed in Komjáth 2006. Cited in 8.6.
Warsaw, University of. 1918–1919. Document “Uniwersytet Warszawski Sekretariat” dated 8
November 1918 and 5 February 1919. In the Archive of the University of Warsaw, Signatura
Teitelbaum-Tarski Alfred: RP 2909. Cited in 1.1.
———. 1921–1939. Skad uniwersytetu i spis wykadów. One volume for each year. Warsaw:
University of Warsaw. The title means University Bulletin and Schedule of Lectures. Cited
in 9.3 and 9.7.
Warszawski Instytut Filozoficzny. 1921–1922. Wykady, odczyty, referaty. Ruch filozoficzny 6:
72, 146. The title means “Lectures, readings, papers.” Cited in chapter 3.
———. 1931. October 1931 minutes. Ruch filozoficzny 12 (1930–1931), 228. Cited in 14.2.
———. 1937. Wykady, odczyty, referaty. Ruch filozoficzny 13: 149. The title means “Lectures,
readings, papers.” Cited in 14.2.
Waszyski, Edmund. 1996. Professor Dr. Rudolf Weigl (1883–1957) i dziaalno jego Instytutu
Tyfusowego we Lwowie w latach 1939–1944. Archiwum historii i filozofii medycyny 59: 77–
84. These pages also include the German translation “Professor Dr. Rudolf Weigl (1883–1957)
und die Tätigkeit seines Instituts für Fleckfieberforschung in Lwów in den Jahren 1939–1944”
and an English summary. Cited in 4.2.
Watt, Richard M. 1979. Bitter Glory: Poland and Its Fate, 1918 to 1939. New York: Simon and
Schuster. Cited in 1.1 and chapter 3.
Wells, Benjamin. 2007. Alfred Tarski, friend and daemon. Notices of the American Mathematical
Society 54: 982–984. Cited in chapter 17.
White, Morton Gabriel. 1999. A philosopher’s story. University Park, Pennsylvania: Pennsylva-
nia State University Press. Cited in 16.3.
Whitehead, Alfred N., and Bertrand Russell. 1910–1913. Principia Mathematica. Three volumes.
Cambridge, England: The University Press. JFM: Jourdain. Cited in 4.4 and 15.6.
Witt-Hansen, Jobs. 1969. Obituary on Jørgen Jørgensen. Logique et analyse, new series, 46:
121–122. Cited in 16.1.
Wilkosz, Witold. 1936. Znaczenie logiki matematycznej dla matematyki i innych nauk cisych.
In Polski Zjazd Filozoficzny 1936, 343–347. The title means “The significance of mathematical
logic in mathematics and other exact sciences.” Contains the original version of the comment
Tarski [1936] 2014b. Cited in 15.9.
Wilson, Trevor M. 2005. A continuous movement version of the Banach–Tarski paradox: a
solution to de Groot’s problem. Journal of Symbolic Logic 70: 946–952. Cited in 8.6.
Wójcicki, Ryszard, ed. 1981. Maria Kokoszyska-Lutman (1905–1981). Studia Logica 40:
313–314. The article is by the editors; Wójcicki was editor-in-chief. Cited in 14.2.
Wojciechowska, Agnieszka. 2001. Alfred Tarski (1901–1983). Matematyka: czasopismo dla
nauczycieli 54(3): 132–135. In Polish. Cited in chapter 17.
———. 2011. Alfred Tarski (1901–1983). Matematyka: czasopismo dla nauczycieli 64(9): 3–9.
In Polish. Cited in chapter 17.
Wójcik, Andrzej. 1985. Alfred Tarski: ycie i dzieo. Pismo literacko-artystyczne 4 (1): 174–178.
The titles mean “Life and work” and Literary-Artistic Writings. Cited in chapter 17.
Wójtowicz, Anna. 1996. Logika dla filozofów nauki. Filozofia nauki 4: 117–122. The title means
“Logic in the philosophy of science.” This is a review of Tarski 1996. Cited in 16.2.
Bibliography 469

Wojtowicz, Wadysaw. [1919] 1926. Zarys geometrji elementarnej do uytku szkó rednich. Sixth
edition. Lwów: Wydawnictwo Zakladu Narodowego im. Ossolinskich. The 1919 edition was
published in Warsaw by Gebethner i Wolff. Reviewed in Kwietniewski 1923. The title means
Outline of Elementary Geometry for Use in Secondary Schools. The third edition has been
revised and modernized by editor Jerzy Norwa and published in the series Gra0ient—Kufer
matematyków, Warsaw: Fundacja Rozwoju Matematyki Polskiej, 2000–2001. Cited in 7.1, 7.2,
7.4, 9.2, 9.3, and 12.11.
Wojtowicz, Wadysaw, and Stefan Straszewicz. 1923. Od redakcji. Przegld matematyczno-
fizyczny 1: 1–2. The title means “From the editors.” Cited in 4.3.
Woleski, Jan. 1984. Alfred Tarski (1901–1983). Studia filozoficzne (2): 3–8. In Polish. Cited
in chapter 17.
———. 1987. Alfred Tarski jako filozof. Wiadomoci matematyczne 27 (1986–1987): 247–259.
MR: Murawski. The title means “Alfred Tarski as a philosopher.” J. Tarski 1994b contains
remarks about this paper. Cited in 18.2.
———. 1989. Logic and Philosophy in the Lvov–Warsaw School. Synthese library 198. Dord-
recht: Kluwer Academic Publishers. Cited in the preface.
———. 1990a. Kotarbiski. Warsaw: Wiedza Powszechna. Contains Woleski 1990b.
———. 1990b. ycie i twórczo . In Woleski 1990a, 7–15. The title means “Life and work.”
This is a biographical sketch of Tadeusz Kotarbiski. Cited in 1.1.
———. 1993. Tarski as a philosopher. In Coniglione, Poli, and Woleski 1993, 319–338. This
is not a translation of Woleski 1987. Cited in 15.2, 15.10, 15.11, and 18.2.
———. 1994. Philosophical Logic in Poland. Synthese library 228. Dordrecht: Kluwer Academic
Publishers. Contains Bocheski 1994.
———. 1995a. Mathematical logic in Poland 1900–1939: People, circles, institutions, ideas.
Modern Logic 5: 363–405. MR: Sinaceur. Reprinted in Woleski 1999, 59–84. This paper has
two sections numbered 4. Cited in chapter 3, 9.4, 14.2, and 14.3.
———. 1995b. On Tarski’s background. In Hintikka 1995, 331–341. MR: Anonymous.
Reprinted in Woleski 1999, 126–133. Cited in chapter 17.
———. 1999. Essays in the History of Logic and Logical Philosophy. Dialogikon volume 8. With
foreword by Christian Thiel. Cracow: Jagiellonian University Press. Contains reprints of
Woleski 1995a and 1995b.
———. 2000–2001. Review of Leniewski 1992. Modern Logic 8: 195–201.
———. 2001a. Alfred Tarski–ksi logików: wzór na prawd. Polityka 2281 (20 January):
76–77. The title phrases mean “Prince of logicians: pattern for truth.” The journal is a weekly
newsmagazine. Cited in chapter 17.
———. 2001b. In defense of the semantic definition of truth. Synthese 126: 67–90. Cited
in 15.11.
———. 2008. Mathematical logic in Warsaw: 1918–1939. In Ehrenfeucht, Marek, and Srebrny
2008, 30–46. Cited in chapter 3.
———. 2009. Alfred Tarski: Logik z fantazj. Gazeta wyborcza (4–5 April): 24–25. The title
phrase means “Logician with flair.” The journal is a daily newspaper with national circulation.
Cited in chapter 17.
Woleski, Jan, and Eckehart Köhler. 1999. Alfred Tarski and the Vienna Circle: Austro–Polish
Connections in Logical Empiricism. Vienna Circle Institute Yearbook 6. Dordrecht: Kluwer
Academic Publishers. Mostly the proceedings of a 1997 congress in Vienna. Contains A.
Feferman [1997] 1999, S. Feferman [1997] 1999, and Tarski [1942–1947] 1999. Cited in 16.3
and chapter 17.
Wolna Wszechnica Polska. 1925. Skad osoby i spis wykadów. Warsaw: Wolna Wszechnica Pol-
ska. The author name and title mean “Free University of Poland” and List of Personnel and
Schedule of Courses. Cited in chapter 3.
470 Bibliography

Woodger, Joseph H. 1937. The Axiomatic Method in Biology. With appendices by Alfred Tarski
and W. F. Floyd. Cambridge, England: Cambridge University Press. JFM: Ackermann. Con-
tains Tarski 1937a. Cited in 14.3.
Woytak, Lidia. 1987. Alfred Tarski—matematyk o jzyku. Przegld Polonijny 1987 (4): 67–75.
The title phrase means “a mathematician on language.” The journal title means Polonia
Review. Cited in chapter 17.
Wynot, Edward D., Jr., 1983. Warsaw between the World Wars: Profile of the Capital City in a
Developing Land, 1918–1939. East European monograph 129. New York: Columbia University
Press. Cited in 1.1, chapter 3, 9.1, and 14.2.

Yanigahara, Kitizi. 1927. On the axiomatic simplicity of the proof of Pythagorean theorem by
dissection and addition. Tôhoku Mathematical Journal 18: 59–64. JFM: Süss. Cited in 7.1.
Yarden, Tama. 2006. Dov Jarden. On the Internet at http://www.shvoong.com/human-
ities/religion-studies/243858-dov-jarden/. Cited in 7.1.
Young, J. W. A., editor. 1911. Monographs on Topics of Modern Mathematics Relevant to the
Elementary Field. London: Longmans, Green and Co. JFM: Jourdain. Reprinted in 1955 by
Dover Publications with an introduction by Morris Kline. Contains Veblen 1911.

Zaanen, Adriaan C. 1958. An Introduction to the Theory of Integration. Amsterdam: North–

Holland Publishing Company. MR: Hildebrandt. Cited in 4.2 and 4.4.
Zagórowski, Zygmunt, editor. 1924–1926. Spis nauczycieli: Szkó wyszych, rednich, zawodo-
wych, seminarjów nauczycielskich oraz wykaz zakadów naukowych i wadz szkolnych. Two
volumes. Lwów: Ksinica Polska, Ksinica–Atlas. The title means List of Teachers in
Higher, Middle, Vocational, and Teacher Training Colleges and List of Seminars, Scientific
Establishments, and Government Schools. Cited in chapter 3, 9.1, 9.2, and 9.9.
Zahorski, Bolesaw. 1920. Wiersze Pisudczyka. Minsk and Warsaw: Wende. The title means
Pisudski Poems. Cited in 9.1.
Zakad Ubezpiecze Spoecznych. 1935. Letter to Alfred Tarski, 22 November. In the Archiwum
Akt Nowych (Central Archives of Modern Records) in Warsaw. The author name means Social
Insurance Institution. Cited in 9.1.
Zalasiewicz, Jan. 2009. The meaning of palaeontology. Palaeontology Newsletter 72: 8–13. Cited
in 14.1.
Zalta, Edward N., editor. 2011–. Stanford Encyclopedia of Philosophy. Stanford: http://
plato.stanford.edu/. Internet website. Contains Betti 2011, Cat 2010, Gómez-Torrente
2006, Segerberg 2009–, and Thomasson 2012.
Zawirski, Zygmunt. [1934] 1936. Bedeutung der mehrwehrtigen Logik für die Erkenntnis und
ihr Zusammenhang mit der Wahrscheinlichkeitsrechnung. In International Congress of
Philosophy 1936, 175–180. Discussed in Tarski [1934] 1936a and [1936] 2003a. Cited in 16.1.
———. 1936. W sprawie syntezy naukowej. In Polski Zjazd Filosoficzny 1936, 347–352. The
title means “Concerning scientific synthesis.” Contains the original version of the comment
Tarski [1936] 2014c. Cited in 15.10.
Ziemiaski, Ignacy. 1933. Praca kobiet w P. O. W.–Wschód. Preface and remembrance by Mel-
chior Wakowicz, cover and archival items by Adolf Buraczewski. Warsaw: W. azarski. The
title means, Efforts of Women in the Polish Military Organization–East. Cited in 9.1.
Zupani, Rihard. 1918–1921. Pogldowa nauka geometryi dla klasy pierwszej; zarys geometryi
dla klasy drugiej; zarys geometryi dla klasy trzeciej. Three separate volumes in series. Adapted
by Ludwik Hordyski from German editions. Lwów: Seyfarth, Ossolinskich. The titles mean
An Intuitive Study of Geometry for the First Class, and Outline of Geometry for the Second and
Third Classes. The original volumes were published in Vienna by F. Tempsky. The German
form of the author’s name is Richard Suppantschitsch. Cited in 9.2.
Bibliography 471

