Lloyd A. Newton Medieval Commentaries On Aristotles Categories Brills Companions To The Christian Tradition 2008

Medieval Commentaries on
Aristotles Categories
Brills Companions
to the
Christian Tradition
A series of handbooks and reference works
on the intellectual and religious life of Europe,
5001700
VOLUME 10
Medieval Commentaries
on Aristotles Categories
Edited by
Lloyd A. Newton
LEIDEN BOSTON
2008
ISSN 1871-6377
ISBN 978 90 04 16752 0
Copyright 2008 by Koninklijke Brill NV, Leiden, The Netherlands.
Koninklijke Brill NV incorporates the imprints Brill, Hotei Publishers,
IDC Publishers, Martinus Nijhoff Publishers and VSP.
All rights reserved. No part of this publication may be reproduced, translated,
stored in a retrieval system, or transmitted in any form or by any means, electronic,
mechanical, photocopying, recording or otherwise, without prior written permission
from the publisher.
Authorization to photocopy items for internal or personal use is granted
by Koninklijke Brill NV provided that the appropriate fees are paid directly to
The Copyright Clearance Center, 222 Rosewood Drive, Suite 910,
Danvers, MA 01923, USA.
Fees are subject to change.
printed in the netherlands
Cover illustration: Avicenna, Opera, Venice, 1508, folio 2 verso. Printed edition. Photograph
by Megan Bickford. With kind permission of Ed Macierowski.
This book is printed on acid-free paper.
Library of Congress Cataloging-in-Publication Data
Medieval commentaries on Aristotles Categories / edited by Lloyd A. Newton.
p. cm. (Brills companions to the Christian tradition, 18716377 ; v. 10)
Includes bibliographical references and index.
ISBN 978-90-04-16752-0 (hardback : alk. paper) 1. Aristotle. Categoriae.
2. Categories (Philosophy).
I. Newton, Lloyd A. II. Title. III. Series.
B438.M43 2008
160dc22
2008014809
CONTENTS
Preface ......................................................................................... vii
The Importance of Medieval Commentaries on Aristotles
Categories ................................................................................... 1
Lloyd A. Newton
The Medieval Posterity of Simplicius Commentary on the
Categories: Thomas Aquinas and al-Frb ............................. 9
Michael Chase
Avicenna The Commentator ...................................................... 31
Allan Bck
Albertus Magnus On the Subject of Aristotles Categories ......... 73
Bruno Tremblay
Interconnected Literal Commentaries on the Categories in the
Middle Ages ............................................................................ 99
Robert Andrews
Thomas Aquinas on Establishing the Identity of Aristotles
Categories ............................................................................... 119
Paul Symington
Reading Aristotles Categories as an Introduction to Logic:
Later Medieval Discussions about Its Place in the
Aristotelian Corpus ................................................................... 145
Giorgio Pini
Simon of Faversham on Aristotles Categories and The Scientia
Praedicamentorum ....................................................................... 183
Martin Pickav
Duns Scotuss Account of a Propter Quid Science of the
Categories ............................................................................... 221
Lloyd A. Newton
vi contents
Fine-tuning Pini s Reading of Scotus s Categories
Commentary ........................................................................... 259
Todd Bates
How Is Scotuss Logic Related to His Metaphysics?
A Reply to Todd Bates ........................................................... 277
Giorgio Pini
John Buridan : On Aristotles Categories ....................................... 295
Alexander W. Hall
A Realist Interpretation of the Categories in the Fourteenth
Century: The Litteralis sententia super Praedicamenta Aristotelis
of Robert Alyngton ................................................................ 317
Alessandro D. Conti
Thomas Maulevelts Denial of Substance ................................. 347
Thomas Maulevelt: Quaestiones super Praedicamenta:
Quaestio 16 ............................................................................... 358
Robert Andrews
Categories and Universals in the Later Middle Ages ................ 369
Alessandro D. Conti
Bibliography ................................................................................ 411
List of Contributors .................................................................... 429
Index ........................................................................................... 433
PREFACE
St. Augustine s account of the ease with which he understood Aristotles
Categories is, as he himself admits, an anomaly. Unfortunately, the rest
of us, like Augustines contemporaries, struggle to understand what
was for centuries, and still is, a fundamental text. On the surface,
Aristotles Categories is a markedly anti-platonic text: things are equivo-
cal, not univocal; individual substances are primary, whereas universals
are secondary; and Aristotle lists ten highest genera or categories of
things, not ve, as Plato does.
1
From its earliest reception, though, many
commentators such as Porphyry and Boethius go to great lengths to
reconcile it with Platonism , with the predictable result that many other
commentatorsOckham comes immediately to mindgo to equally
great lengths to purge it of any remaining traces of Platonism. Such
attempts often hinge on what one takes to be the subject of the book: is
the Categories about words, concepts, or things? Or is it somehow about
all three: words, concepts and things? Regardless of how one answers
this question, the philosophically more important question remains:
to what extent do words, concepts and things parallel or mirror one
another?
Of course, these are not the only questions pertaining to Aristotles
brief text. Given its wide range of topics, from the nature of equivocity
to the different kinds of motion, coupled with its terse and introduc-
tory remarks on a number of issues, Aristotles Categories generated a
disproportionate number of commentaries since its rst appearance in
antiquity. These commentaries, though, are often far from being simple,
literal expositions of the text. More often than not, they are occasions to
1
As far as possible, I have endeavored to distinquish between Categories, which
refers to the name of the work by that title, and the ten categories, which, when not
italicized, refer to the subject matter of that work, namely, to the ten principal genera
of being discussed by Aristotle. Unfortunately, though, the distinction is not always
clear, especially since Aristotles Categories is essentially about the ten categories. This
confusion is particularly evident in subsequent discussions about the subject of the
Categories, which discussions I presume to be about the subject of the book, but which
are also closely related to discussions about the way in which the ten categories are
the subject of a science (which discussions are not about the subject of the book but
about the ten genera).
viii preface
explore philosophical problems, or opportunities to attack a particular
philosophers thesis, or the chance to defend ones own thesis. Bearing in
mind, then, that commentary writing was traditionally a way of doing
philosophy, it is not surprising that hundreds of extant commentaries on
Aristotles Categories exist, and that these commentaries often contradict
one another, explore different topics of concern, and are philosophically
rich, as the ensuing articles amply demonstrate.
While the vast majority of the extant commentaries on Categories
are still not translated and largely inaccessible, things are beginning
to change. The Ancient Commentaries on Aristotle series, published by
Cornell Press, has recently made a number of early commentaries
on the Categories available in English. Likewise, several medieval com-
mentaries have recently been published and others are, I hope, soon to
follow. Correspondingly, there has recently been an increasing interest in
medieval logic in general, and in categories in particular, with a number
of conferences, articles and books devoted to the subject.
When Julian Deahl rst suggested the subject of this book in 2002 to
Jorge J. E. Gracia at a medieval conference, Gracia remarked that, at the
time, he did not think that more than half a dozen people in the eld
were qualied or interested in the subject. Fortunately, Gracia recom-
mended that I be one of the contributors should the project ever get
off the ground. While doing my own research on Scotus s commentary
on Aristotles Categories, I realized that more scholars were interested
in the subject and approached Julian with the current project. To my
surprise, Julian liked the project and asked if I would be interested in
editing it. Consequently, I wish to thank Jorge J. E. Gracia for introduc-
ing and recommending me to Julian, to Julian Deahl for having the
condence in me to edit this work; and to Marcella Mulder and Gera
van Bedaf of Brill, without whose assistance, this book would not be
possible. Most of all, though, I am grateful to my wife, Lori, and to
our three children for their support.
LLOYD A. NEWTON
INTRODUCTION
THE IMPORTANCE OF MEDIEVAL COMMENTARIES
ON ARISTOTLES CATEGORIES
Lloyd A. Newton
Aristotles Categories is the subject of an extensive number of commen-
taries and of an unusual amount of debate, and for good reasons.
1
To
begin with, in spite of its relatively short length, it can be a rather dif-
cult text to understand, even for the trained philosopher, to say nothing
of those who are just beginning their study of philosophy. Yet, because
it laid the foundation for many subsequent philosophical discussions in
general, and for logic in particular, it was, during much of the Middle
Ages, often the very rst philosophical text students encountered. Even
contemporary philosophers who are steeped in philosophy and who have
studied the Categories in depth often nd it difcult, albeit for different
reasons. One difculty, as the ancient commentators on the Categories
recognized, is that Aristotle himself is ambiguous about the subject of
the work. What exactly is he categorizing? Is it things that are or things
that are said or something in between, such as a concept? Furthermore,
depending on how one understands its purpose, the Categories can be
seen in harmony with, in contrast to, or even in contradiction to, Platos
own theory of the ve highest genera. For all of these reasons the
Categories has historically acted like a magnet, attracting commentaries
from Aristotelians, Platonists, and Stoics alike. Quite naturally, some
of these commentaries defend Aristotelianism, whereas others defend
either Platonism or Stoicism by attacking Aristotles Categories. Finally,
still others, especially during the Late Middle Ages, use the Categories
as a means to expound their own philosophical systems in the process
of interpreting Aristotle.
1
According to my count of the texts listed by Charles Lohr, roughly two hundred
extant Latin commentaries on the Categories were written during the Middle Ages. Of
course, this number does not take into account the commentaries that are not extant,
nor the ones written in Greek, Arabic, or Hebrew. Cf. the lists of extant commentaries
cited by Charles Lohr in Traditio, vols. 2329.
2 lloyd a. newton
Though many of the ancient and medieval commentators, such as
Porphyry, Boethius and Albert the Great, did write original treatises
on philosophical issues, their commentaries are in themselves valuable
contributions to philosophy, particularly those from the later Middle
Ages.
2
Consequently, studies of the various commentaries, and especially
those dealing with the Categories, are valuable projects, as the following
essays amply demonstrate. As Robert Andrews points out, medieval
Categories commentaries are the repository of centuries of analyses
of the basic concepts of Western thought, all carefully organized and
awaiting modern rediscovery.
3
And while most of those commentaries
are still awaiting rediscovery, the following essays, I hope, will convince
everyone that the effort is worthwhile.
Originally, I planned to include essays on all three main philosophi-
cal traditions alive throughout the Middle Ages, namely, those written
by Jewish, Christian and Islamic philosophers. Essays pertaining to
the Jewish tradition, however, are noticeably absent due to the lack of
contemporary scholarship in this area. Consequently, the preponder-
ance of the remaining articles focuses mainly on Christian philosophers.
The scope of the project has, however, stayed away from theological
issues, even though discussions of the categories often have tremen-
dous theological implications, especially concerning the doctrines of
Transubstantiation and the Trinity. Consequently, the issues raised in
the following essays are properly philosophical issues, not theological.
What follows is a collection of fourteen original essays,
4
all devoted
to one or more medieval commentaries on Aristotles Categories, writ-
ten by a wide variety of philosophers from Europe, Canada, and the
United States. I will summarize each of them briey.
Michael Chase begins the volume by demonstrating the importance
of Simplicius commentary for two key medieval thinkers, Aquinas
and al Frb. Due in part to Simplicius inuence, and particularly
2
Compare Fr. Wippels description of St. Thomas commentaries: of his theological
commentaries, two are commentaries in the strict sense, i.e., on the De Hebdomadibus
of Boethius and on the De divinis nominibus; the other two offer brief expositions of the
texts of Boethius and of Peter and use them as occasions for much fuller and highly
personal disquisitions by Thomas himself. John F. Wippel, The Metaphysical Thought of
Thomas Aquinas: From Finite Being to Uncreated Being (Washington, D.C.: Catholic University
Press, 2000), p. xviii.
3
Robert Andrews, Question Commentaries on the Categories in the Thirteenth
Century, Medioevo 26 (2001), 265326, p. 266.
4
Not counting the introductory essay or the original work by Thomas Maulevelt.
the importance of medieval commentaries 3
his commentary on the Categories, both gures adopt the Neoplatonic
project of reconciling Plato and Aristotle, in spite of the apparent dif-
ferences between them. Interestingly, though, while both al-Frb and
Aquinas ultimately agree on the harmony between Plato and Aristotle,
they differ in that Aquinas follows Iamblichus, who makes philosophy
subordinate to theology, while al-Frb follows Porphyry, who views
philosophy as alone sufcient for beatitude.
In the second article, Allan Bck argues that it is Avicenna, not Averroes,
who rightly deserves the title The Commentator. In doing so, Bck
rst turns to Aristotle himself to see what the true function of a com-
mentator is. A good commentator, according to The Philosopher, is not
simply one who explicates the views of others, but one who reports and
critiques certain endoxa, namely, the opinions of the experts. Bearing
in mind, then, the peculiar purpose, context, and style of Avicennas
commentaries, it turns out that Avicenna is a better commentator than
Averroes. At best, Averroes is only the commentator for the beginning
student, who seeks to nd out only what Aristotle actually said.
Whether Avicenna really deserves to be called The Commentator or
not, his views on logic denitely inuenced Albert the Great, as Bruno
Tremblay shows in the next article. As is well known, Albert was one
of the rst medieval philosophers in the West to produce a set of com-
mentaries on the entire Organon, and his assimilation of Avicennian
doctrines (especially the distinction between universals ante rem, in re, and
post rem) played an extremely important role in Alberts understanding
of logic. Consequently, in contrast to other contemporary philosophers,
who interpret Albert in such a way that the subject of logic is either
words or things, Tremblay contends that the subject of the categories
for Albert is rst and foremost concepts, and that Albert, therefore, is
responsible for shifting the study of logic away from a more linguistic
approach long favored by Neoplatonic philosophers such as Boethius.
If Avicennas commentaries are on one end of the spectrum of com-
mentary writing, with Alberts commentaries somewhere in the middle,
then the other end of the spectrum is the literal commentary, a genre
hardly appreciated by most contemporary philosophers. Nevertheless,
in sharp contrast to Bcks spirited defense of Avicenna as The
Commentator, Robert Andrews shows the importance of those often
neglected studies at the other end of the spectrum, namely, the literal
commentaries. In doing so, he focuses on three particular issues (the
4 lloyd a. newton
nature of truth, denominative predication, and having) as those
issues are treated by four rather obscure philosophers: Peter of Saint-
Amour, Gerard of Nogent, Siger of Courtrai , and Thomas of Erfurt.
He also lightens things up with what was probably the only joke in all
the medieval commentaries on the Categories.
In the next article, Giorgio Pini begins by observing that the Categories
owes its pride of place within the Organon and within the larger corpus
of Aristotles writings not to Aristotle himself, but to early Neoplatonists
who wanted to structure Aristotles writings into a system comparable
to that of the Stoics. Considering the Categories as the beginning work
of logic, however, posed a number of philosophical problems, not the
least of which was how to reconcile what Aristotle says in this work with
conicting statements that he makes about categories in the Physics and
Metaphysics. More importantly, Pini demonstrates that as the conception
of logic evolved from a consideration of arguments to a consideration
of second intentions, so too changed the conception of the Categories.
Thus, depending on the philosopherand Pini looks at eight differ-
ent philosophers: Thomas Aquinas, Giles of Rome, Radulphus Brito,
Duns Scotus, Peter Olivi, William Ockham, Walter Burley and Robert
Alyngtonthe Categories can be read in any number of ways: as an
introduction to syllogistic reasoning; as a treatise dealing with second
intentions; as a sort of therapy preliminary to philosophy to protect
the student from any lingering traces of Platonism; or as an ontological
treatise establishing the existence of universals.
Although Aquinas did not write a full length commentary on the
Categories, he did comment on the categories in his commentaries on
the Physics and Metaphysics. Furthermore, given the overall importance
and inuence of the Angelic Doctor as well as the scant treatment of
his comments on the categories by most Thomistic studies, the editor
thought it worthwhile to include an article exclusively on Aquinas.
More important, though, in this article Paul Symington challenges the
traditional interpretation of Aquinass treatment presented by John
Wippel and others. Symington notes a number of difculties with
the typical accounts and offers his own interpretation that connects
Aquinas derivation of the ten categories to the three main modes of
per se predication.
While the derivation of the categories is one important issue, another
important topic concerns the scientic status of categories, namely,
whether a single subject, analogous to being qua being for metaphys-
ics, enables the categories to be treated as a science. If so, what kind
of science is it and how does this science of the categories differ from
metaphysics? Consequently, in the following two articles, two authors
look at the scientic status of the categories. In the rst one, Martin
Pickav focuses primarily on Simon of Favershams commentary on the
Categories and shows how Simon defends a certain type of knowledge or
science of the categories. To understand and appreciate Simon better,
Pickav contrasts Simons approach with an anonymous commentary
(the Anonymous of Madrid) and with Henry of Ghent s position on
the compound nature of the categories. As a result, we get a clearer,
though also more complicated, picture of the relationship between
logic and metaphysics as they are conceived by Simon of Faversham
and Henry of Ghent.
In the second article, Lloyd Newton traces the development of the
science of the categories from its inception through Martin of Dacia ,
Peter of Auvergne, and Simon of Faversham to its culmination in Duns
Scotus. In order fully to appreciate what a science of the categories
looks like, and what problems must be overcome in positing such a
science, Newton includes a short overview of the medieval conception
of science and its various kinds, and demonstrates the appropriation of
Avicennian doctrines by Scotus in his unique claim that the categories
are the subject of a propter quid science.
One of the main monographs on categories to appear recently, and
one which is cited approvingly by many of the authors in this volume is
Categories and Logic in Duns Scotus, written by Giorgio Pini and published
by Brill Press. Pinis book traces the development of categories and logic
throughout the thirteenth century, through both major philosophers,
such as Aquinas and Henry of Ghent , and lesser known gures, such
as Radulphus Brito, and culminates in his masterful study on Duns
Scotuss treatment of the Categories. In this work, Pini advances the
rather controversial claim that for Scotus, the categories are not always
the same as they are studied in logic and as they are studied in meta-
physics; that for Scotus there is not always an isomorphism between
language, concepts, and things. Taking up the traditional interpreta-
tion of Scotus, however, is Todd Bates, who provides a brief criticism
of Pinis monograph and argues that for Scotus, common natures are
composite entities made up of a plurality of forms. To accomplish
his goal, Bates rst summarizes Pinis work, then presents texts from
6 lloyd a. newton
Scotus that seem to contradict Pinis interpretation, and nally offers
his own interpretation of Scotuss texts that Pini cites in favor of his
own views.
Given the importance of Pinis work in general
5
to the study of medieval
commentaries on the Categories, and to the generally positive reception
of his monograph in particular, I thought it only fair to allow Pini to
respond to Batess criticism. Thus, in the spirit of a medieval debate,
there follows Pinis reply to Batess criticism. In short, Pini argues that
Bates fails to appreciate the subtle nuances of Scotuss logic, and in
particular that the science of logic, for Duns Scotus, is radically distinct
from metaphysics. According to Pini, the questions that Bates raises
are not the concern of the logician, but of the metaphysician. For the
sake of argument, Pini grants that Scotuss metaphysics may be such
as Bates describes it, but argues that Scotuss logic is indifferent to and
does not imply any specic metaphysical view.
By the fourteenth century, most commentaries on the Categories were
no longer simple, literal expositions of the text, but highly evolved
question commentaries, whose aims were often far different from the
more moderate aims of explication that occupied almost all preceding
commentaries. Consequently, commentaries written during or after the
fourteenth century are often extremely complex, philosophically rich,
and exhibit an almost unbelieveable amount of freedom regarding the
views that are questioned. Thus, of the four remaining essays, the rst
explicates an extremely nominalistic interpretation of the Categories while
the second explicates a radically realist interpretation of the Categories.
The third essay exhibits a freedom of thought scarcely imaginable in
a medieval commentary, and the nal essay focuses on the (in)famous
problem of universals as it relates to the ten categories.
The rst essay, by Alexander Hall, discusses Buridan s nominalistic
explication of the categories along semantic lines, which enables him
to read the Categories as concerned with properties of terms rather than
things in the world, thereby effectively nipping any realist interpreta-
tion in the bud. However, to interpret the Categories nominalistically
5
See the bibliography for Pinis other writings in this area.
requires a complicated array of novel semantic terms pertaining to the
categories; terms which are ostensibly lacking in Aristotles short text.
Consequently, Halls essay focuses on Buridan s interpretation of being
said of and in insofar as those are relations that hold among terms,
explicable via the semantic properties of supposition, signication,
appellation, and connotation, thus presenting a coherent nominalist
reading of Aristotles work.
In contrast to Hall s explication of a nominalistic interpretation of
the Categories, Alessandro Conti s essay focuses on a realist interpreta-
tion. For in response to the various nominalistic interpretations of the
Categories, many philosophers sought a more realistic interpretation
of the categories. Among them, Robert Alyngton, one of the most
important authors of the generation after Wyclif , must be mentioned:
his commentary on the Categories, which relies on Burley and Wyclif, is
the most mature output of this realist interpretative tradition. Alyngton
was able to work out a coherently realist ontology of the categories and
a new semantic theory of second intentions, which developed into the
general strategy adopted by the Oxford Realists after Wyclif, insofar
as he methodically substituted reference to external objective realities
for reference to linguistic and/or mental activities.
The previous two essays amply demonstrate, I think, the wide variety
of what might be termed academic freedom available to medieval
philosophers. However, lest there be any doubt about the freedom phi-
losophers had to put forth controversial subjects and to suggest novel
views, Robert Andrews essay should help settle the issue. Specically,
Andrews focuses on Thomas Maulevelt, a rather obscure English phi-
losopher active in Paris in the early part of the fourteenth century. Like
Ockham, Maulevelt was engaged in a radical reduction in the number
of the categories. One by one all of the lesser categories fall, each
shown to be merely a way of speaking. Maulevelt, however, is willing
to go a step farther than Ockham. According to Andrews, Maulevelt
entertains a unique, radical hypothesis: we have no need to posit that
substance exists! The only category needed to describe the things of
the world is that of qualitya view that is not entertained again until
Hume . Following Andrews essay is question 16 of Maulevelts com-
mentary on Aristotles Categories, in which he entertains the possibility
that substances do not exist.
8 lloyd a. newton
Finally, this book ends where most medieval commentaries on the
Categories begin: with a discussion of that thorny issue of universals rst
introduced by Porphyry in his famous introduction to the Categories,
the Isagoge. Although it is not meant to be an introduction proper, this
last essay, by Alessandro Conti , discusses the complicated relationship
between universals and categories as they were treated in the fourteenth
and fteenth centuries. Specically, he outlines the problem of universals
in the late Middle Ages both from a systematic and from a historical
point of view, indicating the connections with the doctrine(s) of cat-
egories. To that end, he rst provides a short account of the standard
theories of universals worked out between the thirteenth and fourteenth
centuries; he then summarises Ockhams criticism of the traditional
view; and he concludes with a discussion of the main positions about
universals elaborated from 1330 ca. to 1430 ca. in some detail, thereby
showing their increasing relevance for the categories. Hopefully, Contis
essay, as well as the other essays in this book, will spark a new interest
in medieval theories of the categories analogous to the interest once
sparked by Porphyrys Isagoge.
THE MEDIEVAL POSTERITY OF SIMPLICIUS
COMMENTARY ON THE CATEGORIES:
THOMAS AQUINAS AND AL-FRB
Michael Chase
1. Simplicius Commentary on the Categories
in the Medieval West
Simplicius commentary on the Categories, probably written about 538
A.D.,
1
seems to have had little impact on the Latin-speaking world
until March of 1266, when it was translated into Latin by William of
Moerbeke (c. 1215c. 1286), probably at Viterbo. Whatever the rela-
tions may have been between William, a Dominican who ended his
life as archbishop of Corinth, and Thomas Aquinas ,
2
the fact remains
that it was thanks to Williams translation that Thomas appears to have
1
Ilsetraut Hadot, ed. Simplicius , sa vie, son uvre, sa survie; actes du colloque international de
Paris (28 sept.1 oct. 1985), (Berlin, 1987), (Peripatoi; XV), pp. 2021. Simplicius was the
author of one of a number of Neoplatonic Greek commentaries on the Categories that
have come down to us. We have others by Porphyry , Dexippus , Ammonius , Philoponus ,
Elias , and Olympiodorus . The Latin commentary by Boethius is largely based on
Porphyry. On the interrelations between all these commentaries, cf. Ilsetraut Hadot,
Simplicius , Commentaire sur les Catgories: traduction commente, I: Introduction. Premire partie:
(pp. 19, 3 Kalbeisch), sous la dir. de Ilsetraut Hadot; trad. par Philippe Hoffmann ;
avec la collab. de Ilsetraut Hadot & Pierre Hadot; commentaires et notes la trad.
par Ilsetraut Hadot; avec des appendices de Pierre Hadot & Jean-Pierre Mah (Leiden,
1990), (Philosophia antiqua; 50); Ilsetraut Hadot, Simplicius , Commentaire sur les Catgories:
traduction commente, III: Prambule aux Catgories. Commentaire au premier chapitre
des Catgories: (pp. 2140 Kalbeisch), sous la dir. de Ilsetraut Hadot; trad. de Philippe
Hoffmann ; avec la collab. de Ilsetraut Hadot, Pierre Hadot & Concetta Luna ; com-
mentaires & notes la trad. par Concetta Luna. (Leiden, 1990), (Philosophia antiqua;
51); Concetta Luna, Simplicius , Commentaire sur les Catgories dAristote, Chapitres 2
4; trad. par Philippe Hoffmann ; avec la collab. de Ilsetraut Hadot et Pierre Hadot;
commentaire par Concetta Luna (Anagg; 1).
2
The standard view, based largely on the testimony of Peter of Prussias Legenda
Alberti Magni, has it that William carried out his translations at the request of his friend
Thomas. Yet several recent scholars have debunked this account as legendary: cf. Jean-
Pierre Torrell , Initiation Saint Thomas dAquin: Sa personne et son uvre (Paris, 2002), pp.
25558, followed for instance by Mark D. Jordan , Rewritten Theology Aquinas After His
Readers (Oxford, 2006).
10 michael chase
been the rst Latin author to refer to Simplicius commentary.
3
Prior to
Williams translation, the work appears to have escaped the omnivorous
appetite for learning of Thomas master Albert the Great, although this
judgment may have to be revised as a result of the critical edition of
Alberts logical works currently in course of publication at Cologne.
Williams translation, in contrast, was to have a fairly wide circulation.
It is transmitted in twelve manuscripts,
4
several of which date from the
13th14th centuries, as well as in ve printed editions from the 16th
century.
5
Several of these editions also contain a second translation of
Simplicius commentary by Guilelmus Dorotheus . In the prefatory letter
to his translation, Guilelmus fustigates Moerbekes translation as the
work of one who was insufciently skilled in both Greek and Latin;
this, however, did not stop Dorotheus from copying Moerbekes new
translations of the lemmas from Aristotles Categories, which he passed
off as his own.
Pace Dorotheus, Moerbekes translation was a magnicent achieve-
ment. He made no pretence of producing a literary work, but translated
word-for-word, so faithfully that even today, his version is of great
assistance in the critical establishment and translation of the Greek text
of Simplicius commentary. Despite the fact that he appears to have
worked from a corrupt manuscript,
6
and that Latin still lacked much of
the philosophical vocabulary necessary for rendering such a complex
and technical work, William succeeded in producing a work that had
3
A. Pattin , Pour lhistoire du Commentaire sur les Catgories dAristote de Simplicius
au moyen ge, in Arts libraux et philosophie au Moyen Age: IV
e
Congrs international de philoso-
phie mdivale, Montral 1967 (Paris, 1969), pp. 10731078 , at p. 1073. Aquinas citations
of Simplicius commentary on the Categories include In duodecim libros Metaphysicorum
Aristotelis Expositio, lib. 3, lect. 11 eds. Cathala and Spiazzi; cf. In Octo Libros Physicorum
Aristotelis Expositio, lib. 1, lect. 15, section 138 ed. Marietti; De malo, q. 1, a. 1, ad 2, ad
7, ad 11; De unitate intellectus, cap. 3, p. 48 n. 76 ed. Keeler. Simplicius In Cat. is also
one of the primary sources of Thomas treatise De virtutibus.
4
Laurentius Minio-Paluello , ed. Aristoteles Latinus. I:15: Categoriae vel Praedicamenta
(Bruges, 1961), pp. lxxiilxxiii; A. Pattin, Pour lhistoire du Commentaire, I,
xxiiixxxix.
5
Minio-Paluello, Aristoteles Latinus , lxxiii noted four editions published by Hieronymus
Scotus at Venice in 1540/41, 1543, 1550, and 1567/68. To these must be added the
edition published by Paul of Genezano ( Venice 1516), who appears to have revised
Moerbekes translation after the Greek version of Simplicius commentary published
at Venice in 1499. See A. Pattin, Pour lhistoire du Commentaire, p. 1078.
6
According to Moerbekes own testimony, as preserved in the ms. Toledo, Bibl.
Capit. 47.11, fol. 179r, cited by A. Pattin , Pour lhistoire du Commentaire, p. 1077:
Sciat enim, qui hoc opus inspexerit, exemplar graecum valde fuisse corruptum et in
multis locis sensum nullum ex littera potui extrahere.
the medieval posterity of simplicius
commentary 11
a huge inuence on subsequent Western philosophy, not least because
of the large number of citations it transmitted from ancient authors,
ranging from the Presocratics to Simplicius immediate Peripatetic and
Neoplatonic predecessors.
7
In addition to Thomas Aquinas ,
8
Moerbekes translation of Simplicius
commentary on the Categories was used in the 13th century by Siger of
Brabant , Henry of Ghent , Giles of Rome, and Godefroid de Fontaine
(part of whose manuscript is now the ms. latin 16080 of the Paris
Bibliothque Nationale). Duns Scotus refers to it frequently as an
authoritative work, and it is cited by Jean Quidort , Peter of Auvergne ,
Jacques de Thrines , Durand de St. Pourain , Thomas of Strasbourg ,
Thomas Sutton , and James of Viterbo . The work continued to be cited
throughout the 14th century, by such authors as Siger of Courtrai and
the anonymous author of the ms. Erfurt, Amplon. F. 135.
in the Arabic tradition
The name of Simplicius is mentioned rather infrequently in the Arabic
tradition, although he may well have been used rather more often
than he is actually cited.
9
When he is mentioned, he is known as a
commentator on the rst book of Euclid information completely
lacking in Greek and Latin literatureand on Hippocrates, as well as
on Aristotles De Anima.
10
7
On the importance of Simplicius In Cat. as a source for the history of ancient
philosophy, see Wayne Hankey , Thomas Neoplatonic Histories: His following of
Simplicius, Dionysius 20 (2002), 153178; Michael Chase , Simplicius , On Aristotle, Categories
14; trans. with an introduction by Michael Chase (London, 2003), (The ancient com-
mentators on Aristotle), pp. 1 ff.
8
For what follows, cf. A. Pattin , Pour lhistoire du Commentaire, pp. 10731078;
A. Pattin, ed. Simplicius , Commentaire sur les Catgories dAristote, Traduction de Guillaume
de Moerbeke, 2 vols. (Louvain, 19711975), (Corpus Latinorum commentariorum in
Aristotelem Graecorum 12). vol. I, pp. xviiixxiii.
9
Helmut Gtje , Simplikios in der arabischen berlieferung, Der Islam 59 (1982),
631, p. 14, who points out that while Averroes never mentions the name of Simplicius ,
there are numerous parallels in the Cordobans works with passages from Simplicius
commentaries on the Physics and the De Caelo. On the translations of Greek com-
mentaries into Arabic, see Cristina DAncona , Greek into Arabic: Neoplatonism
in translation, in P. Adamson & R. C. Taylor, eds. The Cambridge Companion to Arabic
philosophy (Cambridge, 2005), pp. 1031.
10
He is also mentioned, as are a plethora of other Greek philosophers, as the author
of alchemical and occult works; cf. H. Gtje , Simplikios, 19.
12 michael chase
That Simplicius commentary on the Categories was known in the
Arabic world is not open to doubt.
11
Not only is such a work mentioned
in the bibliographical surveys of al-Nadm
12
and al-Qif ,
13
but over
and above these mentions, extracts from it are found in the Arabic
manuscript Paris, BN ar. 2346 of the Categories and other works from
the Organon, together with marginal glosses, prepared by al-asan ibn
Suwr (ob. 1017),
14
a student of Ab Zakary Yay ibn Ad (ob. 984),
the learned Christian student of al-Frb . Further Simplician material
may well be present in the still unpublished commentary on the Categories
by ibn Suwrs student Abdallh ibn al-ayyib (ob. 1043), and in the
fragmentary commentary by Ab-l-asan al-Amir (ob. 991).
15
The great Arabo-Islamic philosopher Ab Nar Muammad ibn
Muammad ibn arkhn ibn Awzalugh al-Frb (ob. 950) never men-
tions the name of Simplicius . However, as we have seen that Averroes
uses the physical commentaries of Simplicius without feeling obliged
to cite his source, so Philippe Vallat has recently argued persuasively
that Simplicius commentary on the Categories is an important, albeit
unacknowledged, source for the man who, largely because of his vast
output of works on the Stagirites logic, was known in Islamic philosophy
as the second master (al-muallim al-thn ), after Aristotle.
16
in the Works of Thomas Aquinas
The history of the inuence of Simplicius commentary on the Categories
on Medieval thought as a whole remains to be written. Thanks to such
11
F. E. Peters, Aristoteles Arabus (Leiden, 1968) , p. 7.
12
Fihrist, trans. B. Dodge, 1970, II, p. 598. Elsewhere (ibid., p. 640), however, al-
Nadm writes that Simplicius was the author of an Exposition on the fourth book of
Aristotles Categories (Kitb shar Qighriys li-Arisls al-maqla al-rbia). For speculation
on what this may mean, cf. H. Gtje , Simplikios, pp. 22 ff.
13
Ibn al-Qift , Ta rkh al-ukam (written c. 1240), ed. J. Lippert (Leipzig, 1903).
14
H. Gtje, Simplikios, pp. 2021. On this manuscript, which reveals the critical
work carried out on the Organon by Arabic scholars from the beginning of the tenth
century, cf. Henri Hugonnard-Roche , Remarques sur la tradition arabe de lOrganon
daprs le manuscrit Paris, Bibliothque nationale, ar. 2346, in C. Burnett, ed. Glosses
and commentaries on Aristotles logical texts (London, 1993), pp. 1928.
15
Gtje , Simplikios.
16
For a useful orientation to Frbs logical works, see Deborah L. Black, al-
Frb, in S. H. Nasr & O. Leaman , eds. History of Islamic philosophy (London, 1996),
pp. 17897, pp. 17984 with the literature cited at p. 193 n. 5.
commentary 13
scholars as Wayne Hankey ,
17
however, a fair amount of solid recent work
has been done on the reception of this work by Thomas Aquinas , and
I shall concentrate on this aspect of the matter in what follows.
According to Hankey , the main inuence of Simplicius commentary
on the Categories on Thomas Aquinas concerned his conception of the
history of philosophy in general, and more specically the relations
between Aristotle and Plato . Like all later Neoplatonists , Simplicius,
following a doctrinal trend already sketched by Antiochus of Ascalon
and fully expounded by Porphyry , believed that the doctrines of Plato
and Aristotle were ultimately reconcilable. When in his introduction,
Simplicius, following the standard Neoplatonic scheme of points to be
discussed (kephalaia)
18
at the beginning of a commentary on Aristotle,
discusses the qualities necessary for the good exegete of Aristotles text,
19

he stresses that such a master exegete must be thoroughly familiar with
the Stagirites writings and stylistic habits, and be objective, not always
striving to prove that Aristotle is right, as if he had enrolled himself
in the Philosophers school. When it comes to apparent disagreements
between Plato and Aristotle, however, the good exegete
must [. . .] not convict the philosophers of discordance by looking only at
the letter (lexis) of what [Aristotle] says against Plato ; but he must look
towards the spirit (nous), and track down (anikhneuein) the harmony which
reigns between them on the majority of points.
20
Thus, surface disagreements between the two great philosophers are
just that: supercial, and the skilled commentator must not let himself
be fooled by such apparent contradictions between the letter of their
works, but track down the fundamental agreement hidden in the spirit
of such texts.
17
Wayne Hankey , Thomas Neoplatonic Histories: His Following of Simplicius ,
Dionysius 20 (2002), 153178; Wayne Hankey, Aquinas and the Platonists, in The
Platonic Tradition in the Middle Ages: A Doxographic Approach, eds. Stephen Gersh and
Maarten J. F. M. Hoenen, with the assistance of Pieter Th. van Wingerden (Berlin,
2002), pp. 279324.
18
For the importance of these kephalaia in Frb, cf. Philippe Vallat , Frb et lcole
dAlexandrie. Des prmisses de la connaissance la philosophie politique (Paris, 2004), p. 61, citing
in particular the Book of words used in logic (Kitb al-alf al-mustamal l mantq), 51, ed.
Mahdi, where the order of enumeration corresponds to that in Simplicius , In Cat. In
particular, Frbs fourth point corresponds to the fourth point in Simplicus, In Cat.,
p. 8, 1015 Kalb., who is the only Neoplatonic commentator to include it.
19
This is Simplicius s eighth kephalaion, cf. Michael Chase , Simplicius, p. 103 n. 96.
All translations from Simplicus, In Cat. are taken from this work.
20
Simplicius , In Cat., p. 7, 2932 Kalbeisch.
14 michael chase
The basis of Simplicius attitude is the idea, again established by
Porphyry , that the philosophies of Aristotle and Plato , far from being
opposed, are in fact complementary. If they seem to disagree, this is
because, generally speaking, they are not talking about the same thing.
Aristotle writes from the viewpoint of the beginning student of philoso-
phy, whose mentality is determined by the sensible world around him.
His works therefore provide a virtually infallible guide for all subjects
concerned with the sensible world: logic, physics, meteorology, natural
history, and even, to a certain extent, psychology, which has at least
partly to do with the sensible world.
When studied in the proper reading order (taxis), starting with the
Categories and proceeding through the De interpretatione, Prior and Posterior
Analytics, Topics, and Sophistical Refutations, the Aristotelian Organon even-
tually provides the student with the tools of demonstration (apodeixis),
or reliable scientic reasoning. Armed with this indispensable prere-
quisite, he can then proceed to the study of ethics, physics, and even
metaphysics.
21
Yet this entire course of study, which a gifted student like Proclus
took two years to complete, was considered a mere introduction to
the study of the Platonic dialogues, which, when read in an order
inaugurated by Iamblichus ,
22
provided an understanding of the whole
of reality, including the intelligible world. Far from being mutually
contradictory, then, the works of Aristotle and Plato are indispensable
complements to one another. Apparent divergences are to be explained
by the fact that Aristotle was tailoring his instruction to what we may
anachronistically call an undergraduate audience, while Plato intended
his work for graduate students.
This pedagogical approach was especially convenient for explain-
ing the Categories, the work in which Aristotle is prima facie at his least
Platonic. Indeed, what could be more opposed to Plato than to main-
tain, as the Stagirite does in Categories 5, that substance in the primary
sense designates individual sensible things like human beings and
horses, while such intelligible entities as genera and species are merely
secondary? Surely this places Aristotle at the antipodes from his emi-
nent teacher?
21
On the importance of the taxis of the Aristotelian works in Frb, who links it
in an interesting and apparently original way to the parable of the cave in the Republic,
cf. Ph. Vallat, Frb , pp. 13435; 18990.
22
Dominic OMeara , Pythagoras revived. Mathematics and philosophy in Late Antiquity
(Oxford, 1989), pp. 9799.
commentary 15
By no means. As we have seen, Simplicius and the other Neoplatonists
argued that Aristotle intended the Categories to be the rst philosophical
work tackled by students, and so he tailored it to an audience of begin-
ners. When, therefore, he claims individual substances are primary, he
really means primary for philosophical beginners, who, living on the level of
sense perception, are ignorant of the existence of intelligible entities.
An alternative explanation of the Categories restriction of substance
to the sensible world was developed by Porphyry in his minor commen-
tary on the Categories by questions and answers,
23
which was inuential
on Simplicius own Commentary. The skopos or goal of the Categories
is simple words.
24
When, at the dawn of mythical history, the Sage or
council of Sages rst got together to impose names on things,
25
they
began by naming individual sensible objects: chair, man, dog, sun, etc.
Thus, then, is another sense in which sensible substances are primary:
not only are they primary from the viewpoint of beginning philosophy
students, but they were the primary objects of the rst imposition of
names. But the Categories is precisely devoted to the study of the simple
words that were the objects of this rst imposition, as opposed to the
De interpretatione, which studies the results of the second imposition, in
which the Sages returned to the results of the rst imposition and, in
a metalinguistic approach, divided the words they had used into nouns
and verbs.
Such ideas seem to have struck a chord in Thomas Aquinas . In his
Quaestio disputata de spiritualibus creaturis, written in 12671268, he explains
the differences between Plato and Aristotle as follows:
The diversity of these two opinions comes from the fact that when inquir-
ing into the nature of things, some proceeded from intelligible reasons,
and this was peculiar to the Platonists, and others from sensible things,
and this was peculiar to the philosophy of Aristotle, as Simplicius says
in his commentary on the Categories.
26
23
Porphyry , In Cat., p. 91, 1927.
24
Later, the standard Neoplatonic denition of the skopos of the categories is given
as being [. . .] about simple, primary words which signify the primary and most
generic of beings by means of simple, primary notions (Simpl., In Cat., p. 13, 1920
Kalbeisch). On the skopos (Arabic ghara ) of Aristotles works according to Frb, for
whom it constitutes the privileged criterion for judging the correctness of interpreta-
tions of Plato and Aristotle Cf. Ph. Vallat , Frb, pp. 17275; 17680; 242 ff. For
Frb, the skopos of the Categories are the simple notions or primary principles that are
innate within all human beings; cf. Vallat, p. 210.
25
On the institution of language in Frb, cf. Ph. Vallat , Frb, pp. 25566.
26
Aquinas , Quaestio disputata de spiritualibus creaturis, ed. Cos (Rome, 2000) (Commissio
Leonina: vol. 24/2), art. 3, p. 40, lines 27582 : Harum autem duarum opinionium
16 michael chase
Although here he leaves out Simplicius pedagogical explanation,
Thomas agrees with the commentator that Plato s approach is basically
deductive and rationalist, while Aristotles is fundamentally inductive
and empiricist. Thomas may also have been thinking of another pas-
sage from Simplicius Commentary in Cat, where, once again following
the typical Neoplatonic scheme of points to be discussed (kephalaia) in
a commentary,
27
he discusses the difference in style between Plato and
Aristotle:
[Aristotle] always refuses to deviate from nature; on the contrary, he con-
siders even things that are above nature according to their relation to nature,
just as, by contrast, the divine Plato , according to Pythagorean usage,
examines even natural things insofar as they participate in the things
above nature.
28
Likewise, according to Thomas, while Aristotle is a literalist, Plato
transmits divine things poetically and in fables.
29
He says everything
guratively and teaches through symbols, intending something other
through his words than what they themselves say.
30
This provides
another explanation of the apparent disagreement between Aristotle
and Plato: when the former criticizes the latter, he is arguing only
against the literal meaning of Platos words, not, of course, because
the Stagirite himself failed to grasp the true meaning of Platos words,
but becauseand here the pedagogical approach surfaces in Thomas
as wellhe was anxious lest students be led astray by the false surface
meaning of these words.
In the context of the seventh kephalaion of his commentary on the
Categories, Simplicius explains that while all ancient philosophers took
care not to divulge their doctrines to the unworthy and the unprepared,
they used different methods to achieve this goal. Plato preferred the
use of myths and symbols, while Aristotle cultivated deliberate obscu-
rity, perhaps because he rejected the indeterminate hidden meaning
diuersitas ex hoc procedit quod quidam ad inquirendam ueritatem de natura rerum
processerunt ex rationibus intelligibilibus: et hoc fuit proprium Platonicorum, quidam
uero ex rebus sensibilibus: et hoc fuit proprium philosophiae Aristotilis, ut dicit
Simplicius in Commento super Praedicamenta.
27
The seventh kephalaion, cf. M. Chase , Simplicius , p. 103 n. 91.
28
29
res divinas potice et fabulariter tradiderunt, Aquinas , In De Caelo, lib. 1, lect. 22, sect.
227, p. 108.
30
Aquinas , Sententia libri De Anima (Leonina 1984), vol. 45, pars 1, cap. 8, p. 38,
lines 39.
commentary 17
(huponoia) of myths and symbolssuch things can, after all, easily be
understood differently by each interpreter.
31
Similarly, according to Thomas, Aristotle was not against Plato s
understanding, which was sound, but against his words.
32
This cor-
rect approach on the part of most commentators, Thomas notes, is in
contrast to that of Alexander of Aphrodisias, who takes Aristotle to be
arguing not against Platos surface meaning; but against his intellectual
intention. Again, Thomas main source of information on the views of
Alexander is Simplicius commentary on the Categories.
33
Yet Alexanders
view is inadequate. As we have seen, it is not enough to remain at a
supercial level, where Aristotle attacks the surface meaning of Platos
words: instead, the interpreter must delve beneath the letter of Platos
text in search of its hidden spirit. When this is done successfully, it will
be found that most or all of the apparent disagreements between Plato
and Aristotle vanish into thin air.
4. Simplicius in al-Frb
This concordist attitude is shared by al-Frb ,
34
who believed that
the aim of these two philosophers is one and the same and that they
intended to offer one and the same philosophy.
35
Like Simplicius , Frb
also likes to explain apparent instances of discord between Plato and
Aristotle by the fact that they are not talking about the same thing,
and/or that their intentions in writing the passage in question were not
comparable. This is particularly true in the case of substance: the judg-
ment of the two philosophers does not pertain to the same viewpoint
(min jiya wida), or to the same goal (maqd wid ).
36
As Philippe Vallat
has shown,
37
Frb uses the same or similar principles to explain the
31
32
Aquinas , In de caelo, 1.22, sect. 228.
33
Wayne Hankey , Aquinas and the Platonists, p. 293.
34
Who probably derived it at least in part from Porphyry s lost work On the unity of
the philosophical schools of Plato and Aristotle; cf. Richard Walzer, Porphyry and the Arabic
tradition, in Porphyre (Vandoeuvres, 1966), (Entretiens sur lAntiquit classique 12),
pp. 27599, 285 ff.
35
Frb, Kitb tal al-sada, cited by R. Walzer , Porphyry and the Arabic tradi-
tion, pp. 27599, 287.
36
al-Frb , Book of the concordance in view of the two sages, the divine Plato and Aristotle,
eds. & French transl. F. W. Najjar & D. Mallet (Damascus, 1999), 20, p. 81, cited by
Ph. Vallat, Frb , p. 173.
37
Ph. Vallat , Frb, p. 173.
18 michael chase
surface disparities between Plato and Aristotle on the subject of the
validity of mixed syllogisms;
38
on vision;
39
moral dispositions;
40
remi-
niscence;
41
and the eternity of the world.
42
Perhaps the most striking point of convergence between Simplicius
and Frb concerns the mechanism of teaching itself. In his com-
mentary on the Categories, Simplicius explains the metaphysical presup-
positions of Neoplatonic pedagogy as follows.
43
When the soul is in the
intelligible, it possesses knowledge of all things immediately, for there
is not yet any distinction between subject and object: language is thus
unnecessary. When the soul sinks into generation, howeverthat is,
when it becomes incarnate in a human bodyit loses this direct knowl-
edge and can only project notions that approximate the noetic realities
it used to intuit directly. Although the soul still possesses sparks or
embers of its original, innate knowledge, these have been smothered
and cooled down as the soul has become lled with forgetfulness. The
only remedy for this situation is philosophical instruction:
When, however, the soul has fallen into the realm of becoming, it is lled
with forgetfulness, and requires sight and hearing in order to be able to
recollect. For the soul needs someone who has already beheld the truth,
who, by means of language ( phn ) uttered forth from the concept (ennoia),
also moves the concept within [the soul of the student], which had until
then grown cold [. . .] For intellections (noseis) which proceed forth from
other intellections [sc. those of the teacher] also set things in motion
immediately, and they join the learners notions to those of the teacher,
by becoming intermediaries (mesottes) between the two. When intellec-
tions are set in motion in an appropriate way, they adjust themselves to
realities, and thus there comes about the knowledge of beings, and the
souls spontaneous eros <sc. for knowledge> is fullled.
44
The person who has already beheld the truth is the philosophy teacher,
who, by carefully chosen discourse, is able to rekindle the sparks that
lie hidden in the students common notions or innate ideas, thereby
38
al-Frb , Concordance, 3133.
39
Ibid., 3540.
40
Ibid., 4244.
41
Ibid., 52.
42
Ibid., 5357.
43
Simpl., In Cat., p. 12, 1313, 11 Kalbeisch. On this passage, see the magnicent
paper by Philippe Hoffmann, Catgories et langage selon Simplicius . La question
du skopos du trait aristotlicien des Catgories, in Ilsetraut Hadot , ed. Simplicius, pp.
6190.
44
Simpl., In Cat., p. 12, 2613, 4 Kalbeisch.
commentary 19
making him aware of the knowledge he possessed, unbeknownst to
himself, all along. This is, of course, an adaptation of Socratic/Platonic
maieutics, based on the doctrine of recollection or reminiscence (Greek
anamnsis).
In works such as the Political regime (Kitb al-siysa al-madaniyya, al-
mulaqqab bi-mabdii al-mawjdt, ed. Najjar ); the Political specilegium (Ful
muntazaa, ed. Najjar); the Book of the accession to happiness (Kitb tal
al-sada, ed. al-Yasin ) and the Book of recollection of the way leading to hap-
piness (Kitb al-tanbh al sab-l-sada, ed. al-Yasin), Frb sets forth a
doctrine that is strongly reminiscent of this theory, of which Simplicius
is the only commentator on the Categories to provide a detailed account.
For the Second Master, all human beings are provided by nature with
certain innate ( jarziyya) intelligible principles (awi l ),
45
instilled in us
by the Agent Intellect, but which we have forgotten, and it is the task
of the philosophy teacher to remind us of them by means of lessons or
philosophical dialogues in which the words are carefully chosen to t
or be suitable to (aara al) these innate intelligibles.
46
Once they are
appropriately developed under the teachers guidancethat is, classi-
ed and articulatedthese principles can be used as logical premises,
and thus serve as the starting-point for demonstration.
47
In the words
of Philippe Vallat ,
Nothing, then, seems to be lacking from Frbs exposition for us to be
able to pronounce him faithful to the thought of Simplicius , especially in
that there is no doubt that it is at the summit of this ascent that he too
placed the instant in which man intelligizes the Agent Intellect, achieves
felicity, and becomes divine.
48
If Vallat is correct, not only does Frbs adaptation of this Simplician
doctrine conrm the inuence of the latters commentary on the Categories
on the Islamic thinker, but it enables us to deduce details of the function-
ing of this doctrine that are lacking from Simplicius preserved work.
45
These are the equivalent of the Greek koinai ennoiai, which play a vital role in
Neoplatonic thought, especially that of Porphyry ; cf. Michael Chase , tudes sur le com-
mentaire de Porphyre sur les Catgories dAristote adress Gdalios. Diss. cole pratique des
Hautes tudes, 5th section, Paris 2000.
46
These words t with the primary innate intelligibles, and are therefore able to
reactivate them, because they are analogous to them (mutansiba) in a strictly dened
technical sense; cf. Ph. Vallat 2004, 208, and especially his Appendice: lAnalogie de
ltre, pp. 34765.
47
Cf. Ph. Vallat , Frb, pp. 16061.
48
Vallat , ibid., p. 214.
20 michael chase
5. Thomas, Simplicius and Iamblichus :
The interpretation of Wayne Hankey
In view of what we have seen, we can agree with Wayne Hankey when
he summarizes the inuence on Thomas Aquinas of Simplicius com-
mentary on the Categories. Simplicius work, he writes, is important:
49
1. for ideas and distinctions used in Thomass philosophical constructions,
2. for knowledge of the content of the history of ancient philosophy,
and
3. of the Peripatetic and Platonic hermeneutical traditions,
4. for theories about the history of philosophy used by Aquinas [. . .]
5. but also for that which conveys all these, namely, the commentaries
on Aristotles works.
Hankey is also correct, as we have seen, to point out the crucial impor-
tance for Thomas, as for Simplicius , of the notion of the ultimate
reconcilability of the philosophies of Plato and Aristotle. Quite apart
from specic points of doctrine, Thomas absorbed from Simplicius
in general and from his commentary on the Categories in particular a
conception of what philosophy is.
Here, however, I must express some reserves with regard to another
aspect of Hankey s important work. In line with ideas he has expressed
elsewhere, Hankey is anxious to emphasize the continuity between
Iamblichus and Thomas Aquinas .
50
Iamblichus, who played a key
role in the establishment of the Neoplatonic school curriculum, was
also of fundamental importance for Simplicius commentary on the
Categories. This is not open to doubt; what seems more questionable,
however, is the specic respects in which Hankey claims Iamblichus
transformed the Platonism that Thomas eventually absorbed through
Simplicius and his fellow-commentators. Iamblichus, claims Hankey,
turned Neoplatonism
51
49
Wayne Hankey, Thomas Neoplatonic Histories, p. 156.
50
Cf. especially Wayne Hankey , Philosophy as a Way of Life for Christians ?
Iamblichan and Porphyrian Reections on Religion, Virtue, and Philosophy in Thomas
Aquinas , Laval thologique et philosophique (2003), 193224, where the author attempts
to refute Pierre Hadot s views on Christianitys role in the decline of philosophy as a
way of life by showing that philosophy continues to be a way of life for Thomas, who
continues the tradition of Iamblichus . I disagree with this view, but this is not the place
to enter into the controversy in detail.
51
Wayne Hankey, Thomas Neoplatonic Histories, p. 158.
commentary 21
(i) toward a positive relation to the sensible and the material,
(ii) toward Aristotle, and
(iii) toward theurgic (or sacramental) religion.
Points (i) and (iii) fall outside the framework of this study, although I
will say with regard to the rst point that it is not at all clear in what
sense Iamblichus had a more positive attitude towards the sensible and
the material than, say, his predecessors Plotinus author of a lengthy
treatise against the Gnostics, whom he reproaches for their systematic
denigration of the sensible worldand Porphyry . Hankey s point (ii),
in contrast, deserves some brief remarks. At least as far as the doctrine
of the categories is concerned, Iamblichus may indeed have been more
favorably disposed to Aristotle than, say, Plotinus, who subjects the
Stagirites doctrines to a devastating critique in Enneads VI 13.
52
But
Iamblichus was certainly no more pro-Aristotelian than his own teacher
Porphyry, who had been responsible, prior to Iamblichus, for introducing
the systematic study of Aristotles works into the Neoplatonic school
curriculum. We ought not to forget that while it is true that Simplicius
based his commentary on the Categories to a large extent on Iamblichus
commentary on the Categories, Iamblichus in turn had based his own
commentary on the lost commentary on the Categories by Porphyry in
seven books, addressed to Gedalius. In this work, Porphyrys method
had been to refute the objections raised by Plotinus against Aristotles
doctrines of the categories. To be sure, Iamblichus often, or rather
usually, nds Porphyrys refutations inadequate, and spends many pages
driving home and elaborating upon the criticisms Porphyry had merely
sketched. An example may help to illustrate this point.
At in Cat., p. 302, 5 ff. Kalbeisch, Simplicius reports the objection
by Plotinus (Ennead VI, 15) against the Aristotelian category of Doing
(to poiein): Doing, said Plotinus, along with its opposite Being-affected
(to pathein), should be ranged as a sub-genus of Activity (to energein) or
rather Motion (kinsis). Porphyry s response
53
to this criticism of Aristotle
is as follows: in some things motion seems to be one and continuous in
52
Even this point is controversial, with several modern scholars (notably Franz de
Haas and Steven Strange ) arguing that Plotinus attitude to the Aristotelian doctrine
of the Categories is not one of rejection, but constructive criticism and adaptation.
This view has, I believe, been convincingly refuted by the recent work of Riccardo
Chiaradonna . I hope to return to this point elsewhere.
53
Porphyry , Ad Gedalium, fr. 71 in A. Smith, ed. Porphyrius, fragmenta (Stuttgart-Leipzig,
1993).
22 michael chase
the case of both doing and being-affected. When one throws or strikes
an object, for instance, the motion of the thrower is the same as the
motion of the object thrown or struck, but it becomes an affect ( pathos)
of the object, while it is an action ( poisis) of the agent. Therefore,
doing and being-affected cannot pertain to the same genus of motion,
but there is a specic differentia between them.
Iamblichus nds this solution inadequate: Porphyry went too far
aeld in search of it, derived it from premises that are no more clear
than their conclusion,
54
and which are not in accord with Aristotles
beliefs. It is not because one kind of action behaves this way that we
can conclude they all do, nor should Porphyry have chosen the kind
of action that is ontologically last
55
that which takes place through
pushing and shoving. He was also wrong to agree with the Stoics that
all action takes place by contact: contact is accidental in action, the
real cause of which is the mutual suitability of agent and object of the
action, as is shown by many obvious examples. To claim that the agent
has the same substance as the thing being affected is to sow confusion
everywhere, destroying the axioms of physics. If there is such a motion
as Porphyry postulates, made up of the activity of the agent and the
change of the object, then it does not belong in the pure genera (i.e.,
the intelligible world) of the categories, which are pure and unmixed,
but is merely secondary, like all compound things. Instead, motion is
common to doing and being-affected, but it is also separate from and
intermediary between them, proceeding from the agent and accom-
plishing the affect ( pathos) in the object being affected. Thus, just as the
motive agent and the mobile object are two separate things, so the agent
and the things-being-affected have also been divided into two.
Here, both Porphyry and Iamblichus agree that Plotinus is wrong
in his suggestion of a critical modication to Aristotles doctrine of
the categories. Yet while Porphyry apparently restricted himself to
adducing a counter-example to Plotinus argument, Iamblichus sets out
to demolish both Porphyry and Plotinus by a plethora of arguments,
including the highly characteristic one that Porphyrys doctrine of a
motion made up of doing and being-affected could not, as a compound
entity, belong to the intelligible world.
54
Thereby violating Aristotles strictures in the Posterior Analytics, I, 2, 70b2024.
55
That is, presumably, because it is restricted to the merely sensible world, subject
to the laws of physics.
commentary 23
We have here a clear example of Iamblichus tendency to incorporate
his intellective theory
56
throughout his commentary on the Categories. Far
from regarding this treatise, as Porphyry did, as a work for beginning
philosophers that focuses on the sensible world, Iamblichus assumes
Aristotles treatise is applicable to the intelligible world.
For Iamblichus , as followed by Simplicius , the source of Plotinus
error is that he confuses activity and motion, which is far from being the
truth.
57
This being the case, one may question the aptness of Hankey s
account of this passage. According to this scholar, one of the means
used in Simplicius commentary on the Categories to reconcile Plato
and Aristotle
[. . .] is the notion of motionless motion which in this commentary is
approached when considering making and suffering (de facere et pati )
and when treating motion (de motu). In both places Simplicius reects
on Aristotle through Plotinus and his followers, especially Iamblichus .
Simplicius shows Plotinus trying to dissolve the difference between activ-
ity and motion and between rest and motion. In learning, in intellectual
life, for what is without parts and what is perfect, these exist together in
such a way as to bring together apparently opposed teachings of Plato
and Aristotle. Aquinas , sometimes by way of the Arabic commentators,
is an heir of Plotinus, Simplicius, and others like Proclus , who teach this
Neoplatonic commonplace. Thomas does not only use this construction
to reconcile the two greatest philosophical authorities, motionless motion
is profoundly important for the structure and content of his own thought,
including his treatment of the Trinity.
58
In fact, however, while it is true that Simplicius shows Plotinus trying
to dissolve the difference between activity and motion, he also shows
Iamblichus refuting this notion, which he identies as Stoic , in no uncertain
terms. It is, moreover, not immediately clear what Hankey supposes the
connection to be between the idea of motionless motion and that of
abolishing the difference between activity and motion, or that between
rest and motion. They are certainly not the same thing: that motion
56
Cf. Simplicius , In Cat., p. 2, 914: in his Commentary, Iamblichus [f ]or the
most part, [. . .] followed Porphyry right down to the letter, but he picked out some
things and articulated them in order to make them more clear. At the same time, he
contracted the scholastic long-windedness Porphyry had used against the objections; and
he applied his Intellective Theory everywhere, to almost all of the chapter-headings.
On Iamblichus noetic theory, cf. John Dillon , Iamblichus Noera theria of Aristotles
Categories, Syllecta Classica 8 (1997), 6577.
57
Ho pollou dei althes einai, Simplicius , In Cat., p. 304, 3233.
58
Wayne Hankey , Aquinas and the Platonists, pp. 3001.
24 michael chase
is equivalent to activity is a Stoic idea taken up by Plotinus but unani-
mously rejected by the subsequent Neoplatonic commentators, while
motionless motion is an idea absent from Plotinus and fully developed
by Proclus , but which already appears in the Anonymous Commentary on
the Parmenides,
59
which Pierre Hadot has attributed, no doubt correctly,
60

to Porphyry . In sum, it is unlikely that Thomas derived his idea of
motionless motion from Simplicius commentary on the Categories. If he
did, then he was not following Iamblichus but going against his express
view. Hankey had previously suggested that Thomas derived this view,
which is crucially important in his thought, from Arabic Neoplatonism.
61

This may be, but it seems even more likely that the idea of motionless
motion reached Thomas by way of the Pseudo-Dionysius .
62
59
14, 2123, ed. Pierre Hadot, Porphyre et Victorinus , 2 vols. (Paris, 1968) , vol. 2, p. 110:
And since they [sc. existence, life and thought] are all activities; according to exis-
tence the activity would be stationary (Kai pasn ousn energein kai hs kata men tn huparxin
hestsa an ei h energeia). Hadot notes this idea was taken over from Porphyry by Marius
Victorinus , cf. Adv. Ar. III, 2, 36 cessans motus; IV, 8, 26. It has been suggested that
both Porphyry and Victorinus may in fact have taken it over from the Gnostics, and
more specically the treatise Allogenes in the Nag Hammadi treatises; cf. M. Tardieu in
Michel Tardieu-P. Hadot, Recherches sur la formation de lApocalypse de Zostrien et les sources
de Marius Victorinus . Pierre Hadot, Porphyre et Victorinus: questions et hypothses
(Leuven, 1996). (Res orientales; 9), p. 84; but more recently Ruth Majercik has argued
that it was the Gnostics who took over such conceptual schemes from Porphyry, cf. Ruth
Majercik , Porphyry and Gnosticism, Classical Quarterly 55.1 (2005), 27792.
60
See, most recently, Ruth. Majercik, Porphyry and Gnosticism.
61
Wayne Hankey , Review of: Simplicius : On Aristotles Physics 5, translated by J. O.
Urmson, notes by Peter Lautner. The Ancient Commentators on Aristotle (New York, 1997),
Bryn Mawr Classical Review 9.6 (1998), 5858, p. 586: This different kind of motion,
the act of the perfect, Aquinas learns about from the Arab Neoplatonists .
62
See, for instance, De divinis nominibus IX, 6, 910 (What shall we say, when again
the theologians say that the Immobile (ton akinton) proceeds toward all things and is
moved (kai kinoumenon)? In his scholia on this passage, Maximus the Confessor (Migne,
PG 4, 381AD) sets out to identify the sense in which the Immobile God is moved.
After eliminating the eight kinds of motion enumerated by the Ps.-Dionysus, Maximus
concludes: Thus, God is not moved according to any of these motions, but that which
is called motion of the Immobile. . . . is His will which leads to the generation of beings
and the procession of His providence toward all things. These texts, together with their
Byzantine and Arabic posterity, have been brilliantly studied by Marwan Rashed , La
classication des lignes simples selon Proclus et sa transmission au monde islamique,
in C. DAncona & Giuseppe Serra (eds), Aristotele e Alessandro di Afrodisia nella tradizione
araba (= Subsidia Mediaevalia Patavina 3) (Padova, 2002), pp. 25779.
commentary 25
Conclusion
The particular parallels we have noted between Thomas and al-Frb
may be indicative of a deeper similarity, which Simplicius commentar-
ies, including that on the Categories, may help to explain.
In a reversal of traditional viewpoints, recent commentators have
argued that the philosophies of both Thomas Aquinas
63
and Frb,
64

usually considered as followers of the Peripatetic school, are in fact
basically Platonist . Paradoxically, however, the same scholars have also
argued that neither of these philosophers had actually read Plato .
65

This odd situation can be explained by the nature of the sources of
both Thomas and Frb, which present denite similarities. Neither
had access to complete translations of the works of Plato. Both were
consequently forced to rely on the works of Aristotle, but this was an
Aristotelian corpus quite unlike the one studied in the West today. It
included worksthe Liber de Causis was most inuential in Thomas
case, while the Theology of Aristotle may have played an analogous role
in the case of Frbwhich we now know to be apocryphal compi-
lations of Neoplatonic texts deriving from Proclus , Plotinus , and pos-
sibly Porphyry .
66
Equally importantly, however, it included Neoplatonic
commentaries on the genuine works of Aristotle, including those by
Simplicius .
As we have glimpsed, the philosophy of both Frb and Thomas
Aquinas is profoundly influenced by the kind of Neoplatonizing
interpretation of Aristotle that lls the commentaries of Simplicius ,
Ammonius , Themistius and other late antique professors of philosophy.
These commentaries are the source of most of the common elements
in their thought, the most crucial of which is no doubt the idea of
63
Cf. Wayne Hankey , Aquinas and the Platonists, p. 284: Thomas hermeneutical
horizon was profoundly, extensively and subtly Platonist .
64
Ph. Vallat , passim.
65
In the case of Thomas, cf. R. J. Henle , Saint Thomas and Platonism (Hague, 1956),
p. xxi.
66
We might also count the works of the Pseudo-Dionysius Areopagita among these
sources. As Hankey has shown, Thomas began his philosophical career by consider-
ing the Ps.-Dionsyius as an Aristotelian; then, once he had read Proclus Elements
of Theology in Moerbekes translation, he came to realize the Platonic nature of the
Dionysian works. Ph. Vallat has pointed out a number of parallels between the works
of the Ps.-Dionysius and those of Frb: cf. Vallat, Frb, pp. 33; 76; 88 n. 1; 291
n. 1; 307 n. 2; 334 n. 2; 346; 371.
26 michael chase
the ultimate reconcilability of Plato and Aristotle. According to both
Thomas and Frb, both Plato and Aristotle teach that there is a
single divine cause that perpetually distributes being to all entities in a
continuous, graded hierarchy.
There are, of course, also profound differences in the ways Thomas
and Frb interpreted and utilized the doctrines they both received
from the Alexandrian commentators. For Thomas, who (certainly
indirectly) follows Iamblichus in this regard, philosophy occupies a
subordinate position within theology, while for Frb, whatever his
genuine religious beliefs may have been, philosophy remains the nec
plus ultra, capable of providing ultimate happiness through conjunction
with the Agent Intellect.
The contrasting attitudes of Thomas and Frb may, in turn, be
traceable to a similar contrast with late antique Neoplatonism. Porphyry
of Tyre was considered by his successors to have held that philosophy
alone was sufcient for salvation, consisting in the souls denitive
return to the intelligible world whence it came, while Iamblichus was
held to have placed the emphasis on the need for religion, in the form
of theurgical operations and prayers, and the grace of the gods.
67

What seems to have been at stake in the arguments between the two
was ultimately no less than the nature of philosophy: is it the ultimate
discipline, sufcient for happiness, as Porphyry held, or is it merely an
ancilla theologiae, as was the view of Iamblichus? Thomas and Frb, who
had at least some knowledge of these debates through the intermediary
of such sources as Simplicius commentary on the Categories, seem to
have prolonged this controversy, Thomas siding with Iamblichus and
Frb with Porphyry.
Wayne Hankey has written,
68

Not only for both [sc. Iamblichus and Aquinas ] is philosophy contained
within theology, and theology contained within religion, but also, for both
centuries its great teachers are priests and saints. In order to be doing
philosophy as spiritual exercise belonging to a way of life, we need not
engage directly in self-knowledge.
67
In the judgment of Damascius , the teacher of Simplicius , . . . some prefer phi-
losophy, like Porphyry and Plotinus and many other philosophers; others hieratics [i.e.,
theurgy], like Iamblichus , Syrianus , Proclus and all the other hieratics, Damascius, In
Phaed., 1, vol. I, 172, 13 Westerink.
68
Wayne Hankey , Thomas Neoplatonic Histories, p. 160.
commentary 27
Such ideas were anathema to Porphyry , the other great Neoplatonist
whose ideas were transmitted to posterity by, among other sources,
Simplicius commentary on the Categories. For the Tyrian thinker, as
for Frb writing some six centuries after him, philosophy is not
subordinate to religion, nor are its teachers priests or saints, but it is
autonomous and capable, all by itself, of ensuring human felicity both
in this life and the next. Philosophy for Porphyry was indeed a way of
life, an important part of which was reading and commenting on the
philosophical texts of the ancient Masters. For Porphyry, however, who
wrote a treatise On the Know thyself ,
69
as for the entire ancient tradi-
tion which, as Pierre Hadot has shown, considered philosophy to be
a way of life,
70
self-knowledge was the indispensable starting-point for
all philosophy.
71
Indeed, one may question whether this was not the
case for Iamblichus as well: it was he, after all, who established the
First Alcibiades as the rst Platonic dialogue to be read and studied in
the Neoplatonic curriculum;
72
but the skopos or goal of this dialogue,
for Iamblichus, was none other than self-knowledge.
73
Whatever may have been Iamblichus particular view, the Hellenic
tradition on the whole was unanimous on the crucial importance of
self-knowledge as the starting-point for philosophical education.
74
When
69
Addressed to none other than Iamblichus . The fragments are printed by Smith,
Porphyrius, fr. 273275.
70
Pierre Hadot , Philosophy as a way of life: spiritual exercises from Socrates to Foucault.,
ed. with an introduction by Arnold Davidson; trans. Michael Chase (Oxford, 1995);
Pierre Hadot, What is ancient philosophy?, trans. Michael Chase (Cambridge, Mass,
2002).
71
Cf. Porphyry , Sentence 40, 4, p. 5, 56 Lamberz; Letter to Marcella, ch. 32; etc. On
the central importance of self-knowledge for Porphyry, cf. Pierre Hadot, Porphyre et
Victorinus, I, 91 n. 1; 327328; J. Bouffartigue & M. Patillon, Porphyre, De labstinence,
I (Paris 1977), pp. liiilx; and the further literature cited by R. Majercik , Porphyry
and Gnosticism, p. 285 & n. 48.
72
Proclus , In Alc., 11, 1421, vol. I, p. 9 Segonds : the divine Iamblichus assigns
the Alcibiades the rst rank among the ten dialogues which, he believes, contain the
totality of Plato s philosophy.
73
A.-Ph. Segonds , Introduction to his edition of Proclus Commentary In Alc. (Paris
1985), vol. 1, p. xxv. Other texts by Iamblichus emphasizing the importance of self-
knowledge (Protreptic, 21; De myst., 10, p. 286, 11; Life of Pythagoras, 18, 83) are cited
by Pierre Courcelle, Le Connais-toi toi-mme chez les no-platoniciens grecs, in
Le noplatonisme. Actes du Colloque de Royaumont, 913 juin 1969 (Paris 1971), (Coll.
internat. du CNRS), pp. 15366, pp. 15657.
74
As Mme Ilsetraut Hadot points out in a personal communication, Simplicius ,
who may have been writing in arrn, also takes the First Alcibiades as the starting-
point for his Commentary on the Manual of Epictetus (In Ench. Epict., Praef. 82 ff.),
Tout le commentaire, avec ses explications frquentes sur la nature des trois espces
28 michael chase
in 946 the traveler al-Masd visited arrn in Mesopotamia, center
of the pagan bians, he saw, inscribed on the door-knocker of the
central temple, an inscription in Syriac reading He who knows his
nature becomes god,
75
which is, as Tardieu was the rst to recognize,
a reference to Plato s Alcibiades 133 C. When we recall that, according
to some of his biographers, Frb went to Harrn at about the time
of Masds visit to complete his studies of the Aristotelian Organon,
one is not surprised to nd that self-knowledge is as essential for Frb
as it was for Porphyry , with several of whose works the Second Master
seems to have been familiar.
76
In Frbs noetics, the potential intellect (al-aql bil-quwwah) becomes
an intellect in act (al-aql bil-l ) when, by abstracting the forms in mat-
ter from their material accompanying circumstances, it receives these
disembodied forms within itself. Unlike the forms stamped in wax,
however, which affect only the surface of the receptive matter, these
forms penetrate the potential intellect so thoroughly that it becomes
identical with the forms it has intelligized. Once it has intelligized all
such intelligible forms, the intellect becomes, in act, the totality of
intelligibles. The human intellect has thus become an intelligible, and
when it intelligizes itself, it becomes an intelligible in act.
77
dmes qui se trouvent dans lhomme et leurs relations avec le corps, sur la position
de lme humaine dans lunivers, sur les limites de ses possibilits daction etc., montre
que selon lui une amlioration de ltat thique de lhomme ne peut se faire que sur
la base de la connaissance de soi-mme, cest--dire de ses constituants et de sa pos-
sibilit daction dans ce monde. Le connas-toi toi-mme est la conditio sine qua non de
tout progrs thique.
75
Al-Masd, Kitb murj al-dhahab wa-madin al-jawhar, vol. IV, p. 64, 10 ff. Barbier
Meynard, cited by Michel Tardieu, biens coraniques et biens de arrn, Journal
asiatique 274/12 (1986), 144, p. 13. Elsewhere (Kitb al-tanbh wa-l-ishrf, p. 162, 35
De Goeje), Masd cites the Harrnian saying under the following form: He who
knows himself in truth becomes god.
76
Cf. R. Walzer, Porphyry and the Arabic Tradition, p. 281 f.; p. 294 ff.; F. W.
Zimmermann, al-Frb s commentary and short treatise on Aristotles De interpretatione (Oxford,
1991), p. xcii. I have discussed Frbs possible use of an otherwise unattested com-
mentary by Porphyry on the Posterior Analytics in M. Chase Did Porphyry write a
commentary on Aristotles Posterior Analytics? Albertus Magnus, Frb and Porphyry
on per se predication, in P. Adamson, ed., Classical Arabic Philosophy: Sources and Reception,
London/Turin: Warburg Institute/Nino Aragno (= Warburg Institute Colloquia 11),
2007, 2138.
77
Fit igitur haec essentia sibi intellecta in effectu; non fuit autem in nobis anima,
antequam haec intellegeretur, nisi intellecta in potentia; modo autem est intellecta in
effectu, eo quod intelligitur in effectu, et suum esse in seipsa est intellectu in effectu,
Frb, De intellectu et intellecto, 169173, p. 120, in tienne Gilson, Les sources grco-arabes
de lAugustinisme avicennisant (Paris, 1986), rst published in the Archives dHistoire doctrinale
et littraire du Moyen ge IV (192930) .
commentary 29
Thus, for the soul, or rather the souls intellect, to know itself is to
become, quite literally, identical with its essence,
78
and it can henceforth
intelligize all other separate intelligiblesthat is, those that have never
been in conjunction with matterin the same way as it knows its own
essence. This occurs at the third of Frbs four levels or kinds of
intellection, the intellectus adeptus (al-aql al-mustafd ).
Thus, for Frb, self-knowledge plays a crucial role both at the
beginning and at a fairly advanced stage of philosophical progress. At
the outset, the student must, with the help of an experienced professor,
look within himself to nd the rst intelligibles innate within him which,
once elaborated, claried and classied, will serve as the premises of the
syllogisms he will use as the starting-point of his logical deductions. At
a later stage, when through abstraction he has accumulated a sufcient
number of intelligibles, he will know his own intellect, and therefore
himself, thoroughly. This in turn is the precondition for being able to
know the intelligible Forms and separate intelligences which, unlike the
material forms incorporated in the sensible world, have never been in
conjunction with matter. The way is henceforth open for the perma-
nent conjunction with the Agent Intellect which, according to Frb,
constitutes felicity: that felicity which, for Frb as for Simplicius , is
the only goal and justication for doing philosophy.
78
Cf. E. Gilson , ibid., p. 31: cest en quelque sorte elle-mme, prise dans sa propre
actualit, quelle devient en sapprhendant.
AVICENNA THE COMMENTATOR
Allan Bck
A commentator should provide all the premises that are needed, and omit
nothing but the obvious and the superuous, for the most incompetent
commentator is he who uses in his commentary premises more cryptic
than, or as cryptic as, the premises of whatever he is commenting upon.
These commentaries which [purport to] bring us the truth conceal in
fact the theses better than the original texts, while what they conceal
most is errors.
1
Avicenna has never had high standing as a commentator on Aristotle.
In the scholarly world today, he, like any other Islamic medieval philoso-
pher, has the automatic curse of not working from the original Greek
in critical editions. He has the additional stigma of having received as
Aristotles work various spurious works, including Neoplatonist treatises
by Proclus and Plotinus like the Liber de causis.
2

Even in the medieval Islamic culture, his encyclopedic A-hif (the
Healing or Cure), where he did write on many of Aristotles works, was
not viewed as faithful commentary. Consequently, so the story goes,
Averroes was asked by Ab Yaqb to write a set of commentaries
more textually based.
3

Certainly, this assessment of Avicenna has some merit. He often
departs from the text of Aristotle even when commenting upon it.
He does so in various ways. (1) He states sometimes that Aristotle is
just wrong, for instance, in his doctrine that in thinking the thinker
becomes identical to the object thought.
4
Again, when commenting upon
1
Memoirs of a Disciple [of Avicenna ] from Rayy, 10, trans. Dmitri Gutas ,
Avicenna and the Aristotelian Tradition (Leiden, 1988), pp. 712.
2
For this standard story cf. Frederick Copleston , A History of Medieval Philosophy (New
York, 1972), pp. 1067; F. Van Steenberghen 1970, Aristote en occident (Louvain, 1946),
trans. as Aristotle in the West, trans. L. Johnston (Louvain, 1970), pp. 179; . Gilson ,
History of Christian Philosophy in the Middle Ages (New York, 1955), pp. 1812.
3
Dominque Urvoy , Ibn Rushd, in History of Islamic Philosophy, eds. S. Nasr & O.
Leaman (London, 1996), pp. 330345, p. 332. This story cannot be completely true,
as we now know that prior to Avicenna there were many faithful commentaries written
in Arabic on Aristotles works, especially at the bayt al-hikma. See n. 51.
4
On the Soul 420a1920; F Nafs, ed. G. Anawati (Cairo, 1962) 212,12213,8.
32 allan bck
the ontological square, the fourfold division of beings in Categories 2,
Avicenna rejects it and replaces it with a vefold division.
5
In such cases
Avicenna is offering his own views as substitutes. (2) Other times, he
omits discussing what Aristotle says. Thus, in his Al-Ilhiyyt (Metaphysica)
Avicenna ignores some books of Aristotles Metaphysics, like Book IV.
Again, Avicennas Physics hardly has the organization of Aristotles. (3)
Other times, he adds on a lot of material purportedly consistent with
Aristotles text, with the aim of defending or elaborating on it. Thus, in
Al-Ilhiyyt Avicenna adds on discussions about the necessary being and
prophecy. Likewise the organization of the Qys hardly follows that of
the Prior Analytics, although Avicenna does end up covering most of the
material there, while adding much more. (4) In some cases his attempts
at a literal commentary fail ludicrously, as in his discussion of the Poetics
where he attempts to describe Greek tragedies without having ever read
or seen one.
6
(I shall not be discussing this last type as it hardly gives
Avicenna a claim for being a pre-eminent commentator. For it consists
of standard, literal commentary, just done badly.)
The rst three types of cases differ signicantly. In the rst type
Avicenna goes where, he believes, the truth leads him at the expense
of what Aristotle has said. As a result, we have an explanation incon-
sistent with Aristotles doctrines. In the second and third types, we have
material being introduced or omitted so as presumably to increase
our understanding of the material being discussed by extending its
doctrines. Such additions and omissions can remain consistent with
Aristotles doctrines. As for the omissions, Avicenna generally does
discuss that material somewhere: he has just reorganized their presen-
tation. Thus he does discuss the material of Metaphysics IV in various
places: e.g., what is meant by genus, species, difference, etc. appears
in his commentary on Porphyry (the Logica of the Avicenna Latinus); in
his summaries as well he explains the meaning of many terms and
reorganizes Aristotles doctrines quite a lot.
7
We can then nd some
justication for a commentary deviating from the text in the second or
5
Al-Maqlt, eds. G. Anawati et al. (Cairo, 1959), pp. 18,420,3. See Allan Bck ,
Avicenna s Ontological Pentagon, The Journal of Neoplatonic Studies 7.2 (1999),
87109.
6
Avicenna , A-i r, ed. A. Badaw (Cairo, 1966). See my Review of Averroes Middle
Commentary on Aristotles Poetics, trans. Charles Butterworth, Canadian Philosophical Reviews
(1990), 92101.
7
I am thinking of An-Najt (Salvation), Al-Ihrt (Pointers), and the Persian Dnenme
(trans. Morewedge as The Metaphysica of Avicenna ).
avicenna the commentator 33
third way: the commentator is rearranging the material so as to make
it more comprehensible. Yet the rst way, rejecting what Aristotle says
outright, hardly looks like a commentary at all, but instead an inde-
pendent work. So too we might make this assessment for the so-called
Glosses of Abelard on Porphyry.
Nevertheless I wish to explore the option that Avicenna remains
a faithful commentator of Aristotle. For, if a commentary has the
function of helping us to understand the material and the issues being
discussed, and Avicenna presents materials that increase our under-
standing beyond what Aristotle has said, perhaps then he is writing
commentaries of high value. To be sure, a commentary also has the
function of helping us to understand the text being commented upon
and its authors intentions, context etc. in its own terms and on its own
standards. Avicenna does not focus on such tasks, and when he does, he
has no special claim to excellence (as in the example of the Poetics just
mentioned. Such literal commentary amounts to parsing or paraphras-
ing and giving historical background. Yet, on the other hand, if we
understand the issues being raised by a text in terms other than those
given in the text, we might then be able to appreciate that text more so
than if we had stuck to a literal explicatione de texte. We can understand
what, say, Aristotle was struggling with and what he was trying, in a
pioneering way, to get at. Pioneering efforts often have clumsy features,
as the path of terminology, theory, articulation of detail etc. has not
yet been laid down for those coming later to follow.
So I do not nd it obvious just what constitutes a good commentary.
I will start by discussing the attitudes of the person being commented
upon about commentaries. For Aristotle himself commented upon the
work of other philosophers, and wrote, if not commentaries in the
customary sense, at the least critiques like the Peri Iden. Let us consider
what The Philosopher does here.
Aristotle as Commentator
. . . our duty [is] for the sake of maintaining the truth even to destroy what
touches us closely, especially as we are philosophers; for, while both are
dear, piety requires us to honor truth above our friends.
8

8
Nicomachean Ethics 1096a147.
34 allan bck
In what sense would Aristotle himself understand commentator? Look
at how Aristotle himself commented upon his predecessorsnotably
Plato . Certainly in his Metaphysics Aristotle thinks that he is giving the
history of rst philosophy. He comments upon the theories of his
predecessors, mostly with the goal of showing how his own theory
incorporates all their insights without their shortcomings.
9

Earlier Aristotle scholars generally accepted Aristotle as a reliable
chronicler and commentator. For instance, W. D. Ross generally takes
Aristotle to give an accurate history of philosophy.
10
Taylor , Burnet , and
Guthrie may have had occasional doubts but generally concurred.
11

However it has now become a commonplace among scholars of
Aristotles predecessors to assert that Aristotle misunderstands their
views badly. For example, Kirk , Raven and Schoeld claim that his
judgments are often distorted by his view of earlier philosophy as a
stumbling progress to the truth that Aristotle revealed.
12
Thus they
suggest that Aristotle is mistaken about the signicance of Zeno s
arrow paradox.
13
If you look at random for other recent discussions
of Aristotle in the treatment of the Presocratics, you will routinely nd
many such remarks.
Now if this current assessment of Aristotles scholarly abilities holds,
Aristotle fails by his own standards. For he relies heavily on an endoxic
method. We begin our scientic study, he says, from looking at phenom-
ena and endoxa.
14
As with astronomy, all the sciences must preserve the
phenomena.
15
For him the phenomena are the things as they appear
to us, at times contrasted with how they are in themselves. The endoxa
are the opinions that are generally or mostly accepted by anyone and,
even more so, the opinions of the experts.
16

9
Metaphysics 988a1823.
10
Metaphysics, vol. I, pp. xxxvli,
11
A. E. Taylor , Varia Socratica (Oxford, 1911), pp. 847; J. Burnet , Early Greek
Philosophy, 3rd ed. (London, 1920), vol. I, pp. 2307; W. K. C. Guthrie , A History of
Greek Philosophy, vol. 1 (Cambridge, 1962).
12
The Presocratic Philosophers, 2nd ed. (Cambridge, 1983) p. 3.
13
The Presocratic Philosophers, p. 273.
14
De Caelo 306a517; Analytica Priora 46a 1722. Aristotle also seems at times to
identify this mathematical method with a dialectical method. Cf. Physica 204b4, and
Allan Bck , Aristotles Discovery of First Principles, in From Puzzles to Principles, ed.
May Sim (Lanham, Maryland, 1999), 163182.
15
William Wians , Saving Aristotle from Nussbaum s Phainomena, in Essays in Ancient
Greek Philosophy V, eds. A Preus & J. Anton (Albany, 1992), p. 135.
16
Topics 100b213. See J. D. G. Evans , Aristotles Concept of Dialectic (Cambridge, 1977),
pp. 778 on difculties of translating endoxon. I shall just use endoxon.
Even in the sciences Aristotle uses endoxa. As Owen made well known,
Aristotle complicates his conception of phenomena considerably.
17
In
his ethics he begins his discussion of acrasia (incontinence) by rehears-
ing the phenomena. Yet he does not make observations of human
behavior. Rather, he identies these phenomena with the endoxa, the
reputable opinions. [1145b26] On account of this, Owen claims that
Aristotle uses phenomena ambiguously: on the one hand to designate
the empirical, given by sense perception, and on the other to designate
the dialectical, given by widely accepted or reputable opinion. Indeed,
in some texts Aristotle opposes phenomena and endoxa, and in others he
assimilates them, even in his treatises on natural science.
18
At times the
contrast between phenomena and endoxa seems to disappear. Aristotle
himself identies the phenomena with the endoxa in his ethical investi-
gation of acrasia. Again, in his study of the rst principles common to
all sciences in his science of being qua being, Aristotle examines endoxa
dialectically by way of establishing them as rst principles. Also, in
justifying particular concepts in his Physics, Aristotle again appeals to
endoxa, e.g., for his concept of place. [211a411] Aristotle blurs their
difference even more when he distinguishes endoxa from apparent
endoxa ( ).
19

Nussbaum offers a way of thinking of Aristotles conceptions of
phenomena and endoxa consistently.
20
If we understand Aristotle to hold
Quine s view, that even sense perceptions are a theory-laden part of
the fabric of our web of belief, then the empirical and the dialectical
both concern opinion. We have no hard facts, just our beliefs about
the world. Some beliefs may have a stronger tie to the stimulus mean-
ing of sensation than others, and so are more empirical. Still scientic
knowledge, like all other conversation, falls within the hermeneutical
circle of our society and its ideology.
In any event, Aristotle does not equate endoxa in the dialectical
inquiry of philosophy and science with majority opinion. He weights
expert opinion more.
21
Insofar as common people themselves defer to
the opinions of experts, we might say that deferring to experts agrees
17
G. E. L. Owen , Tithenai ta Phainomena, in Logic, Science, and Dialectic, ed. Martha
Nussbaum (Ithaca, 1986), p. 240.
18
E.g., Generation of Animals 725b56; 729b910; 760b2733; Parts of Animals
648a20ff.
19
Topics 100b25.
20
Martha Nussbaum , The Fragility of Goodness (Cambridge, 1986), pp. 244; 2745.
21
C. D. C. Reeve , Practices of Reason (Oxford, 1992), pp. 356.
36 allan bck
with common opinion, especially when the subject concerns something
about which common people have no rm opinion (e.g., the sex life of
nematodes), but when the expert and the common opinions diverge,
Aristotle weights expert opinion more. Yet this weighting narrows the
gulf between the phenomena which are true and the endoxa, which seem
to be true. Too, phenomena themselves are literally appearances, how
the world appears to us. As we judge, categorize, and reect upon our
sensations, the phenomena may well be thought to end up amounting
to give what seems to us to be true, with special reliance on those who
have observed carefully and repeatedly.
In our terms, Aristotle recommends doing a survey of the litera-
ture and critique of past theories if available before proceeding to do
scientic research via direct observation and experiment and theory
construction. Now, if Aristotle so radically misunderstands his predeces-
sors, he is failing in this dialectical task and in his science.
We might use caution in accepting the opinions of current scholars
on Aristotle as commentator, especially when their work focuses on
Aristotles predecessors. One always has the suspicion of people cheer-
leading for the philosopher on whom they have devoted so much time.
Moreover, it is curious, if Aristotle fails so badly in understanding his
predecessors, why those who followed and critiqued him did not seize
on the point more than they do? For many of the Greek commentators
were Neoplatonists . They may modify Aristotles doctrines, but gener-
ally accept his views of his predecessors.
22
Even Plotinus proceeds in
critiquing Aristotle not so much by showing how he misunderstands
Plato but by defending Platos views and picking holes in Aristotles
objections and own positions.
In any case, I am focusing here not on Aristotles actual prowess as a
commentator but rather on the standards he has for good commentary,
regardless of whether he meets his own standards or not. So, putting
caveats on his excellence as a commentator aside, let us concentrate on
what Aristotle offers, or at any rate uses, as criteria for a commentary.
For him a commentary is both a report and critique on certain endoxa,
the opinions of the experts, preferably their opinions as recorded in
written form. We can see Aristotle himself writing such commentaries.
Aristotle went to great lengths to collect descriptions of phenomena,
customs etc. We need only think of his collection of constitutions with
22
The commentaries of Simplicius and Alexander (pseudo and real) for instance.
accompanying analyses, his collections of observations of natural phe-
nomena (including ones that he did not make himself, e.g., on elephants
and the gymnosophists),
23
and his reviews of past theories whenever
starting a scientic inquiry. These reports then should be accurate.
Note, however, that Aristotle, like the historians of his time, tended
not to insist on the historical accuracy of exact quotation, description
of social and linguistic context, archival research etc. required by the
historians of our times. Still, I suppose his accounts of his predecessors
have about as much accuracy as the speeches that Thucydides puts into
the mouths of Spartans and Persians.
Unlike modern historians and commentators, Aristotle insists also
upon critiquing the views being commented upon. For he intends to use
this material to seek the truth on the subject matter, and not ultimately
the truth about who said what. We see a similar situation in his ethics.
He wants to have a theory of what is the highest good but not merely
for the sake of theory: we are inquiring not in order to know what
excellence is, but in order to become good, since otherwise our inquiry
world have been of no use.
24
To be sure, even modern historians
organize their remarks and use the available primary historical data
selectively, according to an ideology or according to a certain method,
making many background foundational assumptions.
25
Yet they tend
to avoid making assessment of value. In contrast, as Aristotle writes
commentaries in order to use their results, his remarks have a certain
focus largely absent in our modern commentaries. We separate theory
and use, pure and applied science, scholarship and philosophizing,
more sharply than he does.
Avicenna as Commentator
How then would Avicenna fare as a commentator in the Aristotelian
tradition?
Let me rst dispose of the standard complaints against Islamic com-
mentators on Aristotle that I have mentioned at the beginning. To be
sure they were Greek-less, although they had some contact with people
23
E.g., the discussion of elephants in History of Animals I.10 and of the naked soph-
ists in Fr. 35if that is not genuine, just use another of Aristotles remarks on India.
24
Nicomachean Ethics 1103b269.
25
Cf. Marc Bloch , The Historians Craft, trans. P. Putnam (New York, 1953).
38 allan bck
uent in Greek. However, I nd the Arabic translations of unayn Isq
quite accurate for most technical points in Aristotles texts. Moreover,
the Islamic philosophers themselves were sensitive about problems of
translation. Even before Avicenna , Islamic philosophers like al-Frb
had long, sophisticated discussions about the various grammatical
structures in Arabic, Greek, Persian, Syriac, Soghdianas far as their
linguistic resources permitted.
26