Zydler, Jan. 1925. Geometrja w zakresie szkoy redniej. Fourteenth edition. Warsaw: Wydaw-
nictwo M. Arcta. The title means Geometry for Secondary Schools. Cited in 9.2, 12.11,
and 13.1.
Zygmunt, Jan. 1991. Mojesz Presburger: Life and work. History and Philosophy of Logic 13:
211–223. MR: Dawson. Cited in 9.3 and 9.4.
———. 1995. Szkic biograficzny. In Tarski 1995–2001, volume 1, vii–xx. Biography of Alfred
Tarski. Cited in 9.4 and chapter 17.
———. 2000. Alfred Tarski. Edukacja filozoficzna 29: 274–307. In Polish. Cited in chapter 17.
———. 2001. Alfred Tarski. In Mackiewicz 2001, volume 1, 342–375. In Polish. Cited in
chapter 17.
———. 2009. Alfred Tarski—logik i metamatematyk. In Dobierzewska-Mozrzymas and Jezierski
2009, 305–327. The title phrase means “logician and metamathematician.” Cited in 18.2.
———. 2010. Tarski’s first published contribution to general metamathematics. In Béziau 2010,
59–66. Cited in 9.4, 15.6, and 16.1.
 Permissions
The editors acknowledge gratefully the cooperation of several archival institutions that
provided access to their resources and permission to publish images or translations of
various items in the indicated sections of the present book:
Archiwum Akt Nowych (Warsaw) . . . . . . . . Tarski [1934] 2014 letter: . . . . . . . §16.3
translation
Tarski [1935] 2014 letter: . . . . . . . §16.3
translation
Archiwum Fotografii . . . . . . . . . . . . . . . . . . Photo: Chwiakowski, from . . . . . §9.9
Orodka KARTA (Warsaw) sygnatura OK_0997_0001_024
Archiwum Polskiej Akademii . . . . . . . . . . . Tarski [1946–1947] 2014 . . . . . . . §15.14
Nauk (Warsaw)
Archiwum Pastwowe Miasta
Stoecznego Warszawy . . . . . . . . . . . . . . Photo: Ul. Koszykowa 51 . . . . . . . §9.1
Archiwum Pastwowe w Lublinie . . . . . . . Poster: crucifix . . . . . . . . . . . . . . . §1.1
Archiwum Uniwersytetu . . . . . . . . . . . . . . Tarski 1924f coursebook . . . . . . . . §1.1
Warszawskiego Photo from coursebook . . . . . . . . . Ch. 3
Pasenkiewicz 1935
Tarski [1918] 2014 CV: . . . . . . . . . Ch. 3, §16.3
signature, translation
Tarski 1920 letter: . . . . . . . . . . . . . Ch. 3
signature
Tarski 1921–1926: letter . . . . . . . Ch. 3
Tarski [1924] 2014a letter: . . . . . . Ch. 3, §16.3
signature, translation
Tarski [1939] 2003 letter: . . . . . . . frontispiece
signature
Photo: Wajsberg, from . . . . . . . . . Ch. 3
sygnatura RP15153
Photo: Kuratowski, from . . . . . . . §4.4
sygnatura RP379
Photo: Presburger, from . . . . . . . §9.4
sygnatura RP16485
Photo: Mostowski, from . . . . . . . . §9.4
sygnatura RP36752
Photo from coursebook . . . . . . . . . §9.9
Schayer 1932
Photo: Hosiassonówna, . . . . . . . . §14.3
from sygnatura RP5224
Photo: Lindenbaum, from . . . . . . §14.3
sygnatura RP11947
Bancroft Library, University of . . . . . . . . . . Quotations from . . . . . . . . . . . . . . Ch. 3
California, Berkeley Pasenkiewicz 1984
Shipboard menu and cover . . . . . . §14.3

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 473
DOI 10.1007/978-1-4939-1474-6, © Springer Science+Business Media New York 2014
474 Permissions

Centralna Biblioteka Matematyczna, . . . . . Photo: 1929 Congress . . . . . . . . . . §9.5

Instytut Matematyczny, Polska members
Akademia Nauk (Warsaw)
Ksinica Kopernikaska in Toru . . . . . . Photo: Henryk Moese . . . . . . . . . . 7.1
Gazeta Gdaska 1935.09.26
Mathematisches Forschungsinstitut . . . . . Photo: Alfred Tarski . . . . . . . . . . . Ch. 16 Intro.
Oberwolfach: Photo Collection
Narodowe Archiwum . . . . . . . . . . . . . . . . . . Photo: Twardowski . . . . . . . . . . . §1.3
Cyfrowe (Warsaw) Photo: Jabloski . . . . . . . . . . . . . . Ch. 3
Photo: Sierpiski . . . . . . . . . . . . . §4.1
Photo: ukasiewicz . . . . . . . . . . . §9.4
Photo: Wilkosz . . . . . . . . . . . . . . . §15.9
United States Holocaust Memorial . . . . . . . Poster: Trotsky as devil . . . . . . . . §1.1
Museum (Washington)
Wikimedia Commons . . . . . . . . . . . . . . . . . . Photos: Amaldi, Enriques . . . . . . §9.2

Springer, the publisher of this book, extended permission for translating or repro-
ducing images or text from their earlier publications:
Translations from Tarski 1986a, . . . . . . . . . . Tarski [1921] 2014 . . . . . . . . . . . . Ch. 2
Collected Papers Tarski [1924] 2014b . . . . . . . . . . . Ch. 5
Banach–Tarski [1924] 2014 . . . . . Ch. 6
Tarski [1931] 2014a . . . . . . . . . . . §7.2
Tarski [1932] 2014d . . . . . . . . . . . §7.4
Tarski [1929] 2014a . . . . . . . . . . . Ch. 10
Tarski [1932] 2014e . . . . . . . . . . . Ch. 11
Tarski [1930] 2014a–f . . . . . . . . . §§12.1–6
Tarski [1931] 2014b–g . . . . . . . . . §§12.7–12
Tarski [1932] 2014b–c . . . . . . . . . §§12.13–1
Tarski [1925] 2014 . . . . . . . . . . . . 4
Tarski [1927] 2014a–b . . . . . . . . . §15.1
Tarski [1929] 2014b . . . . . . . . . . . §§15.2–3
Tarski [1930–1931] 2014 . . . . . . . §15.5
Tarski [1932] 2014 . . . . . . . . . . . . §15.6
Tarski [1936] 2014a–d . . . . . . . . . §15.7
§15.8–11
Quotations fromTarski 1986a, . . . . . . . . . . . . Tarski 1969 . . . . . . . . . . . . . . . . . .§15.6
Collected Papers
From Marchisotto/Smith 2007, Legacy . . . . . Text and figures . . . . . . . . . . . . . .§9.3
of Mario Pieri in Geometry and
Arithmetic (Birkhäuser), §3.10.2
From Givant 1991, “Portrait of Alfred . . . . . Photos: Tarski, Leniewski . . . . .§1.1
Tarski,” Mathematical Intelligencer
From Givant 1999, “Unifying threads . . . . . Photo: Tarski . . . . . . . . . . . . . . . . Frontispiece
in Alfred Tarski’s work,” Mathematical Photo: Szmielew . . . . . . . . . . . . . . §9.4
Intelligencer
From Purkert 2008, “The double life of Photo: Hausdorff . . . . . . . . . . . . .§4.4
Felix Hausdorff/Paul Mongré,”
Mathematical Intelligencer
Permissions 475

From Szmuilewicz-Lachman et al. 1994, . . . Photo: Zawirski . . . . . . . . . . . . . . .§15.10

Zygmunt Zawirski (Kluwer)

Several individuals, publishers, and other organizations graciously extended permission

for translating or reproducing images or text from their Internet websites or earlier
publications:
Prof. Alina Filipowicz-Banach . . . . . . . . . . . . Photo: Banach . . . . . . . . . . . . . . . §4.2
Galeria Grabek (Lublin, Poland) . . . . . . . . . . Poster: Pozna exhibition . . . . . . §9.5
Prof. Jacek Juliusz Jadacki . . . . . . . . . . . . . . Photo: Kotarbiski . . . . . . . . . . . §1.1
Photo: Ul. Sukowski 4 . . . . . . . . §9.1
Polskie Towarzystwo Matematyczne . . . . . . . Photo: Mazurkiewicz . . . . . . . . . §1.1
Uniwersytet Mikoaja Kopernika, . . . . . . . . . ukasiewicz [1928–1929] 2014: . . §15.4
Instytut Filozofii (Ruch filozoficzny) translation
Prof. Jan Tarski . . . . . . . . . . . . . . . . . . . . . . . Photo: Witkowska sisters . . . . . . §9.1
Translations from Geometrja . . . . Ch. 13
Photo: Kokoszyska, Tarski . . . . §14.2
Prof. Andrzej Jerzmanowski . . . . . . . . . . . . . Translations from Geometrja . . . . Ch. 13
The present editors have searched without success for heirs of the third author, Zygmunt
Chwiakowski, of Tarski’s [1935] 1946 text Geometrja. Its original publishers disappeared
in the aftermath of World War II, and we have searched without result for successors.
Should anyone have questions about rights to that publication, please contact the senior
of the editors, James T. Smith.
 Index of Persons

This book has indexes of persons and of subjects. The subject index starts on page 485.
The present index lists persons featured in the main text of the book. It does not cover
the entries in the bibliography. An = sign indicates that a person is listed under a
different name. Personal dates are supplied if possible. Many biographical sketches are
included in this book in boxes or footnotes; those names and the page numbers of
the sketches are italicized in this index. The book contains many portraits; those names
and page numbers are boldfaced. Alphabetization of the index ignored spaces, punctu-
ation, and diacritical marks. Caution: this convention differs from the Polish standard.

Adalbert, Saint = Wojciech, Saint Bernstein, Felix (1878–1956), 126. See also
Adamowicz, Zofia (1950– ), 413 the subject index
Addison, John West, Jr. (1930– ), 180, 399 Betti, Arianna (1970– ), 18, 40, 362
Ajdukiewicz, Kazimierz (1890–1963), Betti, Enrico (1823–1892), 181
36–37, 180, 325, 327, 342, 344, 361, Béziau, Jean-Yves (1965– ), 386
365–366, 370 Biaobrzeski, Czesaw (1878–1953), 33
Alfred = Tarski, Alfred Baszczyk, Leon Tadeusz (1923– ), 320
Allen, Layman Edward (1927– ), 341 Blok, Willem Johannes (1947–2003), 411
Amaldi, Ugo (1875–1957), 45–46, 61–62, Bocheski, Father Innocentius Maria (1902–
90, 181–182, 183, 189, 213. See also the 1981), 17, 368–369, 396
subject index Bocheski, Józef Maria =
Anders, Wadysaw (1892–1970), 34, 224, 320 Bocheski, Father Innocentius Maria
Andréka, Hajnal Ilona (1947– ), 414 Bolyai, Farkas Wolfgang (1775–1856). See
Anellis, Irving Henry (1946–2013), 401 the subject index
Apostol, Tom Mike (1923– ), 163 Bonola, Roberto (1874–1911), 265
Archimedes (287–212 B.C.E.), 213. See also Boole, George (1815–1864), vi
the subject index Borel, Émile (1871–1956), 33, 65
Aristotle (384–322 B.C.E.), ix, 362 Bornstein, Benedykt Leon (1880–1948), 368
Auerbach, Walter (1900–1942), 374 Borowski, Marian (1879–1938), 12
Axler, Sheldon Jay (1949– ), xiv Borsuk, Karol (1905–1982), 379, 381
Bouveresse, Jacques (1940– ), 389
Bacon, Francis (1561–1626), 8 Brandes, Hans (1883–1965), 126
Baley, Stefan (1885–1952), 20 Braun, Jerzy (1901–1975), 367, 370
Banach, Katarzina (1864–19??), 57 Braus, ucja. = Banach, ucja
Banach, ucja (????–1954), 57 Brentano, Franz (1838–1917), 18
Banach, Stefan (1892–1945), xi, xiii, 41, 45, Brouwer, Luitzen Egbertus Jan (1881–
51–56, 57–66, 70, 73–75, 79, 90–123, 165– 1966), v
166, 189, 320, 363–364, 369, 378, 381. See Burda, Józef (1915?–1992?), 251
also the subject index Burkhardt, Heinrich Friedrich Karl Ludwig
Barwiska, Regina = (1861–1914), 189
ukasiewiczowa, Regina
Barwise, Jon = Barwise, Kenneth Jon Caen, Herbert Eugene (1916–1997), 399
Barwise, Kenneth Jon (1942–2000), 407, 419 Cantor, Georg Ferdinand Ludwig Philipp
Beneš, Edvard (1884–1948), 325 (1845–1918), vi, 15–16. See also the
subject index

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 477
DOI 10.1007/978-1-4939-1474-6, © Springer Science+Business Media New York 2014
478 Index of Persons

Carnap, Rudolf (1891–1970), v, 325, 334, Drewnowski, Jan Franciszek (1886–1978),

394, 419 33–34
Carnielli, Walter Alexandre (1952– ), 401 Dryjski, Albert (1889–1956), 20
Carver, Walter Buckingham (1879– Dubiel, Wadysaw (1934– ), 261
1961), 162 Dunin-Borkowski, Wadysaw (1884–
Castelnuovo, Guido (1865–1952), 181 1922), 329
Castrillo Criado, Pilar, 388 Duren, Peter (1935– ), 407, 419
Cauchy, Augustin Louis (1789–1857), vi
Chang, Chen-Chung (1927– ), 414 Edwards, Paul (1923–2004), 416
Chojnacki, Father Piotr Julian (1897–1969), Ehrenfeucht, Ewa Krystyna (1938– ) 331,
367, 370 336, 400, 405
Chuaqui Kettlun, Rolando Basim Einstein, Albert (1879–1955), 33, 406
(1935–1994), 399, 400 Enriques, Federigo (1871–1946), 45–46,
Chwiakowski, Zygmunt (1884–1952?) 181–182, 183–184, 186, 188–189, 213.
((. . 0 E# )]), 210, 214–218, See also the subject index
219–224, 273–318, 475 Erdo
 s, Paul (1913–1996), 377–378, 406
Chwistek, Leon (1884–1944), 16, 32, 328– Etchemendy, John William (1952– ), 412
329, 334, 336, 343 Euclid (325?–265? B.C.E.), vi, 45–47, 188,194,
Chwistkowa, Olga, 329 265–267, 278. See also the subject index
Cmorej, Pavel (1937– ), 398 Eudoxus of Cnidus (410?–355? B.C.E.), 45
Coniglione, Francesco (1949– ), 419 Eves, Howard Whitley (1911–2004), 162
Corcoran, John (1937– ), xiv, 38, 389,
403, 407, 418 Feferman, Anita Burdman (1927– ) and
Couturat, Louis (1868–1914), 36 Solomon (1928– ), ix, 3, 36, 38, 174–179,
Curwood, James Oliver (1878–1927), 245 331–333, 336, 376, 396, 400–404, 409
Czarliski, Tadeusz, 252 Feferman, Solomon (1928– ), 167, 361, 407,
Czelakowski, Janusz Michal (1949– ), 414 414, 416, 419
Czerwiski, Sawomir (1885–1931), 226 Fraenkel, Abraham Adolf Halevi (1891–
Czeİowski, Tadeusz (1889–1981), 343– 345 1965), 200, 226
Czubiska, Irena (1961– ), 399 Frege, Friedrich Ludwig Gottlob (1848–
1925), vi, ix
Davies, Norman (1939– ), 3 Frycz, Stefan, 4
Davis, Anne C. = Morel, Anne C. Furman, Grazina Ula, xiv
Dawson, John William, Jr. (1944– ), 414
Dedekind, Julius Wilhelm Richard (1831– Gabbay, Dov M. (1945– ), 386
1916), 238. See also the subject index García-Carpintero, Manuel (1957– ), 375
Dedekind, Richard = Gawecki, Bolesaw Józef (1889–1984), 340
Dedekind, Julius Wilhelm Richard Gerwien, P., 47, 62, 74. See also the
Dehn, Max (1878–1952), 47, 61, 62, 90 subject index
Desargues, Girard (1591–1661). See the Gilewicz, Kazimierz, 245
subject index Givant, Steven Roger (1943– ), ix, xiii, xiv,
De Willman-Grabowska, Helena (1870– 38, 159, 162, 188, 244, 267, 390–391,
1957), 367 402–403, 415, 421
De Zolt, Antonio (1847–1926), 46–47. See Gödel, Kurt (1906–1978), v, ix, 322, 365, 386,
also the subject index 396, 402, 418. See also the subject index
Dickstein, Samuel (1851–1939), 5, 201 Goetz, Arek, xiv
Dobierzewska-Mozrzymas, Ewa Zofia Goldschmidt, Charlotte =
(1936– ), 420 Hausdorff, Charlotte
Domoradzki, Stanisaw, xiv, 406 Goliska-Pilarek, Joanna, 401–403, 409
Doner, John Elliot, 161, 411 Gómez-Torrente, Mario (1967– ), 415
Index of Persons 479

Grattan-Guinness, Ivor (1941–), v Hosiasson-Lindenbaumowa, Janina,

Greczek, Stefan (1867–1967), 57 (1899–1942), 201, 323, 325–326, 334–
Grelling, Kurt (1886–1942), 330, 334, 336 335, 336, 387
Grelling, Richard (1853–1929), 334 Hosiassonówna, Janina =
Greniewski, Henryk (1903–1972), 341– Hosiasson-Lindenbaumowa, Janina
343, 348 Hrycak, Teodor, 255
Gross, Feliks (1906–2006), 397 Huber-Dyson, Verena Esther (1927– ), 403
Grygianiec, Mariusz, 387 Huntington, Edward Vermilye (1874–
Grzymaa, Aleksandr, 269 1952), 358. See also the subject index
Gupta, Haragauri Narayan, 265 Husserl, Edmund Gustav Albrecht (1859–
1938), 37, 361
Hadamard, Jacques Salomon (1865–
1963), 226 Infeld, Leopold (1898–1968), 38, 59
Hadwiger, Hugo (1908–1981), 166 Ingarden, Roman (1893–1970), 361, 367,
Haefeli-Huber, Verena Esther = 370–372
Huber-Dyson, Verena Esther
Hahn, Hans (1879–1934), 393. See also Jaboęski, Aleksander (1898–1980),
the subject index 33–34
Hájek, Petr (1940– ), 398 Jadacka, Wanda (1974– ), 405
Halamon, Perez, 253 Jadacki, Jacek Juliusz (1946– ), xiv, 3, 38,
Haller, Rudolf (1929–2014), 387 174, 190, 324, 387, 393–394, 405, 415,
Hampel, Roman (1907–1963), 176–177 James, Ioan Mackenzie (1928– ), 405
Hardy, Godfrey Harold (1877–1947), vi Janczewska, Antonina =
Hatcher, William S. (1935–2005), 396 Witkowska, Antonina
Hausdorff, Charlotte (18??–1942) Janczewska, Maria = Witkowska, Maria
Hausdorff, Felix (1868–1942), 48, 52–53, Janczewski, Julian (????–1918), 174
60, 65–68, 69–71, 75, 95, 109, 115, 163– Janiszewski, Zygmunt (1888–1920), 5–6,
164, 381. See also the subject index 7, 14, 40, 48–49, 64, 72
Heath, Sir Thomas Little (1861–1940), 265 Jarden, Dov (1911–1986), 126, 130
Heinrich, Wadysaw (1869–1957), 373 Jaroszyska-Kirchmann, Anna Dorota, xiv
Heller, Celia Stopnicka (1922–2011), 3, 38 Jasinowski, Bogumi (1883–1969), 20, 370
Hempel, Carl Gustav (1905–1997), 333–334 Jakowski, Stanisaw (1906–1965), 195,
Hempoliski, Micha (1930–2005), 407 198, 208
Henkin, Leon Albert (1921–2006), 389–390, Jdrzejewicz, Janusz (1885–1951), 215
404, 414 Jerzmanowski, Andrzej (1946– ), xiv, 221
Hertz, Paul (1881–1940), 386 Jerzmanowski, Wiktor = Schayer, Wacaw
Hessenberg, Gerhard (1874–1925), 40 Jezierski, Adam (1948– ), 420
Hilbert, David (1862–1943), v–vi, 37, Jónsson, Bjarni (1920– ), 379, 412
46–47, 62, 79, 184–185, 186, 188, 264, Jordan, Marie Ennemond Camille
348, 388, 419. See also the subject index. (1838–1922). See the subject index
Hintikka, Jaakko Kaarlo Juhani (1929– ), Jørgensen, Jens Jørgen Frederik Theodor
402, 408 (1894–1969), 325, 387
Hitler, Adolf (1889–1945), 330
Hitchcock, David Lancelot (1942– ), xix, 388 Kac, Mark (1914–1984), 38, 208
Hi, Henryk (1917–2006), 192, 404 Kaila, Eino (1890–1958), 325
Hodges, Wilfrid Augustine (1941– ), 404, 411 Kalecka, Zofja, 38
Horn, Alfred (1918–2001), 166 Kalicki, Jan (1922–1953), 389
Hornowski, Micha (1893–1966), 251 Kalicun-Chodowicki, Bazyli (1878–1942), 184
Kant, Immanuel (1724–1804), 37
Karski, Jan = Kozielewski, Jan Romuald
480 Index of Persons

Katz, Marek = Kac, Mark LeĤniewski, Stanisaw (1886–1939), vii,

Kawaguchi, Akitsugu (1902–1984), 226 6–9, 12–13, 17, 36–37, 39, 94, 160, 189,
Kiselev, Andrei Petrovich (1852–1940) 195, 197, 324, 328, 333, 340, 346–349, 357,
() , ! ' $+'# 5), 180 362, 386, 417. See also the subect index
Klein, Felix (1849–1925), 389 Leszczyski, Jan (1905–1990), 367
Knaster, Bronisaw (1893–1990), 14, 36, 156, Levi-Civita, Tullio (1873–1941), 181
167, 348 Lévy, Azriel (1934– ), 161, 412
Kobrzyski, Zygmunt (1893–1944), 198 Lewin Riquelme de la Barrera, Renato
Köhler, Eckehart, 396, 400 Alfredo (1951– ), xiv
Kokoszyęska, Maria (1905–1981), Lichtenstein, Leon (1878–1933), 200–
326–327, 361, 370, 372, 374–376 201, 226
Kokoszyska-Lutmanowa, Maria = Lie, Sophus (1842–1899), 181
Kokoszyska, Maria Lindenbaum, Adolf (1904–1941), 33, 40,
Konarski, Stanisaw, 172 108, 134, 137, 152, 156, 158, 161–165, 199,
Kostant, Ann (1937– ), xiii 326, 331, 335–336, 377, 380
Kotarbiska, Janina (1901–1997), 9 Lindström, Sten (1945– ), 407
Kotarbięski, Tadeusz Marian (1886– Lipiski, Edward (1888–1986), 349
1981), 6–7, 8–9, 12, 17, 32, 34, 36–37, 189, omnicki, Antoni Marian (1881–1941), 51,
320, 335, 341–343, 397, 417 53, 57, 215, 227
Kowalewski, Jan (1892–1965), 14 Lorenz, Johann Friedrich (1738–1807), 265
Kozielewski, Jan Romuald (1914–2000), xiii o, Jerzy (1920–1998), 416
Kozowska, Anna, xiv Löwenheim, Leopold (1878–1957). See the
Kozowski ( pseudonym) = subject index
Greniewski, Henryk Lowry, John (1769–1850), 47, 62
Kozowski, Roman (1889–1977), 321 Lubicz, Bolesaw Zygmunt =
Kozowski, Witold (1919– ), xiv, 320–321, Zahorski, Bolesaw
331, 405 ukasiewicz, Jan Leopold (1878–1976),
Kreisel, Georg (1923– ), 389 vii, 5, 7, 9, 12, 17, 33–39, 51, 72, 94, 160,
Krysicki, Wodzimierz (1905–2001), 261 195–196, 197, 201, 322–325, 331, 337,
Kunugui, Kinjiro (1903–1975), 226 339–341, 345, 370–372, 394. See also the
Kuratowski, Kazimierz (1896–1980), 4–5, subject index
14, 33, 35, 40, 48, 70–72, 103, 198–201, ukasiewiczowa, Regina, 197
328, 333, 340, 361, 382. See also the Lutman, Roman (1897–1973), 327
subject index Lutman-Kokoszyska, Maria =
Kwietniewski, Stefan (1874–1940), 5, Kokoszyska, Maria
183–184, 187–189 Lutmanowa-Kokoszyska, Maria =
Kwietniewski, Wadysaw (1840–1902), 189 Kokoszyska, Maria
Luzin, Nikolai Nikolaevich (1883–1950)
Laczkovich, Miklós (1948– ), 63, 164 ( -!, # # 5), 49, 200
Lampe, Wiktor (1875–1962), 41, 393
Langford, Cooper Harold (1895–1965), 417 Mackiewicz, Witold (1941– ), 409
Lapointe, Sandra, 416 Maddux, Roger Duncan (1948– ), 400
Lebesgue, Henri Léon (1875–1941), 14, 33, Magidor, Menachem (1946– ), 38
52, 75. See also the subject index Mahlo, Paul (1883–1971), 126
Legendre, Adrien-Marie (1752–1833), 265 Malinowski, Grzegorz, 414
Lejewski, Czesaw (1913–2001), 198 Mancosu, Paolo (1961– ), xiv, 416
Leniewska, Anna Kazimiera = Marchisotto, Elena Anne Corie (1945– ), xiv
Sierpiska, Anna Kazimiera Marciszewski, Witold (1930– ), 392
Leniewska, Zofia (1893–1958), 9 Marcus, Joseph (1923– ), 172
Martel, Karol (1916– ), 173
Index of Persons 481