Moreover, Avicenna is no nave reader, misled by having received
Neoplatonist works by Proclus and Plotinus as written by Aristotle. He
himself expresses doubts about the authenticity of the so-called Theology
of Aristotle, known now to be spurious.
27
I give further examples below.
So, I submit, let us not dismiss Avicenna s commentary on a priori
grounds; instead, let us look at the contents of his works.
Avicenna claims to have written commentaries in his youth explicat-
ing Aristotles thought.
28
These seem to have followed Aristotles texts
fairly closely, giving explications of his views. These early, fairly standard
commentaries were apparently lost. Still we can perhaps get some sense
of what they were like, or even what they actually were, by looking
at some parts of his extant works. For, as Gutas has shown, Avicenna
tended to recycle parts of earlier works into his later works. Thus some
parts of the De Anima material of A-hif seem to have been written
before.
29
Again, the Poetics portion of A-hif entioned above may have
been recycled, for it is a fairly literal commentary that seems to proceed
by using a translation of the Poetics along with its attendant marginalia
so as to produce an explication as best as can be without having any
independent knowledge of Greek theater.
However, Avicenna wrote another sort of commentary later in his
career. These followed the text much less closely and paraphrased far
less. For in Hamman around 1016, Avicennas students had asked
for him to replace these lost early commentaries.
The hope of ever obtaining his lost works having dimmed, we asked him
to write them and he said, I have neither the time nor the inclination
to occupy myself with close textual analysis and commentary. But if you
26
See my Islamic Logic, forthcoming.
27
Letter to Kiy, section 3, in A. Badaw , Aris inda l-Arab (Cairo, 1947), pp.
120,9122,8; trans. Gutas , Avicenna and the Aristotelian Tradition, p. 64.
28
Namely, Al-sil wa-l-Mal (The Available and the Valid ) probably in 1002 or 1003,
following Gutas dating.
29
Gutas , Avicenna and the Aristotelian Tradition, pp. 1045.
would be content with whatever I have readily in mind [which I have
thought] on my own, then I could write for you a comprehensive work
arranged in the order which will occur to me. We readily offered our
consent to this and urged that he start with physics.
30
Avicenna states clearly that in these commentariescomments if you
willhe will not be explicating Aristotles thought and so will not be
writing the usual sort of commentary. Rather, he shall be giving his
own thoughts and theories on the topics and positions brought up by
Aristotle:
. . . if you would like me to compose a book in which I will set forth what,
in my opinion, is sound in these philosophical (sciences), without debat-
ing with those who disagree or occupying myself with their refutation,
then I will do that.
31
Accordingly, in his A-hif, some parts of which constitute the Avicenna
Latinus, Avicenna set out to comment upon a great portion of
Aristotles works, including the whole of the logic, much of the works
on the natural sciences, and the Metaphysics. He completed this mas-
sive undertaking in but a few years, from 101627, if we are to believe
the historical testimonyalthough it is likely that he used some earlier
writings as some parts of A-hif.
In A-hif Avicenna has respect but not reverence for Aristotle. On
some topics, Avicenna thinks that Aristotle has the whole truth. For
instance, concerning Aristotles classication of the fallacies, Avicenna
says:
. . . after almost one thousand three hundred and thirty years, was this goal
reached by anyone who blamed Aristotle for being decient, and who was
right in identifying a defect in him because Aristotle was in fact decient
in such and such a matter? And did there appear after him anybody who
added anything at all to this art [sophistics] beyond what Aristotle said?
Certainly not; for what Aristotle did is complete and perfect.
32

In such parts of A-hif as the Sophistics then it is not surprising that
Avicenna , perhaps using or borrowing from his earlier literal com-
mentaries, if he had them available, offers something resembling a
30
Introduction to A-hif, Section 3, translated by Gutas , Avicenna and the Aristotelian
Tradition, p. 101.
31
The Life of Ibn Sina, ed. and trans. W. Gohlman (Albany, 1974), 54.15.
32
As-Safsaa, ed. A. Ehwany (Cairo, 1958), 114 7; trans. Gutas , Avicenna and the
Aristotelian Tradition, p. 37.
40 allan bck
usual commentary. Yet, even in such cases where Avicenna thinks that
Aristotle has the right theory, he often proceeds to a critical discussion
of it. Here we may agree with Gutas :
Avicenna presents himself in the prologue to the Cure not as an anti-
Aristotelian despite himself, as al-Jzjn would have it, but as a conscious
reformer of the Aristotelian tradition, an attitude which is also apparent
in the Introduction to The Easterners and shared by other disciples of
his . . .
33

However, in other parts of A-hif Avicenna has a much harsher
attitude about what Aristotle says. First, in general, Avicenna holds
Aristotles logic to be defective, at least in the sense of needing to be
supplemented.
34
Unlike Kant, Avicenna did not think that Aristotle
had completed the whole of formal logic. For instance, following the
Stoics, Avicenna devoted a lot of effort to working out the syllogistic
for various sorts of hypothetical statements in his Qys. Second, and
more particularly, on the Categories, Avicenna follows the lead of the
Greek commentators who doubt the authenticity of some passages of
the Categories and who stress its use as a work for the beginner.
35
He too
doubts the authenticity of the Categories as a whole.
36
Yet, more than
his predecessors, Avicenna stresses that the Categories is not so much
for the philosopher as for the common people: not too technical nor
of much use.
37

Even here, though, Avicenna is not rejecting Aristotles views so much
as reorganizing and extending themperhaps, as I shall suggest below,
following the lead of al-Frb . For instance, Avicenna imports much
of the material on substance found in Metaphysics VII into his discus-
sion of substance in Categories 5. Indeed, Avicenna discusses some of
Aristotles doctrines in the Metaphysics, like focal meaning and the rela-
tion of substance and form, more in this commentary on the Categories
than in his commentary on the Metaphysics, Al-Ilhiyyt. Avicenna has
the general habit of reorganizing the sequence of materials found in
Aristotle, and Al-Maqlt is no exception.
33
Gutas , Avicenna and the Aristotelian Tradition, p. 111.
34
Ibid., 189.
35
E.g., Porphyry , in Cat. 134, 289; Ammonius , in Cat. 13, 208.
36
Al-Maqlt 8,115. So did earlier Islamic commentators, and to some extent Greek
ones like Simplicius , in Cat. 18, 79. Cf. I. Madkour , LOrganon dAristote dans le monde
arabe (Paris, 1934), pp. 789.
37
Al-Maqlt. 35,1011; 35,2036,2; 189,89; 264,7; 265,17.
Avicenna viewed his audience to consist solely in the elite philoso-
phers. Rather like Confucius , who wanted only students who could
bring back a square when given only a corner of it [Analects 7:8],
Avicenna would expect his audience to be able to reconstruct his posi-
tion from a few remarks. For him, the common people should not read
philosophy. Like Plato , Avicenna wanted only the philosophers, and the
worthy ones at that, to read his work. Avicenna believed Aristotle to
hold the same view, due to a letter ascribed to Aristotle and written to
Alexander . According to it, Aristotle was deliberately obscure in order
to ward off the common people.
38
Likewise, al-Frb , whom Avicenna
admired greatly, says:
Our style used an obscure way of expression for three reasons: First, to
test the nature of the student in order to nd out whether he is suitable
to be educated or not; second, to avoid lavishing philosophy on all people
but only on those who were worthy of it; and third, to train the mind
through the exertion of research.
39
Religious traditions in Islam too had the custom of withholding knowl-
edge from the hoi polloi and reserving it for the select few:
I have forbidden all my friends who would read [this treatise] to squander
it on an evil or obdurate person, or show it to them, or to deposit it where
it does not belong, and bound them [by an oath] . . .
40

Like Maimonides later, Avicenna writes only for those who deserve
to read. We know that he withheld his writings from many of his
contemporaries.
In short, Avicenna deliberately writes commentaries for the worthy
few, the elite among the philosophers. Such an audience requires com-
mentaries not of the usual sort. Such a commentary need not intro-
duce and explicate the text to be commented upon. Rather, it needs
to illuminate that text, to give a perspective whereby its reader will be
helped in assessing the merits of that text. I have suggested that we
view Aristotle as a commentator in the same way.
38
Gellius , Noctes Atticae 20,5,112, in Aristoteles, Privatorum Scriptorum Fragmenta, ed. M.
Plezia (Leipsig, 1977), 28. Cf. Simplicius , in Cat. 7, 69.
39
Mabdi, Al-Falasafah Al-Qadma (Cairo, 1910), p. 14, trans. Gutas , Avicenna and the
Aristotelian Tradition, p. 227.
40
Avicenna , Al-Mabda wa-l-Mad [The Provenance and Destination], ed. A. Nrr
(Tehran, 1984), Ch. 16.3; trans. in Gutas , Avicenna and the Aristotelian Tradition, p. 33.
42 allan bck
Yet, even if we accept such criteria for a commentary, surely the
text of Avicenna is so difcult and even convoluted so as to rule it out
as a good commentary. Let me then comment upon Avicennas style
and on his writings.
41
Avicenna deliberately takes on an oracular style. We can see this just
from the titles of his works: The Cure, The Salvation, etc. He does so for
various reasons: 1) on his own view, he has achieved an enlightenment
stemming from the activation of his active intellect and its permanent,
actual connection to the intelligibles;
42
2) what he is doing is better than
the popular, vulgar prophecy of religion anyway;
43
3) he is an elitist.
Avicenna has a style a bit like the Greek commentators: a rather
cryptic, oracular, enigmatic utterance followed by some more mundane
explanation. Alexander , Ammonius and Themistius have similar styles.
Or, perhaps better, we may compare the writing style of Avicenna to that
of Plotinus . In both cases too we might wonder how well the text was
corrected and proofed. Likewise, in both cases it is hard to distinguish
when the author is presenting a position given by somebody else and
its implications from when he is giving his own position; to determine
when a question ends and its answer begins.
Moreover, Avicennas cavalier attitude towards his own writings does
not help the quality of the text that we have. He generally would write
extremely quickly. His own account has him writing fty pages per day
of the metaphysics and physics of A-hif.
44
After writing something,
he would give the copy to whom it was promised, or put it away for
showing to the worthy few. Often, due to his frequent moves and the
religious and political turmoil, his writings were lost or damaged.
45

41
Discussed and supported more in the Introduction to my translation of Al-
Maqlt.
42
F Nafs 212,49. Cf. Gutas , Avicenna and the Aristotelian Tradition, pp. 161 n. 30; 175.
43
Like al-Frb , Avicenna held that religious customs for the common people
differ from those for the elite philosophers. For instance, for Avicenna prayer had as
its goal the nding of middle terms; drinking wine is forbidden to the people but
good for those of powerful brains. Cf. The Life of Ibn Sina, 28,25; 54,79; Al-Qnn
I.169.269. For a spirited discussion of this issue, cf. Gutas , Avicenna and the Aristotelian
Tradition, pp. 1818.
44
The Life of Ibn Sina, 58,28.
45
Al-Jzjn , Introduction to A-hif, 2, trans. Gutas , Avicenna and the Aristotelian
Tradition, p. 40: I have heard, however, that these were widely dispersed in that people
who owned a copy of them withheld them [from others]; as for him, it was not his
habit to save a copy for himself, just as it was not his habit to make a copy from his
archetype or transcribe [an archetype] from his rough draft: he would only either
dictate or himself write the manuscript and give it to the person who had commis-
Consequently, we have the situation that Avicenna probably proofread
little, and what copies there were were made haphazardly under hasty
circumstances. Aside from the contingencies of the political turmoil,
we can see here perhaps echoes of Plato s attitude toward written
philosophy as expressed in his Seventh Letter. Written philosophy is
dead philosophy, relics to be discarded as trinkets for those allowed to
be souvenir hunters.
Again, Avicenna was a Persian, and not a native Arabic speaker or
writer. At any rate, let me say that he is quite careless about the ante-
cedents for his pronouns! I should also remark that at points we have
some possible wordplay.
46
Apart from these reasonsthe haste of composition, the state of the
manuscript, these elitist tendencies, possible language problemsthere
are other reasons why reading Avicenna is difcult. To be sure, Avicenna
can write clearly. Yet often he writes quite obscurely, regardless of the
language in which the text is read. Perhaps this obscurity comes in part
from Avicennas trying to say something new, for which there would
not naturally lie ready to hand extant phrasing. We can see similar
obscurities in many original works: Abelard or Aristotle himselfis
a good example.
Above all, in reading works like A-hif, we have the problem of
context. Avicenna is reacting not only to Aristotles text but also to the
other writings on itcommentaries, notes, marginalia, some of which
surrounded the Arabic translation that he was using. Avicenna was
probably using some revised version of unayn Isq s translation,
the standard Arabic one of the time, with lots of marginalia.
47
Avicenna
himself says about his studies in his youth that resulted in the more
sioned it from him. Moreover, he suffered from successive misfortunes, and disasters
destroyed his books.
46
E.g., at Al-Maqlt 248,178.
47
F. E. Peters , Aristotle and the Arabs (New York, 1968), pp. 5963; 160. It seems clear
that, at least in some cases, Avicenna was not following Isqs translation (edited by
Badawi, the eleventh century one with marginalia at the Bibliothque Nationale of
Paris) closely. Cf. Aristotles Categories 1a245, discussed in Al-Maqlt 28,4ff with the
current Oxford translation: By in a subject I mean what is in something, not as a
part, and cannot exist separately from what it is in. If Avicenna is quoting and not
paraphrasing loosely, he is not using Isqs translation. Again, cf. Al-Maqlt 28,45:
the terminology for the category of having is not the same as Isqs either. Likewise,
at 57,19, the translation for the Greek boxer: Isqs translation has boxer; Avicenna
has wrestler. However, Porphyry , in Cat. 135, 911, has the example of the wrestler as
well as that of the boxer, and it is true that wrestling was and is far more prominent
in Islamic cultures than boxing.
44 allan bck
textual commentaries, now lost: Then I began to read the books [of
the Organon] by myself and consult the commentaries until I had mas-
tered logic.
48
So he had read some commentaries, more than marginal
notes. At this point, we can make only educated guesses about what
commentaries Avicenna used. For Avicenna hardly ever cites any oth-
ers by name. Morevoer, he need not have read those whom he does
mention. Likewise, when al-Frb names Archytas [of Tarentum],
he probably picked this name up from Simplicius .
49
As for Al-Maqlt,
the Greek commentaries on the Categories by Porphyry, Simplicius,
Ammonius, Philoponus, and likewise Plotinus (especially Enneads VI ),
insofar as they were translated into Arabic, are plausible candidates for
Avicennas sources, at least indirectly.
50
By the time of Avicenna, there
were also very many Islamic commentators and glossers, most of which
have not been studied carefully yet.
51
By his own testimony, Avicenna
considered the commentaries of al-Frb the most important of these.
52

48
Avicenna , Autobiography 5.
49
al-Frb, Greater Commentary on De interpretatione, eds. W. Kutsch and S. Morrow
(Beirut, 1960), 157, 1920; trans. F. W. Zimmermann (London, 1981), p. 152. Cf.
Simplicius , in Cat. 86,2830; 206,20; 408,112; 409,15. Archytas was a Pythagorean
contemporary of Plato; the commentary on the Categories is presently thought to be a
rst-century (A.D.) forgery. Cf. T. A. Szlezk , Pseudo-Archytas ber die Kategorien (Berlin,
1972).
50
Gerhard Endress , Die wissenschaftliche Litteratur, in Grundrisse der Arabischen
Philologie, ed. H. Gtje , vol. 2 (Wiesbaden, 1987), pp. 41631; Julius Weinberg , Abstraction,
Relation and Induction (Madison, 1965), p. 91. On Greek texts available to Avicenna
and not preserved today, see A. Badaw , La transmission de la philosophie grecque au monde
arabe (Paris, 1968); New Philosophical Texts Lost in Greek and Preserved in Islamic
Translations, in Islamic Philosophical Theology, ed. P. Morewedge (Albany, 1979), pp.
313; H.-J. Ruland , Die arabischen Fassungen zweier Schriften des Alexander von Aphrodisias:
ber die Vorsehung und ber das liberum arbitrium, diss. Saarbrcken 1976; Zwei arabischen
Fassungen der Abhandlung des Alexander von Aphrodisias ber die universalia, Nachr.
der Akademie der Wiss. In Gttingen, 1979 no. 10. Again, Porphyry s lost commentary
Ad Gedalium, was probably available at the time of Avicenna. Cf. Michael Chase , ed.
& trans., Simplicius : On Aristotles Categories 14 (Ithaca, 2003), p. 96.
51
The number of these works is staggering. On those on the Categories, cf. F. E.
Peters , Aristoteles Arabus (Leiden, 1968), pp. 711; H. Daiber , Review of Peters, Aristoteles
Arabus, Gnomon 42 (1970), p. 542; Gerhard Endress , Die wissenschaftliche Litteratur,
in Grundrisse der Arabischen Philologie, vol. 2 (Wiesbaden, 1987), pp. 40234; Gutas , Greek
Thought, Arabic Culture, p. 151. For secondary literature and more updates see Grundrisse
der Arabischen Philologie, ed. H. Gtje , vol. 2, pp. 481502.
52
Letter to Kiy, in Arist inda l-Arab, ed. Badaw (Cairo, 1947), 120,9122,8,
3, trans. Gutas , Avicenna and the Aristotelian Tradition, p. 64: At the present moment it
is impossible for me [to rewrite it]: I do not have the free time for it, but am occupied
with men like Alexander [of Aphrodisias], Themistius , John Philoponus , and their
likes. As for al-Frb , he ought to be very highly thought of, and not to be weighed
in the same skill with the rest: he is all but the most excellent of our predecessors.
Re al-Frb on the Categories, we have chiey (extant today) al-Frbs epitome on
In sum, on current estimates, the commentaries that Avicenna used
most are those by Simplicius and al-Frb.
53
Given Avicennas choice
of issues, I suggest adding Porphyry to the list; Avicenna would have
known of Porphyrys views on homonymy at least indirectly via their
presentation in Simplicius.
54
Consequently, the text of A-hif is not self-contained. Avicenna is
often replying to arguments and doctrines that he does not state fully.
Many of these arguments can be found in the Greek commentaries
or in later Islamic ones. However, we do not yet have many accessible
editions or translations of the commentaries written in Arabic, even
by such as those by al-Frb .
Moreover, Avicenna uses a technical vocabulary, inherited from
the Greek traditions and his Islamic predecessors. This appears most
clearly at rst glance in his use of various prepositions which seems to
out or at least stretch ordinary Arabic usage. Aristotle himself in the
Categories and Simplicius even more in his commentary on it did the
same with Greek.
55
Avicenna continues this tradition by making up
terms or transforming the meaning of existing terms in order to express
his own theory.
56
For Avicenna also is engaged in constructing his own
the Categories (Al-Maniq inda l-Frab, ed. R. Al-am, vol. 3 (Beirut, 1986); trans.
as al- Frabs Paraphrase of The Categories of Aristotle, Islamic Quarterly (1957),
15897, and more original discussions in the Kitb al-urf, ed. M. Mahdi (Beirut, 1970)
and Kitb al-Alfz al-Mustamala f l-mantiq, ed. M. Mahdi (Beirut, 1968). See Gerhard
Endress, Die wissenschaftliche Litteratur, in Grundrisse der Arabischen Philologie, vol. 3
(Wiesbaden, 1992), p. 53, nn. 1857.
53
Ilsetraut Hadot, La vie et oeuvre de Simplicius daprs des sources grecques et
arabes, in Simplicius: sa vie, son oeuvre, sa survie, ed. I. Hadot (Paris, 1987), p. 36; Gutas,
Avicenna and the Aristotelian Tradition, p. 150; Michael Chase , ed. & trans., Simplicius: On
Aristotles Categories 14 (Ithaca, 2003), pp. 23. In regard to the Al-Maqlt proper, it is
suggestive that, like Ammonius but unlike Simplicius, Avicenna does not discuss chapter
15 on having, although Avicenna might have omitted this on his own initiative. Also,
like Philoponus, in Cat., 126,9, Avicenna discusses the Resurrection example, and so
maybe Avicenna is following him at 153,15ff.
54
Concetta Luna , Commentaire, pp. 65; 82. In addition to the extant commentary
in Cat. in the Commentaria in Aristotelem Graeca and translated by S. Strange (Ithaca, 1992),
there is a lost commentary Ad Gedalium, probably available at the time of Avicenna.
Cf. M. Chase , Simplicius , p. 96.
55
Cf. Richard Gaskin , trans. & comm., Simplicius , On Aristotles Categories 915 (Ithaca,
2000), n. 628.
56
Cf. Gutas , Avicenna and the Aristotelian Tradition, p. 10. Afnan , Philosophical Terminology
in Arabic and Persian, p. 33, says, wrongly, A signicant feature which was not of the
language but which cramped philosophical vocabulary in general and fossilized it almost
permanently, was the lack of initiative on the part of the Falsifah to coin special terms
of their own. But then he is following Atkinsons The Greek Language, which says the
same about Aristotle and Greek!
46 allan bck
technical vocabulary for Greek technical expressions in Arabic. He uses
some already made up, say by al-Frb . Still, he seems to be making
up more himself, especially as his theory differs from earlier ones.
57
In sum, Avicenna is not writing a commentary of the usual sort. Like
his later works, we can say of those in the hif that they contain many
Doubts about Aristotlealthough, to be sure, Avicenna also accepts
much of what Aristotle and earlier Aristotelians said.
58
He is writing a
commentary in the sense of following the order of Aristotles texts and
commenting on what is being discussed. He often does not bother to
explicate Aristotles text. Rather, he presents what he thinks on these
topics. Although he does often agree with Aristotle, often he does not:
for instance, in Al-Maqlt he disagrees with Aristotle in the Categories
on homonymy, synonymy, and paronymy, on the ontological square,
and on the number of categories. Still, many of the doctrines used by
Avicenna to correct the doctrine of the categories can indeed be found
in Aristotles works elsewhere. I have given some examples above.
We can nd analogous commentaries in Abelard . His Glosses on
Porphyry are hardly glosses; his commentaries on Boethius stray so far
from the texts that we look in vain in Boethius for most of the doctrines
Abelard puts forward. Like Abelard, Avicenna uses the text as a source
of questions, topics, and problems that he then investigates and for
which he provides the answers. Avicenna does indeed seem to have a
style much like Abelards: always looking for alternatives, contemptu-
ously dismissing views that he nds silly (here more with the verve of
Roger Bacon ). Avicenna differs from Abelard perhaps in having his
own, nal denite position on most issues.
The Case of Categories 1
What Avicenna does in his discussion of homonymy gives a good
instance of his approach. In effect, he takes the materials from the Greek
period and offers a new theory, perhaps with a wider compass than what
57
Cf. Shukri Abed , Language, in History of Islamic Philosophy, eds. S. Nasr & O.
Leaman (London, 1996), pp. 90413.
58
Gutas , Greek Thought, Arabic Culture, p. 153. Cf. Dag Hasse , Avicenna s De Anima
in the Latin West (London, 2000), p. vii: different stages in a continuous process of
reworking his position with the Peripatetic tradition and eventually emancipating
himself from it.
we nd in the (extant) Greek commentaries. For the Greek commenta-
tors tend to relegate parts of their discussion to different passages: the
materials about rhetorical, dialectical, and fallacious homonymies do
not appear much in the Categories commentary. In contrast, Avicenna
incorporates their doctrines into a general theory of homonymy. He
has a continuum, ranging from the merely ambiguous to the completely
synonymous.
Just what did Avicenna get from al-Frbs discussions of the
Categories? If we look at his extant works, we do not nd much origi-
nal doctrine. Moreover, Avicenna does not follow al-Frb, e.g., in
discussing the category of having.
59
Perhaps what he found important
in al-Frbs work was his method. For al-Frb does not follow the
order of the text of Aristotle, even in his so-called Paraphrase. Instead,
he gathers Aristotles doctrines and writes a treatise with a clear, original
organization.
60
So perhaps Avicenna got from al-Frb a new approach.
This would t with his remark that he did not understand Aristotles
Metaphysics until he had read al-Frbs work on it: that is, he did not
understand the overall plan and structure of the Metaphysics.
61
Along these lines, Avicenna begins his commentary on the text of
the Categories not with homonymy as Aristotle and all the Greek com-
mentators did, but with synonymypresumably because synonymy is
for science while homonymy is for sophistry. It is worth noting that in
his later works like Al-Najt and the Ishrt Avicenna does not discuss
homonymy at all, although he does discuss some other topics discussed
in the Categories.
Moreover, al-Frb, like the Arabic translation of Isq [e.g., at Cat. 1
line 5], speaks of expressions in terms of having the same or different
senses.
62
He also will speak of the essences themselves in terms of their
being the same or different in their senses.
63
Such statements may well
have suggested to Avicenna a general structure for homonymy based on
senses. At any rate, as we shall now see, he does have such a scheme.
59
al-Frb, in his Paraphrase of the Categories of Aristotle ed. & trans. D. M. Dunlop,
Islamic Quarterly 5.1 (1959), 24,715 [trans. p. 40 36], comments on the category of
having and even distinguishes two types: a natural having, as a tree has its bark, and
a voluntary having, as a man has his clothes.
60
Likewise in Kitb al-urf, where he presents some of the doctrine of the Categories,
al-Frb uses his own organization of the materials.
61
Autobiography 9. So too Gutas, Avicenna and the Aristotelian Tradition, p. 239.
62
Kitb al-urf 81 110,201.
63
In Kitb al-urf 89 117, 810.
48 allan bck
In order to understand Avicennas text and to see just how he differs
from the earlier commentators, I shall rst discuss their doctrines.
The Greek commentators on Categories 1 all presented the same basic
interpretation of homonymy with some variations. From the view-
point of Aristotles text, they provide too much commentary. For they
develop, or augment, the few lines devoted to homonyms into a full-
blown theory of homonymy incorporating many of Aristotles doctrines
found elsewhere. By comparison, their commentaries on synonymy and
paronymy are much shorter.
The Greek commentators worry a lot over the subject to be discussed
in the Categories. They have various answers, ranging from beings qua
beings, to thought, to predicables, to predications, to signicative expres-
sions, to expressions qua expressions. [Simplicius , in Cat.9,4ff.; Porphyry,
in Cat. 59,1033] Simplicius reports that Alexander of Aphrodisias
proposed that it deals with thoughts. [10,119; 9,3110,2] In contrast
Porphyry, in both his extant commentary and in his lost commentary
to Gedalius, says that it is about predicates, sc., about expressions
signifying things. [57,6; 58,610 & 1820; Simplicius, in Cat. 10,203]
Simplicius ends up concluding that the Categories deals with signica-
tive expressions, but claims that this amounts to dealing with thoughts
insofar as they signify. [10,45; 11,12; 12,14; 13,115; 21,79] At one
point he likens these thoughts to the Stoic concepts, presumably the
lekta. [10,24] As Aristotle holds at the beginning of On Interpretation
that thoughts constitute a mental language that the spoken language
directly, and the written language indirectly, signify, we can see why he
would think these two positions equivalent. Aristotle seems to speak of
words and propositions signifying denitions and meanings on this
basis. [ E.g., An. Po. 93b2935]
64

They all then proceed to divide up these items, the signicative
thoughts, lets call them, into those that signify homonyms and those
that signify synonyms.
65
They nd this division exhaustive for the mental language. They add
polynyms and heteronyms to the homonyms, synonyms, and paronyms
64
David Charles , Aristotle on Names and their Signication, in Language, ed. S. Everson
(Cambridge, 1994), p. 43.
65
In addition to those discussed below, there are also classications by Boethius,
in Cat. 166B-C, who follows Simplicius , and variant ones by Olympiodorus, in Cat.
34,335; David Elias, in Cat. 139,29140,25; Sophonias , in Cat. 1,242,17. All these
seem to have had no inuence on Avicenna.
discussed explicitly by Aristotle to account for features peculiar to the
spoken or written language. Polynyms have the same account but
different names, like two mantles with respect to cloak and cape.
Heteronyms have neither the name nor the account in common, as a
crow is named crow while a dog is named dog. Simplicius dis-
tinguishes such cases, which he calls merely other names, from
heteronyms proper, which have a common substratum. [22,301] In
this sense, there is a single thing, say, a stair or a mountain, with two
names, ascent and descent. [23,313]
66
Simplicius gives two reasons
why Aristotle does not discuss polynyms and heteronyms explicitly in
Categories 1: either he found them obvious or he relegated their discus-
sion to his rhetorical and poetical works. [23,619; cf. 36,2531]
Porphyry presents a classication of homonyms given also by many
other commentators. [in Cat. 65,1866,21] Homonyms are: I) by chance,
as Alexander names both the son of Priam and the son of Philip; II)
by intent, either 1) by similarity, as a man and a picture of a man are
both named by animal, or 2) by analogy, as the term (principle
or beginning) can be applied to numbers and lines and rivers, or 3)
from something ( ), as different items can be called medical as
their names are all said from medicine according to different accounts,
or 4) relative to something ( ), as things that are called healthy
as said relative to health. He notes that some put what is from or rela-
tive to something as intermediate between homonymy and synonymy.
[661521] This would include, for instance, Alexander, where he locates
focal meaning between homonymy taken in a narrow sense and syn-
onymy. [in Metaph. 241,324]
67
Also Porphyry discusses the example of
the foot of a bed or a mountain. He says that some, like Atticus, take
this to come about by metaphor from the foot of an animal.
68
However
he classies it as homonymous by analogy. [66,3467,32]
Simplicius gives the same classication and examples as Porphyry.
[31,2232,11] He adds only the remark that some combine II.3 & II.4
into a single type. [32,123] Still, given Aristotles discussion of the two
examples in Metaphysics IV and the connection between paronymy and
66
Also cf. Clement of Alexandria, Stromateis VIII.8.24.29, ed. Sthlin, p. 95,526;
Ammonius, in Cat. 16,2417,3; Alexander, in Top. V,4 398,14; Luna Commentaire,
p. 52.
67
Also Porphyry, in Cat. 66,18; Syrianus , in Met. 57,1820.
68
Hans Wagner , ber das Aristotelische pollachos legetai to on, Kantstudien 53
(1962), 7591, p. 75, observes that analogy became the main focus of discussions of
homonymy in Latin medieval philosophy.
50 allan bck
focal meaning discussed above, it is hard to see why the two types should
be distinguished in the rst place. To be sure, the medical example
concerns things used in the service of medicine, and so concerns the
efcient or productive cause, while the health example concerns things
for the sake of health and so concerns the nal cause. Yet these differ-
ences look material and not formal. Perhaps we can nd examples of
focal meaning that do not involve paronymy. Of course, we could nd
terms having focal meaning that are not derived from a common base,
like Aristotles own example of excellence [Cat. 10b59], but then we
would not have homonymy but some sort of heteronymy.
Following Alexander, Simplicius says that homonymy can hold
between things under the same genus, like equal for continuous and
for discrete quanta, or between things under different genera, like
equal for quanta and for relata. [in Phys. 403,1319] He is insisting on
its being possible for homonyms to be in the same genus, as Aristotle
says, for instance, that, when differentiae are in different species, there is
homonymy. [Part. An. 643a17] This distinction might be correlated
with a narrow and a broad conception of homonymy. Simplicius says
also that a triangle and triangularity are homonyms with respect to
triangle. [Simplicius, in Cat. 264,710]
69
However this position seems
to come from his Platonist insistence that material triangles, not being
perfect triangles, can be triangles in name only.
70
Ammonius however gives a more elaborate classication. [in Cat.
21,1622,10] Once again, homonyms are either I) by chance or II) by
intention 1) Some are homonyms of one another and paronyms of
what they are called after
71
a) from the efcient cause, as what is said
from something, as with medical b) from the nal cause, as what is
said relative to something, as with healthy 2) Others are homonyms of
one another and also homonyms of what they are called after, where
the two things a) differ in the times where they have the name i) when
one thing is named in memory of the other, earlier one, like calling a
child by the name of his father or teacher ii) when two have the same
name by chance iii) when the later one is named in the hope that it will
have attributes of the earlier thing, like naming someone today Plato
69
Cf. Plotinus, Enneads VI.3.2.24; also VI.II.1.
70
Michel Narcy , Lhomonymie entre Aristote et ses commentateurs noplatoniciens,
Les tudes Philosophiques 1 (1981), 3552, p. 47. S. Marc Cohen and G. Matthews,
trans. & comm., Ammonius: On Aristotles Categories (Ithaca, 1991), n. 42, claim that this
is Platonist.
71
Trans. S. Marc Cohen and G. Matthews, p. 30.
b) do not differ in the times where they have the name, and are named
i) by the similarity of the objects, as when the wise ( phronimos) man is
called wisdom ( phronesis) ii) by participation, like musical in musical
woman and musical knowledge iii) by analogy, like good applied to
the bed and to the mountain aa) in virtue of similarity, as Gorgias and
the river both have the name Gorgias, from moving rapidly bb) from the
similarity of the shape cc) by metaphor, like the feet and crown of a
mountain. Ammonius does not give many details, and may have run
together different classications into one.
Philoponus follows Ammonius somewhat, perhaps because his com-
mentary was, it seems, his notes on Ammonius. Be that as it may, he
offers two divisions.
The rst one is: homonymy either I) by chance II) by intention 1) by
memory 2) in hope 3) by analogy, as foot for an animal and a hill 4)
from something, like medical i) from a cause as a paradigm (i.e., the
formal cause?) as a picture of a man is called a man from the name
man being applied to the man ii) from the efcient cause, as a scalpel
is called healthy 5) relative to one, like healthy [and perhaps 6) not
by proportion, namely, perhaps. a proportion suitable for making an
analogy]. [in Cat. 16,2217,10]
The second one is: homonymy is either I) by chance II) by inten-
tion 1) where one of the homonyms is named paronymously from the
other a) from the efcient cause, as with medical b) from the nal
cause, as with healthy 2) where one of the homonyms is not named
paronymously from the other when one thing is named in memory of
the other earlier one. [21,1422,11] This second classication resembles
Ammonius one, but seems less organized, as Philoponus goes on to
observe that the from something and the relative to something can
be simultaneous or not, and that the from something can be based on
similarity or on the second having the shape of the rst, as the picture
has the shape of the man in the pictured and so both are called man.
He also lists the difference of being said in hope or in memory. What we
have here is basically Ammonius scheme, a bit more disorganized with
a few other distinctions already made appended perhaps in haste.
Whatever its nuances, the interpretation of Simplicius et al. makes
homonymy more a matter of the relationship between concepts about
things, than about the things themselves.
72
Perhaps in this he has
72
Concetta Luna , Commentaire, p. 41: . . . une interpretation conceptualiste
de lhomonymie: pour Simplicius , lhomonymie consiste dans le rapport entre un seul
52 allan bck
become more Platonist, as Plato saw homonymy as more a matter of
different beliefs people have about things than a doctrine about things
and their names.
73
I shall now make some remarks useful for understanding these clas-
sications and Avicennas use of them.
In surveying the corpus of Aristotle, the Greek commentators found
a broad and a narrow conception of homonymy. Sometimes Aristotle
calls any things said in many ways homonyms. Other times he distin-
guishes homonyms from things said to be relative to something. Some
beings do not have the merely accidental unity of a name, but a real
unity, sufcient to ground a science of being qua being as Alexander
notes. In contrast to this scientic use of a sort of homonymy, Aristotle
also names a fallacy one of homonymy.
Still, on account of Aristotles general use of homonymy, the
commentators had reason to see how all these more particular ways
in which things are said in many ways could t into a general scheme
of homonymy.
However, they tended to leave out, or at least deemphasize, some
cases of homonymy from their general classication: analogy, metaphor,
and the homonymy of the fallacy [which we might call, following
Avicenna et al. ambiguity]. Accordingly, they explain that Aristotle
did not discuss metaphor and analogy, and likewise heteronymy and
polynymy, in the Categories as he relegated these topics to rhetorical
and poetical works.
In some of these classications of homonymy we see the appear-
ance of paronymy. Simplicius even goes so far as to make paronymy
intermediate between synonymy and homonymy. [in Cat. 37,34] No
one gives a full explanation why, but we can construct one.
74
Accept
mot et une multiplicit de rpresentations mentales produites par ce mot, pltot que
dans le rapport entre un seul mot et une multiplicit de choses.
73
Concetta Luna , Commentaire, pp. 567: Comme la soulign J. P. Anton , la
diffrence dapproche du problems de lhomonymie, entre Aristote et Platon, consists
en ce que pour Platon lhomonymie est une question dopinions diffrentes (ce qui est
cohrent avec la conception platonicienne de la philosophie comme dialogue), tandis que
pour Aristote les ambiguits quil faut lever ne sont pas celles des opinions: le problme
nat uniquement parce que certaines ralits ont Ie mme nom que dautres. Cf.
J. P. Anton, The Aristotelian Doctrine of Homonymy in the Categories and its Platonic
Antecedents, Journal of the History of Philosophy 6 (1968), 315326, p. 317.
74
This is the charitable interpretation. The other leading candidate is that they con-
fused in what ways the relative to something and the from something are related.
Aristotles doctrine of focal meaning. Then there will be things having
different denitions yet having the same name derived from the focal
one. E.g., walking and having a temperature of 98F are both called
healthy, a name derived from the name health.
75
(Note that their
having the same name is a feature accidental to particular natural lan-
guages, like English and Greek. Indeed Arabic with its more complex
verbal system would tend to use different forms for names for being
productive of health and being a sign of health.)
76
Here then we have
a case of two things sharing the same name yet differing in account.
So they are homonyms. Yet each of them is also a paronym relative
to health. On this interpretation, health, the state itself, would not be
a homonym relative to one of the things called healthy.
Ammonius classication, despite being expressed unclearly, falls along
these lines. Some homonyms have a paronymous name in common;
others do not. Thus Philoponus says, for it (medical) is named par-
onymously from it (medicine), while [holding] homonymously to each
other. [in Cat. 21,201; cf Simplicius , in Cat. 264,710]
77
However, his
use of from something and relative to something would be rather
careless. For all homonyms having a paronymous name are said from
something. But then Ammonius goes on to subdivide these two into
those having things named from the efcient cause, as being produc-
tive of, say, health, and into those having things named from the nal
cause, as for the sake of health. The problem is that he describes the
former as from something and the latter as relative to something.
Yet these expressions also describe the structure common to all such
paronyms. Ammonius himself admits that some combine these two
types [in Cat. 21,212].
75
Cf. Philoponus, in Cat. 15,710, although he is discussing names for things and
for their essence.
76
As Avicenna himself notes at Al-Maqlt, eds. G. Anawati, A. El-Ehwani, M. El-
Khodeiri, & S. Zayed (Cairo, 1959) (Part One, Volume Two of A-hif) 16,1217,14;
Al-Ibra, ed. M. Al-Khudayri (Cairo, 1970), (Part One, Volume Three of A-hif)
19,1621,6. He is following al-Frb, Kitb al-urf, ed. M. Mahdi (Beirut, 1969), e.g.,
1920, 71,215. On the structure of paronymy in Arabic see W. Wright , A Grammar
of the Arabic Language (Cambridge, 1967), Vol. I 195; Wolfdietrich Fischer , A Grammar
of Classical Arabic, 3rd ed. trans. J. Rodgers (New Haven, 2002) p. 35.
77
Concetta Luna , Commentaire, p. 149: La langue grecque na pas deux termes
pour dsigner la triangularit et le triangle, comme elle en a deux pour designer la
courbure . . . et ce qui est courbe . . . On dit donc la fois pour dsigner la
proprit dtre triangle . . . et pour dsigner le triangle crit sur le tableau ou fait en bois
(le poion). Cette double signication de trigwnon en fait un terme homonyme.
54 allan bck
The doctrines of the Greek commentators passed into Islamic phi-
losophy, just as the late Greek commentators themselves were exiled
from a Byzantine Christian court but then invited to establish a school
by a Muslim caliph.
78
Let us now see what Avicenna does with it.
Avicenna begins by discussing synonymy, whereas Aristotle starts
Categories 1 with homonymy. The Greek commentators made a great
defense of Aristotles order of presentation (Cf. Simplicius , in Cat.
21,122,13). In contrast, Avicenna is rejecting it.
Avicenna holds that names have a relation to things via senses or
concepts. We have seen some basis for such a conception of senses in
Aristotle and the Greek commentators, as well as in the Arabic transla-
tion of the Categories and in al-Frb. There is also the Stoic concept
of the lekton, which seems quite a close relative of Avicennas concep-
tion of a sense (manan). Aside from Stoic doctrines presented in the
Greek commentators, it is hard to know what Stoic sources Avicenna
had. Still, given his extensive treatment of hypotheticals in his formal
logic (in al-Qys), he seems to have had some. In any case, for him the
conception of a sense is fundamental. Also suggestive is the fact that
Stoics tended not to distinguish synonymy and pluronymy, nor even
heteronymy and homonymy.
79
This tendency would encourage Avicenna
to see homonymy, synonymy etc. lying on a continuum.
Constructed on a conception of the sense, Avicennas theory of
homonymy is far simpler than those of the Greek commentators.
Seeing its simplicity depends on having a certain interpretation of his
conception of similarity. What makes things similar but not the same?
We may say: because they share some but not all features. This holds
even in a metaphorical comparison. A woman may be likened to a rose
with respect to beauty but not with respect to having thorns or needing
pruning in the winter. So too senses may have some similarity or overlap
but not be the same. Thus, when Avicenna distinguishes homonyms,
he says that homonyms of the second sort do not have the same sense
but have a similarity. Now not having the same sense is compatible with
having and with not having totally different senses. So we get cases like
78
Gerhard Endress, Die Wissenschaftliche Literatur, in Grundrisse der Arabischen
Philologie, ed. H. Gtje (Wiesbaden, 1987), vol. 2, pp. 4025.
79
Cf. Simplicius , in Cat. 36,812. Concetta Luna , Commentaire, p. 115: Dans
lusage stoicien, synonyme a donc la mme signication que polyonyme et, pour-
rat-on dire, il correspond tout fait lusage moderne de ce mot. Simplicius afrme
que lusage aristotlicien est plus appropri . . . Cf. C. Douglas McGee, Who Means
What by Synonymy?, Inquiry 2 (1959), 199212.
similarity, such as Socrates and his picture both being named man.
Avicenna will consider metaphors of the same type, like the leg of
an animal and a bed, and analogies, like the arch of number and
lineapparently with analogy being a type of metaphor.
80

Furthermore, it helps to view Avicennas classication of homonymy
not as discrete but as continuous. In effect, he gives a continuum with
one endpoint limiting the homonymous being the strictly synonymous
and the other being the completely heteronomous (which he mentions at
11,4, and at 15,1616,3 along with polynyms), having neither a common
name, account, or sense. Inside this continuum, within each type of
homonymy, he will discuss cases that satisfy its description more or less
so. Moreover, because having disjoint, discrete types does not concern
him, he does not fuss much over the classication but over the details
of the structure of particular cases. So he has a continuum within his
division, and perhaps implies this at Al-Maqlt 11,57; 14,615,15.
But, for the sake of reference, I shall divide up and number the sorts
that he does discuss.
His classication is threefold:
It may be said that everything that is not by way of agreement [synonymy]
is by coincidence of the name [homonymy], and has three divisions: that
is because either [ I ] the sense is one in itself, even through it differs in
another respect, or that [ II ] it is not one, but between the two of them
there is a similarity, or that [ III ] it is not one and there is also not a
similarity between them.
81
The rst division [ I ] contains things that are not strictly synonymous
but share a single sense. In light of the descriptions and examples, the
rst division includes those things said by Aristotle and the commenta-
tors to be named either from something or relative to something.
82

[ II ] The second division concerns cases where the things being named
have a real basis of similarity for sharing the name, but not the same
sense. From the examples, this type seems to include all cases based
on a real resemblance, including metaphor, analogy, and similarity of
appearance, like the example of Socrates and his picture. [ III ] The
third division concerns cases where the things share only the name.
83

80
Cf. Aristotle, Poetics 1457b616.
81
Al-Maqlt 10,47; all translations from this are mine.
82
Cf. Aristotle, Metaphysics 1030a32b7; Nicomachean Ethics 1147b341148a2.
83
A likely important source for this classication is al-Frbs Short Treatise On Aristotles
De interpretatione, trans. Zimmermann (Oxford, 1981), pp. 4952, pp. 22730.
56 allan bck
Avicenna considers senses, the concepts in the mental language [cf.
Int. 16a38], to exist in the mind via sense perception and abstraction.
Such senses are given conventional names by imposition or stipulation
in the written or spoken language. These senses differ from denitions
or accounts () proper, as they contain all that is common to the
individuals from which they are abstracted. Avicenna gives the example
of people living in the Sudan: from them we would have the sense or
concept of a human being as having black skin, even though the deni-
tion of human being, grounded upon quiddities in themselves accessed
via intuition () does not include that characteristic.
84
I now discuss each type of homonymy distinguished:
[ I ] Given what he says at 11,5, Avicenna divides homonyms sharing
a name and a single sense into A) the absolute and B) the relative. Both
the absolute and the relative sorts concern things with focal meaning.
Avicenna says about all these cases that the two things could have
instead been named by different names or expressionspresumably
synonymous ones. Even in the absolute cases, we could have said,
more precisely, e.g., being that is prior and being that is posterior.
Hence, even though only the relative cases use paronymous names as
commonly understood, even the absolute cases use names having a
whiff of the paronymous.
85

[A] In the absolute sort, the name applies to the two things in the
same sense. The name does not need a different sense when predicated
of each. Yet it will need a different account, or perhaps denition,
if the things are to avoid being synonyms.
1. [10,811] Existent (being; ) can be applied to substances, quan-
tities etc. in the same sense and so is used absolutely.
86
Quantities and
substances exist in re in the same sense. Yet they do differ in terms of
priority and posteriority and this, as Aristotle had said, eliminates strict
synonymy.
87
(Indeed Simplicius had brought up this issue later on in
his Categories commentary. [418,19419,6]) Also they differ in the sense
that they have different modes of existence or being: as a substance,
as a quantity etc. This ts with Avicennas metaphysics where the
84
Al-Burhn, ed. Badawi (Cairo, 1954), 46,116 et passim. Cf. Avicenna, F Al-Nafs,
ed. G. Anawati (Cairo, 1962), II.2. I thank Claude Panaccio for instigating this para-
graph.
85
As in Porphyrys distinction II.2 perhaps.
86
Aristotle, Metaphysics 1030a32b7.
87
Eudemian Ethics 1218a18.
necessary being causes various quiddities to come to exist in re and to
associate with other ones via acquiring differentiae and via some becom-
ing accidental to others.
88

2. [10,1216] Avicenna also says that homonyms may have a name
with the same sense and be simultaneous (being neither prior nor
posterior), while differing in being primary and secondary. He gives
no examples. I offer as possible candidates a triangle and an isosceles
triangle both being named triangle or being named something hav-
ing its interior angles equal to two right angles. As Aristotle had said,
such names hold of both triangle and the isosceles triangle, but of the
former primarily and of the latter not so. [An. Po. I.5]
3. [10,1711,4] Other names keep the same sense and even the same
account but differ in what they signify about the things they describe.
Ivory and snow are not equally white. They are white understood on
the same denition. Still white when applied to ivory means off-white,
lets say, while, when applied to snow, it means arctic white. The Greek
commentators do not mention this case.
89

Why does Avicenna say that such homonyms have the same sense?
Apparently, he is thinking of white as the genus, covering a range of
whitish colors. These colors would be differentiated by specic differentiae
like off- and arctic in the example above. They still share the same
general, Aristotelian denition of whiteness: standing out in sight.
90

Nevertheless, they are not precisely synonyms with respect to white it
seems. For they are more or less white, and so have the name hold of
them in different degrees. They also fall under the priority condition
given by Aristotle,
91
like the rst two cases (IA.1 & 2).
Avicenna also remarks, about this sort but presumably with general
import, that in ordinary language names will be used imprecisely and
their senses must be ascertained from the intention of the speaker.
Indeed Avicenna regularly distinguishes an ideal, technical language
from the ordinary vernacular. Perhaps he is thinking here of such cases
as when white is taken to describe varying colors of human skin.
92
88
See Allan Bck, The Triplex Status and its Justication, Studies in the History of
Logic, eds. I. Angelelli & M. Cerrezo (Berlin, 1996).
89
Possibly: Nicomachean Ethics 1147b341148a2.
90
Cf. Ammonius, in Cat. 45,2; 40, 134.
91
Eudemian Ethics 1218a18.
92
See Allan Bck, The Ordinary Language Approach in Traditional Logic, in
Argumentationstheorie, ed. K. Jacobi (Leiden, 1993), pp. 50730, pp. 5112.
58 allan bck
[ B] [11,57] The relative sort concerns things having the same name
said relative to something in Aristotles account of focal meaning. In
this case the shared sense does not come from the name common to
the two homonyms but from the thing relative to which they have a
relationship. Thus the knife and exercise may have a common, parony-
mous name, medical. Medical however is dened differently when
predicated of the knife and exercise. The shared sense comes from the
presence of medicine in the different denitions.
Avicenna differs fundamentally from the Greek commentators and
perhaps from Aristotle himself about what things he allows to share a
single name in the same sense. He has a set of examples of names rang-
ing from existent to white and philosophy to medical and healthy
and divine. He says that some of these names hold absolutely with
different degrees of priority, primacy, or participation, while others hold
only relatively. The fundamental point lies in his claiming that cases like
being and medical have formal differences. Even Aristotle grouped
these cases together in arguing for the unity of a science of being qua
being. To put the point historically backwards, Avicenna has a position
on transcendental terms more like Scotus than Aristotles.
II. [11,812,7] The type where the sense of the name is not one
but the homonyms have a similarity is homonymy proper, in the nar-
row sense distinguished by Alexander. Avicenna runs together two
examples commonly kept apart by the Greek commentators. (A) The
rst is the basic one from Categories 1, where an animal and its picture
are homonyms with respect to being named animal. (B) The second
is the standard example of a metaphor, of a leg of an animal and a
leg of a bed.
93
Avicenna recognizes that the latter is a metaphor while
the former is not. Still he observes that the metaphor was made on
the basis of a perceived similarityand so he does not use the usual
Arabic word for metaphor but instead calls this a transferred name:
we might call it a simile.
94
In both cases the name is applied to one
homonym primarily and to the other secondarily.
He also remarks that some of these homonyms are (C) xed by itself
(or per se) while (D) others are based on a relationship (nisba).
(C) The rst sort seems straightforward. The very attributes of the
two things involved x the similarity. Thus the form of the animal and
93
Poetics 1457b169; Rhetoric 1401a125.
94
Cf. Zimmermann, al-Frbs Commentary, p. 227, nn. 23.
the form of its picture are similar, and so are the shapes of the legs of
animals and tables because of attributes of the two things involved.
(D) The second sort is more cryptic. Nisba can also mean attribu-
tion or proportion. Avicenna seems to use it in a quasi-technical sense
to signify a relationship between two things that does not fall into the
category of relation.
95
From the Greek commentators we can see that
the relationship here is analogy, which can be understood as a pro-
portion. The example of beginning () is the standard one of
analogy in Porphyry and Simplicius .
Some of the subdivisions of the classications given by the Greek
commentators might apply to this second sort, as I will discuss below. For
example, Avicenna mentions Ammonius distinction [I.2.i
iii] of giving
things a common name by chance, in memory, or in hope. [14,15]
III. [11,813.14] In the third main type, the two things have only
the name in common, with no shared sense or similarity. Avicenna
does not include metaphor here when it has a real basisthe trans-
ferred name or simile. However, when dog is applied to a star and
an animal, in the example from the Rhetoric, Avicenna thinks that there
is no real similarity between stars and dogs, but only a merely mental
one.
96
So Aristotle puts this example here. Paralogisms of the fallacy
of homonymy would belong to this sort too.
97

Avicenna does offer the option that, if such cases like the dog and
the Dog star could be shown to have a real similarity, then they would
belong to the second sort and be similes. [13,1514,6] He uses the
example of the name ayn applied to an eye and to cash. He gives an
etymological way to connect them up so that they have a real similar-
ity, but seems to be ambivalent about whether or not there is indeed a
real similarity. He distinguishes here the three causes for applying the
name of an eye (ayn) to cash that Ammonius had given: coincidence,
memory, and hope. [II.2ai
iii] For Avicenna these would suggest that

the naming had no real similarity. Still, he can also apply his remark
about ordinary language use and the intention of the speaker
here, so as to allow for a loose conception of the second type (II).
Avicenna settles on calling this third sort ambiguity (takk). Thus,
when discussing whether being is a genus for the ten categories or
whether it is said in many ways, he says:
95
Nicomachean Ethics 1148b710. See too 67,23; 79,5; 85,10;145,17. Cf. al-Frb,
Kitb al-urf 3740.
96
Aristotle, Phyiscs 249a1925; Nicomachean Ethics 1147b341148a2.
97
Cf. Al-Safsaa, ed. A. Ehwany (Cairo, 1958), 45,5; 50,14.
60 allan bck
And the multiplication of what is being talked about is in three modes:
Either there is multiplication via agreement in its subjects, or a multipli-
cation via a coincidence in inection comprehending its similarity and
participation [ homonymy], or a multiplication via ambiguity. [59,101]
Apparently Avicenna means by agreement in its subjects here not
synonymy (as agreement normally means for him) but having the
same sense. If so, we have the focal-meaning type of homonymy (I),
homonymy in the narrow sense, the second type (II) of homonymy, and
ambiguity as its third type (III). Avicenna, being interested in science
and not sophistry, would have little use in his own philosophy for this
third sort, and so would separate it out.
He distinguishes the second and third types of homonymy in the
same way at 14,615,15. He goes on to observe that they agree in
not having the same account but using a single name. He then gives
a somewhat confusing discussion of the example where the name leg
is applied to both a bed and an animal. Above he said that this was a
case of the second type. Here he suggests a way that it might, alter-
natively, be taken of the third type and be a case of mere ambiguity.
The main idea is that the things could have the similarity of being
three-dimensional bodies, but differ in being legs. For after all, Aristotle
would say that a leg that is not alive is a leg only homonymously.
[Part. An. 640b35641a5; cf. Metaph. 1035b235; 1036b302] If this
were the only similarity between bed and animal legsand not also
being a support for the main bulk of an object etc.then leg would
indeed be used ambiguously. However the things being compared
do seem to have such additional similarities. Thus the example looks
contrived. It has the further problem that animal legs and bed legs are
legs synonymously with respect to the name bodyunless Avicenna
is thinking of making it an instance of one of the sorts of the rst
type of homonymy (I). This entire discussion would then resemble the
standard one in the Greek commentators about whether homonyms
are synonyms: with respect to the name homonym the things that are
homonyms via another name are synonyms; with respect to that other
name, they are homonyms.
98

(Avicenna also considers but rejects an argument appearing in the
Sophistical Refutations [165a103] that homonymy is inevitable as there
98
E.g., Simplicius , in Cat. 30,1631,21.
are an innite number of things but only a nite number of words.
This entire issue does not appear in the Greek commentators.)
In sum, Avicenna has a eld of homonymy running the full spec-
trum of the phenomena: the linguistic data and the endoxa, the texts of
Aristotle and the others, and the reputable common ways of talking.
Strict synonymy, where the name has no change in inection and the
denition, gives one endpoint or limit to the spectrum. The things hav-
ing the same name in the same sense but in different respects (I) lie on
the boundary of homonymy closest to that endpoint. Mere ambiguity,
where it has just chanced in a particular language that the same sign,
spoken and/or written, happens to be used for two different objects
without any common ground, provides the other boundary. Heteronomy,
where two different things have two different names, senses, and deni-
tions, provides the other endpoint or limit. Various changes of meaning,
along with the respective texts in the Aristotelian tradition, can be tted
along the continuum of this spectrum within those limits.
Avicennas division of homonymy does not match up with the stan-
dard one of the Greek commentators. Take their main one: homonyms
by chance or by intention. Some of each of these types will fall into
his third type, where there is no overlap in sense or in similarity. The
same word may chance to name two different sorts of objects, like
in Greek, ayn in Arabic, and bank in English, with no overlap.
Yet likewise we can have such names intentionally: if I call my child
angel in hope that she will act like one, or my wastebasket Dubya in
memory of a U.S. President. One might object that in the latter cases
there is an imagined similarity. But for Avicenna such fancies are for
sophists; an imagined similarity need not be a real one.
Avicennas scheme for homonymy has the advantage over the earlier
ones of greater compass. He can include in it the fallacious, meta-
phorical, and ambiguous sort of homonymy that were omitted or left
to one side earlier. He also has a greater range of texts from Aristotle,
e.g., Sophistical Refutations 165a103, on the argument that homonymy
is inevitable because things are innite while words are nite.
Avicenna has a very complex terminology for homonymy. The
homonymous he indicates by two expressions, not necessarily similar in
his own theory: participating or sharing (itirk), on the one hand,
and coinciding (ittifq), on the other. The latter appears in Isqs
translation of Categories 1. The former comes from Arabic philology
and was used by al-Frb, who uses participation in the name when
62 allan bck
speaking of two senses of non-existent.
99
These two expressions have
the common notion that the things share the name, or coincide or agree
in it, but not in the denition. In addition Avicenna adds ambiguous
(mutabih) and equivocal (muakuk) to the mix, each of which might
be used to signify homonymy. Many translators have ignored these
differences and just translated most of these expressions by ambigu-
ous or homonymous.
100
Yet Avicenna does take some pains to keep
all these terms distinct.
However, Avicenna perhaps has a technical usage for these differ-
ent expressions, obscured by such translations. (I have noted that the
Revised Oxford translation of Aristotle is not too precise either!) Now
we have seen that the Greek commentators kept distinct various types
of homonymy often signied by different expressions. Following their
tradition, Avicenna distinguishes different types of homonymy too. Does
he have a systematic vocabulary?
In light of Avicennas classication of homonymy, I suggest that
Avicenna generally, albeit with exceptions, has the following usage:
coincident, following Isqs translation, signifies homonymy in
general, and sometimes, in particular the narrow sort of homonymy
discussed in Categories 1, i.e., type II. Likewise equivocal when modi-
fying name would represent some sort of homonymy, such that the
two homonyms have two different denitions or accounts relative to
that name.
101
Participating signies the rst type (I) of homonymy,
based on focal meaning, and ambiguity the third type (III) dealing
with poetry, rhetoric, and sophistry. Thus, for instance he contrasts
synonymy, homonymy by participation, and ambiguity.
102
So perhaps
Avicenna has a systematic vocabulary as well as a systematic theory
for homonymy.
How does Avicennas classication fare as an interpretation of Aris-
totle? Above all, his theory seems to differ, as he uses senses as well as
99
Kitb al-urf 122,4; so too in Avicenna, Safsaa 45,5; 47,1 when discussing
the fallacy of homonymy. Cf. Zimmermann, al-Frbs Commentary and Short Treatise,
p. 228 n.3. This use of participation may not be confusing Aristotle with Plato. Cf.
Ammonius, in Cat. 22,210, on division of homonymy, incl. participation. S. Marc
Cohen and G. Matthews, trans. & comm. Ammonius: On Aristotles Categories (Ithaca,
1991), n. 42, claim that this is Platonist.
100
Goichon in her Lexique and the editors of Al-Maqlt in their Index of terms do
give different translations for most but dont explain the differences.
101
Cf. Al-Maqlt 11,5; 94,13.
102
Al-Maqlt 28,58.
names and accounts (or denitions) in his analysis of homonymy.
Aristotle himself does not mention senses in his accountunless we
accept the new Owenite orthodoxy that homonymy deals with meanings.
So it depends how strongly we take Aristotles claim that homonyms
have only a name in common. [Cat. 1a3]
Avicennas scheme fares better as giving a theory covering Aristotles
uses and remarks on homonymy. The continuum gives a clear sense to
Aristotles distinction between close and remote homonymy
103
in terms
of being closer to or further away from having a relation of synonymy.
That is, we can think of objects as having senses and then consider
to what extent the senses overlap. For modern interpreters also have
claimed that the distinction between homonymy (complete difference
in denitions) and synonymy (identity of denitions) is not a dichotomy
because denitions may be partially identical, partially different.
104

Again, Hintikka also complains that Aristotle is not consistent in
different passages in classifying analogical and metaphorical uses of a
word.
105
Avicennas discussion about how metaphors can be taken as
type II or III has some use here as he offers a way to put analogy and
metaphor on a continuum.
We might compare Avicennas classication of homonymy with a
modern one, given as an interpretation of Aristotle. Christopher Shields ,
who also attributes senses and meanings to Aristotle, takes homonymy
in general to be comprehensive homonymy, where the homonyms
are not completely overlapping in denition.
106
He divides compre-
hensive homonymy into the discrete and the associated.
107
The discrete
has no overlap in denition, while the associated does.
108
Homonymy
103
Aristotle, Physics 249a1925; Nicomachean Ethics 1147b341148a2.
104
Hintikka. Aristotle and the Ambiguity of Ambiguity, p. 141.
105
Ibid., p. 143.
106
Christopher Shields , Unity in Multiplicity (Oxford, 1999), p. 11. Cf. T. H. Irwin,
Homonymy in Aristotle, Review of Metaphysics 34 (1981), p. 255 on conneced
and unconnected homonymy. Likewise, Hintikka. Aristotle and the Ambiguity of
Ambiguity, p. 144: We have to realize that the distinction between homonyms and
terms with merely different applications often amounts to something rather different
from distinction between complete and partial discrepancy of denitions.
107
Christopher Shields , Unity in Multiplicity, p. 29.
108
So too T. H. Irwin, Homonymy in Aristotle, p. 524, with connected and
disconnected homonymyi.e., Avicennas I & II versus III. He goes on, p. 528,
to contrast homonymy where the term is predicated of both things strictly, from the
spurious cases, where only one is, while the second has the predication based on
similarityi.e., Avicennas I versus II.
64 allan bck
of focal meaning is one sort of the associated, which Shields calls the
core-dependent.
109
He also divides the associated into the accidental
and the non-accidental, which seems to include the core-dependent.
110

(He also distinguishes seductive and non-seductive homonymy, as well
as homonymy based on shallow and deep differences in signication,
although he does not want these in his classication proper.
111
The
former distinction contrasts homonymies easy to spot from those that
are not; the latter does something similar by appealing to nominal
versus real denitions.)
In the terms of the Greek commentators (whom he does not cite
much) and Avicenna, Shields comprehensive is homonymy in the
broad sense (IIII), the discrete is homonymy in the narrow sense (I &
II), the associated is that of the focal meaning sort (I), the accidental
is homonymy by chance, and the non-accidental is homonymy by
intention. So Avicenna does not have too different a classication from
Shields, except that his doctrine of shared senses gives more precision
to overlap, and his scheme gives Aristotelian philosophy more theo-
retical coherence. Indeed, if we are to follow the new orthodoxy of
attributing senses to Aristotle, we might as well become followers of
Avicenna. Anthropological trope theory today uses a vertical dimension
of analogy similar to Avicennas continuum.
112
Moreover, as for the
contemporary treatment of fallacy, if Powers is right that all fallacies
can be reduced to the fallacy of equivocation, Avicennas classication
may offer a start to a comprehensive theory.
113
There are some semblances of scholarly and philological scholasticism,
with regard to Aristotelian studies . . . but the hurricane of Avicennas
philosophy quickly swept such tendencies away.
114

Averroes as The Commentator?
The Aristotelian tradition may have slighted Avicenna as a com-
mentator on account of Averroes being given canonical status as the
109
Christopher Shields , Unity in Multiplicity, pp. 37; 27.
110
Ibid., p. 29.
111
Ibid., pp. 3940; 1012. Cf. Aristotle, Sophistical Refutations 182b257.
112
Paul Friedrich , Polytropy, in Beyond Metaphor, ed. J. Fernandez (Stanford, 1991),
pp. 1955, p. 40.
113
Lawrence Powers , Equivocation, in Fallacies, eds. H. Hansen & R. Pinto
(University Park, 1995), pp. 293301, p. 287.
114
Dmitri Gutas, Greek Thought, Arabic Culture, p. 55.
Commentator. In addition, Averroes depicts Avicenna as a deviant
Aristotelian and a lousy commentator. Before assessing Avicennas worth
as a commentator, let me review this historiography.
Averroes is known in the West in two guises: rst, as the Commentator
on Aristotles works (with the exception of a few like the Politics).
115
This
appellation came about because he gave the most literal and extensive
commentary then available, and because in large part the West received
the works of Aristotle in Latin translation incorporated into Averroes
commentary. Second, as the leader of Averroism in the Renaissance,
where, among other things, he was followed for his views on the active
intellect. In his latter guise, his Incoherence of the Incoherence assumed
prominence, where he debates with Al-Ghazl over the orthodoxy of
the views of the philosophers. Yet even here Averroes mostly follows
the Greek commentators and occasionally earlier Islamic commentators
like al-Frb or Avicenna .
Thus the medievals reckoned Averroes as The Commentator of
Aristotle without equal. At any rate, so they appellated him. Why did
he deserve to have this name so imposed?
The usual story is this: Averroes , that is, Abul Walid Muhammad
ibn Ahmad ibn Muhammad ibn Rushd, was born at Cordoba in
Andalusia in the twelfth century into a family of jurists and theolo-
gians of the literalist tradition. He himself became a judge, as well
as a court physician and a philosopher. He was commissioned by the
caliph, Ab Yaqb , to write a series of commentaries on Aristotle, so
that the doctrine could be understood more clearly and used by the
Islamic community.
116

At the least this standard view of Averroes as The Commentator
needs modication. For we know that there were many literal com-
mentaries on Aristotles works available before Averroes and indeed
before Avicenna , notably in Baghdad. Perhaps the point remains that
those in Spain did not have these works accessible, and so needed a
set of Aristotle commentaries locally. However, Muslim Spain was not
that isolated: recall that Maimonides managed to move from Spain to
Africa to Egypt.
At any rate, upon his commission from the caliph, Averroes then wrote
a series of threefold commentaries on most of the works of Aristotle.
115
Oliver Leaman , Averroes and his Philosophy (Oxford, 1988), p. 179.
116
George Hourani , Averroes on the Harmony of Philosophy and Religion (Leiden, 1959),
pp. 123.
66 allan bck
For most of the works, he wrote an epitome, a medium-length or
middle commentary, and a long commentary. For him, these com-
mentaries had different functions: 1) the epitomes were designed for
general use, and accordingly contained many remarks not to be found
in the works of Aristotle, so as to relate the doctrines of Aristotle to the
current conditions in Muslim Andalusia; 2) the middle commentaries
were designed for the religious authorities, those who had a profound
knowledge of Islam but were not philosophers; 3) the long commen-
taries were designed for specialists, sc., for the few, elite philosophers.
This division of labor among the commentaries agrees with Averroes
theory of the threefold levels of truth: the rhetorical, the dialectical,
and the demonstrative.
117

Averroes commentaries follow in the tradition of his native culture
and jurisprudence: they are literalist. Despite the insistence of some
on the brilliance of Averroes,
118
I can nd nothing original in detail in
Averroes that cannot be found in earlier philosophers and commenta-
tors, notably Alexander and Themistius . The exceptions concern general
and not particular features, as I discuss next. We might call him not the
Commentator as the medievals did, but the Paraphraser. However,
I am not certain that this is a criticism. For Averroes was a literalist.
Further, at his time, given that many commentators on Aristotle, like
Porphyry , Simplicius , and al-Frb , had injected Neoplatonic and
Stoic inuences, and others, like Avicenna , had departed greatly from
the text, a literalist interpretation, intent on reclaiming the original
Aristotle, had some value.
Moreover, Averroes did not paraphrase blindly all of Aristotles works
nor the earlier commentaries, especially of Alexander of Aphrodisias.
For instance, take his commentary on the Poetics. Note that Islam had
no tradition of theater. Accordingly, Averroes had no experience of
plays, whether tragic or comic. Now, in writing his commentary on the
Poetics, Averroes did not follow Avicenna s lead, of trying, pathetically,
to explain what tragedies and comedies were on the basis, it seems, of
marginalia in the Arabic translations of the Poetics. Instead, he looked for
counterparts for tragedies and comedies in his own culture. He found
them in the poetical traditions of eulogy and satire. His examples are
not as silly as you might think. Indeed, they give some plausibility to the
117
Hourani , On the Harmony of Religion and Philosophy, pp. 646.
118
E.g., Leaman , Averroes pp. 10; 110.
claim that Aristotles philosophy, or, if you like, philosophy in general,
has some claim to being transcultural.
Another instance of Averroes independent thought concerns politi-
cal theory. Averroes commented on most works of Aristotle. One big
exception is the Politics. In its place, he commented on Plato s Republic.
We can suspect various reasons for this. One is that Averroes is follow-
ing the Islamic tradition of al-Frb et al., who commented upon and
extended the doctrines of Platos Republic to Islam. Another is that he
would have found Platos political theory more relevant to the caliphate
structure of government common in Andalusia. A third is that, given
his own theory of the threefold levels of truth, he would have found
Platos view of the relation of philosophy and religion, with its sup-
porting doctrines, more congenial or, at any rate, more explicit than
Aristotles. Still, on the whole, Averroes does little in his commentaries
besides present Aristotles texts in a close reading, with explications
drawn from earlier commentators.
Averroes may though have had originality in his depiction of Avicenna .
In his debate with Al-Ghazl , largely unknown to the medieval West,
he defended philosophy by offering Avicenna as a scapegoat. In general
he asserts that Aristotle and the like do not have views antithetical to
Islam. Rather only Aristotle as interpreted by Avicenna does. Averroes
concedes the incoherence of a philosopher, sc., of Avicennas views,
both with the text of Aristotle and with the truth, while defending the
coherence of the other philosophers.
In this, Averroes may have been partly responsible for the Latin
medievals having an Avicenne ctif, as well as an Aristote ctif. For
instance, he represents Avicenna as dening the possible as what has
a cause and the necessary existent as what does not have a cause.
119

True, Avicenna does hold that the possible existent needs a cause to
exist while the necessary existent does not.
120
Still, he holds too that
the logically possible itself like the necessary needs no cause for being
possible: it is possible in itself. Averroes makes other mistakes about
Avicennas doctrines, e.g., that he denied that there can be an innite
number of souls.
121
Again, Averroes accuses Avicenna of being wrong
on the nature of the empirical agent, and sides with Aristotle. However,
119
Van den Bergh , Tahfut, pp. 1646 [27780].
120
Al-Ilhiyyt I.6.
121
Van den Bergh , Tahfut, p. 163 [274], but see p. 14 n. 6; n. 1 and Al-Ilhiyyt
IX.34.
68 allan bck
he does not seem to have Aristotles view correctly.
122
Again, Averroes
may be making the same mistake in accusing Avicenna for introducing
matter for incorporeal substances and not following Aristotle.
123
Again, in his Questions on Logic, Averroes sees Avicenna as the odd-
ball among commentators who has committed many errors both of
interpretation and of philosophy. He insists, against Avicenna, that the
copula serves merely to connect subject and predicate and not to make
an assertion of existence.
124
He particularly dislikes Avicennas analysis
(derived from Sosigenes and Alexander ) of necessary and categorical
propositions into different types according to the duration of the exis-
tence of the subject and the time of predication.
125
However, Averroes
own views look peculiar. He says that the existence of individual men
has no relevance to the truth of every man is animal because the
universals are not generable.
126
That is, a universal proposition concerns
universals and not individuals. This surely moves away from Aristotles
insistence on the primacy of individual substances. (Averroes may have
been inclined to this view by his view of the copula.)
One circumstance that might explain Averroes inaccurate portrait
of Avicenna and Aristotle?concerns the philosophical community
in Spain at his time. There were then some rather fervent followers of
al-Frb and Avicenna.
127
Averroes might be attacking their interpreta-
tions rather than the tortuous texts of Avicenna.
We might further explain a lot of Averroes claims in light of his
political motives.
128
He wanted to make philosophy respectable to
122
Van den Bergh , Tahfut, pp. 1089 [1802]; 108 n. 1. Aristotle, Metaphysics
1073a 28
123
Van den Bergh , Tahfut, p. 160 [2701], but cf. Aristotle, Metaphysics 1069b 25
124
Quaestiones in libros logicae, in Aristotelis Opera cum Averrois Commentariis (Venice
156274; repr. Frankfurt-am-Main, 1962), vol. Ib 78B. (I have consulted the Arabic
text in D. M. Dunlop , Averroes on the Modality of Propositions, Islamic Studies 1
(1962), when available: 112, pp. 324).
125
Quaestiones 79L; 80B. He says, 80B-C, that Avicenna has a different view in An-Najt
(so too Nicholas Rescher, Studies in the History of Arabic Logic (Liverpool) 1963, p. 104.
But cf. the notes above: Avicenna seems there only to be summarizing and simplifying
his doctrines. George Hourani , Ibn Sina on Necessary and Possible Existence, The
Philosophical Forum 14.1 (1974), p. 74, agrees.
126
Quaestiones, 80C.
127
Nicholas Rescher , Studies in the History of Arabic Logic, p. 90.
128
Cf. Barry Kogan , Averroes and the Metaphysics of Causation (Albany, 1985), pp. 7; 125;
7223. He says, pp. 523, that Averroes is concealing his real views in the Tahfut.
But where then are they? In his commentaries on Aristotle? Cf. too, Thrse-Ann
Druart , Averroes on Gods Knowledge of Being qua Being, in Studies in Thomistic
Theology, ed. P. Lockey (Houston, 1996), p. 185.
Muslim fundamentalists perhaps at the cost of truth. Too, Averroes
may be attacking some of his Avicennian contemporaries more than
Avicenna himself. For many of Avicennas current followers ascribe some
of Alexander s doctrines to him: the world as a necessary eternal animal;
the divinity of the heavenly animal.
129
Given also his political motives,
Averroes may well be attacking them more than Avicenna proper.
At any rate, clearly Averroes has the goal of making philosophy
respectable to Muslims. E.g., on miracles he says:
The ancient philosophers did not discuss the problem of miracles, since
according to them such things must not be examined and questioned, for
they are the principles of the religions, and the man who inquires into
them and doubts them deserves punishment, like the man who examines
the other general religious principles, such as whether God exists . . .
130

Given Aristotles (and Plato s!) contempt for popular religion, these
claims have little truth.
131
But they would serve to clear Aristotle of
heresy, as Averroes wanted. For, at the conclusion of his Tahfut, Averroes
says that Al-Ghazl had accused the philosophers of heresy, and now
he has cleared them of the charges.
132
Averroes then may clear philosophy of heresy at the cost of making
an Aristotle and an Avicenna ctif. Indeed he may have created one
or more Averroes ctifs too! For his views in his various works do not
agree.
133
So what to make of Averroes ? He did write reliable, literal com-
mentaries paraphrasing Aristotles texts and previous commentaries on
them. He also misrepresents Aristotles views, especially in his debate
129
Van den Bergh , Tahfut, p. 254 [421]. Averroes : Tahfut al-Tahfut, Simon van den
Bergh, trans. & comm. (London 1954). The numbers in brackets are page numbers
from the Arabic, Averroes, Tahafot at-tahafot, ed. Maurice Bouyges (Beirut 1930). Also
I consulted Beatrice Zedler , trans. & comm., Destructio Destructionum Philosophiae Algazels
(Milwaukee, 1961).
130
Van den Bergh , Tahfut, p. 315 [514]. Averroes approves of religion pragmati-
cally, because religion causes virtue in men. Cf. too p. 359 [5802].
131
Alfred Ivry , Towards a Unied View of Averroes Philosophy, The Philosophical
Forum 14.1 (1974), p. 108, says, It seems well nigh impossible for Averroes to modify
his position to accommodate the dogmas of religion or any particularist religion . . . in
Averroes view certain locutions [in the political sphere] are ones that are absurd out-
side that sphere. Ivry, pp. 10910, has the implausible view that Averroes is struggling
towards a view that the philosophical viewpoint is just as metaphorical and inadequate
as the religious one.
132
Van den Bergh , Tahfut, p. 362 [587].
133
Cf. George Hourani , Averroes on the Harmony of Religion and Philosophy (London
1961), pp. 4154; 77, who attributes two different views to Averroes.
70 allan bck
with Al-Ghazl , for the sake of making philosophy respectable in Islam.
In doing so he caricatures Avicenna so as to offer him as a scapegoat
for the devout to sacrice instead of philosophy entire. This caricature
may have distorted later views on Avicennas worth as a commentator
on Aristotle.
Conclusions
I have some sympathy for Hintikka s complaint:
. . . those who have not themselves wrestled with serious conceptual issues
tend to underestimate by orders of magnitude the extent to which a
great philosopher has to struggle against confusion, contradictions, and
other difculties.
134

One advantage medieval commentators have here over contemporary
ones, despite their linguistic and historical poverty, lies in their having
this same pragmatic interest: they seek to do Aristotelian science, not
merely talk about it.
What sort of commentator would Aristotle himself have like to have
had? Judging by how he treats his predecessors, including his friend and
teacher Plato , he would have wanted someone who would comment on
his writings in the way that he did upon Platos writings.
His own student and successor Theophrastus did so upon his writings.
Thus he disagreed with Aristotle on the validity of the mixed modal
Barbara syllogism. He challenged Aristotles account of the intellect
and his astronomy. Hardly a tractable student.
Now certainly, like Theophrastus , Avicenna ts Aristotles criteria for
a commentator far more than Averroes does. At times Avicenna defends
Aristotles entire doctrine. At other times, he will reject Aristotles teach-
ings almost completely.
Would Aristotle like this? Certainly he treated his own teacher and his
other predecessors thus. Yet we may distinguish choosing a successor
of a school and a tradition from selecting someone to comment upon
and preserve the original content of a member of that school. In this
way, commentators are more like curators and historians, and less like
those who wish to continue the tradition of the school.
134
Jaakko Hintikka , The Development of Aristotles Ideas of Scientic Method,
in Aristotles Philosophical Development, ed. W. Wians (Lanham, Md., 1996), p. 84.
Should Aristotle like this? Suppose Aristotle had been put in the
position of designing his own ideal commentator. Should he choose
one dedicated to preserving his legacy and prevent it from being cor-
rupted by later interpolations, corrections and emendations? Or one
who would take his results as starting points for future researchas is
indeed done in the sciences, which themselves have an Aristotelian base
in doctrine as well as in their organization? Let us have Aristotle playing
the Creator of The Commentator. In De libero arbitrio III.16, Augustine
argues that it is better for God to create beings with free will who may
well disobey His precepts and the path that He has set for them than
to create beings who will do the right, divinely prescribed thing without
having to make choices on their own. Now, just on those grounds alone,
Aristotle may well prefer commentators of independent mind. But in
his case, as his own views do not have the divine guarantee of being
the truth, he has the additional ground for preferring commentators
following the truth where it leads them, that his successors might make
better choices and get better results than he did.
So I submit that Aristotle would choose Avicenna , not Averroes , to
comment upon his works. Now we pedants may prefer the style of
paraphrase and footnote. And of course such a style has its uses for
those seeking to become familiar with a textfor students, be they
beginning or advanced. Yet this approach does not make the philoso-
phy being commented upon a living enterprise capable of growth and
progress, but rather an antique curiosity suitable for custodians and
gawkers. Averroes may rightly have been The Commentator for the
Latin medievals just beginning to study Aristotle. Yet, judging by the
citations of Avicenna at key points in discussions by Aquinas , Scotus ,
Ockham et al., Avicenna was The Commentator for these working,
creative philosophers. Perhaps we modernior post-modernialso should
take stock and reconsider what we want in a commentator.
ALBERTUS MAGNUS ON THE SUBJECT
OF ARISTOTLES CATEGORIES
Bruno Tremblay
1
If the natural and metaphysical works written by Albertus Magnus
(c. 12001280) have received attention from a relatively important
number of researchers, it is nevertheless the case that his logical writings
have been largely ignored by twentieth-century scholars.
2
One can only
hope that the publication of the critical edition of his logical commen-
taries, by the Albertus-Magnus-Institut, will help change this situation.
It seems to me that many reasons militate in favour of a closer and
deeper study of his commentaries-paraphrases on Aristotles Organon
and on the treatises that tradition has added to it.
3
It is too early to say
much about the precise degree of originality of his logical writings,
4

but we know that Albert the Great was among the very rst medievals
to produce a complete or nearly complete logical corpus, similar to
1
Thank you to David Zettel, Cynthia Tremblay , and Lloyd Newton , for their kind
and numerous suggestions concerning my use of the English language throughout this
paper. It goes without saying that all remaining awkward wordings are my responsibility.
2
The reasons for this fact are numerous and cannot be explained here. William
A. Wallace enumerates some of them on pp. 11 and 15 of his Albert the Greats
Inventive Logic: His Exposition on the Topics of Aristotle, American Catholic Philosophical
Quarterly 70 (1996), 1139.
3
Besides his commentaries on Aristotles Categories, Peri Hermeneias, Prior Analytics,
Posterior Analytics, Topics, and Sophistical Refutations, we have his commentaries on
Porphyry s Isagoge, on Boethius De divisione, and on the anonymous Liber sex principio-
rum. Some passages from his works might lead us to believe that he also commented
on one of Boethius opuscula on categorical syllogisms and that he wrote a book on
poetics, but these works, if they ever existed, are now lost. See Analytica posteriora 1.1.1,
ed. Auguste Borgnet (Paris, 1890), p. 4A, and Metaphysica 3.2.10, ed. Bernhard Geyer
(Mnster, 1960), p. 127, l.3435, and 5.5.5, p. 280, l.52.
4
The kind of general judgment passed by Sten Ebbesen on Albert s Logica seems
to me to be quite hasty, to say the least, considering how little the details have been
studied. See Sten Ebbesen, Albert (the Great?)s Companion to the Organon, in Albert
der Grosse. Seine Zeit, seine Werk, seine Wirkung, ed. Albert Zimmermann (Berlin, 1981),
pp. 89103. A very different view on Alberts logical commentaries can be found in
Ruth Meyer , Eine neue Perspektive im Geistleben des 13. Jahrhunderts: Pldoyer fr
eine Wrdigung der Organon-Kommentierung Alberts des Groen, in Geistesleben im
13. Jahrhundert, ed. Jan A. Aertsen and Andreas Speer, Miscellanea Mediaevalia 27
(Berlin, 2000), pp. 189201.
74 bruno tremblay
those which already existed among Arabic philosophers, and that he
actually played an extremely important role in the assimilation to the
West of Arabic thought, particularly in the eld of logic.
5
The time in
which Albert made his contribution to logic is also crucial: the writing
of his logical corpus (or at least most of it) probably took place in the
1250s,
6
when Albert was in Cologne after having left the University
of Paris, which situates this corpus just after those of Parisian masters
of arts like Johannes Pagus , Nicholas of Paris , and Robert Kilwardby ,
but before the large number of commentaries from the last third of
the thirteenth century and the beginning of the fourteenth. There is a
denite possibility that Alberts teachings in logic inuenced later com-
mentators,
7
and, combined with his general strength as a philosopher,
it warrants a closer look at his writings in this discipline.
Further, Albert s teachings on the Categories deserves even more atten-
tion, if only because of the importance that has been traditionally given
to this particular treatise within Aristotles thought, and because of the
very serious difculties that its interpretation poses, notwithstanding
Augustine s famous claim that, as a young man, he easily read and
understood it.
8
At the core of these difculties lies the precise identica-
tion of the subject of the book, and whether it is about things, concepts,
or words, or all three. This problem is signicant, both in itself, and
also because of the bearing it has on questions like the method of the
treatise, its division, or the part of philosophy it belongs to. The dif-
culty and signicance of this problem were seen by medieval thinkers,
and in particular by those of the thirteenth century, which of course
does not mean that they were any more unanimous about its solution
than we modern readers of Aristotle are. This presentation is obviously
not the right place for an overall comparison of the different views
on the subject of the Categories that dominated the thirteenth century,
5
Assimilation and inuence that are especially evident in Albert s general introduc-
tion to logic. See Bruno Tremblay , Albert le Grand: De ce qui vient avant la logique,
History and Philosophy of Logic 25 (2004), 165203.
6
Hugo Stehkmper , ed. Albertus Magnus: Ausstellung zum 700. Todestag (Cologne,
198081), pp. 135136. See also the editors introduction to Albert s Super Porphyrium
De V universalibus, ed. Manuel Santos Noya (Mnster, 2004), pp. vivii.
7
I am thinking here about commentators from the second half of the thirteenth
century, but the same can be said about fourteenth- and fteenth-century thinkers. See
for instance E. P. Bos , John Versor s Albertism in his Commentaries on Porphyry and
the Categories, in Chemins de la pense mdivale. tudes offertes Znon Kaluza, ed. Paul J. J.
M. Bakker , Textes et tudes du Moyen ge 20 (Turnhout, 2002), 4778.
8
Augustine , Confessions, 4.16.28, ed. Pierre de Labriolle (Paris, 1950), p. 86.
albertus magnus: on the subject of a.
s CATEGORIES 75
but clearly the position thereon of a giant like Albertus Magnus must
be better known before we can understand the real nature and value
of that centurys contribution to the comprehension of this intriguing
Aristotelian treatise.
The Short Denition of the Subject Found in the
Commentary on the Categories
Like most medieval works of this kind, Albertus Magnus commentary
on the Categories is preceded by a proem in which some very important
questions are briey addressed. They are important because they con-
cern the very foundations of the science to be taught, and the treat-
ment thereof is necessarily brief because it belongs to an introduction
and because a more in-depth consideration would presuppose a level
of knowledge that the beginner, for whom the commentary is written,
does not yet have. Here is how Albert the Great briey describes, in the
proem of his commentary, the subject of the Categories: the subject is
what can be ordered into a relation of predicability or subjectibility,
inasmuch as it stands under the vocal sound that signies this order,
est enim subiectum ordinabile in rationem praedicabilis vel subicibilis, secundum
quod stat sub voce talem ordinem signicante.
9
This denition of the subject, at rst sight and considered by itself, is
not extremely helpful. First, it seems to be very close to a widely used
and almost standard formulation of the time, traces and variations of
which can be found in other commentaries and even in student manu-
als.
10
It remains to be seen how much such a commonly used expression
9
De praedicamentis 1.1, eds. Carlos Steel, Silvia Donati and Manuel Santos Noya
(Mnster, 2008), p. 2, l.79. I am using the (almost complete) rst draft of the forthcom-
ing critical edition, which C. Steel was kind enough to let me consult. When I refer to
passages beyond 7.3, however, I use the Auguste Borgnet edition (Paris, 1890).
10
This is true of texts written both before and after Albert s commentary. See for
example Ezio Franceschini , Giovanni Pago: Le sue Rationes super Praedicamenta Aristotelis e
la loro posizione nel movimento aristotelico del secolo XIII, Sophia 2 (1934), 172182,
p. 176: dicibile incomplexum ordinabile in genere vel ordinatum; E. J. Ashworth , Lanalogie
de ltre et les homonymes: Catgories, 1 dans le Guide de ltudiant, pp. 287288,
in Lenseignement de la philosophie au XIII
e
sicle. Autour du Guide de ltudiant du ms. Ripoll
109, eds. Claude Laeur and Joanne Carrier (Turnhout, 1997), pp. 281295: dicibile
incomplexum ordinabile in genere; Robert Andrews , Petrus de Alvernia, Quaestiones super
Praedicamentis: An Edition, Cahiers de lInstitut du Moyen-ge grec et latin 55 (1987), 384,
76 bruno tremblay
can tell the reader about Albert s own views on the question. Also, were
this denition specic to Albertus Magnus, there would still remain the
problem of its ambiguity. What, exactly, can be ordered? Is it things,
concepts, or words? The fact that modern scholars do not agree on
this is probably the best sign of the ambiguity of the denition: E.
J. Ashworth ,
11
pointing to the second half of Alberts denition and
also to his explicit mention of Boethius well-known description of
the subject of the Categories (the ten vocal sounds that signify the rst
genera of things),
12
thinks Albert has rst and foremost words in mind;
Alessandro Conti ,
13
on the basis of the great similarity between Alberts
denition and those of other contemporary thinkers who often explicitly
mention ens, believes it is things; Giorgio Pini
14
formally rejects a more
conceptual interpretation because Albert describes the subject of the
Categories in terms of predication rather than of second intention, and
because this expression is nowhere to be found in this philosophers
general introduction to logic, even though Pini identies some pas-
sages and distinctions which might actually militate in favour of such
an interpretation. In relation to this problem, what does the last part
of the denitioninasmuch as it stands under the vocal sound that
signies this ordermean? Does it refer to some kind of grammatical
or linguistic consideration of the Categories? Moreover, how does this
denition of the subject relate to the ten supreme genera themselves?
p. 10: ens incomplexum secundum quod ordinabile in genere; Carmelo Ottaviano , Le
Quaestiones super libro Praedicamentorum di Simone di Faversham, Atti della R.
Accademia Nazionale dei Lincei. Memorie della Classe di Scienze morali, storiche e lologiche 3 (1930),
258351, p. 266: ens dicibile incomplexum ordinabile in genere secundum sub et supra. For other
authors, see P. O. Lewry , Robert Kilwardby s Writings on the Logica Vetus, Studied With Regard
to Their Teaching (doctoral thesis, Oxford, 1978), pp. 91 and 265266. These descriptions
of the subject of the Categories are not necessarily and always the only ones used by the
authors who include them in their writings. Robert Kilwardby, Notulae super librum Praedi-
camentorum, proemium, for instance, refers to that subject as ordinabile and dicibile after
formally treating the question of the subject in very Boethian terms. (Thank you to
Alessandro Conti for letting me see the rst draft of the new edition he is preparing.)
11
Lanalogie de ltre et les homonymes, p. 289. This view is somewhat sup-
ported in William E. McMahon , Albert the Great on the Semantics of the Categories
of Substance, Quantity, and Quality, Historiographia Linguistica 7 (1980), 145157,
although in a strongly semantical perspective.
12
De praedicamentis 1.1, p. 2, l.12. See Boethius , In Cat. 1, ed. J. P. Migne (Paris,
1891), pp. 159D160B. A better edition of this passage is provided in Monika Asztalos ,
Boethius as a Transmitter of Greek Logic to the Latin West, Harvard Studies in Classical
Philology 95 (1993), 367407.
13
Alessandro Conti , Thomas Sutton s Commentary on the Categories according to
MS Oxford, Merton College 289, p. 175, in The Rise of British Logic, ed. P. O. Lewry
(Toronto, 1983), pp. 173213.
14
Giorgio Pini , Categories and Logic in Duns Scotus (Leiden, 2002), pp. 2627.
s CATEGORIES 77
Are they the subject of the treatise? If not, what are they? Finally, as
a last example of a question that this denition raises: how does this
denition allow us to distinguish the subject of the Categories from the
subject of the Isagoge? Isnt Porphyry s short treatise also concerned, at
least partially, with something that can be ordered within a genus or
a species?
I shall address the rst problem, i.e., the nature of that ordinabile that
constitutes the subject of the Categories, and in so doing I shall have
the opportunity to touch upon the other problems briey. As a more
complete picture of Albert s views on what the Categories is concerned
with is drawn, it will become clear that Ashworth and Conti were
both right in saying, respectively, that words (voces) and things (entia)
have to be part of the solution, but unlike them I shall argue that the
emphasis needs to be put above all on concepts (intentiones). In order
to show this, I shall rst state a few things about Alberts conception
of logic, and then establish a series of fundamental (and mostly meta-
physical) distinctions that are absolutely necessary to reach a deeper
and better understanding of what Albert has in mind when he writes
his short denition of the subject of the Categories, a denition which
I shall revisit toward the end of this paper. In so doing, I shall rely
upon Alberts De praedicamentis and also upon some of his other works,
whether they belong to logic or not. Among these, Alberts commentary
on the Isagoge, which was likely written right before the commentary
on the Categories and which begins with a general introduction to logic,
deserves special attention.
Albert s Conception of Logic
A few scholarly publications over the last 40 years have tackled the
question of Albert the Greats general views on logic, but usually in a
very brief and partial way, which probably explains why they often reach
contradictory conclusions.
15
The question of how Albert conceives
15
See for example Mauricio Beuchot , La naturaleza de la lgica y su conexin
con la ontologa en Alberto Magno, Dinoia 33 (1987), 235246; Eladio Chvarri ,
El orden de los escritos lgicos de Aristteles segn San Alberto Magno, Estudios
loscos 9 (1960), 97134; Ralph McInerny , Albert on Universals, Southwestern Journal
of Philosophy 10 (1979), 318; Giorgio Pini , Categories and Logic in Duns Scotus , pp. 2425;
Richard F. Washell , Logic, Language, and Albert the Great, Journal of the History of
Ideas 34 (1973), 445450.
78 bruno tremblay
of logic and its subject is complex and would certainly deserve a
separate study, but for now I would like to emphasize two important
points, each of which shows a strong Avicennian inuence on Alberts
logical views.
First, logics practical goal is to help the human mind go from the
known to the unknown.
16
Throughout his works Albert constantly
reminds his reader of this, and although it might sound like empty
and incomprehensible jargon, it actually states something quite simple
and easy to understand. It refers to the fact that the usual mode of
our intelligence is not intuition, but rational process: we usually come
to know things based on other things we already know, be it when we
draw a conclusion based on some premises we already accept or when
we try to understand the nature of a thing based on ideas already pos-
sessed and arranged in a denition.
17
Logic studies this rational mode
or process and the instruments that allow us to complete it, and thus
provides guidance to reason in all philosophical elds.
18
Second, Albert has a very intentionalist view on what the science of
logic studies.
19
He accepts the Avicennian position
20
that the science of
logic considers properties that belong to things as we know them, and
as we use them to know other things. These properties have a purely
intentional being, and are necessarily produced by reason as it goes
from the known to the unknown.
21
Although Albert, much like Avicenna
himself, rarely uses the expression second intention, he does use it
at the beginning and at the end of his careerin his commentaries
on Peter Lombard s Sentences
22
and on Aristotles Metaphysics
23
, and
certainly the doctrine that underlies this expression, taken directly from
Avicennas Logyca, is clearly and explicitly presented by Albert in his
16
De praedicamentis, 1.1, p. 1, l.621.
17
Super Porphyrium De V universalibus 1.1.5, p. 8, l.2853.
18
Super Porphyrium De V universalibus 1.1.1, 5 and 7. See Avicenna , Logyca (Venise, 1508),
fol. 2v, and Bruno Tremblay, Ncessit, rle et nature de lart logique, selon Albert le
Grand, Bochumer philosophisches Jahrbuch fr Antike und Mittelalter 12 (2007).
19
This is a claim that has rightly been made by Norman Kretzmann , although he
has provided no detailed justication for it. See his William of Sherwoods Introduction to
Logic (St. Paul, 1966), p. 21, n.1.
20
Avicenna , Logyca, fol. 2rb. See also Michael E. Marmura , Avicenna on the
Division of the Sciences in the Isagoge of His Shifa, Journal of the History of Arabic
Science 4 (1980), 239251.
21
Super Porphyrium De V universalibus 1.1.2.
22
Super libros sententiarum 1.13.A.1, ad 1, ed. Auguste Borgnet (Paris, 1893), p. 370B.
23
Metaphysica 1.1.1, ed. Bernhard Geyer (Mnster, 1960), p. 3, l.812. See Avicenna , Liber
de philosophia prima sive scientia divina 1.2, ed. S. van Riet (Louvain, 1977), p. 10, l.7375.
s CATEGORIES 79
general introduction to logic.
24
More sophisticated denitions of second
intentions, such as those which can be found in Thomas Aquinas or
Duns Scotus would, of course, come only later.
25
The Categories Is Concerned with Universals Post Rem
Since according to Albert universals are what the mind orders within
a genus or a species,
26
one must conclude that for him the Categories,
whose subject is an ordinabile, is concerned with universals. Yet, saying
this does not take us very far; in fact, it could even be like opening a
Pandoras box, since Alberts doctrine of universals rests on complex
metaphysical distinctions, and in fact has the reputation of being marked
with confusion and contradictions.
27
Although I cannot here engage in
a serious analysis of this difcult doctrine, I will endeavor to state a
few of its basic elements.
The rst thing that one should keep in mind when trying to decipher
Albert s teachings on universals is that for him the word universal
is a very equivocal word that can refer to a considerable number of
meanings, which, although linked to each other by analogy, can be
distinguished.
28
In everything that Albert labels a universal, one
can always nd, one way or another, something one in relation to
some multiplicity,
29
but the manner in which one, relation or
multiplicity is to be understood is not always the same, and if it is,
the manner in which the things named universals relate to them
24
The close link between the doctrine laid out in the beginning of Avicenna s Logyca
and the use of the expression second intentions in his Metaphysica (see supra, note
23) must have been seen by Albert just as it is easily understood by modern readers of
Avicenna. See for example A. I. Sabra , Avicenna on the Subject Matter of Logic,
The Journal of Philosophy 77 (1980), 746764.
25
See Giorgio Pini , Categories and Logic in Duns Scotus , chapters 2, 3, and 4.
26
De sex principiis 1.1, ed. Ruth Meyer (Mnster, 2006), p. 4, l.3841.
27
Ralph McInerny , Albert on Universals, pp. 1718.
28
The way Albert begins his most systematic review of the different kinds of uni-
versals clearly indicates that he does so in the context of analogy of words: Dicimus
igitur universale quattuor modis dici proprie et hoc, quocumque modo dicatur, tribus modis vel quattuor
considerari. (Metaphysica 5.6.5, p. 285, l.1618)
29
Metaphysica 5.6.5, p. 285, l.8486. Albert s choice of words in this passage also
shows that logical analogy, and not univocation, is at stake here.
80 bruno tremblay
changes.
30
Some of the meanings attached by Albertus Magnus to the
word universal are laid out in the famous Neoplatonist division of
the kinds of universals that he found in Eustratus
31
and Avicenna ,
32

the one into universals ante rem, in re, and post rem. This division is
ubiquitous in Alberts works, and it makes sense that the scholars who
have treated the question put emphasis on it.
33
To put it very simply,
universals ante rem are the eternal divine ideas. They are imagined by
us as being many but in fact are one and can somehow be identied
with God.
34
They are exemplary ideas, acting as an origin and a cause
in relation to concrete things that surround us,
35
through an elaborate
causal process which cannot be described here. They can be said to
be universals in the sense that as exemplars they precontain ( praeha-
bentes) and somehow cause a multiplicity of things.
36
Universals ante rem
are, thus, the exemplary origin of the actualized natures or forms of
natural things. These latter forms, or universals in re, exist in the things
that surround us and make natural, material, concrete created things
what they are. A particularized form or nature is universal in the sense
that from it other particularized things can be generated
37
a human
being generates other human beings
38
and in the sense that despite
30
In other words, I think that behind the multiple uses of the word universal
Albert would see cases of both analogies secundum communam proportionem and ad unum.
For a brief consideration of Alberts doctrine of logical analogy, see Bruno Tremblay ,
A First Glance at Albert the Greats Teachings on Analogy of Words, Medieval
Philosophy and Theology 5 (1996), 265292.
31
In primum Aristotelis moralium ad Nicomachum, 1096a1014, ed. H. Paul F. Mercken
(Leiden, 1973), p. 69, l.85p. 70, l.29.
32
Logyca, fol. 2rb and 12rava.
33
See among others Georg Wieland , Untersuchungen zum Seinbergriff im Metaphysikkommentar
Alberts des Groen (Mnster: 1972), pp. 4146; Sophie Wlodek , Albert le Grand et les
Albertistes du XV
e
sicle: le problme des universaux, in Albert der Grosse. Seine Zeit,
seine Werk, seine Wirkung, Band 2, ed. Albert Zimmermann , Miscellanea Mediaevalia 14
(Berlin, 1981), pp. 193207; Alain de Libera , Thorie des universaux et ralisme
logique chez Albert le Grand, Revue des sciences philosophiques et thologiques 65 (1981),
5573; id., Albert le Grand et la philosophie (Paris, 1990), pp. 180213; id., La querelle des
universaux. De Platon la n du Moyen ge (Paris, 1996), pp. 245262; Thrse Bonin ,
Creation as Emanation. The Origin of Diversity in Albert the Greats On the Causes and the
Procession of the Universe (Notre Dame, 2001), pp. 4150.
34
De causis et processu universitatis 2.1.20, ed. Winfried Fauser (Mnster, 1993), p. 85,
l.3944; Super Dionysium De divinis nominibus 4.84, ed. Paul Simon (Mnster, 1972), p. 190,
l.65p. 191, l.20; 5.16, p. 312, l.515; De praedicamentis 7.9, p. 290A.
35
Super Dionysium De divinis nominibus 2.8.3, p. 97, l.2029.
36
Super Porphyrium De V universalibus 2.3, p. 23, l.2638, and Metaphysica 5.6.5,
p. 285, l.5977.
37
Metaphysica 5.6.5, p. 285, l.8792.
38
De sex principiis 1.5, p. 58, p. 11, l.29p. 12, l.11.
s CATEGORIES 81
its particularization it keeps its real capacity of being intellectually
abstracted from matter and of being grasped in its unity in relation to
the actual or at least potential multiplicity of things in which it nds
itself individualized.
39
Finally, one can also call universal something
that is posterior to natural things: the one common notion thanks to
which a non-causative mind grasps a multiplicity of similar individual
things.
40
It is a universal post rem, which exists as a likeness of things in
the human or rational mind, and not in the extra-mental natural world
or in a divine or purely intellectual mind.
Before moving ahead in the description of the logical universal and
of the subject of the Categories, one should take note of the special
effort often made by Albertus Magnus to link universality post rem to
prior universalities, whether they be in re or ante rem, and to insist on the
underlying unity that exists among them all and that has its source
in divine causality.
41
He says very explicitly that universality exists
outside of the human mind, in natura rerum, a way of speaking that
might surprise someone who is used to the more Averroistic approach
and vocabulary that can be found in Aquinas .
42
It is possible that this
insistence is to be explained by the fact that Albert still remembers and
fears the nominalist ghosts of the twelfth century,
43
and that he sees the
danger that would be implied by afrming without any nuance and too
abruptly that universals or universality are pure creations of the human
mind, a preoccupation that Aquinas might not have had to the same
degree.
44
It must also be repeated that for Albert the logical universal
and the other universals are not universals in exactly the same sense,
and that when he mentions
45
that universality exists in natura rerum, he
39
Super Porphyrium De V universalibus 2.3, p. 24, l.3772.
40
Metaphysica 5.6.5, p. 285, l.9396.
41
Super Porphyrium De V universalibus 2.6, p. 34, l.57p. 35, l.20.
42
See for example De ente et essentia 3, ed. Hyacinthe F. Dondaine (Rome, 1976),
l.99101.
43
Super Porphyrium De V universalibus 9.3, p. 146, l.120, and 2.3. See also Manuel
Santos Noya , Die Universalienlehre der Nominales in der Darstellung Alberts
des Groen, in Albertus Magnus. Zum Gedenken nach 800 Jahren: Neue Zugnge, Aspekte
und Perspektiven, ed. Walter Senner, Quellen und Forschungen zur Geschichte des
Dominikanerordens 10 (Berlin, 2001), pp. 171194.
44
Which is not to say that Aquinas himself never refers to the ante rem / in re / post
rem division or that he never speaks of the extra-mental nature as something that has
in itself the property of being communicable to many. See for example Scriptum super
libros sententiarum 2.3.3.2, ad 1, ed. R. P. Mandonnet (Paris, 1929), and Expositio libri
peryermenias 1.10, ed. Ren-Antoine Gauthier (Paris, 1989), l.8789.
45
Including in logic. See Super Porphyrium De V universalibus 2.3, p. 24, l.49p. 25, l.6.
82 bruno tremblay
means that realities analogous to logical universals exist outside of our
mind and can also rightly be called universalsother universals on
which emphasis is placed since they are presupposed by logical univer-
sals and are their foundation.
46
I think that the best proof that Albert
is thinking about universals analogously and not univocally is that for
him predicability is a necessary property of the logical universal,
47

whereas other universals are not, strictly speaking, predicable.
48
Which,
of course, means that the Categories deals directly with universals post
rem, and not in re or ante rem.
The Categories Is Concerned with Notions and
Accidents of Things
The opportunity to afrm that this predicability is proper, strictly
speaking, to the universal post rem presents itself when Albertus Magnus
determines what meaning must be given to the word substance when
it is said that logic, and more specically the Categories, studies substance.
(The distinctions he then makes are quite important because many of
them apply to all the other remaining categories.) After identifying three
main senses of the word, he goes on to say that the substance that logic
is interested in is something to which all things that are substances can
be reduced through predication: whatever a dog, a ower, an animal,
a plant, or a man are, each of them can be said to be a substance,
which plays the role of a common, unifying predicate or predicable.
Now this substance can be such a predicable only because it is the form
of the whole subject of which it is saidwhich ultimately means the
individual substance.
49
This characterization of the logical universal as
46
Metaphysica 5.6.6. It seems that some of Albert s followers from the fteenth century,
for instance Heymeric de Campo , also ascribed some sort of analogical character to
the underlying unity that links the different types of universals. See Sophie Wlodek ,
Albert le Grand et les Albertistes, pp. 1920.
47
48
See, among others, De praedicamentis 2.1 and 3, p. 24, l.1321, and Super Dionysium
De divinis nominibus, 2.83, p. 97, l.2029. According to De causis et processu universitatis
2.1.5, p. 65, l.4749, the classic denition of the logical universal, unum in multis et de
multis, may be applied to other universals, but only secundum quid, again an expression
that clearly shows that Albert s treatment of the problem of the universals rests on
analogical uses of words.
49
De praedicamentis 2.1, p. 20, l.4151.
s CATEGORIES 83
forma totius is often repeated throughout Albert s works,
50
among others
in a commentary that must not be neglected in the interpretation of
the one on the Categories: the commentary on the Liber sex principiorum,
an anonymous treatise that we know to have been written in the twelfth
century, supposedly to compensate for Aristotles extreme concision in
the treatment of the last six categories.
51

The unknown author of this work holds the rather enigmatic view
that the six categories that constitute the subject of the book are forms
and that each of them is contingent to composition, compositioni con-
tingens.
52
This difcult statement is not really explained by its author,
but Albert s interpretation of it does provide some more insight into his
conception of what a treatise like the Categories or the Liber sex principiorum
is about. From a logical point of view, Albert writes,
53
forms like the
categories are intentions or notions, representations that exist in the
human mind and that would not exist without it. Our mind grasps real
things through determinations or forms that exist in themwhether
these determinations be substantial or accidental, necessary or contin-
gent, etc.and through this multiplicity of determinations or forms
it fashions within itself a multiplicity of concepts or intentions, all of
which allow it, although in diverse ways, to apprehend a part of reality.
When the mind relates one of its notions or concepts to the external
world in order to know the latter, it does so through predication and
the use of the copula is, which afrms a kind of unity or identity
between a subject and a form that is predicated of it.
54
If I am right
in thinking and saying that Plato is an animal, then it means that
through my notion of animal I truly conceive the whole of Plato and
not just part of himotherwise I could not identify both Plato and
my concept of animal in a sentence. This identity cannot, obviously,
be without qualication, since different conceptualized forms are truly
50
See for example Super Dionysium De divinis nominibus 2.85, p. 99, l.1318, in which
the universals ante rem, in re and post rem are named and distinguished after three kinds
of forms: forma exemplaris, forma partis and forma totius.
51
Albert himself (De sex principiis 1.1, p. 1, l.1420) attributes this work to Gilbert
of Poitiers , and this seems to have inuenced other authors to think the same. See
P. O. Lewry , The Liber sex principiorum, A Supposedly Porretanean Work. A Study in
Ascription, in Gilbert de Poitiers et ses contemporains. Aux origines de la Logica modernorum,
eds. Jean Jolivet and Alain de Libera (Naples, 1987), pp. 251252.
52
Anonymi fragmentum vulgo vocatum Liber sex principiorum 1.3, ed. Laurent Minio-
Paluello (Bruges, 1966), p. 35, l.34.
53
De sex principiis 1.2, p. 4, l.929.
54
84 bruno tremblay
identied with a same subjectPlato is at the same time truly said to
be a man, an animal, a philosopher, and a Greek, even though the
words that signify those conceptualized forms are not synonyms.
55
It
is possible for such forms to be predicable or identied with a sub-
ject through predication only because they are the form of the whole
subject, although not necessarily in actuality. When I say that Plato
is an animal, not only do I explicitly and actually attribute animality
to Plato, but I also conceive and leave open the possibility that he is
something else: rational or non-rational, a man or a horse, white or
black, etc. This is the difference between, say, animal or white, which
can be predicated of Plato, and animality and whiteness, which afrm
the forms they signify in some kind of absolute sense that excludes any
other form or determination: using the concept or word of animal-
ity, I mean the form of animality alone and I imply no possibility
of anything else, whatever that may be. Animality and whiteness
therefore cannot be said of Plato in a true sentence, since it would
be false to identify Plato with animality or whiteness, which are just
parts of what he is.
56
The Categories (as well as the Liber sex principiorum),
in short, is concerned with the universal concepts or representations
thanks to which we grasp the whole of things or subjects and which
thus can be said of them.
Albertus Magnus also comments on the second part of the Liber sex
principiorums denition, which says that the kind of form this work is
concerned with is compositioni contingens. At rst sight, this could lead
one to believe that these forms are notions that represent mentally only
contingent forms of material beings, but Albert interprets it differently.
His explanation is especially interesting because it links this question
to the distinction of the three kinds of universals:
I say therefore that it [i.e., the kind of form studied in the Liber sex
principiorum] is contingent because the way of receiving itas the notion
of the wholeis accidental and always contingent. For there is a universal
that comes before the thing, and this one is the per se or the accidental
cause of the thing; there is also a universal that is in the thing, and this
one is the substance of the thing or an accident disposed toward its
55
De praedicamentis 2.8, p. 37, l.57p. 38, l.36.
56
Super Porphyrium De V universalibus 3.3, p. 47, l.60p. 48, l.6. See also E. P. Bos and
A. C. van der Helm , The Division of Being over the Categories According to Albert the
Great, Thomas Aquinas and John Duns Scotus , in John Duns Scotus: Renewal of Philosophy.
Acts of the Third Symposium Organized by the Dutch Society Medium Aevum (May 23 and 24,
1996), ed. E. P. Bos (Amsterdam, 1998), 183196, p. 187. These are all distinctions
that Thomas Aquinas would make famous in his De ente et essentia 3, l.245308, a work
likely written at the same time as Albert was working on his logical commentaries.
s CATEGORIES 85
substance; and there is a universal that comes after the thing, and this
one is the universal separated from the thing itself by the intellect, and
this happens to the thing and is accidental to the thing. This, indeed, is
nothing but the apprehension of the universal or the form without this or
that [particular], and without the now, which is the apprehension, simple
and separated from the particular and what makes [things] particular, of
the thing. It is in this [third] sense that the form that the intellect orders
within a genus and a species and whose coordination is reduced to a
predicament or a principle was said to be a universal.
57

The forms that the Categories and the Liber sex principiorum are interested
in are universals post rem and as such are accidents of things that sur-
round us. The notions or concepts that our intellect fashions are not
accidental to things in the sense that they all mentally represent real
accidents of things, but in the sense that the formation and the very
existence of these notions or concepts is not part of the essential nature
of things and is in fact contingent: it happens or does not happen to
things that a human mind knows them, and whether or not this happens
changes nothing to what they are fundamentally.
58
This explains why
for Albertus Magnus both treatises belong to logic, since their subjects
are included in what logic in general is concerned with, as Avicenna
and Albert himself in his introduction to logic had dened it: both deal
with accidents of things that happen to them as we acquire notions of
them and as these become part of a rational process of knowing.
Albert explicitly mentions logical universality and its five main
modes as described in the Isagoge as examples of such accidents. Being
a genus, for example, is a relation that happens to exist in the intel-
lectually abstracted notion of animal as we compare it to man and
horse.
59
Our intellect then grasps animal as something one having a
similar relation to many things, and conceived as forma totius, animal,
unlike animality, can be said of man and horse in order to know and
express what these essentially are.
The Categories Is Concerned with The Principles of Things
Now if the genus is the rst kind of logical universal, as Albert states
in his commentary on the Isagoge,
60
and if the Categories is said to have
57
De sex principiis 1.2, p. 4, l.2741.
58
De natura et origine animae 1.2, ed. Bernhard Geyer (Mnster, 1955), p. 4, l.8992.
59
60
86 bruno tremblay
something to do with genera and especially the ten supreme genera,
then a question arises. Porphyry says in his Isagoge
61
that genera are
principles of things: does this mean that the Categories studies principles
of things? The passage from Alberts De sex principiis which I quoted
above and in which predicament and principle seem to be consid-
ered as synonymous words goes in the same direction.
However, Albert explicitly rejects this possibility in his commentary
on the Categories. Universals, substance for example, can indeed be seen
as principles of being or formal causes in relation to concrete things
that surround usthese are what they are because of their forms or
naturesbut that is not how universals are considered in a logical
treatise like the Categories. Other parts of philosophy study the natures
of things in their extra-mental activity as principles of being; in other
words, they study universals as founded in natures or forms that are
independent of the human mind.
62
But logic, says Albert, considers
them as predicable, that is to say in as much as natures, when grasped
by our intellect and used to grasp other natures to which it relates in
the same way,
63
take on the property of logical universality, which can
be recognized by the fact that these natures are predicable. These are
not principles of being: they only exist in the human mind.
64

Now even if strictly speaking logic does not consider universals post
rem as principles of being (since, as principles of being outside of the
human mind, natures or forms are not even species or genera yet),
65

Albert still thinks that the formal causality that a nature possesses in
the extra-mental world is somehow mirrored in the notion that the
human intellect has of this nature. A genus or a species, for example,
can be said in logic to be a principle of being of an individual thing in
the limited sense that it is the intention or mental representation of its
principle of being. Commenting on a famous passage from chapter 5
of the Categories, Albert writes that it is only in this sense that secondary
substances (e.g., the universal concept of animal or man) could be said
to be ontologically prior to primary substances (e.g., Socrates ).
66
61
Porphyrii Isagoge 2, ed. Adolf Busse (Berlin, 1887), p. 2, l.1213.
62
63
Analytica posteriora 1.2.3, p. 28A.
64
De sex principiis 1.2, p. 4, l.948, and p. 5, l.4353.
65
66
s CATEGORIES 87
But there is another way in which the Categories can be said to study
principles of things, and this way is more important because it lies at
the core of Albert s conception of logic. Rational discourse is opposed
to intuition and consists in knowing things step by step, in knowing some
aspect of reality and then using it to know another one. As mentioned
earlier, Albert likes to describe logic as a study of this passage from the
known to the unknown and of the means to complete this movement,
because it is through this study that reason can obtain a better knowledge
of its own process and thus perform it better. This practical goal of
logic is closely related to the reason the Categories considers universals post
rem and not the prior universals that exist outside of the human mind
and that are the foundations of the former. Here is the key distinction,
made by Albert at the outset of his logical corpus: the known, thanks
to which the knowledge of the unknown is acquired, can be considered
in two ways: inasmuch as it is a thing outside of the soul of the knower
or inasmuch as it is a certain notion in the soul of the knower. Now,
[the known] does not make one know the unknown inasmuch as it is
a thing outside of the soul of the knower, but rather inasmuch as the
notion of the thing exists in the soul of the knower. Indeed, it is thus
that it makes known and sheds light on the unknown.
67
It is only once it is grasped or conceptualized by the human mind that
a thing can immediately play the role of an instrumenta beginning, a
point of departure, a principlein relation to the unknown thing that
one is trying to know rationally. Things, once they are known through
abstraction, can acquire properties that, although accidental to the real
natures of things, make rational discourse possible and are therefore
at the center of interest for logic. First among these logical properties
is universality, a notion which within logic itself is analogical:
68
a form
or nature, once intellectually abstracted from the particular conditions
in which it exists, can be compared to its subjects and then perhaps
relates to them as a genus, thus allowing us to know partially what
these subjects essentially are; or as a species, thus allowing us to know
the complete essence of these subjects; and so on for the three other
modes of universality, i.e., specic difference, property, and accident.
On a more general level, once a nature or form has been intellectually
67
68
88 bruno tremblay
and immaterially grasped by the human mind, it is ready to take on
the properties that are at the core of logics consideration: depending
on how reason uses it and what it compares it to, it can become a
genus, part of a denition, a middle term, etc. It can, in other words,
become a principle of knowing in relation to some unknown. This ts
well with Albert s general description of logic or rational science, as
opposed to sciences that study entia naturae: rational sciences are dif-
ferent from natural sciences, because rational sciences consider things
through their principles of knowing whereas natural sciences consider
them through their principles of being.
69
The Categories can be said to be about principles of knowing in
more than one way. First, and quite simply, because the book is about
universals, and universals are principles of knowing that of which they
are said, by making it known through its essence or its accidents.
70
But
since our intellect cannot have a distinct knowledge of the whole of
reality through one single concept
71
and must in fact multiply them, the
question of the order of these multiple universals poses itself right at
the outset of logic.
72
This means that the attribution of a universal form
or nature to some subject through predication necessarily involves other
logical relations that are either already known, or that this predication
will somehow help us know: when placing dog in the genus animal,
one knows or one is becoming ready to know that a dog is a living
being, since living being, as a more universal genus, comprehends the
universal animal.
73
At the end of the day, such observations lead Albert
to hold the view that the Categories is especially concerned with the ten
fundamental, univocal
74
universals, i.e., the ten supreme genera. They
are, indeed, principles of knowledge like any other universals, but
they are also principles of knowledge specically as supreme genera,
because of their role in the ordering of the multiplicity of universals
and therefore predicables that exist in the human mind, which ordering
69
De homine 60, sol., ed. Auguste Borgnet (Paris, 1896), p. 517A.
70
Metaphysica 5.2.11, p. 248, l.6769. Like any other intentions or mental representa-
tions, universal intentions make known the things of which they are the intentions. See
De Anima 2.3.4, ed. Clemens Stroick (Mnster, 1968), p. 102, l.2836.
71
De bono 5.1.3, sol., eds. Heinrich Khle et al. (Mnster, 1951), p. 274, l.3031.
72
73
74
This precision, i.e., univocation, is of a certain importance, inasmuch as the ana-
logical notions of being and one, for instance, can also be seen as kinds of beginnings
or principles of knowledge. See De praedicamentis 1.7, p. 17, l.3948.
s CATEGORIES 89
Albert usually calls ordinatio praedicabilium. The ten supreme genera are
in fact principles in relation to all ordinabiles, and through the study of
the ten genera the logician in some way reaches and treats the whole
coordination of ordinabiles.
75

Does this distinction between principles of being and knowing cut
off logic, and more specically the Categories, from reality? Does it turn
this treatise into a study that is completely isolated from the external,
extra-mental world, into a consideration of concepts that would not
and could not also refer to things? Obviously not. In order to be at the
root of predications that are true, the known natures or forms that are
ordered within the categories must correspond to something somehow
real, and while the logical relations and order that exist among predi-
cables are a product of the human intellect
76
a product that depends
on the mode of knowing of the latter, i.e., the necessity for the intellect
to go through abstraction, to multiply concepts, to order them so as
to be able to know and speak the truth, etc., the remote founda-
tion remains extra-mental reality. In other words, even though logical
generality and specieity exist only because of the rational activity
of our intellect, the truth of the subordination of dog to the genus
animal, expressed through a predication like a dog is an animal, is
directly measured by the external world: extra-mental reality, and not
the intellect, is the reason why this statement is true.
77
There exists at least one other reason why for Albertus Magnus forms
or things, and not only the accidents or logical properties that their
mental representations or intentiones take on as they are compared to
one another in order to go from the known to the unknown, must in
some way be present in the Categories. I mentioned earlier that Albertus
Magnus accepts the Avicennian doctrine of second intentions and that
he sometimes also uses this vocabulary. In his commentary on the
Metaphysics, he repeats Avicenna s dictum that logic considers second
intentions established in things (. . .), thanks to which one obtains ways
to go from the known to the unknown.
78
This might well mean that
logical entities (i.e., the accidents or more precisely the relations that
75
De praedicamentis 1.3, p. 10, l.1016, and De sex principiis 1.1, p. 1, l.2444, and 2.6,
p. 28, l.1825.
76
De praedicamentis 1.1, p. 1, l.39p. 2, l.5.
77
De praedicamentis 1.5, p. 13, l.4651, and 7.12, p. 296B, and De Anima 3.3.2, p.
210, l.1820.
78
Metaphysica 1.1.1, p. 3, l.911. See Avicenna , Liber de philosophia prima sive scientia
divina 1.2, p. 10, l.7375.
90 bruno tremblay
exist between the already known and conceptualized things that are used
as principles of knowledge and the things that become known thanks
to the former) are always dened concretely in logic. Let us take the
example of a logical being at the core of the treatise of the Categories:
genus. The rst part of Porphyry s denition of genus is not relation or
order, but that which, quod, that is to say the known thing or nature
that is the subject to the accident that is generality.
79
The passage from
the commentary on the Metaphysics gives us the reason why in logic
the accidents of things that constitute the subject of the discipline are
dened concretely, as inherent to things, and not in an abstract way:
second intentions can become instrumental in the passage from the
known to the unknown only if they are incarnated in the things we
conceive. Logical generality, as an abstract accident conceived without
a subject, is of no use to allow us to know partially the essence of the
different kinds of animals: only the genus animal, that is to say the
notion of animal put in relation to its species, can.
Thus, one understands better Albert s concern to remind his reader
that the principles of the being of a thing are also its principles of
knowing:
80
too radical a dissociation would endanger the objectivity
and even the possibility of human rational knowledge. But the truth
of this saying requires for him an important distinction: it is only as it
becomes a mental representation or intentio and takes on the accidents
that are brought about by the rational process that a thing might
become for us an instrument or a principle of knowing and that it can
be studied by logic.
The Categories Is Concerned with Words
For Albert the Categories is concerned both with things and the notions
we have of them. But what about words, the third option that is often
put forth by Greek commentators as the subject of this work?
It is clear that for Albert words are not foreign to the subject of the
Categories, or at least to its consideration. If statements and arguments
are dened as signifying vocal sounds (voces), as they are in Aristotle
and in Alberts commentaries,
81
and if the Categories examines potential
79
80
Super Porphyrium De V universalibus 2.3, p. 23, l.513. See Aristotle, Physics 1.1, ed.
Henri Carteron (Paris, 1961), 184a1213.
81
See Aristotle, On Interpretation 4, ed. Laurent Minio-Paluello (Oxford, 1949),
s CATEGORIES 91
elements of statements and arguments, as Albert tells us they do,
82
then
it seems quite logical to assume that the Categories is about some sort of
words. Moreover, if one describes this subject in terms of predicability,
and if one holds that logical predication takes place within a statement,
83

then it seems necessary that the said subject is somehow related to
words. Finally, although strictly speaking Albert does not adopt Boethius
denition to answer directly the question of the skopos of the Categories,
he does refer to it and he does so with obvious sympathy.
84
Despite all this, we should not conclude too hastily and without any
nuance that Albert identies the subject of the treatise with words. As
a good disciple of Avicenna , he denies explicitly and strongly in his
general introduction to logic that the subject of logic can be dened,
strictly speaking, in terms of words or vocal sounds.
85
Among his many
justications, one is especially worth mentioning here: the purely logi-
cal properties that we ascribe to words and statements do not belong
to them as words or statementsjust as, strictly speaking, they do not
belong to the things that exist in the external world and that the words
indirectly refer to, but only inasmuch as they are signs of what these
logical properties rst and foremost belong to, that is to say things as
conceived or known by us.
86
For example, we divide signifying vocal
sounds into complex and non-complex, but vocal sounds have these
properties only because they signify complex and non complex concep-
tions of the mind.
87
Interestingly, Albert repeats the same thing about the
logical property that is at the core of the Categories: logical universality
is essentially a property of intentiones or things as known, and not things
as named or things as they exist outside of our intellect.
88

Albert does not reject Boethius denition for one reason: it gives
him a chance to refer to a distinction he has already made in his
general introduction to logic. There Albert reminds his readers of
the experiential fact that we are not able to go from the known to the
16b2617a3, and Prior Analytics 1.1, ed. W. D. Ross (Oxford, 1964), 24a16b20, and
Albertus Magnus, Peri hermeneias 1.4.1, ed. Auguste Borgnet (Paris, 1890), p. 406A, and
Analytica priora 1.1.3, ed. Auguste Borgnet (Paris, 1890), p. 461.
82
De praedicamentis 1.1, p. 2, l.2536, and Peri hermeneias 1.1.1, p. 374B.
83
84
De praedicamentis 1.1, p. 1, l.52p. 2, l.2. He also makes explicit use of Boethius
denition in Peri hermeneias 1.1.2, pp. 377B378A.
85
Super Porphyrium De V universalibus 1.4 and 5.
86
87
Super Porphyrium De V universalibus 1.5, p. 9, l.4067, and De praedicamentis 1.5, p. 12,
l.635.
88
92 bruno tremblay
unknown unless we use, externally or in our imagination, vocal signs of
our concepts.
89
Now since the Categories considers universals, and more
precisely universals as principles of knowing, and since universals can-
not concretely act as principles of knowing unless they are signied by
words, then the Categories will study and consider universals in relation
to words, as the property of predicability shows. But it must be kept in
mind that for Albertus Magnus words are not, properly speaking, the
ordinabilia that constitute the subject of the Categories.
As long as it is clearly understood that in logic things are considered
not in themselves but as foundations of our notions, and likewise words
and sentences are considered not in themselves but as signs of our
notions, Albert seems to think that it matters little whether one speaks
of the different subjects studied in logical treatises in terms of words,
concepts, or things. This learned indifference, so to speak, is particularly
obvious in his commentary on the Categories, in which the ten supreme
genera are successively alluded to in terms of things, notions, and
words.
90
This is an attitude which is in sharp contrast with the rigidity
found in some Neoplatonist commentators, such as Boethius .
91
It is
easy to understand that so much diversity in the way the subject of a
book is dened could make a commentator ill at ease, especially if he
wants to see the treatise as being more than a mere junk yard lled
with grammatical, logical, and metaphysical considerations which are
unable to produce a unied science. But one must be reminded of
the important role played by logical analogy in the thought of many
medieval thinkers, and we saw an example of this when briey discuss-
ing Alberts teachings on universality. If it is thanks to analogy that
one can understand the underlying unity of Aristotles Metaphysics,
92

may we not also think that if in the course of his commenting on the
Categories Albert ascribes the logical properties that he studies to things,
notions, and words, it is because they belong to them all, but not in
the same sense? The analogy that is at stake reminds us of the anal-
ogy between three main uses of the word healthy: strictly speaking
only living beings that accomplish their biological functions well are
healthy, but by analogy one can also call healthy what causes health
(e.g., a healthy food) and what manifests health (e.g., a healthy color).
89
90
See, among a great diversity of examples, 2.3, p. 24, l.1821, and 6.1, p. 130, l.3235.
91
See Boethius , In Categorias Aristotelis 1, p. 162AC.
92
Metaphysica 1.1.3.
s CATEGORIES 93
In the same way, the notions of things are the essential subject of the
logical properties that are studied in the Categories, but by analogy one
can also attribute such properties to the extra-mental foundations or
to the vocal signs of these notions.
93

The Denition Of The Subject Of The Categories, Revisited
We are now in a better position to understand the meaning of the brief
and introductory denition of the subject of the Categories that is given
by Albert in the proemium of his commentary: what can be ordered
into a relation of predicability or subjectibility, inasmuch as it stands
under the vocal sound that signies this order. Although much more
could be said, the following distinctions can be established.
First, it should now be clear that, in Albert s mind, that which, strictly
speaking, can be related to something else as a subject or a predicate
is primarily a notion or a form as it exists in our mind, and as logi-
cal universality is added to it by reason when it tries to go from the
known to the unknown. One may also say that it is a thing or being
(or a principle of being), or a word, but such ascriptions are less strict
and presuppose analogy.
Second, the order or relationship that is at stake is necessarily pro-
duced by reason when, using a notion that it already possesses in
order to know something else, it grasps the already known thing as
inherent to the unknown thing and as able to be identied, as forma
totius, with the unknown thing and said of it through predication. The
known, which here is used as a principle of knowing in comparison
to the unknown, becomes then related to the unknown as a predicate,
and that to which the known nature is attributed becomes related to it
as a subject. The use of the notion of order (ordo), which adds to the
notion of relationship (ratio) the idea of before and after, is justied by
the fact that logically speaking a predicate is a principle of knowing
with regard to the subject, and as such it comes before.
94
These rela-
tionships of predicability and subjectibility are obviously actualized
only in the second act of reasoncomposition or division, through
93
De praedicamentis 7.6. See also Ethica 6.2.1, ed. Auguste Borgnet (Paris, 1891), pp.
407A408B.
94
De praedicamentis 1.1, p. 1, l.2238, and p. 2, l.3136; 7.1, p. 133, l.53p. 134, l.9,
and 12, p. 295A. See also Analytica priora 1.2.6, p. 495.
94 bruno tremblay
statements, of grasped natures, and the logical study that depends
on the rst act of reason examines these relationships inasmuch as
they exist in potency.
95

Third, even if in this denition subjectibility and predicability are
both explicitly present,
96
they do not seem to be on the same footing or
equally relevant for Albert , who in fact usually mentions only predica-
bility.
97
This is understandable given that: 1) in the strictest sense only
individual substances, and not natures as known, are subjects,
98
and only
natures as conceptualized or known, and not individual substances,
99

are predicates; 2) the notion of predicate includes all universal things
studied in the Categories whereas the notion of subject, even taken in a
looser sense, does not include the ten supreme genera; 3) the Categories
studies universals as principles of knowledge and order, and therefore
the ten supreme genera or univocal predicates, which are such principles
not only in regard to individual things but also to all other universals
within their category, get the lions share of the logicians attention;
100

4) universals are, as such, dened as predicates and not as subjects.
101

Fourth, the last part of the denition found in Albert s proemium
(inasmuch as it stands under the vocal sound that signies that order)
does not mean that the Categories studies rst and foremost words, or
that universals are examined through the grammatical and conventional
properties of the words that signify them. It merely states that since
we need the sensible signs that words are, whether they be spoken or
simply imagined, in order to think rationally (to go from the known
to the unknown), universals and the logical order that exists among
them can only be discovered and studied through their sensible, vocal
manifestation.
95
96
Probably to emphasize the fact that the Categories will provide material to build
statements, which are always made of a subject and a predicate. See De praedicamentis
1.1, p. 2, l.2536, and Peri hermeneias 1.1.1, p. 374B.
97
See for example Super Porphyrium De V universalibus 2.1, p. 17, l.632, and De
praedicamentis 1.1, p. 2, l.3136.
98
De praedicamentis 2.1, p. 20, l.4151, and p. 21, l.2939.
99
Super Porphyrium De V universalibus 3.4, p. 48, l.42p. 49, l.21, and De praedicamentis
1.5, p. 13, l.2945.
100
101
Super Porphyrium De V universalibus 4.2, p. 59, l.5764. Given everything that I
have tried to establish in this paragraph and in general throughout the paper, I really
cannot understand what E. P. Bos means when he writes that it is important to note
that by a praedicabile Albert means a thing of which something is or can be said (The
Division of Being over the Categories According to Albert the Great, Thomas Aquinas
and John Duns Scotus , p. 185) and that by category, Albert here means a thing of
which something is or can be said (John Versor s Albertism in his Commentaries on
Porphyry and the Categories, p. 56).
s CATEGORIES 95
Even if our knowledge of thirteenth-century logic remains very par-
tial, Albert s general understanding of the subject of the Categories shows
that it is probably quite safe to say that this philosopher is a very clear
example of the shift in emphasis from words to concepts or intentions
that took place during that century,
102
even if words, as well as things,
remain important parts of the solution he proposes.
The Categories, The Isagoge and Metaphysics
I mentioned at the beginning of this paper at least one more question
which Albert s short denition of the subject of the Categories, considered
by itself, does not solve, and which requires further explanations. It is a
problem that, at rst, looks like a rather technical and not very signi-
cant one, but which opens the door to some very serious questioning.
This problem is the following: how does Albert distinguish the logical
consideration of universals found in the Categories from the one made
by Porphyry in his Isagoge?
Albert s usual answer is that the Isagoge studies the different relations
by means of which universals can be ordered, whereas the Categories
studies the ordered things themselves. And this makes sense given the
practical purpose he assigns to logic: to order the things that exist in
our thoughts, we must rst know the ve main relations of universality
or predicabilitythe task of the Isagoge
103
, but we must also know
the things that can be ordered according to these relationsthe task
assigned to the Categories.
104
Albert goes as far as to say that the Isagoge
102
Norman Kretzmann , History of Semantics, in The Encyclopedia of Philosophy,
ed. Paul Edwards (New York, 1967), vol. 7, p. 371. Thus I disagree with Giorgio Pini
(Categories and Logic in Duns Scotus , p. 27), who, in his very interesting survey of pre-Scotus
thirteenth-century views on logic and the Categories, says that Albert did not really apply
the Avicennian doctrine of second intentions to his understanding of the subject of
that work, and that such attempts would only come slightly after Albert. It is true that
Albert never uses the expression secunda intentio in his commentary on the Categories, and
that in Super Porphyrium De V universalibus 1.4 he states that the subject of logic is argu-
mentum and does not refer to that expression there either, and that an explanation must
indeed be found for that fact. As I have tried to show in this paper, however, Alberts
understanding of ordinabile and praedicabile, in his denition of the subject of the Categories,
is very clearly along the lines of Avicenna s description of the kind of being that logic
studies. It comes as no surprise to me to see that Albert would explicitly write, at a
point in his career when he had not yet begun to write commentaries in logic, that the
word genus names a second intention (Super libros sententiarum 1.13.A.1, ad 1).
103
104
96 bruno tremblay
considers accidents of things according to which things are classied in
genera, species, etc., and that the Categories studies the classied things
themselves.
105
This last description, in particular, is quite troubling, and
according to the way Albert himself understands logic, it might be like
saying that the Isagoge is a logical treatise but the Categories is not! Yet,
in every part of his logical corpus, Albert afrms that the Aristotelian
work is a logical treatise.
106
Although Albert is not extremely explicit on this, I think that for him
the difference between the logical consideration of universals conducted
in the Isagoge and that conducted in the Categories must be understood
as a distinction between what is more general and abstract, on the one
hand, and what is more particular and concrete, on the other hand.
107
In
his commentary on Porphyry ,
108
Albert writes that, unlike the Categories,
the Isagoge considers universals as predicable, which in this particular
passage does not simply mean that the Isagoge is not interested in uni-
versals in re or ante rem, but that it is interested in dening the ve main
modes of logical universality or predicability in general, whatever the
things that are universal and predicable may be. This would be why,
in Alberts view,
109
Porphyry begins his denitions with a vague that
which, quod: a genus is that under which a species is placed, is that
which is predicated of specically different things, in saying what they
are, etc. The ve different modes of universality or predicability can
belong to most things the mind conceives, and this is why one cannot
dene what a genus is using the notion of animal or living being,
for instance. The Categories, on the other hand, is interested in logical
properties that belong to precise notions or conceptualized things or
natures. The logical inquiry into substance, for example, studies the
logical properties that belong to substance as this specic universal post
rem, and not necessarily to every genus.
110
So in that sense the Categories
105
106
Super Porphyrium De V universalibus 1.7, p. 15, l.2334; 2.1, p. 18, l.2350; De prae-
dicamentis 1.1; De sex principiis 1.1, p. 1, l.640; Peri hermeneias 1.1.1, p. 374B; etc.
107
Robert Kilwardby might have the same distinction in mind when he writes at
the beginning of his commentary on the Categories that after the consideration of the
universals that took place in Porphyry s Isagoge, the logician must now go down to the
predicaments themselves, which is what is done in the Categories. See Robert Kilwardby,
Notulae super librum Praedicamentorum, proemium.
108
Super Porphyrium De V universalibus 2.1, p. 17, l.632. See also De praedicamentis 1.1,
p. 1, l.2430.
109
110
De praedicamentis 2.6, p. 31, l.16. See also the general statement made in 1.1, p. 2,
s CATEGORIES 97
studies the different known things themselves that are predicable, but
it remains a logical consideration.
But important problems remain, the most important of which is,
I think, that Albert sees in the Aristotelian treatise long passages in
which extra-mental properties of categories are discussed. The very
fundamental chapter on substance, for example, contains, according
to Albert, rst an enumeration of the logical properties of substance,
and then an enumeration of objective or extra-mental properties.
111

He mentions that the former comes rst in the text because it belongs
more properly to a logical work like the Categories, but he accepts the
presence of the latter without protesting. Why? Answering this ques-
tion will require a very careful examination of the commentary, which
cannot be done here.
112
It is interesting to see, however, how strong an
advocate of Avicenna s views on logic Albert was, and how he tried to
apply them to the Categories, which the tradition classies as a logical
work. But at the same time we are told by Avicennian scholars that in
the yet untranslated Categories of his Shifa, Avicenna himself strongly
questions this classication.
113
In such circumstances, one wishes one
could rewrite history and make a Latin translation of Avicennas
commentary available to Albert the Great, just to see what fruits this
confrontation of ideas might have borne.
l.6267, which I think must be interpreted in the same light: (. . .) inasmuch as they
[= the natures or forms that are studied in the Categories] are something which can be
predicated or ordered within a genus according to this or that way of being a predicate
or a subject, (. . .) they have many properties and characteristics ( passiones) that can be
demonstrated of them. And this is how we will treat of them here. (My emphasis.)
111
De praedicamentis, 2.6, p. 31, l.16.
112
Among all the different justications given by Albert for a logical consideration
of apparently metaphysical or natural matterssee for example De praedicamentis 2.12,
p. 45, l.1439; 6.1, p. 130, l.2841; 7.1, p. 133, l.1742; 7.4, p. 278; and 7.13, 297A,
some seem at rst sight a little unsatisfying, and obviously call for a more serious and
in-depth consideration. Just like the short denition of the subject of the Categories
that one nds in Alberts proemium, some of these short explanations seem to be quite
standard for the time (see for example the proemium of Kilwardbys Notulae super librum
Praedicamentorum) and would require a separate study of their own in order for us to
fully comprehend what precise meaning Albert gives them.
113
Dimitri Gutras, Avicenna and the Aristotelian Tradition (Leiden, 1988), pp. 265267;
Ibrahim Madkour , Le trait des Catgories du Shifa, Mlanges de lInstitut dominicain
dtudes orientales du Caire 5 (1958), 253278; A. I. Sabra , Avicenna on the Subject
Matter of Logic, p. 764.
INTERCONNECTED LITERAL COMMENTARIES
ON THE CATEGORIES IN THE MIDDLE AGES
Robert Andrews
Literal commentaries on Aristotles Categories from the Middle Ages
have certain conditions which make them less attractive to the modern
researcher than comparable question commentaries.
1
Literal commen-
taries analyze an authoritative text line by line, so that less interesting
passages often get just as much attention as important ones, and impor-
tant passages may be handled perfunctorily. For the modern investiga-
tor, philosophical debates need to be teased out of the text rather than
presenting themselves organized into questions. For an editor, different
literal commentaries can be confusingly alike, leading to conation and
misattribution. For a translator, literal commentaries are almost impos-
sible to translate smoothly or pleasingly to the modern ear. (Attempts
can be seen in the translations of Aquinas s various Aristotelian com-
mentaries). Finally, the sheer mass of detailed comment, of varying
import, can overwhelm an unaccustomed researcher.
However, these perceived disadvantages carry with them concomi-
tant advantages. Literal commentaries are exhaustive in their analyses,
including even minor points of exegesis which might be of interest to
a scholar. Their format is systematic, so that a researcher knows where
to nd the locus for a philosophical discussion extending over several
centuries. (Discussions of the nature of number, for example, are to be
found in the Categories at 4b26.) Trails of traditions, translations, biases,
and inuences are all laid clear. And literal commentaries still provide
some focus on important issues by means of inserted dubia regarding
disputed topics. Ultimately, the greatest advantage of the medieval
working and re-working of this sort of detailed commentary is that it
1
For the relative advantages of question commentaries, see Robert Andrews ,
Question Commentaries on the Categories in the Thirteenth Century, in Medioevo:
Rivista di storie della losoa medievale 26 (2000), 141183. Silvia Donati proposes a nuanced
classication of questions, in Per lo studio dei commenti alla Fisica del XIII secolo.
I: Commenti di probabile origine inglese degli anni 12501270 ca., Documenti e Studi
sulla Tradizione Filosoca Medievale II, 2 (1991), 3845.
100 robert andrews
results in a thorough understanding of the text, one that is illuminating
of both the original work as well as of the medieval psyche.
For these reasons, to begin to provide entrance to the treasury of
material to be found in medieval literal commentaries, I shall show some
of the features of an interconnected tradition among literal Categories
commentaries of the late 13th and early 14th centuries. I shall do so
by centering on a work by a relatively obscure commentator active at
the University of Paris, Peter of Saint-Amour , showing his style, method,
derivations, originality, traditions, and inuences, and his connection to
other authors within the same historical progression, including Gerard
of Nogent , Siger of Courtrai , and Thomas of Erfurt .
1. Peter of Saint-Amour , Life and Works
Not much is known about the life of Peter of Saint-Amour . The most
established information about his life seems to be that he was a Master
of Arts at the University of Paris, and that in 1281 he was rector
there.
2
In a pamphlet from 1902, M. Perrod
3
tried to argue that Peter
can be identied with the Peter of Saint-Amour who was nephew of
William of Saint-Amour , but this identication has been refuted by
Fr. Gauthier.
4

The works of Peter of Saint-Amour , besides the Sententia supra librum
Praedicamentorum in Paris Bibl. Nat., nouv. acq. lat. 1374 ff. 13rb34rb,
include the anking commentaries on Porphyry s Isagoge (ff. 1ra9rb)
and Aristotles De interpretatione (ff. 34rb48vb). Other Perihermenias
commentary versions are dismissed by Tabarroni .
5
Catalog records
also report commentaries on the Posterior Analytics and De generatione
et corruptione by Peter.
6
Barbara Faes de Mottoni claims to have found
2
Chartularium Universitatis Pariensis I, Heinrich Denie and Emile Chatelain, eds.
(Paris,18891897), pp. 5889 n. 503.
3
M. Perrod , Pierre de Saint-Amour, recteur de lUniversit de Paris, chanoine de Mcon,
archidiacre de Vauxrenard (Besanon, 1902). Perrod is likely the source of Glorieux s note:
Pierre de Saint-Amour. Pierre Bellison. Neveu de Guillaume de Saint-Amour. Ds
1250 chanoine de Saint-Vincent de Mcon. Recteur de lUniversit en 1281. Mort le
1er juin 1295; P. Glorieux, La Facult des artes et ses matres au XIII
e
sicle (Paris, 1971).
4
Thomas de Aquino, Expositio libri Peryermenias. Opera omnia I,1, Editio altera retracta
(Rome: Commissio Leonina, 1989) 74*.
5
Andrea Tabarroni , Lo pseudo Egidio (Guglielmo Arnaldi) e uninedita continu-
azione del commento di Tommaso al Perihermenias, Medioevo 14 (1988), 391403.
6
Charles H. Lohr , Medieval Latin Aristotle Commentaries. Authors: Narcissus
Richardus, Traditio 28 (1972), 371.
interconnected literal commentaries 101
excerpts from the Posterior Analytics commentary in Vatican Bibl. Apost.
Pal. lat. 1009;
7
another manuscript, Vienna Nationalbibl. Pal. 5366,
8
contains a Posterior Analytics commentary explicitly attributed to Peter,
which Tabarroni
9
reports is close to but not identical with the text
printed in Faes de Mottoni.
One sophism, Nihil est contingens, is explicitly attributed to Saint-
Amour in Vatican lat. 14812 ff. 5r6r; he may as well be the author
of the two preceding sophisma on folios 1r5r.
10
2. Peter of Saint-Amour s
Sententia supra librum Praedicamentorum
A surprising number of manuscript copies of commentaries on the
Categories have been attributed to Peter. In his introduction to Thomas
Aquinas on the Perihermenias, Fr. Gauthier
11
suggests several new texts
by Peter: Munich Clm 8002 ff. 88va98va; Vatican Bibl. Apost. Pal.
lat. 1007 ff. 11ra21ra; Paris Arsenal 530 ff. 16ra21va; Oxford Merton
College 296 ff. 7r 23v; and a fragment contained on f. 142r of Munich
Clm 14763. Among these the strongest case lies with the fragmentary
Munich Clm 14763, which is explicitly attributed to Peter: Explicit
liber Praedicamentorum editus a Petro de Sancto Amore. This same
fragment is credited to Peter by Lohr .
12
However, Tabarroni
13
refutes
these attributions; the incipits and explicits that he reproduces do not
correspond with the Peter of Saint-Amour text of Paris Bibl. Nat.,
nouv. acq. lat. 1374.
7
Barbara Faes de Mottoni, Il commento di Pietro di S. Amore agli Analitici Posteriori
ritrovato?, Studi Medievali 27 (1986), 383405.
8
Cf. M. Markowski , Repertorium commentariorum medii aevi in Aristotelem latinorum quae
in bibliothecis Wiennae asservantur. Opera philosophorum medii aevi, Textus et studia (Wroclaw,
1985), p. 264.
9
azione del commento di Tommaso al Perihermenias, Medioevo 14 (1988), 400 n. 42.
10
Cf. Siger de Brabant, crits de logique, de morale et de physique, Bernardo Bazn, ed.
Philosophes Mdivaux xiv (Louvain-la-neuve,1974), pp. 78.
11
(Rome, 1989), pp. 73*74*.
12
13
azione del commento di Tommaso al Perihermenias, Medioevo 14 (1988), 3934.
102 robert andrews
Tabarroni
14
also provides the service of showing that the text of
Paris BN 1374 is indeed authentically by Peter of Saint-Amour . He
nds in a Notabilia super Porphyrium in the same manuscript an allusion to
Peters doctrine:
Anonymus (H. de Brox?) Notabilia super Porphyrium, Paris Bibl. Nat., nouv. acq.
lat. 1374 f. 55vb: Tertio modo potest exponi, ut exponit magister Petrus
de Sancto Amore, et hoc sic: sub gura appellationis, id est sub simili signi-
catione appellantis ad appellatum, quia res signicata est una, quamvis
differenter consideratur, quae importatur et designatur nomine generis,
speciei, et differentiae, et individui.
Petrus de Sancto Amore Sententia supra librum Praedicamentorum, Paris Bibl. Nat.,
nouv. acq. lat. 1374 f. 17vb: Et ideo exponatur sic littera (3b15): autem, pro
sed, videtur in secundis substantiis hoc aliquid signicare sub gura appellationis,
id est sub similitudine termini appellantis ad terminum appellatum, quia
utraque imponuntur in singulari numero.
15
Peter of Saint-Amour s Sententia supra librum Praedicamentorum is thus extant
in a single manuscript, Paris Bibl. Nat., nouv. acq. lat. 1374 ff. 13rb34rb.
The manuscript has been previously described by Delisle
16
and Senko .
17

A marginal note on f. 109v dates an early ownership to 1330.
In structure, the Sententia supra librum Praedicamentorum is a literal com-
mentary with inserted dubia and notabilia.
18
The format is rather invari-
ant: rst, a division of the text (the rst paragraph after the Prooemium
is a division of the entire book), providing explicit lemmata for the sub-
divisions; then each subdivision of the text is analyzedusually both
summarized and reported word for word, with interposed clarifying
paraphrases. For example, in explaining 10a4, the author writes:
14
Ibid., 3989.
15
References to the text of the Categories within my edition are indicated by the
Bekker number in parentheses, thusly: (7a17). My procedure has been to italicize the
interposed text of Aristotle, as done in the edition of the Metaphysics commentary
of Albert the Great (Albertus Magnus Metaphysica, libros quinque priores, ed. B. Geyer,
Alberti Magni Opera omnia XVI,1, Aschendorff, 1960). The spelling presented in the
editions is normalized. The rudimentary punctuation of the manuscripts has served
as a guide without obligation.
16
L. Delisle , Manuscrits divers acquis par la Bibliothque Nationale en 1876, 1877
et 1878, Mlanges de Palographie et de Bibliographie (Paris, 1880), pp. 46971. In edition
it is 61000 words long.
17
W. Senko , Repertorium commentariorum medii aevi in Aristotelem latinorum quae in Bibliothecis
publicis Parisiis asservantur. Opera philosophorum medii aevi, Textus et studia tom. 5 fasc. 12
(Warsaw, 1982), pp. 1036.
18
The colophon entitles the work Sententia et etiam notabilia supra librum Praedica-
mentorum.
And then the Philosopher continues, saying that not only those qualities
which arise from passions are called transient qualities, but Similarly any
aberrations, that is externalities or affections, not naturally,supply are in
the soul, but are made from some other circumstances which are difcult to
change; also such things are called qualities. And he proves this, because
according to those we are called qualities.
The author dissects Aristotles text in great detail, and often parses the
arguments into syllogisms, supplying missing premises or conclusions.
Following each section are dubia and notabilia. The notabilia are the
commentators didactic remarks on the text. The dubia raise problems,
which are immediately answered. These may occur singly or in a series,
and at the most complex there may be a series of counter-objections
and successive responses. Although at the time this commentary was
written the genre of question commentaries on the Categories was well
established, this work maintains the older literal commentary format,
and never rises to the complexity and freedom of the question com-
mentary genre.
Nothing is known about the composition or presentation of the work,
but it may have been presented as a lecture:
Amplius si convenienter (7a31). In parte ista, quia tota die locutus est
de convenienti et inconvenienti assignatione, et ista non multum sunt
manifesta, ideo hic ostendit quid per ipsa intelligat.
Authors of three other commentaries seem to belong to the same tradi-
tion of literal Categories commentaries as Peter of Saint-Amour : Gerard
of Nogent , Siger of Courtrai , and Thomas of Erfurt .
3. Gerard of Nogent s
Glossulae supra librum Praedicamentorum Aristotelis
The attribution to Gerard of Nogent of the Logica vetus commentary
considered here is uncertain; it is in competition with an unspecied
Durandus (and the Perihermenias commentary contained in it was some-
times credited to Thomas Aquinas ).
19
Gerard of Nogent, like Peter of
Saint-Amour , was Rector at the University of Paris, in 1292. However,
his Categories commentary shows signs of being earlier than Peters,
19
Cf. Thomas de Aquino, Expositio libri Peryermenias, Opera omnia I,1, Editio altera
retracta (Rome, 1989), p. 75*.
104 robert andrews
which is comprehensible since it is known that Gerard was resident
earlier in Paris, between 1253 and 1274.
20
The only work which with
certainty can be attributed to Gerard is a commentary Super librum
Posteriorum. The manuscript of the Categories commentary from which
I transcribe is Padova Bantoniana XX.480 ff. 7vb25vb; it is extant
also in Escorial BReal f.III.24, ff. 144v, Milano BTrivulziana 753
(H 67), ff. 174, Oxford Merton College 261, ff. 72112, Paris Bib. Nat.
Mazarine 3523, ff. 1r56v, Paris BN lat. 15005, ff. 206249,
21
and Paris
BN lat. 16618 f.125seqq.
22
A conation in the Milan manuscript with
some material associated with John of Seccheville is discussed by P. O.
Lewry .
23
Despite the uncertainty of attribution I shall continue to refer
to the work as by Gerard of Nogent.
24
4. Siger of Courtrai s Commentarium in Categorias
Siger of Courtrai is well known from the early collection of his works
in Les Philosophes Belges
25
and the complete edition of his Perihermeneias
commentary by C. Verhaak .
26
As well a number of his modistic texts
are available.
27
His Categories commentary discussed here is found
20
Charles H. Lohr , Medieval Latin Aristotle Commentaries. Authors AF, Traditio
23 (1967), 1623.
21
This manuscript is incomplete; it is described in W. Senko , Repertorium commentari-
orum medii aevi in Aristotelem latinorum quae in Bibliothecis publicis Parisiis asservantur, in Opera
philosophorum medii aevi, Textus et studia tom. 5 fasc. 12 (Warsaw, 1982), p. 185.
22
Described in W. Senko , Repertorium commentariorum medii aevi in Aristotelem latinorum
quae in Bibliothecis publicis Parisiis asservantur, in Opera philosophorum medii aevi, Textus et studia
tom. 5 fasc. 12 (Warsaw, 1982), p. 78.
23
P. O. Lewry , Robertus Anglicus and the Italian Kilwardby, in English Logic in
Italy in the 14th and 15th Centuries, Acts of the 5th European Symposium on Medieval
Logic and Semantics, Rome, 1014 November 1980, ed. A. Maier (Naples, 1982),
pp. 345.
24
Bits of this work have been printed in Robert Andrews , Peter of Auvergne s Commentary
on Aristotles Categories: Edition, Translation, and Analysis, Dissertation: Cornell University
1988 and in Ioannes Duns Scotus , Quaestiones in Isagogen Porphyrii et Quaestiones super
Praedicamenta Aristotelis, eds. R. Andrews, G. Etzkorn, G. Gl, R. Green, T. Noone, and
R. Wood (Opera philosophica vol. I) (New York, 1999).
25
G. Wallerand , Les oeuvres de Siger de Courtrai (Les Philosophes Belges, VIII) (Louvain,
1913).
26
Siger de Courtraco, Expositio in librum Aristotelis Perihermeneias, in Zeger van Kortrijk,
Commentator van Perihermeneias, edited with an introduction by C. Verhaak (Brussels, 1964).
27
Siger de Courtraco, Summa modorum signicandi, Sophismata. New edition by Jan
Pinborg . Sophisma IV: Album potest esse nigrum (Amsterdam Studies in the Theory and History
of Linguistic Science. Series III, Studies in the History of Linguistics 14) (Amsterdam, 1977).
in two manuscripts, Erfurt SB Amplon. F.135, ff. 91ra139v,
28
and
Venice BMarc. lat. VI 21, ff. 92r158v. Brief passages from this work
have been published.
29
Siger was a Master of Arts in Paris in 1309,
and died in 1341.
30
Pinborg dates his commentary on the Categories to
13001320.
31
5. Thomas of Erfurt s Expositio
super librum Praedicamentorum
Thomas of Erfurt was active in the rst quarter of the fourteenth
century. His De modis signicandi gained lasting impact because of its
misattribution to Duns Scotus .
32
Little is known of his life; inuences
on his work suggest that he was educated at the University of Paris,
and later documents place him in Erfurt, Germany.
33
His Expositio super
librum Praedicamentorum exists in four manuscripts: Erfurt SB Amplon.
Q.266, ff. 8ra20rb; Leipzig UB 1356, ff. 9ra18rb; Mnchen SB Clm.
4378, ff. 29r 66v; Mnchen SB Clm. 14458, ff. 119ra133ra.
34
28
Described in Mieczyslaw Markowski , Repertorium commentatorium medii aevi in
Aristotelem Latinorum quae in Bibliotheca Amploniana Erffordiae asservantur (Polska Akademia
nauk, Instytut lozoi i socjologii) (Wroclaw, 1987), pp. 1734.
29
Robert Andrews , Peter of Auvergne s Commentary on Aristotles Categories: Edition,
Translation, and Analysis, Dissertation: Cornell University 1988 and in Ioannes Duns Scotus ,
Quaestiones in Isagogen Porphyrii et Quaestiones super Praedicamenta Aristotelis, eds. R. Andrews,
G. Etzkorn, G. Gl, R. Green, T. Noone, and R. Wood . (Opera philosophica vol. I) (New
York, 1999).
30
C. Lohr , Medieval Latin Aristotle Commentaries. Authors: Robertus Wilgelmus,
Traditio 29 (1973), 138; C. Verhaak , Zeger van Kortrijk, Commentator van Perihermeneias
(Brussels, 1964), pp. xxiixxix.
31
Jan Pinborg , Die Logik der Modistae, Studia Mediewistyczne 16 (1975), 3997, p. 42.
32
Thomas of Erfurt , Grammatica Speculativa. An Edition with Translation and Commentary,
ed. G. L. Bursill-Hall (London, 1972).
33
Information about his life may be found in Snke Lorenz , Studium Generale Erfordense.
Zum Erfurter Schulleben im 13. und 14. Jahrhundert (Monographien zur Geschichte des
Mittelalters, 34) (Stuttgart, 1989), pp. 31225. See also Jack Zupko s article online in
the Stanford Encyclopedia of Philosophy: http://plato.stanford.edu/entries/erfurt/.
34
This work is discussed in Robert Andrews , Thomas of Erfurt on the Categories
in Philosophy, in Was ist Philosophie im Mittelalter? Akten des X. Internationalen
Kongresses fr mittelalterliche Philosophie der Socit Internationale pour lEtude de
la Philosophie Mdivale, 25. bis 30. August 1997 in Erfurt, edited by Jan A. Aertsen
and Andreas Speer (Miscellanea Mediaevalia 26) (Berlin, 1998), pp. 8018. Its edited
version is 54000 words.
106 robert andrews
6. Interconnections among the Commentaries
Peter of Saint-Amour s work was not written in isolation. It is demon-
strable that his commentary imitated other writers, and it in turn
was imitated by successors. This is not at all unusual. Medieval literal
commentaries (transgressing against modern taste) are often highly
derivative. In composing a literal commentary, many authors took some
predecessors text as a template upon which to elaborate, embellish, and
expand. The modern disdain for plagiarism is an inverse of medieval
values, for in the medieval world tradition was respected rather than
innovation; what was new was unproven and suspect, while the old
was established, authoritative, and time-honored. This is reected in
the medieval practice of only referring to antiquated authorities like
Aristotle or Boethius by name, referring to contemporaries instead as
some people (aliqui ) or certain ones (quidam).
It is clear that there is an interconnection among the commentaries
by the authors Gerard of Nogent (writing c. 127080), Peter of Saint-
Amour (c. 127080), Siger of Courtrai (c. 130020), and Thomas of
Erfurt (c. 130025). Other commentaries, attributed or anonymous, may
be added to the list. The similarities are such that the works sometimes
have been confused with each other. Several scholars have noticed the
similarities between the Categories commentaries of Siger of Courtrai and
Thomas of Erfurt.
35
Verhaak observes that the Thomas manuscripts are
in many cases identical with Sigers commentary, and says that these
manuscripts represent a different recension of the texts ascribed to
Siger.
36
Verhaaks conclusion is that the two commentaries show traces
of having derived from a common older expositio. Also, despite all the
manuscripts Fr. Gauthier attributes to Peter of Saint-Amour, only one is
authentic, but the other manuscripts exhibit considerable similarities.
In what follows I shall show that these commentaries belong to a
long tradition of imitation and accretion in which each writer builds
upon what has gone before, rening the analysis and adding original
material.
35
For example, Stephan Grotz in his introduction to his translation of De Modis
Signicandis: Stephan Grotz, translator, Thomas von Erfurt, Abhandlung ber die bedeutsamen
Verhaltensweisen der Sprache (Tractatus De modis signicandi) (Amsterdam, 1998), p. vii,
n. 1.
36
Commentator van Perihermeneias, edited with an introduction by C. Verhaak (Brussels,
1964), pp. cxxixcxxx.
7. Truth (2a8)
I shall begin with the history of a typical passage, an inserted note
prompted by Aristotles passage at 2a8 where he remarks that only
complete sentences are true or falseindividual components of sen-
tences (such as the terms which fall under the categories) are neither
true nor false. I set in juxtaposition the treatment of this passage in
four authors, beginning with Gerard of Nogent .
Gerardus de Nogento
Ms. Padova Bantoniana XX.480 f. 9vb
Secundo notandum quod supra passum istum dicit Boethius
37
quod
duplex est veritas. Quaedam est quae est adaequatio rei et intellectus
cognitione.
38
Et loquendo de tali, nullum illorum est verum vel falsum.
Alia est veritas quae est rei entitas. Et de tali dicitur quod verum et ens
convertuntur.
39
Et hoc modo quodlibet illorum potest dici verum.
40
Gerard has the simplest approach. He wishes to note that there is a sense
of truth which derives from the so-called transcendental terms (drawn
from Aristotles Metaphysics II,1 993b3031), a sense in which there is a
convertibility among being, truth, and goodness: whatever has being is
good; whatever is good is true; whatever is true has being. Of course in
this sense even the single terms which fall under the categories may be
said to be true. Another sense is the traditional one of adaequatio rei et
37
Immo Albertus Magnus, Liber de Praedicamentis I,7 (1890) 165b: Et hoc est verum
de hoc vero quod est adaequatio rerum complexarum cum intellectu componente vel
dividente. . . . Possunt tamen alio modo esse vera vel falsa, secundum quod verum dicimus
purum, alienae naturae non permixtum, vel existens et non apparens tantum.
38
Cf. Avicenna, Metaph. I, 8 (1977) 55; Bonav., Sent. I d. 40 a. 2 q. 1 in corp. (I
707b nota 5); Thomas Aquinas , De veritate q. 1 a. 1 (XXII1) 6; Robertus Grosseteste,
De veritate propositionis (1912) 144: Veritas enim sermonis vel opinionis est adaequatio
sermonis vel opinionis et rei; cf. J. T. Muckle , Isaac Israelis Denition of Truth,
Archives dhistoire doctrinale et littraire du Moyen Age 8 (1933), 58.
39
Cf. Aristoteles, Metaph. II,1 993b3031; Jacqueline Hamesse , Auctoritates Aristotelis,
Senecae, Boethii, Platonis, Apulei et quorundam aliorum (Louvain) 1974, p. 118 (42):
Unumquodque sicut se habet ad entitatem, sic se habet ad veritatem.
40
In the second place it should be noted about this passage (2a8) that Boethius
says that there are two kinds of truth. One kind is a correspondence of thing and the
intellect in comprehending; and in this sense none of these is either true or false. The other
kind of truth is the same as the being of a thing, as when it is said that truth and being
are convertible (993b30); and in this way each of them can be said to be true.
108 robert andrews
intellectus, the correspondence between intellect and thing, and in this
sense single terms are not true. This denition has a long history,
41
but
it does not occur in Boethius , to whom Gerard misattributes it.
Petrus de Sancto Amore
Ms. Paris BN 1374 f. 15va
Secundo notandum quod supra istum passum dicit Boethius quod
duplex est veritas. Quaedam est quae est adaequatio rei et intellectus
cognitione. Et dico cum cognitione, quia si intellectus non cognoscaret
rem, non posset apprehendere neque adaequari rei. Et per oppositum,
falsitas est inadaequatione rei et intellectus. Et sic loquendo, nullum
istorum est verum vel falsum.
Alia est veritas quae est rei entitas. Et de tali dicit Philosophus
42
secundo
Metaphysicae quod ens et verum convertuntur. Et hoc modo quodlibet
istorum potest dici verum.
43
Peter clearly bases his analysis on Gerard, but he inserts several clarify-
ing additions. Like Gerard, he falsely credits the distinction to Boethius .
He has the same term cognitione in the traditional denition adaequa-
tio rei et intellectus, but appends the clarication that this is because a
true correspondence depends on cognition, not intellect alone. Then
Peter inverts the denition to apply it to falsity, which is the lack of
correspondence between intellect and thing. Finally, Peter provides
the specic reference of Aristotles Metaphysics Book II for the appeal
to transcendentals. Peters account is basically Gerards, with inserted
clarications.
41
Cf. J. T. Muckle , Isaac Israelis Denition of Truth, Archives dhistoire doctrinale et
littraire du Moyen Age 8 (1933), 58.
42
Aristoteles, Metaph. II,1 993b3031; Hamesse , Auctoritates Aristotelis, p. 118 (42):
43
In the second place it should be noted about this passage (2a8) that Boethius
says that there are two kinds of truth. One kind is a correspondence of thing and the
intellect in comprehending. I say in comprehending, because if the intellect does not
comprehend a thing, it can neither understand nor correspond to the thing. Conversely,
falsity is a lack of correspondence between thing and intellect. In this sense none of
them is either true or false. The other kind of truth is the same as the being of a thing.
In this sense Aristotle in the second book of the Metaphysics says that being and truth
are convertible (993b30); and in this way either of them can be said to be true.
Siger of Courtrai
Commentarium in Categorias
Ms. Venice M L. VI. 21 f. 111vb
Ad evidentiam huius partis est intelligendum quod duplex est veritas,
una scilicet quae est natura rei entitatis, quae est adaequatio rei ad sua
principia. Et isto modo quodlibet ens est verum, quia quodlibet illorum
habet naturam rei entitatem.
Alia est veritas quae causatur ex compositione vel divisione aliquorum
adinvicem. Et illa quae sic sunt vera non sunt in praedicamento, quia
omnia quae sunt in praedicamento sunt incomplexa, ut probatum est.
44
In his rst distinction, Siger conates the transcendental and the cor-
respondence senses of truth: transcendental truth is the adaequatio rei ad
sua principia. For the second limb of his distinction he appeals (without
reference) to the denition of truth found in Aristotles De interpretatione
16a1213: circa compositionem enim et divisionem est falsitas veritasque. A
similar but tripartite distinction is also to be found in Sigers De interpre-
tatione commentary: transcendental truth, correspondence truth, and the
truth arising from composition and division.
45
Compared with Gerard
and Peter, Siger employs much the same terminology, but incorporates
it into a new result.
Thomas de Erfordia
Expositio super librum Praedicamentorum
Mss. Erfurt SB Amplon. Q.266 [= E] f. 9vb; Leipzig UB 1356 [= L] f. 10rb;
Mnchen SB Clm. 4378 [= M] f. 34v; Mnchen SB Clm. 14458 [= N] f. 121rb
Notandum quod aliquis posset instare: verum et ens convertuntur; sed
unumquodque praedicamentorum signicat ens; ergo signicat verum,
ut videtur.
44
In explaining this part (2a8) it should be understood that that there are two
kinds of truth. One kind is the same as the nature of the being of a thing, which is
the correspondence of a thing to its principles. In this sense each thing is true, because
each thing has the nature of the being of a thing. The other kind of truth is that
which is caused by the composition or division [of expressions], and what is true in
this sense is not within a category, because anything within a category is simple, as
has been shown.
45
Commentator van Perihermeneias, edited with an introduction by C. Verhaak (Brussels,
1964) in 16a1213: Et est sciendum quod veritas potest accipi tripliciter. Uno modo
secundum quod est passio entis; et isto modo verum convertitur cum ente, nec isto
modo accipitur hic. Alio modo potest accipi ut est passio intellectus; in isto modo non
pertinet ad librum istum sed magis ad librum De Anima. Alio modo potest accipi ut
signicatur et repraesentatur per vocem; et isto modo accipitur hic cum dicitur circa
compositionem et divisionem etcetera.
110 robert andrews
Dicendum quod duplex est veritas: quaedam est veritas quae est idem
quod pura et impermixta rei entitas, sicut dicimus illud est unum verum
quod est purum et natura aliqua impermixtum. De ista veritate loquitur
Philosophus secundo Metaphysicae
46
quando dicit quod omnino sicut
res se habent ad entitatem, sic se habent ad veritatem; quia sicut ens
et verum convertuntur, sic entitas et veritas. Et in isto modo concludit
argumentum.
Alia est veritas quae causatur ex compositione et divisione praedicati
cum subiecto vel a subiecto, de qua veritate dicit Philosophus primo
Perihermenias
47
quod circa compositionem et divisionem consistit
veritas vel falsitas. Quam veritatem etiam Boethius et Commentator
48

secundo Metaphysicae sic deniunt: quod veritas est adaequatio rei et
intellectus, id est conformitas intellectus cum re extra. Et de illa veritate
intendit Philosophus cum dicit Singula praedicamentorum secundum se
non sunt vera neque falsa (2a5) etc.
49
Thomas passage is the most complex of all. He formulates the issue
as an objection (aliquis posset instare) to Aristotles claim that categori-
cal items are neither true nor false: true and being convert (transcen-
dentally); anything within a category is a being, and so therefore true.
Thomas distinction is basically Sigers : the two types of truth are
transcendental or caused by composition and division. However, in
his description of transcendental truth he (like Peter of Saint-Amour )
gives the reference to Aristotles Metaphysics, as well as the full aucto-
46
Aristoteles, Metaph. II,1 993b3031; Hamesse , Auctoritates Aristotelis, p. 118 (42):
47
Aristoteles, De interp. 1 16a1213: Circa compositionem enim et divisionem est
falsitas veritasque.
48
Cf. Averroes, In Aristotelis Librum II Metaphysicorum Commentarius com. 1, ed. Darms
(Freiburg, 1966) 53.1656.67, sed non ad rem.
49
It should be noted that someone might object, thusly: Truth and being are
convertible; but each category signies being; therefore each category signies truth.
In response to this it should be said that that there are two kinds of truth. One kind
is truth which is the same as the pure and unmixed being of a thing, just as we say that
which is one [and] true is pure and unmixed with another nature. The Philosopher
speaks of this sense of truth in the second book of the Metaphysics when he says that
each thing is ordered to being in just the way that it is ordered to truth, because just
as being and the true are convertible, so are [the abstracts] being and truth. In this
way the objection is resolved.
The other kind of truth is that which is caused by the composition of the predicate
with the subject, or the division of the predicate from the subject. Aristotle speaks of
this kind of truth in the rst book of De interpretatione (16a12) when he says that truth
and falsity belong to composition and division. Also Boethius , and Averroes on the
second book of the Metaphysics, dene this sort of truth as correspondence of thing
and intellect, that is, a conformity of the intellect with an external object. Aristotle
means this sort of truth when he says (2a5) that Individuals within a category are in
themselves neither true nor false.
ritates quotation. Furthermore, he uses language which refers back to
the earliest known proponent of a distinction regarding truth in this
context, Albert the Great, whose language (purum, alienae naturae non
permixtum) was only adumbrated in Siger.
When Thomas explains the type of truth caused by composition and
division, he adds to Siger a reference to and quotation from De interpre-
tatione. As well he identies this sort of truth with the correspondence
theory of truth. Like Gerard and Peter he falsely attributes this denition
to Boethius and adds another false attribution for good measure, to
Averroes ! Indeed, in his comments on Metaphysics II Averroes discusses
truth, but without the phrase adaequatio rei et intellectus. Thomas
account is more than three times as long as Gerards, and seems to
incorporate elements from all of the other writers. Various writers and
commentators may be involved in the chain of inuence, but it is clear
that Thomas nal result is original only in the sense that it uniquely
combines earlier elements.
The preceding passages reect the several different techniques for
adaptation of a predecessor. Clarication may be provided by inserted
annotations and elaborations. Associations might lead to digressions. A
simple allusion to an authority may be lled out with an explicit attribu-
tioncorrectly or incorrectly!and a more complete quotation. Textual
elements may be rearranged or combined. None of these techniques
produce entirely original results, but they are the methods by which
literal commentaries growby synthesis and accretion.
8. Denomination (1a13)
In another series of texts one can see the correlations and incrementa-
tion of the literal commentary process. The issue at hand concerns the
requirements for denomination.
50
The starting point for the discussion is
Boethius : Tria sunt autem necessaria ut denominative vocabula constituantur: prius
ut re participet, post ut nomine, postremo ut sit quaedam nominis transguratio.
51

50
Regarding denomination, see the bibliography in Robert Andrews , Peter of
Auvergne on Denomination, in Meaning and Inference in Medieval Philosophy: Studies in
Memory of Jan Pinborg , ed. Norman Kretzmann (Dordrecht, 1988).
51
Boethius, In Cat. (1860), col. 168: Three conditions are required in order that
words be considered denominative: rst, that they agree in object; next that [they agree]
in name; and nally that there is a certain transformation of the name.
112 robert andrews
A more proximate source is Albert the Great: In denominatione ergo haec
tria necessaria sunt, scilicet natura aliena subiecto extrinsecus aptata et circumposita,
in principali et denominativo eadem res signicata, et diversi modi signicandi.
52

I present our four authors in the same order as above.
Gerardus de Nogento
Ms. Padova Bantoniana XX.480 f. 8vb
Circa istam partem notandum primo quod ad hoc quod aliqua dicuntur
denominativa quattuor requirunturBoethium.
53
Primum est quod com-
municant in principio et differunt in ne. Et per hoc patet quod studiosus
denominative non dicitur a virtute. Secundum est quod signicet eandem
rem formaliter. Tertium est quod differant in modo signicandi, ita quod
unum signicet per modum per se stantis, reliquum per modum adiacentis.
Quartum est quod abstractum differat essentialiter ab eo cui concretum
inheret. Et per hoc habetur quod concretum proprie non est substantiis,
quia humanitas non differt essentialiter ab eo cui homo sit concretum.
54
Ms. Paris BN 1374 f. 14rb
Item nota quod ad naturam denominativorum quattuor requiruntur.
Primum est quod conveniant in voce a parte principali et differunt a
parte nis. Et per hoc removetur quod studiosus non dicitur a virtute
denominative. Secundum est quod signicet eandem rem formaliter.
Tertium est quod differant in modo signicandi, ita quod unum signicet
per modum per se stantis, reliquum per modum adiacentis. Quartum
est quod abstractum differat (determinat ms.) per essentiam ab eo cui
concretum inheret. Et per hoc habetur quod non convenit proprie in
substantias, quia humanitas non differt per essentiam ab eo cui homo
52
Albertus Magnus, Liber de Praedicabilibus (B. Alberti Magni Opera omnia I), ed. A. Borgnet
(Paris, 1890) 158b; Thus these three conditions are required for denomination, namely,
a nature separate from a subject extrinsically attached and situated, the same thing
signied by both a principal [term] and a denominative [term], and different modes
of signifying.
53
Immo Albertus Magnus, Liber de Praedicamentis I,4 158b.
54
Regarding this part it should be noted rstly that four conditions are required
for denomination (according to Boethius ). The rst is that [the terms] agree at the
beginning and differ at the end; because of this studious is not said to be derived
denominatively from virtue. The second condition is that [the terms] signify the
same thing formally. The third condition is that they differ in mode of signifying, so
that one signies in the mode of a noun, the other in the mode of an adjective. The
fourth condition is that the abstract differs essentially from that in which the concrete
inheres; and because of this there is properly no concrete among substances, because
humanity does not differ essentially from that for which man is a concrete.
est concretum; vel oportet quod differant essentialiter. Et quod utrique
sit nomen impositum, propter hoc non possunt denominative dici sicut
homo ab humanitate, quia non differunt essentialiter.
55
Siger of Courtrai
Commentarium in Categorias
Ms. Venice M L. VI. 21 f. 109va
Notandum est secundum Albertum
56
quod ad hoc quod aliquid sit
denominativum tria requiruntur, scilicet identitas vocis a parte principii,
et diversitas a parte nis. Quia si esset totalis identitas, tunc esset aequivo-
cum et non denominativum, sicut apparet de musica prout dicitur de
scientia et muliere habente scientiam. Similiter si esset totalis diversitas
in voce, tunc tale nomen esset synonymum et non denominativum, ut
studiosus. Quod autem differant in ne potius quam in principio, ratio
est quia omnis diversitas est a forma, quia forma dat esse,
57
et distinguit
et separat. Etiam forma et nis coincidunt, intelligendo de ne intrinseco,
propter quod diversitas debuit esse ex parte nis.
Secundo requiritur identitas signicati.
Tertio diversitas modi signicandi, quia principale signicat rem suam
per modum entis absoluti, sed denominativum signicat idem per modum
inhaerentis et dependentis ad subiectum.
58
55
Next note that four conditions are required for the nature of denominatives. The
rst is that [the terms] are alike at the beginning of the word, and differ at the end;
because of this studious is not said to be derived denominatively from virtue. The
second condition is that [the terms] signify the same thing formally. The third condition
is that they differ in mode of signifying, so that one signies in the mode of a noun, the
other in the mode of an adjective. The fourth condition is that the abstract differs in
essence from that in which the concrete inheres; and because of this there is properly
no [denomination] among substances, because humanity does not differ in essence from
that in which man is a concrete (or, it is required that they differ essentially). And it is
not enough for denomination that the [same] name be assigned to them, like man and
humanity, because these two do not differ essentially.
56
Albertus Magnus, Liber de Praedicamentis I,4 158b.
57
Cf. Aristoteles, Metaph. VIII,5 1043a2; Hamesse , Auctoritates Aristotelis, p. 130 (189):
Unde forma dat esse rei.
58
It should be noted according to Albert that three conditions are required for
denomination. One is identity of word at the beginning, and difference at the end. If
there would be complete identity, then this would be a case of equivocation and not
denomination, just as music means both the science of music as well as a [musical]
woman knowing that science. Likewise, if there were complete diversity in the words,
then it would be a case of synonyms and not denominatives, like studious [and virtu-
ous]. The reason why [the terms] must differ at the end rather than at the beginning
is that all diversity arises from form (since form gives being), and form distinguishes
and separates; and form and [end] agree (meaning intrinsic end); so diversity must
arise because of the end.
The second condition is identity of what is signied.
The third [condition is] diversity in mode of signifying, so that the principle [term]
signies a thing in an absolute mode of being, but the denominative [term] signies
the same thing in a mode of inherence and dependence upon a subject.
114 robert andrews
Thomas de Erfordia
Expositio super librum Praedicamentorum
Mss. Erfurt SB Amplon. Q.266 [= E] f. 8vab; Leipzig UB 1356 [= L] f. 9va;
Mnchen SB Clm. 4378 [= M] f. 31r; Mnchen SB Clm. 14458 [= N] f.
119vb120ra
Item secundo notandum secundum Albertum
59
ad hoc quod aliquid sit
denominativum, tria requiruntur. Prima identitas vocis ex parte principii,
et diversitas ex parte nis. Quia si esset totalis identitas, tunc magis essent
aequivocum et non denominativum, sicut apparet de musica prout dicitur
de scientia et muliere habente scientiam musicae. Similiter si esset totalis
diversitas in voce, tunc tale nomen magis esset synonymum et multivocum
et non denominativum, sicut studiosus non dicitur denominative a virtute.
Quod autem magis debeat in ne differre quam in principio, ratio est
quia omnis diversitas est a forma, eo quod forma separat et distinguit;
sed forma et nis coincidunt, ut dicitur secundo Physicorum;
60
propter
quod diversitas debet esse a parte nis.
Secundo requiritur identitas signicati. Et propter hoc lucus non dici-
tur a luce denominative, quia lucus non convenit cum luce in signicato,
quia dicitur lucus quasi minime lucens, quia repugnat rationi lucis.
61
Tertio requiritur diversitas in modos signicandi, ita quod denominati-
vum sive concretum signicet per modum inhaerentis subiecto; principale
autem sive abstractum per modum distincti contra subiectum. Et propter
hoc obliqui casus non dicuntur denominative a recto; quia sicut rectus
signicat per modum adiacentis subiecto, vel distincti contra subiectum,
sic et obliqui ipsius.
62
59
Albertus Magnus, Liber de Praedicamentis I,4 158b.
60
Cf. Aristoteles, Physica II,7 198a247.
61
Cf. The Concise Oxford Dictionary, 5th ed. (1964), p. 726: lucus a non lucendo
Paradoxical derivation; . . . explanation by contraries. Cf. Maurus Servius Honoratus,
In Vergilii carmina commentarii, eds. G. Thilo et H. Hagen (Leipzig, 1881) I, 441: lucus
autem dicitur quod non luceat.
62
Secondly, it should be noted according to Albert that three conditions are required
for denomination. The rst is identity of word at the beginning, and difference at the
end. If there would be complete identity, then this would rather be a case of equivo-
cation and not denomination, just as it is clear that music means both the science
of music as well as a [musical] woman knowing that science. Likewise, if there were
complete diversity in the words, then it would be a case of synonyms or reformulations
rather than denominatives, just as studious is not derived denominatively from virtu-
ous. The reason why [the terms] must differ at the end rather than at the beginning
is that all diversity arises from form, since form separates and distinguishes; but form
and end agree, as is stated in the second book of the Physics (198a247); so diversity
must arise because of the end.
The second condition is identity of what is signied. Because of this condition grove
(lucus) is not derived denominatively from light (lucis), because grove and light do
not have the same signicationrather grove is called so because there is a lack of
light there, and is opposed to the nature of light.
These four passages show a great deal of similarities, but the differences
among them reveal their relative connections. The preponderance of
correlations occurs between Gerard and Peter, and between Siger and
Thomas. All agree that denominative words vary in spelling: for Gerard
and Peter they communicant (conveniant in voce, Peter) in principio et differunt
in ne; according to Siger and Thomas they exhibit identitas vocis a
parte principii, et diversitas a parte nis. As a second requirement all agree
that denominative words signify the same thing: for Gerard and Peter
they signicet eandem rem formaliter; according to Siger and Thomas they
exhibit identitas signicati. In describing the third requirement they all
adopt a version of Albert s phrase: for Gerard and Peter they differant
in modo signicandi; according to Siger and Thomas they exhibit diver-
sitas modi signicandi or diversitas in modos signicandi. Only Gerard and
Peter add a fourth requirement, that abstractum differat essentialiter ( per
essentiam, Peter) ab eo cui concretum inheret. In each of the two pairs, the
later author adds a few clarifying phrases to his predecessor. Even when
little of importance is changed, there seems to be a conscious effort to
re-word or to rearrange the phrasing. Small ourishes of originality
are valued even amid the extensive duplication.
Information does not always survive the transmission. Only Gerard
correctly identies the originator of the discussion, Boethius . Once
Peter drops the reference, it is left to Siger and Thomas to add credit
to the more proximate source, Albert .
9. Having a Wife (15b28)
Finally, I turn to two anonymous manuscripts which have been attrib-
uted to Peter of Saint-Amour . One is a fragment, Mnchen CLM
14763 f. 142, which Charles Lohr thought to be a version of Peter of
Saint-Amours Categories commentary.
63
The other one, Vatican Bibl.
Apost. Pal. lat. 1007 ff. 11ra21ra, was among those credited to Peter
The third requirement is diversity in mode of signifying, so that the denominative
or concrete [term] signies in a mode of inhering in a subject, while the principle or
abstract [term] signies in a mode distinct from the subject. Because of this oblique
cases are not said to be derived denominatively from the nominative case, because just
as the nominative case might signify its subject in the mode of an adjectivedistinct
from the subjectso might the oblique case signify it.
63
116 robert andrews
by Fr. Gauthier while cataloging the manuscripts of Thomas Aquinas
on the Perihermenias.
64
That these are misattributions can be shown by
reference to that rarest feature in medieval philosophya joke.
The humor arises in connection with the very end of the Categories
(in fact, the nal few lines of Aristotles text)perhaps when the
commentators were in a celebratory mood. Aristotle there offers eight
different senses of the word having (habere), such as to have a qual-
ity, have a quantity, have a possession, and so forth. The nal type of
having is the way in which a husband is said to have a wife (15b28).
Most commentators in the genre give a standard explanation of why
Aristotle lists this type of having last: because it is the least important.
This explanation is given in all four of the commentaries considered
aboveGerard of Nogent , Peter of Saint-Amour , Siger of Courtrai ,
and Thomas of Erfurt . However, only Peter of Saint-Amour adds that,
according to a certain truffator (that is, joker or fool), this is placed last
because when a man believes that he has a wife, he doesntrather,
some other man does.
Ms. Paris BN 1374 f. 34ra
Aut etiam vir (15b28). Hic ponit octavum modum habere. Et dicit quod
octavus modus ipsius habere est habere uxorem. Et secundum istum modum
illud quod habet etiam habetur, ut vir habet uxorem, et habetur ab uxore.
Et ideo dicit Philosophus de isto modo in comparatione ad alios, quod
iste est alienissimus in eo quod est habere. Et subdit rationem, quia nihil aliud est
habere uxorem quam cohabitare cum ea. Sed sicut vir cohabitat cum muli-
ere, ita mulier cum viro. Et ita utrumque dicitur habere secundum istum
modum; secundum autem alios, non; et ideo dicitur alienissimus.
Sed quidam truffator (tryfator var.) dicit quod iste est alienissimus
propter hoc quod quando vir credit habere uxorem, non habet; immo
quidam alius.
65
64
(Rome: Commissio Leonina, 1989). The misattribution is discussed in Andrea Tabarroni ,
Lo pseudo Egidio (Guglielmo Arnaldi) e uninedita continuazione del commento di
Tommaso al Perihermenias, Medioevo 14 (1988), 371427.
65
A man also [is said to have a wife]. Here [Aristotle] presents the eighth sense of to
have. And he says that the eighth sense of to have is to have a wife. In this sense he
who has is also had, as a man has a wife, and is had by a wife. Therefore Aristotle
says that this sense in comparison to the others is the most remote meaning of to have. He
appends the reason for this, because to have a wife is nothing other than to live with her.
In the two anonymous manuscripts we see a procedure typical for
literal commentaries, in which an original comment is elaborated and
revised. Not only do they repeat but also invert the jokeperhaps in
keeping with equality between the sexes. Here, a certain truffator says
that when a man believes that he has a wife, he doesnt have her,
because some other man does; and similarly, when a wife believes that
she has a husband, she does not have him because, sometimes, some
other woman does.
Anonymus
Expositio super librum Praedicamentorum (fragm.)
Ms. Mnchen CLM 14763 f. 142r
Dicit quidam trumphator dicaxque (dycitque? ms.) quod iste modus
alienissimus est; dixit quidam trumphator propter hoc <quod> quando
vir credit habere uxorem, non habet eam, sed alius (aliis ms.). Similiter
et uxor, cum credit habere virum, non habet ipsum, sed quaedam alia
mulier quandoque habet. In aliis autem modis non sic est; et ideo est
alienissimus.
66
Anonymus
Rationes super librum Praedicamentorum
Ms. Vatican Bibl. Apost. Pal. lat. 1007 f. 21ra
Et quidam truphator dixit quod ille modus alienissimus est propter
id quod quando vir credit habere uxorem, non habet eam, sed aliquis
alius. Similiter et uxor, cum credit habere virum, non habet (is add. et del.)
ipsum, sed aliqua alia mulier quandoque habet. In aliis autem modis non
sic est; ideo alienissimus est.
67
Just as a man lives with a wife, so does a wife live with a man. Each is said to have
the other reciprocally in this sense, but not in the other senses; therefore this sense is
said to be the most remote.
However, a certain joker says that this is the most remote sense because when a man
believes that he has a wife, he doesnt, because some other man does!
66
A certain joker and wit says that this is the most remote sense; and the joker said
this because when a man believes that he has a wife, he doesnt have her, but some
other man does. Likewise when a wife believes that she has a husband, she does not
have him, but sometimes some other woman does. This is true in no other sense, so
this is the most remote.
67
A certain joker said that this is the most remote sense because of the fact that
when a man believes that he has a wife, he doesnt have her, but some other man
does. Likewise when a wife believes that she has a husband, she does not have him,
but sometimes some other woman does. This is true in no other sense, so this is the
most remote.
118 robert andrews
This is surely one of the only jokes to be found in medieval philosophy.
That it is a joke is signaled by the appearance of the vernacular term
truffator, which also proves that all of the versions belongs to the same
tradition. Truffator (spelled here as tryfator, trumphator, and truphator)
was current in ancient French as trufeor,
68
and survives in English as both
trufe and trie. The expansions of the two anonymous manuscripts
are sufcient evidence that the works are not the same as the simpler
version of Peter of Saint-Amour . Indeed, their elaborations suggest
that they are subsequent to Peter, although all could have arisen inde-
pendently from an even earlier source. But certainly there is enough
evidence to treat all of these works as part of the same tradition.
Upon contemplating this commentary passage, I realized that perhaps
the truffator has it right, and that Aristotle himself intended his conclud-
ing remark to be a joke. It seems only natural that, after a long lecture,
one would end with a light-hearted remark, perhaps even a somewhat
risqu one. So the reason that Aristotle listed having a spouse as the
very last type of having may well have been his wry insight that this
is indeed a tenuous sort of possession.
The recognition of the interconnectedness of related literal com-
mentaries can help in dating, assigning provenance, and identifying
philosophical traditions. Understanding standard treatments can aid
in several respects, such as lling out illegible or missing script. Some
passages can even be collated across different authors. Most importantly,
literal philosophical commentaries cannot be appreciated in isolation.
Their value lies not in pure originality, but in accretion and the subtle
reworking of the materials of tradition.
68
A. J. Greimas , Dictionnaire de lancien franais jusquau milieu du XIV
e
sicle (Paris,
1987), p. 647.
THOMAS AQUINAS ON ESTABLISHING THE IDENTITY
OF ARISTOTLES CATEGORIES
Paul Symington
Providing a philosophical justication for the specic number and
identity of Aristotles categories is a task that dates back at least to
Simplicius s commentary on Aristotles Categories (ca. 6th century A.D.).
1