Mates, Benson (1919–2009), 389 Neumann, Johann von =

Mathé, Svätoslav (1943– ), 398 Von Neumann, John
Matuszewska, Maria Irena = Neurath, Otto (1882–1945), 341, 393
Maria Mostowska Neyman, Jerzy (1894–1981), 39
Mazurkiewicz, Stefan (1888–1945), 5–6, Nikliborc, Wadysaw (1889–1948), 320
7–9, 12, 14, 33, 35, 40, 48–49, 62, 64, 72, Nikodym, Otton Marcin (1887–1974),
94, 132, 200–201, 379, 394. See also the 193, 200
subject index Nowak, Leszek (1943– ), 365
McFarland, Andrew (1978– ), iii–iv, viii, xiv
McFarland, Joanna (1977– ), iii–iv, viii, xiv Ovchinnikov, Sergei V., xiv
McFarland, Maria Anna, ii
McKenzie, Ralph Nelson Whitfield (1941– ), Paderewski, Ignacy Jan (1860–1941), 12, 197
ix, 390 Pasch, Moritz (1843–1930). See the subject
McNulty, George Frank (1945– ), 412 index
Mehlberg, Henryk (1904–1979), 344, 367 Pasenkiewicz, Kazimierz (1897–1995), 31–
Melnick, Norman (1928–1997), 405–406 32, 33, 36
Mendelson, Elliott (1931– ), 391 Patterson, Douglas Eden, 416–417
Mendez, Edith Prentice, xiv Pawlikowska-Broek, Zofia (1941– ), 406
Menger, Karl (1902–1985), 164, 200, 226, 394 Peano, Giuseppe (1858–1932), vi, 204, 369.
Meschkowski, Herbert (1909–1990), 63 See also the subject index
Mianowski, Józef (1804–1879), 189 Peirce, Charles Sanders (1839–1914), vi
Mihuowicz, Jerzy (1877–1950), 237 Petraycki, Leon (1867–1931), 7
Mnatsakanian, Mamikon A. (1942– ), 163 Piekowski, Stefan (1883–1953), 7–8, 394
Moese, Henryk (1886–19??), xi, 43, 45, Pieri, Mario (1860–1913), 184, 186–188, 194
125–126, 129–130, 131–134, 142–152, Pigozzi, Don Leonard (1935– ), 411
155–157, 162, 206, 211, 248, 263 Pisudski, Józef Klemens (1867–1935), 4–5,
Mongré, Paul = Hausdorff, Felix 8, 10, 14, 175
Monk, James Donald (1930– ), 390, 412 Pincherle, Salvatore (1853–1936), 181
Monoide, S. P., 399 Piotrowski, Tadeusz (1940– ), 224
Montague, Richard Merritt (1930–1971), 333 Pla i Carrera, Josep (1942– ), 417
Montlak, Bronisawa, 199 Playfair, John (1748–1819). See the
Montlak, Dawid, 199 subject index
Moore, Gregory H. (1944– ), 416 Powa, Franciszka (1845–1926), 57
Morawiec, Adelina, 403 Polonsky, Antony (1940– ), xiv
Morel, Anne C., 167 Popoff, Kyrille (1880–1966) ($#%# , '),
Morse, Anthony Perry (1911–1984), 381 200–201, 226
Mostowska, Maria Irena (1910?– ), 199 Porbska-Srebrna, Joanna (1950– ), 403
Mostowski, Andrzej Stanisaw (1913– Preller, Anne, 389
1975), 161, 198–199, 328, 331, 335–336, Presburger, Mojz esz (1904–1943?), 190,
380, 389, 413, 416 195–196, 200–201
Murawski, Roman (1949– ), 416 Prewysz-Kwinto, Zofia = Leniewska, Zofia
Prussak, Róa = Teitelbaum, Róa
Nagel, Ernest (1901–1985), 325 Purdy, Robert, 386
Nawroczyski, Bogdan (1882–1974), 4, 12 Purkert, Walter (1944– ), 59
Nelson, Leonard (1882–1927), 37, 334 Putnam, Hilary Whitehall (1926– ), 389
Németi, István (1942– ), 390 Pythagoras (569? B.C.–475? B.C.) See the
Nerode, Anil (1932– ), 400–401 subject index
Nešet ilová, Helena (1948– ), 402
Neumann, János Lajos = Quine, Willard Van Orman (1908–2000),
Von Neumann, John 333, 397
482 Index of Persons

Raabe, Henryk Wacaw (1882–1953), 173, Sierpięski, Wacaw Franciszek (1882–

177, 180 1969), 7–14, 31, 33, 37, 39–40, 48–49, 54,
Rav, Yehuda (1930– ), 419 57, 60, 62–64, 70, 72, 94, 161, 164–165,
Reichenbach, Hans (1891–1953), 325, 193, 200–201, 226, 320, 335, 339,
334, 387 376–382, 397. See also the subject index
Reidemeister, Kurt Werner Friedrich Simmons, Keith Eric George (1952– ), 418
(1893–1910), 393 Sinaceur, Hourya Benis (1940– ), 388, 400–
Réthy, Móricz (1846–1925), 62 401, 406–407, 418–419
Rickert, Heinrich (1863–1936), 365 Skabowski, 10
Rickey, V. Frederick (1941– ), xiv Skolem, Thoralf Albert (1887–1963). See the
Robinson, Raphael Mitchell (1911–1995), subject index
165, 381 Supecki, Jerzy (1904–1987), 198
Rodríguez-Consuegra, Francisco A. (1951– ), Smith, Helen Marie Patteson (1940– ), ii, xiv
389, 418 Smith, James Thomas (1939– ), iii–iv, xiii,
Rosen, Saul (1922–1991), 406 125, 163, 475
Rougier, Louis (1889–1982), 325 Smoleska, Agnieszka (1904–1995), 174–175
Rowe, David E. (1950– ), x Sobociski, Bolesaw (1906–1980), 198
Róycki, Ludomir (1884–1953), 201 Sosnowski, Pawe (1859–1947), 38
Rudniaski, Jarosaw (1921–2008), xiv, 320 Srebrny, Marian (1947– ), 401–403, 409
Rusiecki, Antoni Marjan (1892–1956), 59, Srzednicki, Jan Tadeusz Jerzy (1923–
169, 201, 204–208, 209, 221, 243–248 2008), 386
Russell, Bertrand Arthur William Stachniak, Zbigniew (1953– ), 386
(1872–1970), vi, 335, 418 Steckel, Samuel, 245
Ruziewicz, Stanisaw (1889–1941), 193 Steinberg, Witold, 367
Steinhaus, Wadysaw Hugo Dyonizy
Sadowska, Joanna, 172 (1887–1972), 57, 59, 329
Scanlan, Michael (1957?– ), 418 Steinhausówna, Olga = Chwistkowa, Olga
Schayer, Alicja (1931–1944), 221 Stpniewski, Jarosaw, 367, 370
Schayer, Genowefa = Stojda, Wadysaw, 251–253
Zajc-Jerzmanowska, Genowefa Stone, Marshall Harvey (1903–1989), 165
Schayer, Krystyna (1933– ), 221 Stoek, Wodzimierz (1883–1941), 57, 320
Schayer, Stanisaw (1899–1941), 214 Straszewicz, Stefan (1889–1983), 58, 169,
Schayer, Wacaw Maria (1905–1959), 169, 189, 192–193, 204–209, 210, 243–248
190, 208–210, 214–220, 221–224, 273–318 Stroiska, Maria Magdalena, xii, 388
Schinzel, Andrzej Bobola Maria (1937– ), xiv Sujczyska, Monika, 392
Schlick, Moritz (1882–1936), 325 Suppantschitsch, Richard =
Scholz, Heinrich (1884–1956), 197 Zupani, Rihard
Schönflies, Arthur Moritz (1853–1928), 60 Suppes, Patrick (1922– ), 407, 412, 419
Schröder, Friedrich Wilhelm Karl Ernst Suslin, Mikhail Yakovlevich (1894–1919)
(1841–1902), vi ((-)!, 1 D#  5), 64. See
Schwabhäuser, Wolfram (1931–1985), also the subject index
267, 390 Süss, Wilhelm (1895–1958), 126
Scott, Dana Stewart (1932– ), 167, 333 witalski, Kazimierz Stanisaw (1886–
Sergescu, Petre (1893–1954), 226 1962), 201
Shilleto, James Robert (1941– ), xiv, 134, 163 Szczerba, Leslaw Wlodzimierz (1938–2011),
Sica, Giandomenico, 418 162, 267, 412
Sierpiska, Anna Kazimiera (1888–1961), 49 Szmielew, Borys, 199
Sierpiski, Mieczysaw (1912–1983), 49 Szmielew, Wanda Montlak (1918–1976),
198–199, 267, 328, 336, 390, 412
Index of Persons 483

Sztejnbarg-Kamiska, Dina = Voronoy, Georgy Fedoseevich (1868–1908)