Scholastics from the thirteenth century onward addressed this issue,
which they called sufcientia praedicamentorum, mostly in commentaries on
Aristotles Categories.
2
Two related questions were pertinent. The rst
asked whether Aristotle provided an adequate list of categories and the
second asked whether a philosophical justication could be given for
the specic items on the list.
3
Although the latter task predates Albertus
Magnus (ca. 120880), he is credited as being the rst scholastic to
attempt it.
4
Albertus established a method of arriving at a list of the
1
For a recent translation of Simplicius s commentary on Aristotles Categories, see
Simplicius: On Aristotles Categories 14, trans. Michael Chase (Ithaca, 2003). See esp. pp.
7491.
2
Robert Andrews identifies other texts that offered opportunity for medieval
commentators to address the topic of the sufcientia, such as Aristotles Metaphysics V,
Physics III, and Topics I, in Question Commentaries on the Categories in the Thirteenth
Century, Medioevo 26 (2001), 292. Although Aquinas may be the rst scholastic to
refer to Simplicius , he does not seem to be familiar with Simpliciuss justication of
the number and identity of the categories. However, Radulphus Brito shows famil-
iarity with Simpliciuss treatment. See William E. McMahon , Radulphus Brito on
the Sufciency of the Categories, Cahiers de linstitut du moyen-Age grec et latin 39 (1981),
86. For a topical discussion of the various philosophical questions generated around
Aristotles Categories, see Jorge J. E. Gracia and Lloyd Newton , Medieval Theories
of Categories, Stanford Encyclopedia of Philosophy, 14 April 2006, <http://plato.stanford
.edu/entries/medieval-Categories> (8 August 2006).
3
See Giorgio Pini , Scotus on Deducing Aristotles Categories, La tradition mdivales
des Catgories (XII
e
X
ve
sicles): XIIIe Symposium europen de logique et de smantique mdivals,
eds. Jol Biard and Irne Rosier-Catach (Louvain, 2003), p. 24.
4
Albertus Magnus, Liber de Predicamentis 7.1, Alberti Magni Opera omnia, 1, ed. A.
Borgnet (Paris, 1890), pp. 27072. For a discussion of Albertus Magnus on the catego-
ries, see William E. McMahon , Albert the Great on the Semantics of the Categories
of Substance, Quantity, and Quality, Historiographia Linguistica 7, 1/2 (1980), 14557
and E. P. Bos and A. C. van der Helm , The Division of Being over the Categories
According to Albert the Great, Thomas Aquinas and Duns Scotus , in John Duns Scotus:
Renewal of Philosophy: Acts of the Third Symposium Organized by the Dutch Society for Medieval
Philosophy Medium Aevum (May 23 and 24, 1996), ed. E. P. Bos, (ELEMENTA: Schriften
zur Philosophie und ihrer Problemgeschichte) 72 (Amsterdam, 1998), pp. 18396.
120 paul symington
categories from the modes of predication. Thomas Aquinas followed
Albertuss lead and tried to establish an exhaustive categorial division of
being by showing that there are only so many ways in which predicates
are said of subjects.
5
Although Aquinas never wrote a commentary on
Aristotles Categories, he provided two closely related justications for the
list of categories ( praedicamenta) in his commentaries on the Metaphysics
(Book V, lect. 9) and the Physics (Book III, lect. 3).
6
Moreover, Aquinass
account was historically more inuential than Albert s, even though
Aquinass discussions of the topic are relatively brief.
Although the division of the categories is metaphysically fundamental
and interesting, as well as hotly debated among scholastics, the second-
ary literature on Aquinas s contribution to the problem of sufcientia
praedicamentorum is scant. In fact, not only is the literature devoted spe-
cically to it meager, but the discussion is often altogether ignored in
broad discussions of Aquinass metaphysics.
7
This is not surprising given
Aquinass brief and elliptical treatment of it. The issue, however, has
not escaped the attention of a few scholars, including John Wippel , who
has provided the most detailed treatment of Aquinass view to date.
The purpose of this article is to offer an alternative interpretation
of Aquinas s view. The key passage is his statement that those things
that are said to be secundum se which signify in every manner the g-
ures of predication [i.e., the modes of predication].
8
I argue that this
passage is crucial to a proper understanding of Aquinass derivation
5
In Categories in Aristotle, Michael Frede interprets the Aristotelian categories
as kinds of predications rather than kinds of predicates. In Studies in Aristotle, ed.
Dominic J. OMeara (Washington, D.C., 1981), pp. 124. He also claims that in
Aristotles works there is not any sign of a systematical derivation of the categories,
e.g., in terms of a set of formal features (p. 22). In contrast, Aquinas holds 1) that the
categories ( praedicamenta) are substance and nine accidents, and 2) the list of categories
can be established according to formal features of propositions (namely, the modes of
predication or guras praedicationis).
6
Thomas Aquinas , In duodecim libros Metaphysicorum Aristotelis expositio, eds. M. R.
Cathala and R. M. Spiazzi (Turin, 1950); hereafter cited as In Met. Through the
chapter, I either provide my own translations of the text or modications of John P.
Rowans translation, Commentary on Aristotles Metaphysics: St. Thomas Aquinas (Notre
Dame, 1995). Thomas Aquinas, In octo libros Physicorum Aristotelis expositio, ed. P. M.
Maggilo (Turin, 1954); hereafter cited as In Phys.
7
See, for example, Eleonore Stump , Aquinas (The Arguments of the Philosophers), (New
York, 2005) and Robert Pasnau and Christopher Shields , The Philosophy of Aquinas
(Boulder, 2004). Etienne Gilson also does not mention it in Being and Some Philosophers
(Toronto, 1952), p. 55.
8
In Met. 5.9, n. 889: quod illa dicuntur esse secundum se, quaecumque signicant
guras praedicationis. For the passage of Aristotle to which Aquinas refers, see
Metaphysics 5.7.
aquinas on the identity of a.s categories 121
of the categories and that Wippels account does not adequately take
it into consideration. This passage indicates a key feature of Aquinass
approach: that the categories are identied and distinguished from
each other based on essential propositions, i.e., secundum se or per se
propositions, which have predicates that are essentially related to their
subjects. Consequently, in the rst section of this chapter, I present
Wippel s interpretation of what Aquinas means by secundum se as it
relates to the modes of predication and I point out some difculties
with it.
9
To be more precise, problems arise with Wippels interpreta-
tion because he does not take into account Aquinass focus on per se
modes of predication. In the second section of this chapter, I offer an
interpretation of what Aquinas means by secundum se when he describes
the modes of predication from which the categories are established.
Specically, I interpret secundum se to refer to three of the four per se
modes of predication as they are discussed in the Posterior Analytics:
primo modo, secundo modo, and quarto modo per se predication.
10
My claim
is that Aquinas determines the number of categories by reecting on
the ways in which the predicates of per se propositions are related to
the subjects of other such per se propositions. Finally, in the third sec-
tion I show how Aquinas establishes the categories from the modes of
per se predication. For the sake of brevity, I focus mainly on substance,
quantity, and quality and provide only a sketch of how Aquinas deals
with the remaining categories.
9
John F. Wippel , Thomas Aquinas s Derivation of the Aristotelian Categories
(Predicaments), Journal of the History of Philosophy 25 (1987), 1334 and The Metaphysical
Thought of Thomas Aquinas (Washington, D.C., 2000), pp. 20828. Other scholars with
similar interpretations of Aquinass view are Giorgio Pini , Scotus on Deducing
Aristotles Categories, and E. P. Bos and A. C. van der Helm , The Division of
Being over the Categories According to Albert the Great, Thomas Aquinas and Duns
Scotus.
10
Thomas Aquinas , Expositio Libri Posteriorum, Sancti Thomae de Aquino Opera omnia,
Leonine edition (Rome, 1989), 1.2; hereafter cited as Post. An. The fact that the com-
mentary on the Posterior Analytics is believed to have been written roughly at the same
time as the commentary on the Metaphysics (between 126972) makes it pertinent for
our discussion. See The Cambridge Companion to Aquinas, eds. Norman Kretzmann and
Eleonore Stump (New York, 1993), p. 283. In addition, both texts discuss modes of
predication. Also, although these three modes of predication are awkwardly named (I
refer to three modes of predication but there is no tertio modo), I follow the convention
in the secondary literature of naming each specic mode according to the Latin ordinal
used in the text. See William M. Walton , The Second Mode of Necessary or Per Se
Propositions According to St. Thomas Aquinas, The Modern Schoolman 29 (195152),
293306. Aquinas identies tertio modo in Post. An. 1.10 as not a mode of predication
but rather one way to understand the term per. For a similar interpretation of secundum
se, see Ralph McInerny, Being and Predication, in his Being and Predication: Thomistic
Interpretations (Washington, D.C., 1986), pp. 173228.
122 paul symington
I. John Wippel s Interpretation
In this section I focus exclusively on Wippel s interpretation of Aquinas s
derivation as it is presented in his commentary on the Metaphysics V,
9, focusing specically on Wippels understanding of Aquinass state-
ment that those things are said to be secundum se that signify in every
manner the modes of predication.
11
I also criticize Wippels inter-
pretation insofar as it is neither sufcient for establishing the number
and identity of the categories from modes of predication, as Aquinas
claims, nor compatible with the general context of Aquinass discus-
sion. To be fair to Wippel, however, it should be noted that he offers
an accurate overview of the text in which Aquinas establishes the list
of categories. He notes that Aquinas identies three ways in which
a predicate can be related to its subject in a proposition. In the rst
way, (1) the predicate is really identical with that which serves as
the subject,
12
and these propositions signify substance. The example
that Wippel provides to illustrate this is Socrates is an animal. In a
second way, (2) a predicate may be taken from something which is in
the subject.
13
If the predicate is absolutely in the subject and follows
from the matter (2a), then the category of quantity results, but if it
follows from the form (2b) of the subject, then the category of quality
results. Wippel does not give examples of, nor discusses how, a predicate
is understood to be in a subject either according to the matter of the
subject or according to its form. It is possible that he would say that
Socrates is ve-feet tall and Socrates is bald are examples in which
the predicate is said to be in the subject because the former signies
how much the subject is and the latter how the subject is. If, however,
the predicate is taken not absolutely but in relation to something other
than the subject (2c), then relation is expressed. Although Wippel does
not provide an example, he might say that Socrates is the teacher of
Plato is an example of this.
In a third way, (3) a predicate may be derived from something which
is realized outside the subject.
14
Again, Wippel provides no examples to
illustrate this, nor does he discuss how a predicate is understood to be
realized outside the subject. Rather, he simply claims that the various
11
In Met. 5.9, n. 889.
12
Wippel , Metaphysical Thought of Thomas Aquinas , p. 213.
13
Ibid., p. 214.
14
Ibid.
ways in which the predicate is denominated by that which is realized
outside the subject yield the remaining categories. For example, if the
predicate in any way measures the subject (3a
c), in combination with

other conditions, then the categories of place (e.g., Socrates is in the
marketplace), time (e.g., The play is at dusk) and position (e.g.,
Socrates is sitting) result. If the predicate does not measure the subject
(3d), then habit (e.g., Socrates is shod) results. The last two categories,
action (e.g., Socrates is cutting) and passion (e.g., The paper is being
cut) are signied by those predicates that are only partly present in
the subject (3ef ).
15
Wippel cites an earlier passage in which Aquinas states that those
things are said to be secundum se that signify in every manner the modes
of predication.
16
He notes that Thomas begins [the derivation] with
the observation that according to Aristotle those things are said to
be in the proper sense (secundum se rather than only per accidens) which
signify different gures of predication [i.e., modes of predication].
17

However, Wippel does not expand any further on this passage, leaving
us with three questions. First, what does he take gures of predication
to mean? Second, what is the antecedent to those things, i.e., to what
do those things refer? Do they refer to predicates, subjects, the
various signications of the copula, propositions, or something else?
Third, what does it mean for something to be said to be secundum se (in
a proper sense) and how is this different from that which is said to be
per accidens? I will address each question in turn.
Figures of Predication
In discussing the meaning of gures of predication, Wippel refers to
both modes of predicating (modus praedicandi ) and modes of predica-
tion (modus praedicationis). Wippel does not distinguish between these
two terms and seems to use them interchangeably.
18
He simply says
that the categories are distinguished according to different modes or
ways of predicating.
15
Ibid., p. 215.
16
In Met. 5.9, n. 889.
17
Wippel , Metaphysical Thought of Thomas Aquinas , p. 212.
18
In keeping with Aristotles and Aquinass general theory of psychology, I take it
that a proposition is the object of the second act of the intellect, and that predicating
is the activity itself. Thus, predication would be the abstract form of predicating.
124 paul symington
Of those names which are predicated, continues Thomas, some signify
what something is, that is to say, substance. Some signify how it is (qual-
ity), and others how much there is (quantity), and so on. Therefore,
in accord with each of these supreme modes of predicating, esse must
signify the same thing, i.e., what something is, or what kind it is, or how
much there is, etc. For instance when we say that man is an animal, the
term is signies substance. When we say a man is white, the verb is
signies quality.
19
In this passage, Wippel identies modes of predicating as the basic
ways in which a predicate says something of its subject and points
out that each category is distinguished according to a different mode
of predicating. A predicate that expresses what the subject is signies
substance, one that expresses how the subject is expresses quality, etc.
For example, since both Socrates is an animal and Bucephalus is
a horse have predicates that say what the subject is, the esse of each
proposition must signify substance. Likewise, Socrates is white and
Socrates is bald signify quality because they have predicates that
show how the subject is. To put it another way, white and bald
are answers to the question How is Socrates? Thus, two propositions
that have predicates that relate to their subjects in the same way (e.g.,
both propositions have predicates that signify what the subject is) signify
the same category. When Wippel refers to the signication of esse, he
means the copula is in a proposition. In a simple subject-predicate
proposition the copula functions to unify the subject and predicate
in a single signication, so that a proposition as a whole can signify
a substance or accident according to the sense of the predicate. For
example, the proposition Socrates is an animal signies substance,
and Socrates is white signies an accident (viz., quality).
The Antecedent of those things
This understanding of what Wippel means by modes of predication
or modes of predicating helps us to understand the second question
posed above concerning how Wippel understands those things in
Aquinas s statement that those things are said to be secundum se that
signify in every manner the modes of predication.
20
Although Wippel
does not identify the antecedent of those things, his brief treatment
19
Ibid.
20
In Met. 5.9, n. 889.
of the issue seems to suggests that the modes of predication are the
various ways in which predicates are related to subjects. Therefore, it
would seem that Wippel understands those things (illa) to mean the
modes of predication that he identies, such as what the subject is,
how the subject is, how much the subject is, etc. However, if this
were the case, Aquinass passage would have the following interpreta-
tion: the specic modes of predication are said to be secundum se that
signify in every manner the modes of predication (and the categories
are established from them). That is, the modes of predication that
Wippel identies 1) are the modes of predication that are said to be
in a proper sense and 2) are the modes of predication from which the
categories are derived.
Secundum se
This interpretation in turn provides us with an answer to the question
concerning what it means for modes of predication to be said secundum
se. Wippel points out that secundum se is opposed to per accidens and trans-
lates the former as in a proper sense. Thus, it would be reasonable
to infer from this that per accidens in this context may be translated as
in an improper sense. Consequently, in the passage in question, in a
proper sense qualies modes of predication in that only propositions
that have predicates that are related to their subjects according to one
of the ten supreme modes of predication are secundum se. For example,
a proposition that is secundum se is one in which the predicate says what
the subject is or is one in which the predicate says how the subject is,
etc. Also, those propositions that have predicates that are not related
to their subjects according to one of these ways are secundum accidens.
This indicates how, for Wippel, categories are distinguished according
to modes of predication. Socrates is a man, Bucephalus is a horse,
and A man is rational, signify substance because they have predicates
that show what their subjects are and Socrates is white signies qual-
ity because the predicate says how the subject is.
Now let me turn to some problems with Wippel s understanding of
Aquinas s view. First, Wippel seems to present two independent ways of
distinguishing or deriving the categories.
21
This could be called the
21
It may be controversial to interpret Wippel as equating the terms deriving and
distinguishing. However, there are several reasons why I believe he does so. 1) He
126 paul symington
problem of two derivations. On one hand, he says that the categories
are distinguished according to the different modes or ways of predi-
cating,
22
and then identies the mode of predication specic to each
category. For example, propositions with predicates that express what the
subject is signify substance (e.g., man is an animal) and propositions
that express how the subject is signify quality (e.g., Socrates is bald).
On the other hand, he presents the derivation of the categories
from propositions that express a relationship between the subject and
predicate according to three ways: In one way, the predicate is really
identical with that which serves as the subject;
23
in a second way, a
predicate may be taken from something which is in the subject;
24
and
in a third major way, a predicate may be derived from something which
is realized outside the subject.
25
Thus, it appears that he has a problem
of two derivations. That is, he seems to have two accounts for how
the list of categories is established. This problem arises because there
is no discernable relation between the so-called ten supreme modes of
predication and the three ways in which a predicate is related to its
subject. This is problematic because Aquinas only presents one way
of establishing the categories in the text. Wippel does not discuss how
the modes of predicating that show how the subject is or how much
the subject is are related to that in which the predicate is taken from
something in the subject. That is, if Socrates is white distinguishes
the category of quality because it shows how the subject is, what is
the signicance of Aquinas saying that white as understood in the
never denes derivation but he does use both terms in very similar contexts: Thomas
would have us appeal . . . to diversity in modes of predication in order to render explicit
the distinction between substance and accident in general, and also to derive the nine
supreme classes of accidents (p. 211). It would make sense to say that just as sub-
stance is distinguished from accident, so too the nine categories are distinguished from
each other as well. 2) Aquinas never uses a term that corresponds to Wippels term
derivation. Rather, Aquinas uses terms such as distinguuntur and dividitur to discuss the
so-called derivation of the categories. Thus, when Wippel uses the term distinguish in
the context of the categories it is reasonable to hold that he uses it interchangably with
the term derive. 3) He states that the categories are distinguished according to different
modes or ways of predicating (p. 212) and immediately after identifying these modes
of predicating through which the categories are distinguished he proceeds directly to
discuss the derivation of the categories. It is difcult to understand how the categories
are distinguished and derived from the modes of predication if distinguished and
derived do not have the same meaning.
22
Ibid., p. 212.
23
Ibid., 213.
24
Ibid., 214.
25
Ibid.
subject Socrates according to the form of the subject is a way of
distinguishing the category of quality from other categories? Indeed,
it appears that the rst account presented by Wippel obviates the need
for the second one.
A second problem with Wippel s interpretation specically pertains
to the way that Aquinas establishes the category of substance. Wippel
suggests that substance is derived when a predicate is really identi-
ed with the subject.
26
But it seems that this is the same as the mode
of predication in which the predicate says what the subject is. For
example, in Socrates is an animal, Socrates is really identied with
animal and animal says what the subject Socrates is. Given these
factors, according to Wippel, the proposition signies substance. This
is in contrast to Socrates is white, which does not signify substance
because the predicate is neither identied with the subject nor does it
show what the subject is.
However, not only is the condition in which the predicate is really
identied with the subject not mentioned by Aquinas , but it is also
insufcient for establishing the category of substance. For it cannot be
used to distinguish propositions signifying substance from propositions
that do not. This is because there are propositions that have predicates
that are identied with their subject and yet do not signify substance.
Consider the following examples: White is a color, A surface is a
continuous quantity, or This patch of color is red. In these propo-
sitions the predicate is identied with the subject and shows what the
subject is. White is really identied with color in the sense that
white really is a color. Also, a particular patch of color that is red is
such that red is identied with that patch of color. But none of these
propositions directly signies substance. Thus Wippel s suggestion is not
sufcient for establishing the category of substance. This is a problem
because Aquinass procedure aims to distinguish propositions from each
other that signify different categories from a consideration of the way
in which predicates are related to subjects.
27
The fact that Wippels
interpretation does not take this into account leads one to infer that
he has misinterpreted Aquinass position.
26
Ibid., p. 213. All Aquinas says is that in one way a predicate states what the
subject is, as when I say Socrates is an animal. In Met 5.9, n. 891.
27
In Met. 5.9, n. 890: Et propter hoc ea in quae dividitur ens primo, dicuntur esse
praedicamenta, quia distinguuntur secundum diversum modum praedicandi.
128 paul symington
Third, Wippel does not offer an interpretation of how Aquinas estab-
lishes the categories that is consistent with the overall meaning of the
text. As a consequence, Wippels interpretation fails to show how all ten
categories are said to be entia secundum se, which is a fundamental claim
of Aquinas in the text. In lect. 9, Aquinas is commenting on Aristotles
text in which he discusses the various ways that being is said.
28
The
text is divided into the following distinct sections. 1) In n. 885, Aquinas
provides an overarching division in which being is said (ens dicitur) either
secundum se or secundum accidens.
29
Furthermore, it is important to note that
in this section Aquinas says that these two ways that being is said are not
the same as the division of being into substance and accident.
30
Rather,
he says that the division of being secundum se and secundum accidens is
understood according to whether something is predicated of something
else either essentially [ per se] or accidentally [ per accidens].
31
2) Next,
in nn. 88688, Aquinas subdivides the latter division and discusses the
nature of being that is said accidentally (secundum accidens); namely, that
which is predicated accidentally. A man is musical and Socrates is
white are beings that are said secundum accidens because they predicate
accidents of substances and their predicates are not essentially predi-
cated of their subjects.
32
3) Finally, in nn. 88994, Thomas turns to
the former division and discusses that which is said to be secundum se.
It is in this context that he states that those things that are said to be
secundum se in every manner signify the modes of predication
33
and it
is in this context that he establishes the list of categories (n. 892).
An understanding of how the text is divided and how the so-called
derivation text fits into it reveals some significant problems with
Wippel s interpretation. On the one hand, Wippel states that the mode
of predication that reveals the category of quality is that in which the
predicate shows how the subject is. Therefore, since Socrates is white
and Socrates is musical are propositions that have predicates that
show how the subject is, they must signify the category of quality. On
28
In Met. 5.9, n. 885: Hic Philosophus distinguit quot modis dicitur ens.
29
Ibid.: ens dicitur quoddam secundum se, et quoddam secundum accidens.
30
Ibid.: Sciendum tamen est quod illa divisio entis non est eadem cum illa divisione
qua dividitur ens in substantiam et accidens.
31
Ibid.: Unde patet quod divisio entis secundum se et secundum accidens, attenditur
secundum quod aliquid praedicatur de aliquo per se vel per accidens.
32
Ibid., n. 886: Ostendit quot modis dicitur ens per accidens . . . cum accidens
praedicatur de subiecto, ut cum dicitur, homo est musicus.
33
Ibid., n. 889.
the other hand, Aquinas states that being secundum se is divided into
the ten categories, of which nine are accidental kinds
34
and being
secundum se refers to essential predication. However, both Socrates is
white and Socrates is musical are said secundum accidens. Therefore,
since propositions that have predicates that say how the subject is have
predicates that are predicated accidentally (secundum accidens), this cannot
be the mode of predication from which quality is established. Rather,
the mode of predication from which all the categories are derived must
be related to essential predication.
35
Finally, I question Wippel s overall understanding of Aquinas s
approach to establishing the categories. I disagree with Wippels sug-
gestion that Aquinass text provides a way to discover the categories,
if by discover Wippel means that the technique for distinguishing the
categories is in itself a way of determining the meaning of each of
the categories. That is, since the term discover has the connotation
of being made known, the suggestion that through the technique
presented by Aquinas one discovers denitions or characteristics of
specic categories is misguided. Rather than discovering the categories,
Aquinass technique merely distinguishes or divides propositions that
signify different categories according to the logical properties of propo-
sitions. Wippel seems to hold that for Aquinas Socrates is human
is a way of understanding the meaning of substance signied by the
proposition because the predicate says what the subject is. In contrast, I
34
Ibid., n. 885: Quod ex hoc patet, quia ipse postmodum, ens secundum se dividit
in decem praedicamenta, quorum novem sunt de genere accidentis.
35
Pini follows Wippel in his interpretation in Scotus on Deducing Aristotles
Categories, p. 26: in the predication man is white, is signies a quality, and so on.
Since there are ten kinds of predicatesomething Aquinas demonstrated too . . . there
are ten different meanings of the verb to be. . . . Hence, since predicates are classied
into ten genera according to what they signify, he could conclude that there are ten
genera of being. Earlier in the text Pini states that Since a predicate can be attrib-
uted to its subject in ten different ways, he could conclude that there were ten differ-
ent modes of being (pp. 25, 26). The same criticism applies to Pinis interpretation:
although Aquinas holds that the is in the proposition Socrates is white expresses a
metaphysical accident, such a proposition cannot be used to deduce the category of
quality. This is similar to the interpretation given by Bos and van der Helm in The
Division of Being over the Categories, pp. 18789. Two other treatments of Aquinass
on the sufcientia praedicamentorum are of note. They follow a similar line of interpreta-
tion as Wippel: M. Marina Scheu in The Categories of Being in Aristotle and St. Thomas
(Washington D.C., 1944), pp. 606; and Stanislas Breton , La dduction thomiste
des catgories, Revue philosophique de Louvain 60 (1962), 532. Scheu stresses Aquinass
logical approach in deriving the modes of being (p. 63). However, Scheu treatment is
brief and does not expand on the various modes of predication.
130 paul symington
hold that Aquinas identies Socrates is human as signifying substance
in the following way: 1) every proposition that has a given logical form
signies substance (what this logical form specically is will be discussed
in the third section); 2) Socrates is human has that given logical form;
3) therefore, Socrates is human signies substance. Aquinas does not
provide a way of discovering what propositions signifying categories
mean (e.g., that Socrates is human signies substance, which means
what the subject is) but rather a way of distinguishing propositions
from each other in a way that parallels the categorial division of being.
In support of this, Aquinas states that that into which being is rst
divided (dividitur) are called categories because they are distinguished
(distinguuntur) according to different modes of predicating.
36
In addition
to this, I contend that Aquinas emphatically does not establish the divi-
sion of being into the categories by identifying the ten supreme modes
of predicating (what the subject is, how the subject is, etc.) but
rather establishes the categories through the three ways that predicates
may be related to their subjects.
II. Per se Predication and the Identity of the Categories
The starting point for my interpretation of the way that Aquinas estab-
lishes the categories is based on the passage in which he says that being
said secundum se and secundum accidens is understood according to whether
something is predicated of something else either essentially [per se] or
accidentally [per accidens].
37
I claim that the categories are established
through the modes of per se (i.e., essential) predication.
38
In this section
I develop some points crucial to understanding Aquinass procedure for
establishing the categories from per se modes of predication; namely,
that a successful way to establish the categories is by identifying dif-
ferences in the predicational structure of per se propositions.
39
Next, I
36
In Met. 5.9, n. 890. My emphasis.
37
In Met. 5.9, n. 885. Later in his commentary (n. 1054), Aquinas specically identies
the various meanings of secundum se with the various meanings of per se predication
as it is found in the Posterior Analytics.
38
For a discussion of the various senses of modus in Aquinas , especially as it plays a
role in his metaphysics, see John Tomarchio , Aquinass Division of Being According
to Modes of Existing, Review of Metaphysics 54, 3 (2001), 585613.
39
For the difference between per se propositions and per se predication, see fn. 18
above.
introduce the various per se modes of predicationnamely, primo modo,
secundo modo and quarto modo per seand distinguish them from the per
accidens modes. Each mode of predication indicates how a predicate
is related to its subject. Furthermore, I show that each per se mode of
predication satises the two criteria of per se predication; viz.: 1) the
predicate is predicated universally of the subject and 2) the subject,
or something possessed by the subject, is the cause of the predicate.
This provides a foundation for the third section of this chapter (III) in
which I discuss how Aquinas establishes the various categories from per
se modes of predication.
40
As previously mentioned, Aquinass derivation of the categories
that I am focusing on is found in the fth chapter of his commentary
on the Metaphysics. In order to understand Aquinas properly, however,
one must consider the larger context of the passage. The general issue
that Aquinas addresses in this section of his commentary concerns the
various signications of being; that is, the basic senses of the term.
41

He begins by noting that for Aristotle the term ens signies either ens
per se or ens per accidens.
42
This is the division of being into substance
( per se) and accidents ( per accidens). However, Aquinas makes another
distinction of ens into being that is said secundum se and that which is
said secundum accidens. This latter division of being refers to the various
senses of the copula (is) in a proposition. Because the copula of a
proposition is a kind of being (is is a cognate of being), being that is
said secundum se refers to propositions in which the predicate is essentially
or per se predicated of its subject.
There are two important features of the copula. First, it signies
what the predicate and subject taken together signify.
43
For example, the
copula of Socrates is an animal signies Socrates-the-animal (the spe-
cic unied extramental being). Second, the copula is classied or char-
acterized according to the way in which the predicate of a proposition
40
The textual basis for my interpretation is mainly the Commentary on the Metaphysics
V, 9. Due to its similarity with the passage in Physics III, 5, I use the latter parallel text
only for purposes of amplication.
41
In Met. 5.9, n. 885.
42
Aristotle, Metaphysics 5.7 (1017a8).
43
See: Gyula Klima , Aquinas Theory of the Copula, Logical Analysis and History of
Philosophy 5 (2002) and The Semantic Principles Underlying Saint Thomas Aquinass
of Being, Medieval Philosophy and Theology 5 (1996), 87141; Pini , Scotus on Assertion
and the Copula: A Comparison with Aquinas, in Medieval Theories on Assertive and Non-
Assertive Language. Acts of the 14th European Symposium on Medieval Logic and Semantics, Rome,
June 1115, 2002, eds. A. Maier and L. Valente (Firenze, 2004), 30731.
132 paul symington
is related to its subject. For example, the copula in Socrates is an
animal is an ens secundum se because the proposition has a predicate
that is essentially related to its subject. The reason why the copula can
be characterized in this way is because it has the specic function of
relating the predicate to the subject in a proposition. The ways in which
the copula relates the predicate to the subject can be characterized in
different ways. An understanding of the two features of the copula
helps us to interpret Aquinas s statement that Aristotle divides being
secundum se into the ten categories, of which nine are accidents.
44
By
this, Aquinas means that even though the nine accidents of being are per
accidens, they are entia secundum se in the sense that they can be signied
by propositions that have secundum se copulas. That is, all ten categories
can be signied by being that is said secundum se because any proposi-
tion in which an essential predication occurs has a copula that signies
any one of the ten categories, including any of the nine accidental
categories. For example, White is a color has a copula that is both
secundum se and signies the category of quality (which is an ens per
accidens), whereas Socrates is white is a proposition that has a copula
that signies a per accidens being and is itself secundum accidens. Similarly,
Socrates is an animal has a copula that is secundum se and signies a per
se being (substance). Although a substance is an ens per se and accidents
are ens per accidens, both substance and accidents can be signied by
propositions with copulas that are entia secundum se. To simplify matters,
instead of referring to a propositions copula, I will use the phrase a
per se predication or a per se proposition to refer to a proposition in
which the predicate is essentially related to its subject.
45
When Aquinas states that those things that are said secundum se signify
the gures of predication [i.e., modes of predication], he means that
per se propositions or predications provide an understanding from which
the categories are established. When this interpretation of secundum
se (meaning per se predication) is coupled with Aquinass statement
that being must be narrowed down (contrahitur) to diverse kinds (viz.,
categories) according to a different mode of predication [from that of
genus and species] because being is said in as many ways as the ways
44
In Met. 5.9, n. 885.
45
Following Wippel, I will use these two terms interchangeably.
in which something is predicated,
46
the result is that the categories are
distinguished from each other in the following way: propositions that
signify different categories are divided from each other by identifying
differences among the modes of per se predication. This procedure
shows how being is divided into categories by dividing propositions
that signify different categories by distinguishing per se propositions from
each other according to different ways that a predicate can be per se
predicated of subjects.
Although it may be counter-intuitive to interpret secundum se as refer-
ring exclusively to propositions in which the predicate is essentially
or per se related to its subject, there are several reasons to do so. The
rst is that it heeds Aquinas s explicit admonition that being is said
in a certain way according to itself and in a certain way according to
accidents, nevertheless, it must be noted that that part of being is not
the same with that division in which being is divided into substance
and accidents.
47
That is, it avoids conating the meaning of ens per se
with ens secundum se; per se being refers to substance and being secundum
se refers to a proposition in which there is an essential relation between
subject and predicate.
Second, it resolves the issue of the sense in which the nine accidents
can be understood as being secundum se. This interpretation allows us to
understand how the nine accidents are on the one hand not entia per se,
because this applies to substance alone (and accidental being is depen-
dent on the being of substance), and on the other hand, because the
nine categories are still fundamental extramental entities, not reducible
to anything else (e.g., not reducible to entia per se); they are fundamental
entities that can be identied through, and correspond to, fundamental
differences among per se propositions.
Third, it allows Aquinas s argument for the categories to avoid the
charge of a non sequitur fallacy. Aquinas would indeed be guilty of such
a charge if he held that The categories are established from the modes
of predication, follows from The categories are established by seeing
which propositions happen to signify which categories; for example,
that Socrates is an animal, establishes the category of substance
because it signies substance. Merely stating that a given proposition
46
In Met. 5.9, n. 890: Unde oportet, quod ens contrahatur ad diversa genera
secundum diversum modum praedicandi, qui consequitur diversum modum essendi;
quia quoties ens dictur, idest quot modis aliquid praedicatur.
47
In Met. 5.9, n. 885.
134 paul symington
signies substance is not the same as establishing the category of
substance from a mode of predication. For in this case, no appeal is
made to the way in which the predicate is related to its subject; rather,
appeal is made only to the metaphysical category a particular propo-
sition happens to signify. Rather, in order for the categories to be
established, it must be shown how propositions having copulas which
signify a specic mode of per se predication can be used to distinguish
different categories.
In contrast to Wippels interpretation, Aquinas establishes the ten
categories by reecting on the three per se modes of predication as they
are established in the Posterior Analytics I.
48
There he gives two criteria
of per se predication. According to one, a predicate must be universally
predicated of its subjectmeaning that the predicate is found in each
of the things that are included in its subject.
49
According to the second
criteria, the subject or something possessed by the subject [must be]
the cause of the predicate.
50
This second criterion is based on the fact
that since the term per (by) signies a causal relation it is necessary
that there be a causal feature for all per se predications.
51
In contrast,
accidental predication obtains when at least one of these two criteria
are not met.
Aquinas goes on to say that there are three general modes of predi-
cation that meet these two criteria: primo modo, secundo modo and quarto
modo.
52
Primo modo per se predication obtains when the denition or some-
thing posited in the denition is predicated of something.
53
Examples
of this mode occur in Socrates is a man, White is a color and
Humans are animals. This mode meets the two criteria because every
48
As previously mentioned, Aquinas will subsequently identify secundum se predication
with per se predication in his Commentary on the Metaphysics, nn. 10541567.
49
Post. An. 1.9, lines 47, 48.
50
Post. An. 1.10, lines 1924: Sicut autem hec prepositio per designat habitudinem
cause quando aliquid extraneum est causa eius quod attribuitur subiecto, ita quando
subiectum uel aliquid eius est causa eius quod attribuitur ei, et hoc signicat per se.
51
Post. An. 1.10, lines 8, 9: Circa primum sciendum est quod hec prepositio per
designat habitudinem cause.
52
Aquinas does mention another sense in which something is per se (viz., tertio modo).
However, this mode is not a mode of predicating, but rather a mode of existing. For
in this sense, the per se signies something that is alone, as something singular in the
genus of substance. Post. An. 1.10, lines 11721.
53
Post An. 1.10, lines 2530: Primus ergo modus dicendi per se est quando id
quod attribuitur alicui pertinet ad formam eius, et quia difnitio signicat formam et
essentiam rei, primus modus eius quod est per se est quando predicatur de aliquo
difnitio uel aliquid in difnitione positum.
proposition of this form has a predicate that is predicated universally
of its subject insofar as the predicate is contained in the denition of
the subject. Also, the predicate is linked to its subject causally insofar
as the predicate signies the essence of the subject.
Secundo modo per se predication obtains when the subject is posited in
the denition of a predicate, which is a proper accident of the subject.
54

This mode occurs in A surface is colored and Humans have the
capacity to laugh. A characteristic of this per se mode of prediction is
that it involves predicates that are property terms ( propria). This mode
of predication meets the two criteria as well. First, the predicate is
universally predicated of its subject; for example, every surface has the
predicate colored truly predicated of it. Second, the predicate is caus-
ally linked to the subject in that even though the predicate is not in the
denition of the subject, the subject is signied in the denition of the
predicate; in A surface is colored, the denition of colored includes
the notion of surface. Because of this, this mode of predication is
determined by a relationship between the predicate and the subject in
which the predicate is understood to be in the subject.
Quarto modo per se, the most obscure of the modes, obtains when the
preposition per designates a condition of efcient cause or other . . . [ but]
the predicate is in fact in the subject on account of itself.
55
Aquinas
says that this mode of predication occurs in Having been slaughtered,
it died. This mode of predication can also be understood to meet the
two criteria of having a predicate that is universally predicated of its
subject and having a predicate that is understood to be causally linked
to its subject. First, in the example given, the predicate it died is
universally predicated of that which has been slaughtered. Second,
the predicate is linked to the subject according to an efcient cause: the
slaughtering is understood as the efcient cause of the death of the
animal. This per se mode of predication is said to signify an extrinsic
54
Post An. 1.10, lines 6467: Unde secundus modus dicendi per se est quando
subiectum ponitur in difnitione predicati quod est proprium accidens eius.
55
Post. An. 1.10, lines 12235: Deinde cum dicit: item alio modo etc., ponit quartum
modum, secundum quod hec prepositio per designat habitudinem cause efcientis uel
cuiuscunque alterius <extrinsece>. Et ideo dicit quod quicquid inest unicuique propter
se ipsum, per se dicitur de eo, quod uero non propter ipsum inest alicui, per accidens
dicitur, sicut cum dico: Hoc ambulante coruscat: non enim propter id quod ambulat,
coruscauit, set hoc dicitur secundum accidens. Si uero quod praedicatur insit subiecto
propter ipsum, per se inest, ut si dicamus quod interfectum interiit: manifestum est
quod propter id quod interfectum est, interiit, et non est accidens quod interfectum
intereat.
136 paul symington
cause between subject and predicate even though it is stated in such
a way that it is per se.
Although Aquinas is clear that the categories are established through
per se predication (this is evident when one considers the statements
being secundum se is divided into ten categories, of which nine are
accidental kinds and the division of being secundum se and secundum
accidens follows according to which something is predicated of another
either per se or per accidens),
56
he does not explicitly mention the three
specic modes of per se predication in either of his commentaries on
Metaphysics V, 9 or Physics III, 5. However each mode can be identi-
ed in the text. For example, secundo modo predication is understood as
having predicates that are in the subject because the predicates signify
the subject itself, and Aquinas describes such a relationship between
subject and predicate as that mode of predication from which quality
and quantity are derived.
57
Likewise, he refers to a way in which a
predicate is related externally to its subject, and this is how he charac-
terizes quarto modo per se predication.
Furthermore, the modes of per se predication are contrasted by
Aquinas with the modes of per accidens predication.
58
He identies
three modes of per accidens predication. The rst obtains when a term
signifying an accident is predicated of another term that also signies
an accident; for example, The just is musical. The second type of
per accidens predication obtains when a term signifying substance has a
term signifying an accident predicated of it; for example, The man is
musical. The third type of per accidens predication obtains when a term
signifying substance is predicated of a term that signies an accident;
for example, The musical is a man. These modes of predication do
not meet the criteria of per se predication.
These differences between per se and per accidens predication clarify
the procedure Aquinas follows to establish the number and identity of
the categories. How he does this is the subject of the next section (III).
However, here I offer an overview of Aquinass procedure. First, he
focuses on per se modes of predication; that is, he focuses on propositions
that are characterized by one of the three per se modes of predication.
56
In Met. 5.9, n. 885.
57
In Met. 5.9, n. 892: Secundo modo ut praedicatum sumatur secundum quod
inest subiecto: quod quidem praedicatum, vel inest ei per se et absolute, ut consequens
materiam, et sic est quantitas: vel ut consequens formam, et sic est qualitas.
58
In Met. 5.9, nn. 88688.
For example, he would group Socrates is a man, Bucephelus is a
horse and A human is a rational animal together because they have
the primo modo per se predicational form. He would group A surface is
colored and A human has the capacity to laugh, together because
they each have the secundo modo per se predicational form. Finally, he
would identify Having been slaughtered, it died, as having the quarto
modo per se predication. Separating propositions such as these from each
other according to the kind of per se modes of predication that they
have is Aquinass procedure for distinguishing the categories from each
other. Any two given per se propositions that have different predicational
structures signify different categories. No two per se propositions that
have different per se modes of predication signify the same category. In
short, Aquinas establishes substance through primo modo predication;
quantity, quality and relation through secundo modo predication; and the
remaining six categories through quarto modo predication.
Second, after he groups propositions according to the different per se
modes of predication that they have, he further subdivides propositions
in the same group according to differences in the way that predicates are
related to their subjects in the group. For example, a given proposition
that has secundo modo predication can signify quantity, quality or rela-
tion. Thus, Aquinas identies the distinctive way in which predicates
are related to subjects in propositions that signify quantity from the
way in which predicates are related to subjects in propositions that
signify quality. It is by way of division and subdivision of propositions
according to their predicational features that Aquinas establishes the
identity of the categories.
III. The Derivation of the Categories from per se predication
In nn. 889894 of the text, Aquinas presents a justication of the list of
categories and refers enigmatically to the modes of predication. Since
Aquinas presupposes so much about how the modes of predication
are used to establish the list of categories, in this section I present in
detail how Aquinas establishes specic categories from per se modes of
predication. Specically, I show how substance, quantity and quality
are derived and then indicate how the remaining categories could be
established.
Although Joseph Owens suggests that because of its ontological
priority substance is too striking to need defense in a metaphysical
138 paul symington
context, Aquinas seeks to establish it based on the per se modes of
predication.
59
A predicate can be per se related to its subject in a rst
way when the predicate expresses what the subject is, just as when
I say, Socrates is an animal, for Socrates is that which is an animal.
And this predicate is said to signify rst substance since it is a particular
substance, about which all things are predicated.
60
Aquinas is referring
to those propositions that are primo modo per se, in which the denition
[of the subject] or something posited in the denition [of the subject]
is predicated of the subject.
61
By reecting on propositions that have
this mode of predication one can see a further relationship between
the terms of the subjects and predicates of such propositions. One can
identify those propositions that have subjects that are not predicated
of another subject but other things are predicated of them.
62
This
yields per se propositions that signify primary substance. For example,
Socrates is an animal signies substance because it is a primo modo
proposition in which the subject cannot be primo modo predicated of
anything else.
However, it could be asked whether this mode of primo modo predi-
cation distinguishes those propositions that signify primary substance
from those that signify primary accidents.
63
For example, are Socrates
is an animal, and This patch of color is white (if this patch of color
really is white) both primo modo per se propositions in which the subject
cannot be primo modo per se predicated of anything else? The answer is
that although both propositions are primo modo per se propositions, there
is a further difference between them that distinguishes substances from
other categories. Indeed, both propositions have subjects that cannot
be per se predicated of another, for neither Socrates nor this patch of
color can be predicated of anything else. However, whereas Socrates
is an animal is such that the predicate animal cannot be secundo modo
predicated of any other subject, This patch of color is white does
have a predicate (white) that can be secundo modo per se predicated of
another subject; namely, a surface. That is, in secundo modo predication
59
Joseph Owens , An Elementary Christian Metaphysics (Milwaukee, 1963), p. 145.
60
In Met. 5.9, n. 891: Uno modo cum est id quod est subiectum, ut cum dico, Socrates
est animal. Nam Socrates est id quod est animal. Et hoc praedicatum dicitur signicare
substantiam primam, quae est substantia particularis, de qua omnia praedicantur.
61
Post. An. 1.10, lines 2530.
62
In Met. 5.10, n. 898.
63
This is discussed by Aristotle in Categories 3 (1a161b9) where he makes the four-
fold distinction between being predicated of and being in another.
the predicate is related to a subject in two ways. In one way, it is related
to the subject of the proposition as a determination of the subject. This
is in virtue of the fact that it is a predicate in a proposition; Aquinas
held that the predicate is compared to the subject as form is to mat-
ter.
64
In the second way, the subject is itself signied by the predicate
independently of the subject. For example, in This patch of color is
white (if this patch of color really is white) surface is signied by
the predicate white. Consequently, this proposition is distinguished
from per se propositions that signify substance because no proposition
with a predicate that can be secundo modo predicated of another signi-
es substance. Therefore, primary substance is sufciently identied
through primo modo predication insofar as per se propositions that signify
substance have predicates that are exclusive to primo modo predication.
From the derivation of primary substance, Aquinas moves to the
derivation of quantity, quality and relation. He indicates that a predicate
is related to its subject in a second way, when a predicate is understood
according to what is in [inest] a subject. For instance, a predicate is in
the subject per se and absolutely as following from its matter, and in this
way it is quantity. Or, a predicate is in the subject per se and absolutely
as something following from its form; in this way it is quality, or it is
not in it absolutely but in respect to another, and then it is relation.
65

Quantity, quality and relation are established through propositions that
signify secundo modo per se predication. Aquinas states in the Commentary
on the Posterior Analytics I, that the per of per se predication, because
it is a causal notion, designates a condition of material cause, just as
when it is said that a body is colored by way of its surface because
a surface is the proper subject of color.
66
Thus, the subject signied
64
Aquinas , Expositio Libri Periermenias, 1.8, n. 11.
65
In Met. 5.9, n. 892: Secundo modo ut praedicatum sumatur secundum quod
inest subiecto: quod quidem praedicatum, vel inest ei per se et absolute, ut consequens
materiam, et sic est quantitas: vel ut consequens formam, et sic est qualitas: vel inest
ei non absolute, sed in respectu ad alius, et sic est ad aliquid. To borrow a phrase
from Walton (supra, fn. 10, pg. 306) quantity and quality are related to substance as
matter and form respectively in an ontological sense, whereas both quantity and quality
are formal determinations of a substance, where substance is the matter, in a logical
sense. Although it seems that Aquinas believes that relation is established through
secundo modo predication, he does not show how it is established through this mode in
either In Met. 5.9 or 5.17 where he discusses the sense of relation.
66
Post. An. 1.10, lines 1417: quandoque autem habitudinem causae materialis,
sicut cum dicitur quod corpus est coloratum per superciem, quia scilicet proprium
subiectum coloris supercies est. See also lines 5167: Secundus modus dicendi per
se est quando hec prepositio per designat habitudinem cause materialis, prout scilicet
140 paul symington
by the subject of a secundo modo per se proposition is the material cause
(logically speaking) of the predicate.
Quantity and quality are derived from an examination of the relation-
ships between subjects and predications of secundo modo per se propositions
insofar as they exist absolutely in their subject. However, since both
of these categories are established from secundo modo per se propositions,
what further predicational understanding can be used to determine how
propositions that signify quantity can be distinguished from those that
signify quality? Aquinas offers a clue for dividing quantity from quality
in V, lect. 15 (where he specically discusses the senses of quantity).
He says that only in the genus quantity are some things signied as
subjects [of properties] and others as properties [themselves].
67
The
category quantity is signied by secundo modo per se propositions that
have predicates that not only are secundo modo predicated of its subject
but also the term of the predicate can itself be the subject of a differ-
ent property. For example, in A body is surfaced (i.e., A body has a
surface) the term surface of the predicate itself can be the subject of
the secundo modo per se predicate colored (i.e., The surface is colored.).
Thus, The body has a surface signies quantity. Another example of
this could be The body has mass. The term mass of the predicate
is secundo modo predicated of body and it also can be the subject in
the secundo modo proposition The mass is heavy. Therefore, according
to this mode of secundo modo predication The body has mass signies
quantity. In this way quantity is established through secundo modo per se
predication.
On the contrary, this criterion does not apply to terms that signify
quality. For example, in The surface is white, white signies a
property because it signies its proper subject surface in its deni-
tion. However, white itself cannot be the subject of a property. For
id cui aliquid attribuitur est propria materia et proprium subiectum ipsius. Oportet
autem quod proprium subiectum ponatur in difnitione accidentis, quandoque quidem
in obliquo, sicut cum accidens in abstracto difnitur, ut cum dicimus quod simitas est
curuitas nasi; quandoque uero in recto, ut cum accidens denitur in concreto, ut cum
dicimus quod simus est nasus curuus. Cuius quidem ratio est quia cum esse accidentis
dependeat a subiecto, oportet etiam quod difnitio eius signicans esse ipsius contineat
in se subiectum. Vnde secundus modus dicendi per se est quando subiectum ponitur
in difnitione predicati quod est proprium accidens eius.
67
In Met. 5.15, n. 983: Nam sola quantitas habet divisionem in partes proprias
post substantiam. Albedo enim non potest dividi, et per consequencs nec intelligitur
individuare nisi per subiectum. Et inde est, quod in solo quantitatis genere aliqua
signicantur ut subiecta, alia ut passiones.
example, in the secundo modo per se proposition The white is colored,
the proper subject in the denition of colored is not white but rather,
surface. Secundo modo per se propositions that signify quantity can be
distinguished from those that do not according to this predicational
criterion. For example, A human being is capable of laughter does
not signify quantity even though a human being is divisible into parts
and capable of laughter is a property of human beings. In this propo-
sition, capable of laughter cannot be the subject of a property and
a human being cannot be predicated secundo modo per se of anything
else. Therefore, propositions such as A human is capable of laughter,
and A surface is colored both signify the category quality because
both capable of laughter and colored are properties that cannot
be the subject of properties. In this way, quality is derived from the
mode of secundo modo per se predication insofar as it is made distinct
from substance and quantity.
Aquinas says that quantity is derived from per se propositions in which
the predicate is in the subject according to a material cause (which is
the second criteria of per se predication). This relates to the character-
istic of per se propositions that signify quantity in which the term of
the predicate can both be the subject of another property and itself be
a property in the following way: a quantity can be the material cause
of another property. Insofar as quantities ow from the matter of a
substance, it can be the material cause or subject of qualities whereas
qualities cannot be. For this reason, Aquinas states in his Commentary
on the Physics that qualities are founded upon quantity.
68
Aquinas holds
that quantity is similar to substance in that it shares the characteristic of
being able to be the subject of properties.
69
Qualities are said to follow
from the form of a substance because they are similar to substantial
forms insofar as they cannot be the subject of properties.
Aquinas says that a predicate is referred to a subject in a third
way when the predicate is taken from something extrinsic to the sub-
ject.
70
This is directly related to quarto modo per se predication in the
Commentary on the Posterior Analytics I, in which the per of per se may
even designate a condition of extrinsic cause.
71
The remaining six
categories are derived through this mode of predication.
68
In Phys. 3.5, n. 322.
69
Ibid.
70
In Met. 5.9, n. 892.
71
Post. An. 1.10, lines 1724: designat etiam habitudinem causae extrinsece.
142 paul symington
Quarto modo per se may involve a relation of efcient causation between
subject and predicate. The example that Aquinas gives of such a per
se proposition is Having been slaughtered, it died. He says that it
is clear that on account of that which has been slaughtered, it [the
animal] died, and it is not accidental that that which has been slaugh-
tered should die.
72
Although the proposition that Aquinas cites as an
example is rather awkward, he says that a relation of external cause is
signied in the relationship between the predicate and subject. There
are three major components in this proposition: the animal, having
been slaughtered and died. The animal is implicit in this proposition.
A relationship of efcient cause is understood between the predicate
and the subject: the slaughter was the efcient cause of the animals
death. At the same time, the predicate is universally predicated of the
subject: everything that has been slaughtered has died. From per se
propositions such as these, Aquinas supposedly understands that the
remaining categories can be derived. Unfortunately, Aquinas leaves us
in the dark about how specic modes of quarto modo per se predication
signify and differentiate propositions that signify one or another of the
remaining six categories. However, it would seem that the additional
categories are derived from quarto modo per se propositions in which the
external causes signied in the per se relationship between the subject
and predicate are understood to be in common to other quarto modo
per se propositions. For example, there may be quarto modo propositions
in which the external relationship between the subject and predicate
is understood in such a way that the predicate measures the subject
in some way, and then one could derive the categories of time, place
and position.
73
Now that I have presented Aquinas s view, an interesting observation
can be made regarding the relationship among the categories. An inter-
esting result of Aquinass view concerns the question of how accidents
72
Post. An. 1.10, lines 12235: Deinde cum dicit: item alio modo etc., ponit quartum
modum, secundum quod hec prepositio per designat habitudinem cause efcientis uel
cuiuscunque alterius <extrinsece>. Et ideo dicit quod quicquid inest unicuique propter
se ipsu m, per se dicitur de eo, quod uero non propter ipsum inest alicui, per accidens
dicitur, sicut cum dico: Hoc ambulante coruscat: non enim propter id quod ambulat,
coruscauit, set hoc dicitur secundum accidens. Si uero quod praedicatur insit subiecto
propter ipsum, per se inest, ut si dicamus quod interfectum interiit: manifestum est
quod propter id quod interfectum est, interiit, et non est accidens quod interfectum
intereat.
73
In Met. 5.9, n. 892.
can be understood to be both per accidens according to its metaphysical
sense and, at the same time, are independent from each other as the
most basic kinds of being. Accidents, although they are not per se beings
(this is the status of substance only) they are secundum se in the sense
that each category corresponds to a unique per se mode of predica-
tion. However, a discernable dependent relationship can be observed
between per se propositions that signify accidents and those that signify
substance. On one hand, color is not formally reducible to the form
or essence of rational animal because color is not predicated primo
modo per se of Socrates . On the other hand, colored is secundo modo
per se predicated of surface. Surface is secundo modo per se predicated
of body. Body is primo modo per se predicated of Socrates. Therefore,
in one way, color is dependent on substance because it is the ultimate
subject of predication. In another sense, color is a secundum se being. It
has quality primo modo per se predicated of it and in this sense quality
is not in any way reducible or dependent on any substance or subject
insofar as it is not primo modo per se predicated of any substance.
74
This
interesting characteristic of accidents is mentioned by Aquinas at the
end of his discussion. He says that on one hand, the word white
signies a subject in as much as it signies whiteness as an accident.
Thus, it is necessary that as a consequence it includes in it the notion
of a subject. For the being of an accident is to be in. . . . [On the other
hand] although whiteness signies an accident, it does not do so only as
an accident but according to the mode of a substance.
75
That which
signies according to the mode of substance is signied according to
primo modo per se predication, which is true of Whiteness is a color.
IV. Conclusion
In order to understand Aquinas s justication for the list of categories,
I have provided an interpretation of the text that tries to make most
sense of it according to its textual context rather than focusing on only
74
In Language and Logic, E. J. Ashworth points out that a characteristic of con-
crete accident terms (e.g., white) is that they have a double relation, on one hand to
substantial things, for only substances can be literate or white, and on the other hand
to the qualities of literacy or whiteness. The Cambridge Companion to Medieval Philosophy
(New York, 2003), pp. 7396, p. 86.
75
In Met. 5.9, n. 894.
144 paul symington
an aspect of the text. For example, I present an account that tries to
understand Aquinass words at the beginning of the text that states
both that all ten categories are secundum se and that secundum se is not
understood to mean substance (being per se) but rather per se predica-
tion. This interpretation has been illuminated by an examination of the
varieties of per se predication in Commentary on the Posterior Analytics I,
9 & 10. However, given the elliptical treatment of the topic by Aquinas,
my task of trying to establish a more unied understanding of it within
its textual context has come at the cost of engaging in a certain amount
of reconstruction of Aquinass thought on the matter.
76
76
Special thanks to Jorge J. E. Gracia , Lloyd Newton , Giorgio Pini , and Jeff Brower
for their careful and insightful comments and suggestions throughout various stages
of this article.
READING ARISTOTLES CATEGORIES
AS AN INTRODUCTION TO LOGIC:
LATER MEDIEVAL DISCUSSIONS ABOUT ITS
PLACE IN THE ARISTOTELIAN CORPUS
Giorgio Pini
Some trends in the history of philosophy are so well established that,
even when it is nally acknowledged that they originated from a one-
sided and somehow fanciful reconstruction of the events, it may still
be difcult to appreciate all the consequences of this discovery and
rearrange our views accordingly. The place and role of the Categories in
the Aristotelian corpus is a particularly good example. We are so used
to nding the Categories as the rst work in any collection of Aristotles
works that we may take this fact for granted and not pay particular
attention to it. As a consequence, we may still be secretly inuenced
by the very specic interpretation of Aristotles philosophy that this
apparently innocent ordering conveys. In that case, we would be just
the last in a very long series of readers assuming the place of the
Categories as uncontroversial. For several centuries, generations of fresh-
men started their philosophical training by reading this short treatise
together with its companion work, Porphyry s Introduction to the Categories.
The position of the Categories as the rst work in Aristotles collection
of logical writings, the so-called Organon, which was in turn posited at
the beginning of the whole Aristotelian corpus, greatly contributed to
its enormous success. As it happens, this short treatise (only 15 pages
in Bekker s edition) was the only philosophical treatise that has been
uninterruptedly read, studied and commented on since Antiquity. No
other single work had a comparable inuence in Western philosophy.
1