Kotarbiska, Janina ( #'#!#, #'  .# #)? 5), 49
Szulczyski, Zygmunt (1897–1967), 193
Wagon, Stanley (1951– ), 54, 63, 66, 164
Tajtelbaum, Alfred = Tarski, Alfred Wajsberg, Mordchaj (1902–1942), 33– 34,
Tajtelbaum-Tarski, Alfred = Tarski, Alfred 198, 346–347
Tarska, Ewa Krystyna = Wallace, William (1768–1843), 47, 62
Ehrenfeucht, Ewa Krystyna Waraszkiewicz, Zenon Jan (1909–1945), 132,
Tarska, Maria (1902–1990). See the 134, 152, 156, 162
subject index Warmus, Mieczysaw (1918–2007), 252
Tarski, Alfred (1901–1983), i, 13, 327, 384. Watt, Richard M. (1930–), 3
See also the subject index Wglowska, Danuta, 406
Tarski, Jan (1934– ), xiv, 31, 33, 177, 331, Weierstrass, Karl Theodor Wilhelm (1815–
336, 374, 388–389, 391–393, 396, 400, 1897), vi
405, 407–408 Weigl, Rudolf Stefan (1883–1957), 57
Tarski, Janusz Andrzej = Tarski, Jan Wells, Benjamin Franklin, III, 408
Tarski, Wacaw (1903–1944), 3, 336 Weryho, Wladyslaw (1868–1916), 20
Teitelbaum, Alfred = Tarski, Alfred White, Morton Gabriel (1917– ), 396
Teitelbaum, Ignacy (Isaac) Mayer Whitehead, Alfred North (1861–1947), vi
(1869–1942), 3, 174, 336, 392 Wiegner, Adam (1889–1967), 367
Teitelbaum, Róa (1879–1942), 3, 174, Wilkosz, Witold (1891–1941), 326, 367, 369
336, 392 Wilson, Trevor Miles, 164
Teitelbaum, Wacaw = Tarski, Wacaw Wirszup, Izaak (1915–2008), 125–126, 130,
Thaler, Michael (1935– ), xiv 135, 144, 151, 212
Thales (c. 624 B.C.E.–c. 546 B.C.E.) See the Witkacy = Witkiewicz, Stanisaw Ignacy
subject index Witkiewicz, Stanisaw Ignacy (1885–1939),
Tonelli, Leonida (1885–1956), 226 329, 400
Trotsky, Leon (1879–1940) ( *'#3,  ), Witkowska, Agnieszka =
8, 10 Smoleska, Agnieszka
Twardowski, Kazimierz (1866–1938), 9, Witkowska, Antonina (????–1957), 175
12, 17–18, 36–37, 197, 201, 327–328, Witkowska, Helena = Woroniecka, Helena
344–345, 373–374, 393, 396, 416 Witkowska, Jadwiga (1909–1988), 174
Witkowska, Józefa Maria = Zahorska, Józefa
Ulam, Adam Bruno (1922–2000), 333–334 Witkowska, Maria (1873–1911), 174
Ulam, Stanisaw Marcin (1909–1984), Witkowska, Maria Józefa = Tarska, Maria
333–334 Witkowski, Antoni (18??–????), 174
Witkowski, Antoni (1905–1984), 174
Van den Dries, Lou P., 412 Witkowski, Wincenty (1841–1919), 174–175
Van Heijenoort, Jean Louis Maxime Witwicki, Tadeusz (1902–1970), 361, 374
(1912–1986), 400 Wojciech, Saint (c. 956–997). See subject
Vaught, Robert Lawson (1926–2002), 361, index
412, 419 Wojciechowska, Agnieszka, 408
Veblen, Oswald (1880–1960). See the Wójcik, Andrzej (1958–2009), 408
subject index Wojeski, Teofil (1890–1963), 173
Vega Reñón, Luis, 388 Wojtowicz, Wadysaw Baltazar (1874–1942),
Vitali, Giuseppe (1875–1932), 52, 75, 95, 115 58, 127–128, 180, 183–184, 188–189,
Vivanti, Giulio (1859–1949) 204, 245
Von Neumann, John (1903–1957), 164, 333– Woleski, Jan (1940– ), ix, 341, 374, 388, 396,
334, 378, 381 400–402, 408–409, 416, 419
484 Index of Persons

Woodger, Joseph Henry (1894–1981), 331, Zarzecki, Adam, 208

368–369 Zawirski, Zygmunt Michal (1882–1948),
Woroniecka, Helena (1911– ), 174 325–326, 370–373, 387
Woytak, Lidia Teresa (1946– ), 409 Zermelo, Ernst Friedrich Ferdinand (1871–
Wycech, Czesaw (1899–1977), 221 1953), 33, 66, 95, 210, 334. See also the
subject index
Yanigahara, Kitizi, 126 eromski, Stefan (1864–1925), 173
Young, William Henry (1863–1942), 226 Zimmermann, Robert von (1824–1898), 18
Znamierowski, Czeslaw (1888–1967), 20
Zaanen, Adriaan Cornelis (1913–2003), ótowski, Adam (1881–1958), 367
54–55 Zupani, Rihard (1878–1949), 180
Zahorska, Józefa (1900–1957), 174, 175–177, Zydler, Jan (1867–1934), 180, 219
331, 336 Zygmund, Antoni (1900–1992), 130
Zahorski, Bolesaw (1887–1922), 175–177 Zygmunt, Jan, xiv, 194–195, 386, 391, 398,
Zajc-Jerzmanowska, Genowefa, 221 409, 420
Zaremba, Stanisaw (1863–1942), 49,
329, 369
 Index of Subjects

This index lists subjects considered in the main text of the book. It does not cover the
entries in the bibliography. Some index headings are personal names that identify
subject-matter content. For other individuals, consult the index of persons that begins
on page 477. Because mathematicians often give commonplace words special meanings,
this index includes headings for some adjective as well as substantive phrases. Subhead-
ings under Poland, Tarska, and Tarski are ordered topically. All headings and all other
subheadings are alphabetized, ignoring spaces, punctuation, diacritical marks, and initial
articles and prepositions.

absolute full, 278

value, exercise on, 250 interior of, 264–265
See also under measure nonconvex, 265
absolutely measurable. See under set straight, 278
Academy of Sciences vertex of, 264
See under Polish; Soviet antinomy
action, theory of, 341–342 of heterological concepts, 334, 357
Adalbert, St. See Wojciech of the liar, 357, 362
additivity. See under measure antireflexivity. See under axiom; theorem
adequacy, material, 362–363 antisemitism, ix, 8, 10–11, 36, 69, 181, 328,
algebra 330–331, 334–336, 387, 401–402
cardinal, 379–380, 412 See also Holocaust
of classes, 358 antisymmetry. See under axiom
symbols for expressions in, 359 apagogic reasoning, 138
cylindric, 390, 411–412, 417 apothem, 310
ordinal, 411 arc, 275
universal, 377–382, 411–412 Archimedes, 213. See also under axiom
algebraic surface. See surfaces, algebraic archives, xiv, 377, 473
almost equivalent. See under denumer- area
able decomposition; symbol theory, xi, 45–48, 67, 79–83, 125,
alphabetization, xiii 219–220, 222–223, 301–318
altitude. See under parallelogram; unit of, 304
rectangle; trapezoid; triangle See also under axiom system; circle; cone;
Amaldi, Ugo. See under axiom system cylinder; parallelogram; polygon; poly-
and in the index of persons gonal regions; rectangle; rhombus;
American Mathematical Monthly, 203 trapezoid; triangle
Amersfoort. See under congresses arithmetic
Amsterdam, xxii–xxiii cardinal, 161, 194, 200, 377–382
Anders Army, 34, 224, 320 integer, 247, 251, 255
angle of natural numbers, 196, 386
arm, 264 rounding in, 236
bisector, 275 of segments, 216–217, 296
central, 275 See also under axiom system; completeness;
concave, 289 decidability; incompleteness; number:
convex, 264–265 irrational, natural

A. McFarland et al. (eds.), Alfred Tarski: Early Work in Poland—Geometry and Teaching, 485
DOI 10.1007/978-1-4939-1474-6, © Springer Science+Business Media New York 2014
486 Index of Subjects

arm. See under angle Baden School. See philosophy: neo-Kantian

art theory, 329 Banach, Stefan
assimilation, 38 space, 57
Association for Symbolic Logic, 334 See also under measure; paradox; theorem
Austria, xxiii and in the index of persons
Empire, xxii base. See under parallelogram; rectangle;
axiom, 15–16 trapezoid; triangle
antireflexivity, 29 beam of a ship, 315
antisymmetry, 16, 21 Berkeley, 125, 339, 376, 397
of Archimedes, 45–47, 127, 216 Cragmont Avenue, 376
of choice, vi, 40, 48, 52, 54, 65–67, 71–72, See also under congresses; universities
75, 79, 95, 161, 199, 377, 380–381, 417 Berlin, xii–xiii, 330
circle, 277–278 See also under philosophy; universities
congruence, 186 Bernstein, Felix. See under theorem and in
continuity, 46, 213, 230, 238–239, 278 the index of persons
of De Zolt, 46, 48, 51–52, 61, 79, 83, 90, 95 betweenness, 183, 186, 194, 266–267
incidence, 186 bibliography
order, 16, 21, 25, 30 citations, xii, 421
parallel conventions, xiii, 421
Euclid’s version, 265 biographical sketches, xiii
exercise on, 264–267 Biographical Dictionary of Polish
Playfair’s version, 265, 267 Mathematicians, 428
Tarski’s version, 248, 264–267 biology, foundations of, 331, 368–369
of Pasch, 267 Birkhäuser. See Springer
transitivity, 16, 21 bisector. See under angle
trichotomy, 16 Bologna, xxii–xxiii. See also under congresses
well-ordering, 21, 25, 30 Bolyai, Farkas Wolfgang. See under theorem
axiomatic method. See postulate theory Bonn. See under universities
axiom system boundary. See under set
for algebra of classes, 358 bounded. See under set
for area, 220, 305 boxed text, xiii, 1, 170, 243, 274, 337, 421
for arithmetic, 58 Breslau, xxii–xxiii
for Boolean logic, Hilbert’s, 348 broadcasting, 369
for geometry Budapest, xxii–xxiii
of Enriques and Amaldi, 183
Euclid’s, 264–267 Cambridge
Hilbert’s, 46–47, 184–186, 194, 264 England, xxii–xxiii (see also under
Pieri’s, 186, 188, 195 congresses)
Tarski’s, 194–195, 208, 267, 390 Massachusetts (see universities: Harvard)
Veblen’s, 194 Cantor, Georg. See under theorem
for inclusion, Huntington’s, 358 and in the index of persons
independent, 16, 22, 25 capitalization, xiii
interpretation of, 27 cardinal. See under algebra; arithmetic;
for magnitude, 51 equivalence
for proper inclusion, Tarski’s, 358 case-ridden arguments, 247, 256–257,
for set theory, 199 260, 268–270, 272
relatively weakest, 29–30 categorical. See under theory
for well-ordering, Tarski’s, 15–17, category
19–30, 161 first (see under set)
See also postulate theory semantic, 9
Index of Subjects 487

Catholicism, 34, 179, 321, 326, 369 1923 1st Polish Philosophy, Lwów, 41,
See also Tarski: religion 53, 324
causal relation, 343–344 1927 1st Polish Mathematics, Lwów, 195,
causality, 324, 343–344 226, 324
ceiling, 127. See also under symbol 1927 2nd Polish Philosophy, Warsaw, 195,
center. See under circle 323, 341–345
central. See under angle 1928 8th International Mathematicians’,
Chicago. See under congresses; universities Bologna, 167, 226, 324
choice. See under axiom 1929 Slavic Mathematicians, Warsaw/
chord. See under circle Pozna, 171, 196, 200–202, 225–227, 322
Cipher Bureau. See code breaking 1931 2nd Polish Mathematics, Vilnius,
circle, 277 207, 324
area, 237 1934 8th International Philosophy, Prague,
center, 277 177, 324–325, 387
chord, 275 1934 Unity of Science Vorkonferenz, Prague,
circumference, 206, 212–213, 222, 229–241 177, 324–325
diameter, 275 1935 1st Unity of Science, Paris, 177,
exterior of, 277 324–325
interior of, 277 1936 2nd Unity of Science, Copenhagen, 326
line, relationship to a, 278–284 1936 3rd Polish Philosophy, Cracow,
secant, 282 323–324, 326, 365–376
tangent, 282 1937 9th International Philosophy, Paris,
radius, 277 324, 326
circles 1937 3rd Unity of Science, Paris, 324, 326
intersecting, 216, 277, 280 1938 4th Unity of Science, Cambridge, 326
relationship of two, 280 1938 Modern Reasoning, Amersfoort,
tangent, 280 324, 330
circle-squaring. See under Tarski 1939 5th Unity of Science, Harvard, ix, 324,
circular polygon, 157 326, 333–336
circumcenter. See under triangle 1946 Princeton Bicentennial, 379, 388, 406
citations. See under bibliography 1955 Reasoning in Mathematics, Paris, 389
closed system of theorems, 216–217, 1957 Summer Institute, Cornell, 402–403
223, 283–284 1965 Gödel Theorems, Chicago, 389
closure operator, 72 1965 Set Theory, London, 389
code breaking, 8–10, 14, 49 1971 Tarski Symposium, Berkeley,
commensurable. See under segments 404–405, 414
commutation, law of, 348 1983 Tarski Commemoration, Stanford,
compass and straightedge. See construction 407, 419
completeness, 377, 388, 418 2001 Tarski Centenary, Warsaw, 405,
of real arithmetic, 194, 196, 331, 418 413, 415
computer-assisted design, 261 2007 State of Jefferson Mathematics,
computer science, 167, 196, 343, 346, Whiskeytown, xiii
363, 391, 402, 406, 408, 414 congruent, 65–66, 80, 96, 183, 186, 194
concave. See under angle See also under symbol
conditional. See under probability; sentence conjunction
conditions, necessary and sufficient, 291 Tarski’s definition of, 39
cone, area and volume of, 237 See also under symbol
congresses connected. See under polygonal regions
1921–1939 annual Polish secondary-school
teachers’, 203
488 Index of Subjects