It is a well-known fact in ancient philosophy, however, that the very
notion of the Organon owes nothing to Aristotles intentions. By now,
1
M. Frede , The Title, Unity, and Authenticity of the Aristotelian Categories, in Id.,
Essays in Ancient Philosophy (Oxford, 1987), pp. 1128: 11; J. Brunschwig , Aristote de
Stagire. LOrganon: tradition grecque, in Dictionnaire des Philosophes Antiques, vol. 1 (Paris,
1989), pp. 482502: 492: Places par la tradition en tte du corpus aristotlicien, les
Catgories ont t parmi les textes les plus lus, recopis, tudis, traduits et comments
de toute lhistoire de la philosophie.
146 giorgio pini
scholars have shed some light on the philosophical signicance of the
order in which Aristotles writings have come down to us. It is now
clear that the prominent position the Categories was made to occupy in
the Aristotelian corpus is the result of a systematic interpretation of
Aristotles writings which is likely to have taken shape in the Neoplatonic
circles between the third and fth centuries A.D. and whose remote
origins may possibly be traced to the Hellenistic period, when some of
Aristotles interpreters came to read their masters writings in the light
of Stoic doctrines. As a result of this interpretation, the Categories was
seen as a work of logic dealing with the simple terms out of which
propositions are made. Again, we are so used to such a conception
that we may forget how odd this interpretation may have sounded to
Aristotles very rst readers and to Aristotle himself. For sure, it should
sound quite odd now to us, in the light of contemporary scholarship.
For considering the Categories as a work of logic is an awkward view,
one that surely requires some justication and that, on the face of it,
is very likely to be wrong. All the same, that was the way in which the
Categories was interpreted up to some decades ago.
What is the philosophical interest of all this? Because of some his-
torical accident, Aristotles Categories was considered as a work of logic
and was posited at the beginning of Aristotles corpus. This did not
reect Aristotles intention but the willingness of his editors to present
Aristotles thought according to a plan that was originally extrane-
ous to him. But one may ask: does this change anything as far as the
philosophical interpretation of the Categories is concerned? One may
suspect that the way the Categories was presented in ancient, medieval
and modern editions, interesting as it may be, is nothing more than
an historical accident of no concern for the evaluation of the argu-
ments contained in either the Categories or the immense philosophical
production that developed around and out of Aristotles treatise. In this
paper, it is my intention to show that, even from a purely philosophi-
cal point of view, it is important to evaluate the place assigned to the
Categories in the Aristotelian corpus. For one thing, the assessment of
the arguments present in the Categories depends on the role that it is
supposed to play in the Aristotelian system. If the Categories is considered
as a work of logicand as such it is given the opening place in the
Aristotelian corpus, then the statements and arguments it contains
must be assessed in the light of what logic is taken to be. This means
that what generations of interpreters during the Middle Ages read
in the Categories was not intended to be (and consequently should not
reading the CATEGORIES as an introduction to logic 147
be taken to be) statements and arguments about the world but about
concepts. Second and consequently, what medieval interpreters read
in the Categories depended on their own conception of logic. So, in
order to assess correctly what a medieval author said on the Categories,
we must take into account his own idea of what logic is. According
to which conception of logic was adopted, the Categories could be seen
as a treatment either of arguments and their premises or of the basic
concepts by which we represent and classify the extramental world.
Since Aristotles Categories undeniably exerted a fundamental inuence
on the thought of virtually any medieval philosopher and theologian,
no matter whether he wrote a commentary on that work or not, it is
evident how vital this preliminary assessment is in order to give a fair
reconstruction of the opinions of many medieval thinkers.
I proceed as follows. First, I briey recall how it came about that
the Categories was considered as the rst treatise of Aristotles Organon.
Second, I turn to the main focus of this paper, i.e., the thirteenth
century, in order to consider the inuence that the massive translation
of many of Aristotles writings hitherto lost had on the interpretation
of his Categories. Third, I consider the impact of the notion of logic as
a study of concepts (and specically of a particular class of concepts,
the so-called second intentions) on the interpretation of the Categories in
the second half of the thirteenth century in authors such as Thomas
Aquinas , Giles of Rome, Radulphus Brito , and John Duns Scotus .
Fourth, I turn to the contrasting interpretations of the Categories in Peter
of John Olivi , William Ockham , Walter Burley and the Oxford realist,
Robert Alyngton . Fifth, I focus on the interpretations that Thomas
Aquinas, Duns Scotus, William Ockham, Walter Burley and Robert
Alyngton gave of the distinction between primary and secondary sub-
stances (Cat. 5, 2a1119). By the analysis of this specic case, I hope
to clarify the relevance of the view of the Categories as the rst work of
logic for the assessment of the meaning and value of the arguments
that medieval authors read into that work.
1. Aristotles Categories and the Origins of the Organon
Scholars have carefully scrutinized the authenticity, unity, title and
nature of the Categories. If no one now seriously challenges Aristotles
authorship, it is usually recognized that this short treatise consists of
two or three different sections probably assembled together by the rst
148 giorgio pini
editors, that the title Categories only refers to one of its sections and
is not the oldest title under which the work was known, and that the
treatise is to be seen as closely connected with the Topics.
2
This last
conclusion is particularly illuminating as far as the content of the trea-
tise is concerned. In this short work, Aristotle did not mean to present
the foundations of his ontology. Even less did he intend to provide a
treatment of simple terms as a preliminary to the study of propositions
and syllogism. The whole notion of logic as a discipline concerned with
deductions (syllogism) and the elements out of which deductions are
made (i.e., propositions and terms) seems to be extraneous to Aristotle.
The very idea of a unitary Organon was imposed on Aristotle by suc-
cessive scholarship. Aristotle did not even have a single word for what
we call logic. What constituted his logical legacy actually consists of
separate studies of deduction (Prior and Posterior Analytics), of the rules to
follow in a dialectical discussion (Topics and its foreword, the Categories),
of fallacies (Sophistical Refutations) and of notions such as noun, verb and
statement (De interpretatione).
3
So when did the notion of a unitary Organon including all the
writings from the Categories to the Sophistical Refutations rst appear?
Probably between the third and the fth century A.D. as a result of the
systematization of the study of Aristotelian philosophy as part of the
standard philosophical curriculum in Neoplatonic schools.
4
Recently, it
has been interestingly argued that the core of a unitary Organon should
be traced back to the Hellenistic period: it was probably the inven-
tion of Aristotelian scholars willing to structure their masters writings
into a system comparable to the one that the Stoics had devised after
Aristotle. In order to achieve such a result, it was necessary to sort out
Aristotles writings according to the three parts of the Stoic system, logic,
physics, and ethics. It is a consequence of this Stoic inuence that the
2
Frede , The Title; R. Bods , Introduction, in Aristote, Catgories. Texte tabli
et traduit par R. Bods (Paris, 2001), pp. XXIVXLI, LXIVLXXX, XCCX.
3
Brunschwig , Aristote de Stagire, pp. 486488; M. Burnyeat , A Map of Metaphysics
Zeta (Pittsburgh, 2001), pp. 8789. Burnyeat actually argues that the Organon does
constitute a unity and that it can be described as logical, but in a completely different
sense from that of the tradition, which he explicitly criticizes.
4
F. Solmsen , Boethius and the History of the Organon, American Journal of Philology
65 (1944), 6974; Ilsetraut Hadot , Deuxime point du premier schma introductif:
la division noplatonicienne des crits dAristote, in Simplicius , Commentaire sur les
Catgories. Traduction commente sous la direction de Ilsetraut Hadot, fascicule I
(Leiden, 1990), pp. 6393.
different treatises that now constitute the Organon were constructed as
the elements of a single study, i.e., logic. Furthermore, it is because the
Stoics conceived of logic as the study of deductions and their elements
(propositions) that Aristotles own logic was constructed around the
notion of syllogism and its elements (propositions and the terms out of
which propositions are made, i.e., subjects and predicates). In this process
a major role was probably played by Andronicus of Rhodes, who in the
rst century B.C. provided the most famous edition of Aristotles corpus
in ancient times. Because, in order to ll out the Stoic scheme of logic,
there was a need for a treatise devoted to the study of propositions as
parts of syllogism and of another treatise devoted to the study of terms
as parts of propositions, the De interpretatione and the Categories came to
play such a role, even though Aristotle did not apparently design them
for that function.
5
As it happens, the Categories is particularly ill-chosen
as an introduction to the study of syllogism. Admittedly, its central part
is devoted to uncombined expressions classied into ten kinds accord-
ing to their meaning. But such a classication is completely irrelevant
for the study of syllogism, whose terms play their role in deductions
independently of their meaning, as is evident from the fact that Aristotle
views them as symbolic variables expressed by letters.
6
All the same, the place of the Categories in the Aristotelian system
went undisputed from Andronicus of Rhodes onwards. The sequence
Categories (study of simple terms)De interpretatione (study of proposi-
tions made of simple terms)Prior Analytics (study of syllogisms made
of propositions) became standard. No matter how articial this triad
was and how extraneous to the others each one of its element was
originally Andronicuss edition provided the denitive and standard
ordering of Aristotles writingsan ordering that conveyed an inter-
pretation of Aristotles corpus as a system whose rst part was a logic
structured around the notion of deduction.
The elements of this picture are already present in Plotinus s pupil,
Porphyry . By the fth and sixth century, the Categories was commonly
regarded as the rst work to be studied in the Neoplatonic philosophical
5
R. Bods , Linuence historique du stocisme sur linterprtation de luvre
philosophique dAristote, Revue des Sciences philosophiques et thologiques 79 (1995), 553586;
Id., Introduction, in Aristote, Catgories, pp. XIXLI. See already Frede , The Title,
pp. 2021, for the origins of the Organon in the rst century B.C. and the role of the
Categories in it.
6
Bods , Linuence, p. 585.
150 giorgio pini
curriculum.
7
The Neoplatonic legacy was received in the sixth cen-
tury by Boethius , who handed down to the Latin West the sequence
CategoriesDe interpretationeAnalytics together with the implicit system-
atic interpretation of Aristotles philosophy that had generated it:
Since Aristotle wanted to hand down to his successors the structure and
the knowledge of both dialectical and demonstrative science, he realized
that both the dialectical and the demonstrative argumentation was orga-
nized according to the same structure, i.e., syllogism. Thus, he wrote the
Prior Analytics, which the Greek call so that students should
read them before they got any knowledge of either the dialectic or sci-
entic discipline [. . .]. But since it is necessary that syllogism be made
out of propositions, he wrote the book whose title is , on
propositions. But all propositions are made out of expressions signifying
something. Consequently the book that he wrote on the ten categories,
which in Greek are called , is about the rst names of things
and their meanings.
8
Thanks to Boethius , Latin medieval authors became familiar with
the ancient opinion concerning the place and role of the Categories in
Aristotles Organon. So it happened that the role of the Categories as an
introduction to syllogism was regarded as an undisputed truth even in
an age when the largest part of Aristotles corpus (including the last
and most important part of the logical triad, the Prior Analytics) was
unknown to the Latin West.
9
Admittedly, it was open to debate whether
7
Porphyry , In Cat., ed. A. Busse, CAG IV,1 (Berlin, 1887), p. 56,2030. See the
English translation in Porphyry, On Aristotle Categories. Translated by S. K. Strange
(London, 1992), pp. 3132. See Solmsen , Boethius , pp. 71, 73; Hadot, Deuxime
point, pp. 8084.
8
Boethius , In Isagogen Porphyrii Commenta, eds. G. Scheppss S. Brandt (Vindobonae-
Lipsiae, 1906), Editio prima, pp. 1214: Aristoteles enim quoniam dialecticae atquae apo-
dicticae disciplinae uolebat posteris ordinem scientiamque contradere, uidit apodicticam
dialecticamque uim uno syllogismi ordine contineri. Scribit itaque priores Resolutorios,
quos Graeci uocant, qui legendi essent antequam aliquid dialecticae uel
apodicticae artis attingerent [. . .]. Sed quoniam syllogismum ex propositionibus con-
stare necesse est, librum qui inscribitur, de propositionibus adnotauit.
Omnes uero propositiones ex sermonibus aliquid signicantibus componuntur. Itaque
liber quem de decem praedicamentis scripsit, quae apud graecos dicuntur,
de primis rerum nominibus signicationibusque est. See also In Categorias Aristotelis
libri quatuor, ed. J.-P. Migne, PL 64 (Paris, 1847; reprint Turnhout, 1979), col. 161 B:
Hinc est quod ad logicam tendentibus primus hic liber legendus occurrit, idcirco quod
cum omnis logica syllogismorum ratione sit constituta, syllogismi vero propositionibus
jungantur, propositiones vero semonibus constent, prima est utilitas quid quisque sermo
signicet, propriae scientiae difnitione cognoscere.
9
Boethius also translated the Prior Analytics, Topics and Sophistical Refutations, but these
translations had virtually no circulation. See L. Minio-Paluello , Note sullAristotele
the Categories dealt primarily with words or things. All the same, no one
seems to have doubted that this writing was a treatment of simple ele-
ments (whether terms or things) to be read before the De interpretatione,
which dealt with propositions. Consequently, no one seems to have seri-
ously doubted that the Categories belonged to logic.
10
Interestingly, this
agreement is a peculiarly Western story. Among the Arabs, Avicenna
clearly realized that there was no good reason to assign the Categories
to logic and consequently he maintained that it was not to be seen as
part of the Organon.
11
2. The Thirteenth Century: The Impact of the New Aristotle
By the rst decades of the thirteenth century, most of Aristotles works
that had been unknown for centuries in the West started circulating
again in new translations.
12
What inuence did this new wave of texts
have on the interpretation of the Categories, which together with the De
interpretatione had long been the only evidence of Aristotles teachings?
Interpreters of the Categories were now confronted with two sets of newly
translated works. First, readers of the Categories could now compare it
to the Prior and Posterior Analytics, whose translation made available to
the West the last and most important part of the logical triad of which
the Categories was meant to be the rst element. Accordingly, the logical
character of the Categories as a study of simple terms ultimately aimed
at syllogism was emphasized. It may have been a consequence of the
integration of the Categories in the newly complete Organon that the
contrast between realistic and nominalistic interpretations of the treatise
rapidly declined.
13
In the thirteenth century, the logical character of the
latino medievale VI. Boezio, Giacomo Veneto, Guglielmo di Moerbeke, Jacques Lefvre
dEtaples e gli Elenchi Sostici, Rivista di Filosoa Neo-Scolastica 44 (1952), 398411,
repr. in Id., Opuscula. The Latin Aristotle (Amsterdam, 1972), pp. 164177; Id., Les
traductions et les commentaires aristotliciens de Boce, Studia Patristica 64 (1957),
567584, repr. in Id., Opuscula, pp. 328335.
10
On early medieval commentaries on the Categories see J. Marenbon , Vocalism,
Nominalism and the Commentaries on the Categories from the Earlier Twelfth Century,
Vivarium 30 (1992), 5161.
11
I. Madkour , LOrganon dAristote dans le monde arabe, 2
e
d. (Paris, 1969), pp. 7983;
Bods , Linuence, p. 585.
12
See F. Van Steenberghen , Aristotle in the West. The Origins of Latin Aristotelianism,
translated by L. Johnston (Louvain, 1955).
13
For one of the last commentaries on the Categories in the twelfth-century tradition,
152 giorgio pini
Categories as a work dealing not with things but with logical objects (no
matter what they were taken to be) was unanimously stressed. Second,
readers of the Categories could now become familiar with the Physics and
Metaphysics, whose translation made available again a treatment of cat-
egories that was in many ways different from the one with which Latin
authors had long been familiar. The attempt to reconcile the old and
the newly discovered treatment of categories resulted in the doctrine
according to which categories can be studied in two ways, in logic and
in metaphysics. In the Categories, categories are studied not as kinds of
extramental things but as logical objects. By contrast, in the Physics and
Metaphysics, categories are considered as kinds of things. This position
solved any possible contrast between the treatments of the Categories,
on the one hand, and of the Physics and Metaphysics, on the other hand.
It was usually maintained that, in the two sets of works, Aristotle does
not contradict himself, because he does speak of the same things (i.e.,
categories) but from two different points of view and insofar as categories
play two distinct roles, as logical objects constituting the simple terms
out of which propositions and syllogisms are made or as real objects
out of which the extramental world is constituted.
14
Thus, the translation of both the logica nova and the physical and
metaphysical works resulted in a renewed emphasis on the logical
character of the Categories. Robert Kilwardby , Robertus Anglicus ,
the anonymous author of the students guide contained in the ms.
Barcelona, Arxiu de la Corona dArag, Ripoll 109all writing before
or around the middle of the thirteenth centuryare among those who
consider the Categories as a logical treatise dealing with simple terms
insofar as they are remote parts of syllogisms, whose proximate parts
are the propositions studied in the De interpretatione, which in turn are
constituted of simple terms. So all these authors consider the Categories
as the rst part of logic, which in its integrity is the study of syllogistic
deduction.
15
Their conception of logic is a noteworthy formal one (i.e.,
see Sten Ebbesen , Anonymous DOrvillensis Commentary on Aristotles Categories,
Cahiers de linstitut du moyen-age grec et latin 70 (1999), 229423.
14
Giorgio Pini , Categories and Logic in Duns Scotus . An Interpretation of Aristotles Categories
in the Late Thirteenth Century (Leiden, 2002), pp. 1927.
15
Robert Kilwardby , Notulae super Praedicamenta, ed. P. O. Lewry in Robert Kilwardbys
Writings on the Logica Vetus, Studied with Regard to their Teaching and Method (D.Phil. Thesis,
University of Oxford, 1978), p. 370; Robertus Anglicus, Scriptum super libro Porphyrii, ed.
D. Pich in Le problme des universaux la facult des arts de Paris entre 1230 et 1260 (Paris
logic as the study of deductions) and their idea is that the Categories
studies the simple elements out of which propositions are constituted,
i.e., subjects and predicates. Accordingly, all that is said in the Categories
must be seen as referring to terms insofar as they act as subjects and
predicates in a proposition. It is neither the structure of the world nor
the structure of the concepts by which our intellect knows the world
that is the focus of Aristotles treatise; it is the elements out of which
formal arguments are constructed.
Around 1235, however, the Paris arts master, Jean le Page ( Johannes
Pagus ), linked the logical interpretation of the sequence CategoriesDe
interpretationePrior Analytics to the ontological structure of the world in
a remarkably comprehensive picture. The logical sequence of terms,
propositions and arguments studied in the three rst works of the
Organon is grounded on a corresponding sequence of real unities: the
unity of simple things such as angels, the unity of compound things such
as material substances and the unity of causality which gives coherence
to the whole universe. The standard theme of the place of the Categories
in the Organon and in the Aristotelian corpus is accordingly integrated
in a powerful and all-encompassing view of the universe:
It is written in the second book of the Physics that Art imitates nature
as far as it can. Therefore, corresponding to the threefold unity that is
found in things, there is a threefold unity in form. For there is a unity
by indivision, such as in the angels: another unity is that by the union of
matter with form, such as in compound things such as the things made
out of elements; the third unity is the one that occurs because of order
and causality, and it is in this way that the parts of the universe have
unity, for they are ordered [to one another] and the one causes the other.
To this threefold unity there corresponds a threefold unity in speech.
To the rst unity that occurs in simple things there corresponds the
term, for the term has unity by indivision, since the logician proceeds
in his divisions no further than terms. Moreover, there is a second unity
in speech that corresponds to the second unity in things, and this unity
occurs in propositions, for a proposition is composed of a name and a
verb as of matter and form, and this unity corresponds to the unity that
2005), pp. 270272; C. Laeur , Le Guide de ltudiant dun matre anonyme de la Facult
des arts de Paris au XIII
e
sicle (Universit Laval, Qubec, 1992), par. 503514. See E. J.
Ashworth , Lanalogie de ltre et les homonymes: Catgories, 1 dans le Guide de
ltudiant , in Lenseignement de la philosophie au XIII
e
sicle. Autour du Guide de ltudiant
du ms. Ripoll 109, d. par C. Laeur J. Carrier (Turnhout, 1997), pp. 281295:
286287.
154 giorgio pini
occurs in things that are compounded of matter and form, such as the
things that are compounded of elements. And there is a third unity in
speech, which occurs according to order and causality, and this unity
occurs in syllogism, for between premises and conclusions there is order
and causality, and this unity corresponds to the unity of the universe.
Thus, from what has been said, the division of rational philosophy,
which is called logic, is clear. For with regard to the rst unity, we
have the Categories, in which the uncombined is dealt with, which has
unity by indivisibility. With regard to the second unity we have the Peri
hermeneias, in which the proposition is dealt with, which has unity by the
composition of form with matter. As far as the third unity is concerned,
there are the other books where syllogism is dealt with, whose unity is
by causality and order.
16
3. Categories and Concepts
By the middle of the thirteenth century, a new conception of logic
became more and more widespread among Latin authors. Largely as
a result of the inuence of al-Frb and Avicenna , logic came to be
considered as the study of a particular class of concepts, the so-called
second intentions. It is not necessary to dwell on the details of this
16
Jean le Page ( Johannes Pagus ), Rationes super Predicamenta Aristotelis, ed. E. France-
schini , in E. Franceschini, Giovanni Pago: le sue Rationes super Predicamenta
Aristotelis e la loro posizione nel movimento aristotelico del secolo XIII, Sophia 2
(1934), 172182, 329350, 476486: 175176: Ars imitatur naturam in quantum potest,
scribitur in secundo phisicorum; secundum igitur triplicem unitatem in rebus repertam
triplex est unitas in forma: est enim unitas per indivisionem ut in angelis, alia est unitas
ex unione materie cum forma ut est in rebus compositis ut in elemen[ta]tis, tertia est
que attenditur penes ordinem et causalitatem et sic partes mundi habent unitatem;
ordinantur enim, et una causat aliam. Huic triplici unioni correspondet triplex unitas
in sermone; unitati prime que est in rebus simplicibus correspondet terminus; terminus
enim habet unitatem per indivisionem, cum loicus ultra terminum non dividat. Item
est secunda unitas in sermone que respondet secunde unitati in rebus, et hec unitas
est in enunciatione; enunciatio enim componitur ex nomine et verbo sicut ex materia
et forma, et hec unitas correspondet unitati que est in compositis ex materia et forma
sicut in elementatis. Et est tertia unitas in sermone que attenditur penes ordinem et
causalitatem, et hec unitas est in sillogismo: est enim causalitas et ordo premissarum
ad conclusionem; et hec unitas respondet unitati universi. Ex predictis ergo patet
divisio rationalis philosophie que loica appellatur; nam quantum ad primam unitatem
accipitur liber predicamentorum ubi determinatur de incomplexo quod habet unitatem per
indivisibilitatem; quantum ad secundam accipitur liber peryermenias in quo determinatur
de enunciatione que habet unitatem per compositionem forme cum materia; secundum
tertiam unitatem acci[piuntur] alii libri in quibus de sillogismo determinatur in quo
est unitas per causalitatem et ordinem.
new conception of logical objects.
17
What is relevant for the study
of Aristotles Categories is that logic was seen not only as the study of
deductions and arguments but also and especially as the study of cer-
tain conceptual entities, called second intentions, which were taken to
represent the relationships holding among concepts representing things
in the world. Examples of second intentions are species (which is a
concept representing the relationship holding between an individual
such as Socrates and the concept of the kind to which it belongs, such
as man), proposition (which is the concept representing the relation-
ship holding between a concept acting as subject and a concept act-
ing as predicate) and syllogism (which is the concept representing the
relationship holding between the two premises and the conclusion of
a syllogistic argument).
The place and role of the Categories in the Organon was accordingly
reinterpreted in the light of this new doctrine of logic. Several authors
of the second half of the thirteenth century (including Thomas Aquinas ,
Peter of Auvergne , John Duns Scotus , Thomas Sutton , and Radulphus
Brito ) still regarded the Categories as aimed at the study of syllogism,
because in that treatise Aristotle deals with simple terms constituting
propositions, which in turn constitute syllogisms.
18
But these authors
conceived of terms, propositions and syllogisms not as linguistic but as
conceptual entities. So, the Categories was seen as focusing on the simple
concepts by which our intellect represents the extramental world and out
of which mental propositions are made. As Thomas Sutton observes,
these simple concepts are not second intentions themselves, because
they represent things in the extramental world, not relationship holding
between concepts. All the same, the Categories is a work of logic and not
of metaphysics because it studies simple conceptual terms not insofar
as they represent the world but insofar as they act as the foundations
17
See Pini , Categories and Logic, pp. 45137.
18
Thomas Aquinas , Expositio libri Posteriorum, in S. Thomae de Aquino Opera omnia,
I* 2, ed. Commissio Leonina (Rome, 1989), p. 4; Peter of Auvergne , Quaestiones super
Porphyrium, ed. A. Tin, in Le questioni su Porrio di Pierre dAuvergne, Archives
dhistoire doctrinale et littraire du Moyen Age 64 (1997), 253333, q. 3, pp. 272275; John
Duns Scotus , Quaestiones in librum Porphyrii Isagoge, eds. R. Andrews et al., in Ioannis Duns
Scoti Opera philosophica I (St. Bonaventure, 1999), q. 3, n. 20, pp. 1617; Radulphus
Brito , Quaestiones super librum Porphyrii, in Quaestiones subtilissimae magistri Rodulphi Britonis
super arte veteri (Venetiis, [1499]), q. 3, f. 6vab; Thomas Sutton , In Cat, ed. A. Conti
in Thomas Sutton s Commentary on the Categories according to ms. Oxford, Merton
College 289, in The Rise of British Logic: Acts of the Sixth European Symposium on Medieval
Logic and Semantics, ed. P. O. Lewry (Toronto, 1985), pp. 173213: 191, 193.
156 giorgio pini
for second intentions. So, for example, the simple concept of substance
is studied in logic not insofar as it represents substances in the world
but insofar as the logical concept of species is founded on it, as species
represents the relationship between two substances of different degree
of universality. In this way, Sutton recognizes some difference between
the Categories on the one hand and the other treatises of the Organon on
the other hand. Unlike the other logical treatises, the Categories does not
deal with second intentions but with the foundations of second inten-
tions, and so with concepts that, by themselves, represent extramental
things and not a relationship between concepts. All the same, Sutton
maintains that the Categories belongs to logic as a sort of preliminary
treatment necessary for the study of second intentions that, properly
speaking, constitutes logic.
19
Thus, according to this conception, the Categories is a sort of preface
to logic: it is the study of the simple concepts by which our intellect
represents the world and out of which mental propositions are made. As
I have said, late thirteenth-century authors maintain that logic itself is
not so much the study of valid deductions (as it was for authors writing
in the rst half of the century) as the study of conceptual relationships
holding among conceptual entities. Accordingly, their conception of
logic is closer to what we would consider cognitive psychology than to
what we call formal logic.
Typically, however, the new conception of logic did not replace com-
pletely the old view according to which logic is the study of deductions.
For example, both Duns Scotus and Radulphus Brito still consider syl-
logism as the subject matter of logic, because syllogism is what gives
logic its unity. Moreoever, both Brito and Scotus still maintain that the
Categories is a part of logic because it is aimed at syllogism, since it deals
with its ultimate constituents, i.e., terms. But they also consider both
syllogisms and their constituents as conceptual as opposed to linguistic
entities.
20
Consequently, in their approach to logic they emphasize not
the necessary nature of the link between premises and conclusions
in a valid argument but the conceptual nature of arguments and of
their parts.
19
Thomas Sutton , In Cat., ed. Conti , p. 191. See Pini , Categories and Logic, pp.
3840.
20
See above, n. 15.
Because of the new conception of logic as a study of concepts,
several authors propose a reconsideration of the old triad accord-
ing to which the Organon is organized as the study of simple terms
(Categories), propositions (De interpretatione) and syllogisms (Analytics). Not
only is this triad interpreted as concerning mental terms or concepts,
as we have seen; it is also reduced to a more fundamental threefold
division, which it is taken to reect. From some elements present in
Aristotles De Anima,
21
thirteenth-century authors developed a theory
of three operations of the intellect: simple apprehension, composition,
and argumentation. These three operations were regarded as the basic
ways in which our intellect works: by forming and apprehending simple
concepts, by combining simple concepts to form mental propositions,
by arranging mental propositions in deductive chains. Consequently,
the sequence CategoriesDe interpretationePrior Analytics was seen as
mirroring the three basic operations of the intellect. Thomas Aquinas
provides a clear statement of how the old triad was reinterpreted in a
psychological way:
As the Philosopher says in the third book of On the Soul, there are two
operations of the intellect: one that is called comprehension of the indi-
visibles, by which namely, the intellect apprehends the essence of each
thing in itself; the other is the operation of the intellect that combines and
divides. To these two operations, the third operation of arguing is added,
according to which reason proceeds from what is known to the inquiry
of what is unknown [. . .]. Now, since logic is called rational science, it
is necessary that it consider what pertains to the three aforementioned
operations of reason. Thus, in the Categories Aristotle deals with what
pertains to the rst operation of the intellect, namely, what is understood
by simple apprehension. By contrast, in the Peri hermeneias he deals with
what pertains to the second operation of the intellect, namely, afrma-
tive and negative statements. Finally, he deals with what pertains to the
third operation of the intellect in the Prior Analytics and in the following
treatises, which deal with syllogism in general and with the various species
of syllogisms and arguments by which reason proceeds from one thing to
another. Therefore, according to the aforementioned order of the three
operations of the intellect, the Categories is aimed at the Peri hermeneias,
which is aimed at the Prior Analytics and at the following works.
22
21
Aristotle, De Anima III, 6, 430a2628.
22
Thomas Aquinas , Expositio libri Peryermenias, in S. Thomae de Aquino Opera omnia
I*, 1, ed. Commissio Leonina (Rome, 1989), L. I, p. 5: Sicut Philosophus dicit in III
De Anima, duplex est operatio intellectus: una quidem que dicitur indiuisibilium intel-
ligencia, per quam scilicet intellectus apprehendit essenciam uniuscuiusque rei in se
ipsa; alia est autem operatio intellectus componentis et diuidentis; additur autem et
158 giorgio pini
Basically the same scheme is present in Giles of Rome and Duns
Scotus , both of whom are clearly dependent on Thomas Aquinas in
this respect.
23
This shows how an old conception of the Organon based
on purely formal criteria (the distinction between terms, propositions
and syllogisms) was reinterpreted in a psychological way when some
previously unknown works (such as Aristotles De Anima) were inte-
grated in the system of sciences and when a new conception of logic
appeared.
By the end of the thirteenth century, then, the Categories came to be
interpreted as the study of the simple concepts by which our intellect
knows the world. This meant that most of what Aristotle says in the
Categories was interpreted as referring neither to things in the world nor
to linguistic terms but to the way in which we conceive of the world.
Such an interpretation is particularly evident in Duns Scotus .
24
He read
the discussion of equivocity, univocity and denomination as well as the
description of the predicative relationship holding between species and
genera as referring to concepts. Furthermore, he presented the very
description of the ten categories which Aristotle gives in chapters 49
as referring to the way in which our intellect understands the world by
way of ten kinds of concepts.
Scotus s interpretation of the Categories is merely a variant of the
standard thirteenth-century doctrine that categories are studied both
in logic as concepts and in metaphysics as things. In this approach,
a question remained implicit. Are the categories considered in logic
the same items as those considered in metaphysics? Properly speak-
tercia operatio ratiocinandi, secundum quod ratio procedit a notis ad inquisitionem
ignotorum [. . .]. Cum autem logica dicatur rationalis sciencia, necesse est quod eius
consideratio uersetur circa ea que pertinent ad tres praedictas operationes rationis; de
hiis igitur que pertinent ad primam operationem intellectus, id est de hiis que simplici
intellectu concipiuntur, determinat Aristotiles in libro Preadicamentorum; de hiis uero que
pertinent ad secundam operationem, scilicet de enunciatione afrmatiua et negatiua,
determinat Philosophus in libro Peryermenias; de hiis uero que pertinent ad terciam
operationem determinat Aristotiles in libro Priorum et in consequentibus, in quibus agitur
de sillogismo simpliciter et de diuersis sillogismorum et argumentationum speciebus,
quibus ratio de uno procedit ad aliud; et ideo secundum praedictum ordinem trium
operationum, liber Praedicamentorum ad librum Peryermenias ordinatur, qui ordinatur ad
librum Priorum et consequentes. See also Expositio libri Posteriorum, p. 2.
23
Giles of Rome, Expositio super libros Elenchorum Aristotelis (Venetiis, 1500; reprint
Frankfurt am Main, 1968), f. 2rbvb; John Duns Scotus , Quaestiones in duos libros
Perihermenias Aristotelis, eds. R. Andrews et al., in Ioannis Duns Scoti Opera philosophica
II (St. Bonaventure, 2004), Proemium, p. 135, nn. 12.
24
John Duns Scotus , Quaestiones super Praedicamenta Aristotelis, q. 1, pp. 249256.
ing, there is only one sets of things, i.e., categories in the extramental
world. For categories as studied in logic are not things but concepts.
But categories, insofar as they are studied in logic, have certain prop-
erties that they do not have when they are considered in metaphysics.
For example, substance is studied in logic insofar as a logical relation
such as being predicated or being divided into species is grounded on
itbecause more universal substantial concepts are predicated of less
universal substantial ones and generic substantial concepts are divided
into specic ones. Consequently, it seems that categories in logic (i.e.,
as they are studied in the Categories) and in metaphysics are not the
same but different items: in logic, categories are entia rationis (concepts),
in metaphysics they are entia realia (things). This view of considering
categories as things and concepts becomes particularly evident in Scotus
and Brito . Brito even devotes an entire question of his commentary on
the Categories to determine whether the notion of category is a real or
a conceptual being. Symptomatically, he concludes that it is both, for
categories can be considered both as the things to which the intellect
attributes conceptual properties or as modied by such conceptual
properties. In the former case, a category is a real being; in the latter,
it is a conceptual one.
25
This interpretation of the Categories has an obvious consequence. Since
the logician studies concepts and not things whereas the metaphysician
considers things and not concepts, there is the possibility of a gap
between logic as the study of concepts and metaphysics as the study
of things. Specically, the Categories says much concerning the way in
which we understand the world. But is it legitimate to infer from the
analysis of our concepts any conclusion concerning the structure of
the world? Unlike most of his predecessors, Scotus is particularly aware
of the problem implicit in the view that categories can be studied both
in logic as concepts and in metaphysics as kinds of things. He is wary
of bridging the gap between the two approaches. This is all the more
problematic, because he maintains that the ten categories, which are
studied in the Categories as ten concepts, are also ten kinds of extramental
things. But what assures us that there are as many kinds of concepts as
there are kinds of things? In principle, nothing, as Scotus knew well. So
he needs to argue that there are ten kinds of thing without relying on
25
Radulphus Brito , Quaestiones super Praedicamenta, in Quaestiones subtilissimae magistri
Rodulphi Britonis super arte veteri, q. 4, f. 39vb.
160 giorgio pini
the way in which our concepts are structured. In this respect, Scotus can
be regarded as the one who took the doctrine of the twofold study of
the categories (in logic and in metaphysics) to its extreme consequences
and who accordingly showed the limits of such an approach. Basically,
he claims that there is a close parallelism between categories as studied
in logic and categories as studied in metaphysics but that this paral-
lelism cannot be demonstrated. This is of course a problem. It is this
problem which will set the philosophical agenda for those writing on
the Categories at the beginning of the fourteenth century.
26
4. The parting of the ways: Olivi , Ockham , Burley
4.1. Peter of John Olivi
Scotus is among the last to maintain that a coordinated and organic
treatment of categories, as both concepts and things, is possible. As a
matter of fact, his treatment should probably be seen as a bold attempt
to rescue the thirteenth-century doctrine of the twofold consideration of
categories against the criticisms that had already been leveled against it.
As I have mentioned, the weak point of that approach was the possible
gap between categories as concepts and categories as things. Before
Scotus, Peter of John Olivi had already noticed that there was no good
reason to maintain that there are as many concepts as kinds of things.
He claimed that the famous division into ten categories should not be
seen as a distinction of kinds of things but of real accounts, i.e., of the
different ways in which the same thing can be described. For the same
thing can be accurately described in several ways. For example, a man
can be described as the permanent subject of changing properties or as
a thing of a certain extension or as a thing of a certain color or as the
son of his father and mother and so on. This does not mean, however,
that in the world there are different things corresponding to each of
these different descriptions. In the world, there is only one and the same
26
See Duns Scotus , Quaest. super Praed., q. 10, pp. 333342; Quaestiones super libros
Metaphysicorum Aristotelis. Libri IV, eds. R. Andrews et al., in Ioannis Duns Scoti Opera
philosophica III (St. Bonaventure, 1997), L. V, q. 56, nn. 73103, pp. 464471. On
the demonstration of the number of Categories, see G. Pini , Scotus on Deducing
Aristotles Categories, in La tradition mdivale des Catgories (XII
e
XIV
e
sicles). Actes du
XIII
e
Symposium europen de logique et de smantique mdivales (Avignon, 610
juin 2000), ds. par J. Biard I. Rosier-Catach (Louvain, 2003), pp. 2335.
thing, i.e., a man, which can be described in several ways. Olivi calls
these descriptions rationes reales, probably because each of them is a
concept by which we know and describe the world (and so it is a ratio)
and at the same time it is not a ction made up by our intellect but
an accurate description of something in the world (and so it is realis).
Olivi knows that the supporters of the standard view that there are ten
kinds of concepts and ten parallel kinds of things base their opinion
on Aristotles Categories and Metaphysics. Against this claim, Olivi notes
that in both works Aristotle merely assumes the distinction among
categories as something known but he does not seem to consider it as
a distinction among different real essences. Even less does he give any
argument in support of the claim that the ten categories are actually
ten and no fewer than ten. Actually, Olivi notices that Avicenna already
seems to have maintained that there is no demonstration of the number
of categories. Accordingly, Olivi maintains that what Aristotle says in
the Categories can be interpreted in two ways. The rst possibility is that
Aristotle is speaking not of kinds of things but merely of concepts, i.e.,
the ways we describe things; so he would not be drawing an ontologi-
cal distinction but merely a distinction of concepts. It is the fault of
Aristotles interpreters to infer a plurality of things from a plurality
of concepts. The second possibility is that Aristotle himself made the
mistake of inferring an ontological distinction among essences from a
conceptual one among descriptions. Such a mistake, Olivi observes, was
made by many ancient philosophers. The implicit suggestion seems to
be that there is no justication for philosophers and theologians who
still make the same mistake today.
27
27
Peter of John Olivi , Quaestiones in secundum librum Sententiarum, ed. B. Jansen, I (Ad
Claras Aquas, 1922), q. XXVIII, pp. 483484: Quidam enim sequaces Aristotelis
credentes quod Aristoteles voluerit decem praedicamenta esse decem genera rerum
inter se essentialiter distinctarum acceperunt quasi pro primo principio quod hoc ita se
habent, quamquam Aristoteles non inveniatur multum expressisse quod dicant semper
necessario diversas essentias et multo minus inveniatur aliquam rationem sive neces-
sariam sive probabilem ad distinctionem et numerum praedicamentorum probandum
alicubi adduxisse. In Praedicamentis enim nihil facit super hoc nisi variare, acsi quicquid
dicat esset per se notum; nec in Metaphysicis aliquid de hoc probat, sed potius tanquam
notum supponit [. . .]; q. XIV, p. 264: Haec autem distinctio conveniens esse potuit
secundum modum logicalem qui plus considerat rationes rerum et intentiones quam
res; et ideo rationes quae secundum modum intelligendi habebant se ad alias per
modum accidentium convenienter potuerunt ponere inter praedicamenta accidentium.
Aut si hoc non movit eos, dicunt quod Aristoteles fuit deceptus arguens ex pluralitate
rationum realium diversitatem essentialem, quam deceptionem dicunt fuisse in multis
antiquis philosophis.
162 giorgio pini
Olivi s main point is not that Aristotles doctrine in the Categories is
wrong. Moreover, he agrees with the view commonly adopted by his
contemporaries that the Categories introduces not ten kinds of things
but ten kinds of concepts. Starting from this standard point, however,
Olivi argues that it is not legitimate to infer a plurality of things from
a plurality of descriptions or concepts. Accordingly, it is the traditional
doctrine of the parallelism between categories as concepts and things
that in Olivis eyes must be abandoned. Scotus and Brito made a last
effort to defend it but they both were aware that the parallelism between
concepts and things cannot be simply assumed; it must now be defended
with good arguments and put on solid new ground.
Scotus s and Brito s attempts to defend the traditional view of the
categories seem to have been isolated and, ingenious as they were,
short-lived. Confronted with the crisis of the traditional conception,
philosophers took two contrasting attitudes. First, the way was open for
them to maintain that the Categories deals with concepts while denying
that there is a parallel number of kinds of things. This is basically
Olivi s opinion. It is also that of William Ockham (who maintains that
there are only particular substances and qualities) and John Buridan
(who maintains that there are only particular substances, quantities and
qualities). Alternatively, it was possible to reject the traditional claim that
the Categories, being a logical work, deals with concepts, and to maintain
instead that Aristotles treatise should be seen as a work of ontology.
This is the position of Walter Burley and of the Oxford realists of the
late fourteenth century ( John Wyclif and his followers, Robert Alyngton ,
John Sharpe , William Milverley , William Penbygull , Roger Whelpdale,
and John Tarteys ).
28
It must be noted that, although the latter tended
to present themselves as the defenders of the traditional opinion con-
cerning the Categories, their interpretation was no less innovative and
radical than that of Ockham and Buridan. Both nominalists and real-
ists rejected the old doctrine of the twofold consideration of categories
and based their interpretation of the Categories on such a rejection. So
thinkers of both afliations agreed that the Categories is not to be seen
as the logical counterpart of a parallel ontological treatment of cat-
egories (given in the Physics and Metaphysics), as it was commonly held
in the thirteenth century. By contrast, the Categories must be considered
28
See Alessandro D. Conti , Studio storico-critico, in Johannes Sharpe , Quaestio
super universalia, a cura di A. D. Conti (Olschki, 1990), pp. 295336.
as either a logical treatise to which there corresponds a dramatically
simplied ontology (according to nominalists) or an ontological work
whose logical nature consists merely in its introductory and general
character, not in the fact that it deals with categories insofar as they
are concepts (according to realists).
4.2. William Ockham
So let us now consider Ockham s position in his commentary on the
Categories, written between 1321 and 1324.
29
As is well known, Ockham
endorses a form of ontological reductivism. He maintains that there
are only two kinds of things, i.e., individual substances and individual
qualities. All the other categories are not things distinct from substances
and qualities but result from a certain way of signifying a particular
substance and a particular quality while signifying something else (i.e.,
another particular substance or another particular quality) in a second-
ary way.
30
This is the gist of Ockhams famous doctrine of connotative
terms, both linguistic and mental. The main result of this doctrine is that
Ockham can posit ten distinct kinds of words and their corresponding
concepts, which are properly speaking the categories, while at the same
time holding a noteworthy ontological parsimony. Thus, in Ockhams
interpretation, the Categories is a treatise devoted to the study of terms
insofar as they signify. Even though Ockham rst describes these terms
as vocal expressions (voces), he explicitly adds that they include both vocal
and mental words (intentiones or conceptus). Accordingly, the Categories deals
with terms both vocal and mental insofar as they signify extramental
things. Ockham quotes Boethius as the source for his reading of the
Categories: terms are considered as signs of extramental things, i.e., of
individual substances and qualities, which for Ockham are the only
constituents of the world. Consequently, Aristotles Categories primar-
ily concerns terms and only secondarily things, insofar as things are
29
Here I do not take Buridan into consideration. See J. Zupko , John Buridan Portait
of a Fourteenth-Century Arts Master (Notre Dame, IN, 2003), pp. 5458.
30
On Ockham s ontological program and the connection with his semantic views, see
M. McCord Adams , William Ockham (Notre Dame, IN., 1987), pp. 143167; C. Panaccio ,
Les mots, les concepts et les choses. La smantique de Guillaume dOccam et le nominalisme daujourdhui
(Montral, 1991), pp. 2367; Id., Connotative Concepts and Their Denitions in Ockhams
Nominalism, in La tradition mdivale des Catgories, pp. 141155; G. Klima , Ockhams
Semantics and Ontology of the Categories, in The Cambridge Companion to Ockham, ed.
P. V. Spade (Cambridge, 1999), pp. 118142.
164 giorgio pini
signied by terms. Specically, this implies that most of the statements
that we read in the Categories refer to words and concepts, not to things.
Ockham notes that many modern authors do not realize that Aristotles
intention is to deal with terms and not with things; as a consequence,
these authors endorse a realist interpretation of Aristotles treatise which
is completely off target.
31
This is probably a reference to Burley s rst
commentary on the Categories, which precedes Ockhams commentary
by some years.
What is the function of the Categories and why is it the rst work of the
Organon and more generally of the entire series of Aristotles writings?
Ockham considers the short treatise as a sort of introduction to the study
of philosophy. He remarks that the study of verbal and mental terms
insofar as they signify things is necessary for any kind of philosophical
dispute, because any philosophical dispute takes place through terms.
Specically, the study of terms according to their signication as car-
ried out in the Categories allows one to distinguish between two classes
of terms, i.e., absolute and connotative. Absolute terms signify what
they supposit for. Connotative terms primarily signify what they sup-
posit for but also have a secondary signication; for example, a relative
term like son primarily signies all the individual men (except Adam)
and secondarily signies all the individual men who are fathers of the
men primarily signied. Since the Categories studies terms according to
their signication, the distinction between absolute and connotative
signication pertains to the Categories. Since the confusion between these
two modes of signication causes many mistakes (such as the mistake
of positing as many kinds of things as there are signifying terms), the
Categories performs a sort of therapy preliminary to any philosophical
dispute. Curiously, Ockham does not seem to notice that Aristotle
himself does not draw any distinction between absolute and connota-
31
William Ockham , Expositio in librum Praedicamentorum Aristotelis, ed. G. Gl, in Opera
philosophica II (St. Bonaventure, N.Y., 1978), pp. 135136: Circa primum sciendum
estsequendo dictum Boethii in Commento suoquod in hoc opere haec intentio est, de
primis rerum nominibus et de decem vocibus res signicantibus disputare [. . .]. Unde
sciendum est quod principalis intentio in hoc libro est de vocibus res signicantibus
determinare. [. . .] Verumtamen non obstante quod Philosophus in isto libro tractet de
vocibus, tractat tamen simul cum hoc de rebus; hoc est, propositiones multas ponit
accidentaliter in quibus termini non pro nominibus sed pro rebus supponunt. [. . .]
Hoc est, secundo loco tractat de iis rebus pro quibus supponunt. Et ignorantia istius
intentionis Aristotelis in hoc libro facit multos modernos errare, credentes hic multa
dicta pro rebus, quae tamen pro solis vocibuset proportionaliter pro intentionibus
seu conceptibus in animavult intelligi.
tive terms either in the Categories or elsewhere. Since such a distinction
is central to his own philosophical project and since he maintains that
the Categories is a study of verbal and mental terms according to their
signication, Ockham just claims that the distinction between the two
kinds of terms is the main aim of the Categories itself.
So what is the role and place of the Categories in the Organon? What
about the old triad of CategoriesDe interpretationePrior Analytics?
Ockham still mentions it approvingly. Its inuence can still be easily
detected in the structure of his Summa logicae. And he still maintains
that the Categories studies simple terms out of which the propositions
analyzed in the De interpretatione are made and that in turn these propo-
sitions constitute the syllogisms studied in the Prior Analytics.
32
This
schema, however, does not play a very important function in Ockhams
thought. There seem to be at least two reasons for this. The rst reason
is Ockhams rejection of syllogism as the subject matter of logic. For
Ockham maintains that logic, like all the other sciences, does not have a
single subject matter. In any science, there are as many subject matters
as there are things studied. Logic does not have a single subject matter
at which all its parts are aimed. Consequently, syllogism cannot play the
role of the ultimate end of logic and the whole sequence CategoriesDe
interpretationePrior Analytics loses its signicance.
33
The second reason for
this is Ockhams rejection of the traditional formulation of the doctrine
of the three operations of the intellect. As I have mentioned, at the
end of the thirteenth century it was customary to link the sequence
CategoriesDe interpretationePrior Analytics with the three operations of
apprehending, combining and arguing. Specically, the Categories was
taken to study the objects of apprehension, i.e., simple concepts. For only
simple concepts (such as man and animal, for example) can be appre-
hended by the rst operation of the intellect and successively combined
by the second operation in the proposition Man is an animal. Now,
Ockham follows Scotus in distinguishing a mental content, which has
a propositional form, from its being asserted as true or false. For not
only simple concepts but also propositions can be apprehended without
32
William Ockham , Exp. in librum Praed., p. 164: [. . .] Philosophus principaliter
in isto libro determinat de incomplexis, ex quibus unt propositiones et syllogismi.
Ockham also constructs his Summa logicae out of three parts respectively devoted to the
study of termsstudy of propositionsstudy of arguments, which clearly reproduces
the sequence CategoriesDe interpretationePrior Analytics.
33
William Ockham , Exp. in librum Praed., p. 135; cf. Proemium Expositionis in libros Artis
Logicae, ed. G. Gl, in Guillelmi de Ockham, Opera philosophica II, pp. 37.
166 giorgio pini
being the object of assertion by the second operation of the intellect.
This happens when a propositional content such as mans being an
animal is merely contemplated without taking any stance concerning
its being true or false. Ockham concludes from this possibility (which
apparently was not part of the standard picture of predication endorsed
until the end of the thirteenth century, for example by Thomas Aquinas )
that not only simple concepts such as man but also complexes such as
mans being an animal can be the object of the rst operation of the
intellect.
34
As a consequence, the link between the rst operation of
the intellect and the categories is loosened. Simple apprehension is not
necessarily turned to the simple concepts studied in the Categories; even
mental propositions can be its object.
All the same, even for Ockham the Categories keeps its place as the
opening work of Aristotles Organon. But this is due not so much to
the old schema according to which the Analytics are the culmination
of logic as to the fact that, as we have seen, Ockham considers the
Categories as a sort of therapy preliminary to the rest of philosophy,
insofar as any philosophical dispute relies on signifying terms. The
Categories is necessary in order to get rid of the fallacy that derives from
not distinguishing between absolute and connotative terms. So, the
main utility of the Categories is to make clear that there is no one-to-
one correspondence between kinds of terms (verbal or conceptual) and
kinds of things and that accordingly our conceptual apparatus of ten
categories is compatible with an ontology comprising only individual
substances and individual qualities.
35
34
See William Ockham , Scriptum in librum primum Sententiarum ordinatio. Prologus et distinc-
tio prima, ed. G. Gl, in Guillelmi de Ockham, Opera theologica I (St. Bonaventure, N.Y.,
1967), Prol., q. 1, pp. 1617; Quodlibeta septem, ed. J. C. Wey, in Guillelmi de Ockham
Opera theologica IX (St. Bonaventure, N.Y., 1980), Quodl. III, q. 8, pp. 232235; Quodl.
IV, q. 16, pp. 376380; Quodl. V, q. 6, pp. 500503. See G. Nuchelmans , Theories of the
Proposition. Ancient and Medieval Conceptions of the Bearers of Truth and Falsity (Amsterdam,
1973), pp. 197199; G. Pini , Scotus on Assertion and the Copula: A Comparison
with Aquinas , in Medieval Theories on Assertive and non-Assertive Language. Acts of the 14th
European Symposium on Medieval Logic and Semantics, Rome, June 1115, 2002,
eds. A. Maier L. Valente (Olschki, 2004), pp. 307331.
35
William Ockham , Exp. in librum Praed., pp. 136137: Utilitas istius libri est scire
quae nomina quas res signicant. Et hoc multum necessarium est ad omnem dispu-
tationem, cum omnis disputatio signicata vocabulorum praesupponat. Et etiam ista
scientia specialiter utilis est ad notitiam fallaciae gurae dictionis, cuius ignorantia
multos involvit. Nam cum aliquae voces aliquas res absolutae signicent et aliquae
in comparatione ad alias, et alias res connotando, arguere a primis ad secundas
frequenter est facere fallaciam gurae dictionis, quod est valde commune; et talia
4.3. Walter Burley and Robert Alyngton
Walter Burley s second commentary on the Categories (written in 1337)
must be seen as a reaction to Ockham s views.
36
Burley, however, does
not revive the old doctrine of the twofold consideration of categories
and of the parallelism between categories as concepts and categories
as things. By contrast, he maintains that Aristotles Categories deals
with things. Burley is aware of going against no lesser authorities than
Boethius and Simplicius , who both claimed that the Categories deals with
things only in a secondary way, namely, only to the extent in which it
is inescapable to speak of things in a treatise concerned with signify-
ing expressions. So he is aware of putting forward an innovative view
of the Categories. But he too has some authorities in his support. He
refers to Avicenna and Averroes . His main argument is that terms are
necessary only for the sake of communication, but logic could dispense
with them if human beings were capable of non-verbal communica-
tion.
37
It is remarkable that Burley, while considering the Categories de
facto as an ontological treatise, does not deny its belonging to logic.
He only modies the conception of logic: logic itself appears to be
a sort of study of the ontological structure of reality. As a matter of
fact, this is not surprising in an author who speaks of such things as
propositiones in re.
38
So contrary to an old tradition that goes back to Porphyry and
Boethius , Burley claims that the main focus of the Categories is things
and that terms are only secondarily dealt with. Indeed, Burley has a
very good argument in his support: the very content of the Categories.
sophismata a multis demonstrationes creduntur. Sed qualiter hoc debeat declarari, in
libro Praedicamentorum ostendetur.
36
See Alessandro D. Conti , Ontology in Walter Burley s Last Commentary on the
Ars Vetus, Franciscan Studies 50 (1990), 121176.
37
See Walter Burley , Expositio super Praedicamenta Aristotelis, in Gualteri Burlei Expositio
super artem veterem Porphyrii et Aristotelis (Venetiis, 1509), f. 17va: Si enim esset possibile
solo intellectu dicere, non indigeremus verbis. Si enim doctor artis posset illud revelare
quod est in anima sua alio modo quam loquendo, semper supersederet a verbis. Unde
si logica posset docere alio modo quam per voces, ut puta per signa digitorum vel per
nutum oculorum, logicus non indigeret vocibus. Quod autem logicus considerat de
vocibus, hoc non est ex primaria intentione, sed secundaria et ex consequenti, quia
scientia logicalis non aliter commode doceri potest. Unde si aliquis surdus a nativitate
posset invenire artem silogizandi et demonstrandi, tunc posset logicam sine cognitione
vocum habere [. . .].
38
See L. Cesalli , Le ralisme propositionnel de Walter Burley , Archives dhistoire
doctrinale et littraire du Moyen Age 68 (2001), 155221.
168 giorgio pini
Those who maintain that the Categories deals with terms and not with
things had always been embarrassed when trying to explain the central
chapters of the treatisedevoted to the study of the ten categoriesas
dealing with terms and properties of terms rather than with things.
Burley thinks that there is a much more straightforward reading. As
he remarks, most of what Aristotle says in his short treatise is about
properties of things, not properties of terms: This view [i.e., that the
Categories deals primarily with things] seems to agree with Aristotles
way of proceeding. For in this book he discusses the properties and
attributes of things, such as the properties of substance, quality and
quantity. Now, a science is mainly about the things whose properties
and attributes it mainly discusses. But in this book it is not properties
of expressions that are discussed but of things. Therefore etc.
39
Burley s interpretation of the Categories as a treatise dealing with the
basic structure of reality is the source of the realist interpretations of the
Categories adopted by the so-called Oxford realists of the late fourteenth
century. For example, Robert Alyngton (. ca. 1390) maintained that in
the Categories Aristotle deals mainly with things signied by terms, even
though sometimes he does speak of the expressions signifying things.
Again, the traditional remark by Porphyry and Boethius is plainly con-
tradicted. But like Burley, Alyngton does not conclude that the Categories
is a metaphysical treatise. Quite the contrary, he claims that we must
distinguish between the order in which logicians carry out their task
and their aim. It is true that, as far as the order is concerned, logi-
cians rst deal with words then with things. Their aim, however, is to
speak of the things as they are in the extramental world. So, what we
read in the Categories about predications of genera and species must be
interpreted as referring not to relationships among concepts but to real
relationships holding among things of different degrees of universal-
it y.
40
Again, there emerges a conception of logic as a sort of ontology,
39
Walter Burley , Exp. super Praed., f. 17vb: Istud videtur concordari processui
Philosophi, quia Philosophus determinat in hoc libro de proprietatibus et passionibus
rerum, ut de proprietatibus substantiae, qualitatis et quantitatis. Nunc autem scientia
videtur de istis esse principaliter, de quorum proprietatibus et passionibus principaliter
determinat in illa; sed in hoc libro non determinatur de proprietatibus vocum sed
rerum; ergo et cetera.
40
Robert Alyngton , Litteralis sententia super Praedicamenta Aristotelis, in Alessandro D.
Conti , Linguaggio e realt nel commento alle Categorie di Robert Alyngton, Documenti e
studi sulla tradizione losoca medievale 4 (1993), 179306: 251: Quantum ad primum vide-
tur mihi quod Aristoteles hic mixtim tractat de signis et rebus signicatis; sed principalius
de rebus signicatis. Nam cum dupliciter potest dici prius: aut in ordine procedendi,
dealing with the structure of reality from a general point of view. In
this perspective, the old sequence CategoriesDe interpretationePrior
Analytics is to be reinterpreted as regarding simple and complex items
in the world. It is the world itself that is divided into simple objects,
propositional objects and complex objects linked by a relationship of
implication. It is the world, not merely our conceptual framework,
that is logically structured. For this reason, logic deals with things, not
with concepts.
So the Categories deals with simple items in the world. But then, given
that Aristotle deals with categories not only in the treatise of that name
but also in the Metaphysics, what are we to say about the difference
between these two treatments? For there are indeed several differences,
even some contradictions between what Aristotle says in the Categories
on the one hand and in the Metaphysics on the other hand. Alyngton is
probably not aware of the old doctrine of the twofold consideration of
categories, so he cannot explain away the contradictions between the
two works as the result of treatments carried out from different points
of view. By contrast, Alyngton claims that the difference between the
Categories and the physical and metaphysical treatises is that the former
work is not scientic but only probable. Indeed, the whole difference
between logic and metaphysics is merely one of certitude: the logician
speaks according to what is likely, the real philosopher speaks scienti-
cally. But they speak of exactly the same things.
41
What we read in the
Categories must be interpreted as a sort of introductory treatment of
the same issues that will be studied more carefully in the Metaphysics. The
difference between the two works can also be seen as one concerning
sicut pars est prior toto cuius est pars, aut secundo modo prius intentione, sicut totum
est prius sua parte. Prius ergo in ordine procedendi debet logicus tractare de vocibus
quam de rebus. [. . .] Principalior tamen et prior in intentione est logici determinatio
de rebus distinctis contra signa, quamvis posterior in ordine procedendi.
41
Robert Alyngton , Litt. sent. super Praed., ed. Conti , p. 252: Notandum secundo quod
aliter determinat hic de substantia, qualitate, quantitate etc. et aliter in V Metaphysicae.
Nam secundum Avicennam, in principio Metaphysicae suae, metaphysicus procedit scien-
tice circa singula quae conveniunt in ente analogo, logicus autem probabiliter. Et
secundum sanctum Thomam, in IV Metaphysicae, logica potest considerari uno modo
ut est scientia docens modum probabiliter procedendi, aut modum sciendi; et tunc
considerat de rebus sub ratione qua eis insunt intentiones secundae, ut sub ratione qua
est commune vel singulare, genus etc. Et tunc procedit scientice. Et tunc habet pro
subiecto, ut dicit Avicenna et Lincolniensis, I Posteriorum, intentiones secundas prout
oppununtur primis. Sed secundo modo potest logica considerari ut utens. Et tunc
considerat circa subiectum aeque universale sicut metaphysicus, sed non scientice. Et
sic, secundum Lincolniensem, logica est instrumentum metaphysicae.
170 giorgio pini
their didactic destination: the Categories is a work for beginners, the
Metaphysics is for advanced students.
Both the idea that logic deals with things at a higher level of generality
than metaphysics and the connected presentation of the Categories as a
work for absolute beginners are quite old. They rst appeared among
Neoplatonic commentators eager to reconcile Aristotles logical and
metaphysical works. Curiously, this view has been recently revived in
contemporary scholarship on Aristotle.
42
Its merit is to allow for an
ontological reading of the Categories while recognizing the contradictions
with the Physics and Metaphysics. The alternative view is that the Categories
and the physical and metaphysical treatises have only apparently the
same subject, for the former adopts a logical point of view, the latter
a metaphysical one.
43
This illustrates an obvious point which we may
forget: contemporary interpretations of Aristotle are a continuation and
quite often a reconsideration of ancient and medieval ones.
5. A Case Study: The Distinction Between Primary and Secondary Substances
So far, I have argued that the idea that the Categories is the rst work of
the logical syllabus continued to be entertained at least until the end of
the fourteenth century, no matter what logic was taken to bea formal
study of arguments, a study of concepts, a study of mental language
or a beginners guide to the study of the basic structure of reality. The
view that the Categories is a work of logic had a lasting inuence on
both the interpretation of that work and more in general the history
of logic. On the one hand, according to which conception of logic
was adopted, it was possible to read different things in the Categories: an
introduction to arguments, a study of simple concepts, an analysis of
the structure of reality. On the other hand, logic itself was construed
as to include what could be read in the Categories. For example, the
42
See M. Burnyeat , A Map of Metaphysics Zeta (Pittsburgh, 2001), pp. 87125. The
idea that the Categories is a work for beginners originated from the second-century
Aristotelian, Herminus, and ourished in the Neoplatonic circles from Porphyry
onwards. See P. Moraux , Der Aristotelismus bei den Griechen. Von Andronikos bis Alexander
von Aphrodisias, vol. 2 (Berlin, 1984), pp. 363374.
43
See for example Joseph Owens , Aristotle on Categories, The Review of Metaphysics
14 (1960), 7390: 8384; Bods , Introduction, in Aristote, Catgories, p. XCVI. Owenss
positionwhich Bods quotes approvinglyis directly indebted to Thomas Aquinas s
commentary on the Metaphysics (see below n. 46).
distinctions between individuals and universals and between substance
and accident were considered as parts of what logic had to account
for because Aristotle introduces them in the Categories. The two-way
inuence of logic on the Categories and of the Categories on logic is of
great concern to the evaluation of several arguments and positions put
forward by thirteenth- and fourteenth-century authors in the realm of
logic and metaphysics. In order to illustrate this point, I will consider
how some of the late medieval thinkers I have already mentioned
interpreted the very same lines of chapter 5 of the Categories. These
thinkers are Thomas Aquinas , John Duns Scotus , William Ockham ,
Walter Burley and Robert Alyngton .
5.1. Thomas Aquinas
In the rst lines of Cat. 5, Aristotle introduces the famous distinction
between primary and secondary substances:
A substancethat which is called a substance most strictly, primarily, and
most of allis that which is neither said of a subject nor in a subject, e.g.,
the individual man or the individual horse. The species in which the things
primarily called substances are, are called secondary substances, as also
are the genera of these species. For example, the individual man belongs
in a species, man, and animal is a genus of the species; so theseboth
man and animalare called secondary substances.
44
Aristotles point is straightforward. He distinguishes two kinds of objects,
some particular and others universal. As it happens, Aristotles endorsing
of universal substances is usually regarded as one of the main points
of contrast between the ontology of the Categories and that of the
Metaphysics, where Aristotle holds that no universal is a substance.
45
In modern times, scholars have proposed several solutions to this
predicament. The difculty, however, was already known in the Middle
Ages. Thomas Aquinas wrote no commentary on the Categories but in
his commentary on the Metaphysics he tries to resolve the contradiction
between the latter work and the Categories. It is in that context that he
44
Aristotle, Cat. 5, 2a1119 (Akrills translation). See Aristotles Categories and De inter-
pretatione. Translated with Notes and Glossary by J.L. Ackrill (Oxford, 1963), pp. 56.
45
Aristotle, Met. Z 13. See Bods , Introduction, in Aristote, Catgories, pp. XCIV
XCV. For an analysis of various contemporary interpretations see G. Galluzzo , Met.
Z 13 in the Contemporary Debate and in Aquinas s Interpretation, Documenti e studi
sulla tradizione losoca medievale 14 (2003), 159226: 161185.
172 giorgio pini
provides his interpretation of the distinction between primary and
secondary substances as introduced in Categories 5.
Aquinas maintains that Aristotles admitting of universal substances
in the Categories must be interpreted in the light of the logical character
of this work. Since logicians deal with things not insofar as they exist in
the extramental world but insofar as they are understood and present
in the mind, it follows that they consider substances insofar as they are
understood by the intellect, i.e., insofar as substances are conceived of
as universal concepts. For example, let us consider the meaning that the
metaphysician and the logician give to the word man. For the meta-
physician, man is a real essence constituting each individual belonging
in a certain natural kind. By contrast, the logician considers such an
essence insofar as the intellect understands it as a universal concept,
i.e., as the universal concept of man. It is this universal concept that
the logician calls secondary substance. Since our intellect attributes
this universal concept (for example, the universal concept of man) to
the real extramental thing that has been understood by way of that
concept, it follows that according to the logician secondary substances
are predicated of primary substances, i.e., universal concepts are
predicated of the real things existing in the world. In the Categories,
it is this logical consideration that is taken into account. Accordingly,
in the Categories Aristotle says that secondary substances (i.e., universal
concepts) are predicated of primary substances (i.e., real individuals
in the world):
But it must be said that in the Categories the Philosopher speaks according
to a logical consideration. The logician, however, considers things insofar
as they are in the mind; accordingly, he considers substances to the extent
that they are subject to the intention of universality insofar as the intellect
conceives of them. Therefore, concerning predication, which is an act of
the intellect, the logician says that secondary substances are predicated
of a subject, i.e., of the substance existing extramentally.
46
46
Thomas Aquinas , In duodecim libros Metaphysicorum Aristotelis Expositio, edd. M.-R.
Cathala R. M. Spiazzi (Rome, 1964), L. VII, lect. XIII, n. 1576: Sed dicendum
quod secundum logicam considerationem loquitur Philosophus in Praedicamentis. Logicus
autem considerat res secundum quod sunt in ratione; et ideo considerat substantias prout
secundum acceptionem intellectus subsunt intentioni universalitatis. Et ideo quantum
ad praedicationem, quae est actus rationis, dicit quod praedicatur de subiecto, idest
de substantia subsistente extra animam.
Things are different in the Metaphysics, where Aristotle focuses not on
our concepts but on the way the world is. Thus, in the Metaphysics
Aristotle only considers real individuals belonging in certain natural
kinds. From that point of view, there is no room for predicating a
substance of another substance, for only concepts, not things, are
predicated. Consequently, the only relationship that Aristotle takes into
account in the Metaphysics is the relationship of being in or inhering
in a substance. He ignores the other relationship he had introduced in
the Categories, i.e., being said of, which holds not between two things
but between a concept and a thing or between two concepts of differ-
ent universality.
47
From this passage, it appears that Thomas Aquinas interprets the dis-
tinction between primary and secondary substances drawn in Categories
5 as a distinction not between two kinds of extramental objects (i.e.,
individuals and universals) but between the real objects existent in the
world (i.e., primary substances) and the universal concepts by which
our mind understands them (i.e., secondary substances). So Aquinas
takes Aristotle as distinguishing between things (individual and real)
and concepts (universal and mental).
5.2. Duns Scotus
Scotus shares with Thomas Aquinas the view that the Categories deals
with concepts. Some twenty years after Aquinas, he gives a similar
interpretation of the distinction between primary and secondary
substances. He states that the distinction drawn at Cat. 5, 2a1119 is
not to be interpreted as the division of a genus (i.e., the category of
substance) into its species but as the distinction of a subject accord-
ing to its accidents. Scotus is here implicitly referring to Boethius s De
divisione, where several kinds of divisions are introduced and described.
In particular, Boethius distinguishes the way in which a genus such as
animal is divided into its species from the way in which a subject is
divided according to which accidents inhere in it, for example when
men are divided into black, white and neither black nor white accord-
ing to the color of their skin. The latter division is clearly not of a
genus into different species, because all men belong to the same spe-
cies no matter what the color of their skin is; it is merely a division
47
Ibid.
174 giorgio pini
carried out according to whether an accidental feature does or does
not inhere in a subject.
48
Similarly, Scotus maintains that primary and
secondary substances are not two species of the same genus; the divi-
sion of substances into primary and secondary is merely the division
of a subject according to which accidents inhere in it. Scotus has an
argument for his claim. He recalls that substance is considered in logic
insofar as it is understood and not insofar as it is a real kind of thing.
Now, he notices that substance, when considered as understood (and
so as a concept and not as a thing), is divided into different varieties
according to which kind of conceptual properties (intentiones) pertain to
them: Note that the division of substance into primary and secondary
is not the division of a category into species but of a subject according
to its accidents, because substance insofar as it is understoodaccord-
ing to which consideration it is considered by the logicianis divided
into intentions pertaining to it in an accidental way.
49
Thus a substance, for example, can be considered by the intellect
according to its individual features. In that case, it is regarded as an
individual within a certain kind. Alternatively, the intellect can consider
the same substance by neglecting its individual features. In that way,
that substance is regarded as a certain notion describing more than
one individual. These two ways of considering the same substance do
not have any bearing on the very essence of substance, which remains
the same essence no matter how we consider it. Scotus maintains that
the division of substances into primary and secondary captures exactly the
two ways in which our intellect can consider substances. Granted, he
does not think that the existence of individual and universal items in
the world depends on the way we consider the world. He only maintains
that the way in which our intellect can conceive of substances does
not imply any correspondent distinction in reality. As a consequence,
Scotus holds that when Aristotle distinguishes between primary and
secondary substances he should not be interpreted as positing two kinds
of substances, particular and universal. What Aristotle is saying is that
substance can be considered in two ways and that substances can be
(accidentally) divided into two classes according to whether our intellect
48
Boethius , De divisione liber, ed. J. Magee (Leiden, 1998), pp. 810.
49
Duns Scotus , Quaest. in Praed., q. 12, n. 31, p. 363: Nota quod divisio substantiae
in primam et secundam non est divisio generalissimi in species, sed subiecti in acciden-
tia, quia substantia secundum quod intelligitursecundum quam considerationem
pertinet ad logicumdividitur in intentiones sibi accidentes.
consider them according to the properties that pertain to them insofar
as they are conceived as individuals of a certain kind or as universal
concepts abstracting from individual features. Accordingly, Cat. 5,
2a1119 cannot be taken as evidence of the existence of universals in
addition to individuals. The argument for the existence of universals
as real constituents of the world (if there are any such things) must
be based on a different kind of evidence, not logical (i.e., depending
on the way our intellect considers the world) but metaphysical or, in
alternative, concerning the very grounds and legitimacy of our con-
ceptual framework.
5.3. Ockham
Ockham s interpretation of Cat. 5, 2a1119 is not so different from that
of Scotus . Both Scotus and Ockham maintain that the Categories deals
not with things but with concepts. Consequently, both of them hold
that the distinction between primary and secondary substances must
be accounted for as a distinction between two kinds of concepts and
ways of considering the world, not between two kinds of things. Of
course, Ockham also thinks that in the world there are only particular
substances and qualities. By contrast, Scotus maintains that there are
things belonging to all the ten categories and that individuals are not the
only furniture of the world. But this difference in their metaphysics has
no immediate bearing on their logical conceptions as expressed in their
reading of Aristotles Categories. So Ockham agrees with Scotus that in
Categories 5 Aristotle is not dividing the category of substance into two
kinds, universal and individual substances. In the extramental world,
there is only one kind of substances, i.e., individual substances. Aristotle
is merely dividing the logical category of substance into its contents.
Now, Ockham holds that in the logical category of substance there are
not only things (i.e., the individual substances existing in the world) but
also the names signifying them. These names are either individual, if
they signify only one individual substance, or universal, if they signify
many individual substances. What Aristotle calls primary substances
are either the real individual substances existing in the world or their
names. What Aristotle calls secondary substances are the concepts
and the names signifying many individual substances:
First of all, it must be noted here that every substance that is a real thing
contained in the category of substance is simply a primary, individual
and singular substance. So, the genera, species and very category of
176 giorgio pini
substance are not real extramental substances, nor are they part of the
essence of particular substances. By contrast, they are concepts in the
soulwhich in no way exist extramentallyand expressions or signs
referring to real substances. Therefore, this division of substance is not
a division of substance into extramental things, each of which is a real
thing, but a division of substance into what is contained in it, i.e., into
what is less universal than the category of substance. And not only things
but also concepts in the soul or expressions are said to be contained in
the category of substance or under that category because they are less
common [than the category of substance]. So primary substances are
said to be either the extramental substances or their names and second-
ary substances are said to be the concepts or names referring to more
than one substance.
50
Thus, the reason why Aristotle introduces this distinction between
primary and secondary substances has nothing to do with metaphys-
ics. He is not interested in distinguishing two kinds of substances but
in sorting out two kinds of names and concepts, i.e., those signifying
only one thing and those signifying many things. Since the difference
between these two kinds of simple terms (both linguistic and mental)
is philosophically signicant and since the Categories is concerned with
simple terms, Aristotle draws such a distinction in this place. But he
certainly does not intend to claim that secondary substances are real
items in the world. To say this would result in a basic misunderstanding
of Aristotles aim in his Categories.
51
5.4. Walter Burley and Robert Alyngton
In the second version of his commentary on the Categories, Walter
Burley extensively quotes and criticizes Ockham s interpretation of Cat.
50
William Ockham , Exp. in Praed., p. 164: Notandum est hic primo quod omnis
substantia quae est vera res contenta in genere substantiae est simpliciter substantia
prima et individua et singularis. Ita quod genera et species et ipsummet genus genera-
lissimum substantiae non sunt verae substantiae extra animam, nec sunt de essentia
substantiarum particularium, sed sunt quaedam intentiones in anima, nullo modo
extra exsistentes, et voces vel signa importantia veras substantias. Et ideo ista divisio
substantiae non est divisio substantiae in res extra animam, quarum quaelibet sit vera
res, sed est divisio substantiae in contenta in inferiora ad genus generalissimum, et talia
contenta non sunt tantum res, sed sunt etiam intentiones in anima vel voces importantes
veras res et substantias, quae ideo dicuntur contenta in genere generalissimo vel sub
tali genere, quia sunt minus communia, ut sic substantiae primae dicantur vel ipsae
substantiae extra, vel nomina illarum rerum, et secundae substantiae dicantur intentiones
vel nomina importantia multas substantias. See also ibid., p. 152.
51
See Ockham s arguments against universals ibid., pp. 164ff.
5, 2a1119. Burleys own interpretation of the distinction between pri-
mary and secondary substances is clearly ontological. By distinguishing
between primary and secondary substances, Aristotle introduces a clas-
sication of two kinds of real items existing in the world: individual and
universal substances. What Aristotle is talking about is not a distinction
between things and concepts or between two ways of considering things;
he is plainly speaking about the world and its structure.
52
For this reason,
Burley must explain the passage at 3b1023, where Aristotle says that
primary substances signify something particular (a hoc aliquid) whereas
secondary substances signify something qualied (a quale quid). The
reference to primary and secondary substances as signifying something
was easy to explain for those who maintained that the distinction was
not between two kinds of things but two kinds of names or concepts:
things do not signify, names and concepts do. By contrast, Burley has
to give some account of this passage, which seems to run counter to his
ontological interpretation of the distinction between primary and sec-
ondary substances. Quite simply, he explains that Aristotles reference to
primary and secondary substances as signifying something must be taken
as the shortening for names of primary and secondary substances: it
is the name of a primary substance that signies something particular,
i.e., a primary substance, and the name of a secondary substance that
signies something qualied, i.e., a secondary substance. Both primary
and secondary substances, however, are items of the world and as such
they signify nothing but are just what they are:
It must be understood that when it is said a primary substance signi-
es something determinate, by primary substance the Philosopher
understands the name of a primary substance. Therefore, the name
of a primary substance signies something particular, and a primary
substance is something particular. Similarly, when it is said that a sec-
ondary substance signies something qualied, by secondary substance
the Philosopher means the name of a secondary substance. It must be
understood that by something particular the Philosopher means a singu-
lar, namely, something numerically one that is not predicable of several
things. Hence to signify something particular is the same as to signify
something particular and manifest, such as a singular manifest to the
senses. But something qualied is the same as common; and to signify
something qualied is the same as to signify something common [. . .].
Therefore, man and animal and such names of secondary substances
signify something qualied, because they signify something common. But
52
Walter Burley , Exp. super Praed., f. 18ra.
178 giorgio pini
they do not signify something particular, because they do not signify a
certain determinate individual.
53
Some fty years afterwards, Robert Alyngton gives a particularly strik-
ing rendering of the realistic interpretation of the distinction between
primary and secondary substances. A secondary substance is a common
form, i.e., a real item in the world. For example, humanity is a real
form common to all men. It is the existence of such common forms
in the world that explains why we can attribute the predicate man to
all individual men in our predications. The nominalistic view that the
community on which our predications are based merely consists in our
attributing the same predicate to many individuals is plainly wrong, as
is the view that secondary substances are not common forms in the
world but terms or concepts:
Subsequently, he [scil., Aristotle] describes secondary substancewhich
some call substantial common term; but we maintain that it is a com-
mon form. He says that secondary substances are said to be the species
in which primary substances are as subjective parts or as things that are
by themselves under them, such as a certain man is in the species of
man and the genus of man is an animal. Therefore, these species and
their generasuch as the common form humanity, which Porphyry
calls common man, and the general form animalityare secondary
substances. For just as all men have in common their being men, so in
all of them there is the same human nature or the same human species
or the same humanity in commonwhich is to say the same. And such
a thing is called secondary substance. For such a specic commonality is
not grounded merely on a common term or on its predicability, because
even when there is no common term such a specic commonality in
human nature is not less strong. Therefore the human species is not a
term or a concept.
54
53
Ibid., ff. 25vb26ra, quoted in Conti , Ontology, p. 156: Intelligendum quod,
cum dicitur prima substantia signicat hoc aliquid, per primam substantiam intel-
ligit Philosophus nomen primae substantiae. Nomen ergo primae substantiae signicat
hoc aliquid, et prima substantia est hoc aliquid. Similiter, cum dicitur quod secunda
substantia signicat quale quid, per secundam substantiam intelligit nomen secundae
substantiae. Intelligendum est quod per hoc aliquid intelligit Philosophus singulare
sive unum numero non praedicabile de pluribus. Unde signicare hoc aliquid est sig-
nicare hoc aliquid demonstratum, cuiusmodi est singulare demonstratum ad sensum.
Sed quale quid idem est quod commune; et signicare quale est signicare commune
aliquid [. . .]. Ideo homo et animal et huiusmodi nomina secundarum substantiarum
signicant quale quid, quia signicant quid commune. Sed non signicant hoc aliquid,
quia non signicant aliquod unum individuum determinatum.
54
Robert Alyngton , Litt. sent. super Praed., ed. Conti , p. 267: Consequenter describit
What is the conclusion of this brief review of some readings of the
distinction between primary and secondary substances in Categories 5,
2a1119? At rst sight, the moral could be depressing: Aristotle can
be bent to any interpretation, no matter how far it is from the original
sense of his words. What matters is not so much what Aristotle actu-
ally said but the philosophical presuppositions of his interpreters. This
is not the only possible conclusion, however. A much more interesting
possibility is to consider the vitality of Aristotles text. So fundamental
are the problems raised by Aristotle (such as the distinction between
universals and particulars) that philosophers have never ceased to devise
new solutions to the old problems. It must be credited to the vitality
and originality of Aristotelianism that each of these solutions has a
legitimate claim to restitute the sense of Aristotles own words. If we
take into account that the Categories was considered as a logical work
and that the conception of logic changed all along the thirteenth and
fourteenth century, the various readings of Cat. 5 appear not as mere
distortions of Aristotles words but as intelligible and philosophically
interesting possibilities of developing his ideas in different directions. By
now it should also be clear how important it is to consider the place and
role that each thinker gave to the Categories in order to assess his own
interpretation of the philosophical arguments contained in that work.
Conclusion
In this paper, I have been considering the way Aristotles Categories
was interpreted as the rst work of the Organon until the end of the
fourteenth century. A fact has clearly emerged: the longevity of the
substantiam secundam quam aliqui vocant terminum communem substantialem, sed
nos dicimus eam formam communem, dicens substantiae secundae dicuntur species
quibus substantiae primae insunt tamquam partes subiectivae aut per se inferiora, ut
aliquis homo est in specie hominis et genus hominis est animal. Hae ergo species et
talium specierum genera, cuius modi sunt forma communis, quae est humanitas, quam
Porphyrius vocat hominem communem, et forma generalis, quae est animalitas, sunt
substantiae secundae. Sicut enim omnes homines in hoc conveniunt quod sunt homines,
sic est eadem natura humana, vel eadem species humana, aut eadem humanitas in
communiquod idem estin omnibus illis; et illa vocatur substantia secunda. Non
enim est talis convenientia specica fundata solum in termino communi vel in eius
praedicabilitate, cum nullo existente tali termino non eo minus forte talis convenientia
in natura humana; non ergo species humana est terminus vel intentio.
180 giorgio pini
idea that the Categories is devoted to the study of simple items and that
as such it is the rst part of a sequence whose other elements are the
De interpretatione (devoted to the study of propositions made of terms)
and the Prior Analytics (devoted to the study of syllogisms made out of
propositions). This conception was completely extraneous to Aristotles
original intentions. Its origins are to be traced to the Neoplatonic circles
between the third and fth century A.D. or perhaps even earlier to a
Stoic inuence on Hellenistic Aristotelianism. It is remarkable how much
this idea inuenced the interpretation of Aristotles Categories both in
Antiquity and in the Middle Ages. The conception of the nature of logic
changed but the idea that the Categories is a logical treatise devoted to the
study of simple items survived all these changes and was adapted each
time to a new conception of logic. So when logic was seen as the study
of arguments expressed in linguistic forms, the Categories was interpreted
as the study of simple linguistic terms aimed at the study of syllogism.
When logic was seen as the study of the relationships holding among
concepts (i.e., the so-called second intentions), the Categories was read as
a treatise devoted to simple concepts out of which mental propositions
and mental arguments are made. The conception of the Categories as
the rst treatise of the Organon devoted to simple items even survived
the contraposition between nominalists and realists in the fourteenth
century: nominalists thought that the simple items were mental terms,
realists thought that they were simple extramental things.
As a consequence, the Categories uninterruptedly occupied the open-
ing position in Aristotles corpus. Since this editorial ordering reected
a didactic scheme, the Categories, together with its companion work,
Porphyry s Introduction, was also the rst treatise that students of philoso-
phy read for several centuries. Accordingly, its inuence on the molding
of the Western philosophical worldview was overwhelming. This inu-
ence did not cease after the fourteenth century. As a matter of fact, it
extended well into the modern era. The idea that logic is divided into a
part devoted to simple terms, another one devoted to propositions and
a third one devoted to arguments, which lies behind works so diverse as
Peter of Spain s Summulae logicales
55
and Ockham s Summa logicae, survived
even the sinking of Aristotelian logic, although that idea was originally
55
On the structure of Peter of Spain s Summulae logicales and specically on the
presence of a section devoted to Categories, see L. M. de Rijk , Introduction, in Peter
of Spain, Tractatus called afterwards Summule logicales, ed. L. M. de Rijk (Assen, 1972),
p. XCII.
devised to give a systematic form to the Aristotelian corpus. The same
schema is still clearly detectable in the layout of the rst part of John
Stuart Mill s System of logic, published in 1843.
56
To summarize, I tried
to show that the idea that the Categories should be read as a work of
logic and as the rst part of a longer sequence according to which logic
was organized had important consequences even from a purely philo-
sophical point of view, as the example of the different interpretations
of the distinction between primary and secondary substances between
the thirteenth and fourteenth century should have made clear.
56
John Stuart Mill , A System of Logic: Ratiocinative and Inductive, eds. J. M. Robson
R. F. MacRae in Collected Works VIIVIII (Toronto, 1973).
SIMON OF FAVERSHAM ON ARISTOTLES CATEGORIES AND
THE SCIENTIA PRAEDICAMENTORUM
Martin Pickav
I.
The question of whether Aristotles short treatise on the so-called
categories is about words or concepts or thingsor in other words,
whether the Categories belongs to logic (or even grammar) rather than
to metaphysicsis a notorious one in the history of philosophy. Later
medieval authors inherit the relevant debate from the ancient com-
mentators on Aristotles Categories, but their answers often differ from
that of their predecessors in at least two fundamental respects: (1) Later
medieval philosophers (from the 13th century onwards) are typically
more inclined towards a realist understanding of the categories; for the
most part they accept the idea of ten kinds of being. And although
not all of them have a realist understanding of the subject-matter of
the Categories itself, they regard the treatise as a work in which Aristotle
addresses these ten kinds of being from a logical point of view.
1
(2) Later
medieval authors tend to broaden the problem of the scientic status
of the Categories insofar as they not only discuss whether the Categories
belongs to this or that area of knowledge, but whether the doctrine
described in this work can count as knowledge at all.
2
1
For a brief survey of the different ways of interpreting the Categories cf. Giorgio
Pini , Categories and Logic in Duns Scotus : an Interpretation of Aristotles Categories in the Late
Thirteenth Century (Leiden, 2002), pp. 613; Sten Ebbesen , Les Catgories au moyen ge
et au dbut de la modernit, in Otto Bruun/Lorenzo Corti, eds. Les Catgories et leur
histoire (Paris, 2005), pp. 245274, esp. pp. 257sqq.
2
Many later medieval commentaries on the Categories start with questions like Utrum
de praedicamentis possit esse scientia, utrum scientia praedicamentorum sit una scientia, etc. Cf. for
instance, Peter of Auvergne , Quaestiones super Praedicamentis, ed. Robert Andrews , Cahiers
de lInstitut du moyen-ge grec et latin [= CIMAGL] 55 (1987), 384, qq. 13; Anonymus
Matritensis, Quaestiones super librum Praedicamentorum, ed. Robert Andrews, CIMAGL 56
(1988), 117192, qq. 13; Thomas Sutton , Expositio super Praedicamentis, ed. Alessandro
D. Conti , in Paul Osmund Lewry , ed. The Rise of British Logic (Toronto, 1985), pp.
173213, prologus. Their late ancient predecessors, for example Simplicius , do of
course undertake with great pains to determine the skopos of the Categories, but they
184 martin pickav
Simon of Faversham (before 12601306), one of the most prolic
English commentators on Aristotle at the end of the 13th century,
is a good example of this combination of a realist approach and of
an approach that focuses on the status of our knowledge of the cat-
egories. His conception of the categories will be at the focus of this
paper because it helps to understand to what extent, according to some
medieval philosophers, the categories are located right at the intersec-
tion between logic and metaphysics and why the question regarding
the proper scientic place of the categories (i.e., whether they properly
belong to logic or to metaphysics) is not just a question which can simply
be answered in one or the other direction. I chose Simon as the primary
object of my paper because his commentary on the Categories and his
other related texts are original in the way in which they synthesize dif-
ferent inuences and because they provide an interesting example of
a doctrine of the categories from around 1280, i.e., the years before
Scotus s masterly commentary and before the appearance of more
nominalist accounts of the categories.
The paper will begin with a discussion of Simon s general under-
standing of logic and its scientic status. We shall then move on to
Simons treatment of the scientic character of the doctrine developed
in Aristotles Categories. Like medieval commentators on Aristotles
Metaphysics, he asks in particular how scientic knowledge (scientia) of
the categories (or predicaments, to use the medieval term) is able to
match the general requirements for knowledge set out in the Posterior
Analytics: What is the subject-matter of such a piece of knowledge and
what are the subjects parts, properties and causes? Simons answers
will provide us with a better understanding of what he means by
category or predicament ( praedicamentum). We will then focus on a
related problem, namely, Simons contention that predicaments are
composed of a res praedicamenti and a ratio praedicamenti. How are we to
understand this theory and what role does it play? The nal section
of this article is devoted to answer the central question regarding the
relationship between logic and metaphysics as far as the categories are
concerned.
One of my aims in this article is thus expository. Somewhat undeserv-
edly Simon s teaching on the categories is almost unknown to modern
do not discuss how the Categories relates to the account of knowledge provided in the
Posterior Analytics.
simon of faversham on the CATEGORIES 185
historians of philosophy. I hope to show that his Categories commentary,
like many other commentaries of the same period, is philosophically
intriguing and worth studying.
3
Unfortunately, though, my paper has
to be limited to those commentaries that have been edited up to now.
I am convinced that a study of the many commentaries that are yet
unavailable in print would conrm the philosophical importance of
this genre of texts.
4
My other aim is explanatory. I want to examine
what kind of knowledge we can expect, according to some 13th-century
philosophers, when we pursue knowledge of the categories, the most
general kinds of things that surround us in life. This examination will
help to clarify the relationship between logic and metaphysics, between
two disciplines which deal with the same subject-matter, namely, every-
thing there is.
II.
There can hardly be any doubt that Simon of Faversham considers
Aristotles Categories as a work of logic. Logic is a rational science
(rationalis scientia) not just because every disposition of the rational soul
is by denition rational or because logic proceeds (like any other sort
of knowledge) according to reason, butand here Simon follows a
long traditionbecause logic has to do with the different operations
of reason (ratio). The rational powers of human beings are character-
ized by discursivity, as everyone can experience. We rarely experience
moments in which we fully grasp an item of knowledge or belief
intuitively, and although this might happen in the case of some items
of knowledge (e.g., analytical truths), we normally reason by moving
from one bit of knowledge or belief to another. However, there are,
so Simon contends, different operations according to which reason
3
For an overview of some of these commentaries cf. Robert Andrews , Question
Commentaries on the Categories in the Thirteenth Century, Medioevo 26 (2001), 265326.
Simon s teaching on the categories gets incidentally discussed in general articles on the
categories in the middle ages and, most importantly, in a series of works by Giorgio Pini
(on John Duns Scotus ) but to date there is no study that is mainly devoted to Simon.
The most relevant studies by Pini are: Categories and Logic in Duns Scotus, and Scotuss
Realist Conception of the Categories: His Legacy to Late Medieval Debates, Vivarium
43 (2005), 63110. The interpretation proposed in this article is very much indebted
to Pinis pioneering work.
4
Some information about those commentaries that are yet unedited can be gathered
from Andrews , Question Commentaries on the Categories.
186 martin pickav
proceeds . . . in moving from one thing to another.
5
Each of the vari-
ous acts of reason is treated in one corresponding part of logic, a fact
that also explains why Aristotle left us a whole series of logical works:
works on rhetorical reasoning (Rhetoric) as well as on poetical (Poetics),
dialectical (Topics), sophistical (Sophistical Refutations), demonstrative rea-
soning (Posterior Analytics) and on reasoning tout court (Prior Analytics).
6
But
as already granted there are also other acts of reason in which we are
not (or less) discursively moving from one thing to another. There is
one operation of reason by which reason grasps the simple quiddity of
a given thing or object and another by which reason combines simple
quiddities in an afrmative or negative proposition. Aristotles Categories
and his treatise On Interpretation are about these two remaining operations,
which are the basis of all the other discursive rational acts.
7
If logic has to do with acts of reason, then it can hardly count as
real knowledge in the strict sense. Consequently, medieval philosophers
distinguish between scientiae reales and scientiae rationales. The natural sci-
5
Simon of Faversham, Quaestiones super libro Porphyrii, prol. in Magistri Simonis Anglici
sive de Faverisham Opera omnia, vol. 1: Opera logica, ed. Pasquale Mazzarella (Padova,
1957), p. 14: Cum autem logica sit rationalis sciencia, quia est de ipso actu rationis
secundum quod ratio diversimode procedit ab uno discurrendo in aliud. Cf. also ibid.,
q. 2, p. 19: Unde non dicitur esse rationalis scientia quia per rationem procedat,
quia sic omnis scientia esset rationalis; sed dicitur rationalis, que est de hiis que sunt
causata ab intellectu; Quaestiones veteres super libro Elenchorum, prol., edd. Sten Ebbesen
et al. (Toronto, 1984), p. 28: Unde intelligendum primo quod tota logica est de actu
rationis . . . Secundum igitur diversitatem actuum rationis accipiendae sunt diversae
partes logicae . . .; cf. ibid., q. 2, p. 32. Please note that throughout this article I do
not necessarily respect the punctuation of the editions I use. Cf. Thomas Aquinas ,
Expositio libri Posteriorum, book I, lect. 1, ed. Leonina, vol. I* 2 (Rome, 1989), p. 3sq:
Et hec ars est logica, id est rationalis, sciencia. Que non solum rationalis est ex hoc
quod est secundum rationem, quod est omnibus artibus commune, set ex hoc quod
est circa ipsum actum rationis sicut circa propriam materiam . . . Oportet igitur logice
partes accipere secundum diuersitatem actuum rationis. For Aquinass conception of
logic cf. Robert W. Schmidt , The Domain of Logic According to Saint Thomas Aquinas (The
Hague, 1966), esp. pp. 49sqq.; for Simon of Faversham, cf. also Costantino Marmo ,
Suspicio: A Key Word to the Signicance of Aristotles Rhetoric in Thirteenth Century
Scholasticism, CIMAGL 60 (1990), 145198, esp. pp. 170175.
6
Simon of Faversham, Quaest. super lib. Porph., prol., p. 14.
7
Cf. Quaest. super lib. Porph., prol., p. 14sq.: Quando autem ratio non procedit discur-
rendo ab uno in aliud, sed absolute accipit simplicem quidditatem rei, considerando
ipsam ut est ordinabilis in genere secundum sub et supra, et de tali actu rationis est liber
Predicamentorum, cuius subiectum est ens dicibile incomplexum ordinabile in genere.
Quandoque autem ratio simplicia comprehensa ad invicem componit vel dividit, et de
tali actu rationis est liber peryhermeneias, cuius subiectum est enunciacio afrmativa
vel negativa. Cf. id., Quaest. vet. super lib. Elench., prol., p. 28.
ences and metaphysics are examples of the rst group since they deal
with real extra-mental objects; rational sciences only deal with some-
thing in the mind. Simon adopts this distinction,
8
yet he also seems to
radicalize it. For there is a sense in which logic is not real knowledge
because it is not really knowledge at all! Knowledge (scientia) is a type
of intellectual cognition
9
and as such it must be related to the object
of the intellect. Yet the primary objects of the human intellect, Simon
explains, are things that really exist outside the soul. Nothing inside
the intellect can therefore be the object of the intellect. According to
Simon (who quotes Aristotle in support), acts of the intellect are only
perceived insofar as something external is perceived. There is no self-
cognition prior to the cognition of some external object. Logic cannot,
therefore, be knowledge because its objects do not straightforwardly
fall under the object of the intellect. So Simon concludes, logic can
only be called knowledge (scientia) if we allow for a less rigid use of the
label knowledge.
10
At this point, however, someone might raise an obvious objection:
Even if acts of the intellect exist in the intellect they are nevertheless
real things in a strict sense because they are real qualities inhering in
the soul. There seems to be no reason for regarding mental states as
having a diminished degree of being. Moreover, my intellectual acts
may not be something outside of my soul, yet other persons acts are
(outside of my soul). From this perspective it looks as if there could,
at least in principle, be knowledge in the proper sense of the objects
of logic. Since even logic seems to be about things that fall under the
object of the intellect. Although Simon himself does not address this
objection, there can be no doubt about his reply: Intellectual acts or
8
Cf. Quaest. vet. super lib. Elench., q. 2, p. 32.
9
Cf. Quaest. super lib. Porph., q. 2, p. 19: Sciencia quedam cognicio intellectualis
est; Quaestiones super libro Praedicamentorum, q. 1, ed. Pasquale Mazzarella in Magistri
Simonis Anglici . . . Opera omnia I, p. 72.
10
Quaest. vet. super lib. Elench., q. 2, p. 31: Scientia enim proprie est de hiis quae
possunt esse obiectum intellectus; talia autem sunt res verae existentes extra animam.
Cum vero intellectus intelligat alia <et> intelligendo alia intelligit se, ut potest haberi
ex III De Anima [III 4, 429b9], nihil quod est in intellectu poterit esse obiectum intel-
lectus. Cum igitur intentiones non sunt vera res, non poterunt esse proprium obiectum
intellectus. Et ideo sophistica . . . est modus sciendi et non scientia . . . Extendendo tamen
nomen scientiae, scientia est. Although Simon develops these ideas in answering the
question of whether the discipline described in the Sophistical Refutations can be a sci-
ence, it is obvious that these thoughts apply to logic as a whole.
188 martin pickav
other states of the mind are indeed real existing qualities of the mind.
Yet, insofar as they are representative, i.e., insofar as they are intentional
states of the mind, they are mind- or soul-dependent in a particular
way and not just insofar as they inhere in the cognizing faculty as in
a subject. In this sense intellectual acts and their content are not just
simple extra-mental objects like this lamp in front of me, this tree out-
side my window, or this color of the wall in my room. A characteristic
feature of cognitive acts is that they are intentional; they produce or
express what Simon and other medieval thinkers call intentions (inten-
tiones) and these intentions are different in kind from any other form of
extra-mental existence. They cannot simply be reduced to extra-mental
objects.
11
Moreover, that acts of reason are linked to the production
of intentions explains why Simon sees no difculty in saying both that
logic has to do with the operations of reason and that it deals with
intentions. Both descriptions mean the same insofar as every act of
reason involves the production of intentions.
But if logic is just a scientia rationalis and not about objects of our
extra-mental world, does this not mean that logic is just about inventions
of the mind? Is logic ction and completely arbitrary? Of course Simon
denies this conclusion. The human intellect is fundamentally receptive,
i.e., it is normally determined by its objects. The acts of the intellect
must therefore be taken as to follow the objects and to be grounded in
them. When logic studies the acts of the intellect it considers products
of the mind that are nevertheless not wholly independent of the extra-
mental world (even though the intentions that are produced in the mind
are completely external to the objects themselves). Simon even goes so
far as to say that the intentions the logician examines are attributed by
the intellect to its objects because of certain properties of these objects.
In Simons understanding, these intentions are some sort of second
order properties of the extra-mental objects themselves. The whole
of logic is thus taken from the properties of the objects.
12
11
For Simon s denition of intention or more precisely rst and second inten-
tions cf. his Notulae super Summulas Petri Hispani, ed. Lambert Marie de Rijk , Vivarium
6 (1968), 69101, esp. p. 94.
12
Quaest. super lib. Porph., q. 1, p. 17: Logica considerat intenciones, quas intel-
lectus fundat in rebus et <que> sunt extranee rei; ibid., q. 2, p. 19: Cum autem
intellectus causat tales intentiones, et movetur ab apparentibus in re; et propter hoc
intellectus diversas intenciones logicales attribuit diversis rebus propter diversas propri-
etates. Unde logicus non diceret hanc esse veram, Homo est genus, sed hanc, Homo
est species. Ideo tota logica accipitur a proprietatibus rerum, quia aliter logica esset
But despite these last claims about logics foundation in extra-mental
objects, logic does not deal with these objects in any direct way. The
logician, Simon emphasizes, does not consider the nature of things but
only intentions. Or if the logician were to consider extra-mental things,
then [he does so] only insofar as they are related to intentions.
13
This
seems to settle the question of the scientic status of the categories:
as a part of logic, namely, that part of logic that is related to the rst
act of the intellect, the kind of knowledge described in the Categories
considers primarily intentions. If the Categories contains a metaphysical
doctrine, then it does so only incidentally.
III.
In the rst question of his commentary on the Categories, Simon ex-
plicitly addresses the problem of whether the doctrine laid out in the
Categories qualies as proper knowledge (scientia). From the point of view
of his general ideas about logic we expect him to give a negative or,
at least, a qualied answer. Moreover, to know something means to
have a clear grasp of it and a grasp of why it is this or that. Proper
knowledge entails knowledge of causes. Someone who knows that x is
y can say why x is y, otherwise we can, strictly speaking, only attribute
to her the belief that x is y. But what could count as the causes of the
categories or predicaments? This line of reasoning leads, for instance,
Peter of Auvergne to conclude that there cannot bein the strict
senseany knowledge of the categories. When we speak of a scientia
praedicamentorum we use scientia merely as an expression signifying a
cognition that is certain (certa cognitio), but not as referring to (proper)
knowledge through causes.
14
gmentum intellectus, quod non dicimus. For a more detailed discussion of Simon s
views about intentions cf. Pini , Categories and Logic in Duns Scotus , pp. 7280.
13
Ibid., q. 2, p. 19: Logicus secundum quod logicus non considerat de naturis
rerum, sed intenciones solas; vel si res consideret, hoc solum est ut sub intencionibus
sunt. Quaest. super lib. Praed., q. 23, p. 98: . . . cum logicus non consideret essentias
rerum . . .
14
Peter of Auvergne , Quaestiones super Praedicamentis, qq. 1+2, p. 10: Nihilominus
tamen potest esse scientia de aliquo, quae idem est quod certa cognitio ipsius, et haec
datur per denitiones et descriptiones; et hoc modo potest esse scientia de praedica-
mentis; cognoscuntur enim per quasdam denitiones et descriptiones. Et cum non
habeatur de praedicamentis scientia per causam, cum ipsa non habeant causam, tamen
isto modo per denitiones et descriptiones potest de hiis esse scientia.
190 martin pickav
However, Simon s evaluation of the scientic status of such a scientia
is quite positive: He unambiguously afrms that there can be proper
knowledge of the predicaments (scientia de praedicamentis). Simon lays out
three criteria any piece of cognition has to fulll in order to count as
proper knowledge (scientia) and he then goes on to show that knowledge
of the predicaments satises all of them. Simon does not inform his
reader of whether he thinks this list of criteria is exhaustive, but there
seems to be a progress from more general criteria to more specic ones.
The rst criterion for knowledge proper demands that (1) knowledge has
to be about something that exists, i.e., something that is a being (ens). We
cannot positively know anything of what is non-existent. Knowledge,
he explains, can only be had of something we grasp intellectually.
Yet, if there is nothing to be grasped, there can be no knowledge.
15

(2) Knowledge is necessarily of something intelligible. This criterion
is derived from the fact that knowledge is an intellectual habit and a
kind of perfection of the intellect. But only what is accessible to the
intellect can be involved in causing such a habit or perfection.
16
(3) The
object of knowledge has to have parts and properties. Properties are
what we know of the object or what we pursue or try to demonstrate
when we acquire knowledge of a given object. Arithmetical knowledge,
for instance, is about the properties and features of numbers. Without
properties there would simply be nothing left to know. It is less obvious
why the object of knowledge is supposed to have parts. Simon explains
this requirement by pointing to denitions. The pursuit of knowledge
aims at denitions. Only if I really know what x is, am I able to give
a denition of x; and only if I am able to give a denition of x do I
really know what x is. A denition, however, is composed out of parts,
for instance, a genus and a difference. But it is not instantly clear why
it follows that the objects of knowledge have to have parts too. Simon
provides his readers with no further argument for this conclusion. He
15
Simon of Faversham, Quaest. super lib. Praed., q. 1, p. 71sq.: Ad hoc enim quod
de aliquo possit esse scientia tria requiruntur, scilicet quod ipsum sit ens et quod ipsum
sit intelligibile et quod habet partes constituentes ipsum, proprietates et passiones que
possunt probari de hoc. Quod oporteat ipsum esse ens apparet, quia quod non est non
contingit scire; et ratio huius est quia nihil scitur nisi quod intelligitur; omne autem
quod intelligitur est ens; et ideo quod quid scitur necesse est esse ens; et ideo bene
dixit Philosophus quod non est non contingit scire.
16
Ibid., p. 72: Item quod oporteat illud de quo est scientia esse intelligibile,
patet quia scientia habitus quidam intellectualis est; est enim perfectio intellectus;
et ideo scibile oportet esse intelligibile; et ideo illud de quo est scientia oportet esse
intelligibile.
might have had in mind that a satisfactory denition requires that the
denition corresponds to the thing to be dened. Thus, the parts men-
tioned in the denition have to be real or at least conceptual parts (partes
rationis) of the thing to be dened.
17
Yet, this seems to contradict what
he says on other occasions where he explicitly holds that a denition
is not an aggregation of things but merely of concepts.
18
Whatever we make of this last problem, it is obvious for Simon that
the scientia de praedicamentis ts all three criteria. Such a knowledge is (1)
about something that exists or, to be more precise, something that exists
outside the mind. According to Simon, the ten categories or predica-
ments really exist outside the mind; they do not just exist because of
the mind.
19
Naturally, this kind of knowledge is (2) also about something
intelligible. For the intellect assigns genera and species to all the ten
predicaments. That is, there are genera and species of substances (e.g.,
animal, human being), qualities (e.g., color, whiteness) and of all the
remaining predicaments. But the intellect can only assign intentions
such as genus and species to the predicaments if the predicaments
themselves are accessible for it. Or in Simons more technical language:
everything in which the intellect can ground the intentions genus and
species must be intelligible to the intellect.
20
Last but not least, (3) the
17
Ibid.: Item oportet quod tale habeat partes constitutentes ipsum. Illud enim
quod scitur oportet quod sciatur per diffinitionem; diffinitio autem est sermo habens
partes . . .; et ideo illud quod scitur oportet quod habeat partes rei vel partes rationis
(quod dico propter intelligentiam). Item oportet quod illud de quo debet esse scientia
habeat proprietates et passiones demonstrabiles de ipso, si ipsum per demonstrationem
sciatur. In stipulating that the object of knowledge has to have properties and parts
Simon is basically following Aristotle. Cf. Analytica Posteriora I 10 (76b1116) and I 28
(87a3839). For the medieval understanding of these passages cf. Albert Zimmermann ,
Ontologie oder Metaphysik? Diskussionen ber den Gegenstand der Metaphysik im 13. und 14.
Jahrhundert: Texte und Untersuchungen, 2nd ed. (Leuven, 1998), pp. 130sqq.
18
Quaest. super lib. Porph., q. 21, p. 41: . . . in difnicione non est agregacio plurium
rerum, sed plurium intellectuum . . .
19
Quaest. super lib. Praed., q. 1, p. 72: Nunc autem quodlibet predicamentum est ens,
non quidem ens in anima, sed verum ens extra animam. Dicit enim Philosophus quinto
[sic] Metaphysice [VI 2, 1026a33] quod ens verum extra animam est quod dividitur in
decem guras predicamentorum. Voco autem illud ens verum extra animam, cuius esse
non ab anima dependet; illa autem voco encia per animam, quorum esse ab anima
dependet, cuiusmodi sunt intentiones logice, sicut genus, species et talia.
20
Ibid.: Item quodlibet eorum quod ponitur in predicamento est aliquid intelligibile;
et huius probatio est quia genera et species solum ponuntur in predicamento; illa ergo
que ponuntur in predicamento oportet esse talia quod supra ipsa intellectus fundet
intencionem generis et speciei. Ex hoc arguo: omne illud supra quod intellectus fundat
intencionem generis et speciei est intelligibile ab intellectu; intellectus enim in nulla re
fundat intencionem generis et speciei nisi cum est in actu; sed singulum eorum que
192 martin pickav
predicaments have properties and parts. Simon does not provide his
reader with examples of such properties, he just broadly refers to the
examples Aristotle gives in the Categories. There, Aristotle mentions, for
instance, the property not to have contraries and the property not to
be in a subject as characteristic properties of substances.
21
Finally pre-
dicaments have parts for they are composed, according to Simon, of two
elements: a predicamental thing (res predicamenti ) and a mode of being
(to which he also refers to as ratio predicamenti ). For example, an accident
is a thing, to which the mode of being to be in something else (as a
subject) is added, whereas a substance is a thing with the superadded
mode of being to be not in something else (as a subject).
22
Before we have a closer look at Simon s three criteria for proper
knowledge, we should rst try to get a clear idea of what Simon means
with the term predicament. This is of course crucial for our under-
standing of what a scientia de praedicamentis really is about. Praedicamentum,
however, is a notoriously ambiguous term.
23
In his description of the
rst criterion, Simon insists that every predicament is a being, not
however a being in the soul, but a real being outside the soul.
24
One
way to understand this statement is to take him as speaking about the
members or tokens of classes of beings. These singular things exist
indeed outside the soul. But we can also understand Simon here as
making a claim about the classes, or better, types or modes of beings.
According to this sense he would emphasize that each of these types
or modes of beings is real, they are not just an invention of the soul
or the mind. It is obvious that Simons remark cannot be about classes
(genera) of predicates (or predications). Predicates or predications always
presuppose an activity of the mind and the properties of praedicamenta
that, according to Simon, are known in a scientia de praedicamentis (e.g.,
ponuntur in predicamento est tale quod supra ipsum intellectus fundat intencionem
generis et speciei.
21
Ibid., p. 72sq.: Quod habeant proprietates manifestum est ex processu Philosophi.
Dat enim proprietates substantie, quantitatis et aliorum. For properties of substances,
for instance, cf. Categoriae 5 (3a7sqq.). Cf. also Simon s discussion of the properties of
substance in Quaest. super lib. Praed., q. 10, p. 79sq.
22
Ibid., p. 73: Iterum quodlibet predicamentum habet partes constituentes ipsum:
quodlibet predicamentum constituitur ex duobus, scilicet ex re et ex modo essendi
sibi superaddito. Unde res cui competit talis modus essendi, qui est esse non in alio,
est substantia; res autem cui competit iste modus essendi, qui est esse in alio, est
accidens . . . Sic igitur de predicamentis potest esse scientia.
23
Cf. the illuminating remarks in Pini , Categories and Logic in Duns Scotus , pp. 1416.
24
Cf. supra fn. 19.
not to be in a subject or not to have contraries) are not properties
of classes of predicates (or predications). The scientia de praedicamentis
cannot, therefore, be about concepts or other items that require a
productive role of the mind. However, the description of the second
criterion makes it clear, I think, that what Simon has primarily in mind
when he talks about praedicamenta are types or modes of beings (and
not particular categorical beings). For what sense could it make to say
about the tokens of a class of beings that genera and species are only
posited in each of those?
25
But it makes perfect sense to understand him
stating that genera and species are only posited in the ten different
types or modes of beings, i.e., that they apply only to them.
But if the focal meaning of praedicamentum is something like type
of beings, why are these types of beings referred to by an expression
that is so obviously derived from the purely mental act of predicat-
ing? Simon s answer is very sketchy but he seems to subscribe to the
one given by Thomas Aquinas : He assumes that there is an intimate
relationship between the modes of being (modi essendi ), the modes of
predicating (modi praedicandi), and the modes of signifying (modi signi-
candi ). Language mirrors reality, so there should be no surprise that the
ten fundamental modes of being are reected in ten distinct modes in
which x can be said to be y. Because of this relationship and because
of the fact that our encounter with reality and its foundations happens
through language, the fundamental types of being are rst known to us
by means of knowing the modes of predication. This is the reason why
we refer to these ten types of being as the ten predicaments, although
the different ways of predication are strictly speaking derived from and
ontologically later than the modes of being.
26
25
Cf. supra fn. 20.
26
Cf. Simon of Faversham, Quaest. super lib. Praed., q. 12, p. 85: Predicamenta
enim distinguuntur penes modos essendi, quia distinguuntur penes modos predicandi;
propter hoc enim distinguitur substantia ab aliis. Sed modi predicandi sumuntur a
modis essendi sicut modi signicandi; et propter hoc predicamenta distinguuntur
penes modos essendi et non penes quoscumque modos essendi, sed penes tales modos
essendi qui in nullo communicant. Cf. Thomas Aquinas , In V Metaph., lect. 9, edd.
M.-R. Cathala/R. M. Spiazzi (Turin, 1950), p. 238, n. 890: Unde oportet quod ens
contrahatur ad diversa genera secundum diversum modum praedicandi, qui consequitur
diversum modum essendi; quia quoties ens dicitur, idest quot modis aliquid praedicatur,
toties esse signicatur, idest tot modis signicatur aliquid esse. Et propter hoc ea in
quae dividitur ens primo dicuntur esse praedicamenta, quia distinguuntur secundum
diversum modum praedicandi.
194 martin pickav
This last remark also leads us to the solution of another puzzle. As
we have just seen, Simon considers the scientia de praedicamentis as real
knowledge because it has to do with something extra-mental. But if
predicaments are types of beings, then what keeps us from saying that
they are classes or genera of being? Indeed, Simon constantly follows
the traditional usage and refers to the different categories as (highest)
genera.
27
Now, Simon also holds that it is the intellect which causes
universality and that in order to avoid Plato s mistake we are bound
to say that nothing universal exists outside the soul.
28
But would positing
real genera of being outside the soul not contradict his conviction that
nothing that is properly universal exists outside the soul? The reply to
this puzzle is clear from Simons distinction between two kinds of mean-
ings of the term universal. According to the rst meaning, universal
stands for the intentio universalitatis, i.e., that which makes something
universal; according to the second meaning, universal signies that
very thing which underlies the intentio universalitatis and which is made
universal by it.
29
Insofar as the intentio universalitatis is something that is
produced by the mind alone, it cannot exist outside the soul. A universal
concept dog, a concept of a species, includes an intentio that makes it
into a species. The concept dog thus does not exist outside the soul.
But for a realist like Simon there is of course something extra-mental
that corresponds to such a (true) concept, i.e., something like dogness.
This is the basis for the intellects forming a universal concept.
30
In itself
something like dogness, a real property all dogs in the world share, is
not a universal. As something on which a universal is founded it might,
however, be called a universal too. I take it that Simon has exactly
this in mind when he calls substance and other predicaments genera.
27
Cf. e.g., Simon of Faversham, Quaest. super lib. Praed., q. 13, p. 86: Intelligendum
est hic quod substantia est genus primum et est genus unum . . .; ibid., q. 33, p. 113:
. . . quantitas est genus generalissimum et est genus unum . . .
28
Cf. Quaest. super lib. Porph., q. 5, p. 26sq.: Unde non est ponere hominem univer-
salem quia universaliter existat, sed est universale quia universaliter intelligitur circum-
scribendo omnes condiciones individuantes. Ideo dicit Philosophus quod homo et asinus
et que universaliter in singularibus dicuntur non sint encia, sed sint intellecta, quia que
sunt encia extra, particularia sunt, unde universalia sunt alia, quia sunt intellecta.
29
Cf. ibid., q. 2, p. 18sq.: Dicendum est quod universale potest nominari duo,
scilicet intencionem universalitatis et rem subiectam intencioni. For Simon s account
of universals cf. also ibid., qq. 3, 12, and 20. Cf. also Pini , Categories and Logic in Duns
Scotus , pp. 75sqq.
30
Cf. ibid., 19: Nunc autem res subiecta intencioni universalitatis est per se intel-
ligibilis ab intellectu, aliter enim intellectus non fundaret in ea intencionem univer-
salitatis.
Furthermore, when predicaments are referred to as highest genera this
simply means that they are irreducible types of being.
31
However, at this point we are left with yet another puzzle. If for
Simon logic is a scientia rationalis but the scientia de praedicamentis is knowl-
edge in the proper sense, then we seem to be forced to conclude that
knowledge of the categories or predicaments is not a part of logic!
Applied to Simons understanding of the works of Aristotles Organon
this amounts to saying that the rst work of the Organon is not a work of
logic but of metaphysics. But how then can the different works making
up the Organon represent the different operations of reason?
IV.
Simon is not the rst who came forward with the idea of settling the
question of whether there can be a scientia de predicamentis by putting
forward a list of criteria. The anonymous author of a Categories com-
mentary in ms. Madrid BN 1565 [= Anonymous of Madrid ] offers
a list that is strikingly similar. He too mentions intelligibility of the
object of knowledge and the requirement that the object has to have
properties. But the Anonymous of Madrid adds as a third condition
that there have to be causes by means of which the properties can be
demonstrated to belong to the object.
32
This last condition is obviously a reference to the kind of objection
against the scientic status of the scientia de predicamentis that we found
expressed by Peter of Auvergne . In light of the Anonymous position,
however, Peter must be regarded as overshooting the mark when he
denies that there can be proper knowledge of the predicaments because
there are no causes at hand through which we know.
33
The Anonymous
answers by putting forward a couple of clarications. (1) In talking
about causes we have to be more precise. What knowledge requires
are strictly speaking causes by means of which the properties of the
31
For the aspect of irreducibility cf. the quotation supra in fn. 27.
32
Anonymus Matritensis, Quaest. super lib. Praed., q. 1 , p. 122: Sciendum quod tria
requiruntur ad hoc quod de aliquo sit scientia, et si illa tria habuerit, de illo potest
esse scientia. Primum est quod sit intelligibile. Secundum est quod proprias habeat
proprietates per se inhaerentes illi, et non per accidens . . . Tertium est quod habeat
causas per quas illae proprietates probantur de illo.
33
Cf. supra fn. 14.
196 martin pickav
object of knowledge can be demonstrated to inhere in the object.
34

Other causes are irrelevant. But why shouldnt it be possible that there
are causes which we use in demonstration? Denitions, for example,
can play this role. And by means of them we can demonstrate that
predicaments have certain properties.
35
(2) But even if there were no
robust causes, i.e., real causes of the object of knowledge itself: it is
simply not true that those are required for knowledge. The Anonymous
reminds his readers that there are two kinds of knowledge: propter quid
knowledge and quia knowledge. Only the former is knowledge through
real causes. The latter is the kind of knowledge that we have of some-
thing through its effects.
36
Knowledge that God exists, to use a famous
medieval example, is a case of quia knowledge. Since God himself has
no causes there can be no proper causal explanation of his existence
and therefore no propter quid knowledge. But it can be demonstrated
from Gods effects that there has to be a rst cause etc. By this line
of reasoning we only understand that (quia) God exists and not why
(propter quid) he exists. (3) There is a point of view from which Peter of
Auvergnes objection is simply wrong. The predicaments have causes
or have at least one cause, namely, God who is the efcient cause of
there being something like predicaments. In this sense we might con-
sider propter quid knowledge of the predicaments as possible.
37
But the
Anonymous admits that God is only prior to the predicaments insofar
as God is outside the line of predication (linea praedicamentalis). First,
God is not a genus that can account for properties of things and spe-
cies that fall under it. Second, there is nothing in the line of predica-
tion that is prior to the predicaments. For the ten predicaments are
ten highest genera.
38
To sum up, according to the Anonymous, proper
34
Cf. supra fn. 32.
35
Anonymus Matritensis, Quaest. super lib. Praed., q. 1 , p. 122: Habent etiam causas
ut denitions per quas tales proprietates et passiones de ipsis probantur.
36
Ibid., p. 122sq.: Dicendum est ad hoc quod duplex est scientia, scilicet propter
quid et quia, ut habetur primo Posteriorum [I 13, 78a2223]. Scientia propter quid
procedit per causam; scientia quia per effectum. Et ideo absque hoc quod generalis-
simum non haberet causam, non sequeretur propter hoc quod de ipso non possit esse
scientia saltem quia, per effectum.
37
Ibid., p. 123: Ergo [lege etiam?] de praedicamentis potest esse scientia propter
quid, quia habent causam saltem effectivam, ut quoniam deus est causa efciens omnium
et non est in aliquo praedicamento . . . Et sic de ipsis poterit esse scientia propter quid.
This hypothetical idea is however already mentioned by Peter of Auvergne himself: cf.
Quaest. super praed., qq. 1+2 , p. 10.
38
Anonymus Matritensis, Quaest. super lib. Praed., q. 1 , p. 123: Modo quamvis [enim]
praedicamenta non habeant aliqua priora in linea praedicamentali, tamen habent
aliqua priora quae non sunt in linea praedicamentali.
knowledge of the categories is possible. There are causes; maybe not
causes that are absolutely prior to the object of knowledge, but causes
that are prior for us (i.e., for us rational beings) and known prior. This
is enough for the scientia predicamentorum to qualify as knowledge
39
and
not just as certain cognition.
If we compare the Anonymous of Madrid to Simon of Faversham
we are tempted to say that Simon deliberately drops the requirement
for causes from his list of criteria for knowledge. But Simon does not
ignore this requirement. In the rest of the question, in which he pro-
vides the list, he deals extensively with it and he gives a reply to Peters
objection that looks very similar to that of the Anonymous. Knowledge,
Simon begins, is a habit acquired through demonstration and since there
are two kinds of demonstration there are consequently two kinds of
knowledge. All demonstration proceeds from what is prior and what is
better known to what is posterior and less known; but since there are
things that are by nature prior and better known and things that are,
on the opposite, prior to us and better known for us, we get two kinds
of demonstrations, which Simon, with their traditional names, calls
demonstrations from causes and demonstrations from effects.
40
Thus far,
Simon is very much in accord with the Anonymous. He seems to agree
that there is proper knowledge, although not propter quid knowledge, of
the predicaments. But Simon also partially consents to Peters view. If
our knowledge of the predicaments is not by means of causes and thus
is not strict propter quid demonstration, what can serve as the basis of
such knowledge? Peter holds that we have such knowledge only on the
basis of denitions and descriptions.
41
On this point Simon seems
to agree with Peter. Although Simon doesnt explicitly conclude that
the two kinds of knowledge we can acquire following the two kinds of
demonstration are, on the one side, demonstrative and, on the other
side, denitive or descriptive, he comes very close to this when he
39
Cf. ibid.
40
Simon of Faversham, Quaest. super lib. Praed., q. 1, p. 73: Cum scientia sit habitus
ageneratus per demonstracionem, sicut est duplex demonstracio sic est duplex sciencia.
Quedam est demonstracio que procedit ex prioribus et nocioribus simpliciter, alia que
procedit ex prioribus et nocioribus quoad nos. Prima demonstratio t per causam,
secunda per essentiam. Cf. also Pini , Categories and Logic in Duns Scotus , pp. 164166.
41
Peter of Auvergne , Quaestiones super Praedicamentis, qq. 1+2 , p. 10: . . . et hoc modo
potest esse scientia de praedicamentis, cognoscuntur enim per quasdam denitiones
et descriptiones. Et cum non habeatur de praedicamentis scientia per causam, cum
ipsa non habeant causam, tamen isto modo per denitiones et descriptiones potest de
hiis esse scientia.
198 martin pickav
remarks that Aristotle here in his Categories does not proceed demon-
stratively but rather by means of explanations and examples.
42
Simon of Faversham and Peter of Auvergne are by no means the
only medieval authors who draw a distinction between knowledge
(scientia) that is demonstrative and knowledge that falls short of being
demonstrative and is therefore only descriptive or denitive. The
anonymous Metaphysics commentary in Escorial h.II.1, for instance,
makes a similar distinction with regard to the scientic status of meta-
physical knowledge. There the commentator, sometimes believed to be
Henry of Ghent , has particular problems with Aristotles claim accord-
ing to which every knowledge is about the properties and parts of its
proper object. But what is a property of being, the subject-matter of
metaphysics? Properties are normally outside the object in the sense
that the property adds something to the object. The object of a sci-
ence thus has to leave behind (derelinquere) something. To have angles
equal to two right angles is, for instance, a property of triangles. Yet,
insofar as this property is not already expressed by the denition of
the triangle one might say that it is added. But what could be added
to being? Moreover, the subject to which a property belongs can nor-
mally never be predicated of its property. It is plainly wrong to say
that to have angles equal to two right angles is a triangle. But insofar
as properties of beings exist, they have to be said to be. This nally
suggests that being cannot be the subject-matter of metaphysics.
43
The
Escorial commentator proposes a radical solution to these problems.
He plainly denies that the requirement to have parts and properties, a
requirement for demonstrative sciences, applies to metaphysics and its
42
Simon of Faversham, Quaest. super lib. Praed., q. 1, p. 73: Unde Philosophis hic
non procedit demonstrative, sed magis nominative et exemplariter.
43
For further information regarding the Escorial commentary on the Metaphysics
and a partial edition, cf. Pasquale Porro , Le Quaestiones super Metaphysicam attribuite a
Enrico di Gand: elementi per un sondaggio dottrinale, Documenti e studi sulla tradizione
losoca medievale 13 (2002), 507602. Cf. also the appendix in my Heinrich von Gent ber
Metaphysik als erste Wissenschaft. Studien zu einem Metaphysikentwurf aus dem letzten Viertel des
13. Jahrhunderts (Leiden, 2007), pp. 360374. Anonymus Escorial (Ps.-Henry of Ghent ),
Quaestiones super Metaphysicam, q. 119 [IV, q. 1], in supra, p. 576: Omnis scientia extra
ambitum sui subiecti aliquid debet derelinquere. Extra ambitum entis simpliciter nihil
potest derelinqui: ergo ens simpliciter non potest esse subiectum alicuius scientiae, et
sic idem quod prius. Maior patet, quia scientia una debet esse unius generis subiecti
partes et passiones considerans. Passio autem est aliud a subiecto.
object. According to him, metaphysics is simply not a kind of demon-
strative knowledge but merely denitive knowledge.
44
The Escorial commentator on the one hand and Simon of Faversham
on the other hand might have very different reasons for why they regard
the particular knowledge they are dealing with as merely denitive.
45

But what they both have in common is the idea that there are more
forms of proper knowledge than just strictly demonstrative knowledge.
This is, I think, one of the main reasons why Simon dropped the cause-
requirement from the requirements for scientia. Why drop this require-
ment (and not others)? Well, to restrict the label scientia to propter quid
knowledge seems most at odds with the medieval views about actual
scientic disciplines. Obviously, being qua being, for instance, does not
have propter quid causes either, nor are there any real causes of God,
the subject-matter of theology. But metaphysics and theology are surely
to be counted among the theoretical sciences. This last point is worth
being emphasized. Medieval authors from the 13th century onwards
are very keen to show how the many different scientic disciplines
match Aristotles criteria for demonstrative knowledge from the Posterior
Analytics. That is, the Posterior Analytics became especially important for
medieval accounts of the scientic status of metaphysics and theology
where it is less clear how these two could count as knowledge proper.
46

Texts such as the Escorial commentary and the commentaries on
Aristotles Categories that are discussed in this paper show a growing
awareness that our knowledge of what is rst and most fundamental
cannot be strictly speaking of the same kind as our knowledge of what
is less fundamental. These works counterbalance the optimistic views
expressed in many 13th-century Metaphysics commentaries.
44
Ibid., p. 578: Duplex est modus scientiae principalis, demonstrativus scilicet
et denitivus. De scientiis vero demonstrativis solum vericatur illa propositio, unde
sumitur ex illo verbo I Posteriorum scientia una est unius generis subiecti partes et pas-
siones considerans [An. Post. I 28 87a38sq.], ubi principaliter docet modum scientiae
demonstrativae. Talis autem non est ista scientia de ente simpliciter: non enim est
demonstrativa sed denitiva, docens investigare quidditates substantiarum et acciden-
tium, unde nihil considerat nisi quod sub suo subiecto includitur. Cf. also ibid., q. 122
[IV, q. 4], p. 587; q. 303 [VI, q. 2], p. 595.
45
It should be clear that Simon does not regard the scientia praedicamentorum as deni-
tive knowledge because there are no real properties or parts of the predicaments!
46
For the relevant discussion regarding metaphysics cf. Zimmermann , Ontologie oder
Metaphysik?; for theology cf. the classical study by Marie Dominique Chenu , La thologie
comme science au XII
e
sicle, 3rd ed. (Paris, 1969).
200 martin pickav
V.
There are two further points that become clearer when we compare
Simon of Favershams account of a scientia praedicamentorum with that
of his contemporaries. Lets rst look at the reasons why, according to
Simon, there can be no proper demonstrative knowledge of the catego-
ries. One of Simons arguments reminds us that the predicaments have
no (common) genus.
47
Notice that he does not says that predicaments
are the highest genera and that there is thus by denition no higher
genus which could be like a cause for the genus of each predicament.
Simons ontological understanding of predicaments as types of being
(and not as highest logical genera) precludes this line of reasoning. For
Simon propter quid knowledge ultimately provides us with an insight
into what the thing known really is. In other words: such knowledge
provides us with a denition of the thing we know. Now, there cannot
be any denition of the predicaments, for the predicaments are ten
different and irreducible types of beings. They have nothing really
or intrinsically in common. The commonness of the different types,
i.e., the reason why they are called types of being, is grounded in the
fact that they form a unity by attribution:
48
Substances are beings in
the proper sense; qualities, quantities etc. are things that are said to
be (beings) because they exist in something else etc. So, according to
Simon, the ontological diversity of the subject-matter of a science of
the predicaments rules out any propter quid knowledge of them. This
argument nicely underlines Simons ontological understanding of the
predicaments: predicaments are types of being.
49
In his list of the three criteria for knowledge, Simon further empha-
sizes his ontological understanding of the predicaments and the scientia
47
Simon of Faversham, Quaest. super lib. Praed., q. 1, p. 73: Predicamenta autem non
habent genus; et ideo de eis non est scientia ex prioribus et nocioribus simpliciter.
48
Ibid., p. 74.
49
In his rejection of the possibility of propter quid knowledge of the predicaments,
Simon also argues that there cannot be such knowledge because the essences of the
predicaments are hidden to us (occultae) like the essences of every other thing (ibid.,
p. 73). I dont think this is a helpful line of reasoning. If the principal inaccessibility of
the essences of things is an argument against propter quid knowledge in this case, then
it seems to rule out every possible kind of propter quid knowledge. Or maybe Simon
simply wanted to say the essences are unknown to us, i.e., the logicians who study the
categories. This reading, however, is equally problematic, for in the context there is
nothing that indicates that the claim is limited to the perspective of the logician.
praedicamentorum. Remember that the rst criterion demanded that the
object of knowledge has to be something existing outside the mind. I
already mentioned that this criterion does not appear on the similar
list of the Anonymous of Madrid . But its appearance on Simons list
gives the author the occasion to explicitly state that predicaments are
extra-mental entities and that the knowledge concerned with them is
about an extra-mental reality.
My second remark is related to Simon s contention that the scientia
praedicamentorum also deals with the parts of its subject-matter, i.e.,
the parts of the predicaments. Here again, a comparison with the
Anonymous of Madrid is interesting. The Anonymous mentions in
an argument in oppositum that knowledge is also concerned with the
parts of its object but he never takes this up again.
50
In fact, the whole
commentary is quiet about what could possibly be regarded as parts of
predicaments. As we already noticed earlier, Simon takes the idea of
parts of predicaments seriously. For him predicaments are composed
of two elements: a res praedicamenti and a ratio praedicamenti (a mode of
being). This compositional view of the categories can be found not only
in Simon but also in the works of Henry of Ghent and, to a certain
lesser extent, in the writings of Thomas Aquinas .
51
Henry of Ghent , however, is not concerned with the scientic status
of knowledge of the categories. His starting point is the compositional
nature of creatures. For Henry, creatures are not only composed out
of being (existence) and essence; essences themselves are composed. In
order to make this last claim more convincing, Henry goes on to develop
a theory according to which every being falling under a category (i.e.,
50
Anonymus Matritensis, Quaest. super lib. Praed., q. 1, p. 122: Oppositum per
Aristotelem et etiam per rationem arguitur. Quia de illo quod habet partes et pro-
prietates, et causas et principia per quae illae proprietates de partibus probantur, est
scientia.
51
One of the most obvious differences between Henry of Ghent and Thomas
Aquinas on this matter is terminology: Whereas Henry (cf. infra) distinguishes between
a res praedicamenti and a ratio praedicamenti, Aquinas talks about a composition of ratio
and being. Cf. e.g., Summa theologiae I, q. 28, a. 2, the emphasis is here on the acciden-
tal predicaments, but the basic idea also applies to substances: Ad cuius evidentiam,
considerandum est quod in quolibet novem generum accidentis est duo considerare.
Quorum unum est esse quod competit unicuique ipsorum secundum quod est accidens.
Et hoc communiter in omnibus est inesse subiecto, accidentis enim esse est inesse.
Aliud quod potest considerari in unoquoque est propria ratio uniuscuiusque illorum
generum. On this passage cf. Mark Henninger , Relations: Medieval Theories 12501325
(Oxford, 1989), pp. 1317.
202 martin pickav
every creature) is composed out of res praedicamenti and a mode of being
or ratio praedicamenti.
52
But why are predicamental beings composed at
all? For Henry this follows from the fact that predicaments are neither
only pure modes of being nor merely different things or essences.
53
To understand this better, let us begin by focusing on the rst part of
this disjunctive claim: that predicaments cannot be understood as pure
modes of being. In a way this seems to contradict Aristotles account
of the categories, for Aristotle denes substance, for instance, as that
which does not exist in something else.
54
The properties esse in subiecto
and non esse in subiecto characterize accidents and substances. The main
reason why the division of the predicaments cannot be derived solely
from the modes of being is that being itself is not a genus. Although
this argument is only alluded to in Henrys work,
55
it is fairly well
known from the works of Thomas Aquinas . Predicaments are types of
being. However, were the properties esse in subiecto and non esse in subiecto
meant to divide being adequately into substances and accidents, then
it would follow that being not in a subject (ens non in subiecto) would be
considered as something like the denition of substance. Now every
denition is composed of a genus and a specic difference. But being,
(ens), as Aristotle argued, does not have the character of a genus (or
of a most general type).
56
For there is nothing outside of being that
would be able to play the role of a restricting differentia. Likewise,
the other part of the stipulated denition, i.e., not in a subject (non
in subiecto), can also not constitute a genus or type of being. For it is
merely a negative qualication, and such a qualication cannot stand
52
Cf. Henry of Ghent , Quodlibet V, q. 2, ed. J. Badius Ascensius (Paris, 1518, repr.
Leuven, 1961), fol. 154rDvE: Res creata in simplicitate essentiae suae absque essentia
actuali considerata rationem compositionis habet ex eo quod est et quo est, sive ex
essentia inquantum est ipsa realitas dicta et ipso esse participato . . . Et secundum illa
duo quae cadunt in creatura secundum esse quidditativum considerata, constituitur
natura praedicamenti, et hoc ex illis duobus quae sunt res praedicamenti et ratio
praedicamenti. For Henrys understanding of the categories cf. Henninger , Relations,
pp. 4852 and my Henry of Ghent on Categories (forthcoming).
53
Ibid., fol. 154vE: Res enim sive essentia, inquantum res ist, non est praedica-
mentum, neque esse neque ens inquantum ens est.
54
Categoriae 5 2a1113.
55
Henry of Ghent , Summa quaestionum ordinariarum, a. 32, q. 5, ed. Raymond Macken
(Leuven, 1991), p. 79sq.
56
Cf. Thomas Aquinas , Summa contra Gentiles I, cap. 25: Dicendum est ex dictis quod
in denitione substantiae non est ens per se. Ex hoc enim quod dicitur ens non pos-
set esse genus: quia iam probatum est quod ens non habet rationem generis. For the
Aristotelian proof of the claim that being is not a genus cf. Metaph. III 3 998b2228.
for a genus, since then the genus would not express what something
is but only what it is not.
57
Consequently, in order for there to be a
genus substance, an essence has to have generic features; the mode
of being (for instance, to be in a subject) is then only added to this
essence or thing (res). The same is true for the denition of accidents.
Aquinas can thus conclude:
Substance is what exists by itself is not a true denition of substance,
and Accident is what exists in something else is not a true denition of
accidents, but only a circumlocution of the true account. The true account
has to be understood like this: Substance is a thing (res) whose nature
requires that it does not exist in something else, An accident, however, is a
thing (res) whose nature requires that it does exist in something else.
58
Henry of Ghent is in complete agreement with Aquinas s analysis and
it is on this background that we have to understand his claim that it is
not just in virtue of being a thing (res) or some nature that something
has the character of a type of being, nor is it in virtue of its being in
a subject or not being in a subject, but insofar as it is composed out
of both aspects.
59
The meaning of this last statement becomes clearer if we look at why
it is impossible for Henry to give a distinct account of each of the ten
predicaments only by their res praedicamenti. Aquinas seems to hold that
each predicament has its own characteristic thing or essence, ratio in
57
Thomas Aquinas , Summa contra Gentiles I, cap. 25: Similiter nec ex hoc quod dici-
tur per se. Quia hoc non videtur importare nisi negationem tantum: dicitur enim ens
per se ex hoc quod non est in alio; quod est negatio pura. Quae nec potest rationem
generis constituere: quia sic genus non diceret quid est res, sed quid non est. Cf. also
Henry of Ghent , Summa quaestionum ordinariarum, a. 26, q. 2 ad 1, ed. J. Badius Ascensius
(Paris, 1520, repr. St. Bonaventure, N.Y., 1953), fol. 159rT.
58
Thomas Aquinas , Quodlibet IX, q. 3, a. 1, ed. Leonina, vol. 25 (Rome, 1996), p. 99,
8390: Et ideo haec non est vera denitio substantiae: Substantia est quod est per
se, vel accidentis: Accidens est quod est in alio, sed est circumlocutio verae descrip-
tionis, quae talis intelligitur: Substantia est res cuius naturae debetur esse non in alio;
Accidens vero est res, cuius naturae debetur esse in alio. For Aquinass denitions
of the categories cf. John Wippel , The Metaphysical Thought of Thomas Aquinas. From Finite
Being to Uncreated Being (Washington, D.C., 2000), pp. 228237; Etienne Gilson , Quasi
Denitio Substantiae, Armand Maurer et al., eds. St. Thomas Aquinas 12741974.
Commemorative Studies (Toronto, 1974), vol. 1, pp. 111129.
59
Summa quaestionum ordinariarum, a. 32, q. 5, p. 79: Non enim ex hoc quod aliquid
sit res et natura aliqua, sive substantiae sive accidentis . . ., habet rationem generis prae-
dicamenti, neque similiter ex hoc quod est esse non in subiecto vel esse in subiecto,
quoniam esse ex se nullo modo potest habere rationem generis . . ., sed natura generis
praedicamenti ex utroque simul, scilicet ex re et ratione praedicamenti constituitur.
204 martin pickav
his terminology.
60
This would allow us, at least, to distinguish the ten
predicaments by their essences. For Henry, however, this is impossible
because there are only three different kinds of res. Only the categories
of substance, quality, and quantity, which he calls absolute categories,
contain their own res praedicamenti; the remaining seven relative categories
(relation, location, time etc.) do not possess any real content on their
own, rather they owe their reality to a res of the rst three predica-
ments, on which they depend as their foundation. Henry tries to defend
his view of the unrealness of relations (and the relational categories)
by a series of arguments. I will only focus on two of them here.
One of Henrys arguments advances the idea that there is nothing
specic to understand when our intellect understands relations them-
selves: If the thing (res) of substance, quantity, and quality is completely
set aside by the intellect and if a notion of a relation or a respect
towards something else is grasped precisely as such, then it is impossible
to assign to the relation any specic difference.
61
Relations, according
to Henrys view, are not by themselves relations of a specic type, say,
relations of similarity, identity, temporal priority, etc. They owe their
specic character to the entities which they relate to one another. In this
respect he treats relations as similar to certain second order concepts
such as the concept of species: In the same way in which being a spe-
cies does not differ in character regarding man, donkey, and cattle . . . so
being-towards-something-else does not differ in character regarding
master and servant. What is responsible for the fact that the master is
related to the servant by a specic relation, the relation of dominion,
is not the relation itself, but the power to dominate in which the rela-
tion is founded in the master and in the inability to resist the master,
in which the relation is founded in the servant.
62
But since a relation
as such (i.e., simply as a relation) lacks any features that determine it
60
Cf. supra fn. 51.
61
Quodlibet VII, qq. 1+2, ed. Gordon A. Wilson (Leuven, 1991), p. 23: Quod clare
patet ex tribus rationibus philosophicis. Primo quoniam si circumscribatur per intellec-
tum omnino res substantiae, quantitatis, et qualitatis et concipitur intellectus relationis
sive respectus ad aliud praecise, sub illa impossibile est assignare aliquas differentias
specicas. Ipsum enim ad aliud esse, quantum est ex se, eiusdem rationis est, si nullam
contrahat differentiam ab eo super quod fundatur.
62
Ibid. Sicut enim esse speciem non est alterius rationis circa hominem, circa
asinum et circa bovem . . . sic esse ad aliquid [lege aliud] non est alterius rationis circa
dominum et circa servum . . . circumscripta realitate aliorum praedicamentorum quae
importatur nomine domini et servi, quae est dominandi potentia in domino et impo-
tentia resistendi in servo, quae pertinent ad secundam speciem qualitatis.
to something specic, and since entities belonging to a category are
specied by something real, it follows that there is nothing real in the
category of relation except that which belongs to another category on
which a relation is founded.
63
Another argument against the idea that relations have their own
thing (res) runs like this: Lets assume that relations have their own
thing (res) or reality in things they relate and that this realitas is dif-
ferent from the reality of the related thing. In this case, an object not
related to something elsesay, the white Socrates who is not related to
the black Plato cannot enter in a relation if there is not something
newly produced in it that hasnt been there previously. There has to be
some kind of transmutation, a real change; Socrates who is not similar
to Plato cannot become similar without him being transformed. For
Henry, however, this conclusion is wrong. For imagine that a quality in
Plato is changed so that he, Plato, becomes white. Socrates will then
immediately become similar to him, and this without any kind of real
change in Socrates.
64
This argument shows that a too realist concep-
tion of relations would lead to the absurd consequences that things are
constantly changing and being transformed because they constantly
enter into new relations.
For Henry of Ghent it is thus clear that relation and relational cat-
egories do not possess their own res. These relations are real insofar
as they are grounded in the res of their foundations (i.e., in the res of
one of the absolute predicaments). Considered in themselves, relations
only express a mode of being or a ratio praedicamenti, namely, a being
toward something else (esse ad aliud).
65
Henry adopts a very restricted
63
Ibid., p. 24: Quare, cum nihil est reale in praedicamento, nisi illud penes quod
accipiuntur differentiae specicae in illo et specierum distinctio, nihil igitur procul dubio
reale est in praedicamento relationis, nisi id quod est res alterius praedicamenti.
64
Ibid., p. 24sq.: Si relatio poneret propriam realitatem a realitate sui fundamenti
in suo subiecto relato, tunc de se non relatum aliqua relatione, puta similitudine, quia
nullum habet in se respectu alicuius alterius, puta Sortes albus in respectu ad Platonem
nigrum, non eret relatum ad illud similitudine quam haberet in se de novo et prius
non habuit, nisi quia res aliqua de novo facta est in eo, quae prius in eo non fuit. Hoc
autem non est possibile eri in aliquo sine propria sui transmutatione. Sortes ergo
similis Platoni non posset eri similis eidem sine sua transmutatione reali. Consequens
falsum est, quia Platone transmutato . . ., ut de nigro at albus, statim Sortes sine omni
sua transmutatione de novo factus est similis Platoni.
65
Quodlibet IX, q. 3, ed. Raymond Macken (Leuven, 1983), p. 56: Propter quod
saepius alibi diximus quod relatio realitatem suam contrahit a suo fundamento et quod
ex se non est nisi habitudo nuda, quae non est nisi modus quidam rem habendi ad
aliud. For more on Henrys understanding of relations cf. Henninger , Relations, pp.
206 martin pickav
use of the term res. Of course he does not deny that relations are real;
relations are not nothing, they are just not things (res).
66
It is therefore
wrong and exaggerated to conclude Henry is cutting down the list of
categories to just three when he says that there are, in the totality of
creation, only three genera of things, namely, substances, quality, and
quantity.
67
This statement, however, emphasizes that the division of the
ten predicaments or categories cannot solely be based on the division
of the res. For a complete classication of the predicaments we need a
combination of res praedicamenti and ratio praedicamenti.
VI.
To my knowledge, Simon of Faversham is the rst and only commen-
tator in the 13th century who uses Henry of Ghent s doctrine of the
categories afrmatively in a Categories commentary.
68
Simon too thinks
that the different predicaments cannot be distinguished on the basis of
their respective essences alone but only insofar as they form a composi-
tion of essences and modes of being. He even goes beyond Henry and
interprets the composition along the lines of the composition between
matter and form. The mode of being is of course not a form but it can
be considered as a formal constituent in the composition, whereas the
underlying res or essence is like a material constituent. And like Henry
4058; Rolf Schnberger , Relation als Vergleich. Die Relationstheorie des Johannes Buridanus
im Kontext seines Denkens und der Scholastik (Leiden, 1994), pp. 86102; and the many
articles by Jos Decorte , e.g.: Giles of Rome and Henri de Gand on the Reality of
a Real Relation, Documenti e studi sulla tradizione losoca medievale 7 (1996), 183211;
Avicenna s Ontology of Relation. A Source of Inspiration to Henry of Ghent , in Jules
Janssens/Daniel De Smet, eds. Avicenna and his Heritage. Acts of the International Colloquium,
Leuven Louvain-la-Neuve, September 8September 11, 1999 (Leuven, 2002), pp. 197224;
Relatio as Modus Essendi: The Origins of Henry of Ghents Denition of Relation,
International Journal of Philosophical Studies 10 (2002), 309336.
66
According to Henry they can only be called res if we allow for a broader use
of the term, namely, one that applies res also to modes of being. Cf. Quodlibet IX, q. 3,
p. 56: . . . nisi extendendo rem ut etiam modus rei dicatur res.
67
Quodlibet V, q. 6, fol. 161vO. The misunderstanding mentioned can, for instance,
be found in Wippel , The Metaphysical Thought of Thomas, p. 225 and fn. 98.
68
But maybe I should be more cautious given that in this paper I only study edited
commentaries.
(and unlike Aquinas ) he refers to the two constituents of predicaments
as res praedicamenti and ratio praedicamenti.
69
Further evidence that Simon follows Henrys stricter version of the
compositional view of predicaments can be gathered from Simons
statements on relations.
70
Like Henry, Simon denies that relations import
their own distinctive thingness. He offers three main arguments for this
claim. His third argument is absolutely identical with one of the above
mentioned arguments Henry gave for the same conclusion: Whenever a
thing is added to something else we say a transformation or change has
occurred. But something can become related to something else without
itself being changed. Suppose that Socrates is white right now and
that tomorrow the white Plato will be generated; what will happen
in this case is that Socrates will then become related to Plato by the
relation of similarity, but nothing really happened in Socrates himself.
Thus, such a relation does not add a thing to Socrates, although it
adds a real relation to him. For he is now really (and newly) related to
Plato and this is not just a matter of an observing mind (i.e., a rational
relation).
71
The other two arguments seem to be Simon s own creations. The
rst runs like this: Suppose the opposite, namely, that a relation, e.g., the
relation of similitude, adds some thing to that in which it is founded, e.g.,
the whiteness. Of what kind is this added thing (res)? Simon assumes that
it would be a quality. But every quality can be the basis of a relation
69
Simon of Faversham, Quaest. super lib. Praed., q. 13, p. 86sq.: Predicamentum enim
substantie constituitur ex duobus, scilicet ex re et ex modo essendi sibi superaddito; illa
enim requirantur in omni predicamento, per que predicamenta ad invicem distinguun-
tur; sed predicamenta non distinguuntur solum penes essencias, quia videmus diversas
essencias contineri sub uno predicamento . . . requiruntur ergo diversi modi essendi per
quos distinguantur, et iste modus essendi est formalior in predicamento quam res, et
dicitur ratio predicandi [lege predicamenti]; et ideo quodlibet predicamentum constituitur
ex re et ex ratione predicandi [lege predicamenti]. Cf. also q. 20, p. 93.
70
Although Aquinas too holds the view that the predicaments are composites,
Henrys version of this view is more radical, for, unlike Aquinas (cf. text supra in fn.
51), he does not hold that there is one specic essence (ratio in Aquinass words) for
each predicament. There are fewer essences than predicaments.
71
Quaest. super lib. Praed., q. 42, p. 135: Nulla res potest de novo advenire alicui
<sine> transmutatione facta in eo; sed similitudo fundata supra albedinem potest
advenire alicui absque aliqua transmutatione facta in eo. Ponamus enim quod Socrates
nunc sit albus, et cras generetur Plato albus; Socrates t similis Platoni, et tamen nulla
transmutacio t in Socrate. Similitudo ergo fundata supra albedinem non est res et
natura albedini addita; et quamvis illa similitudo non sit res et natura, est tamen aliquod
ens reale. For Henys argument cf. supra fn. 64.
208 martin pickav
of similitude. So, if a relation adds a quality to the thing that it relates
to something else, then this quality can itself become the basis of
another relation and so forth. According to Simons argument, the realist
account of relations leads into an innite regress and must therefore
be wrong.
72
I am not convinced that any defender of the view that real
relations add some thing to their foundations would be convinced by
Simons reasoning, since it assumes the highly controversial claim that
such an added entity would be a quality, i.e., something belonging to
a different category. But why should this be so? Proponents of a strong
realist view of relations characteristically hold that the added thing is
a res relativa, an entity which cannot be reduced to the thing of any
other category.
Simon s second argument is based on the idea of composition. If
relations add real things to their foundations, then they cause a greater
degree of composition in them. To use the example of the arguments
discussed before: if the relation of similitude adds a thing to the white
Socrates , then Socrates is no longer simply the entity composed of
matter, form, and a series of characteristic accidents (snub-nosed, bald
etc.), he is now this whole composition plus the new thing. But this, so
Simon contends, is counter-intuitive. Socrates himself does not seem to
have changed with regard to the things out of which he is composed.
The same is true in other cases. Once an agent causes an effect, the
agent is related to its effect. But the agent is then not more composed
than in the moment before the agent acted.
73
Imagine a natural agent
like the sun, which has many different effects. It should exhibit a very
high degree of composition were it true that relations themselves add
things to their foundations.
72
Ibid., p. 134sq.: Si illa similitudo fundata super albedinem esset res addita albe-
dini, illa res esset de genere qualitatis; sed supra omnem rem de genere qualitatis potest
fundari similitudo, quia proprium est qualitatis secundum eam simile vel dissimile dici;
ergo supra eam similitudinem fundatur similitudo. Quero tunc de illa similitudine, aut
illa similitudo sit res addita priori similitudini per quam fundatur, aut non. Si sic, erit
de genere qualitatis, et cum supra omnem rem de genere qualitatis contingat fundare
similitudinem, queram de illa similitudine; et sic erit procedere in innitum, aut erit
dare quod relatio fundata supra rem aliquam non est res addita ei.
73
Ibid., p. 135: Omnis res addita rei maiorem facit compositionem in re; si ergo
similitudo esset res aliqua addita albedini, tunc similitudo adveniens alicui maiorem
compositionem causaret in eo; nunc autem similitude adveniens albedini non causat
maiorem composicionem in re; Socrates enim similes et albus non est composicior in
re quam si sit albus solum; sicut aliquod agens non est composicius ex hoc quod habet
formam et agit per illam quam ex hoc quod habet illam solum.
No doubt, Henry of Ghent and Simon of Faversham are propo-
nents of the same view of the predicaments. So far I tacitly assumed,
in accordance with the traditional opinion, that Simons teaching on
this issue is posterior to and dependent on Henrys. But nothing rules
out that the dependence is in the opposite direction. Simons Categories
commentary is generally believed to have been composed in the years
around 1280, but this dating is pretty vague.
74
The relevant works of
Henry of Ghent can be dated with a bit more precision: The fth
Quodlibet was held either Christmas 1280 or Easter 1281, the seventh
Quodlibet Christmas 1282, and art. 32 of the Summa quaestionum ordina-
riarum is considered as having been disputed in the year 1279;
75
thus
the dates do not preclude any of the alternatives. Of course, this whole
question about possible sources and dependencies needs some further
investigation, but for now I favor the view that Henry was at the origin
of the strong compositional account of the predicaments. The theory is
much more developed in his works and the bits and pieces in Simons
commentary are only fully comprehensible against the background of
Henrys works.
If that is true, then we can see Simon not just borrowing a philo-
sophical theory of Henrys; he also engages critically with what he
takes from his contemporary. This is obvious in Simons discussion of
the difference between a relation and the absolute thing (res) in which
the relation is founded. For Henry it was obvious that this distinction
could not be a real distinction. If the relation does not add another
thing to its foundation, then relation and foundation cannot be really
distinct, i.e., distinct in the way one thing is distinct from another. In
the case of a real distinction the two items which are distinct bring
with them their own natures or essences. But a relation does not have
its own essence besides the essence of its foundation.
76
Neither is the
distinction between a relation and its foundation a purely rational
distinction. Its not a matter of how our intellect looks at relations and
74
Cf. Introduction in Quaestiones super libro Elenchorum , p. 5; I nd it highly unlikely
that both authors came up on their own with their respective teaching on the categories.
But it should not be ruled out that both depend on an yet unknown source.
75
Cf. Matthias Laarmann , Deus, primum cognitum: Die Lehre von Gott als dem Ersterkannten
des menschlichen Intellekts bei Heinrich von Gent ( 1293) (Mnster, 1999), p. 50sq.
76
Cf. Quodlibet V, q. 6, fol. 161rK: Unde re differunt quaecumque <quae> diver-
sas naturas et essentias important secundum rem, sive fuerint simplicia, ut materia et
forma, sive composita, ut homo et asinus . . . Nullus enim respectus aliquid rei ponit
praeter rem eius super quod fundatur.
210 martin pickav
their foundations. The distinction is real in the sense that it is not com-
pletely mind-dependent. The distinction between them must, therefore,
be weaker than a real distinction but stronger than a purely rational
one. Henry allows for such an intermediate distinction; he calls it an
intentional distinction.
77
Simon of Faversham takes up this line of argument. Using in part
the very same words as Henry, he agrees with Henry that relations are
not really distinct from their foundations.
78
However, Simon goes on
to deny that there is, in this case, an intentional distinction. His descrip-
tion of the intentional distinction follows closely that of Henrys fth
Quodlibet. There, Henry explains that different intentions are found in
the very same thing (or essence). Intentions are, so to speak, different
eidetic aspects of the very same thing, such as animal and rational
of an individual human being. Animal and rational are not really
distinct since both elements do not carry along their own essences.
They are simply parts of the one essence human being. Without the
intellect we would, of course, never be able to distinguish animal and
rational; it doesnt follow, however, that their distinction is purely a
matter of the mind. There is something ex parte rei that corresponds to
this distinction. That is the reason why Henry distinguishes the inten-
tional distinction from the rational distinction, which imports a distinc-
tion that is completely mind-dependent. Yet, the role of the intellect in
discovering intentions (although not in producing them) explains
why Henry denes intentions by introducing a reference to concepts:
An intention is something that really belongs to the simplicity of an
essence and that can be conceived in a precise way without conceiv-
ing of something else from which it doesnt differ as a real absolute
77
Henry of Ghent explains this using the example of a relation of similitude
(between two white objects); Quodlibet V, q. 6, fol. 162vS: Similitudo ergo sic supra esse
qualitatis fundatur, ut sit omnino extra conceptum esse qualitatis secundum rationem
essentiae suae, ut qualitas est simpliciter. Talia autem . . . necesse est saltem secundum
intentionem differe inter se. For Henrys famous intentional distinction cf. John Wippel ,
Essence and Existence, in Norman Kretzmann et al., eds. The Cambridge History of
Later Medieval Philosophy. From the Rediscovery of Aristotle to the Disintegration of Scholasticism
11001600 (Cambridge, 1982), pp. 385410; for Henry on the distinction between
relations and their foundations cf. Henninger , Relations, p. 54sq.
78
Quaest. super lib. Praed., q. 42, p. 136: Illa differunt secundum rem que impor-
tant diversas essentias secundum rem, sive sint simplicia sicut materia et forma, sive
composita ut homo et equus . . . Modo dico quod relatio et res supra quam fundatur
non differunt secundum rem . . . relatio non est res et natura diversa a re supra quam
fundatur. Cf. Henrys words in fn. 76.
thing.
79
It is interesting to see that Simon does not simply reject the
very idea of an intentional distinction; he just denies that it applies to
the distinction in question. In order to show this he uses Henrys own
description of the intentional distinction. For Henry, as we just saw,
the intentional distinction is a distinction that applies to items that
have the same essence. But, for Simon, the foundation of a relation is
the cause of the relation, while the relation itself is something like the
effect of the foundation in a different relata. Cause and effect, he adds,
have their own essences.
80
Thus relations and their foundations cannot
differ intentionally.
81
Now the only remaining possibility seems to posit
a rational distinction between the two items in question. And this is
exactly the way Simon goes. But he rejects the traditional understand-
ing of the rational distinction as a distinction between different ways
of understanding one and the same thing. In his version of the rational
distinction the distinction applies to things and their modes and is thus
in a sense real because it is not just caused by the intellect alone.
82
Simon s account of the distinction between relations and their foun-
dations is a bit odd for at least two reasons. (1) Simons idea about the
nature of a rational distinction as a distinction between things and
their modes seems to be ad hoc. It departs radically from the way other
people in his time used the concept of a rational distinction. According
to this traditional understanding, a rational distinction is a distinction
79
Quodlibet V, q. 6, fol. 161rL: Sed appellatur hic intentio aliquid pertinens realiter
ad simplicitatem essentiae alicuius natum praecise concipi absque aliquo alio a quo
non differt re absoluta quod similiter pertinet ad eandem . . . De eis enim quae sunt
idem re in eodem, aliquando sic formantur conceptus diversi, ut neutrum eorum in
suo conceptu alterum includat, ut sunt conceptus diversarum differentiarum quae
concurrunt in eodem . . . et similiter conceptus generis et differentiae.
80
Think of the case of parents and their children. The parents and the children
have their own individual essences. Otherwise they would be identical!
81
Quaest. super lib. Praed., q. 42, p. 136: Secundum intencionem differunt illa quae
inportant diversam [lege eandem] essenciam sub diversis conceptibus, quorum unum
non includitur in alio; et isto modo genus et differentia differunt . . . Nec differunt secun-
dum intencionem, quia illa differunt secundum intencionem que inportant eandem
unam essenciam et naturam sub diversis rationibus; sed similitudo et res supra quam
fundatur non important eandem essenciam, quia causa et effectus per se aliam et aliam
essenciam important; nunc autem relatio et res supra quam fundatur differunt sicut
causa et effectus per se. Cf. supra fn. 79 for a comparison with Henrys account of
the intentional distinction.
82
Ibid.: Oportet ergo ponere quod relatio et res supra quam fundatur differant
secundum rationem, quia illa differunt secundum rationem quorum unum comparatur
ad aliud sicut modus rei ad rem; sed sic se habet relatio et res supra quam fundatur . . . Et
dico quod relatio et res differunt secundum rationem, non ita quod differant secundum
rationem intelligendi solum, sed differunt secundum rationem, quia unum non est de
essencia alterius.
212 martin pickav
applying to the different ways something is grasped by the intellect. A
human being and a rational animal, for instance, are two things that
are only rationally distinct, for it is exactly the same thing that is both.
But a rational animal is not simply identical with a human being. For
then everybody who intellectually perceives a human being would also
right away perceive his denition. It is obvious why this distinction is
rational, but what is the reason for calling the distinction between thing
and mode of thing rational? (2) Furthermore, Simons argument is
inconsistent. He rejects the intentional distinction on the basis that it
applies only to objects that have the same essence, whereas in relations
and their foundations there must be more than one, because they relate
to each other as effect to cause. This, however, contradicts what he says
elsewhere about relations. And, even more problematic, this seems to
rule out the rational distinction as well. For Simon explicitly says in
his initial description that a rational distinction only applies to things
with one and the same essence.
83
But if this very fact counts against the
intentional distinction it should also count against the rational one.
It is hard to imagine how an author committing such crude mistakes
could have been at the origin of the elaborate compositional view of
the predicaments that we nd in Henry of Ghent s writings. If Simon
were the origin of this theory, the weakness of this argument would
surely not have gone unnoticed in Henrys Quodlibet V, q. 6, a text that
has so many resemblances to q. 42 of Simons Categories commentary.
I assume, though it is little more than just an assumption, that it is
therefore more plausible to regard Simons discussion as dependent
on Henrys.
VII.
We still have not answered one of the central puzzles of Simon s under-
standing of a scientia de praedicamentis. Does such knowledge qualify as
logical knowledge or does it belong to metaphysics? Furthermore, can
Simons understanding of the place of the Categories in logic, which is a
scientia rationalis, be harmonized with his insistence that the knowledge
83
Ibid.: Illa differunt secundum rationem que important eandem essenciam; sed
unum se habet ad alterum sicut modus rei ad rem. Compare with the text supra in
fn. 81.
of the predicaments is real knowledge? I can imagine some reader
objecting at this point, for someone might consider the problem in
question a pseudo-problem and one that is purely my own construction.
After all, Simons description of logic as a discipline that falls short of
being proper knowledge is mostly developed in his commentaries on
Porphyry s Isagoge and on the Sophistical Refutations. How can I assume
that ideas developed in those works apply equally to Simons Categories
commentary? Maybe Simon changed his mind? I acknowledge I do
not take such a consideration seriously, nor do I think that it should
be taken seriously by others. It is justied only if there is absolutely
no other (and better) way to make sense of Simons understanding of
the categories. If such a consideration is meant as a starting point for
an interpretation, it destroys every need for a thoroughly philosophical
interpretation of the kind I intend to provide here.
84
So, in this sense,
there is a real problem here: The way in which Simon describes the
scientia de praedicamentis (i.e., as knowledge about something extra-mental
that has real properties and parts etc.) makes it sound as if it is at least
partially identical with metaphysics. Of course, the subject-matter of
metaphysics is, according to most 13th-century philosophers, being qua
being and not the predicaments as such. But insofar as the predicaments
are referred to as parts (or modes) of being, the study of the predica-
ments is part of the study of being and thus part of metaphysics. From
this perspective it looks as if the scientia de praedicamentis is something
like a metaphysica specialis, to use a term that became fashionable in the
early modern period.
The impression that the scientia de praedicamentis belongs to metaphysics
is almost reinforced by Simon s treatment of its unity. Knowledge is not
about particular items as such but of something universal. It applies
to particulars only insofar as they fall under a certain universal genus.
There is no knowledge of this individual tree in front of my window, but
knowledge of trees also covers this individual tree. In other words, the
object of knowledge has to be some universal genus, and from the unity
84
There is, however, a more sophisticated version of this objection. It is based on
the story, often repeated at the end of the 13th century, according to which Aristotle
wrote the Categories as a young man and not everything in the Categories has to be
taken as his own considered opinion. On this legend (and its origin) cf. Ebbesen , Les
Catgories au moyen ge, p. 247sq. Simon quotes it in Quaest. vet. super lib. Elench., q. 15,
p. 67sq. Whatever one might think of this story, I dont consider it as crucial for Simons
understanding of the Categories. I expect he would have mentioned it in his Categories
commentary itself, had he really given it more than just anecdotal value.
214 martin pickav
of this genus derives the unity of the corresponding knowledge. But
how can there be any unied knowledge of the predicaments if there
is no common genus for them? Some commentators on the Categories
simply avoid this problem by denying that real being is the subject-
matter of the scientia praedicamentorum. For Martin of Dacia , for instance,
the subject-matter of this scientia is an intellectual being considered
by the soul, a logical concept. Only this kind of being, but not real
being, matches the criterion to be common to all the predicaments.
More precisely, that which is sayable without a connection (dicibile
incomplexum) is the subject-matter of the knowledge exposed in the
Categories.
85
But for Simon, as well as for Peter of Auvergne and the
Anonymous of Madrid , there is a simple response to the problem of
the non-existing generic subject-matter. It is not true in every case that
knowledge has to have a subject-matter that has the unity of a genus
(i.e., is unum genere). There are forms of knowledge, such as metaphysics,
for instance, which have a non-generic object. The object of metaphysics
is being (ens), which, however, is not a genus. Nevertheless metaphysics
is one unied science of being. The unity of metaphysics derives from
the fact that everything that is, is said to be by attribution to substance,
the rst kind of being. In this way being has quasi-generic features and
is an object that is at least one by attribution (unum secundum attributio-
nem). Consequently the knowledge of being is one unied knowledge,
for everything that is considered in metaphysics is considered insofar
as it can be related by attribution to substance, the primary kind of
being.
86
In the same way, we are to understand that there is one uni-
ed form of knowledge of the predicaments even though there is no
common genus.
87
85
Martin of Dacia , Quaestiones super librum Praedicamentorum, q. 1, ed. Heinrich Roos
(Copenhagen, 1961), p. 155: Duplex est ens, scilicet ens commune et intentionale
apprehensum ab anima et ens speciale et reale ponens aliquid in praedicamento.
Unde cum dicitur: omnis scientia est de ente, concede quod <non> est de ente aliquo
reali, quod ponit aliquid in praedicamento; sed est de ente intellectuali considerato
ab anima, et istud ens nihil aliud est quam ratio quaedam logica. Ibid., q. 2, p. 156:
Subiectum cuiuslibet scientiae logicalis debet esse res logica. Sed dicibile incomplexum
est huiusmodi . . . Et hoc est subiectum libri Praedicamentorum.
86
Quaest. super lib. Praed., q. 1, p. 74: Non requiritur quod subiectum sit unum in
genere, quia scientia divina [i.e., metaphysics] est una, tamen subiectum eius non est
unum genere, cum eius subiectum sit ens; ens autem non est unum genere. Sed ad
hoc quod aliqua scientia sit una, sufficit quod subiectum sit unum secundum attribu-
tionem. Hoc modo scientia divina est una, quia omnia que ibi considerantur habent
attributionem ad unum primum ens quod est substantia.
87
Ibid. Cf. also Peter of Auvergne , Quaestiones super Praedicamentis, qq. 1+2 , p. 10;
Anonymus Matritensis, Quaest. super lib. Praed., q. 2, p. 124.
There is, however, a traditional view, according to which the pre-
dicaments can be considered in (at least) two ways. The metaphysician
considers them according to their being and their essences, whereas
the logician considers them insofar as second intentions, with which
logic deals primarily, are grounded in them.
88
In one form or another,
this view gets repeated in most of the Categories commentaries in the
13th century
89
and we will not be surprised to nd it in Simon of
Favershams commentary too. Note, however, that Simon turns the
idea of a twofold consideration of the categories in a special direction.
When he mentions the idea, his exact words are:
And from this it is clear how this scientia and the scientia divina (i.e., meta-
physics) consider the predicaments in different ways. For metaphysics
considers them insofar as they are certain essences and parts of being;
in this book however the predicaments are not treated insofar as they
are things of some absolute matter (?), but insofar as they have this or
that mode of predication. And therefore we speak of the Book of Pre-
dicaments.
90
There are two things worth mentioning: (1) Simon is more precise
than many of his contemporaries insofar as he tells us exactly how the
predicaments are considered in metaphysics. They are not the subject-
matter of metaphysics, but they fall under the study of being because
the predicaments are conceived of as parts of being.
91
(2) Surprisingly,
Simon does not mention logic at all, but instead talks about this scientia,
i.e., the scientia de praedicamentis, and the way this scientia is concerned
about the predicaments in opposition to metaphysics. I believe Simons
88
Cf. e.g., Peter of Auvergne , Quaestiones super Praedicamentis, q. 3 , p. 11: De praedi-
camentis possumus loqui dupliciter: aut secundum sui entitatem et essentiam, et sic sunt
de consideratione ipsius metaphysici; aut secundum quod in ipsis fundantur secundae
intentiones, et sic ea logicus considerat. Cf. Avicenna , Liber de philosophia prima, tract.
I, cap. 1, ed. Simone van Riet (Leuven, 1977), p. 10. for the view that logic considers
second intentions as its subject-matter.
89
On the origin of this distinction cf. Pini , Categories and Logic in Duns Scotus , pp.
4044 (with an explicit discussion of this distinction in Peter of Auvergne and the
Anonymous of Madrid ).
90
Quaest. super lib. Praed., q. 1, p. 74: Et ex hoc apparet quomodo diversimode
considerat ista scientia de predicamentis et scientia divina [i.e., metaphysics], quoniam
scientia divina considerat de eis ut sunt quedam essencie et partes entis; in isto autem
libro non determinatur de predicamentis secundum quod sunt res materie absolute
[sic], sed secundum quod habent talem modum predicandi vel talem; et ideo dicitur
liber Predicamentorum.
91
Peter of Auvergne (cf. supra fn. 88) and the Anonymous of Madrid (p. 125), for
instance, simply say metaphysics considers the predicaments according to their being
and their essence (Peter) or insofar as they are beings.
216 martin pickav
choice of words is intentional. For if logic is primarily about what is
caused by the intellect,
92
then this scientia, whose subject-matter is
something that is explicitly said to be mind-independent, cannot be
identical with logic!
But what exactly does the scientia de praedicamentis study? According to
Giorgio Pini , the passage just quoted is the key for understanding how
Simon conceives of logics approach to the categories:
Logic deals with categories insofar as they have a certain mode of predi-
cation, but Simon also maintains that categories are composed of an
essence and of a mode of being, which is either identical to or is parallel
to a mode of predication, as we have seen. It can be said, therefore, that
logic studies categories not according to their essences but according to
their modes of being, and such modes of being are as real as the essences
of categories.
93
To fully appreciate Pini s point we have to go back to the place where
Simon draws the connection between modes of being and modes of
predication. Like Aquinas , Simon considers modes of being (modi essendi)
and modes of predication (modi praedicandi) to be parallel. And because
the former are made manifest by the later, we refer to the different types
of being by the term predicament.
94
According to Pini, Simon even
identies the modes of predication with the modes of being, and if we
replace the one by the other, the quote mentioned above makes a clear
case: metaphysics deals with essences of the predicaments, logic with
their being. We might still go further and say that this twofold examina-
tion of predicaments is required by the composition of predicaments out
of res praedicamenti and ratio praedicamenti. At any rate, Pinis interpretation
seems to make excellent sense of the fact that the scientia de praedicamentis
is real knowledge, i.e., knowledge of an extra-mental reality.
However, I do not think Pini s account of what the scientia de praedica-
mentis studies is completely adequate. One problem is the identication
of modes of being and modes of predication on which the account at
least partially rests. I can see no clear passage in which Simon really
identies the two. The one passage at which Pini points contains most
likely a misreading on the part of the editor.
95
But even if the text in the
92
Cf. the texts supra in fn. 5.
93
Pini , Categories and Logic in Duns Scotus , p. 154.
94
Cf. supra fn. 26.
95
Cf. the text supra in fn. 69. The text almost denitely has . . . ideo quodlibet
predicamentum constituitur ex re et ratione predicamenti [instead of Mazzarellas
praedicandi ].
passage in question is correct there is a further problem. For the study
of the modes of beings seems already included in the task assigned to
metaphysics. In the quotation Simon does not limit metaphysics to the
study of the essences of the predicaments; he says too that metaphys-
ics studies predicaments insofar as they are parts of being ( partes entis).
Yet, it seems articial to exclude the modi essendi from the inquiry into
the parts of being. Moreover, if we compare Simons teaching about
the composition of the predicaments with Henry of Ghent s, then it
becomes clear that the modes of beings out of which predicaments are
composed belong to the side of essences as opposed to existence. As
part of the essence of a predicament, the modes of being were then,
according to Pinis division of labor, to be studied together with the
essence in metaphysics and not in logic. Finally, if we look carefully at
what Simon exactly says, he neither says that metaphysics studies the
categories according to their essences nor is it immediately clear (and
I rather doubt) that Simon really intends to talk about metaphysics as
opposed to logic.
VIII.
In order to get a clearer idea of what the scientia de predicamentis is sup-
posed to be about for Simon I, once more, propose to start by looking at
another contemporary Categories commentary. When the Anonymous of
Madrid puts forward the double consideration view, the commentator
ascribes to logic (or in his words dialectics) the consideration of the
predicaments insofar as there are things falling under the predicaments
(res praedicamenti) that underlie intentions and insofar as they are things
which can become a subject [in a predication] or can be predicated [of
a subject] (subicibiles vel praedicabiles).
96
The exact meaning of this pas-
sage is not very clear and if we consider the context of this statement
we may be tempted to render it as: . . . insofar as they (i.e., the pre-
dicaments) are predicamental things (res praedicamenti ) . . . Despite the
96
Anonymus Matritensis, Quaest. super lib. Praed., q. 3 , p. 125: Praedicamenta
possunt dupliciter considerari. Uno modo inquantum sunt entia, et sic pertinent ad
metaphysicum, alio modo inquantum sunt res praedicamenti subiectae intentioni et
secundum quod sunt res subicibiles vel praedicabiles . . . Et sic pertinet considerare de
hiis ad dialecticum. The Anonymous then goes on to talk about a third mode of
considering predicaments (namely, as principles of movement and rest), a mode that
is proper to natural philosophy.
218 martin pickav
fact that this breaks up the series of analogous statements regarding
metaphysics and natural science, I chose the translation above because
it is otherwise hard to make sense of the expression res praedicamenti in
the Anonymouss explanation and because the phrase . . . insofar as they
are predicamental things . . . is awkwardly redundant.
97
But whatever
we make of this, the Anonymous clearly hints at an understanding
of concepts (and especially universal concepts) that we have already
encountered earlier. In ordinary concept formation the intellect produces
a concept (or an act of understanding) on the basis of an extra-men-
tal object. There are thus two related items, the extra-mental reality
and the intention in the intellect. In the case of a universal concept,
Simon of Faversham (as we saw earlier) refers to these two items as the
intentio universalitatis and as the (extra-mental) thing underlying such an
intention (res subiecta intentioni).
98
Employing this use of expression we
are now able to say that, according to the Anonymous, logic considers
the categories only insofar as the things falling under the categories
serve as the basis for our concepts and for propositions in which we
predicate one thing of another.
99
This brings us closer to what proper
knowledge of the predicaments is about but it does not lead us all the
way. For if all we can say is that there is a twofold consideration of the
predicaments, then the scientia de praedicamentis should strictly speaking
either belong to metaphysics or to logic (since there is apparently no
room for a third way of considering them). Both alternatives, however,
seem unpromising.
In the passage quoted above, the Anonymous of Madrid employs
the expression res subicibiles et praedicabiles to characterize logics way of
looking at the predicaments. This expression is often used in apposi-
tion to a classic answer concerning the subject-matter of the Categories,
according to which the Categories studies that which is sayable without
a connection and orderable within a genus (dicibile incomplexum ordina-
bile in genere).
100
This leads me to the central passage in which Simon
describes the scientia de praedicamentis:
97
Although this passage might remind the reader of the distinction between res
praedicamenti and ratio praedicamenti it has to be noted that there is no such distinction
in this commentary.
98
Cf. supra p. 194 and fn. 29.
99
This passage also conrms that the Anonymous of Madrid , like Simon , has a
realist understanding of predicaments. Predicaments are extra-mental things.
100
This formula concerning the subject-matter of the Categories goes back at least to
John le Page. Cf. Ebbesen , Les Catgories au moyen ge, pp. 257267.
All things which are considered here [i.e., in the scientia de praedicamentis]
are attributed to one thing, and this is being which is sayable without a
connection and orderable within a genus according to below and above.
Therefore, we here deal with substance and the other predicaments
insofar as in each of them something orderable according to below and
above can be found.
101
Even if Simon does not explicitly refer here to the dicibile incomplexum as
the subject-matter (subiectum) of the scientia de praedicamentis as it is laid
down in Aristotles Categories, he does so elsewhere.
102
The rst thing
to notice is, again, that for Simon the knowledge of predicaments is
about (extra-mental) being. Nothing here forces us to consider the ens
dicibile incomplexum as a linguistic expression or a mental entity.
103
This
denitely rules out that the scientia de praedicamentis is a proper part of
logic. But, as Simon goes on to say, in this scientia the modes of being,
like substance, quality, quantity and so on, are dealt with insofar as in
each of them there is a certain property, namely, insofar as we are able
to nd in them something that can be ordered in a genus according
to below and above and that can serve as the basis of a predication.
Of what kind is this property or these properties by means of which
we create a hierarchical order? Are they mental products? Surely not.
According to Simons view regarding the origin of logical concepts, the
extra-mental objects serve as a basis for both rst-order (rst intentions)
and second-order concepts (second intentions). This leads us to the
conclusion that the scientia de praedicamentis is knowledge of extra-mental
objects, but knowledge that is concerned with those properties of reality
which are the basis of predication and acts of thought. The knowledge
of the predicaments is thus a scientia realis in the strict sense, for it stud-
ies features of reality. However, it does not study being insofar as it is
being; therefore it is not identical with metaphysics. But it studies being
insofar as it is the basis for the acts of the intellect and insofar as it can
be referred to in meaningful speech and predication.
104
101
Quaest. super lib. Praed., q. 1, p. 74: Omnia que hic considerantur, habent attribu-
tionem ad unum, et illud est ens dicibile incomplexum ordinabile in genere secundum
sub et supra. Unde determinatur hic de substantia et de aliis secundum quod in quolibet
eorum invenitur aliquid ordinabile secundum sub et supra.
102
Cf. Quaest. super lib. Porph., prol., p. 14sq. (text quoted supra in fn. 7).
103
This is in line with how other authors understand the meaning of the formula
dicibile incomplexum. Cf. Ebbesen , Les Catgories au moyen ge, p. 259sq.
104
Cf. also Quaest. super lib. Porph., prol., p. 15: . . . liber predicamentorum qui est de
ente ut in eo intellectus fundat intencionem suam subiecti et predicati . . .
220 martin pickav
This allows us now to appreciate better the small yet crucial dif-
ference between the Anonymous and Simon regarding the double
consideration view. If the logician, as the Anonymous believes, were
really examining the res subiectae intentioni, the extra-mental features that
underlie concepts, then the logician would also examine the origin of
logic itself. Simon, however, denies this conclusion: the logician is only
concerned with intentions and not with the nature of things, and an
inquiry into the origins of logic would be an inquiry into the nature
of things.
105
The examination of the res subiectae intentioni is business
of a different discipline, i.e., the scientia de praedicamentis, and for this
reason Simon applies the double consideration view no longer to the
predicaments with regard to metaphysics and logic, but with regard to
metaphysics and this scientia.
For the very same reason for which knowledge of the predicaments
is not strictly speaking part of logic, such knowledge has an intimate
connection to logic. One cannot study those properties which are at
the basis of intellectual apprehension without talking about the fun-
damental acts of understanding and similarly one cannot examine the
extra-mental foundations of predication without studying the funda-
mental concepts of predication such as subject and predicate etc.
106

It is in this respect that the Categories and the knowledge laid down in
it overlap with logic proper, a fact which also explains its inclusion in
the Organon. Moreover, this last line of thought can also be taken as
an explanation why it is nally in vain to ask whether the categories
belong to logic or metaphysics.
Simon of Faversham might generally not have been a very inuential
and original philosopher. Yet, I hope to have shown that his under-
standing of the categories is more original than expected. Maybe his
conception of a scientia de praedicamentis cannot withstand further scrutiny,
but at least he has tried to provide a more sophisticated response to
the notorious question about the content of the Categories than many
of his contemporaries.
105
Cf. Quaest. super lib. Porph., q. 2, p. 19: Qui logicam invenit, logicus non fuit; ex
quo enim consideravit naturas rerum, logicus non fuit, cum logicus, secundum quod
logicus, non considerat de naturis rerum, sed intenciones solas.
106
This is how I understand the quotation supra in fn. 104.
DUNS SCOTUS S ACCOUNT OF A PROPTER QUID
SCIENCE OF THE CATEGORIES
Lloyd A. Newton
Far from detracting from the study of the Categories, the introduction
of the rest of Aristotles writings in the West provided new venues
for studying what was, for many medievals, one of the rst works in
philosophy. Consequently, the literal commentaries that were common
at the beginning of the 13th century quickly gave way to question
commentaries, which allowed philosophers a greater opportunity to
explore issues than was available with a literal commentary.
1
Examples
of those issues included the derivation of the categories; the relation
between logic and metaphysics; whether being is univocal, or the extent
to which there is a science of the categories. It is this last topic that I
am interested in here: whether there is a science of the categories, and
if so, what kind of science it is. By looking specically at this ques-
tion, I hope to show the ways in which philosophical debates typically
progressed in medieval works and the originality of Scotus.
As Martin Pickav previously points out in his article on Simon
Faversham,
2
it is typical for late medieval commentaries on the Categories
to start with a question concerning the scientic nature of the catego-
ries. Although this particular question was not asked by late ancient
philosophers, such as Simplicius , the question most likely developed
out of their consideration of the skopos of the text. More obviously, the
question is formed by topics discussed in the Posterior Analytics regarding
the nature of science, and hence is subsequent to a thorough familiarity
with this latter text. However, it is the science of metaphysics, more
than any other science, that denitively frames what a science of the
1
The exibility available in a question commentary can be seen in both the variety
and nature of the questions raised (virtually none of the questions is designed to help
a beginning student better understand Aristotles text, since a basic mastery of the text
is taken for granted) and in the sheer size of the commentary (e.g., Scotuss question
commentary is over 250 pages in the critical edition, whereas Aristotles short text is
less than 25 pages).
2
Cf. Pickav s article supra, fn. 2.
222 lloyd a. newton
categories will look like.
3
As we will see in the following sections, the
science of the categories is often conceived after, and modeled upon,
the science of metaphysics, especially insofar as both sciences resort to
a unity of attribution when a generic unity is unavailable. Be that as it
may for his contemporaries, Scotus will have none of it. Rather for him,
the science of the categories will have a greater unity, a unity proper
to a propter quid science. It is this last claim, I think, that stands out as
one of the most salient features (among many) that is philosophically
important in Scotuss commentary on the Categories.
4
And even though
Scotus specically denies that being is univocal in this early commen-
taryin contrast to his more (in)famous claim to the contraryit is
quite clear that he is already moving in that direction.
In order to understand and appreciate Scotuss unique understand-
ing of a science of the categories, I will proceed as follows. In the rst
section, I quickly cover the requirements for a science, the difference
between a science and a syllogism, and the two main kinds of science.
5

In the second section, I chart the development of the science of the
categories through three of his contemporaries: Martin of Dacia , Peter
of Auvergne, and Simon of Faversham. In the third section, I indicate
what Scotus takes to be the subject of logic as a whole, and how the
categories are understood to be parts of that generic science. In the
fourth section, I show how Scotus adopts and tweaks the Avicennian
doctrine of the threefold understanding of essence. Finally, in the last
section I demonstrate how Scotus employs these various distinctions to
advance his novel claim that the ten categories are the single subject
of a proper quid science.
3
It is perhaps unfortunate that the Posterior Analytics and the Metaphysics had such an
inuential role in the ways in which medieval philosophers conceived of the Categories
(to the near exclusion of any inuence from the Physics), especially concerning the deri-
vation of the Categories. For had the Physics been more inuential, it would, I suspect,
have been easier to defend a more robust realist ontology, especially insofar as it would
be nearly impossible to conate or reduce quantity to substance.
4
An English translation of Scotuss commentary on the Categories, as well as a dis-
cussion of many of the issues raised in this article, is available in: Lloyd A. Newton ,
Duns Scotuss Questions on the Categories of Aristotle: A Translation of the Whole, Together with a
Philosophical Analysis and Commentary on Questions 18, Dissertation: University of Dallas,
2003.
5
I apologize for the elementarey nature of some of the material presented in this
section. I have included it for two reasons: First, it is helpful to understand the sub-
sequent sections, especially since our notion of science, which is drawn largely from
the empirical disciplines, is radically different from the ancient / medieval view; and
second, while some of it may be fairly common knowledge, I know of no other short
article that summarizes the relation between science and demonstration.
scotuss PROPTER QUID science of the categories 223
1. The Nature of Science
By the time Scotus writes his Categories commentary, the sciences are
typically divided into two broad classes: practical and speculative.
6

The practical sciences are then subdivided into: A) moral philosophy,
which is further divided into 1) ethics, 2) economics, and 3) politics;
and B) mechanics. In a similar way, the speculative or theoretical sci-
ences are divided into: A) a science of the real, which is subdivided into
1) natural science (physics), 2) mathematics, and 3) metaphysics; and
B) a science of reason, i.e., logic.
7
Furthermore, logic can be considered
in two ways: as it is applied (logica utens), and as it is pure (logica docens).
Like many of his contemporaries, Scotus thinks that only in the latter
way is logic truly a science in itself.
It must be understood that logic is considered in two ways. In one way
insofar as it is pure [docens], and in this way it proceeds from necessary
and proper principles to necessary conclusions, and thus it is a science.
In another way, insofar as logic is used, by being applied to those things
in which it is used, and in this way it is not from proper [principles] but
from common [principles], nor is it a science.
8
6
There are several different ways to locate the study of logic within the various
sciences. For example, depending on whether that which is taken into account is the
differences of logic from all other sciences or the similarities of logic to all the other
sciences. The division provided here is based on a consideration of logic as one of the
speculative sciences, which is how Scotus portrays it in his commentaries on Porphyry s
Isagoge and Aristotles Categories. Although a bit dated, Robert Schmidts work on logic
is still very informative. See in particular: Robert W. Schmidt , The Domain of Logic (The
Hague, 1966), pp. 2631. For a more recent discussion, see Marmos article on Rhetoric
(Costantino Marmo, Suspicio: A Key Word to the Signicance of Aristotles Rhetoric
in Thirteenth Century Scholasticism, CIMAGL 60 (1990), 145198, pp. 148152.) At
the beginning of the thirteenth century, according to Marmo, at least four alternative
general divisions of the sciences were advocated by different philosophers.
7
Noticeably absent in this division, as well as in all the classications provided by
Schmidt , is a science of grammar. The rise of a science of grammar, the so-called
Logic der Midistae, which was inuenced by the reception of the Arabic division of
logic, is discussed by Marmo (ibid., p. 158) and by Pinborg. Cf. Jan Pinborg , Die Logik
der Modistae, Studia Mediewistyczne 16 (1975), 3997; Costantino Marmo, A Pragmatic
Approach to Language in Modism, in Sprachtheorien in Sptantike und Mittelalter, ed. Sten
Ebbesen (Tbingen, 1995), pp. 16983; Costantino Marmo, The Semantics of the
Modistae, in Medieval Analyses in Language and Cognition, Acts of the Symposium, The
Copenhagen School of Medieval Philosophy, January 1013, 1996, eds. Sten Ebbesen
and Russell L. Friedman (Copenhagen, 1999), pp. 83104.
8
Quaestiones in librum Porphyrii Isagoge, eds. Robert Andrews et al. (St. Bonaventure,
1999) q. 1, n. 7: Intellegendum tamen quod logica dupliciter consideratur. Uno modo
in quantum est docens, et sic ex necessariis et propriis principiis procedit ad necessarias
conclusiones, et est scientia. Alio modo in quantum utimur logica, applicando ad illa
in quibus est usus, et sic non est ex propriis sed ex communibus, nec est scientia.
224 lloyd a. newton
Insofar as it is pure, Scotus adds:
logic is a science since those things which are taught in it are concluded
by a demonstration, just as in other sciences; therefore they are known
since a demonstration is a syllogism that makes one to know.
9
Thus, logic is both a tool used in the other sciences and something
known in itself. In the former way, it is an art, in the latter, it is one
of the sciences. Before we look at the science of logic in particular, it
may be helpful to dene each of these technical terms.
In English, the word science does not have the same associations as
its Latin cognate, scientia.
10
In Latin, scientia is an abstract noun derived
from the word scio, scire, to know. Like its Latin cognate, science is
an abstract noun; in English, its best synonym is the word knowledge,
an abstract noun derived from to know. In its broadest sense, to know
something is to have a knowledge of something that could not be oth-
erwise. In this sense, knowledge includes both scientia and intuitive
cognition of something.
11
In its narrower sense, to know something, to
have scientia of something, is to have a certain cognition of that thing by
means of a syllogism. Accordingly, in its narrower sense, science is the
knowledge of something derived from self-evident principles, whereas
the self-evident principles are not said to be known demonstratively
but intuitively. Thus, knowledge, in its narrower sense, is a mediated
cognition of something that could not be otherwise, as opposed to an
immediate cognition of it.
12
More importantly, one needs to distinguish
between science as the certain cognition of something and a syllogism,
which is a mode of acquiring knowledge or science. In either sense of
9
Ibid., n. 6: Dicendum quod logica est scientia, quia quae docentur in ea dem-
onstratione concluduntur sicut in aliis scientiis; igitur sciuntur, quia demonstratio est
syllogismus faciens scire. Sunt etiam in logica omnia requisita ad demonstrationem: ut
subiectum, passio demonstrabilis de subiecto per medium, quod est denition.
10
From here on out, I will be using the terms science and knowledge in the
Aristotelian / pre-Cartesian sense of the terms, which include an understanding of
all four causes. This conception of knowledge or science is in sharp contrast with the
typical post-Cartesian conception derived from the empirical disciplines.
11
I am using the term intuitive cognition in the broad, non-technical sense for the
understading of any self-evident statement (such as Equals added to equals are equal)
and not in the more technical sense, which is in contrast to abstractive cognition, and
for which Scotus is famous.
12
For a more analytic treatment of the difference between mediate and immediate
cognition, see Scott MacDonald, Theory of Knowledge, in The Cambridge Companion
to Aquinas, eds. Noman Kretzmann and Eleonore Stump (Cambridge, 1993), pp.
160195, pp. 168174.
the word, the opposite of knowledge is either opinion or belief, both
of which lack certainty.
The fact that knowledge is a mediated cognition of something raises
the question of the way in which it is mediated, which is through a
syllogism, or deductive reasoning.
13
A syllogism is dened as a set of
statements, one of which, the conclusion, follows of necessity from the
other two, which are the premises. Based on Aristotles Posterior Analytics,
one can also dene a statement as a complex enunciation, which is
resolved into two simple or non-complex terms, i.e., the subject and
the predicate.
14
Finally, every categorical syllogism consists of only three
terms:
15
A) a minor term, which is the subject of the conclusion; B) a
major term, which is the predicate of the conclusion; and C) a middle
term, which connects the minor to the major term. A commonplace
example of a rst gure syllogism is:
All living things are mortal. (major premise)
All men are living things . (minor premise)
Therefore, all men are mortal . (conclusion)
What we are interested in here is one specic type of deductive argu-
ment: a demonstration, which is distinguished from other types of
syllogisms by the fact that its middle term denes its subject. For by
means of a demonstration, one has certain knowledge of something that
could not be otherwise, which knowledge is obtained through knowing
the essence of that thing. As Aristotle states it, By demonstration I
mean a scientic deduction; and by scientic I mean one in virtue of
which, by having it, we understand something.
16
In his commentary
on Porphyrys Isagoge , Scotus adds that logic is a science since those
things which are taught in it are concluded by a demonstration. . . . [for]
in logic, there are also all the requisites for a demonstration: a subject
and attributes that are demonstrated of the subject through a middle
term, which is a denition.
17
As one can see from this last statement,
13
The word syllogism is etymologically derived from the Greek word .
It is synonymous with deductive argument, which is derived etymologically from the
Latin word deductus.
14
Cf. Posterior Analytics, I, ch. 1, 24b18.
15
A categorical syllogism is one that consists of simple declarative statements, as
opposed to hypothetical or disjunctive statements.
16
Posterior Analytics, I, ch. 2, 71b 1819.
17
Quaestiones in librum Porphyrii Isagoge, q. 1, n. 6: Dicendum quod logica est scientia,
quia quae docentur in ea demonstratione concluduntur sicut in aliis scientiis; igitur
sciuntur, quia demonstratio est syllogismus faciens scire. Sunt etiam in logica omnia
226 lloyd a. newton
Scotus mentions three elements required for a demonstration: 1) a sub-
ject; 2) attributes; and 3) these attributes being demonstrated through
a middle term, which is a denition of the subject. Let us briey look
at each of these three elements.
The subject of a demonstration is the minor term. It is also that
of which the proper attributes, or properties, are predicated. In fact,
it is often said that properties are predicated of a subject in the same
way as the nine accidents inhere in a substance. The second element,
on the other hand, is an attribute, or affection ( passiones), caused by
the subject. Due to the causal relationship between the subject and
the attribute, there is also a corresponding necessity for the inherence
of a property. As Schmidt describes it, A property necessarily follows
from the nature or essence of a being and is caused by the nature or
essence. But science is knowledge from causes. Hence a property will
be shown to belong to a subject by expressing its cause, the essence of
the subject.
18
The third element of a demonstrative syllogism is that
the middle term must be a denition of the subject. Again, Schmidt
describes it this way: The middle term in this syllogism must therefore
be the denition of the subject. Since in the conclusion the property
will be predicated of the subject, the property will be the major term
and the subject will be the minor term.
19
However, to say that the
middle term is a denition of the subject does not necessarily mean
that the middle term must consist of a genus and a differentia. Rather,
the middle term may be taken from any one of the four causes, even
though it is generally taken from the formal cause.
20
As we will see
later, the middle term may even be taken from an effect nearer and
better known to us. To return to the previous example, the subject of
the conclusion, men , is the subject of the argument. The predicate
of the conclusion, mortal, is the attribute demonstrated of the subject
by a middle term dening the subject.
requisita ad demonstrationem: ut suiectum, passio demonstrabilis de subiecto per
medium, quod est denitio.
18
Schmidt , The Domain of Logic, p. 267.
19
Ibid., p. 268.
20
Cf. Paul Symingtons article on Aquinas, supra, for a discussion of the three dif-
ferent types of essential or per se predication and how those three ways are related to
the ten categories.
A demonstration, though, is not the same thing as science. Science
is the knowledge of something, whereas a demonstration is the means
of acquiring knowledge. Nevertheless, there are similarities between
knowledge or scientia on the one hand and demonstration on the other.
Knowledge also has three requirements that parallel the three require-
ments of a syllogism. These requirements are: 1) that which is known,
2) that from which it is known, and, 3) that about which it is known,
which is the subject of the science.
21
The rst requirement, that which
is known, is composed of a subject and a predicate.
22
The predicate,
Schmidt adds, stands for a proper accident (id quod per se inest) or
property ( propria passio) of the subject.
23
However, since that which is
known is principally the attribute or property, this rst requirement of a
science is parallel to the major term of a syllogism. For the major term
of a syllogism is a property demonstrated to inhere in its subject.
The second requirement for a science is that it have axioms or com-
mon principles, a requirement similar to the middle term of a syllogism.
The difference between the two is that an axiom or common principle
includes denitions as well as the general laws of logic, whereas a
middle term is, strictly speaking, only a denition.
24
For example, every
science is based on the law of non-contradiction, the law of identity,
and the law of excluded middle. However, these laws are neither proper
to a demonstration nor are they denitions. Rather, a demonstration
presupposes these general laws. One example, taken from mathemat-
ics, is that geometry requires certain common notions, such as equals
added to equals produce equals, as well as axioms or postulates, such
as all right angles are equal to one another. The former are common
21
For more on these three requirements, see Schmidt , The Domain of Logic, pp.
1012.
22
Ibid., p. 11.
23
Ibid.
24
Cf. the following two syllogisms taken from Euclid , The Thirteen Books of the Elements,
ed. Thomas L. Heath, 2nd ed. (New York, 1956), bk. I, prop. 1: The radii of a circle
are equal to one another, line A and line B are the radii of a circle, therefore, line
A and line B are equal to one another. Things equal to one and the same thing are
equal to one another, line A and line B are equal to one and the same thing, therefore
line A and line B are equal to one another. The former syllogism is a demonstration,
since the two lines are dened as the radii of a circle, whereas in the latter syllogism
the subject is not dened. There is a parallel, though, between the two insofar as the
second example uses the rst common notion, which is an axiom, as the predicate
of the middle term, whereas the former uses the middle term as a denition of the
minor term.
228 lloyd a. newton
to all the mathematical sciences, whereas the latter are proper to the
science of geometry.
Most importantly, the third requirement for a science is that it have
an appropriate subject.
25
But, what exactly is the subject of a science?
Unfortunately, the term is rather ambiguous. In a broad sense, the
subject of a science is whatever enters into the consideration of the
science.
26
In this sense, the subject of a science includes the axioms
and properties demonstrated of the subject. In a more restricted sense,
however, it signies what the science principally considers. It is this
latter sense that interests us here.
Because a science is the knowledge of something by means of a
demonstration, the subject of a science is often identical to the subject
of a demonstration. That the subject of a science need not always
be identical to the subject of a demonstration is so only because a
subject of one demonstration may well be the predicate of a previous
conclusion. As a result, a property or attribute demonstrated of one
subject may be a cause or subject of another attribute of a subsequent
demonstration. For example, mortality, the property demonstrated of
men above, is the subject of the following argument:
To be in need of a redeemer is to be in need of Christ. (major premise)
Mortals are in need of a redeemer. (minor premise)
Therefore, mortals are in need of Christ. (conclusion)
This example illustrates that the mortality concluding the previ-
ous argument is now used as the subject of another demonstration.
Consequently, in a science one could connect the two arguments to
show that men are in need of Christ. This kind of logical progression
is best seen in geometry, in which one proposition after another forms
the premises in subsequent demonstrations. Aquinas explains the rea-
son for this logical progression as follows: But in those sciences which
bear upon accidents, nothing prevents a same thing from being taken
as a subject in reference to one proper attribute, and as an attribute in
reference to a more basic subject.
27
A property, then, is essentially an
effect that is itself the cause of something else.
25
The subject of a science for Scotus is what modern terminology often refers to
as the object of a science.
26
Schmidt , The Domain of Logic, p. 14.
27
Thomas Aquinas , Commentary on the Posterior Analytics of Aristotle, trans. Larcher, I,
lect. 2 (New York, 1970).
Although the property of one demonstration may be the cause of
a subsequent demonstration, and vice versa, nevertheless, there must
be some one subject that is itself not caused by a previous argument.
For just as there cannot be an innite regress of causes in the natural
order, so there cannot be an innite regress of subjects in the order
of predications. Rather, some subject must be the ultimate cause of
all the other attributes in subsequent demonstrations. Nevertheless,
Aquinas points out, this [progression] must not develop into an innite
process, for one must arrive at something which is rst in that science
and which is taken as a subject in such a way that it is never taken as
a proper attribute, as is clear in the mathematical sciences, which treat
of continuous or discrete quantity. For in these sciences those things are
postulated which are rst in the genus of quantity; for example, unit and
line and surface and the like.
28
Thus, the primary subject of a science
is, so to speak, essentially the rst uncaused cause of that science.
What is the ultimate or rst subject of a science? It is the rst or pri-
mary subject to which all subsequent demonstrations are nally reduced.
In his commentary on Porphyry s Isagoge, Scotus adds that there are
three requirements for a primary subject of a science: 1) the existence
and the nature of the subject must be better known than the attributes;
2) the attributes must be demonstrated through the nature or essence
of the subject in that science; and 3) all the things demonstrated in that
science must be reducible to that subject as to a rst principle, which is
required for the unity of the science.
29
In his work on medieval logic,
Schmidt describes it this way: It is the subject about which knowledge
is sought. And though we come to understand a thing by knowing its
causes, and thus in a science seek causes, still it is not the causes which
we are seeking to know, but the subject of our science.
30
The primary subject of a science, then, is that which species the
science and gives it its true nature. Furthermore, the subject of a sci-
ence is what distinguishes one science from another science, since the
28
Ibid.
29
Quaestiones in librum Porphyrii Isagoge, q. 3, n. 13: Ad videndum quae positio sit verior,
notandum quod tres condiciones sunt requisitae subiecti in scientia, videlicet quod sit
notum quid est et quia est, nam haec duo oportet praesupponere demonstrationi,
ut dicitur I Posteriorum. Secunda est quod per eius quid est demonstrentur in scientia
passiones illius subiecti de eo. Tertia est quod ad ipsum omnia alia determinata in
scientia reducantur et propter ipsum considerentur. Aliter enim ab eius unitate non
esset unitas scientiae.
30
Schmidt , The Domain of Logic, pp. 1112.
230 lloyd a. newton
subject of a science stands in the same relation to a science as the
formal object of a power or habit stands in relation to that power or
habit. For example, although the power of sight has, when considered
materially, various objects, such as a stone, a person, or a house, each
of these objects has only one specic form that is the proper object of
sight, namely, color. Likewise, the proper object of hearing is the formal
object of sound, regardless of the material diversity of the objects that
produce sounds. Finally, the formal object of the habit of moderation
is easily obtainable pleasure, even though, considered materially, there
may be many pleasurable objects that are easily obtainable. These
examples illustrate that just as the unity of a power or habit is taken
from its formal object, so the unity of a science depends on the unity
of the subject: a science is one to the extent to which its subject is
one. As we will see in the next section, the requirement for a unied
subject presents a special obstacle to the claim that the categories are
the subject of a science.
These three requirements of a science may also be illustrated with a
couple of examples taken from the science of geometry. A geometer can
demonstrate that the interior angles of a triangle are equal to two right
angles. In this example, being equal to two right angles is an attribute
of the interior angles of a triangle. However, in a given demonstration
the subject is a particular gure, often referred to simply as ABC, which
is then dened by the middle term as a triangle.
31
In this demonstra-
tion, the geometer makes use of other propositions, some of which in
turn make use of common notions and postulates. He also makes use
of the notion triangle, which, in an earlier demonstration, is shown to
be constructed such that its three angles are equal to a given straight
line. In this latter case, the existence of the triangle is said to be the
attribute and the straight line the subject.
32
Again, in this instance, the
subject is a particular gure, referred to simply as AB, which is then
dened by the middle term as a straight line. In these two examples,
though, neither the angles of all triangles nor straight lines are the
ultimate subject of geometry. Rather, the ultimate subject of geometry,
as a science, is continuous quantity, to which both the straight lines
and the angles of triangles are nally reduced.
33
31
For an example of this demonstration, see Euclid , The Thirteen Books of the Elements,
bk. 1, prop. 32.
32
For a fuller discussion of this point, see Thomas Aquinas , Commentary on the Posterior
Analytics of Aristotle, bk. I, lect. 2
33
Ibid.
Having claried the differences between science and demonstra-
tion, there is one last distinction that needs to be made: that between
a demonstration of the fact (quia) and a demonstration of the reasoned
fact ( propter quid). The latter proceeds from a knowledge of causes to a
knowledge of their effect, whereas the former, (quia), or demonstration
of the fact, proceeds from a knowledge of effects to a knowledge of
their causes.
34
In a demonstration quia, the middle term is still said to
dene the subject since it is that which is better known to us. Aquinas
explains the reason for this possibility: in the case of two things equally
predicable, i.e., convertible, one of which is the cause and the other
the effect, there is nothing to preclude that now and then the better
known will not be the cause but the effect.
35
In a demonstration quia,
the attribute is not said to be an effect caused by the subject but rather a
cause known by the effect. Aristotle himself provides the classic examples
illustrating the differences between scientia quia and scientia propter quid.
36

Here are the two examples in syllogistic form:
37
scientia propter quid scientia quia
Near (things) are non-twinkling. Non-twinkling (things) are near.
The planets are near (things) (cause). The planets are non-twinkling (effect)
The planets are non-twinkling (effect). The planets are near (cause)
As one can see, in the rst example the fact that planets are non-
twinkling is demonstrated by their being near, whereas in the second
example, the fact that planets are near is explained from the fact that
they are non-twinking. Obviously, the nearness of the planets is not
caused by their failure to twinkle but rather their failure to twinkle is
caused by their nearness. These two types of demonstration correspond
to two types of science: scientia propter quid and scientia quia. As we will
see in the next two sections, Scotus and his contemporaries disagree
as to which type of science pertains to the categories.
34
Technically speaking, scientia quia is also any knowledge that proceeds from mediate
principles as opposed to proceeding from immediate principles. See Thomas Aquinas ,
Commentary on the Posterior Analytics of Aristotle, bk., I, lect. 2.
35
Ibid.
36
Posterior Analytics, I, ch. 13, 78a 305.
37
For the sake of clarity, I have placed these in the form of Barbara. Two things
should be kept in mind: First, all the statements are to be understood as afrmative
declarations, even when the predicate is negative, i.e., non-twinking is the predicate.
Second, the attribute of non-twinkling must be construed as a property of nearness,
such that they have the same extension and hence are convertible.
232 lloyd a. newton
2. The Science of the Categories According to Scotus s Contemporaries
By the time Scotus composes his Categories commentary, logic is typi-
cally subdivided into three different parts or sciences, each with its own
specic subject, corresponding to the three acts of the mind: 1) the act
of simple apprehension, which grasps simple concepts, corresponds to
the ten categories; 2) the act of judgment, which combines two simple
concepts into a judgment, corresponds to statements; and 3) the act of
reasoning, from which one proceeds from two known statements to a
third unknown statement, corresponds to the syllogism. The following
examples, taken from Scotuss commentary on the Categories, exemplify
a demonstration corresponding to each of the three acts of the mind:
First, every term, which is grasped by simple apprehension, is orderable
within a genus. Animal is such a term, therefore, animal is orderable in
a genus. Second, every statement, which is a product of judgment, is
either true or false. Sentence A is such a statement, therefore, sentence
A is either true or false. Third, an invalid syllogism, which is produced
by the act of reasoning, does not produce a necessarily true conclusion.
That particular syllogism is invalid, therefore that particular syllogism
does not produce a necessarily true conclusion.
As one can see from these instances, attributes are demonstrated of
the subject term by means of a rst gure syllogism. More importantly,
the attributes only pertain to things insofar as those things are acts of the
mind.
38
That is, being orderable in a genus, true or false, or invalid
does not pertain to either things considered as individuals or to common
natures, which are neither individual nor universal. Since this chapter is
principally about Scotuss commentary on the Categories, let us focus on
the science of the categories to the exclusion of the other two sciences,
viz., a science of judgments and a science of reasoning.
As previously mentioned, most medieval commentaries on the
Categories raise a question or two concerning whether or not a science
of the categories is possible, to which manyMartin of Dacia , Peter of
Auvergne , Simon Faversham, Radulphus Brito , and Duns Scotus all
respond afrmatively. As we will see, Scotuss position is quite unusual
among these others in several ways.
39
To appreciate the uniqueness of
38
I will return to this point at the end of this chapter.
39
Ockham is unique for a completely different reason, namely insofar as he denies
that there is a single subject of the categories. Cf. William of Okcham, Expositio in
librum Praedicamentorum Aristotelis, in Guillelmi de Ockham Opera philosophica et theologica, Opera
his position, we will rst look at the different conceptions of the science
of the categories put forth by three of his contemporaries: Martin of
Dacia, Peter of Auvergne, and Simon of Faversham.
Martin of Dacia composed his commentary on the Categories some-
time between 1270 and 1275.
40
Although he does not specically ask
whether the categories are the subject of a science, he does ask two
related questions. The rst is whether being is the subject of the catego-
ries. The rst of two arguments for the negative is that the subject of a
logical science is a second intention. However, based on the authority
of Avicenna , second intentions are not real things, whereas being (ens)
is a real thing, not an intention. The conclusion is that the subject of
this work is not being. The second argument is that a subject of a sci-
ence must be univocal; being is not univocal; therefore, being is not the
subject. The one opposing argument is that for something to be the
subject of a science, it must exist. In his solution, Martin concedes to
the arguments for the negative, namely, that being is not the subject of
the categories since it is not predicated univocally of them. In response
to the opposing argument, he points out that being can be considered
in two ways: 1) insofar as it is common and intentional, which is
apprehended by the intellect; and 2) insofar as it is specic and real,
positing something in one of the ten categories. In the latter way, so
Martin concedes, being is not the subject of the categories. However,
he contends that being is the subject of a science in the former way.
Martins second question is whether the subject of the categories is
an incomplex, univocal entity that is sayable and orderable within a
genus (ens incomplexum univocum dicibile ordinabile in genere). Although Martin
does not spell out what this phrase means, Andrews notes that it is a
version of a traditional summarizing expression.
41
Other commenta-
tors sometimes drop the word univocum, so that the whole expression is
simply ens incomplexum dicibile ordinabile genere. According to Lewry , John
Pagus , Nicholas of Paris , and Gerard Nogent rst formulated this phrase
philosophica II, ed. Gedeon Gl (St. Bonaventure, 1978), Prooemium: Unde suppositis
quibusdam communibus omnibus libris philosophiae in libro Porphyrii determinatis
sicut nec aliorum librorum, ita nec scientia praedicamentorum est unus habitus numero;
nec habet unum subiectum, sed multa; nec habet plures causas essentiales quam duas,
proprie loquendo de causis.
40
Martins commentary on the Categories is in Martin of Dacia , Opera omnia, ed.
Roos, pp. 153231, and his rst two questions are on pp. 1556.
41
Robert Andrews, Peter of Auvergnes Commentary, p. 14.
234 lloyd a. newton
to account for the material cause of a science of the categories.
42
The
phrase is based on the standard position, rst articulated by Porphyry ,
that the Categories is about the ten utterances signifying the ten genera.
It entails the following: rst, ens, or being, signies that the subject
is something that exists as opposed to being a gment of someones
mind. Furthermore, being is posited because the rst requirement for
a science is that the subject must exist prior to, and be better known
than, the attributes demonstrated of it. The second word, incomplexum
signies that the subject is the ten categories which are apprehended by
the rst act of the intellect. The third word, dicibile, is said to reect
Boethius description of logic as words signifying things.
43
However,
dicibile also implies that what is signied is predicable of others, thus
this word is added to the phrase to exclude proper names which are
not predicable of others.
44
Finally, the words, ordinabile in genere signify
the fact that the categories, taken as the whole hierarchy of the ten
various kinds of things as opposed to simply the ten highest genera, are
ordered to one another as above and below within those most common
genera. For example, the word animal is ordered to the word living
above it and to the word man below it.
To return to the question of the proper subject of the categories,
Martin provides only one dubious argument on each side of the ques-
tion. His solution is equally dubious: the subject of a logical science, he
says, ought to be a logical thing. An incomplex expression is a logical
thing, therefore, an incomplex expression is the subject of this science.
45

As one can see, Martins two questions concerning the science of the
categories leave much to be desired. Taken together, he offers only
three arguments for the negative and two arguments for the afrma-
tive. His responses leave even more to be desired. Overall, he indicates
no awareness in his commentary concerning the various requirements
of a science nor how those requirements pose specic problems for a
science of the categories.
42
P. O. Lewry , Robert Kilwardbys Writings on the Logica Vetus, Studied with
Regard to their Teaching and Method, (Unpublished D.Phil. thesis, University of
Oxford, 1978), p. 91.
43
Andrews , Peter of Auvergne s Commentary on Aristotles Categories, p. 14.
44
De Rijk , Logica Modernorum, vol. 2, part I, p. 470.
45
I follow Martin of Dacia s lead in reducing this phrase simply to dicibile incom-
plexum.
Within a few years of Martins commentary on the Categories, Peter
of Auvergne composed his own commentary (circa 1275).
46
Unlike
Martin, though, Peter is aware of two key difculties concerning the
claim that there is a science of the categories, and he raises them in the
rst two questions. The rst question is whether there can be a science
of the categories. He gives only one argument for the negative, viz.,
that the categories do not have a cause, and therefore a science of them
does not exist. His opposing argument is an appeal to the authority of
Aristotle. In his solution to the question, Peter argues that even though
the categories do not have a cause, there is still a science of them, just
not an unqualied demonstration based on causality. Rather, the sci-
ence of the categories, according to Peter, is only a certain cognition
of them based on their denitions and descriptions.
The second question he raises is whether there can be one science
of the categories. Again, the one argument for the negative is that the
categories are not one but many, so there cannot be one science of
them. The reason supporting this argument, as we saw earlier, is that
in order for a science to be one, its subject must also be one. However,
since there is nothing common to the ten categories, they cannot be the
subject of one science. The opposing argument claims that the nine
accidents are one insofar as they are all attributed to substance, and
therefore some sort of unity inheres in them. In his own personal solu-
tion to the argument, Peter adopts the main argument to the opposite
with a slight modication. According to the argument to the opposite,
the categories are the subject of one science since they are all attributed
to substance. According to Peters own solution, the categories are the
subject of one science since they are all attributed to a single subject,
although not to substance. What is this subject to which all the categories
are attributed? Peter says it is an incomplex entity orderable in a genus
(ens incomplexum ordinabile in genere), a solution essentially the same as the
one adopted by Martin of Dacia . For reasons not explained, though,
Peter drops the words dicibile and univocum from the phrase.
Peters solutions present three important elements. First, he does not
think that the categories are the subject of an unqualied demonstration,
but simply says the categories are the subject of a descriptive science.
Second, the unity of the categories is a unity based on attribution to a
46
Peter of Auvergne s commentary is translated as part of Andrews dissertation.
See Andrews, Peter of Auvergnes Commentary on Aristotles Categories, pp. 46970.
236 lloyd a. newton
rst subject, which is an incomplex entity orderable in a genus. Finally,
although he does not use the word analogy, Peters argument is similar
to the way in which being is said to be predicated analogously of the
ten categories in the science of metaphysics.
47
The difference is that in
the latter science, the ten categories are each attributed to a primary
substance, whereas here they are each attributed to an incomplex entity
which exists only objectively, not really. Nevertheless, it is important to
note that the science of the categories is modeled on, and parallels,
the science of metaphysics. That is, even though the two sciences may
be really distinct from one another, the two are isomorphic to one
another.
Although Martin and Peter ask different questions, their basic posi-
tions do not essentially differ. Both agree that there is a science of the
categories whose subject is either an incomplex entity orderable in a
genus, or an incomplex univocal entity, sayable and orderable in a
genus. Overall, though, Peters arguments are more developed since he
is at least aware of the fact that a science must have a cause and that
its subject must have an appropriate unity, items which Martin either
is unaware of or overlooks. Finally, Peters solution to these arguments
has the effect of making the science of the categories similar to the
science of metaphysics, whereas Martins basic position seems to keep
the two sciences more separated.
Within a decade or so of Peters and Martins commentaries, Simon
of Faversham composes his own commentary on the Categories (circa
1280).
48
Although Simon raises only one question concerning whether
a science of the categories exists, that one question shows a marked
development in the awareness of the various problems associated with
a science of the categories. To the question: whether there can be a sci-
ence of the categories?, there are two arguments to the negative which
claim that the categories are not the subject of a science. We have seen
the rst argument before in Peters commentary, viz., a science must
have a cause. The second argument, however, is new: every science is
had through a demonstration in which the middle term is a denition.
Since the categories are the highest genera, they do not have a denition,
for every denition is through a genus and a differentia. The conclu-
47
Ibid., pp. 1218.
48
Simon of Faversham, Quaestiones super librum Praedicamentorum, in Opera omnia,
vol. I, ed. P. Mazzarella (Padova, 1957), q. 1, pp. 2637.
sion is that there cannot be a science of the categories. Like Peters
and Martins commentaries, Simon only gives one opposing argument.
Unlike Peters argument, though, his opposing argument does not simply
appeal to the authority of Aristotle but actually indicates some original
thinking on the issue. According to this argument, there is a science
of everything that can be done correctly or incorrectly. Things within
a category can be ordered correctly or incorrectly when compared to
those things above and below them. Therefore, there must be a science
of the categories. Whatever its strengths may be, it is certainly better
than Martins claim that the subject of a logical science ought to be
a logical thing.
More important still, Simon provides eight separate arguments on
behalf of a science of the categories. Out of the eight separate argu-
ments, four things stand out in his solution to this initial question.
First, he is aware that a science must meet three conditions: 1) its
subject must exist; 2) it must be intelligible; and 3) it must have constitu-
tive parts. These parts, Simon says, are properties demonstrated of the
subject.
49
Not surprisingly, Simon thinks that the categories meet these
three requirements. Regarding the rst requirement, Simon emphati-
cally argues that every category is a true being that exists outside the
soul, ens verum extra animam est .
50
By true being, Simon refers to the
categories as they are studied in metaphysics, which he calls divine sci-
ence. In contrast, those things whose being depends on the soul and
whose mode of being are logical intentions, such as genus and species,
he calls beings, entia, not being, ens. Nevertheless, whether they exist
outside the soul or depend on the soul, the fact remains that they are,
i.e., they exist in some sense.
Regarding the second requirement, i.e., intelligibility, Simon maintains
that the categories are intelligible because genus and species are posited
in the categories. The intellect, however, does not base the intention of
49
Simon is not too clear on the three conditions required for a science. The Latin
is: Ad hoc enim quod de aliquo possit esse scientia tria requiruntur, scilicet quod
ipsum sit ens et quod ipsum sit intelligibile et quod habeat partes constituentes ipsum,
proprietates et passiones que possunt probari de hoc q. 1, p., 2634. Based on the
construction of the Latin, it seems that constituent parts are synonymous with proper-
ties and attributes. However, in the arguments that follow this paragraph, it seems as if
there are four requirements for a science: 1) being; 2) intelligibility; 3) constituent parts,
which he identies with a denition; and 4) properties and attributes. In his previous
article, Martin Pickav takes parts and denitions to be identical. See, supra, p. 265.
50
Simon of Faversham, Quaestiones super librum Praedicamentorum, q. 1, p. 264.
238 lloyd a. newton
genus and species on something unless it is intelligible; therefore, the
categories must be intelligible.
Regarding the last requirement, that it have parts, Simon claims that
the categories have properties attributed to them. The proof of this
claim is that Aristotle gives various properties of substance, quantity,
quality, etc. What are these properties? In a subsequent argument,
Simon gives two examples taken from Aristotles Categories pertaining to
the category substance: 1) it receives more and less; and 2) nothing
is contrary to substance.
51
Second, Simon next points out that since a science is a habit based
on a demonstration, there are as many types of sciences as there are
types of demonstration. According to Aristotle, there are two types of
demonstration: one that reasons from effects to causes, and one that
reasons from causes to effects. As Simon states, science is twofold: one
that proceeds from things that are nearer and prior in themselves, and
another which proceeds from things nearer and prior with respect to
us.
52
The former, which is an unqualied demonstration, is not the
kind of science that pertains to the categories, for that kind of science
requires a denition constituted by a genus and differentia, and the cat-
egories, as the highest genera, do not have a higher genus. Accordingly,
the science of the categories is a science of the fact, a scientia quia, not
a science of the reasoned fact, a scientia propter quid. Proof of this claim
is that we do not know the essences of the categories in themselves.
Rather, what we know about the categories are their properties, which
are nearer to us.
53
51
Contrary to Simons claim, however, it would seem that these two properties of
substance pertain to the category of substance as it exists extra animam, not insfar as
it is a being of reason.
52
Simon of Faversham, Quaestiones super librum Praedicamentorum, q. 1, p. 265: Similiter
duplex est scientia, una que procedit ex prioribus et nocioribus simpliciter, et alia
ex nocioribus et prioribus quoad nos, et nociora nobis quandoque sunt ignociora et
posteriora quoad nos.
53
The importance of this claim, and the extent to which Scotus account differs
from Simon s, is highlighted by Pini , who states that: In the Categories, much space is
devoted to the properties pertaining to categories, but not much else is said concerning
each category. Why does Aristotle consider the properties pertaining to categories in
more detail than he considers categories themselves? Simon of Faversham maintains
that the science of categories deals with the properties of categories because their
essence is unknown to us. Consequently, in order to know something about categories,
the only viable option is to study their properties. Scotus, however, does not agree with
this position, for he maintains that the logical consideration of categories has nothing to
do with the study of their essence. The fact that the essence of categories is unknown
to us does not explain why the Categories studies the properties of categories more than
Third, Simon maintains that each of the ten categories consists of
two parts: a thing, or essence, and a mode of being.
54
The essence
of a category is studied in metaphysics, or, as he prefers to call it,
divine science. A categorys mode of being, which is either identical
to or is the basis of its mode of predicating, is studied in logic.
55
If a
categorys mode of being is not to be in another, then that category is
a substance. If a categorys mode of being is to be in another, then it
is an accident. Not surprisingly, in his twelfth question Simon thinks
that he can deduce the ten categories from the different ways in which
the categories are predicated of something.
56
Finally, Like Peter of Auvergne , Simon also thinks that the unity
of the science of the categories is a unity of attribution. In the divine
science, where the essences of categories are studied as things in them-
selves, the categories are attributed to God, not simply to substance. In
logic, where the categories are not studied as things in themselves but
in light of their mode of predicating, all things are attributed to one
subject. That subject, as one might suspect, is an incomplex entity say-
able and orderable within a genus. As Simon states, all things which
are considered here have an attribution to one thing, and that is an
incomplex being that is sayable and orderable in a genus according to
categories themselves. Pini, Categories and Logic in Duns Scotus, p. 163. In the preceding
paragraph, Pini also points out that Scotus thinks that those properties [of the indi-
vidual categories] are intentional whereas others maintain that they are real.
54
The belief that a category consists of a thing and its mode is shared by Thomas
Aquinas and Henry of Ghent as well. For a detailed discussion of Simon s view on
the categories, see Pini , Categories and Logic in Duns Scotus , pp. 1524. Scotus, on the
contrary, does not think that a category consists of a thing and a mode. As Pini points
out, According to Scotus, categories are not constituted by two metaphysical aspects.
Instead, categories, metaphysically speaking, are simple things and essences. Everything
is identied as something thanks to its essence, which is an essence of a certain type.
Scotus seems to maintain that it is not possible to divide a thing from its mode, and
this may be the reason for Scotuss suspicion concerning the distinction between being
and essence. p. 147.
55
Pini points out the inconsistency in Simon s account: Simon adds, however, that a
categorys mode of being is identical to its ratio praedicandi, that is to the mode in which
that category is predicated of the things that are contained in it. Other times, he says
that the mode of predicating is not identical to the mode of being, but is nonethe-
less derived from it. Be that as it may, it is clear that Simon maintains that categories
consist of modes of predication that are either based on or identical to the modes of
being. Pini, Categories and Logic in Duns Scotus , p. 153.
56
For a discussion of how this is carried out by Aquinas, see Symingtons article
above.
240 lloyd a. newton
above and below.
57
Because each of the categories is attributed to this
incomplex entity, he adds,
the diverse ways in which this science and the divine science considers
the categories is obvious, for the divine science considers them insofar as
they are certain essences and parts of being; however, in this book the
categories are not considered insofar as they are simply material things,
but insofar as they have some mode of predicating, which is why this
book is called Of the Predicaments.
58
Thus, Simon maintains that the categories are composed of an essence
and of a mode of being, which is either identical to or is parallel to a
mode of predication, as we have seen. It can be said, therefore, that
logic studies categories not according to their essences but according
to their modes of being, and such modes of being are as real as the
essences of categories.
59
In this way, Simon moves closer than either
Martin or Peter in holding to a similarity or parallel between both a
logical and a metaphysical consideration of the categories.
From Martin of Dacia to Simon of Faversham, the science of the
categories has made a marked development in a relatively short period
of time. Martin seems to be unaware of the problems concerning the
claim that the categories are the subject of a science. Peter, on the other
hand, is aware of the two fundamental problems, viz., that they have
neither a cause nor an appropriate unity to satisfy the basic require-
ments of a science. Although he does not offer any original contribu-
tions, Peters solutions to these problems are obviously taken from the
science of metaphysics, with the consequence that the science of the
categories resembles the science of metaphysics. This similarity is not
57
Simon of Faversham, Quaestiones super librum Praedicamentorum, q. 1, p. 266. The
quotation in question comes at the end, the context of which is Simons comparison
of the unity found in the science of logic to the unity found in theology or divine
science: Sed ad hoc quod aliqua scientia sit una, sufcit quod subiectum sit unum
secundum attributionem. Hoc modo scientia divina est una, quia omnia que ibi con-
siderantur habent attributionem ad unum primum ens quod est substantia, et non (2)
ad quamcumque substantiam, sed ad primam que est Deus, ut apparet ex suo pro-
cesso.Similiter omnia que hic considerantur, habent attributionem ad unum, et illud
est ens dicibile incomplexum ordinabile in gnere secundum sub et supra.
58
Ibid.: Et ex hoc apparet quomodo diversimode considerat ista scientia de pre-
dicamentis et scientia divina, quoniam scientia divina considerat de eis ut sunt quedam
essencie et partes entis; in isto autem libro non determinatur de predicamentis secundum
quod sunt res materie absolute, sed secundum quod habent talem modum predicandi
vel talem; et ide dicitur liber Praedicamentorum.
59
surprising given that he is an early representative of the Modistae.
60