consequence, 366–367 See also under definitions

direct and indirect, 366–367 defect. See under triangle
infinitary concept of, 367 definitions, 15–16, 388
logical, 195, 326, 344, 365, 388–389, 412 recursive, 165, 167, 340, 359, 362
material, 344 in deduction theory, 339, 346–349
consistency proofs, 340–341, 361, 367, 388 degree
construction problems, 188, 216–217, of equivalence (see under finite decomposi-
279–281, 285–291, 301 tion: into polygons; symbol; tau)
compass and straightedge, 285 of irregularity, 157 (see also under symbol)
set square, 285 Delta, 126
contiguity. See phenomenon deltoid. See kite
continuity. See under axiom denumerable decomposition
continuum hypothesis, 200 almost equivalent by, 75, 117, 123
contrapositive, 291 equivalent by, 65–66, 74–76, 95,
converse, 291 115, 163, 377
convex. See under angle; polygonal regions into closed sets, 123
coordinates, 186 into Lebesgue measurable sets, 123
Copenhagen Desargues, Girard. See under theorem
See under congresses; universities De Zolt, Antonio. See under axiom
Cornell. See under congresses diagonal of a square, 274, 292–295
coryphaeus, 12 diameter. See under circle; polygon; symbol
Cracow, xxii–xxiii, 38, 57 difference. See under polygonal regions;
Circle (see under philosophy) sets; symbol
See also under congresses; Mathematical direct. See under isometry
Society; universities disjoint. See under sets; symbol
curriculum disjunction, 347. See also under symbol
of Polish schools, 207, 210, 214–215, disk, 277
222–223, 230–231, 278, 288 equidecomposibility with a square, 62, 90
at Third Boys’ Gimnazjum, 179–180 displaced persons. See under World War II
curve. See under length distance
cuts equal, between pairs of points, 80
Dedekind, 230, 238 point to line, 217, 273
method of, 213, 230, 238, 241 point to point, 217, 273–274
cylinder, area and volume of, 237 docent, 41
cylindric. See under algebra double-subdivision method, 60, 81, 138
Czechoslovakia, 330, 398 draft of a ship, 315
duel, Chwistek’s, 329
daemon, 408
Danzig, 193 Economists and Statisticians Association, 349
decidability, 199, 377, 388, 411 editorial comments, xii
of real arithmetic, 196, 412, 414, 418, 420 Ekonomista, 349
decision procedure. See decidability elementary. See under language
Dedekind, Richard. See cuts empiricism, 370–373, 396, 400
decomposition empty. See under set; symbol
equivalence by, xi Encyclopedia of Unified Science, 393
See also denumerable decomposition; Enriques, Federigo. See under axiom system
finite decomposition and in the index of persons
deduction entirety. See floor
theorem, 160 equality. See under symbol
theory of, 36, 160, 339–341, 344, 346–349 equicomplementability, 47
Index of Subjects 489

equidecomposibility. See finite decomposi- into polygons, 341

tion: into polygons: equivalence by and equivalence by, xi, 45–47, 59
under finite decomposition: into disjoint degree of, 45, 125–129, 132–158,
sets: equivalence by 162–163, 206, 211–212, 263
equidistance (see also tau and under symbol)
ternary, 186 methods for, 157
from three lines, exercises on, 268–271 See also area: theory
equidistant curve, exercise on, 256 into polyhedra, equivalence by, 90
equivalence into spherical polygons, 112
of axiom systems (see postulate theory) finiteness, definition of, 40–41, 161
cardinal, 70 first element, 15, 21–22, 25
connective, 346–347 flea-feeder, 57
material, 39, 160 floor, 247, 253. See also under symbol
See also under decomposition; denumer- Formism. See art theory
able decomposition; finite decomposi- formula
tion; symbol definition, 340
Erlanger Programm, 389 well-formed, 341
Euclid. See under axiom; axiom system See also under quadratic
and in the index of persons foundations. See under geometry;
Europe, Central, maps of Leniewski; mathematics; physics;
1914, xxii psychology; theology
1924, xxiii fractal, 69
event, 341 function
generalized, 341–342 measurable, 54
excluded middle, law of, 357, 360, 365 of names, 359
exercise, use of the word, 243 nonextensional, 357
exhaustion, method of 45 propositional, 195
Exposition, General Polish. See congresses: sentential, 358–359
1929 Slavic Mathematicians’ Tarski’s usage, 357
exterior. See under circle functional 54–56
Fundamenta Mathematicae, vii, 14, 62,
factorization, exercise on, 251–253 64, 93–94
figures
geometric, 80 Gdask. See Danzig
numbering of, xx Gdynia, xxiii, 193, 333
filament. See under Tarski Geometrja (coauthored by Tarski), xiii,
filter, prime, 165 214–224, 273–318, 475
finite decomposition, 43, 45 contents, 216–219, 223, 274–275, 291, 301
into disjoint sets later editions, 224
equivalence by, xi, 62, 66, 95–105, prerequisites, 215–217
163, 165–166 style, 291
in dimension 1, 73 geometry
in dimension  2, 95, 106–109, 363 analytic, 268–271
in dimension 2, 53, 58–63, 65, 77–91 based on notion of solid body, 40, 195
in dimension 3, 90–91, 381 foundations of, 162, 181, 184–188, 194, 199,
in dimension  3, 65, 73–74, 93, 267, 390, 412
95, 109–112 non-Euclidean, 188, 265
transitivity of, 60 texts, 180, 183–184
on a sphere, 74, 95, 112–114 transformational, 186, 210
using Lebesgue measurable sets, 123 See also under axiom system
490 Index of Subjects

Germany, xxiii hyperbolic paraboloid, 268, 271

Empire, xxii hyperinflation, 35
Gerwien, P.. See under theorem and
in the index of persons idealism, transcendental, 365–367
gimnazjum, xiii, 3, 172 identity law, 348
Kalecka, 38 implication connective, 30, 347
Limanowski, 221 See also under symbol
Mazowieckie, 4–5, 12, 392 incidence, 183, 186
State, Number 4, 3, 392 incircle. See under triangle
Third Boys’, 37, 173–174, 176–180, 203 inclusion relation. See under axiom system
Zamoyski, 218–219, 343 incommensurable. See under segment
eromski, 177, 320, 328, 331 incompleteness of integer arithmetic
goals of this book, x, 319 322, 365, 367, 386, 388, 402
Gos nauczycielski, 203 independence. See under axiom system;
Gödel, Kurt. See incompleteness of integer primitive notions
arithmetic and in the index of persons index conventions, xiii, 477, 485
Göttingen, xxii–xxiii individual, 341
See also under universities generalized, 341–342
grosz, 318 inductive reasoning. See under logic
group inequalities, 247, 250, 253, 256, 262
amenable, 164 exercises on, 257–260
of direct isometries, 55 inequality. See under symbol
measurable, 164 inequivalence. See under symbol
of rotations, 67–68 inference rules, 367
symmetry, 272 influenza, 14
transformation, 164–165, 181 integer. See under arithmetic;
of translations, 55 insurance. See under mathematics
theory, 199 integral
Vierergruppe, 272 Lebesgue, 54–55
gymnasium. See gimnazjum See also under symbol
interior, 85
habilitation. See under Tarski: as researcher See also under angle; circle; symbol
Hahn, Hans. See under theorem interpretation. See under axiom system
and in the index of persons intersection. See under circles; symbol
Handbook for Self-Education, 183, 189 interval, multidimensional, 74, 115
Hanover, 224 intuitionism, v
Harvard. See under congresses; universities invariant. See under measure
Hauber’s law, 217, 283 irrational. See under number
Hausdorff, Felix. See under paradox irregularity, degree of, 157
and in the index of persons isometry, direct, 80
height. See under parallelogram
hemisphere, 129, 133, 146–147 Jahrbuch über die Fortschritte der
heterological. See under antinomy Mathematik. See JFM.
Hilbert, David. Jerusalem, 224
problems, 47 Jews, 3, 5, 8, 34, 36, 38, 57, 69, 130, 173, 181,
See also under axiom system 330, 333–336, 345, 387, 405, 409
and in the index of persons JFM, v, 16, 63, 421
Holocaust, 34, 57, 69, 334–336, 345, 374 Jordan, Camille. See under Peano
Huntington, Edward V. See under Journal of Symbolic Logic, 404, 411
axiom system
Index of Subjects 491

Kpno, 130–131, 144, 263 logic, 5, 12, 33

Kiev, xxii–xxiii. See also under universities algebraic, 34, 412, 414
kite, 157 Aristotelian, 197
area, 313 Boolean (see deduction: theory of )
Königsberg, xxii–xxiii California school of, ix
Koszykowa Street. See under Warsaw deductive, 372
Kraków. See Cracow fundamental concepts of, including those of
Kristallnacht, 331 metamathematics, 36–37, 160, 196, 200,
Kuratowski, Kazimierz. See under theorem 322–326, 335, 356–363, 367–368, 386,
and in the index of persons 402, 411–412, 414, 417–419
game, 340–341
language inductive, 335
closed, 40 intuitionistic, 34
elementary, 195 Leniewski’s, 9, 39, 160, 324, 349, 386
equational, 390 modal, 34
formal, 357, 362–363 multivalued, 34, 197, 325, 348, 372,
order of, 363 386–387
natural, 40, 357 nonclassical, v
object, 357, 362–363 philosophical, 37, 160, 327, 339, 345, 404,
philosophy of, 9 412–413, 415–417
See also metalanguage and semantics Polish school of, vi–vii, 12, 14, 18, 49, 192,
Latino sine Flexione. See Lingua Peano 195, 197, 322, 340, 345, 374, 393, 419
lattice, complete, 167 propositional. See deduction: theory of
Lebesgue, Henri. See under measure sentential. See deduction: theory of
and in the index of persons in teaching geometry, 216, 223, 247, 291
length See also under Tarski: as researcher
of curve, 213 logical
unit of, 219, 294, 296 notion, 372, 389
See also under segment positivism, 323, 325, 373, 387, 393
Leniewski, Stanisaw See also truth
death, 333 logicism, vi
seminar, 21, 33, 36, 161 logistic. See logic: Leniewski’s; Peano:
system of foundations, 9, 36, 39, 386 school of
Tarski, intellectual debt to, 39 London, xxii–xxiii
See also under logic; Tarski See also under congresses
and in the index of persons Löwenheim, Leopold. See under theorem
letterspacing, e x p a n d e d, xii, 380 Lublin, xxii–xxiii
liceum, 172, 215, 222 ukasiewicz, Jan
limits See also under Tarski: as researcher;
method of, 212, 230–231, 241 theorem and in the index of persons
theory of, vi, 212–213, 231, 238–239 Lviv. See Lwów
line, as primitive notion, 183, 186 Lwów, xxii–xxiii, 38, 53
Lingua Peano, 204 See also under congresses; Mathematical
Lithuania, xxiii Society; mathematics; Philosophical:
locus, 277 Society; universities
ódz, xxii–xxiii, 3
See also under university magnitude. See under axiom system
logarithms, 247 mark, Polish, 35
exercise on, 253 Marszakowska Street. See under Warsaw
Marxist commentary, 208
492 Index of Subjects