Simon of Faversham then works out the most sophisticated discussion
of the issue. In addition to the problems raised in Peters commentary,
Simon is aware of the three conditions required for a science and of the
distinction between a scientia propter quid and scientia quia. To no surprise,
Simons solution to these difculties borrows heavily from the science
of metaphysics, with the result that a science of the categories is almost
identical to the divine science. The reason for this parallel is that, for
Simon, categories consist of two things: an essence and its mode of
existence. The essences of the categories are considered in the divine
science, while their modes are considered in logic. Accordingly, Simon
also claims that the logician can deduce the number of categories based
on an analysis of their modes of being or being predicated.
When one turns from Scotuss contemporaries accounts of a science
of the categories to Scotuss own account of this same issue, one can
see that Scotus makes by far the most remarkable progress and has the
most sophisticated view on this issue. Unlike his contemporaries, who
devote at most ve pages to this issue in their commentaries on the
Categories, Scotus devotes twenty pages to the science of the categories
and another twenty pages to a related question concerning the univoc-
ity of being. Moreover, these pages do not include his discussion of
the three requirements of a science, which are addressed in his com-
mentary on the Isagoge.
61
To begin with, Scotus criticizes the standard position that maintains
the subject of the categories is an incomplex entity sayable and order-
able in a genus. He also thinks that the categories are the subject of
a scientia propter quid, not just a scientia quia. However, the nature of the
categories and the requirements for any kind of science are such that
a number of difculties face Scotus in demonstrating successfully that
the categories are the subject of science. Consequently, in the remaining
sections of this chapter I focus on three portions of Scotuss argument.
First, I look at how the categories are part of the overall science of
logic. Second, I show how Scotus appeals to the role of the intellect as
the cause of the various properties, thereby establishing a foundation for
the unity of this science. Finally, I focus on the unity of the categories
as second intentions. By covering these points, we will see that Scotus
60
Andrews , Peter of Auvergne s Commentary on Aristotles Categories, p. 12.
61
Cf. q. 13, which are another eighteen pages.
242 lloyd a. newton
successfully demonstrates that the categories are the subject of a scientia
propter quid, a science that he calls simply the categories.
3. The Science of Logic according to Scotus
As I previously mentioned, by the time Scotus composes his commentary
on the Categories, it is commonly accepted that logic is not only a specu-
lative science but also that it is divided into three separate parts com-
parable to the three operations of the intellect. In spite of this general
agreement, considerable disagreement remains over what constitutes the
primary subject of logic, to which all the various properties are nally
reduced. In his commentary on the Isagoge, Scotus reminds us that the
nature and existence of this primary subject must be better known than
all of its properties, which are demonstrated of it. Furthermore, all the
attributes demonstrated of that subject must ultimately be reduced to
that primary subject as to a rst principle. Scotuss contemporaries,
however, disagree on what the primary subject of logic is. Prior to
writing his logical commentaries, various philosophers proposed six
different possibilities. In his commentary on Porphyry s Isagoge, Scotus
rejects ve of these possibilities on grounds that they are insufcient.
62

Let us rst look at these six possibilities.
The rst position, held by Aquinas , contends that the acts of reason
are the rst subject of logic. Just as the acts of reason are threefold
namely, the act of simple apprehension, judgment, and reasoningso
the parts of logic are threefold, which are treated by Aristotle in three
different books, i.e., the Categories, On Interpretation, and Prior Analytics,
respectively. The second opinion, which Scotus incorrectly attributes to
Boethius but was actually put forth by Avicenna , is that the subject of
logic consists of second intentions applied to rst intentions. The third
position, also held by Aquinas, maintains that beings of reason are the
subject of logic. Scotus thinks that these three positions are equivalent
to one another and are all insufcient on the same grounds, namely,
that they do not satisfy the rst two requirements of a subject of a sci-
ence.
63
That is to say, the nature and existence of these three things is
not better known than their attributes, nor are properties demonstrated
62
Quaestiones super Porphyrii Isagoge, q. 3, nn. 720.
63
Ibid., n. 1317.
of them through their denitions. In fact, says Scotus, their denitions
are not even given in logic.
The fourth opinion, rst articulated by Boethius , is that the subject of
logic is speech (sermo).
64
According to Scotus , though, speech is not the
subject of logic, for it fails to satisfy the second and third requirements
of the subject of a science. That is, the denition of speech is not given
in logic nor are the attributes reduced to it as to a rst principle. As
Pini describes it, logical demonstrations are not about the properties
of speech and what is studied in logic cannot be reduced to speech.
65

Scotus conrms this point in his commentary on the Categories, since
the properties demonstrated of the categories exist even if one never
speaks.
66
The last opinion Scotus rejects is the one held by Albert the Great
the subject of logic is reasoning or argumentation.
67
Scotus mercilessly
disproves this position: it fails to satisfy all three conditions. Neither is
the existence and nature of argumentation known in logic, nor are
properties demonstrated of it through a denition, nor are the attributes
demonstrated in logic reducible to argumentation as its rst subject.
According to Scotus , the only possible candidate that satises the
necessary and sufcient conditions for the primary subject of a science
is the syllogism, since only it satises all three requirements.
68
The
64
Ibid., n. 18: Quarto, scilicet orationi, decit secunda condicio et tertia. Et quando
accipitur quod convenit syllogismo, dico quod nonnisi aequivoce ei et enuntiationi; et
aequivocum non est subiectum, quia nec denibile; sumitur enim in denitione syl-
logismi pro argumentatione.
65
66
Q. 1, n. 11: Ad quaestionem dici potest quod iste liber non est de decem vocibus
ut de primo subiectonec aliqua pars logicae est de voce, quia omnes passiones syllo-
gismi et omnium partium eius possunt sibi inesse secundum esse quod habent in mente
etsi non proferantur, ut patet inductive, sed est de aliquo priore, quod respectu vocis
signicativae tantum habet rationem signicati.
67
Quaestiones in librum Porphyrii Isagoge, q. 3. n. 19: Quinto, scilicet argumentationi,
deest primum. Non enim ab Aristotele usquam denitur argumentatio in communi,
nec ponitur eius denitio ut principium scientiae; sed Boethius denit eam in Topicis
suis. Secundum etiam deest sibi, ut manifestum est secundum rationem gereralem.
Tertium etiam probo: quia vel est aequivocum, et tunc manifestum est quod ab eius
unitate non est scientia una, cum tunc non sit unum scibile nec unum intelligibile; vel
est analogum ad contenta sub ipso, et tunc scientia non erit una ab ipso sed ab unitate
primi ad quod alia attribuuntur,et illud est syllogismussicut unitas metaphysicae
est ab unitate substantiae ad quam alia entia attribuuntur.
68
Ibid., n. 20: Dicendum ergo quod subiectum primum et proprium est syllogismus.
Habet enim primam condicionem, quia statim post determinationem de partibus eius
in verteri logica, in principio Priorum, praemittit eius denitioem. Et secundum, quia
244 lloyd a. newton
existence and nature of a syllogism are better known than the attributes
demonstrated of it. Whether they are trained in logic or, as Aristotle
says, simply attempt to discuss statements and to maintain them, all
men proceed from the known to the unknown.
69
Accordingly, the exis-
tence and nature of syllogisms are obvious. Those who study logic also
know that the subject can plainly be handled systematically,
70
such
that properties are shown to inhere in various syllogisms, e.g., to be
from three terms, to be valid or invalid, and to have a middle term.
These properties demonstrated in logic are demonstrated through the
denition of a syllogism. Finally, all things studied in logic are reduced
to a syllogism as to a rst subject.
This last criterion is of interest here. According to this condition,
everything considered in logic must either be a syllogism or something
that can be reduced to a syllogism, as a part is reduced to a whole.
71

As we saw above, though, the immediate parts of a syllogism are propo-
sitions, namely, the two premises and the conclusion. As the immedi-
ate parts of a syllogism, each proposition also has its own attributes,
such as being either true or false.
72
Since a proposition is itself not a
simple part, but is resolved into subject and predicate terms, the parts
of propositions also have properties in their own right. Scotus claims
that, that which is treated here [in the Categories] is essentially [ per se]
part of the subject considered in the book On Interpretation, and that
[subject] is part of that which is considered in the book Prior Analytics,
which is the syllogism.
73
Thus, categories and statements are studied
in logic to the extent to which they are parts of syllogisms.
in eodem multas passiones de illo ostendit per illam denitionem, ut modum et gu-
ram, in illis de inesse et modout habere tres terminosin illo capitulo Quoniam
ititur in his per hoc quod est de necessitate accidere conclusioni etc. Et tertiam, quia
propter ipsum determinatur de partibus eius, scilicet de incomplexo et enuntiatione et
integralibus partibus eius subiectivis in libro Priorum et Posteriorum; et de aliis speciebus
argumentationis, quia illa reducuntur ad ipsum ut imperfectum ad perfectum; et de
syllogismo sophistico, ut de privatione eius, quia eiusdem est cognoscere habitum et
privationem. Sic ergo penes eius divisionem et attributa illi patet divisio logicae.
69
Rhetoric, I, ch. 1, 1354a 56.
70
Ibid., 1354a 910.
71
72
Duns Scotus, Quaestiones super Praedicamenta Aristoteles, eds. Robert Andrews et al.
(St. Bonaventure, 1999), q. 1, n. 9: Ad ista duo potest dici quod esse ex tribus termi-
nis est passio syllogismi ut est in voce, sicut verum vel falsum signicare est passio
enuntiationis, et sic de ceteris passionibus huiusmodi.
73
Ibid., n. 7: Item, illud de quo hic agitur est per se pars subiecti considerati in libro
Perihermenias, et illud est pars eius quod consideratur in libro Priorum, quod est syllogismus;
sed illud omnes suas passiones habet sibi inhaerentes, ut manifestum est, nulla voce
According to many medieval Aristotelians, the primary subject
stands in the same relation to a science as an object stands in rela-
tion to a power. For example, there is a generic power of sensation,
which includes the particular powers of seeing, hearing, smelling, etc.
Moreover, as particular powers, each one has its own formal object
distinguishing one sensation from another. Nevertheless, the ability to
see and to hear share in the common power of sensing. In the same
way, the subject of logic as a whole is the syllogism, but since the syl-
logism is comprised of various parts, there are also various subordi-
nate sciences, such as a science of complex concepts and a science of
incomplex concepts.
74
The former is dealt with in Aristotles work On
Interpretation, while the latter is treated in the Categories.
One may argue, however, that just as Aristotle discusses a part of
logic in the Categories and another part in On Interpretation, so he dis-
cusses a part of logic in the Prior Analytics, namely, the formal nature
of the syllogism. If so, it would seem to follow that the syllogism is
the subject of a part of logic and not of the whole of logic. Would
it not be unreasonable to have the same subject for a part and for a
whole? Such an objection is actually raised in Scotus s commentary
on Porphyry s Isagoge: Nothing is the subject of the whole and of the
part; but a syllogism is the subject of a part of logic, namely, of the
book Prior Analytics, therefore, etc.
75
Scotus responds that the syllogism
is the subject of a part of logic as it is considered formally. However,
it is the subject of the whole of logic with respect to all the attributes
in it or in its integral parts or in its integral subjects, or in the parts
reducible to it. Thus the syllogism, considered formally, is a part of
exsistente nec possibile esse. Similiter, enuntiatio suas habet. Ergo utrumque illorum est
prius naturaliter voce signicativa Et quod hic consideratur, est prius naturaliter utroque
istorum, sicut pars integralis toto; ergo quod hic consideratur primo est naturaliter prius
voce signicativa quacumque. Ergo non est hic subiectum vox signicativa.
74
Ultimately, Scotus thinks that there is a specic science corresponding to each
of the books in Aristotles Organon. Altogether, these separate sciences comprise the
whole science of logic, which has the syllogism as a whole for its proper subject. Cf.
Quaestiones in librum Porphyrii Isagoge, q. 3, n. 24: Ad secundum argumentum principale
dico quod syllogismus quoad proprietates formaliter ipsum consequentes est subiectum
libri Priorum. Est autem subiectum totius logicae quoad omnes passiones in se vel in suis
partibus integralibus vel subiectivis vel reducibilibus ad ipsum. Nec oportet subiectum
praedicari de omnibus consideratis in scientia, sed esse propter quod alia considerantur,
ut patet de subiecto naturalis scientiae, quod est corpus mobile, ubi tamen agitur de
motu et natura, quae non sunt corpus mobile.
75
Quaestiones in librum Porphyrii Isagoge, q. 3, n. 4: Item, nihil est subiectum totius et
partis; sed syllogismus est subiectum partis logicae, scilicet libri Priorum, itigtur, etc.
246 lloyd a. newton
logic and considered as a whole it is the primary subject of logic. With
these distinctions in mind, let us now look more closely at the science
of the categories.
4. The Cause of Second Intentions
Unlike the other parts of logic, a science of the categories presents
several major problems that Scotus must solve. Those problems include
the following: 1) The categories, as the highest genera, do not have a
higher genus. 2) The categories are not specically, or even generically,
one since there is nothing common to them. 3) The categories are not
denable by a genus and differentia since, again, they do not have a
higher genus. 4) The categories, as the principles and causes of every-
thing else, do not have a cause. Although these problems take different
forms, they all essentially amount to one problem: the categories, as
extramental beings, are not one but many, yet a primary subject of
a science must essentially be one, or have an essential unity.
76
To see
how Scotus solves these basic problems, it will be necessary to review
what Scotus has to say about the three considerations of an essence
and about concepts.
Like many of his contemporaries, Scotus accepts the Avicennian
doctrine of a threefold consideration of an essence: 1) as it exists in a
physical thing, in which case it is a material being; 2) as it exists on its
own, in which case it is simply a common nature; and 3) as it exists in
the mind, in which case it is a universal.
77
In his commentary on the
Isagoge, Scotus states:
Nevertheless it must be known that a signicate of a common term,
which signies a true nature, can be considered in three ways. In one
way, according to its being which exists in its supposits, which is said to
be its material being, and in this way, common accidents inhere in it.
In a second way, it is considered simply according to its essential being,
and in this way the essential predicates inhere in it. In a third way, [it is
considered] as it is apprehended by the intellect through an intelligible
76
If the categories had a higher genus, or could be dened by means of a genus
and differentia, or had a cause, they would all have an essential unity. However, the
diversity or plularlity of the ten categories is essentially related to the fact that they do
not have a higher genus, or a single denition, or a cause.
77
For Avicennas inuence on Albertus Magnus, compare Bruno Tremblay s article
above.
form, which is its cognitive being, and in this way intentions are attrib-
uted to it.
78
In the second way, i.e., in itself, an essence or common nature is nei-
ther particular nor universal. As that nature exists in individual things,
however, it is particular and has a numerical unity to it corresponding
to its material being. Finally, as it exists in the mind, it is universal and
has an objective or quidditative being.
Just as one can consider an essence in three ways, one can consider
its attributes in the same three ways. For example, the essence consid-
ered in itself has certain essential properties, namely, those properties
that are contained in a denition, such as the property of risibility as
a necessary attribute of the nature of man. Then there are the prop-
erties or accidents that are present in individuals which pertain to the
essence insofar as it is considered to exist in individuals. For example,
the accident of being white pertains to the essence of man as it exists
in a particular person. Finally, there are also certain properties or attri-
butes that pertain to the essence insofar as it is considered to exist in
the mind, such as the attribute of being predicable of many. For Scotus,
though, the ability to be predicated of many is neither a property of
the common nature as it exists in itself nor as it exists in a particular
person. Rather, it is only a property of the common nature insofar as
that nature is apprended by the intellect.
As an essence exists in the mind, it is a concept or intentionthat
is, the thing considered insofar as it is understood. Without going into
Scotus s theory of knowledge in detail, it is important to remember
that Scotus accepts that there is an identity between the knower and
the known.
79
Thus, when the essence of something is presented to
the knower by way of an intelligible species, which is a real property
inhering in the soul, it exists in the soul as a universal. The essence, as
a universal, however, does not have real, extramental being but only
cognitive, or objective, being. That is, an essence so considered only
exists to the extent that it is an intention. In this way, the essence is said
78
Quaestiones in librum Porphyrii Isagoge, q. 911, n. 16: Sciendum tamen quod signica-
tum termini communis, signicantis veram naturam, tripliciter potest considerari. Uno
modo secundum esse in suppositis, quod dicitur esse materiale eius, et hoc modo insunt
ei accidentia communia. Secundo modo consideratur absolute secundum esse quidita-
tivum, et sic insunt ei praedicata essentialia. Tertio modo ut per formam intelligibilem
ab intellectu apprehenditur, quod est esse cognitum, et sic insunt ei intentiones.
79
248 lloyd a. newton
to be an intention or concept. To be more precise, it is a rst intention,
since that which is rst known is a real, extramental being. According
to Scotus, however, one must distinguish between an essence, which
is neither universal nor individual, and its mode, which is universal.
As Pini states it, Scotus draws a distinction between essence and uni-
versal: the essence is the object of the intellect and what the intellect
understands, whereas the universal is the mode in which the intellect
understands an essence. Since the object and its mode of understanding
are essentially distinct, the intellect can acquire a concept of each of
them and can understand each of them distinctly.
80
In his commentary
on the Isagoge, Scotus states:
It must be said that a universal is per se intelligible. This is evident thus:
The rst object of the intellect, namely, the what it is is understood
under the aspect of a universal. However, that aspect is not essentially
the same as that what it is, but is its accidental mode. Therefore, the
intellect can recognize a difference between its rst object and that mode,
since it can distinguish between all things which are not essentially the
same. But every power, recognizing a per se difference between two other
things, recognizes each extreme under its proper aspect.
81
When the intellect considers not the essence of the extramental thing
but the mode in which that thing exists, it forms a new concept of
this second object. This new concept is a second intention, which is
essentially a concept of a concept. As a second intention, that which
is known is not an extramental being, but the mode of knowing an
extramental being, which mode is really distinct from the essence of
the rst object known.
The formation of second intentions is described by Pini as follows:
when the intellect considers an essence as universal, it distinguishes the
essence and the mode in which that essence is understood. The prop-
80
Ibid., p. 108.
81
Quaestions in librum Porphyry Isagoge, q. 5, n. 4: Dicendum quod universale est per
se intelligibile. Quod patet sic: primum obiectum intellectus, scilicet quod quid est
intelligitur sub ratione universalis. Illa autem ratio non est idem essentialiter cum
illo quod quid est, sed modus eius accidentialis. Igitur intellectus potest cognoscere
differentiam inter suum primum obiectum et illum modum, quia potest distinguere
inter omnia quae non sunt essentialiter eadem. Sed omnis virtus, cognoscens per se
differentiam inter aliqua duo, cognoscit utrumque extremum sub propria ratione, per
Aristotele II De anima. Per hoc enim probat sensum communem esse. Igitur intellectus
potest cognoscere illum modum sive rationem univeraslis per se et sub propria ratione.
Hoc modo, reectendo, cognoscit intellectus se et suam operationem et modum ope-
randi et cetera quae sibi insunt.
erty that moves the intellect to cause a second intention is a property
pertaining to an essence only as conceived under the mode in which
it is understood. It is such a property that moves the intellect to cause
a second intention.
82
To illustrate the formation of second intentions,
consider the category of substance. As an extramental thing, substance
is neither particular nor universal. When this essence is conceived in
such a way that it can be predicated of many it has the attribute of
being universal. The property of being predicated of many, though, is
an accidental mode that is caused by the intellect.
To say that the intellect causes second intentions, however, is not to
maintain that these intentions are ctions of the mind. On the other
hand, as Pini emphasizes, because extramental things are not the
causes of second intentions, there is no one-to-one correspondence
between intentions and extramental things or properties.
83
It must be
asked, then, how these intentions are related to real beings. According
to Scotus , an extramental thing is not the total cause of a second inten-
tion, but is only an occasion, namely, insofar as it moves the intellect
so that it considers in act, while the intellect is the principle cause.
84

To illustrate, when the intellect understands that man is a substance,
it is the common natures of man and substance, as extramental things,
that are the principle causes of the intellects rst intentions, i.e., of
the concepts man and animal. When it comes to items of second
intention, however, the extramental object is only the occasion, not
the primary cause. The primary cause is the intellect itself. In his
commentary on the Isagoge, Scotus describes it thus: For the intellect,
considering the nature of man as one in many and said of many, is
moved by some property found in the nature so considered to cause
an intention. And that cause is attributed to the nature of which it is
a property and from which it is taken.
85
Thus, unlike the primary
cause of rst intentions, the primary cause of second intentions is the
intellect, not an extramental thing.
82
83
Ibid., p. 110.
84
Cf. also, John Duns Scotus , God and Creatures: The Quodlibetal Questions, trans. Felix
Alluntis and Allan B. Wolter (Princeton, NJ, 1975), q. 15, nn. 4063; and Scotus,
Questions on the Metaphysics, VII, q. 18, nn. 478, and q. 19, nn. 1719.
85
Quaestions in librum Porphyry Isagoge, q. 911, n. 17: Intellectus enim, considerans
naturam hominis unam in multis et de multis, ab aliqua proprietate reperta in natura
sic considerata movetur ad causandum intentionem; et illam causatum attribuit illi
naturae cuius est proprietas a qua accipitur.
250 lloyd a. newton
Because the act of reection is involved in the formation of second
intentions, Scotus maintains that a lesser unity sufces in the thing than
the unity that is associated with the rst intention. As Scotus states it,
the intellect considering through that one species, can reect innumer-
able times upon itself by considering its own operation, and whatever
consideration there is, it has nothing extrinsic corresponding to itself,
except only the rst object for the occasion, insofar as that [thing]
moves the intellect initially to consider it.
86
The following example
helps illustrate Scotuss point. Suppose that while visiting a zoo, a man
sees a number of particular animals that he had never seen before.
Since the sign on the cage informs him that these particular animals
are ferrets, he recognizes the general truth that ferrets are animals.
Then, at some time after he has left the zoo, this same man reects
upon his initial observation that ferrets are animals, and notices that
a relation exists between the subject and the predicate, namely, that
the subject points to or signies fewer things while the predicate points
to more things. Or, to use the technical terms, the subject is a species
and the predicate a genus. If this same man were then to reect upon
similar statements, such as all triangles are shapes, and every shade
of blue is a color, he would realize that this same relation is involved
in all these propositions, even though these statements are about dif-
ferent categories. As this example illustrates, the terms species and
genus have a greater unity than the objects of rst intention, since the
objects of rst intention are in three separate genera. To paraphrase
his response in his commentary on the Categories, it sufces for the
intellect to be moved by something extrinsic for the causing of many
notions through the simple consideration of that which is extrinsic, to
which nothing corresponds in the thing.
87
In other words, the reason
for this unity, so Scotus argues, is that the objects of rst intention
are only the occasion that moves the intellect to form these terms. As
Scotus states, the intellect, by considering one concept as related to
another, establishes a relationship between them. That relationship is
the second intention.
88
86
Ibid.
87
Quaestiones super Praedicamenta Aristoteles, q. 3, n. 13: Ad tertium dico quod res non
est tota causa intentionis, sed tantum occasio, scilicet in quantum movet intellectum
ut actu consideret, et intellectus est principalis causa. Ideo minor unitas sufcit in re
quam sit intentionis, quia sufcit intellectum ab aliquo extrinseco moveri ad causandum
multa per considerationem, quibus non correspondent aliqua in re.
88
That second intentions are relations formed by the intellect, how-
ever, is not new to Scotus . Aquinas , as Pini documents, had already
maintained that the intellect forms second intentions by an act of self-
reection. He had maintained that the intellect, after a rst operation
by which it turns towards extramental things, reects on itself and
compares the concepts it has formed to the extramental things those
concepts represent. By this comparison, the intellect understands and
forms an intention following from the mode in which it understands.
89

For Aquinas, a second intention is essentially a comparison between the
intellects understanding of something and that something itself.
90
By
contrast, Scotus maintains that a second intention is not a comparison
of the intellect and an extramental thing, but simply a comparison
between two things insofar as they are both understood.
91
That is,
second intentions, such as genus and species, are relations between two
concepts, not between a concept and an extramental thing, as Aquinas
holds. Stated another way, what is new to Scotus is his claim that the
intellect is the cause of the properties that pertain to essences as they
exist in the mind. For, according to Scotus, the attribute of being
universal is not something that pertains to the common nature as a
common nature, but rather it is something that is caused by the intel-
lect when it compares two concepts, i.e., to things considered insofar
as they are understood.
Thus far, then, we have seen that Scotus accepts the Avicennian
notion that an essence, along with its properties, can be considered
in three ways: as it exists in itself, as it exists in an individual, and
as it exists in the mind. As is exists in the mind, the essence must
be distinguished from the mode in which it exists. That distinction
allows the intellect to form second intentions, which are concepts of
concepts. While these second intentions do not have real, extramental
89
Ibid., p. 107.
90
Although Aquinas typically describes second intentions as relations between the
intellect and the extra mental thing, Gyula Klima points out that in De potentia, Aquinas
does claim that a second intention is either a relation between a concept and an extra
mental thing, or between two concepts formed by the intellect [emphasis added]. See,
Thomas Aquinas, De potentia, in Quaestiones disputatae, vol. 2, (Rome, 1953), q. 7, a. 11:
Uno modo secundum quod iste ordo est adinventus per intellectum, et attributus ei
quod relative dicitur; et huiusmodi sunt relationes quae attribuuntur ab intellectu rebus
intellectis, prout sunt intellectae, sicut relatio generis et speciei: has enim relationes ratio
adinvenit considerando ordinem eius quod est in intellectu ad res quae sunt extra, vel
etiam ordinem intellectum ad invicem.
91
Pini , Categories and Logic in Duns Scotus, p. 125.
252 lloyd a. newton
being, they are nevertheless not ctions created by the mind. For real
beings are the occasion of these second intentions, which intentions
are caused by the intellect reecting on itself. Finally, because it is the
intellect, when it compares two things as they are understood, that
causes second intentions, these second intentions have a greater unity
than the external objects that serve as the occasion for their existence.
With these distinctions in mind, let us now see how Scotus solves these
fundamental problems relating to a science of the categories.
5. The Unity of the Categories
As I said at the beginning of the last section, a science of the categories
presents several major problems that Scotus must address. In one way
or another, these problems can be reduced to the fact that the catego-
ries, as extramental beings, are not one but many, whereas a primary
subject of a science must essentially be one. As we have seen, Scotuss
contemporariesMartin of Dacia , Peter of Auvergne , and Simon
Favershamsolve this problem by claiming that the categories have
a unity of attribution or an analogous unity; a position that is similar
to their view regarding the science of metaphysics.
92
Scotus, however,
rejects the similarity between logic and metaphysics for the following
three reasons.
First, while the categories of quantity, quality, relation, etc., exist
metaphysically or ontologically only in substances, the logical concepts
of quantity and quality do not inhere in a substance in the same way.
In other words, as beings of reasonas logical concepts, the categories
exist as products of the intellect. This is seen most evidently in the
category of substance, which, as a logical concept, depends upon the
activity of the intellect, even though ontologically a substance is what
exists in itself. As second intentions, all ten of the categories inhere
equally in a subject, namely, in the one who conceives of them. As
92
Following Aristotle, most medievals claim that being is said analogously of the
ten categories, not univocally, both insofar as they are studied in metaphysics and
insofar as they are studied in logic. In this early commentary on the Categories, Scotus
will grant that, metaphysically speaking, being is said analogously of the various
categories. However, he will argue that they have a greater unity, a univocal unity,
when they are studied in logic. As is well known, Scotus will later afrm that being is
said univocally of the various categories, both when studied in logic and when studied
in metaphysics.
an item of rst intention, however, substance signies something that
is more fundamental than the other nine accidents since these latter
inhere in substances. Nevertheless, as second intentions, they all are
equal since their properties are caused by the intellect.
The second reason why Scotus rejects the similarity between meta-
physics and logic is that an appeal to a unity of attribution will not
work in logic as it does in metaphysics. Metaphysically speaking, the
existence of accidents and the existence of substance is not the same
kind of existence. Accidents exist in a substance, whereas a substance
is something that exists in and of itself. Thus, the kind of unity that
the categories share metaphysically is a unity of attribution or a unity
of analogy, since the accidents are attributed to substance or exist in
a substance. However, when someone predicates genus of substance,
quantity, and quality, etc., this predicate term is not simply analogous
to each subject. That is, the nine accidental genera are not attributed,
as logical entities, to substance. Rather, substance (and quantity, quality,
etc.) is a highest genus in exactly the same way as every other category
is a highest genus. As Scotus points out, the term genus is univocal
to all the ten categories since each one of them is a highest genus in
exactly the same way as every other one is a highest genus. Thus, it
is true that, metaphysically speaking, the categories, as items of rst
intention, do not have anything predicated of them univocally, all that
can be said is that they have a unity of analogy or a unity of attribu-
tion. As items of second intentions, however, they have a greater unity,
namely, a generic, or perhaps even a specic unity. For when I say that
substance is a highest genus and quantity is a highest genus, there is
a univocal term that is predicated of each of these intentions.
Third, Scotus rejects his contemporaries view in part because he does
not agree that the mode of signifying is parallel to the mode of being,
a position which is intricately related to many of his contemporaries
belief that a thing is constituted by two really distinct metaphysical
principles, i.e., essence and existence.
93
In fact, where Thomas Aquinas,
Henry of Ghent , Simon of Faversham, and others all think that the
categories are constituted by two metaphysical principles, an essence or
ratio studied in metaphysics and existence or mode of being studied in
93
It is not entirely clear whether the lack of a parallel between signifying, under-
standing, and being is caused by or results from Scotus denial of a real distinction
between essence and existence. Nevertheless, that the two are related is suggested by
Pini , Categories and Logic in Duns Scotus, pp. 1479.
254 lloyd a. newton
logic, Scotus thinks that the categories, insofar as they exist apart from
the intellect, are simple essences in themselves.
94
According to Scotus,
though, the categories can also be considered in logic: not insofar as
they are simply diverse, but insofar as some property caused by the
intellect is attributed to them.
For Scotus , then, categories can be considered in two ways: insofar as
they are extramental beings and insofar as they have objective being.
95

As simple essences, the categories are studied in metaphysics, and have
a unity of attribution.
96
Considered as they are understood, categories
are studied in logic, and have, as we saw, a greater unity than the
unity of attribution. For Scotus, the categories are not constituted by
two metaphysical principles but are simply considered in two different
ways. As Pini says, Scotus maintains that categories can be considered
in two ways, logically and metaphysically, and not that they are consti-
tuted by two metaphysical components. According to Scotus, Aquinas
and Henry confuse the way in which we consider categoriesi.e., as
classifying notionsand the way categories arei.e., as essences and
mind-independent things. Categories, metaphysically considered, differ
from one another by themselves and as immediately diverse things, not
because of a mode of being that somehow pertains to and is separated
from them.
97
94
Ibid., pp. 14454. According to Wolter , a real distinction between essence and
existence is, as Scotus says, a ction of which I know nothing. Quoted in Wolter,
Transcendentals and their Function, p. 66; see also Joseph OBrian , Duns Scotuss Teaching
on the Distinction Between Essence and Existence, New Scholasticism 38 (1964), 6177.
As Paul Symington points out in his article above, the parallel between signifying and
the modes of being allows for many medieval philosophers to deduce the number
of categories from their modes of being. In this case, the mode of being seems to
function as a univocal genus from which the ten categories are ultimately derived by a
process of division. For a detailed discussion of this topic , see Giorgio Pini, Scotus on
Deducing Aristotles Categories, in La tradition mdivale des Catgories XII
e
XIV
e
sicles.
Actes du XIII
e
Symposium europen de logique et de smantique mdivales Avignon,
610 juin 2000, ds. J. Biard and I. Rosier-Catach (Louvain, 2003), pp. 2335.
95
Quaestiones super Praedicamenta Aristoteles, q. 2, n. 5: Dicitur ad quaestionem quod
decem praedicamenta possunt dipliciter considerari: uno modo in quantum sunt entia;
alio modo in quantum considerantur a ratione, sive in quantum aliqua proprietas
causata ab intellectu eis attributur.
96
Ibid., q. 2, n. 6: Primo modo de eis considerat metaphysicus; subiectum enim eius
primum est ens in quantum ens; q. 2, n. 8: Sed cum ista decem non sint subiectum
unius scientiae realis, scilicet metaphysicae, nisi in quantum est in istis unum primum
ad quod alia attribuntur, ut habetur IV Metaphysicae; et in quantum considerantur a
ratione nullum est ad quod attribuantur, quomodo sunt unum subiectum.; Cf. also
q. 4, n. 37.
97
In order for him to explain how the categories have a univocal subject
without reducing them to some higher genus, Scotus makes a crucial
distinction between something predicated univocally of many and a
univocal term predicated of many. As he states it in question three of
his commentary on the Categories, it is one thing to say something is
predicated univocally of many, and [another thing to say] there is a
univocal predicate of many.
98
According to Scotus, in the rst way
a denominative [term] can be predicated univocally, in the latter way
it is not.
99
Scotus provides several arguments to substantiate his claim
that a denominative term can be univocally predicated of a subject,
even though it is not a univocal predicate. Here, it will sufce to look
at two of these arguments.
The rst one is taken from a comparison to the denominative term
white. According to Scotus, white can be predicated equally of a
man and of a surface, insofar as it is true to say both that a man is
white and that the mans skin, or surface, is white. In this example, the
term white signies the same common nature, and hence is predicated
univocally. However, in this example the two things of which white is
predicated are of two different genera, i.e., substance and quantity,
respectively. Because it is predicated of things in other genera, it is
also a denominative term. Nevertheless, in these two examples the
term white signies one and the same thing and hence is predicated
univocally. However, it is true that neither a man nor his skin is a color,
which would follow if the term were a univocal predicate, as opposed
simply to being predicated univocally. Applying this distinction to the
categories, Scotus argues that highest genus is also a denominative term
since it is univocally predicated of ten diverse genera. Nevertheless, the
term is not univocal insofar as it is predicated of these ten categories in
exactly the same way, even though it is predicated of them univocally.
Once again, the reason that it is predicated univocally is that it signies
the same common account when it is predicated of each of the ten
categories. The reason that it is denominative is not that it receives its
name from an abstract accident, as grammarian does from grammar,
nor from the fact that it has a different ending, as Aristotle claims, but
98
Quaestiones super Praedicamenta Aristotele, q. 3, n. 14: Aliud est dicere aliquid univo-
cum praedicari de multis, et esse univocum praedicatum de multis.
99
Quaestiones super Praedicamenta Aristoteles, q. 3, n. 14: Ad quartum dico quod aliud est
dicere aliquid univocum praedicari de multis, et esse univocum praedicatum de multis.
Denominativum primo modo est univocum, secundo modo non, ut ostendit ratio.
256 lloyd a. newton
that it is predicated of things in diverse genera in the same way that the
quality white inheres in the categories of substance and quantity.
The second argument in behalf of his claim that a denominative
term can be predicated univocally is taken from a comparison between
a circle and the matter of a circular object. The remote genus of a
circle is quality, while its proximate genus is shape. What differentiates
a circle from other plane gures is that it is contained by one line such
that all the straight lines falling upon it from the center among those
lying within the gure are equal to one another. Of course, a particular
circle may exist in copper as well as in silver. The difference between
a circle that exists in copper, however, is purely accidental to one that
exists in silver, for silver and copper are material differences, whereas the
formal differentia is that which differentiates a circle from other plane
gures. Furthermore, the material differences of various circular things
do not essentially inhere in the nature of a circle, whereas a formal
differentia does necessarily inhere in the essence of a circle. In a similar
way, Scotus argues, the difference between the categories as extramental
beings and the categories as they are studied in logic is analogous to
the material differences of various circular objects and the nature of a
circle in itself. For as they are in themselves, the categories are simple
essences that are completely diverse from one another. In fact, they are
more different from one another than any two material things are from
one another. However, as they are studied in logic, there is a formality
to the ten categories. That formality is the nature of genus, which is
a second intention. For insofar as they are conceived, the categories
have a formal unity to them. What is that formal unity? To the extent
that they are all conceived, they are all concepts, or second intentions.
Consequently, the fact that the categories are essentially diverse does
not prevent something from being predicated of them univocally.
In these last two sections, we see a number of distinctions, some
borrowed from the tradition and some that are new to Scotus , that are
foundational to his claim regarding the knowledge of the categories.
These distinctions include: 1) the threefold consideration of an essence;
2) a distinction between an essence as a simple thing and its mode of
existence in the mind; 3) the reection of the intellect on its concepts,
thereby causing second intentions; 4) second intentions as relations
between things understood; 5) a greater unity in second order concepts
than exists in their occasions; 6) the twofold consideration of the cat-
egories; and nally 7) the distinction between a univocal predicate and
something predicated univocally. With these distinctions in mind, let
us briey return to some of the various questions involved in a propter
quid science of the categories.
To the claim that the categories, as highest genera, do not have a
higher genus, Scotus responds by pointing out that the categories can
be considered in two ways: in metaphysics and in logic.
100
In the former,
they do not have a higher genus, since they are simply diverse. In the
latter, though, they do have a genus, namely, the genus category. To
the claim that the categories are not specically, or even generically, one
since there is nothing common to them, Scotus responds by claiming
that they do have a unity, one caused by the intellect when it reects
upon rst intentions.
101
Considered as second intentions applied to rst,
the ten categories have the common aspect of second intention. To the
claim that the categories are not denable by a genus and differentia,
Scotus responds that they are denable as one subject, namely, as pre-
dicaments or categories.
102
To the claim that the categories, as the
principles and causes of everything else, do not have a cause, Scotus
responds that although the categories themselves do not have a cause,
the properties inhering in them do have a cause, namely, the intellect.
103

Taken together, these claims amount to the fact that the categories is
the subject of a science, not just a descriptive science or a scientia quia,
but a scientia propter quid. What is the name of this science? According
to Scotus, it should simply be called the Categories.
100
Quaestiones super Praedicamenta Aristoteles, q. 2, n. 5: Dicitur ad quaestionem quod
decem praedicamenta possunt dipliciter considerari: uno modo in quantum sunt entia;
alio modo in quantum considerantur a ratione, sive in quantum aliqua proprietas
causata ab intellectu eis attributur.
101
Ibid., n. 9: Oportet dicere quod maior est unitas eorum in aliqua proprietate
causata ab intellectu quam in quantum sunt entia. Et ita, cum haec scientia non sit
una unitate analogiae, oportet assignare aliquid intentionale quod sit commune istis,
et primum subiectum, quia de solo tali per se considerat logicus.
102
Ibid., n. 29: Ad secundum patet quod primum subiectum hic est aliquid univocum
istis decem, ut praedicamentum.
103
Ibid., n. 19: Ideo dici potest quod hic consideratur de decem praedicamentis in
quantum eis attribuitur aliquid causatum a ratione, quia aliter non possunt considerari
a logico. Et illo modo non habent tantum unitatem analogiae, sed etiam univocatio-
nis; et illud univocum istis illo modo est aliquod intentionale, quod est hic primum
subiectum. Illud potest nominari praedicamentum vel generalissimum, quia omnes
proprietates quae per se determinantur hic de istis, determinantur de eis in quantum
habent rationem generalissimi vel praedicamenti.
258 lloyd a. newton
6. Summary
Considered in metaphysics, the categories, as extramental beings, are
simply diverse from one another and have only a unity of attribution.
Considered in logic, the categories have a unity caused by the intellect
when it considers them as highest genera. The difference between items
of rst intentions and items of second intention, however, should not
be construed to imply that there is absolutely no parallel between the
orders of being, understanding, and signifying, simply that one should
not presume that such a parallel exists. As Pini puts it, Scotus does
not deny that sometimes there may be a correspondence between an
intentional property of the categories and a real thing in the extramental
world. His point is that, since such a correspondence is not something
regularly found, it would be misleading to rely on a general parallelism
between the intentional and real plain.
104
To see how extensive this
lack of a parallelism is, one needs to examine his commentary on the
Categories in its entirety.
104
FINE
-
TUNING PINI S READING OF SCOTUS S
CATEGORIES COMMENTARY
1
Todd Bates
In his Categories and Logic in Duns Scotus , Giorgio Pini offers a cogent
and persuasive interpretation of Scotuss thorny and difcult Categories
commentary. Nevertheless, I shall argue that his interpretation is awed.
Pini claims that Scotus distinguishes the domain of logic from that of
metaphysics in a different manner from his most notable modist con-
temporaries, such as Martin of Dacia , who argue that there is a strict
isomorphism between words, concepts and things. In particular, Pini
reads Scotus as saying that categories taken logically, like species and
genera of substances, do not correspond to distinct proper parts in the
thing cognized; I shall contend that they do.
On my interpretation, Scotus holds that some distinct common entity
in the thing corresponds to each of our generic or specic logical con-
cepts. For instance, I hold that Scotus believes there is both a foundation
for our concept of the genussay, animal in the substance cognized
and a distinct foundation in the thing cognized for our concept of the
most specic species, say, human. The foundation in neither case simply
has the universality belonging to the respective logical category. For
instance, there is no generic proper part with universaslity in the thing
that corresponds to the logical genus universality. Indeed, with Pini , I
read Scotus as saying there is a common proper part that corresponds
to the logical genus in the intellect. I depart from Pini by adding that
Scotus implies that a universal species in the intellect, human, has its
foundation in a corresponding common entity distinct from the entity
corresponding to the genus, animal.
Scotus is not concerned to establish a foundation in the thing cog-
nized with a universality that matches the universality of the concept.
Rather, I argue that Scotus is concerned to show there is a foundation
1
Editors note: The following article by Bates is directed against Pinis book Categories
and Logic in Duns Scotus (Boston, 2002), not his article in this book. To the extent, how-
ever, that the previous article by Newton draws heavily on Pinis work, Bates is thus
also indirectly attacking part of the thesis set forth in that article.
260 todd bates
in the thing cognized that can be the very same item as that which is
present in the cognizing intellect. Scotus desideratum is not a matching
correspondence with respect to universality, but rather unity between
an entity in the thing cognized and the concept in the intellect.
1. Pini s Reading of Scotus
I shall begin by explaining Pini s interpretation of Scotus , with emphasis
on the problem of how Scotus portrays the relationship between logical
features of our concepts and features of things apart from the intellect.
Pini argues that Scotus does not claim something universal in the thing
corresponds to any of our specic or generic conceptsPini successfully
makes that point. However, I contend that he has not succeeded in two
other parts of his argument. First, Pini locates the essence of the thing
as the foundation for the universality of all of our logical concepts;
I disagree. Second, he claims that no proper parts of the essence cor-
respond with our logical conceptshere I disagree as well.
Pini points out that Scotus shared with his contemporaries a con-
viction that Aristotles Categories was primarily a work of logic, not of
metaphysics. Logic considers properties, for example the universality
of generic concepts, attributed to items from the activity of the intel-
lect. Scotus writes:
Between a thing and speech, or voice, is a concept. Thus, just as there is
a science per se of things, and some sciences per se of signicant speech,
like Grammar and Rhetoric which consider the passions of the voice . . . so
there can be a science per se of the concept, and this is logic. Whence it
follows necessarily that logic is called a rational science not only as it is
carried on through reason, as is every other science, but because of this:
that it is about concepts formed through an act of reason.
2
As the logician seeks knowledge of properties of our concepts, and
our concepts belong to the intellect, the properties that the logician
studies are mental entities. These latter are to be distinguished from
the properties that the metaphysician studies, the properties of beings
independent of the operations of our intellects.
3
2
Duns Scotus , Quaestiones super Praedicamenta Aristoteles, eds. Robert Andrews et al.
(St. Bonaventure, 1999), q. 1, n. 4.
3
In their understanding of the Categories, Scotus and his contemporary logicians
diverge from our contemporariesamong whom there is a consensus around taking
the work as a piece of Aristotles early ontology. In addition, what Scotus and his
pinis reading of scotuss CATEGORIES commentary 261
While Scotus and his contemporaries share a reading of Aristotles
Categories as a work of logic rather than metaphysics, Pini argues that
they did not share Scotuss understanding of logic, which stood out
from among his contemporaries as carrying relatively modest ontological
commitments.
4
Scotuss contemporaries understood logical properties
attributed to items from the activity of the intellect to be causally tied
to a foundation in properties those items had independently of the
intellect. In contrast, according to Pini, Scotus posits no such ties: in
particular, there need be no correspondence between the modi intelligendi
and the modi essendi.
For instance, Scotus s modist contemporaries saw the universality of a
species-concept as requiring for its foundation in reality some universal-
ity in the thing. They required a correspondence between the modes of
conceiving, or modi intelligendi, and the modes of being, or modi essendi.
In addition, Scotuss modist contemporaries were characterized by their
giving a special ontological interpretation to the modi signicandi, one
that determines the ontological interpretation of logic.
5
Modi signicandi
were types of grammatical distinctions in actual spoken language like
declensions of nouns and tenses of verbs. According to them, modi
signicandi correspond to modi intelligendi, which in turn correspond to
modi essendi.
6
For instance, a certain type of concept corresponds to a
singular nominative noun, and another type of concept to a singular
accusative noun. To each type of concept, there corresponds some type
of being in the thing. For the modists, metaphysics proper study took
up the being of things outside the intellect; logic considered the features
of types of conceptswhich were called second intentions. However
according to the logician, genus and species are not to be explained
only as concepts of concepts . . . they reect features of the object to
contemporaries call logic would not be called that today. In the shadow of Frege,
most philosophers today would reject this conception of logic as mistakenly tying it to
the realm of human psychology.
4
Thus, Pini says I have also made a special effort to avoid general tags such as
nominalism and realism, which may be helpful in other contexts but here may
mislead. For example, Scotus is usually classied as a realist, but his logical doctrine
of categories turns out to have remarkably weak presuppositions. (2)
5
Christian Knudsen , Intentions and Impositions, in Norman Kretzmann et al.,
eds. The Cambridge History of Later Medieval Philosophy. From the Rediscovery of Aristotle to the
Disintegration of Scholasticism 11001600 (Cambridge, 1982), pp. 47995, p. 486.
6
Jan Pinborg, Speculative Grammar, in Norman Kretzmann et al., eds. The
Cambridge History of Later Medieval Philosophy. From the Rediscovery of Aristotle to the
Disintegration of Scholasticism 11001600 (Cambridge, 1982), pp. 25470, p. 262. For
more on the modistae, cf. Irne Rosier, La grammaire spculative des modistes (Lille, 1983).
262 todd bates
which they refer.
7
For the modists, [a]ll the intentional entities, rst
intentions as well as second intentions, are drawn from the properties of
the object signied.
8
Thus, the universality of a logical species would
be drawn from a corresponding universal species in the object.
It may seem that on Pini s reading Scotus , in contrast to the modists,
cuts logical features of concepts like universality loose from reality. Pini
writes Scotus maintains that the object of the intellect, considered
insofar as it is understood, is a mental entity, a sort of internal object,
whose being is identical to its being understood.
9
Pini here seems to
claim that there is nothing more to the universality of the concept ani-
mal than its being understood, such that any correspondence to things
outside the intellect is accidental, like the broken clock that happens
to be right twice each day. Again, Pini writes: Species and genera, by
themselves, however, are logical notions depending not on the way the
essence is, but on the way it is understood.
10
Pini seems to say that
the intellect just makes up universality by instituting a purely mental
relation of predicability between conceptual predicates and subjects.
One might worry that inasmuch as this universality matches nothing
about objects, it is ctitious.
Pini thinks Scotus has a reply to this worry. Scotus nds a foundation
in the thing for our specic and generic concepts: the things essence.
Categories admit what Pini calls a twofold consideration for Scotus,
as logical and as metaphysical entities. In support of this view, Pini
cites Scotuss commentary on the Categories when he claims: [f ]irst,
metaphysics studies categories as types of beings. Second, logic studies
categories insofar as our intellect understands them and attributes some
properties to them.
11
Pini elaborates that Scotus identies a [meta-
physical] category with a type of essence, which is neither individual
nor universal,
12
i.e., what becomes known in Scotuss commentaries
on the Sentences as the common nature.
In his logical work, a theory of common natures is not fully articu-
lated, yet a form of the theory is present in Scotus s theory of the
threefold signication of the term human, in which Scotus recognizes
13

7
Ibid., p. 262.
8
Knudsen , Intentions and Impositions, p. 487.
9
Pini , Categories and Logic, p. 124, emphasis mine.
10
Ibid., p. 148, emphasis mine.
11
Ibid., p. 19.
12
Ibid., pp. 1478.
13
Cf. his Quaestiones in librum Porphyrii Isagoge, q. 911, n. 16: Sciendum tamen
quod signicatum termini communis, signicantis veram naturam, tripliciter potest
three ways in which the human essence can be signied. First, it may
be signied according to its material being, i.e., its actual existence
in the thing to which it belongs. Thus, the term human may signify the
essence of man as it exists in actual humans. Second, the term human
may signify the essence of man according to the essences quiddita-
tive being. In this sense the term human signies the essence of man
taken absolutely apart from its actual existence in the thing or its
existence in the intellectit is taken simply as it is in itself. Third, the
term human may signify the essence of a man insofar as it is a cog-
nized being, i.e., the existence it has in the intellect, with an accident
of universality. Here the term human signies the essence human insofar
as it is a concept.
The theory of the threefold signication of the term human implies
that the essence, taken simply as it is in itself, is indifferent to being
universal in the intellect or being in the thing. That is, it can exist in
either the intellect or the thing, but need not exist in the intellect or
the thing. Existence in the thing is accidental to the essence, as well as
existence in the intellect. Thus, to conceive the essence as universal is
not to have a false or ctitious cognition of it, since the essence taken
absolutely can have universality as an accident. Consequently, Pini has
a reply ready to anyone who might worry that his account of Scotus
would render our concepts ctitious. He writes, [t]his [there being a
thing as a necessary condition for the production of a second inten-
tion] is enough to give some real ground to second intentions and to
differentiate them from ctitious concepts.
14
As one can see, according to Pinis understanding of Scotus, the
intellect adds universality to the essenceuniversality is not taken from
the essence absolutely as it exists in the thing apart from the intellect.
Thus, Pini writes: [a]n intention [like the universality of a species or
genus] is founded not on a property of an extra-mental thing but on a
property of a thing as understood by the intellect, which for Scotus is
a concept and a mental entity, as we have seen.
15
The essence exists
in the thing apart from the intellect, and need not be actually universal
considerari. Uno modo secundum esse in suppositis, quod dicitur esse materiale eius,
et hoc modo insunt ei accidentia communia. Secundo modo consideratur absolute
secundum esse quiditativum, et sic insunt ei praedicata essentialia. Tertio modo ut
per formam intelligibilem ab intellectu apprehenditur, quod est esse cognitum, et sic
insunt ei intentiones.
14
Pini , Categories and Logic, p. 113.
15
Ibid., p. 110.
264 todd bates
in the thing. The essence in the thing is the occasion for the intellect to
produce universality. Occasion is Scotuss term; e.g., he writes, a thing
is not the whole cause of the intention, but only an occasion, inasmuch
as it moves the intellect, as it considers in act, and the intellect is the
principal cause.
16
According to Pini , Scotus holds that the existence of
an essence in a thing is merely a necessary condition for the essence to
be universal in the intellect, but not a sufcient condition.
17
Thus, the intellect is the efcient cause of the universality of the logi-
cal concept, and the essence in the thing is merely the occasional cause
of that universality. Again, Pini writes, Although Scotus s doctrine of
the threefold consideration of a thing is not new, what is peculiar is his
insistence on the intellects role in forming second intentions.
18
That is,
the intellect forms second intentions of universality from reecting on
its understanding of essences. The intellect without the essence in the
thing actually produces the logical concepts universality when it takes
up the essence in the thing for its consideration. I concede to Pini that
in places Scotus seems to talk this way, as when Scotus writes
Similarly, the intellect considering a thing, through that one species, can
reect thousands of times on itself about its operation, in considering, and
any consideration is something having nothing extrinsic corresponding
to it, except only the rst object for the occasion, inasmuch as that rst
[occasion] moves the intellect to consider.
19
Moreover, as Pini sees it, Scotus holds that one and the same essence
in the intellect may receive different intentions of universality produced
by the intellect. Uncovering additional properties of the thing is not
requiredonly reection on the intellects own operation: it [the intel-
lect] reects on its activity and takes into account the mode in which
it understands extramental things. The point is, as Pini claims, Scotus
16
Scotus , Quaestiones super Praedicamenta Aristoteles, q. 3, n. 13: Ad tertium dico quod res
non est tota causa intentionis, sed tantum occasio, scilicet in quantum movet intellectum
17
Thus, Pini writes These adverbs suggest that a thing is a material or necessary
condition for the production of intentions. Giorgio Pini, Categories and Logic, p. 110.
18
Ibid., p. 104.
19
Scotus , Quaestiones super Praedicamenta Aristoteles, q. 3, n. 13: Similiter, intellectus
considerans per illam unam speciem, potest millesies reectere se supra suam operatio-
nem considerando, et quaelibet consideratio aliquid est, nihil habens extrinsecum sibi
correspondens, nisi tantum primum obiectum pro occasione, in quantum illud movet
primo intellectum ad considerandum.
holds the intellect can form several second intentions to which there
corresponds one and the same extramental thing.
20
For example, consider substance and accident. In metaphysics,
according to Pini , a substance is an entity which exists in itself and
not in another. Likewise, accidents are said to exist in another. In
these cases, existing in itself and existing in another are features
that items in the world have apart from any inuence of the intellect.
In logic, however, the study of substance and accident does not involve
being in itself or being in another apart from the intellect; rather, logic
takes up inherence as a relation between subjects and predicates in
the intellect. Thus, Pini argues that according to Scotus there are two
relations of inherence, one real and one logical. The logical relation of
inherence between a predicate and a subject differs from the relation
of inherence that exists between accidents and substances. For in a
predication, such as Socrates is human, the predicate concept human
inheres in the subject concept Socrates. This merely logical inherence
is expressed in the relation of predicability between the predicate and
the subject: we say the concept human is predicable of the concept
Socrates. But the concept human bears predicability to many distinct
substance-concepts; for instance, we may predicate the concept human
of the subject concepts Xanthippe, Plato , and Alcibiades.
For a concept to be predicable of many distinct concepts of individual
substances is for it to be universal. Scotus calls the feature of univer-
sality possessed by the species human a second intention. Pini writes,
According to Scotus, however, not only does the intellect understand, it
also causes a second intention as a relation. Thus, a second intention is
a relation and a comparison established by the intellect.
21
Scotus writes
Every second intention is a rational relation . . . pertaining to the act
of the intellect composing and dividing one extreme and another.
22

The intellect considers the concept human. Noting that the concept
human can be related to many subjects such that it is predicable of
each, the intellect possesses a universal concept of humanity. The
universality of the species is just its predicability, its being able to be
related by the intellect to numerous subject concepts in predications
by the intellect. The intellect did not discover the universality of the
20
Pini, Categories and Logic, p. 112.
21
Ibid., p. 116.
22
John Duns Scotus , Ordinatio, in Opera omnia (Vaticana, 1950), I, d. 23, n. 10.
266 todd bates
concept as something universal existing in the external thing. As Pini
reports, . . . the relation of predicability the intellect establishes does
not depend on the nature of animals and men taken as extramental
things; it depends on the intellects understanding of animals and men
in a certain manner (specically, animal is understood as a universal as
compared to men).
23
Other mental items, say the concept of Socrates ,
cannot be related by the intellect in predication like the concept animal
canthe concept Socrates is not predicable, and so cannot be universal.
Quite apart from the intellect deriving its understanding of a genus
universality from the consideration of things and their real properties,
the intellect forms second intentions, like the species with its note of
universality, remaining within its own mental realm.
Thus, although there is, on Pini s reading of Scotus , a metaphysical
correlate in the substantial thing for our specic and generic concepts,
it would be wrong to infer that there are distinct metaphysical correlates
in the thing corresponding to each of our distinct concepts of species
and genera. Apparently supporting Pinis interpretation, Scotus writes,
Thus, a lesser unity sufces in the thing than the unity of the inten-
tion. For it sufces that the intellect is moved by something extrinsic
to cause many things through consideration, to which there do not
simply correspond many things in the thing.
24
No universal entity
animal in the thing corresponds to the universal concept animalto all
these concepts in the intellect there corresponds only one entity in the
thing, the essence. Thus, according to Pini, There is, however, only
one extramental thing corresponding to all intentions, for each one of
these intentions represents neither a different extramental thing nor a
different mode of being but the mode in which the intellect understands
an extramental thing.
25
To sum up, as Pini sees it, Scotus says that the universality of logical
categories does have a metaphysical presuppositionan essence existing
in the thing cognized. That essence in the thing cognized is not itself
universal, and does not contain universal parts corresponding to the
other universal concepts the intellect may form while cognizing specic
23
Giorgio Pini , Categories and Logic, p. 119.
24
Scotus , Quaestiones super Praedicamenta Aristoteles, q. 3, n. 13: Ad tertium dico quod res
non est tota causa intentionis, sed tantum occasio, scilicet in quantum movet intellectum
25
Pini , Categories and Logic, p. 107.
and generic concepts. And it does not contain any proper parts cor-
responding to the various logical concepts of a thing we may form in
the intellect. As Pini writes, On the other hand, because extramental
things are not the causes of second intentions, there is no one-to-one
correspondence between intentions and extramental things or proper-
ties, and it may happen that one and the same thing corresponds to
many intentions.
26
2. An Alternative Reading of Scotus
My reading of Scotus overlaps Pini s in our agreement that the essence
of the thing cognized is a foundation for our generic and specic con-
cepts. Nevertheless, I regard Pinis interpretation of Scotus as awed in
arguing that the essence in the thing is the only foundation of our specic
and generic concepts. Indeed, I hold that Pinis interpretation produces
a tension between Scotuss logic, on which the essence is without proper
parts, and a plausible reading of his metaphysics which I shall present,
on which the essence has proper parts. On my reading of Scotus, there
is one entity in the thing that is a foundation for the concept of the
species, and another distinct entity that is the foundation for the concept
of the genus. For each higher generic concept, there is still another
entity serving as a foundation in the thing. For instance, the essence
in humans is the foundation for our concept of the species human, the
generic entity animal in humans is a foundationa inasmuch as it
could have a foundation as well in other species of animalsfor our
generic concept animal, and the generic entity substance in humans in a
foundation for our generic concept substance. These foundational entities
are all proper parts of the substanceand the foundational entities for
our generic concepts are all proper parts of the essence.
I argue that in Scotus s metaphysics, the logical, generic concept in
the intellect has a foundation in a common entity, the metaphysical
genus that is a proper part of the substances essence. When Scotus
writes, And this must be understood so as to make clear what it is in
the singular [the thing external to the intellect] that corresponds to the
common concept of the genus, as was explained in reading the text
27
26
Ibid. , p. 110.
27
John Duns Scotus , Quaestiones super libros Metaphysicorum Aristotelis, Libri IIX, in Opera
philosophica, vols. 3 and 4, eds. R. Andrews, G. Etzkorn, G. Gl, R. Green, F. Kelley,
G. Marcil, T. Noone, and R. Wood (St. Bonaventure, 19971998), bk VII, q. 19, n. 58.
268 todd bates
He implies that there is some item in the thing corresponding to the
generic concept. No mention is made of its corresponding to the con-
cept of the essence. This item in the external thing I take to be the
metaphysical genus.
Again, in a text from Scotus s Ordinatio (Vaticana, 1950; I, d8, n16),
he writes: . . . in some creatures the genus and difference are taken
from one and from another thing [res], just as in positing a plurality of
forms in a man, animal is taken from the sensitive and rational from
the intellective [items in the soul]. Then, that thing from which the
genus is taken is really potential to and perfectible by that thing from
which the difference is taken. Consider the essence human for example,
and let it be dened as rational animal, with animal being the genus. We
move merely within the logical realm: the genus animal as it is a part
of the denition is a logical genus. Consider the generic concept animal
alone. As we saw in the two previous texts, Scotus implies that some
item in the external thing corresponds to the logical genus in the intel-
lectthat is, the logical genus is taken from an item which he calls
a reality. Presumably the reality is the item in the external thing to
which the logical category corresponds. I read Scotus to be referring
with reality to a metaphysical genus, animal, which is a proper part
of the external thing. On my reading, we may put Scotuss point by
saying that the real, metaphysical category, the generic entity animal in
Socrates , corresponds to the logical category, the genus concept animal
in the intellect.
On Pinis interpretation, the item in the external thing to which the
concept of the genus corresponds would simply be the essence. Indeed,
on Pinis reading, the essence in the external thing also corresponds
to the concept of the species. Making my case against Pinis reading,
I will sketch Scotuss metaphysics of the composition of substances.
Let us interpret his Categories commentary against the background of his
metaphysics of composition. If we do so, we shall see that by following
Pinis interpretation of Scotuss logic rather than my own a tension is
introduced between Scotuss logic and his metaphysics.
Note additional texts from Scotus where he explicitly implies the
essence in the thing is complex and includes its genus as an entity and
proper part. First, consider Scotuss discussion of the common nature
in his Ordinatio; Scotus writes:
I say that composition can be understood either properly, as is a com-
position out of an actual thing and a potential thing, or else less properly,
as is the composition out of an actual reality and a potential reality in
the same thing. In the rst sense, the individual is not a composite with
respect to the specic nature. For it adds no reality, because it adds neither
matter nor form nor a composite, as the argument goes on to show. In the
second sense, the individual is necessarily a composite. For the reality
the specic difference is taken from is potential with respect to the reality
the individual difference is taken from, just as if they were two things.
For the specic difference does not have the wherewithal to include
through an identity the individual reality. Instead, only some third factor
includes both of these through an identity.
28
Note that in this text, Scotus recognizes a type of composition proper
to realities, such as the generic and difference entities. This is a
composition in the external, substantial thing, not a logical relation in
the intellect. In this long quote, Scotus implies that the generic real-
ityrealitas in Scotuss textis a proper part of the external thing along
with the difference-reality.
According to Scotus these realities are related to each other as potency
to actthe generic entity animal is in potency to the entity of the dif-
ference, rational, so that the two may be unied, made one. Animal is
perfected by being unied with rational in that the two realities together
compose a third item, a thing (res). Unied, neither reality is actual with
its own proper entity. For instance, in an actual composition, even the
difference rational, which is actual with respect to the generic reality
animal, is in turn potential in that it can stand as a proper part in a
whole with the genus. This whole, in the case of rational and animal, is
the essence human.
29
Having argued that Scotus considers the genus of a metaphysical
essence to be a proper part of that essence, I shall make a second
argument from Scotuss texts in the Ordinatio showing that he considers
not only the immediate genus in the denition of the essence to be a
proper part of that essence, but also that he takes a series of genera
up to the highest genus substance to be proper parts of the essence.
Scotus means to extend his picture of the composition of the essence
28
Scotus , Ordinatio, II, d. 3, qq. 56, n. 189, trans. Spade , Five Texts on the Medieval
Problem of Universals (Indianapolis, 1994).
29
There is no absurdity here that would block my interpretation of Scotus text
such that one and the same item is simply both potential and actual. For instance, the
difference is potential with respect to the whole of which it is a constituent, in that it
has the capacity to be part of such a whole. In addition, the difference is actual with
respect to the genus with which it composes a whole. Thus, the difference-reality is
not both actual and potential with respect to the same item.
270 todd bates
to include higher genera. That is, the metaphysical genus animal is in
turn a species with respect to its genus, call it G1. That is, animal
as a proper part of the essence is itself a composite consisting of G1
and some difference, call it D1, as proper parts. G1 in turn will be
a composite entity, consisting of a different genus, G2 and difference,
D2, as its proper parts. This scheme of composition continues, but not
innitelyafter a nite series of steps, we reach a stopping point with
the highest genus; in the case of material substances, this would be the
genus substance. That genus is a proper part of both the genus animal
and the common nature human. I take such a scheme of composition
to be what Scotus refers to when he speaks of a categorical hierarchy, a
Porphyrian tree consisting of species and genera: So I say the ultimate
distinction in a categorial hierarchy is the individual distinction, and
it occurs through the ultimate act that pertains of itself to the catego-
rial hierarchy. But the actual existence doe not by itself pertain to this
hierarchy. Actual existence is the ultimate act, but it is posterior to the
whole categorial hierarchy.
30
He distinguishes the items of the Porphyrian tree, the real genera
and species that are proper parts of external things, from what actually
exists. The items of the tree go to make up what he calls the whole
hierarchy of quidditative being
31
to which actual existence only con-
tingently belongs. Thus, Scotus writes This distinction [between quid-
ditative and actual existence] is accidental in a certain sense. Granted,
it is not truly accidental.
32
Thus, his discussion of the categorical
hierarchy of quiddiattive being is consistent with his threefold distinc-
tion of the signication of human. That is, the items in the quidditative
hierarchy are each specic or generic entities taken as they individuate
in themselves rather than as they exist individuate in things or as they
exist logically in the intellect.
Scotus gives some real status to items taken quidditativelythey are
not merely artifacts of the intellect. It follows, the items on this hierarchy
that are proper parts of the essence of a substance have a real status
as proper parts apart from the intellect.
Using a general principle which I take to conrm my interpretation
of him on metaphysical genera, Scotus notes that: Every quidditative
30
Scotus , Ordinatio, II, d. 3, q. 2, trans. Spade.
31
Ibid.
32
Ibid.
entity (whether partial or total) in some genus is of itself indifferent
as a quidditative entity to this or that individual entity and that one,
in such a way that as a quidditative entity it is naturally prior to this
individual entity insofar as it is this. As naturally prior, just as it does
not belong to it to be a this, so the opposite of being a this is not
incompatible with it from its very notion.
33
That is, metaphysical genera and species both have the feature of
indifference or equivalently, community, according to which they have
their own proper entity and may exist as proper parts in different
composite substances, or as concepts in the intellect. More importantly
for my case, the genus as indifferent is given its own natural priority.
That is, the genus taken absolutely as it is in itself is naturally prior to
the genus in the intellect, just as the essence taken absolutely as it is
in itself is naturally prior to the essence in the intellect. In both cases,
Scotus means to imply that the naturally prior item has its own being
apart from whatever else it may otherwise exist in, whether that be the
intellect or substances. In addition, on the basis of the genus natural
priority, I infer we must see Scotus as holding that genera are real enti-
ties in external things, just as the essence taken absolutely.
Finally, arguing again that the genus exists as a real proper part of
the metaphysical essence, I shall refer to a discussion
34
from Scotus of
genera and intrinsic modes. I do not wish to defend the soundness of
his rather arcane reasoning, but merely wish to use it to point out how
he seems to have thought of metaphysical genera. If the distinction
between a genus and difference were the same type of distinction that
obtains between a color and its intrinsic mode of luminosity, then there
would be no real composition between the genus and the difference.
At most, their distinction would be an artifact of the intellect, and we
would be confused if we took the genus as a real, distinct entity in its
own right. In fact, Scotus is careful to point out that the difference
which combines with the genus is not an intrinsic mode of the genus;
he writes (n. 16), the differentiae are not said to be an intrinsic mode of
a genus, since regardless of the grade in which animality is understood,
it is not on account of that grade that either rationality or irrationality
is found with it. . . .
33
Ibid., d. 3. qq. 56, n. 187, trans. Spade.
34
Scotus, Ordinatio, (Vaticana, 1950), I, d. 8, q. 3, nn. 1617, my translations.
272 todd bates
Scotus uses the example of a color such as a shade of white; he writes
(n. 17) [f ]or example, if there were white under the tenth grade of
intensity, however much it was simple in itself, it could be conceived
merely under the nature white. . . . In cases where the intellect distin-
guishes an item X from its intrinsic mode, Y, as with a shade of white
and its intensity, the intellect is able to conceive of X, the shade of white,
apart from its intrinsic mode. But according to Scotus , this conception
of X is confused; he writes (n. 17), if there were a distinction in the
thing as that which holds between a reality and its intrinsic mode, the
intellect could not have a proper concept of that reality and have a
[proper] concept of that intrinsic mode of the thing. . . . for while one
intellect is able to distinguish X and Y, in reality apart from the intel-
lect, there is no distinction between X and Y. Between X and Y there
is merely a rational distinction. Scotus uses the example of a color such
as a shade of white. That shade of white admits varying intensities,
such that it may be brighter or dimmer. For Scotus, the intensity of a
shade of white is an intrinsic mode of that shade of white, but there is
no real distinction in the color itself between the shade and its intensity.
Let us accept his account for the sake of argument.
In contrast, the intellects apprehension of the genus apart from the
difference is not confused. Scotus claims that rational and irrational,
which he treats here as differentiae of the genus animal, are not intrinsic
modes of animal. Rather he claims that animal, as it is in itself, is in
potentiality to either. That is, animal as a genus would stand in potency
to the differentia rationaland though they are distinct, they could
combine to make a composite. Where there is a potentiality between
items for a composition, that composition is not merely an artifact of
the intellect. Thus, Scotus implies that genera are real, not merely
conceptual, proper parts of external things and of essences. That is,
for instance, the essence human in a substance has the genus animal as a
proper part, and that genus constitutes the essence with the differentia
in a reality/reality composition.
On the sketch of Scotus s metaphysical picture of the composition
of material substances I have offered, Scotus believes the essence in the
substance is not simple, but contains the differentia taken absolutely and
its genus taken absolutely as proper parts. Not only that, the essence in
the thing contains a set of genera as proper parts, from the rst genus
to the highest genus, substance. Given this metaphysical complexity on
the side of the external thing, it is odd to hold, as Pini does, that there
is no correspondence between our logical generic concepts and the
metaphysical genera in things.
One might have received the impression that the essence in the thing
lacks proper parts from Pini s reading of Scotus s logic. Pini nds that
the essence alone corresponds to all of our logical concepts, whether
generic or specic. On Pinis interpretation, Scotuss logic implies no
metaphysical complexity in the essences of external things. Nevertheless,
as I have argued, Scotus considers the essence in the substance to be
complex, consisting of genera and differences as its proper parts. We
would do well, I suggest, to interpret Scotuss logic with charity, in such
a way that we do not introduce any tension between his logic and his
metaphysics, provided such tension can be avoided on an alternative
interpretation.
What prevents us from admitting a correspondence between logical
and metaphysical genera, thereby eliminating the tension between his
logic as Pini sees it and his metaphysics? We took note above of several
texts in which Scotus seems explicitly to afrm Pinis interpretation. I
shall offer an alternative reading of Scotuss texts congruent with my
reading of Scotus, in hopes that under my alternative reading, we may
not be obstructed by these problematic texts from afrming a corre-
spondence between logical and metaphysical genera.
First, consider this text: Similarly, the intellect considering a thing,
through that one species, can reect thousands of times on itself about
its operation, in considering, and any consideration is something having
nothing extrinsic corresponding to it, except only the rst object for the
occasion, inasmuch as that rst moves the intellect to consider.
35
It may seem that Scotus means to say here, in line with Pini s reading,
that the essence alone is a sufcient foundation for all of our logical
concepts, including those we have of genera. However, there is another
way to read Scotus. I concede that no generic thing (res) corresponds
to the logical genus in the intellect. As we saw above, the genus taken
absolutely is not in Scotuss sense a thing, but a reality (realitas). It is a
different type of item in Scotuss ontology.
If we take Scotus to mean there is only one thing that is a founda-
tion for our concepts, we may agree, while insisting that Scotus is not
implying that there is no other foundation. Yes, he implies no other
thing or res is a foundation, but not that there is no other founda-
tionhe leaves open that there may be another foundation not in any
thing or res, but in what Scotus calls reality. As a genus is a reality in
Scotus ontology, there is no inconsistency in his maintaining that both
35
Scotus , Quaestiones super Praedicamenta Aristoteles, q. 3, n. 13.
274 todd bates
the essence and the genus are together the foundation of our logical
concept of a genus. The essence is the thing which is the foundation
of our logical concept, while a metaphysical genus is the reality which
is the other part of the foundation of our concept.
Second, consider this text: Thus, a lesser unity sufces in the thing
than the unity of the intention. For it sufces that the intellect is moved
by something extrinsic to cause many things through consideration,
to which there do not simply correspond many things in the thing.
36

Presumably, we should read Scotus to mean here that the metaphysical
essence moves the intellect as an occasional cause. Again, I emphasize
that the genera I claim to be real proper parts of the essence are not
things on Scotuss ontology, but a different type of entity: realities. It
may be true that no thing corresponds to our logical concept of a genus,
while some item that is not a thing does correspond to the logical genus,
a reality. Provided we do not accept the notion that according to Scotus
an entity must be a thing, such passages as the ones above can be read
so as to be consistent with my interpretation. That is, I can reply in
the face of such passages by conceding that no thing corresponds to
the logical concept, but only an item that is a reality.
Having argued that Scotus accepts the reality of metaphysical genera
as proper parts of essences and removed certain textual impediments to
accepting a correspondence between logical and metaphysical genera,
I recall Scotuss answer in his Categories commentary to the worry that
our logical concepts are ctions, condent that I can offer a positive
reason to accept the correspondence I propose as Scotuss intention. The
charge that our logical concepts are ctions arose from their possessing
a universality matching nothing outside the intellect; the community
essences and genera possess in themselves is not universality for Scotus.
In response, Scotus drew on the theory of a threefold signication of
the term human. According to that theory, the essence human taken
absolutely as it is in itself has neither universality nor particularity.
However, it can have either universality or particularity as accidents,
though not in exactly the same way. Thus, the essence in the intellect
is the very same entity as the essence in the substance, despite the fact that
as it is individuated in the thing the essence is individual and in the
intellect the essence is universal. As the essence in the intellect is the
36
Ibid.
same entity as the essence in the thing, and the essence in the thing is
not a ction, neither is the essence in the intellect.
But on Pini s interpretation of Scotus , this reply will not work for
our concepts of genera. If there is no generic entity in the thing to
which a generic concept may correspond, but only the essence, there
is no entity outside the intellect with which our generic concept can be
one. Thus, on Pinis interpretation, it seems to me that Scotus does not
have a reply to the charge that our generic concepts are ctions, or at
least he does not have the reply that he makes on behalf of essences.
If we admit that Scotus intended our generic concepts to correspond
with generic entities, we may retain his reply to the charge of ction
for our generic concepts.
HOW IS SCOTUSS LOGIC RELATED TO HIS METAPHYSICS?
A REPLY TO TODD BATES
1
Giorgio Pini
It is usually contended that one of the central tenets of Scotuss meta-
physics is what I will refer to as the formalistic view. According to
this view, reality is sliced up in items that closely parallel the concepts
by which we understand it. So for example an individual such as John
is constituted of an individual property that he does not share with
anything else (call it Johneity) and a specic component that he has in
common with other human beings (humanity). The specic component
humanity is in turn constituted of a generic component that John shares
with other animals (animality) and an item that differentiates John as
an animal of a certain kind (rationality). The claim of the formalistic
view is that we can go on with this analysis until we reach the ultimate
kind to which John belongs, i.e., substance. To this ultimate kind there
corresponds in John a certain item called substantiality. The reason
I call this view the formalistic view is that each of Johns components
(his individual property, humanity, animality, rationality, and so on up
to substantiality) is called, in Scotuss jargon, a formality. Formalities
have two dening features. The rst dening feature of a formality
is that it cannot exist independently, i.e., without the individual that
it constitutes. The second dening feature of a formality is that it is
nevertheless distinguished from the individual that it constitutes and
from the other formalities within that individual no matter whether we
think of it as a separate constituent of reality or not. In Scotuss terms,
each formality has some degree of mind-independent unity.
The formalistic view is a metaphysical view, i.e., it speaks of how
reality is structured independently of the way we understand it. Todd
Bates argues that Scotus subscribes to this view in the form that I have
described. As I will make clear at the end of this paper, I think that
1
I wish to thank the editor of the present volume, Lloyd Newton , for giving me the
opportunity to answer Todd Batess criticism to my book Categories and Logic in Duns Scotus.
An Interpretation of Aristotles Categories in the Late Thirteenth Century (Leiden, 2002).
278 giorgio pini
Batess arguments in favor of Scotuss endorsement of the formalistic
view are not conclusive.
2
But this is a secondary point in the disagree-
ment between Bates and me. The main point of disagreement is that
Bates maintains that Scotuss logic, and specically his interpretation of
Aristotles Categories, implies this metaphysical view; by contrast, I think
that it does not. So what is at issue in Batess criticism of my inter-
pretation of Scotus is not whether or not Scotus holds the formalistic
viewa point about which I did not take a position in my interpretation
of Scotuss Categories commentary. For the sake of discussion, in what
follows I will assume that this is indeed Scotuss position. As I said,
this is not the issue of our disagreement. What is at issue is whether
Scotuss logic implies this view.
On the face of it, Batess argument has a strong appeal. On the
one hand, there is the world, which is constituted of mind-indepen-
dent formalities. Accordingly, these formalities can be seen as modes
of being (modi essendi ). On the other hand, there are our concepts, by
which we understand the world. The concepts can be seen as modes
of understanding (modi intelligendi ), i.e., the way we understand the
world. Now it is the claim of the formalistic view that to each concept
in our mind (e.g., the generic concept animal ) there corresponds a
certain formality in reality (i.e., the formality animality). So we can say
the modes of being closely parallel the modes of understanding. Why
is there such a parallelism? Clearly, our modes of understanding are
shaped to mirror the modes of being on which they are grounded. If
our modes of understanding did not mirror the modes of beingit
is claimed, they would not be representations of reality and as a
consequence they would not fare better than ctitious concepts such
as chimera, which have no correlate in extramental reality. Since logic
is the science that deals with our modes of understanding, we seem to
be entitled to conclude that, according to Scotus, logic is grounded on
metaphysics. Specically, Scotuss logic implies the formalistic view.
Bates admits that some passages in Scotuss Categories commentary
seem to favor my interpretation. But he also maintains that, in order to
avoid the alleged tension between Scotuss logic and his metaphysics, we
had better adopt a charitable reading of his Categories commentary.
2
I also think that Scotus endorsed a metaphysical view of the way in which real-
ity is structured more sophisticated than the rather crude form of Platonism that the
formalistic view amounts to. But I will not argue for this point in this paper, as it is
irrelevant to the defense of my interpretation of Scotuss Categories commentary.
scotuss logic and his metaphysics? reply to bates 279
The charitable reading that Bates upholds is supposed to eliminate any
tension and to eliminate the apparent anomaly of Scotuss Categories
commentary in the light of his metaphysical teachings.
So the issue that Bates raises is serious and central to the interpreta-
tion of Scotuss thought, as it concerns nothing less than the relation-
ship between Scotuss logic and metaphysics. The question is: Is Scotus
committed to a specic metaphysical view in his logic? My claim is that
he is not, and that this is one of the great attractions of his view. By
contrast, Bates argues that Scotus, even when doing logic, is commit-
ted to a specic metaphysical view, i.e., the formalistic view, and that
if we fail to recognize this, there arises a problem of coherence both
within Scotuss logic (i.e., generic concepts fall short of any foundation
in reality and must be considered as ctitious) and, more in general,
between his logic and metaphysics.
Before addressing Batess criticism of my interpretation of Scotuss
Categories commentary and of his logic in general (or what Scotus and
his contemporaries called logic), I have to address a preliminary
problem, which will turn out to be quite revealing of what I take to be
Batess misunderstanding of both my position and, more importantly,
Scotuss position. This problem concerns the way Bates presents my
interpretation of Scotuss logic. As it happens, Bates presents two
different accounts of my position. At times, he says (correctly) that I
maintain that what Scotus says in his Categories commentary does not
commit him to any metaphysical view concerning the way the world
is. Specically, I maintain that Scotus, in his logic, does not need to
deal with the problem whether essences are simple entities or whether
they are divided into formalities (e.g., whether human being is a simple
entity or whether it is constituted of two mind-independent formalities,
animality and rationality). Scotuss logic, we can say, is metaphysically
neutral. However, the view against which Bates argues in most of his
paper is not this one. What he apparently attributes to me in most of
the paper is the claim that Scotuss logic implies a metaphysical view
that is incompatible with the metaphysical views that he presents in
his most celebrated works, i.e., his commentary on the Sentences and,
we can add, his commentary on the Metaphysics. So this time the claim
that Bates attributes to me is that Scotuss logic implies the view that
extramental essences are simple and not divided into further mind-
independent formalities. It is this claim, not the claim that Scotuss
logic is metaphysically neutral, that posits a tension between Scotuss
logic and metaphysics.
280 giorgio pini
The two claims that Bates attributes to me are clearly different. As
a matter of fact, I am ready to defend the rst, but not the second
one. It is one thing to say, as I say, that Scotuss logic does not require
that essences are compound (for it does not require that essences are
simple either; in his logic Scotus is non-committal with regard to this
issue). It is another thing to say, as Bates has me saying, that Scotuss
logic requires that essences are not compound. Saying that a does not
entail b is clearly different from saying that a entails not-b.
So what are we to make of this preliminary problem? Does it mean
that Batess criticism is directed against a straw man and that I can
ignore it as irrelevant to my interpretation of Scotus? In fairness to
Bates, I think that, should I take this line of defense, I could be charged
with avoiding the real issue. For it can be contended that what Bates
actually means to say is that, even though I explicitly endorse the
view that Scotus, in his Categories commentary, is not committed to
any metaphysical view, what I attribute to Scotus as a matter of fact
commits him (I suppose, unbeknown to me) to a certain metaphysical
view, which is at odds with his metaphysical positions as he states them
elsewhere. So, I will have to argue that my reconstruction of Scotuss
Categories commentary does not surreptitiously imply any commitment
to a specic metaphysical view on Scotuss part. Specically, Scotus
does not have to decide whether essences are simple or compound of
formalities, contrary to what Bates thinks.
Accordingly, I construe Batess criticism as consisting of the follow-
ing three claims:
(1) My reading of Scotuss Categories commentary implies a certain
metaphysical view, namely, the view that extramental essences are
simple entities.
This reading must be rejected for the following two reasons.
(2) If Scotus is indeed committed to the view that essences are simple
entities, he does not have an answer to the charge that generic
concepts are ctitious (and so a concept such as animal would be
grounded in reality no more than a concept such as chimera).
(3) If Scotus is indeed committed to the view that essences are simple
entities, what he says in his Categories commentary is at odds with
his standard metaphysical views as stated in other works, where
he seems to be committed to the idea that essences are entities
compounded of distinct formalities.
Accordingly, it will be enough for me to argue against point (1). If it
is not the case that my interpretation of Scotuss logic commits him
to any specic metaphysical view (and so it is compatible with more
than one, including the formalistic view), the second and third point
lose the premise on which they are grounded and so do not follow. All
the same, for the sake of clarity, I will also say something concerning
point (2) and (3).
1. Logic as a Second-Order Activity
So let me begin with the rst point. Bates thinks that my reading of
Scotuss Categories comentary commits him to the metaphysical view that
extramental essences are simple, so that, for example, the essence human-
ity is not composed of two distinct formalities, rationality and animality.
In my opinion, Batess criticism arises from his failure to appreciate
the signicance of Scotuss view that logic is a second-order activity,
concerned with second-order concepts such as species, genus, and universal.
First-order concepts such as human being and animal interest the logician
only to the extent that they are the subject of second-order concepts,
not as they are applied to extramental things. Thus, in Scotuss and his
contemporaries opinion, the logician is typically concerned with state-
ments such as Human being is a species (homo est species), Animal is a
genus (animal est genus), Rational is a differentia (rationale est differentia),
not with statements such as Socrates is a human being, Human being
is an animal, Human being is a rational animal. To the extent that
logic is a second-order activity dealing with second-order concepts, it
can remain non-committal about the specic way in which rst-order
predications are carried out, i.e., the way in which rst-order concepts
such as human being, animal, and rational are said of extramental things
and capture aspects of reality.
3
In order to see that this is the case, we must recall that Scotus, like
most of his contemporaries, subscribes to the view that Aristotles
3
The distinction between rst and second intentions is clearly drawn by Scotus,
among other places, in his Questions on the Metaphysics, VII, q. 18, nn. 3839 (OPh IV,
p. 347), with regard to the term universal. Logicians are concerned with the second-
order concept universal, metaphysicians are concerned with rst-order concepts such as
human being and animal, insofar as they capture an aspect of reality.
282 giorgio pini
Categories is a work of logic.
4
Similarly, and again like most of his con-
temporaries, he thinks that this implies that the Categories deals neither
with things in the world (such as human beings or horses) nor with
our concepts of such things insofar as they represent such things. The
latter point is worth stressing, as I think that it is its misunderstanding
that caused Batess criticism of my interpretation. Scotus subscribes to
a view that was pretty standard in his time. This view, derived from
Avicenna and ultimately from Alexander of Aphrodisias, held that what
we think of can be considered in two ways.
5
First, it can be considered
as the object of our thought and as what we think of, independently of
the fact that we think of it. In this rst consideration, what we think
of is viewed as something in the extramental world. Second, what we
think of can be considered not as it is something in the world but just
insofar as it is the object of our thought, i.e., insofar as we think of it.
For example, let us posit that we think of what human beings have in
common (without being concerned with its specic ontological status).
We can consider such an entity in two ways. First, we can consider
what we think of, i.e., the common nature or property human being as
something independent of the fact that we think of it or not. Second,
we can consider the same nature or property just with respect to the
fact that we think of it. This distinction, commonly adopted in Scotuss
times, is quite important to appreciate how metaphysics differs from
logic in the understanding of Scotuss contemporaries. For the same
object has different sets of properties according to whether it is con-
sidered in the rst or in the second way. Specically, when the nature
4
See Categories and Logic, pp. 1944, as well as my contribution to the present
volume.
5
See G. Pini, Absoluta consideratio naturae: Tommaso dAquino e la dottrina avicen-
niana dellessenza, Documenti e studi sulla tradizione losoca medievale 15 (2004): 387438.
In what follows I give a simplied description of this view, which Scotus presents,
among other places, in his Isagoge commentary, q. 911, n. 21 (OPh I, p. 46). See Pini,
Categories and Logic, pp. 100102. In the complete picture, what we think of can be
considered in three, not just two ways: as it is exists in individuals, as it is in itself, and
as a concept in the mind. Here only the last two considerations are relevant. It must
also be noticed that Bates takes the passage from Scotuss Isagoge commentary I have
just referred to as evidence that Scotus subscribes to his famous doctrine of essence,
as classically stated in his commentaries on the Sentences, already in his logical writings.
This is not correct. What Scotus provides in the passage from his Isagoge commentary.
is the standard doctrine taken from Avicenna, which can be found in most authors
of the thirteenth and early fourteenth century, including Thomas Aquinas. What is
peculiar to Scotus in his mature writings is the claim that the essence, as considered in
itself, has a less-than-numerical unity quite independently of our minds consideration.
I nd no trace of this latter claim in Scotuss logical writings.
human being is considered independently of whether we think of it or
not, it has properties such as being mortal, being rational, and being
an animal. By contrast, when the nature human being is considered with
respect to the fact that we think of it it has the property of being a
species and being universal. These two sets of properties are clearly
different. On the one hand, we have properties such as being John, being
a human being, being an animal, being a substance. On the other hand, we
have properties such as being an individual, being a species, being a genus,
being a category. Can we account for the diversity between these two sets
of properties? Thirteenth-century thinkers thought that we can, as we
have seen, on the basis of the subject to which these properties are
attributed. The rst set of properties is attributed to what we think of
as it is a component of the extramental reality and independently of
the fact that we think of it. The second set of properties is attributed
to what we think of insofar as we consider it as what we think of. The
rst consideration is what Scotus and his contemporaries call a real or
metaphysical consideration, and the properties connected to it are real
or metaphysical properties. The second consideration is what Scotus
and his contemporaries call logical consideration, and the properties
connected with it are called logical properties. The same points can
be made by using the sometimes discredited vocabulary of rst and
second intentions. Our thoughts concerning the world are called rst
intentions. These thoughts can be considered either as they concern
the world or as they are thoughts that we have. According to the rst
consideration, rst intentions are considered by the metaphysician,
who studies them as representations of items in the mind-independent
world. According to the second consideration, rst intentions are con-
sidered by the logician, who regards them insofar as they are concepts
and as they are the subjects of properties that pertain to them as they
are concepts. These logical properties are called second intentions.
Accordingly, the metaphysician studies rst intentions (i.e., concepts)
such as human being as they represent the extramental reality. The logi-
cian studies rst intentions such as human being only to the extent that
they are subjects of second intentions, i.e., of properties such as being
a species and being universal.
If we apply all this to the Categories, Scotus and his contemporaries
maintained that, since the Categories is a work of logic and not of meta-
physics, whatever Aristotle says about substances such as human beings
or accidents such as white should be interpreted as referring to these
substances and accidents insofar they are known by us and, moreover,
284 giorgio pini
not as they represent something in the extramental world but as they
are the subjects of logical or second-intention properties such as being
a species, being a genus, being universal. As I have said, it was commonly
agreed that the latter properties are properties not of the human beings
or animals we think of but of the concepts in our minds by which we
think of human beings or animals. In other terms, second-intention
properties such as being a species or being a genus were considered not as
properties of things or modes of being (modi essendi), but as ways in
which we look at things or modes of understanding (modi intelligendi ).
So far, there was agreement among thirteenth-century thinkers. The
disagreement started when the cause of these second-intention proper-
ties was inquired. Some thought that the main cause of properties such
as being a genus is the way things are. Our mind merely abstracts the
concept of genus by considering the real properties in things.
6
Others
thought that the main cause of properties such as beings a genus is our
mind. In extramental reality there are human beings and animals, not
species and genera. Species and genera are concepts that our mind
produces when it focuses on the way in which it conceives and classi-
es things in the extramental world. Admittedly, things may be classi-
ed into the same species or into the same genus because they share
some common real properties. But the supporters of this second view
contended that this is not relevant when we are doing logic. When
we say that both horses and human beings are animals we are indeed
taking into account their common real properties. In that case, we are
doing metaphysics, and we may indeed wonder about the ontological
status of those properties. In logic, however, we are not interested in
these properties, but in the fact that the concept of human being is
a species and the concept of animal is a genus. Why is this the case?
The supporters of this view maintained that the answer does not have
anything to do with the fact that all human beings existing in the world
have some real properties in common. By contrast, they maintained that
the correct answer is that the concept of human being is predicated of
some entities (individuals) while at the same time some other concepts
(generic concepts) are predicated of it. Even though the ultimate reason
why a concept is predicated of something may lie in the metaphysical
structure of things, all we need to know in logic is not why a concept is
predicated of something, but just that and how (i.e., with which degree
6
See Categories and logic, pp. 6898.
of universality) it is predicated of something. And we can know that
this is the case without inquiring into the metaphysical structure of
the world, but merely by comparing concepts and their predicability.
So whether a concept is the subject of the second-intention concept
of species is decided on the basis of a comparison among concepts,
not a consideration of real properties of things. For this reason, the
supporters of this view maintained that our intellect, and not the way
things are, is the main cause of second-intention concepts.
7
Scotus is among the supporters of this view (but he is not alone,
contrary to what Bates seems to think; as I stressed in my book, he
shares company with Thomas Aquinas and Peter of Auvergne). This
view has a very interesting consequence. We can speak of concepts
such as species and genus while remaining metaphysically neutral. In
other terms, when we are doing what Scotus and his contemporaries
call logic, we do not have to take a stance concerning the way things
are in the world, for we are talking about the modes or ways things
are understood, not the modes or ways things are in the world inde-
pendently of the way we think of them.
It is this point that Bates does not accept. He contends that, even
when doing logic, Scotus cannot remain metaphysically neutral. But
I think that this criticism stems from a failure to understand what the
subject matter of logic is, according to Scotus and his contemporaries.
Logicand consequently, Aristotles Categoriesis not concerned with
concepts such as human being and animal insofar as they represent
extramental things. These concepts are dealt with in logic only to
the extent that they serve as foundations of concepts such as species
and genus. So Scotus (and in this respect he is in complete agreement
with all his contemporaries) thinks that the real focus of Aristotles
Categories is on concepts such as individual, species, and genus. John, human
being and animal are dealt with only to the extent that they serve as
foun-dations to such concepts. It is consequently not relevant to say
as Bates doesthat if there is no common formality corresponding
to our generic concept of animal such a concept is ctitious. This is
not what Scotus is concerned with in his Categories commentary. If he
were, he would be doing metaphysics, not what he calls logic. What
Scotus is interested in it is not generic concepts such as animal, but
second-intention concepts such as genus. First-intention concepts such
7
Ibid., pp. 4567, 99137.
286 giorgio pini
as animal are interesting for him only insofar as they play the role of
foundations of second-intention concepts. But it is quite irrelevant to
consider their real foundation in extramental reality (i.e., whether there
is a mind-independent formality or instead there is something else cor-
responding to the concept of animal).
Bates seems to be confusing rst-intention concepts such as animal
with second-intention concepts such as genus. This is clear from his
frequent reference to the distinction between logical genus and meta-
physical genus. From what I can understand, he thinks that this dis-
tinction amounts to the distinction between a generic concept such as
animal (which, as a concept, is in our mind, but nonetheless represents
something outside our mind) and its foundation in reality, i.e., what
Bates takes to be the formality animality. So the distinction between
logical and metaphysical genus for Bates amounts to the distinction
between a concept representing something outside the mind and the
feature in reality that corresponds to that concept. To my knowledge,
however, there is no talk of this kind in Scotus. He does talk of the
distinction between a generic concept and its correlate in reality. But
such a distinction is not drawn in terms of logical versus metaphysi-
cal genus.
8
When Scotus speaks of genus in logic, he usually intends
to refer to the second-intention genus, not the concept animal. Also,
Scotus does sometimes speak of a logical and a metaphysical consid-
eration of terms, for example when he speaks of the term being or
the term body as considered logically or metaphysically.
9
But when
Scotus speaks in this way, he is not contrasting two distinct entities (a
concept in the mind and a thing in reality), as Bates maintains. He is
merely distinguishing between two ways of considering the same term,
i.e., insofar as it signies something outside the mind and insofar as it
signies something in our mind. It is because Bates fails to appreciate
this point that he thinks that Scotus should address the question of the
real foundation of generic concepts in his logical works. Bates thinks
that Scotus is talking about concepts such as animal. Consequently,
8
See Scotuss Questions on the Metaphysics, VII, q. 19 (OPh IV, pp. 356379). The
contrast between metaphysical and logical genus can indeed be found in the De rerum
principio. But this work, present in Waddings and Vivss editions of Scotuss Opera
omnia, is not by Scotus. Even there, however, the distinction concerns being and the
categories, not the relationship between a generic concept and its real correlate. See
C. R. S. Harris, Duns Scotus (Oxford, 1927), vol. 2, p. 59.
9
See for example Scotuss Questions on the Sophistical Refutations, q. 15, n. 20 (OPh II,
pp. 337338).
the question naturally arises, what does correspond in reality to such
concepts? But this is not what Scotus is interested in in his logical
works. Scotus is interested in concepts such as genus and in concepts
such as animal not as they represent something outside our minds but
only to the extent that they serve as foundations for concepts such as
genus. Now, Scotus explains in clear terms that a concept such as genus
is caused by the intellect by way of a comparison between concepts,
e.g., animal and human being. These two concepts are compared only to
the extent that they have a more or less universal applicationanimal
has a wider extension than human being. But the reason why animal has
a wider extension than human being is of no concern to the logician.
To ask why animal has a wider extension than human being is indeed a
very good question. But it is a metaphysical question. The logician can
be content with sorting out concepts into generic and specic with no
concern for the metaphysical reason why this is so. The logician only
needs to know that a concept is generic if it is predicated of other
concepts that are classied as specic. And a concept is specic if it
is predicated of other entities that are classied as individual and if
concepts classied as generic are predicated of it. But the logician does
not have to take any stance about the deep metaphysical reason why a
concept such as animal is predicated of more items than a concept such
as human being. Granted, there must be a reason. But to ascertain what
this reason is, is the concern of the metaphysician. As a consequence,
Scotus does not have to commit himself to any specic metaphysical
view in his logic. Of course, he does have to assume that there is an
ultimate metaphysical reason why a concept such as animal has a larger
extension than a concept such as human being. But, while doing logic, he
does not have to concern himself with any of the available explanations.
It is enough for him to know that there is an explanation. It is when
doing metaphysics that he will have to do so. This is most clear in this
important passage from Scotuss Categories commentary:
I say that the <extramental> thing is not the whole cause of an inten-
tion. It is only its occasion, i.e., to the extent that it moves the intellect
to the actual consideration <of something>. It is the intellect that is the
principal cause. Therefore, in reality there need not be as much unity
as the intention has, because it will be sufcient that the intellect will be
moved by something external to cause several <concepts> by its own
act of considering <that same external thing>. And to <each of> these
concepts there does not correspond a separate entity in reality (aliqua in
re) . . . the intellect, when it considers <something> in virtue of the one
and only image <that it has received from the extramental thing>, can
288 giorgio pini
turn to its own operation a thousand times by its act of considering <that
operation>. Each act of considering is something, but it has nothing external cor-
responding to it except the rst object as an occasion (quaelibet consideratio aliquid
est, nihil habens extrinsecum sibi correspondens, nisi tantum primum obiectum pro
occasione), to the extent that <that object> rst moves the intellect to the
act of considering.
10
(Italics mine)
Scotus does not need to posit a separate entity in the extramental
reality corresponding to each separate second intention. So there
need not be separate entities corresponding to the concepts species,
genus, differentia. All Scotus needs, in order to rule out the charge that
these concepts have no ground in reality and are therefore ctitious,
is that there is some extramental thing corresponding to them. But no
additional requirement is needed concerning the internal structure of
this external thing (the occasion of second intentions). To say, as Bates
does, that Scotus does not need a separate external thing correspond-
ing to each of these concepts but nevertheless he needs an external
formality is to miss the point. This may be true when Scotus is talking
about rst-intention concepts such as human being and animal. But this
is not what Scotus is talking about here. In my opinion, Batess criti-
cism fails to take into account the subtle but not at all gratuitous way
in which Scotus and his contemporaries distinguished between logic
and metaphysics. When Bates thinks that he is talking about logic, he
is actually talking about what Scotus and his contemporaries would call
metaphysics. The reason why he thinks that Scotuss logic is committed
to a specic metaphysical view is that he mistakes what Scotus would
call metaphysics for his logic.
Batess difculty with grasping the full extent of the idea that logic
is a science concerned with second-intention concepts and that as such
it is not committed to any specic metaphysical framework is not new.
As it happens, Scotus himself reports and criticizes a view that is very
similar to Batess argument.
11
Some could arguesays Scotusthat
logic deals with concepts, but, since concepts are in turn signs of
things, logic must also consider the way in which things are in the
world. In other words, the claim is that logic, even though it is a sci-
ence of concepts or modes of understanding, cannot remain neutral
10
Scotuss Categories commentary, q. 3, n. 13 (OPh I, 270271). I have modied the
translation I provided in Categories and Logic, p. 112.
11
I am grateful to Lloyd Newton for bringing this point to my attention.
with regard to modes of being, namely, the structure of the things of
which logical concepts are signs. The argument in favor of this claim
is straightforward. A sign is a relative item; a relative item is known
only if its correlate is known; since logicians study concepts as signs of
things, they must assume some specic explanation about how things
are. Interestingly, Scotus remarks that this argument is an objection to
the very view that logic is a science dealing with concepts. We may say
that this claimwhich bears a striking resemblance to Batess criticism
to my interpretation of Scotuseliminates the distinction between logic
and metaphysics. Scotuss reply is that this claim is based on a fallacious
argument. If we intend to know a relative item we must indeed know
its correlative term. However, we do not need to know the correlative
term according to all its features, as the objection assumes, but only
with regard to the properties that inhere in it insofar as it is correlated
to the relative item we are studying. Specically, logic deals with con-
cepts that are ultimately signs of things. It is true, therefore, that logic
must deal with things. But this does not mean that logic must deal with
the structure of things considered in themselves and is consequently
committed to a certain metaphysical view. Things are considered only
with regard to the properties that pertain to them as they are signied
by concepts. These properties are second-intention properties such as
being universal, being predicable, being a subject. Even though it
can be argued that the extramental things to which these properties
pertain must have a certain metaphysical structure, Scotuss point is that
it is the task of the metaphysician, not of the logician, to study that
structure. Logicians can and must remain metaphysically neutral unless
they intend to stop doing logic and start doing metaphysics.
12
12
Scotuss Categories Commentary, q. 1, nn. 2226 (OPh I, 255256): Contra: con-
ceptus est ulterius signum rei; igitur adhuc sequitur quod oportet rem cognisci. Istud
argumentum non est tantum contra Boethium, sed etiam contra dicentes logicam esse
de conceptibus. Ideo potest dici quod non oportet propter unius relativi cognitionem
alterum cognosci quantum ad omnia quas sibi insunt in se, sed tantum quantum ad illa
quae insunt ei in quantum refertur ad aliud. Hoc autem modo non est inconveniens
rem cognosci in logica in quantum est signicatum per conceptum. Batess position
is strikingly similar to Radulphus Britos view, which is quite different from Scotuss
one, as I argue in my book. See Pini, Categories and Logic, pp. 168169.
290 giorgio pini
2. Why Logical Notions Are Not Fictions
Batess second point concerning my reading of Scotuss Categories com-
mentary can be dealt with more briey. Once it is clear that it is not
Scotuss job in his logic to say anything about the real foundation of
our rst-intention concepts, it follows that we should not look there for
a justication that generic concepts such as animal are not ctitious, i.e.,
have no foundation in reality.
Bates contends that, in my reading of Scotuss Categories commentary, I
cannot rescue Scotus from the charge that a generic concept is ctitious.
Why a generic concept and not, for that matter, a specic concept?
Because Bates thinks that my reading implies that Scotus is committed
to the view that the essence of an extramental thing is a simple entity,
not composed of further formalities. So, in the extramental world,
there should be just essences such as humanity, and that these essences
are not further composed of formalities such as animality and rationality.
As a consequence, a specic concept such as human being is grounded
in reality, for there is a real entity corresponding to it, i.e., the essence
humanity. But a generic concept such as animal is not grounded on any
such real foundation, because (in Batess rendering of my interpretation)
there are no distinct components of the world corresponding to it.
As is clear, Bates thinks that what Scotus is talking about is rst-
intention concepts such as animal. Moreover, he assumes that if there
is no separate entity corresponding to such concepts, they fare no bet-
ter than ctitious concepts such as chimera. This latter claim seems to
be unjustied, but I do not intend to focus on it. What I would like
to stress again is that, in his logical writings, Scotus is interested in
generic concepts such as animal not as they represent real features of
the world but as they play the role of foundations of second-intention
concepts such as genus. Consequently, what he is worried about is that
second-intention concepts, being representations not of ways things
are but of ways of understanding, may turn out to be ctitious just as
the concept chimera. It is this worry concerning second intentions that
Scotus evokes in his Isagoge commentary. And it is to this worry that
he answers with regard to the second-intention concept universal (but
what he says can be easily applied to other second intention concepts
such as genus and species). Such second-intention concepts are not cti-
tious, Scotus claims, because, unlike a concept such as chimera, they
are founded on rst-intention concepts, which, in turn, are grounded
on realityeven though it is not insofar as these latter concepts are
grounded in reality that they are the subjects of second intentions.
13