Matematikai és physikai lapok, 203 See also Peano: Peano– Jordan content
Matematyka: Czasopismo dla nauczycieli, measurement theory, 320
208, 248, 408 median. See under triangle
Matematyka i szkoa, 207, 210 mereology, 195
material. See adequacy and under See also logic: Leniewski’s
equivalence metalanguage, v, 9, 40, 357, 362–363
Mathematica, 274 metalogic, v, 368
Mathematical Gazette, 203 metamathematics, v–vi, 341, 388, 419
Mathematical Reviews. See MR. See also logic: fundamental concepts
Mathematical Society metaphysics, 370–372
Cracow, 57 method. See action, theory of
Polish, in Lwów, 322–324 metric, 65
Polish, in Warsaw, 161, 167, 208, Mianowski Fund, 7, 12, 36, 187–189
210, 321, 403 Ministry of Religion and Education, 12, 20,
mathematics 94, 179, 189, 192–193, 197, 204, 207–210,
characteristically Polish, vi, 4–5, 14, 214–215, 221, 224, 226, 243?, 246, 345,
49, 64, 72, 204, 227, 379 369, 394–395
education Minsk, xxii–xxiii, 10, 174–176
in Italy, 47, 59, 181–183, 186–188 Mody matematyk. See Parametr
in Poland, v, 53, 57–59, 126–127, 130, model theory, 34, 160, 389, 412, 414, 418–419
137, 169, 171–172, 183–184, 188–189, moon, trip to the, 397
191–192, 202–210, 219, 221, 230–231, Moscow, xxii–xxiii
241, 243–246, 248, 273–274, 369 MR, 421
summer courses, 192–193, 207 Munich agreement, 330
in the United States, 130, 350 Münster, 197. See also under universities
foundations of, 5, 14, 49, 60, 64, 184, 368, mutual convergence. See under sequence
390 (see also under Leniewski)
insurance, 14, 18, 69, 196, 343, 349–356 names
Lwów school of, 57 personal, xii
social history of, x, xii, 223 place, xii
See also under philosophy nationalism in academia, 402
maximal principle, 54, 72 National Pedagogical Institute, 36–38,
Mazurkiewicz, Stefan. See under paradox 184, 210
and in the index of persons National Schoolbook Publisher, 214–215,
mean, arithmetic, 232 273, 276
meaning, 366, 374–376, 417 Nauka polska, 189
measurable. See under function; set negation, 347
measure, 52, 60, 67, 75, 83, 165 Tarski’s definition of, 39
absolute, 166, 364 See also under symbol
Banach’s, 165, 363–364 neo-Kantianism. See under philosophy
finitely additive, 52–55, 60, 67 New York City, 332–334, 396
fully additive, 52 See also under universities
invariant, 53, 164–165, 363 nonconvex. See under angle
Lebesgue, 52, 106, 117, 123, 363 nonmeasurable. See under set
Lebesgue inner, 76, 122–123 nowhere dense. See under set
problem, 51–57, 60, 65, 67, 69, 75, Nowi Targ, 192
95, 164–166, 331, 364 Nowolipki Street. See under Warsaw
theory, vi, xi, 8, 12, 45, 48, 51–54, numbers
75–76, 164–166, 420 irrational, vi, 49, 217, 222–223, 274,
zero (See under set) 292–300
Index of Subjects 493

approximation of, 294–297 Peano– Jordan content, 166, 364

arithmetic with, 297–300 school of, vi, 51, 323
equality of, 297 See also Lingua Peano and
natural, 16 in the subject index
undefinability of, in real arithmetic, 163 pensions, Polish, 339, 349–356
ordinal, 16–17, 54, 161, 165 perimeter. See under polygon
rational, 273, 294 Periodico di matematica, 203
real, 46, 183, 222–223, 238, 274, 418 (see permissions, 473–475
also under completeness; decidability) perpendicular
number theory, 8, 49 lines, 216
to two lines, 271
Ogniwo, 177, 203 See also under segment
Olympiad, 208, 210, 219 phenomenon, 343–345
ontology, 361 Philosophical
See also logic: Leniewski’s Institute (see under Warsaw)
open sentence, 195 Society, Polish, in Lwów, 51, 322,
See also function: sentential 324, 326, 356, 396
order. See betweenness and under philosophy
axiom; language: formal analytic, 9, 18, 34
ordered pair, 72 Berlin school of, 334
Order of Polish Rebirth, 221 Cracow Circle, 34
ordinal. See under algebra; number deductive, 323
orientation, 80 empiricist, 334
history of, 369, 396
S. See pi neo-Kantian, 365
pansomatism, 407 of mathematics, 181, 188
paradox, 164–166 Polish school of (see under logic)
Banach–Tarski, 53, 57, 62, 90, 93, 164, 378 of science, 8, 181, 329, 343–344,
Hausdorff, 65, 67–69, 73, 109, 164–165 370–373, 387, 412
Mazurkiewicz–Sierpiski, 40, 48, 51 See also under language
parallel physics, foundations of, 368
lines, 265, 275 pi ( S)
projection, 219 approximation, 212, 236
See also under axiom; strip definition, 212, 236, 240
parallelogram, 127–128, 156 Pieri, Mario. See under axiom system
altitude, 127–128, 308, 314 and in the index of persons
area, 45–47, 308–309 Pisudski, M/S, 332–333
base, 308 plan. See action, theory of
height, 314 plane, 186, 216
Parametr and Mody matematyk, 126, 130, Playfair, John. See under axiom
135, 169, 204–213, 408 Pock, iv, xxii–xxiii
problem sections, 205, 243–246, 250 point, as primitive notion, 183, 186, 188, 194
difficulty (see under Tarski: as educator: point-set theory. See topology
exercises) Poland, xxiii
prizes, 245, 250 unification, vii, 4–5, 8, 10, 49
parenthesis-free notation, 340–341 Military Organization—East, 175
Paris, xxii–xxiii, (see also Polish: Legion)
See also under congresses; universities economy, 35, 172, 179, 189, 273,
Pasch, Moritz. See under axiom 349–356, 395
Peano, Giuseppe, 59 history and culture, xii, 3, 174, 223
494 Index of Subjects

school system, 171–172 postulate. See axiom

(see also under curriculum) postulate theory, 15–17, 29–30, 51, 162,
during World War II 186, 188, 331, 365–368, 418
Army in the East, 224 Pozna, xxii–xxiii, 200–202, 227, 328
clandestine education, xiii, 9, 37, 49, 72, See also under universities
132, 197, 199, 208, 210, 221, 345, 373 Prace matematyczno-fizyczny, 189
government-in-exile, xiii, 224 Prague, xxii–xxiii, 177.
underground government, xiii, 32, 221 See also under congresses
People’s Party, 221 precedence. See under relations
after World War II, 72, 177, 379 prime filter, 165
See also under World War II primitive notions, 15–16, 80, 183, 186,
Polish 188, 194–195, 346–349
Academy of Arts and Sciences, 406 independence of, 324
Academy of Sciences, 407–409 Princeton. See under congresses; universities
Association of Forced Emigration, 224 Principia Mathematica, 346, 359
emigré community (Polonia), 397, 404, 409 prizes for solutions. See under Parametr
Exposition, General, in Pozna, 201–202 probability
Legion, 14, 329, 369 conditional, 69, 387
spelling, xii theory, 14, 66, 197, 325, 334–335, 386–387
See also parenthesis-free notation; problem, use of the word, 243
Philosophical: Society; Poland and under product, Cartesian. See under symbol
congresses; Mathematical Society projection. See under parallel
Polish–Soviet War, x, 8–11, 14, 32, 34, 49, propaganda in 1920, 8, 10–11
174–177, 210, 221, 401, 405 proper inclusion. See also set: proper part
Vistula, Miracle of the, 8–10, 49 and under axiom system; symbol
politics, 7–9, 32, 173, 177, 200, 221, 226, propositional logic. See deduction: theory of
328–331, 334–335, 343, 350, 393, 397 propositional variable, 39, 160
Polonia. See Polish: emigré community protothetic. See logic: Leniewski’s
polygon Przegld filozoficzny, 12, 19–20, 189, 321
circular, 157 Przegld matematyczno-fiziczny, 58–59,
diameter, 126, 138 189, 204, 210
perimeter, 163, 213, 236 Przegld pedagogiczny, 189, 203, 207, 215
regular, 212–213, 217, 236–237 psychology, foundations of 334, 372
area, 310 Puck, 192–193
similar, 275 Pythagoras. See under theorem
simple, 240
spherical, 112 quadratic
See also polygonal regions formula, 195
polygonal regions, 46, 125 functions, 247, 249, 259–260, 313
area, 45–47, 53, 59–60, 77–91, 106–109, quantifier, 195
162–163, 219–220, 304–318 elimination, 195–196, 247, 250, 257,
in the broader sense, 137 259–260, 261–262, 412, 414, 418
connected, 127, 133, 139 existential, 195
convex, 127, 133, 138–139, 240 logic, 340
difference and sum, 302 universal, 39, 160, 195
See also finite decomposition: See also under symbol
into polygons; polygon quotation marks, 40, 356
polyhedral region, volume of, 45, 47,
61, 90–91 radio. See broadcasting
positivism. See under logical radius. See under circle
Index of Subjects 495

rank. See under set semantics, 160, 323, 326, 356–363, 374–
ratio. See under segments 376, 393, 396, 407, 417, 419
rational. See under numbers in computer science, 167, 363, 414
real. See under numbers See also under category
rectangle sentence, 195, 357–358
altitude, 304 conditional, 368
area, 304–305 open, 195 (see also function: sentential)
base, 304 sentential logic. See deduction: theory of
in a square, exercise on, 261–263 sequence, 231
recursive. See under definitions exercise on a, 255
reduction to absurdity, 283 mutually convergent, 231
refugees. See World War II: displaced set
Regina, 125 absolutely measurable, 166, 364
region. See polygonal regions; polyhedral boundary, 105, 112, 119
region; triangular region bounded, 84
regularity. See under theorem empty, 81
relations first-category, 70, 105
D , E-, 66, 70, 75, 103 measurable, 52–53, 76, 106
precedence, 21 of measure zero, 70, 75
theory of, 377, 379, 391, 406, 414, 420 nonmeasurable, 52, 74–75, 123
relativity. See under truth nowhere dense, 70, 105
Reno. See universities: of Nevada ordered, 16, 21
rhombus, area of 313 of Peano– Jordan content zero, 70, 105
Righteous among the Nations, 57, 345 proper part of, 80, 358
Riveon lematematika, 126, 130 rank of, 381
Rockefeller grant. See under Tarski: as square (see under construction problems)
researcher well-ordered, 12, 15–21, 58, 161, 165, 331
rounding. See under arithmetic See also sets; set theory
Ruch filozoficzny, 18, 321 sets
Ruch pedagogiczny, 203 difference of, 80
Russia disjoint, 80
Empire, xxii, 3–4, 210 identical, 80
Revolution of 1905, 49 sum of, 80
Revolution of 1918, 8, 10 union of, 80
See also Soviet Union See also under symbol
Rysia Street. See under Warsaw set theory, vi–vii, 5–8, 12, 33, 40, 49, 60, 64,
69, 72, 79, 328, 331, 333, 335, 379, 412,
San Francisco. See under universities 417, 420
satisfaction, 359–360, 362 geometric application of, 33, 39–41, 43, 45,
scalar, 186 53, 58–59, 162, 186
secant line. See under circle without variables, 390
segment Zermelo’s, 17
length of, 217 See also algebra: of classes and
perpendicular bisector of, 275 under axiom system
See also under arithmetic side of a line, 264
segments, 216, 271–272 Sierpiski, Wacaw. See under paradox
commensurable, 217, 291 and in the index of persons
exercise on two, 271–272 similarity, 219–220, 301, 316–318
incommensurable, 217, 274, 292–297 singleton. See under symbol
ratio of, 296 Skolem, Thoralf. See under theorem
496 Index of Subjects

Slovakia, 398 equality, 21, 80 . . . . . . . . . . . . . . . . . . x = y

snow job, 399 equivalence by denumerable
A= B
Social Insurance Institution, 179, 343, decomposition, 66, 96 . . . . . . . . . . . d

351, 395 equivalence by finite

Society of Sciences and Letters. See under decomposition, 60 . . . . . . . . . . . . . . A { B
Warsaw 66 . . . . . . . . . . . . . . A = B
f
Society of Scientific Courses, 210 equivalence connective, 39 . . . . . . . ) @ <
solid body. See under geometry existential quantifier, 195 . . . . . . . . } x )( x)
Sonoma. See under universities floor, 253 . . . . . . . . . . . . . . . . . . . . . . E x, O xP
Soviet Union, xxiii, 10, 226 image of a set by a function, 99 . . . . . . M( A)
Academy of Sciences, 200 implication, connective, 30 . . . . . . . . ) H <
sphere, 186–188, 195, 208 347 . . . . . . . C ) <
area and volume, 237–238 inequality, 21 . . . . . . . . . . . . . . . . . . . x = / y
See also under strip inequivalence of sets, 81 . . . . . . . . . . A {  B
Springer, xiv, 474 integral, 54 . . . . . . . . . . . . . . . . . . . . . . . . . ³
square, set (see under construction problems) interior of a set, 93 . . . . . . . . . . . . . . . . . . A(
Stanford. See under congresses; universities intersection of sets, 93 . . . . . . . . . . . . . A  B
St. Petersburg, xxii–xxiii limit, 229 . . . . . . . . . . . . . . . . . . . . . . . . lim
straight. See under angle minimum number of strips of width x,
strip 129, 145 . . . . . . . . . . Vw( x)
parallel, 129, 145, 152 129 . . . . . . . . . . . . . . V W (x)
spherical, 146 147 . . . . . . . . . Vk( x), Vk w( x)
area of, 147 negation
subset. See under symbol of a formula, 39 . . . . . . . . . . . . . . . . . . . ¬ f
Sukowski Street. See under Warsaw 347 . . . . . . . . . . . . . . . . Nf
sum. See under polygonal regions; sets 358 . . . . . . . . . . . . . . . . . f
summer courses. See under of a relation, 21 . . . . . . . . . . . . . . . . . . . RU
mathematics: education pi, 212, 236 . . . . . . . . . . . . . . . . . . . . . . . . . S
superset. See under symbol product, Cartesian, 40 . . . . . . . . . . . . A × B
surfaces, algebraic 181, 189 proper part of, 80 . . . . . . . . . . . . . . . A ‚ B
Suslin’s problem, 382 358 . . . . . . . . . . . . . . A G B
See also in the index of persons singleton, 93 . . . . . . . . . . . . . . . . . . . . . . { x}
symbol subset, 93 . . . . . . . . . . . . . . . . . . . . . . A I B
almost-equivalence by denumerable A = B superset, 93 . . . . . . . . . . . . . . . . . . . . A J B
decomposition, 75, 117 . . . . . . . . . . p union (sum) of sets, 80, 93 . A F B, also U
betweenness, 267 . . . . . . . . . . . . . . . . EOPQ unique existence, 16 . . . . . . . . . (}! x) )( x)
ceiling, 127 . . . . . . . . . . . . . . . . . . . . . . . . M xN universal quantifier, 195 . . . . . . . (~ x) )( x)
congruence of sets, 66 . . . . . . . . . . . . A ~ =B 358 . . . . . . . . 3 x )( x)
conjunction, 39 . . . . . . . . . . . . . . . . . . ) & < width of a set, 133, 152 . . . . . . . . . . . . Z( A)
degree of equivalence, 126, 137 . . . . V( A, B) See also under translation
Tarski’s functions, 127 . . . . . . . . . . W( x) symmetry, 247, 250, 253, 256–257,
163 . . . . . . . . . Wm( x) 262, 271–272
156 . . . . . . . . . 7( M)
degree of irregularity, 157 . . . . . . . . . . U(W ) T, condition, 362
diameter of a set, 126, 138 . . . . . . . . . . G( A) W, 7, functions. See tau
difference of sets, 80 . . . . . . . . . . . . . A – B tangent. See under circle
disjointness, 80 . . . . . . . . . . . . . . . . . . A ][ B Tarska, Maria
disjunction, 347 . . . . . . . . . . . . . . . . . A ) < childhood and family, 174
358 . . . . . . . . . . . . . . . . . ) Z < in the Polish– Soviet War, 175, 405
empty set, 93 . . . . . . . . . . . . . . . . . . . . . . . I marriage, 174
Index of Subjects 497

as educator, 174–176, 326 Rockefeller grant, 177, 325, 394

pregnancies, 177, 331 logic course, introductory, 191–192,
life and character, 174–176, 179, 320– 326, 331, 404
321, 325, 328, 330, 396 logic text, 192, 217, 223, 283,
portraits, 175, 329, 400 391–392, 407
reunited with Alfred, 336, 376 publication delays, 196, 267, 377, 380, 393
Tarski, Alfred philosophical position, 389, 402,
childhood and youth, 3–5, 392, 400 407–408, 413, 416–419
languages, 3–4, 321, 331, 369, 396, 406 pragmatism, 407
biology, interest in, xi, 4–5, 331, 392 salary, 179
university student search for professorship, 326, 328, 333
first years, x, 6–8, 12, 19, 40 workload, 176, 180, 190–194, 319, 326, 331
enrollment record, 6 political activity (see under politics)
military service, 5, 8 recreation, 176, 192–193, 201, 227, 320, 398
first published paper, 12, 15, 19–30 emigration, 165, 167, 330–334,
first scientific presentation, 36, 160, 322 389, 396, 408
doctoral study, ix, 31, 33–36, 40, family
48, 58, 77, 160–161 parents’, 3, 36, 336, 392, 404
dissertation, 38, 160, 386 wedding, 174
physics, interest in, 393 Tarski’s, 174, 176–179, 336,
nationalism, 4, 8, 38 376, 395–396, 403
religion, 38, 174, 179, 321, 405 apartments, 176–179, 203, 395
name changes, 8, 12, 15, 19–20, 38– reunited, 336, 376
39, 174, 328, 392 in Berkeley, 336, 376–377, 389,
signatures, i, 38, 41, 392–393 396, 406–407
as educator, v, vii, x–xi, 43, 45, 59, 125, personality, 320, 399, 401,
225–227, 229–241, 350, 401, 405 403, 407, 409, 411
first teaching jobs, 36, 38, 41 portraits, i, 13, 329, 384, 400, 407–409
1929 congress report, 225–227 writing style, 17, 126, 222–223, 246, 380
at the Third Boys’ (= eromski) Gim- obituaries
nazjum 127, 173, 179–180, 320, 405 San Francisco Chronicle, 405–406
training teachers, 190–191 San Francisco Examiner, 407.
exercises, published biography, ix–x, 3, 399–409
xi, 243–272 bibliography, ix–x, 385–398, 390–391, 421,
difficulty of, 246–248 other collections, 369, 385, 390–391, 415,
geometry text (see Geometrja) 436, 464–466
as researcher translations into Polish, 385, 391
threads and filaments, 159 archives, ix, 201, 333, 392–397,
habilitation, 41, 161, 190 409, 473–474
logic and set theory, 160–162, 339 See also under axiom; axiom system;
Leniewski, intellectual debt to, 40 conjunction; function; Leniewski;
applying set theory to geometry, 43, Geometrja; negation; paradox;
45, 58, 64–66 symbols; theorem
circle-squaring problem, 62–63, 90, 164 tau ( W, Wn , 7), 127–129, 132–134, 142–145,
with ukasiewicz 148–152, 154–158, 163, 211, 263
as assistant in seminar, 160, 190–192, See also under symbol: degree of
194–198, 247, 267, 322–323, 326, equivalence
328, 331, 335, 356, 419 teachers’ journals, Polish, 203–205
as collaborator, 160, 346 See also Delta; Matematyka i szkoa;
as adjunkt, 191–192, 198, 328, 395 Matematyka; Mianowski Fund
498 Index of Subjects

teachers’ organizations, Polish, 203, area, 309

207, 219, 221 base, 309
Trade Union of Polish Secondary School circumcenter, 275
Teachers, 173, 177, 225 defect, 265
Thales. See under theorem incircle, 275
theology, foundations of, 368 inequality, 250
theorems, 15–16, 357, 361 median, 279
antireflexivity, 22–23 triangular region, 46
Banach’s fixed-point, 57 triangulation, 46
Banach’s, on measure, 53–55, 60, 71, trichotomy. See under axiom
83–84, 91, 106, 363–364 trigonometry, 213, 230, 232–241
Banach–Tarski, 64, 90, 93 tables, 189, 212, 236
Bolyai–Gerwien, 47, 125 truth
Cantor–Bernstein, 66, 70, 74–75, 167, 381 Aristotelian concept, 362
deduction (see deduction, theory of) bearer, 362
of Desargues, 186 classical concept, 362
Gerwien’s, about spherical polygons, 74 function, 39
Gödel’s (see incompleteness) logical, 344, 371–372, 387, 396
Hahn–Banach, 54–55 relativity of, 374–376
Kuratowski’s, 70–71, 103, 167 semantic concept of, 362
of ukasiewicz, 30 theory of, v, 40, 160, 322–323, 333, 356–
of Pythagoras, 46, 128, 217, 222 363, 374–376, 387, 399, 404, 407, 412, 419
regularity, 76, 119 value, 39
Skolem–Löwenheim, 419 tuberculosis, 5
upward 196 tuition. See under universities: Warsaw
Tarski’s fixpoint, 167 type theory, 329, 342, 358, 363
of Thales, 219, 301 typhus, 5
triangle similarity, 301 vaccine production, 57
theory, 195 Twardowski, Pan, 201
categorical, 324, 388
decidable (see decidability) ultrafilter. See filter, prime
undecidable, 412, 420 undecidability. See under theory
thread. See under Tarski, Alfred. undefined notion. See primitive notions
time underground. See under Poland: during
continuity of, 344 World War II
theory of, 344, 373 union. See under sets; symbol
topology, 5, 14, 49, 60, 69, 72, 80, unique existence, 291
195, 335, 379 See also under symbol
torus, 54 unit. See under area; length
transitivity. See under axiom Unity of Science. See under congresses
translation, xii, xiv universal. See under algebra; symbol
conventions, xii, 273, 356, 388 universities and similar institutions
of symbols, xii Berkeley, ix, xiii–xiv, 198–199, 336,
trapezoid 376, 400, 404, 407
altitude, 310 Berlin 387, 393
area, 310 Bonn, 69
base, 310 Chicago 125–126, 130, 212, 334, 344, 426
right, 289 Copenhagen, 325–326, 387
triangle Cracow 343, 368–369, 373
altitude, 309 Göttingen, vi, 14, 36, 53, 189, 329, 334
Index of Subjects 499

Harvard, 324, 333, 388, 396 Völkischer Beobachter, 330

Kiev, 208 volume. See under cone; cylinder;
Leipzig, 69 polyhedral region
Lwów, vii, 4, 9, 14, 18, 36, 38, 49, 53, 57,
177, 197, 327–329, 333, 344–345, Warsaw, ix, xxii–xxiii, 3, 8–10, 14, 49, 72, 130,
361, 368, 373 132, 171–179, 189, 196–201, 208–210,
Lwów Polytechnic 51, 53, 57, 72, 373 214, 218–221, 273, 321–322, 330–331,
Münster, 197, 331 335–336, 343, 392, 404–405, 415
of Nevada at Reno, 382 history and culture, 171–172, 330, 401, 404
New York, City College of, 334, Koszykowa Street, 3, 35, 178, 394, 404
379, 396, 406 Marszakowska Street, 173
Paris, 14, 329, 373, 378 Nowolipki Street, 173
Pozna, 201, 328, 345, 373 Philosophical Institute, 9, 12, 36, 38,
Princeton Institute for Advanced Study, 321–322, 340, 346, 386
333–334, 379 Rysia Street, 173
Rome, 181 Smolna Street, 219
of San Francisco, 408 Society of Sciences and Letters, 164,
San Francisco State, ix, xiv 321–322, 356, 363, 415
Sonoma State, xiii Sukowski Street, 176, 178–179
Stanford, 400, 407 oliborz district, 176–177, 179
Turin, vi, 59, 369 See also under congresses; Mathematical
Vienna, 199, 322, 324 Society; universities
Mathematisches Kolloquium, 324 weakest, relatively. See under axiom system
Warsaw, vii, 3–9, 12, 14, 21, 31–38, 49, 72, Wektor, 189
132, 176, 189, 190–191, 194–199, 208, well-ordering. See under axiom; axiom
210, 214, 221, 267, 321, 326, 328, 335, system; set
345, 350, 379, 392–396, 404 wff. See under formula
academic year, 7, 190 Wff ’N Proof. See logic: game
tuition, 7–8, 35 Whiskeytown. See under congresses
Warsaw Free, 41, 210 white space, xii
Warsaw Polytechnic, 200–201, 210 width, 133, 152. See also under symbol
Yale, 341 Wilno. See Vilnius
Zurich, 14, 189, 199, 210 wine, exercise on diluting, 249–250
Urelemente, 199 Wojciech’s Bookstore, St., 204, 207
World War I, ix, xii, xxii, 4–5, 10, 14, 32, 37,
Veblen, Oswald. See under axiom system 49, 72, 132, 171–175, 329, 334, 369
venia legendi. See habilitation World War II, ix, xiii, 14, 32, 34, 37, 49, 53,
Versailles, treaty of, 10 57, 69, 72, 132, 161, 334–336, 343–345,
vertex. See under angle 373–376, 404
Vienna, xxii–xxiii, 18, 177, 191, 200, displaced-person camps, 197, 224
322–325, 345, 356, 393 See also under Poland
Circle, 325, 327, 331, 341, 369–372, Wrocaw. See Breslau.
374, 393, 396, 400, 419
See also under universities Zakopane, xxii–xxiii, 176, 329
Vierergruppe. See under group Zermelo, Ernst. See under set theory and in
Vilnius, xxii–xxiii, 10, 174 the index of persons
See also under congresses zloty, 35
Vistula, Miracle of the. See Polish– oliborz district. See under Warsaw
Soviet War Zurich. See under universities