For example, the concept genus is not ctitious, because it is grounded
on the rst intention concept animal, which in turn is grounded on
real animals. How is the rst intention concept animal grounded on
real animals? What is the relationship between our concept (generic,
specic, or whatever) and extramental reality? Scotus does not approach
this topic here, nor does he need to. All he needs to say to differentiate
second intentions from ctitious concepts is that there is something in
reality that plays the role of a necessary condition for their existence.
This necessary condition is indeed a real property existing in the world
independently of the way in which we consider it. But as to what this
property is (i.e., whether it is really, formally or only rationally different
from the reality from which the species and the specic difference are
taken), Scotus does not have to say anything. This is an issue for the
metaphysician to tackle. The logician only needs to know that there
is something, but whether it is an essence considered in abstraction
from some of its feature or a separate formality or anything else, this
is not something that is of any relevance for him. This is the reason
why Scotus, in his logical writings, does not deal with the question of
what corresponds in reality to rst-intention generic concepts such as
animal. As a matter of fact, he does not deal with the question of what
corresponds in reality to rst-intention specic concepts either. Likewise,
there seems to be no mention, in Scotuss logical writings, of his doctrine
of the less-than-numerical unity pertaining to essences independently
of our consideration.
3. Batess Evidence for the Formalistic View
I think that what I have said so far is enough to defend my reading of
Scotuss Categories commentary from Batess criticisms. Nonetheless, Bates
briey sketches a view that, even though it is in no way opposed to what
I say in my book, is worth considering in itself. This is the metaphysical
position that I have called the formalistis view. Bates thinks that there
is good evidence that Scotus is committed, especially in his commentar-
ies on the Sentences, to the view that each individual is constituted not
13
See Scotuss Isagoge commentary, q. 4, n. 12 (OPh I, 25); Pini, Categories and Logic,
p. 110.
292 giorgio pini
only of two formalities, i.e., an essence and an individuating feature,
but of many others, for the essence is in turn constituted of a formal-
ity corresponding to the genus into which that individual falls and of
the relevant specic difference, and the formality corresponding to the
genus is in turn constituted of other formalities, up to the formality
corresponding to the category to which the individual belongs. So
for example John is constituted of his individuating difference and of
humanity, which is in turn constituted of animality and rationality, and so
on until we reach the ultimate formality substantiality. At this point, I
hope that it should be clear that this view is perfectly compatible with
my reading of Scotuss Categories commentary, as it is a metaphysical
view concerning the structure of the world and its items, and Scotuss
logic, in my interpretation, is not committed to any specic metaphysi-
cal view nor is it incompatible with any (at least, with any of the most
common Aristotelian metaphysical views, and certainly not with the
formalistic view). But it is a problem in itself worth investigating to see
whether Scotus does endorse the formalistic view. Here I will not try
to give a full treatment of this interesting topic. I will just limit myself
to a few remarks concerning the evidence Bates uses to argue that this
is the case.
It seems to me that the evidence that Bates brings in support of
Scotuss endorsement of the formalistic view is not conclusive. Bates
lays the burden of proof on an important passage in Ord., II, d. 3,
qq. 56, n. 189, which he quotes extensively. In that passage, Scotus
says that composition can be intended as a composition of different
things or as a composition of two components within the same things.
As we know from other passages, although these components of the
same thing cannot exist the one without the other, they are indeed
distinguished from each other independently of whether we think of
them as distinct or not. This is the gist of Scotuss famous doctrine
of the formal distinction. In the passage Bates quotes, Scotus argues
that the reality from which the individual difference is taken and the
reality from which the specic difference is taken are distinguished in
this way within the same individual. So, for example, in John we must
posit a mind-independent distinction between that reality from which
his individual difference ( Johneity?) is taken and the reality from which
rationality is taken. No mention is made of the relationship between
the reality from which the specic difference is taken and the reality
from which the genus is taken. But this is what Bates needs in order
to argue for his point. He may have been confused by Scotuss talk
of the specic difference, but in this passage the specic difference is
contrasted with the individual difference, not with the genus. Thus, I
cannot see how this passage, in itself, can support the formalistic view.
What it supports is clearly a composition of a specic element and
an individuating element within the same individual thing. More than
this cannot be argued for on the basis of this passage. Of course, this
passage is not incompatible with the view that more formalities are to
be found within the same thing. But that Scotus subscribed to such a
view must be argued for on the basis of evidence other than this.
14
This does not mean, of course, that no evidence can be given for
Scotuss endorsement of the formalistic view. There may be other pas-
sages more cogent than the one on which Bates bases his interpretation.
I only want to stress that no such passage is to be found in Scotuss logi-
cal commentaries. Quite the contrary, in his Isagoge commentary Scotus
holds that a generic concept such as animal and a specic difference such
as rational differ only to the extent that the intellect abstracts matter from
the difference whereas it does not from the genus. So, the difference
between the two concepts seem to amount to a distinct consideration
of the intellect (this is what Thomas Aquinas had maintained).
15
4. Conclusion
In this paper, I have responded to the main criticism that Bates moves
against my interpretation of Scotuss Categories commentary, i.e., that
my reading commits Scotus to a specic metaphysical view which is
incompatible with what I have called the formalistic view. Moreover,
I have showed that the evidence on which Bates grounds his claim that
Scotus endorsed the formalistic view is not conclusive. I hope that, in
the light of what I have said, Scotuss view on the relationship between
14
Bates also refers to Scotuss claim that genera are not distinguished among them-
selves by way of intrinsic modes, i.e., as a color is distinguished into different shades.
Bates maintains that the distinction between a color and its intrinsic mode is not real
(in the sense of mind-independent, I assume). This is at best controversial. From this,
Bates concludes that the distinction between a genus and its specic differentiae is real
(in the sense of mind-independent). I think that here Bates is making a mistake. From
if a, then not-b (i.e., if a genus and its difference are distinguished as a color and its
mode, then it is not the case that they are distinguished mind-independently) does not
follow if not-a, then b (i.e., if it is not the case that a genus and its difference are dis-
tinguished as a color and its mode, then they are distinguished mind-independently).
15
See Scotuss Isagoge commentary, q. 20, n. 5 (OPh I, pp. 128129).
294 giorgio pini
logic and metaphysics has emerged as more sophisticated and, possibly,
defensible than it is usually assumed. Scotuss logic does not imply his
metaphysics. Of course, it presupposes that some metaphysical account
of reality can be given. Accordingly, logic is not a merely conventional
discipline whose concepts are merely ctitious. But, in logic, we do not
have to commit to any specic metaphysical account of the structure
of reality. To inquire how the world is structured is the task of the
metaphysician, not of the logician. Scotus, like most of his contem-
poraries, carefully distinguishes between logic and metaphysics. Logic
is a metaphysically neutral tool that we can use well no matter where
our metaphysical preferences lie.
JOHN BURIDAN : ON ARISTOTLES CATEGORIES*
Alexander W. Hall
Inspired by William of Ockham , fourteenth-century nominalists denied
the existence of extramental universals and generally avoided reifying
things in the accidental categories.
1
This raises difculties in the face of
Aristotles ambiguous claim that things that are (ta onta)
2
comprise sub-
stances and the nine categories of accidents. Aristotle describes accidents
as dependent on substances for their existence, though the converse does
not hold. Since they are inherent in substances, accidents along with
their genera and species are said to be in substances. Again genera and
species are described as said of the particulars that are under them, thus
genera in the category of substance are said of particular substances,
while genera in the category of quality are said of particular qualities,
etc.
3
Commentators have long debated whether Aristotles things that
are are types of entities, signicative utterances, or a combination
of both.
4
For his part, Buridan maintains that Aristotles categories
comprise certain types of signicative utterances and he interprets
being said of and in as relations that hold among terms, explicable via
the semantic properties of supposition, signication, appellation, and
* I am grateful to Gyula Klima for suggestions on the orientation of Buridan s
approach, and especially to Jack Zupko for encouraging me to undertake this work,
commenting on several earlier versions, and giving freely of his time and expertise on
numerous occasions.
1
In some cases, this was not possible. For instance, the subsistence of the accidents
of bread and wine apart from the host during the miracle of transubstantiation was
generally held to require that certain accidents could exist apart from a subject through
divine agency. See Paul J. H. M. Bakker , Aristotelian Metaphysics and Eucharistic
Theology: John Buridan and Marsilius of Inghen on the Ontological Status of Acci-
dental Being, in The Metaphysics and Natural Philosophy of John Buridan, eds. Johannes
M. M. H. Thijssen and Jack Zupko , Medieval and Early Modern Science (Leiden,
2001), pp. 24764.
2
Categories (Cat) 1
a
20.
3
Cat 1
a
16
b
19.
4
See section 2. For arguments in favor of granting extramental existence to genera
and species in the accidental categories, see Gyula Klima , Buridan s Logic and the
Ontology of Modes, in Medieval Analyses in Language and Cognition, eds. Sten Ebbsen and
Russell L. Friedman, Acts of the Symposium of the Copenhagen School of Medieval
Philosophy (Copenhagen, 1999), pp. 47395.
296 alexander w. hall
connotation (Summulae de dialectica (S) 3.1.5), thus presenting a coherent
nominalist reading of Aristotles work.
As concerns these semantic properties, a terms supposition is what
it picks out or refers to,
5
and its signication is what it brings to mind.
Yet sentential context delimits signication, as wealthy in Socrates is
wealthy does not signify being wealthy in general but the state of being
wealthy as it relates to Socrates, or Socrates wealth. This signication
in relation to some individual, a product of sentential context, is a
terms appellation. Appellation thus brings to mind some general char-
acteristic inasmuch as that characteristic is related to or joined with an
individual, as white in Socrates is white brings to mind being white
as this characteristic relates to Socrates, or Socrates whiteness. Finally a
terms connotation seems to be its indirect or oblique reference, which,
in relation to a terms appellation, is the general characteristic that a
term signies independent of sentential context. Thus, in Socrates is
happy, happy supposits for Socrates while appellating the quality of
being happy (or simply happiness) as it is joined with Socrates (i.e., while
appellating happiness with respect to Socrates or Socrates happiness)
and connoting or having indirect reference to happiness in general (the
signication of happy divorced from sentential context):
Every term connoting something other than what it supposits for is called
appellative and appellates that which it connotes as pertaining to that
which supposits for, as when white appellates whiteness as pertaining to
that which the term white is apt to supposit for (S 4.5.1).
6
Prcis
This essay discusses Buridan s explication of being said of and in along
semantic lines, which enables him to read the Categories as concerned
with properties of terms rather than things in the world, effectively
nipping realist interpretations in the bud. Substance terms are essential
predicates, conceiving only their supposita
7
and thus functioning as rigid
5
This is leaving aside cases of material supposition which I discuss below (see infra
fn. 38).
6
All translations of Buridan s Summulae are quoted with permission from Gulya
Klima , trans. and ed. John Buridan: Summulae de Dialectica (New Haven: Yale University
Press, 2001), copyright 2001 by Yale University.
7
This comment only holds in the context of a proposition. Divorced from sentential
context, terms such as animal may signify a number of things, but not in the proposi-
tion A man is an animal: Although the term animal signies more [things] than the
john buridan: on aristotles CATEGORIES 297
designators.
8
Unlike accident terms they cannot cease to supposit for
their suppositum unless their suppositum ceases to exist. The syntactic
sign of this is that substance terms lack appellation, suppositing only
for their subject, they bring nothing else to mind. Thus appellation or
the lack thereof is the criterion for distinguishing between accident and
substance terms, respectively. On the other hand, generic and specic
terms are quidditatively predicable of individual terms and thus said of
the latter, which will require a looser sense of quidditative predication
in the case of generic and specic accident terms.
Section 1: Nominalism
In denying the existence of real universals, Buridan is committed to
the thesis that universal terms ultimately refer to individuals that are
conceived indistinctly under general concepts, and is opposed to real-
ists such as John Duns Scotus and Walter Burley who believe in the
existence of general natures that are in some way common to a variety
of individuals.
9
Thus for Buridan, generic and specic substantial terms
do not pick out shared general natures, instead they supply different
ways to speak of one and the same thing in terms of greater or lesser
degrees of generality:
Since universals (that are principally so called according to predication) do
not exist outside the soul, they are nothing but the concepts of the soul,
term man, nevertheless, it does not appellate over and above the signication of the
term man anything having to do with man, i.e., as something pertaining to man [ per
modum adiacentis homini ]. . . . Therefore the predication A man is an animal is essential.
Buridan , Summulae de Dialectica (S ) 2.5.2.
8
See Klima , Buridan and Modes; and Buridans Theory of Denitions in his Sci-
entic Practice, in The Metaphysics and Natural Philosophy of John Buridan, pp. 2947.
9
For a discussion of the development of medieval nominalism, see Calvin G.
Normore , The Tradition of Medieval Nominalism, in Studies in Medieval Philosophy,
ed. John F. Wippel , Studies in Philosophy and the History of Philosophy 17 (Wash-
ington, DC, 1987), pp. 201217. For an overview of some of the disputes between
fourteenth-century nominalists and realists, see Marilyn McCord Adams , Universals
in the Early Fourteenth Century, in Norman Kretzmann et al., eds. The Cambridge
History of Later Medieval Philosophy. From the Rediscovery of Aristotle to the Disintegration of
Scholasticism 11001600 (Cambridge, 1982), pp. 41139; and Adams, William Ockham ,
1 (Notre Dame, 1987). Of course the general natures in which these realists believed
were not Platonic universals, i.e., eternally existent, self-subsistent, immaterial entities
or forms, which are themselves the perfect and only true instances of certain gen-
eral properties such as goodness, wisdom, humanity, etc. Medieval thinkers in general
believed that Aristotle had disproved the existence of Platonic forms in his Metaphysics
(see, e.g., Met 1, 7, 13, and 14).
by which the soul conceives indifferently of several things, as it conceives
indifferently all men by means of the concept whence the name man
is imposed, and the same is the case with the concept of animal with
respect to animals. Therefore, since species and genera are universals
according to predication, it is evident that species and genera are such
concepts of the soul (S 4.3.2).
Whereas Ockham s efforts to establish the legitimacy of his nominalist
framework met with opposition by realists such as Walter Burley , Buridan
writes during a period when nominalism had gained currency
10
and
is freer to apply its tenets without arguing in favor of his approach.
11

Yet given his denial of real universals, Buridan needs to explain how
we form general concepts. An overview of his response may clarify his
nominalist perspective.
12
Buridan draws on psychology and metaphysics to explain why we
have universal concepts, providing both a mechanism whereby we
abstract these concepts along with an external ground for our ability
to relate these concepts back to the multiplicity of individuals that they
characterize.
Buridan s psychological thesis maintains that:
When the intellect receives from the phantasm the appearance or under-
standing of Socrates as fused together with size and location, making
the thing appear in the way something exists within the prospect of
the cognizer, the intellect understands him in a singular manner. If the
intellect can untangle that fusion and abstract the concept of substance
or of whiteness from the concept of location, so that the thing is no
longer perceived in the way something exists within the prospect of
10
Which is not to say that realism was no longer a viable alternative. See E. A.
Moody , Ockham , Buridan , and Nicholas of Autrecourt, in Inquiries into Medieval
Philosophy, ed. James F. Ross (Westport, Conn., 1971), pp. 275314; Jennifer Ottman
and Rega Wood , Walter Burley : His Life and Works, Vivarium 37 (1999), 123; and
Mieczyslaw Markowski , Johannes Buridans Polemik gegen die Universalienlehre des Walter
Burleigh, Mediaevalia Philosophica Polonorum 26 (1982), pp. 717.
11
Note though that Buridan does not reject plausible realist interpretations out
of hand. See, e.g., S 3.2.9 (discussed below). Moreover, the beginning of his career is
marked by several polemical treatises: De dependentiis, diversitatibus et convenientiis (1332),
Quaestiones de universale (also known as De differentia universalis ad individuum) (1332), Defen-
siones determinationis de diversitate generis ad speciem (1333), and De relationibus (1334) (see
Dirk-Jan Dekker , John Buridans Treatise De dependentiis, diversitatibus et convenientiis: An
Edition, Vivarium 42 (2004), 109149, esp. p. 109; and Markowski , Johannes Buridans,
pp. 78).
12
The following discussion is indebted to King , John Buridan s Solution to the
Problem of Universals, in The Metaphysics and Natural Philosophy of John Buridan, pp.
127.
the cognizer, then it will be a common concept. Accordingly, once the
concept of Socrates has been drawn out abstractly from the concepts of
whiteness and of location and of other accidents or extraneous features,
it will then no more represent Socrates than Plato : it will be a common
concept, one from which the name man is derived. Quaestiones in libros
Aristotelis De anima [tertia sive ultima lectura], ed. Zupko (Ithaca, N.Y., 1989),
3.8, ll. 391403, trans. King , Universals, pp. 1415.
Our experiences are with particular individuals characterized by unique
features. Via abstraction we peel away features specic to one individual
and thereby conceive of traits that this individual shares with others to
whom it bears certain resemblances. For the realist, this commonality
suggests the existence of shared general natures, yet Buridan grounds
it in specic and generic likenesses:
Now if it were the case that there are many items similar to one another,
then anything similar to one of them, with respect to the feature in which
they are similar, is similar to any one of them. Hence if all asses have in
reality an agreement and likeness with one another, when the intelligible
appearance represents some ass in the intellect by means of a likeness, it
must simultaneously represent any given ass indifferently. . . . An intention
becomes universal in this way (Quaestiones super octo Physicorum libros Aristotelis
(Paris, 1509), 1.7, fol. 8vb, trans. King , Universals, 2001, p. 18).
We thus attain generic and specic concepts through a process of
abstraction that relies on resemblances. Yet given Buridan s denial
of shared general natures, he needs to explain why these particulars
resemble one another so as to give rise to these concepts. As King
suggests, Buridan may look on resemblances as primitives or unex-
plained brute facts.
13
If this is so, such reliance would seem to highlight
Buridans conviction that the realist position is untenable owing to its
needless hypostatization of universals: It is pointless to hold that there
13
King , Universals, pp. 2527. King also suggests that Buridan s approach may
be modal, grounding similarities in how things exist rather than what they are (ibid.).
For an examination of Buridans modal approach, see Klima , Buridan and Modes,
p. 490, where Klima discusses Buridans consideration of the mode of existence of a
spherical piece of wax (in Quaestiones Physicorum, 2.3). Quoting Klima: It may seem
obvious that a quantity which is now a sphericity may remain in existence without
remaining a sphericity, on account of simply changing the distance of its outermost
points from a given point in space. However, this of course will not cause its perishing,
it will only change the way it is arranged in space, its modus. But this modus does not
have to be another thing over and above the quantity of a body thus and so arranged
in space. By parallel reasoning, Socrates may resemble Plato with respect to arrange-
ment and this arrangement need not itself be any real thing.
are universals distinct from singulars if everything can be preserved
without themand indeed it can.
14
Ultimately, we live in a world
of individuals, generic and specic concepts (i.e., concepts of second-
ary substances) arising from how we attend to impressions, which are
the medium through which we arrive at the concepts by which those
[universal] names are imposed:
15
Secondary substances signify the same substances as primary ones, but dif-
ferently, for primary substances signify one substance in a singular manner,
whereas secondary substances signify many commonly, and indifferently.
Since when the word this is added to a common term, it turns it into a
singular term, therefore we say that a primary substance more properly
signies this something than does a secondary substance, because it signi-
es a singular substance existing in a singular manner (S 3.2.6).
Section 2: Aristotles Categories
Aristotles Categories divides things that are into four classes: [1] Some
are said of a subject but are not in any subject. . . . [2] Some are in a
subject but are not said of any subject. . . . [3] Some are both said of a
subject and in a subject. . . . [4] Some are neither in a subject nor said
of a subject (1
a
20
b
9).
16
This classication suggests that things that are fall into four distinct
groups.
17
Yet, commentators have long debated whether Aristotle
intended to discuss actual things or simply the utterances with which we
signify them.
18
The difculty is compounded by the fact that Aristotles
14
In Metaphysicen (Paris, 1518), 7.16, fol. 51vb (trans. King , Universals, p. 1). For
an analysis of Buridan s critique of realism, see King, Jean Buridan, in Individuation
in Scholasticism: The Later Middle Ages and the Counter-Reformation; 11501650, ed. Jorge
J. E. Gracia , SUNY Series in Philosophy (Albany, N.Y., 1994), pp. 397430, esp. 4.
15
Quaestiones in Isaogogen Porphyrii 11:173, ll. 185354 (trans. Zupko , in Zupko, John
Buridan : Portrait of a Fourteenth-Century Arts Master (Notre Dame, 2003), p. 341, n. 35).
16
Brackets mine. The passage is translated from the Greek by J. L. Ackrill , in
Aristotles Categories and De interpretatione, trans. and ed. J. L. Ackrill, Clarendon
Aristotle Series (Oxford, 1963). All translations from the Greek are noted as such and
taken from Ackrill.
17
J. L. Ackrill , for instance, believes that this is the case: It is perfectly clear that
Aristotles fourfold classication is a classication of things and not names, Aristotles
Categories, p. 75.
18
Porphyry for example holds that the work concerns signicative utterances (see
below). For discussion of competing ontological and linguistic interpretations, see Mohan
Matthen , The Categories and Aristotles Ontology, Dialogue 17.2 (1978), 22841; and
language appears to allow for either interpretation, or a combination
of both,
19
and even for those who believe that the Categories concerns
signicative utterances, several paths unfold. Yet, before we explore
these, let us consider the motivation behind Aristotles work.
20
In Categories 2
b
46, Aristotle writes: All the other things are either
said of the primary substances as subjects or in them as subjects. So
if the primary substances did not exist it would be impossible for any
of the other things to exist.
21
This statement looks to be an attack
on Plato s middle-period theory of forms, which has it that universals
possess real existence apart from particulars, with respect to which the
forms are ontologically prior.
22
The Categories reverses this picture by
granting ontological priority to primary substances. Essences are inher-
ent, existing only in subjects, and it is these subjects that are responsible
for the essences existence, not the other way around. Aristotle eshes
this out with the notion of priority by nature (t phusei ): Of things
which reciprocate as to the implication of existence, that which is in
some way the cause of the others existence might reasonably be called
prior by nature (Cat 14b1113).
23
Thus, not only are forms inseparable
from substances, but it is the latter that are responsible for the existence
of the former.
J. M. E. Moravcsik , Aristotles Theory of Categories, in Aristotle: A Collection of Critical Essays,
ed. Moravcsik, Modern Studies in Philosophy (Garden City, 1967), pp. 12545.
19
In any case, the Categories is generally held to t with Aristotles logical works,
supplying a discussion of the relations that hold among terms, which are the building
blocks of both propositions and demonstrative syllogisms (see Marc S. Cohen , Aristotles
Metaphysics, The Stanford Encyclopedia of Philosophy (Winter 2003 Edition), Edward N.
Zalta (ed.), URL = http://plato.stanford.edu/ archives/win2003/ entries/aristotle-
metaphysics, esp. 2). Regarding the medieval ordering of the Categories in relation to
Aristotles other logical works, see Ebbesen , Ancient Scholastic Logic as the Source of
Medieval Scholastic Logic, in Norman Kretzmann et al., eds. The Cambridge History of
Later Medieval Philosophy. From the Rediscovery of Aristotle to the Disintegration of Scholasticism
11001600 (Cambridge, 1982), pp. 101127.
20
The following discussion draws primarily on Giorgio Pini , Categories and Logic in
Duns Scotus : An Interpretation of Aristotles Categories in the Late Thirteenth Century (Studien
und Texte zur Geistesgeschichte des Mittelalters) 77 (Leiden, 2002), pp. 118; and
Steven K. Strange , introduction to Porphyry On Aristotles Categories, trans. Strange
(London, 1992), pp. 112.
21
Translated from the Greek.
22
See, e.g., Republic 504e518c and 596e597a, Phaedo 100b102a3, and Phaedrus
247c3247e6. For the dating of these works in Plato s middle-period (c. 365c. 347), see
Richard Kraut, ed. The Cambridge Companion to Plato (Cambridge, Eng., 1992), p. xii.
23
Translated from the Greek. Cf. Met 1019
a
14.
Aristotle introduces the categories themselves as a list of things
said [legomenn] without any combination.
24
This language suggests
that the work is concerned with signicative utterances. Such a read-
ing survived into the Middle Ages in Boethius Categories commentary
(written c. 509511), which in turn was greatly inuence by the sur-
viving commentary of Porphyry , who believed that the Categories had
to do with the primary imposition of expressions. . . . For [the work]
concerns simple signicant words insofar as they signify things.
25
Still,
even allowing that the Categories discusses signicative utterances, it is
unclear what these utterances signify. Duns Scotus , for example, believes
that the expressions under consideration ultimately refer to extramental
essences.
26
Buridan , however, denies the existence of general essences
that are somehow common to a variety of individuals, and this denial
is reected in his claim that the Categories treats the relations that arise
among signicative utterances owing to the semantic properties of
terms, which rules out a realist interpretation of the work.
Buridan wrote two works on the Categories: De Praedicamentis, which
is the third treatise of his Summulae de Dialectica, and his Quaestiones in
Praedicamenta. This study follows Buridans treatment of Cat. 1a20b9
in Summulae 3.1.5, supplementing the discussion with material from a
parallel passage in De Praedicamentis, where Buridan offers substantially
the same account (QC 4).
The rst version of the Summulae de dialectica was probably not writ-
ten before 1325, when Buridan began his career, and the editor of
the critical edition of the De praedicamentis suggests a possible terminus
ante quem for the Summulae de dialectica as sometime between 1325 and
1360.
27
The work is a commentary on Peter of Spain s Tractatus or
24
The exact meaning of without any combination (kata mdemian sumplokn) is unclear.
Aristotle may have in mind terms that pick out just one member of a category, or he
could simply be distinguishing single words from groups of words. See Ackrill , Aristotles
Categories, p. 73.
25
In Aristotelis Categorias Commentarium 58, 46, ed. A. Busse (Berlin, 1887), (trans.
Strange , p. 34). See above, n. 20.
26
As Pini notes with reference to Scotus : To say that logic considers categories
as the most common genera is to say that logic considers categories not as essences
but as the most universal univocal concepts by which our intellect understands the
extramental essences, Categories and Logic in Duns Scotus, p. 143.
27
E. P. Bos , ed. Johannes Buridanus Summulae in Praedicamenta, (Aristarium) 103
(Nijmegen, 1994), p. xvi.
Summulae Logicales
28
(composed around 1250),
29
yet Buridan thoroughly
revises Peters text in light of new developments in logic during the
rst half of the fourteenth century, regrouping material, on occasion
providing his own source text (as may be the case with the gloss of Cat
1
a
20
b
9), and even adding entire treatises (such as that on supposition,
discussed below).
30
Section 3: Buridan s Summulae commentary
Buridan s aforementioned tendency simply to advance a nominalist
interpretation of canonical texts rather than push for the legitimacy
of his approach is evident early in the Summulae commentary, when he
comments on a gloss of Cat 1
a
16
b
9, a passage central to Aristotles
text. Here, Buridan suggests a dichotomy between his own reading
and an unlikely alternative, never considering a realist interpretation.
The passage in question presents Aristotles aforementioned claim
that things that are fall into four classes. Buridan (or his source text)
summarizes Peters gloss of the passage as follows: Of those [things]
that there are, some are said of a subject, but are in no subject, others
are in a subject, but are said of no subject, still others are said of a
subject and are in a subject, and yet others neither are in nor are said
of a subject. (S 3.1.5)
31
Buridan construes this passage as explicating the role of primary
substance terms as the ultimate locus of predication (i.e., as principally
subjected),
32
general terms being said of and accident terms being in
subject terms, while subject terms themselves are neither said of nor
in anything. Thus Aristotles things that there are are of four types:
Some [things] are said of a subject and are in no subject, namely, com-
mon terms from the category of substance. . . . Others are in a subject
and are said of a subject, as are common terms in the categories of
28
See Peter of Spain , Tractatus: called afterwards Summule logicales, rst critical edition
from the manuscripts with an intro. by L. M. De Rijk (Assen, 1972), 6.
29
See Bos , Summulae in Praedicamenta, p. xiii.
30
See Zupko , John Buridan , pp. 4448; Klima , Summulae, pp. xxxxxxii; and Bos ,
Summulae in Praedicamenta, pp. xivxv. On Buridans authoring the treatise on supposi-
tion, see Zupko, John Buridan, p. 59.
31
Cf. Tractatus 3.3, pp. 2829.
32
See S 3.2.1 and 3.2.9.
the accidents. . . . Others are in a subject, but are said of no subject,
namely, singular terms of accidental categories. . . . Still others neither
are in nor are said of a subject, namely, singular terms of the category
of substance. (S 3.1.5)
Underpinning this explication is Buridan s interpretation of Cat
1
a
16
b
9 as having to do with the properties of terms:
[ This passage] is easy to expound if it is interpreted as being about
signicative terms. But it cannot be correctly expounded as being about
other things, since a man is no more said of a subject than is Socrates ,
for a man is the same as Socrates, but the term man is said of a subject,
namely, of an inferior term, for example, of the term Socrates. Further, it
is reasonable that the whole [passage] be interpreted as concerning signi-
cative terms. . . . Therefore, by those that there are we understand nothing
but signicative terms that can be subjects or predicates (S 3.1.5).
Terms are not individuals, only the former are predicable. While one
cannot say a esh and blood man, one can say man of him. Buridan
thus concludes that Aristotle is discussing terms. Of course a realist
might contend that in the sentence Socrates is a man, man signi-
es a shared general nature, viz., humanity, and that there is nothing
odd about claiming that this immanent universal is said of subjects.
Yet Buridan omits mention of this possibility and closes abruptly: If
Aristotles examples appeared prima facie to require another exposition,
I would reply that examples need not be veried [non requiritur vericatio],
as Aristotle himself says in bk. 1 of the Prior Analytics.
33
Buridan distinguishes between substance and accident terms as only
the latter are in subject terms, and will develop the technical sense of
being in a subject term along the lines of Aristotles discussion of denom-
inative expressions: When things get their name from something, with a
difference of ending, they are called denominatives. Thus, for example,
the grammarian gets his name from grammar, the brave get theirs
33
Prior Analytics, 1 41, 49
b
3437. Compare Buridan s question commentary: It is
necessary that by being in a subject we understand denominative predication, and
by being said of a subject we understand essential predication. And if you do not
wish to agree, go your own way, because names are conventional (oportet, quod per esse
in subiecto intelligamus praedicationem denominativam, et per dici de subiecto intelligamus praedica-
tionem essentialem. Et si tu non vis hoc recipere, vadas vias tuas, quia nomina sunt ad placitum)
(QC 4, ll. 5558, p. 32, trans. mine). See 3.45 infra for denominative and essential
predication.
from bravery.
34
In addition, Buridans use of this passage to interpret
the groupings that lie at the heart of the categories employs signication,
supposition and appellation, which we may consider in this order.
3.1 Signication
Terms are utterances imposed to bring to mind or signify concepts.
35

What terms bring to mind are their immediate signicata (signicata
immediata) (S 4.3.2), and when terms signify extramental entities,
these entities are called their ultimate signicata (signicata ultimata)
(S 4.3.2).
3.2 Supposition
A terms supposition, on the other hand, is what it picks out or that for
which it stands. Buridan s nominalist ontology allows for the existence of
certain substances, viz., God, angels, souls, and composites. In addition
the accidents of bread and wine,
36
certain magnitudes and motions, and
prime matter are capable of existence through divine agency;
37
apart
from these, there are no independent, extramental entities for a term
to pick out. Buridan describes the difference between signication and
supposition as follows:
Any word that is a part of a proposition not taken materially
38
signies
and gives rise to some concept in the person hearing it according to the
34
Cat 1
a
1215. I have replaced Ackrill s paronymous with denominatives to bring
his translation of the Greek parnuma in line with Klima s translation of denominativa,
its Latin counterpart.
35
Of course concepts do not have to be of extramental entities. Buridan thus
distinguishes between categorematic and syncategorematic terms, denying that the
latter refer to extramental entities. The function of syncategorematic terms is not to
represent something by themselves but to modify the representative function of per se
representative concepts [i.e., the representative function of categorematic concepts]
(Klima , Summulae, xl, brackets mine).
36
See n. 1.
37
See King , Universals, p. 3. The addition of magnitudes and motions is taken
from Calvin Normore , Buridan s Ontology, in How Things Are: Studies in Predication
and the History of Philosophy and Science, eds. James Bogen and James E. McGuire, Philo-
sophical Studies Series in Philosophy 29 (Dordrecht, 1981), p. 197.
38
The notion of material supposition is related to (though not identical with) our
contemporary distinction between a terms use and mention; for, on some occasions,
when a term supposits materially, it is mentioned rather than used, i.e., it stands for
itself or a similar token-term, as the term man in Man has three letters (See
signication conventionally given to it. But not every such word has sup-
position, for only such a term is apt to supposit that, when something is
pointed out by the pronoun this, or when some things are pointed out
by the pronoun these, can truly be afrmed of that pronoun. Therefore
the term chimera cannot supposit, for whatever is pointed out, it is false
to say This is a chimera, and whichever things are pointed out, it is false
to say These are a chimera. (S 4.1.2)
3.3 Appellation
Sentential context delimits a terms signication via the terms supposi-
tion, that is, that to which a term refers limits what the term brings
to mind. For example, in the proposition Socrates is white the word
white leads one to think of the quality of being white as it pertains
to or is in Socrates rather than simply the quality of being white in
general. This delimited signication is a terms appellation, while the
general quality that the term signies independent of sentential context
is obliquely referred to via the terms connotation:
Some terms are appellative and others are not. For substantial terms in
the nominative case
39
or terms not connoting anything at all beyond the
things for which they supposit are not appellative terms properly speaking.
Klima , Summulae, xlix). However, Buridan s notion of material supposition conveys more
than just this. For he believes that a term also stands in material supposition when it
picks out its immediate signicate, which is the concept according to which it was
imposed to signify, as the term man in the proposition Man is a species (S 4.3.2).
Other thinkers, such as Ockham , refer to this latter type of supposition as simple, but
Buridan views this designation as a holdover from realist semantics, which requires a
special type of supposition to pick out immanent universals (ibid.). Material supposition
is contrasted with personal supposition, in which the subject or the predicate of the
proposition supposits for its ultimate signicates or for its ultimate signicate, as the term
man supposits for men in the proposition A man runs (ibid.). Personal supposition,
in turn, admits of further divisions which need not concern us here. Henceforth, my
comments pertaining to supposition refer exclusively to personal supposition.
39
The reason for this qualication is that a substantive oblique term appellates what
its nominative form would supposit for as pertaining to [adiacens] what the nominative
term governing it supposits for (S 4.5.4). For example, If I say: Socrates donkey
runs, the whole [phrase] Socrates donkey supposits for a donkey, and not for any-
thing other than a donkey, but the term Socrates signies something other than a
donkey; therefore the term Socrates appellates Socrates as pertaining to [adiacentem]
a donkey, i.e., as its owner, though not as something inhering in it (S 4.1.4). On the
other hand, why substantive terms in the nominative case do not appellate will become
clearer when we turn to Buridan s discussion of the phrase in a subject, which phrase
distinguishes between substantial and accidental terms. For now we may note that sub-
stantial terms pick out unique individuals and thus conceive only that for which they
supposit. What makes Socrates uniquely Socrates belongs to no other, hence a term
But every term connoting something other than what it supposits for is
called appellative and appellates that which it connotes as pertaining
to [adiacens] that which it supposits for, as when white appellates white-
ness as pertaining to that which the term white is apt to supposit for.
(S 4.5.1)
40
Appellation then may be described as a property of terms whereby they
evoke a general characteristic in relation to a particular individual, as
Socrates is healthy evokes health in relation to Socrates, or Socrates
health.
3.4 Essential predication
Appellation, supposition, and signication are linked with essential and
non-essential predication. A predication is essential when, with respect
to its subject and predicate terms: Neither of these two terms adds
some extrinsic connotation to the things they supposit for. Therefore,
although the term animal signies more [things] than the term man,
nevertheless [in the proposition A man is an animal], it does not
appellate over and above the signication of the term man anything
having to do with man, i.e., as something pertaining to man [ per modum
adiacentis homini ]. (S 2.5.2, second brackets mine)
In an essential predication, neither term appellates, for in Buridan s
nominalist framework essences are particular and there is nothing for
an essential term to bring to mind apart from the entity it uniquely
characterizes. Moreover, as essential predicates most appropriately
answer the question What is it?,
41
and thus signify characteristics
that uniquely determine an individual, essential predicates refer to
an individual in all circumstances in which that individual exists, and
while a subject can subsist apart from the signicata of its non-essential
predicates, the destruction of the ultimate signicatum of an essen-
tial predicate entails the destruction of its subject.
42
By contrast, in a
that refers only to Socratesan essential predicatecannot bring to mind anything
other than Socrates himself.
40
Cf. S 4.1.4: Every term that supposits for something and signies something else
for which it does not supposit appellates everything it signies other than that for which
it supposits as pertaining [adiacens] to that for which it supposits.
41
S 2.1.3. Strictly speaking, there are two species of essential predication. Predication
in quid most properly answers the question What is it?. Predication in quale, on the
other hand, answers the question What is it like? by supplying a specic difference.
42
Thus Klima points to the correspondence between non-essential and essential
predicates, on the one hand, and non-rigid and rigid designators, on the other: A
non-essential predication, one term of it adds some extrinsic con-
notation over the signication of the other (S 2.5.2), as when white
supposits for a person while appellating whiteness in relation to that
person and connoting whiteness in general. Of course these accidental
terms can pick out individuals, though as Peter King notes those for
which it does supposit may be picked out more directly by an absolute
(i.e., non-appellative term).
43
Yet, when a non-essential predicate ceases
to apply to an individual, it is not because an accident has passed out
of existence, rather the individual has ceased to be included in the
accidental terms ultimate signicata owing to an alteration in the way
or mode in which that individual exists,
44
and discussions of change
need not reify accidental predicates. Let us now consider how Buridan
applies these notions to account for Aristotles distinction between what
is said of and in things that are.
3.5 Denominative terms
The Categories opens with a distinction between three types of utter-
ances: equivocal, univocal, and denominative. Equivocal utterances
bring to mind diverse concepts, e.g., bank can signify either a nancial
institution or a steep incline, and thereby supposits differently in dif-
ferent contexts. Likewise, things themselves can be termed equivocals
because they are equivocally signied by some utterance (S 3.1.1).
Univocals, on the other hand, signify many things under one concept,
connotative term is what in the modern parlance we would call a non-rigid designator
of its supposita: it can cease to supposit for its supposita without their destruction, as
a result of removing its connotata. Take away a wealthy mans wealth, and the man,
while he will still go on existing, will cease to be supposited for by the term wealthy
man (Klima, Buridan s Theory of Denitions, p. 38). See also Klima, Buridan
and Modes.
43
Between Logic and Psychology: Jean Buridan on Mental Language (paper,
presented at John Buridan and Beyond, Copenhagen, September, 2001), 89; accessed
at http://individual.utoronto.ca/pking/presentations/Buridan_on_Mental.pdf.
44
See n. 13. Compare Ockham s characterization of the realist position as mis-
takenly multiplying entities in correspondence with the diverse ways in which we use
language: A column is to the right by to-the-rightness, God is creating by creation, is
good by goodness, just by justice, mighty by might, an accident inheres by inherence,
a subject is subjected by subjection, the apt is apt by aptitude, a chimera is nothing by
nothingness, someone blind is blind by blindness, a body is mobile by mobility, and so
on for other, innumerable cases, Summa Logicae (St. Bonaventure, N.Y., 1974), p. 169,
trans. Klima , Buridan and Modes, p. 475. As to whether the realists are genuinely
committed to this overpopulated ontology, see Klima, Buridan and Modes, and
Summulae, intro.
as when Plato and Socrates are subordinated to the concept man.
As with equivocals, entities can be called univocals when univocally
signied by some utterance.
Denominative terms are crucial to Buridan s distinction between
substance and accident terms. Denominatives are those [terms] that
differ from some [other] only in casus
45
in respect of [the formation of
the] name, [and] have appellation [from the other], as literate from
literacy and strong from strength, are called denominatives.
(S 3.1.3)
Terms must meet two conditions in order to be denominative. First,
a denominative term differs only morphologically from its abstract
counterpart, as white differs from whiteness. Second, a denominative
term must appellate, or connote, some external disposition besides
what it supposits for (ibid.), meaning denominatives conceive a general
characteristic in relation to the subject that they pick out, as Socrates
is happy conceives the general characteristic of being happy in rela-
tion to Socrates, or Socrates happiness (while obliquely referring to or
connoting happiness in general).
Speaking of the grammatical connection between concrete terms
and their abstract counterparts, Buridan claims that the concrete form
derives with respect to its formation from the abstract form (ibid.),
which seems to disagree with our understanding of the derivation of
abstract terms: The grammarian reasonably derives whiteness from
white, namely, the abstract [form] from the concrete. For we have the
concept of white or of just naturally prior to the concept of white-
ness or of justice. . . . Therefore it is understandable that the names
white and just were the ones earlier imposed to signify. (ibid.)
Buridan does not want to challenge the grammarian. Rather he
is exploiting the sense of derivation (casus) where concrete accident
terms derive their predicability from the signication of their abstract
counterparts, viz., the general non-essential characteristics that concrete
45
Casus may be translated as case or ending. But, as Klima notes, Buridan draws
on a broader sense of the term, where its etymological connection with the verb cadere
(to fall) brings to mind derivation. Both senses are at play in this passage. The sense of
casus as ending means that denominatives must be morphologically related to their
abstract counterparts (as white and whiteness), while casus as derivation provides
Buridan the opportunity to note that concrete accident terms derive their appellation
from their abstract counterparts (S 3.1.3), as, in sentential context, the appellation of
white is whiteness in relation to the subject for which white supposits. Klima thus
leaves casus untranslated (see Klima, Summulae, p. 145, n. 3).
accident terms relate to particular subjects: This [talk of derivation]
does not contradict the grammarian, for we are not talking about
casus in the sense of [a words] derivation from [the rst] imposition;
rather, by casus we mean its application to the subject in predicating it
thereof. For an abstract term is not posited in order to be predicated
of a subject, but a concrete term is; thus a man is said to be white,
but not whiteness. (ibid., rst brackets mine)
Denominative terms supposit for their subjects only via non-essential
predicates such as skin color or position in space. These non rigidly
designating characteristics are the denominatives appellata, which may
be understood as general characteristics signied in relation to particu-
lar individuals, as wealthy in Socrates is wealthy brings mind not
just wealth in general (its connotation) but particularly the wealth of
Socrates. In this sense, denominatives derive their supposition, and thus
their predicability, from their abstract counterparts, which signify the
characteristics by which denominatives are related to their subjects.
In some cases, a concrete term is not morphologically related to its
abstract counterpart; when this happens the concrete term is not the
denominative form of the abstract term: For example, even if someone
is virtuous [studiosus] because of his virtue [virtutem], the name virtuous
[studiosus] is not the denominative of the name virtue [virtus] (ibid.).
46

Moreover, not all concrete terms that are morphologically related to an
abstract counterpart are denominative. Buridan considers homo (man)
in relation to its abstract counterpart humanitas (humanity). Despite
their obvious morphological similarity, man is not denominative, for
it is an essential predicate and hence does not appellate.
3.6 Terms said of and in
Immediately following his discussion of denominatives, Buridan turns
to terms said of and in a subject.
47
Here again is the gloss: Of those
46
The use/mention quotes and rst brackets are mine. Klima brackets the English
and has the Latin in the text.
47
Note that here and throughout Buridan is speaking not of ontological but of
grammatical subjects: In connection with the whole chapter, we should take good
notice of the fact that substance is taken in two ways. In one way it supposits for real
substances that subsist on their own, and for their parts. . . . In the second way, substance
is taken to supposit for terms in the category of substance, i.e., for terms that signify
substances in the rst sense without extraneous connotation. . . . Correspondingly, the
term accident is taken as it supposits for terms in the Categories of accidents. . . . It is
[things] that there are, some are said of a subject, but are in no subject,
others are in a subject, but are said of no subject, still others are said
of a subject and are in a subject, and yet others neither are in nor are
said of a subject. (S 3.1.5)
He interprets this text in light of the notions of denominative and
essential predication: By being said of a subject we understand being
predicated quidditatively of an inferior term, and by being in a subject
we understand being predicated denominatively, and not quidditatively
of a subject. (ibid.)
These distinctions are then applied to the four classes of terms in
order to explain the difference between substance and accident terms
on the one hand, and common and particular terms on the other.
3.6.a Terms in a subject
Substance terms do not appellate, which is to say that in the context of
a proposition, they signify only that for which they supposit. Accord-
ingly substance terms cannot be in a grammatical subject. Accident
terms do appellate, conceiving general non-essential characteristics
in relation to particular individuals, as in Socrates is white, white
brings to mind the quality of being white in relation to Socrates, or
Socrates whiteness. Hence accidental terms can be denominative, and
thus in a subject (provided they likewise meet the rst requirement of
morphological similarity to an abstract counterpart).
Where does this leave abstract accident terms? Failing to meet the
requirement of derivation, they are non-denominative, yet as accident
terms they should be in a subject. Buridan recognizes this difculty and
responds that such predications are denominative from the point of
view of intention but . . . not denominative properly speaking (S 3.1.3).
Strictly speaking, for such predications to be denominative we must
understand these abstract terms as being predicated via their concrete
counterparts. Buridan acknowledges that such predications are improper
(impropriae) but allows that when an abstract accident term is predicated
of some subject, we can understand its concrete form in its stead: Every
accidental term is in a subject, for it is predicated denominatively of
in this second way that substance is taken here. And [what is so called] is not called
substance by reason of its underlying [sub-standing, substare] in the sense of sub-
sisting on its own. . . . Rather, it is called substance because of its underlying in the
sense of being subjected, for substantial terms in denominative propositions are subjected
[i.e., serve as subject terms with respect] to accidental terms (S 3.2.1).
a substantial term, at least, either itself, if it is a concrete term, or its
concrete form, if it is an abstract term, such as whiteness (S 3.1.5).
3.6.b Terms said of
To be said of a subject is to be predicated quidditatively of an infe-
rior term (ibid.). Quidditative predication is the species of essential
predication designed to answer the question What is it (quid est)?
48

Accordingly in quidditative predications neither term adds some
extraneous connotation to the signication of the other (S 2.1.3). In
the case of common terms in the category of substance, it is easy to
see how this condition applies. In sentential context substantial terms
are non-connotative and thus function as rigid designators, conceiving
nothing apart from their subjects. Yet generic and specic terms in
the accidental categories conceive general non-essential characteristics.
This seems to render them unt to answer the question What is it?
with respect to their subordinate concrete accident terms. The latter
supposit for individuals yet in the context of a proposition an abstract
accident term will retain oblique reference to the general quality that
it signies in relation to this individual, that is the abstract term will
always appellate some disposition in relation to its subject and thus
should be unable to function as an essential predicate. For example, in
the proposition A white thing is a thing colored by whiteness, white
thing supposits for an individual, while thing colored by whiteness
appellates something over and above this individual, viz., whiteness in
general. Buridan circumvents this difculty by allowing for a broader
sense of quidditative predication: In this connection we have to note
that in the case of connotative terms, the genus is not predicated of its
species in quid in the strictest sense, but broadly speaking it is (S 8.2.5).
While it is true that thing colored by whiteness appellates whiteness,
nonetheless the predicate indicates what the subject is (quid est) with
respect to the subjects suppositum and connotatum, for predicate
and subject both supposit for the same entity (some white thing) and
appellate the same quality (whiteness). Thus the predication is broadly
construed as quidditative.
48
S 2.1.3. On the various species of essential predication, see n. 42.
3.7 Summary
To review, Buridan presents the Categories as a work about signicative
utterances and thus explains the divisions between substances and acci-
dents on the one hand and general and concrete terms on the other
with reference to appellation, supposition, and signication. General
and concrete terms are distinguished insofar as the former are quid-
ditatively predicable, and hence said of the latter, either owing to the
general terms being essential predicates or, in the case of general terms
from the accidental categories, because the accident term indicates
what the subject is (quid est) as concerns both the subjects suppositum
and connotatum. Substance and accident terms, on the other hand,
are distinguished because accident terms signify non-essential and thus
denominative predicates. This creates a difculty, as abstract accident
terms, which are in their subjects are derived for the purpose of sig-
nifying general non-essential characteristics rather than an individual
in relation to such characteristics. Buridan addresses this by allowing
that these terms can be denominative in an improper sense as their
concrete counterparts may be taken in their stead. The end product is
a coherent reading of Cat 1
a
20
b
9 as a text concerned with semantic
properties:
Some [things] are said of a subject and are in no subject, namely, common
terms from the category of substance, such as man or animal; others
are in a subject and are said of a subject, as are common terms in the
categories of the accidents, e.g., white or colored, or even whiteness
and color, which do not denominate substances by themselves but by
means of their concrete forms; others are in a subject but are said of no
subject, namely, singular terms of accidental categories, e.g., this white,
this runner; and still others neither are in nor are said of a subject,
namely, singular terms of the category of substance, as with the terms
Socrates , Plato , this man, this horse (S 3.1.5).
Section 4: The Sixth Property of Substances
In closing, let us consider Buridan s reading of Cat 4
a
1021. His com-
mentary on this passage not only showcases his nominalist semantics,
but likewise draws attention to the fact that he is not so close-minded
as completely to dismiss an apparently straightforward realist interpreta-
tion. This section of the Categories discusses the sixth and most distinctive
property of substances: The sixth property is what Aristotle states in
the following terms: it seems to be most proper to substance that while
it remains numerically one and the same, it is susceptible of contraries
by its own change. (S 3.2.9)
Buridan is quick to admit that this passage seems to refer to real
substances and not the terms that signify them, for it seems odd to
say that a term receives contraries by its own change: It is difcult
to expound this property correctly. . . . It seems to be obvious that this
sixth property is assigned to substance insofar as the name substance
supposits for true, per se subsisting substances. For the term man is not
changed . . . when a man now becomes white and then black, or warm,
or lukewarm, or cold. (ibid.)
But before explaining how this phrase can be taken as applicable
to terms, Buridan rst argues against this seemingly obvious realist
interpretation. He does this by noting that the term most (maxime) in
the phrase most proper to substances (maxime substantiae proprium) can
be construed in only two ways, and in either case the ascription of the
ability to receive contraries turns out to be false. First, most can be
understood as an adjective in the genitive case modifying substance.
In this case, Aristotle is to be read as asserting that Of what is most
a substance, it seems to be proper, etc.. What is most, or maximally,
substance is God or the heavens, and neither is susceptible of contrar-
ies. Accordingly we should not suppose that Aristotle is discussing per se
subsistent substances. On the other hand, most can be taken adverbially
with proper, in which case Aristotle is saying that it is most proper for
substance to receive contraries, etc. Yet, again the proposition would
be false because there are substances to which this property does not
apply, viz., God and the heavens. Thus, it appears that Aristotle is not
speaking of per se subsistent substances.
Nevertheless, Buridan allows that a realist could account for these
difculties by restricting the supposition of substance to generable
and corruptible substances.
49
Then, without criticizing this account,
Buridan offers his own interpretation, suggesting that most be taken
adverbially, while modifying the sense of proper ( proprium). Rather
49
Buridan undoubtedly has in mind Aristotles own examples of one and the same
generable and corruptible substance receiving contrary qualications at different times
while remaining numerically one and the same, e.g: An individual manone and the
samebecomes pale at one time and dark at another, and hot and cold, and bad and
good (Cat 4
a
1922, translated from the Greek).
than understand proper in its technical sense where what is proper,
viz., a property ( proprium), applies to all and only [those things that
are contained under a species] (S 2.5.1), Buridan suggests that here
proper ( proprium) means applicable (conveniens), and that what is most
proper is most applicable. Then, understanding substance to stand
for a primary substance term, Buridan points out that: [What is most
applicable to a primary substance term] is that it is susceptible of con-
trary predications by denominative predication, as when, for example,
of the term Socrates rst [the term] white is predicated, and then
black, or [rst] warm and [then] cold, because of its own change,
i.e., on the basis of the subjective change of that for which it supposits.
(S 3.2.9, rst brackets mine).
50
Against the objection that accidental as well as substantial terms are
thus susceptible of contrary predications, as, e.g., when a white thing
is now warm and later cold, Buridan notes that an accidental term
is not principally subjected (ibid.). A term is subjected when it stands
in a propositions subject position. Earlier, Buridan explains that: [A
primary substance term is] what properly, primarily, and maximally
underlies, i.e., is subjected, both in quidditative propositions, for then it
is subjected to secondary substances, and in denominative propositions,
for then it is subjected to accidental terms. (S 3.2.1, brackets mine)
Thus propositions properly concern substances. Buridan illustrates this
in a paraphrase of Met 5.7, 1017
a
710 (quoting Buridan): If someone
educated [musicum] is said to be white, then this is because both, namely,
educated and white are said of another, namely, substantial, subject,
as of Socrates or Coriscus (S 3.2.9). That is, as accidents do not
exist independently of substances,
51
terms signicative of accidents
cannot have extramental signication unless they are predicated of
50
Cf. QC 6, ll. 100106, p. 46: And because we explain this characteristic with
respect to terms, and terms do not change, it is necessary to provide the sense of by
a change with respect to itself , i.e., with respect to its formal signication, in this
way, viz., that it agrees appropriately with substantial terms to receive predications
of contrary terms. And this is said to be proper as concerns these terms not because
it pertains to all of them, but because it does not pertain to others, viz., accidental
terms (et quia de terminis exponimus hanc proprietatem, et non oportet terminum mutari, oportet
exprimere secundum sui mutationem, id est secundum mutationem eius, quod signicat secundum suam
formalem signicationem, ita quod terminis substantialibus appropriate convenit recipere praedicationes
terminorum contrariorum. Et dicitur hoc proprium illis terminis, non quia conveniat omnibus eis, sed
quia non convenit aliis, scilicet terminis accidentalibus (trans. mine).
51
Again, leaving aside miraculous cases.
substance terms, which terms ultimately supposit for the per se subsistent
substances on which accidents depend. Thus while we may say that
a white [thing] is now warm and later cold, it is not a quality which
is undergoing change, but rather a substance of which that quality is
a mode.
52
Accordingly, it is substantial rather than accidental terms
which admit of contraries.
52
See n. 13.
A REALIST INTERPRETATION OF THE CATEGORIES
IN THE FOURTEENTH CENTURY:
THE LITTERALIS SENTENTIA SUPER PRAEDICAMENTA
ARISTOTELIS OF ROBERT ALYNGTON
Alessandro D. Conti
Introduction
In the Categories Aristotle draws the main features of his metaphysics and
semantics, since he (1) sets out the basic items of the world (individual
and universal substances, individual and universal accidents) in their
mutual relationships, and (2) shows their links to language. So the cat-
egorial table is a division both of things and the signs signifying them.
Unfortunately, Aristotle does not clarify whether such a division is rst
of all a partition of things made on the basis of ontological criteria
and only secondarily a classication of (mental, written, and spoken)
terms, or vice versa.
1
Consequently, from late antiquity onwards there
were many disputes about whether the treatise was primarily concerned
with existent things or with their signs. According to the answer, it is
customary to classify medieval philosophers as being Realists (things)
or Nominalists (signs) concerning categories. This categorization is
questionable. On the one hand, some authors, following Boethius ,
2

support a nominalist solution of the problem of the intentio of the book
(which would deal with non-compounded utterances in their capac-
ity for being signicantvoces res signicantes in eo quod signicantes sunt
according to Boethiuss formula), while they (1) offer a realist reading
1
See J. L. Ackrill , Aristotles Categories and De interpretatione, translated with notes by
J. L. Ackrill, Oxford 1963, pp. 8081.
2
Cf. Boethius , In Categorias Aristotelis libri quattuor, PL 64, 159C and 160A. On
Boethiuss interpretation of Aristotles Categories see James Shiel , Boethius Com-
mentaries on Aristotle, Mediaeval and Renaissance Studies, 4 (1958), 217244; Henry
Chadwick , Boethius. The Consolations of Music, Logic, Theology and Philosophy (Oxford,
1981); Sten Ebbesen , Boethius as an Aristotelian Commentator, in Aristoteles: Werk
und Wirkung, ed. Jrgen Wiesner, 2 vols. (Berlin, 1987), pp. 286311; Alessandro
D. Conti , Boezio commentatore e interprete delle Categorie di Aristotele, in Scritti in
onore di Girolamo Arnaldi offerti dalla Scuola Nazionale di Studi Medievali, eds. A. Degrandi
(Rome, 2001), pp. 77102.
318 alessandro d. conti
of all the crucial points of the treatise and (2) judge the division into
ten categories of the fourth chapter to be a division of things, and
only derivatively of their signs.
3
In fact, we have to wait for Ockham,
Buridan , alii to get a consistently nominalist exposition of the Categories.
In detaching himself from the real Aristotelian intention, the Venera-
bilis Inceptor was (1) to consider the categorial table to concern terms
alone and not things, and (2) to translate Aristotles statements on the
ontological and physical status of substances, quantities and so on,
into rules for the correct use of terms, so that the level of language in
the Categories was raised a stepthe necessary presupposition of any
consistently nominalist interpretation of the tract. On the other hand,
many realist thinkers, such as Henry of Ghent , Simon of Faversham,
the rst Burley (before 1324), and Franciscus de Prato, held a reduc-
tionist position about the question of the number of real categories,
as they judged only the items falling into the three absolute categories
(substance, quantity, and quality) to be things (res), and considered the
remaining ones real aspects (respectus reales) proper to the former.
4
Few
medieval authors developed a fully consistent realist interpretation of
the Categories, by defending the thesis of the real distinction of all the
ten categories, which would be real things irreducible one to another.
5

3
Cf. e.g., Robert Kilwardby, Notulae super librum Praedicamentorum, prooem., and lectio
5, Cambridge, Peterhouse, ms. 206, fols. 41ra, and 44vb45ra; Thomas Sutton , Expo-
sitio super librum Praedicamentorum, prooem., and cap. de numero praedicamentorum, Oxford,
Merton College, ms. 289, fols. 3rbva and 7ra (transcription in Alessandro D. Conti ,
Thomas Sutton s Commentary on the Categories according to the Ms Oxford, Merton
College 289, in The Rise of British Logic, ed. P. O. Lewry (Toronto, 1985), pp. 173213,
pp. 189191 and 196).
4
Cf. Henry of Ghent , Summa quaestionum ordinariarum, a. 32, q. 5, in Opera omnia,
vol. 27, pp. 7980; Quodlibet V, q. 6, ed. (Parisiis, 1518), 2 vols., vol. 1, fols. 238r240v;
Simon of Faversham, Quaestiones super librum Praedicamentorum, q. 12, in Opera omnia, vol. 1,
ed. Pasquale Mazzarella (Padua, 1957), pp. 8285; Walter Burley , Tractatus super librum
Praedicamentorum, cap. de numero et sufcientia praedicamentorum (Cambridge, Peterhouse),
ms. 184, fol. 175rb176rb; Franciscus de Prato, Logica, pars I, tr. 5, a. 1, in Fabrizio
Amerini , La logica di Francesco da Prato. Con ledizione critica della Loyca e del Tractatus de
voce univoca (Florence, 2005), p. 381.
5
Cf. e.g., John Duns Scotus , Quaestiones super Praedicamenta, q. 1, in Opera philosophica,
vol. 1, pp. 249256, especially pp. 250251; Walter Burley , Expositio super Praedicamenta
Aristotelis, cap. de numero et sufcientia praedicamentorum, in Expositio super Artem Veterem Porphyrii
et Aristotelis, ed. (Venetiis, 1509), fol. 21rab; Jacobus de Placentia , Scriptum super librum
Praedicamentorum (Venice), Biblioteca Marciana, ms. lat. VI.97 (2594), fol. 30rbva. On
Duns Scotuss conception of the categories see Peter King , Scotus on Metaphysics,
in The Cambridge Companion to Duns Scotus, ed. Thomas Williams (Cambridge, 2003), pp.
2838; Giorgio Pini , Scotuss Realist Conception of the Categories: His Legacy to
Late Medieval Debates, Vivarium 43 (2005), 63100. On Burleys categorial doctrine
alyngtons interpretation of the CATEGORIES 319
Among them, Robert Alyngton
6
( 1398), one of the most important
authors of the generation after Wyclif , must be mentioned: his com-
mentary on the Categories, which relies on Burleys last commentary on
the Categories (A.D. 1337) and Wyclif s De ente praedicamentali (A.D. 1369),
7

is the most mature output of this stronger interpretative tradition.
Alyngton was able to work out (1) a coherently realist ontology of the
categories, (2) a new semantic theory of second intentions,
8
and (3)
the general strategy adopted by the Oxford Realists after Wyclif,
9
as
see Alessandro D. Conti , Ontology in Walter Burley s Last Commentary on the Ars
Vetus, Franciscan Studies 50 (1990), 121176, pp. 145174; and Marta Vittorini , Predicabili
e categorie nellultimo commento di Walter Burley all Isagoge di Porrio, Ph.D. diss., University
of Salerno, academic year 200405, pp. 335397. On Jacobus de Placentias commen-
tary on the Categories see Alessandro D. Conti, Il commento di Giacomo da Piacenza
allIsagoge e alle Categorie, in L insegnamento della logica a Bologna nel XIV secolo, eds. Dino
Buzzetti Maurizio Ferriani Andrea Tabarroni (Bologna, 1992), pp. 441460.
6
Not a great deal is known of Robert Alyngtons life. Most of the information
about him comes from A. B. Emden , A Biographical Register of the University of Oxford to
A.D. 1500, 3 vols. (Oxford, 195759), vol. 1, pp. 3031. From 1379 until 1386, he was
fellow of Queens College (the same Oxford college where Wyclif started his theological
studies in 1363 and Johannes Sharpe taught in the 1390s); he became Magister Artium
and, by 1393, doctor of theology. He was chancellor of the University in 1393 and
1395. In 1382 he preached Wyclif s religious and political ideas in Hampshire (A. K.
McHardy , The Dissemination of Wyclif s Ideas, in From Ockham to Wyclif, eds. Anne
Hudson Michael J. Wilks (Oxford, 1987), 361368, pp. 361362). He was rector of
Long Whatton, Leicestershire, where he died by September 1398. Alyngton was of
considerable repute as a logician (see E. Jennifer Ashworth Paul V. Spade , Logic in
Late Medieval Oxford, in The History of the University of Oxford, eds. Jeremy I. Catto
Ralph Evans (Oxford, 1992), vol. 2, pp. 5062, passim). Among his extant works: the
Litteralis sententia super Praedicamenta Aristotelis (henceforward In Cat.), partially edited in
Alessandro D. Conti , Linguaggio e realt nel commento alle Categorie di Robert Alyng-
ton, Documenti e studi sulla tradizione losoca medievale 4 (1993), 179306, pp. 242306
(all references are to the pages of this edition or, for the unedited portions, to the
ms. London, Lambeth Palace 393); a treatise on the supposition of terms (Tractatus de
suppositionibus terminorum); a commentary on the Liber sex principiorum; a treatise on the
genera of being (Tractatus generum).
7
On Wyclif s categorial doctrine see Alessandro D. Conti , Logica intensionale e
metasica dell essenza in John Wy cli f, Bollettino dellIstituto Storico Italiano per il Medioevo
e Archivio muratoriano 99.1 (1993), 159219, pp. 197209; Alessandro D. Conti, Wyclif s
Logic and Metaphysics, in A Companion to John Wyclif, ed. Ian C. Levy (Leiden, 2006),
67125, pp. 103113.
8
See Alessandro D. Conti , Second Intentions in the Late Middle Ages, in Medieval
Analyses in Language and Cognition, eds. Sten Ebbesen Russell L. Friedman (Copenhagen,
1999), pp. 453470.
9
For an analysis of Oxford Realistss main doctrines and information on their
lives and works see Alessandro D. Conti , Studio storico-critico, in Johannes Sharpe,
Quaestio super universalia, ed. Alessandro D. Conti (Florence, 1990), pp. 211238, and
295336; Alain de Libera , La querelle des universaux. De Platon la n du Moyen Age (Paris,
1996), pp. 403428; and Alessandro D. Conti, Categories and Universals in the Later
Middle Ages, in this volume.
he methodically substituted reference to external objective realities for
reference to linguistic and/or mental activities.
In what follows I shall offer a short analysis of Alyngtons doctrine of
categories. This will enable us to appreciate the novelty of his thought
and to gauge his importance within late medieval realism. He was one
of the rst authors whose commentary on the Categories shows that
partial dissolution of the traditional doctrine which is the distinctive
feature of the theories evolved during the very last period of the Middle
Ages. In the rst section, I shall sketch Alyngtons position on being
and categories as well as his theory of analogy. The second section
will deal with Alyngtons conception of universals and predication, the
most innovative section of his philosophical system. In the third one, I
shall discuss his ideas about substance. The fourth section will expound
his central theses on the nature, reality, and mutual distinctions of the
three main kinds of accidents: quantity, quality, and relation. In a nal
section, I shall draw some conclusions about the general signicance
of Alyngtons commentary on the Categories trying to place it within
the historical and doctrinal context of the late medieval commentaries
on that text.
Being and Categories
The point of departure of every realist interpretation of the Categories
is the notion of being (ens) in its relationship to the ten categories, as
Realists considered the categorial table to be a division of beings. Thus,
it is not surprising that, like Burley , Alyngton afrms that (1) the division
into ten categories is rst of all a division of things existing outside
the mind, and only secondarily of the mental concepts and spoken
or written terms which signify them, and (2) things belonging to one
categorial eld are really distinct from those belonging to anotherfor
instance, substances are really distinct from quantities, qualities, and
relations; quantities are really distinct from substances, qualities, and
relations, and so on.
10
Unfortunately Alyngton does not dene being;
yet, what he says about (1) the subiect of the book (the real categorial
10
Cf. Alyngton, In Cat., cap. de numero et sufcientia praedicamentorum, pp. 251, 252253,
255, and especially p. 258.
being which can be signied by atomic expressionsens in praedicamento
reali signicabile per signum incomplexum),
11
and (2) analogy
12
seems to entail
that, like Wyclif ,
13
he also thinks of being (ens) as a sort of an extra-
mental reality proper to everything (God and creatures; substances and
accidents; universal and indi vidual items; things, collections of things,
and states of affairs) according to different modes and degrees.
14
In the thirteenth and fourteenth centuries, all the realist authors, with
the only remarkable exception of Duns Scotus, (1) regarded categorial
items as composed of two main aspects: the inner nature or essence,
and their peculiar mode of being or of being predicated (modi essendi
vel praedicandi ); and (2) maintained that the categorial table divides
those categorial items according to their modes of being (or of being
predicated) and not according to their inner natures (or essences). In
more detail, Thomas Aquinas
15
and Thomas Sutton
16
(1) based their
method of nding the ten Aristotelian categories on the differences
of modes of being predicated, and (2) recognized three fundamental
modes of being predicated: (i) essentially, proper to substances, when
the predication indicates what a given res is; (ii) accidentally, proper to
quantities, qualities, and ad aliquid, when the predication indicates that
something inheres in a subject; and (iii) externally, proper to the remain-
ing six categories, when the predication indicates that something which
does not inhere in the subject nevertheless affects it. On the contrary,
11
Cf. Alyngton, In Cat., cap. de numero et sufcientia praedicamentorum, p. 252.
12
Cf. Alyngton, In Cat., cap. de aequivocis, fol. 70rvsee note 31, below.
13
Cf. Wyclif , De ente in communi, cap. 3, in Summa de ente libri pimi tractatus primus et
secundus, ed. Samuel H. Thomson, Oxford 1930, p. 36; De ente praedicamentali, cap. 1,
ed. Rudolf Beer (London, 1891), pp. 2 and 5.
14
Cf. Alyngton, In Cat., cap. de complexo et incomplexo, p. 249; cap. de numero et sufcientia
praedicamentorum, p. 255.
15
Cf. Thomas Aquinas , Sententia super Metaphysicam, liber V, lectio 9, eds. M.R. Cathala
R. M. Spiazzi (Turin, 1950), nn. 889891. On Aquinass derivation of the Categories,
see John F. Wippel , Thomas Aquinass Derivation of the Aristotelian Categories
(Predicaments), Journal of the History of Philosophy 25 (1987), 1334.
16
Cf. Sutton , Expositio super librum Praedicamentorum, cap. de numero praedicamentorum,
fol. 7rab (partial transcription in Conti , Thomas Sutton s Commentary on the
Categories cit., p. 196).
Albert the Great
17
and Simon of Faversham
18
(1) based their method
of nding the categories on the differences of modes of being, and (2)
admitted two fundamental modes of being: (i) being by itself, proper to
substance; and (ii) being in something else, proper to the nine genera
of accidents(2.1) the latter subdivided into (i) being in something else
absolutely, proper to quantities and qualities, and (ii) being in something
else in virtue of a relation to a third res (esse ad aliud), proper to the
remaining seven categories. In his last commentary on the Categories,
Walter Burley
19
recalled two other ways of deducing the ten Aristotelian
categories (aliqui acceperunt sufcientiam praedicamentorum sic . . . aliter accipi-
unt alii sic), both based on the various levels of similarity among their
own modes of being. In turn, Alyngton follows very closely Burleys
rst way of deriving the ten categories. In his opinion, there are two
fundamental modes of being proper to things: being by itself, which
characterizes substances, and being in something else (alteri inheaerens),
which characterizes accidents. The latter is subdivided into three less
general modes: being in something else in virtue of its matter; being
in something else in virtue of its form; and being in something else in
virtue of the whole composite. Something can be in something else in
virtue of its matter, form, and composite according to three different
ways: from inside (ab intrinseco), from outside (ab extrinseco), and partially
from inside and partially from outside ( partim ab intrinseco et partim ab
extrinseco). If something is in something else in virtue of its matter and
from inside, then it is a quantity; if from outside, it is a where (ubi );
if partially from inside and partially from outside, it is an affection
( passio). If something is in something else in virtue of its form and
17
Cf. Albert the Great, Liber de praedicamentis, tr. 1, cap. 7, in Opera omnia, ed.
A. Borgnet (Paris, 1890), vol. 1, pp. 164165. Todays scholars unanimously attribute
to Albert the idea that Categories are divided according to their modes of being predi-
cated, since he himself afrms that his own method of deducing the ten Aristotelian
Categories is based on their modes of being predicated (nos, quantum possumus, studebimus
ex propriis horum generum modis praedicandi ostendere hujus numeri rationem); but, in point of
fact, he does not utilize modes of being predicated in drawing the ten categories from
being. On the contrary, he constantly employs modes of being, as he speaks of ens
per se, ens in alio, ad aliud se habere, absolute inesse secundum materiam and secundum potentiam
formae and so on.
18
Cf. Simon of Faversham, Quaestiones super librum Praedicamentorum, q. 12, p. 84.
19
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de numero et sufcientia praedicamen-
torum, fol. 21rbva. On Burleys solution of the problem of the sufcientia praedicamentorum
see Mischa von Perger , Understanding the Categories by Division: Walter Burley vs.
William of Ockham, in La tradition mdivale des Catgories (XII
e
XV
e
sicles), eds. Joel
Biard Irne Rosier-Catach (Louvain-la-Neuve, 2003), pp. 3752.
from inside, then it is a quality; if from outside, it is when (quando vel
quandalitas); if partially from inside and partially from outside, it is an
action (actio). If something is in something else in virtue of the whole
composite and from inside, then it is a relation; if from outside, it is a
possession (habitus); if partially from inside and partially from outside,
it is a position ( positio).
20
Alyngtons choice implies an anti-reductionist approach to the matter,
which is conrmed by the solution of the problem of what properly
falls into the categorial elds. Unlike most medieval authors, Alyngton
was aware of the importance of the question, which he discussed at
length. According to the standard realist conception, not only the acci-
dental forms, such as whiteness, but also the compounds they cause
when inhering in substances, such as a white-thing (album), fall into the
categories. Burley thought that whilst the accidental forms properly fall
into the categories, the aggregates (aggregata) made up from a substance
and an accidental form do not, since they are beings (entia) per accidens,
wanting in real unity. In his opinion, such an aggregate may be said
to fall into a certain category, the category into which its accidental
form falls, only by reduction, in virtue of the accidental form itself.
21

On the contrary, Wyclif maintained that the aggregates built up by a
substance and an accidental form fall per accidens into both (1) the cat-
egory of substance and (2) the category which the accidental form at
issue belongs to.
22
Alyngton combines in an original way the two slightly
different opinions of Burley and Wyclif. He afrms that a thing (res) can
be said to belong to a category in a threefold way (tripliciter): by itself
( per se), by accident ( per accidens), and by reduction ( per reductionem). (1)
Something is in a category by itself if and only if the supremum genus of
20
Cf. Alyngton, In Cat., cap. de numero et sufcientia praedicamentorum, pp. 252253.
21
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de relatione, fol. 34rb: Illud
quod est aggregatum ex rebus diversorum generum non est per se in aliquo genere
uno. . . . Et ideo illud quod signicatur per huiusmodi nomina pater et lius non est
per se in genere, quia quod est per se in genere debet esse per se ens et per se unum,
sed aggregatum ex rebus diversorum generum non est per se ens nec per se unum;
and cap. de qualitate, fol. 41rb: Dico quod quamvis album vel nigrum non sit per se
in genere, tamen est in genere qualitatis per reductionem.
22
Cf. Wyclif , De ente praedicamentali, cap. 1, pp. 34.
that category is predicated
23
per se in recto of it,
24
that is, if and only if
the highest genus of that category is one of the constitutive elements
of the essence of the thing at issue. Accidental forms belong per se to
the nine categories of accidents.
25
(2) Something is in a category by
accident if and only if it is an aggregate from a substance and an acci-
dental form. Such aggregates belong per accidens both to the category of
substance and to the category in which its accidental form is by itself.
26

(3) Something can be in a category by reduction in two different ways:
in a broad sense (large) and in a strict sense (stricte). (3.1) Something is
in a category per reductionem large if and only if (i) it is an aggregate,
and (ii) the highest genus of a certain category is predicated only in
obliquo of it, that is, only indirectly. (3.2) Something is in a category per
reductionem stricte if and only if (i) it is not an aggregate, and, (ii.a) like
differences, it is a component of the reality of a thing which is in a
category by itself, but the highest genus of that category is not predi-
cated of it; or (ii.b) it is the privation correlated to a certain property
which, in turn, is in a category by itself; or (ii.c), like extra-categorial
principle such as God, the unity, and the point, it somehow instantiates
the mode of being proper to a certain categorial eld, but the highest
genus of that category is not one of the constitutive elements of the
essence of that thing.
27
Fundamental to Alyngtons deduction of the categories and solution
of the problem of what falls into the categorial elds (and how) is a
close isomorphism between language (mental, written, and spoken) and
the world. Like Burley and Wyclif , he was rmly convinced that our
thought is modelled on reality itself, so that it reproduces reality in all
its elements, levels, and inner relations. Therefore, one of the best ways
of understanding the world lay for him in an accurate investigation
of our notions and conceptual schemes, as they show the structure of
the world. A logical consequence of this conviction was his strong
propensity towards reication: he hypostatises the notion of being
23
According to Alyngton, predication is a real relation between things. Such a real
predication ( praedicatio realis) is matched by a corresponding predicative relation between
the signs (mental, written, and spoken) which signify those thingscf. In Cat., cap. de
regulis praedicationis, p. 247; cap. de substantia, pp. 273 and 287.
24
Cf. Alyngton, In Cat., cap. de numero et sufcientia praedicamentorum, p. 259.
25
Cf. Alyngton, In Cat., cap. de relativis, p. 300.
26
Cf. Alyngton, In Cat., cap. de numero et sufcientia praedicamentorum, p. 259; and cap.
de relativis, p. 300.
27
and considers equivocity, analogy, and univocity not only as semantic
relations between terms and things, but also as real relations between
extra-mental items.
28
According to the standard interpretation of the opening passages
of the Categories (1, 1a 112) equivocal terms are correlated with more
than one concept and refer to a multiplicity of things sharing different
natures, whereas univocal terms are correlated with only one concept
and refer to a multiplicity of things sharing one and the same nature. In
his last commentary on the Physics (after 1324), Burley had maintained
that the term being is at the same time univocal and equivocal with
respect to the categories, as (1) a single concept corresponds to it (broadly
speaking univocity), but (2) the categorial items share it in different ways,
according to a hierarchy of value (broadly speaking equivocity).
29
In
turn, Wyclif had admitted three main types of equivocity: by chance
(a casu), analogical, and generic, the second of which applies to the
relationship between being and categories.
30
Alyngton recognises four
main kinds of equivocity: by chance, deliberate (a consilio), analogical,
and generic. (1) Equivocals by chance are those things to which it just
happens that they have the same name, but with different meanings
and/or reasons for imposing the name. (2) Those things are deliberate
equivocals which have distinct natures but the same name, and are
subordinated to different but correlated concepts. (3) Those things are
analogical which share the nature signied by their common name in
various degrees and/or ways. (4) Generic equivocals are those things
which share the same generic nature in the same way, but have distinct
specic natures of different absolute value.
31
So, within Alyngtons
28
Cf. Alyngton, In Cat., cap. de aequivocis, fols. 69v70v; cap. de univocis, fols. 71v72r;
and cap. de numero et sufcientia praedicamentorum, pp. 255256.
29
Cf. Burley , Expositio in libros octo Physicorum, lib. I, tr. 2, cap. 1, ed. (Venetiis, 1501),
fols. 12vb13ra.
30
Cf. Wyclif , De ente praedicamentali, cap. 2, pp. 1617, 1819, and 21. On Wyclif s
theory of analogy see Conti , Wyclif s Logic and Metaphysics, pp. 103107.
31
Cf. Alyngton, In Cat., cap. de aequivocis, fol. 70rv: Ubi primo notandum, secun-
dum sententiam aliquorum, quod quadruplex est ae quivocum. Est enim aequivocum
a casu, aequivocum a consilio, aequivocum analo gum et aequivocum generale. Est
autem a casu quod imponitur casualiter ad signicandum di versa secundum rationes
dispares quarum nulla habet habitudinem ad aliam. Et illud est pure aequivocum.
Ut iste terminus Robertus a casu imponitur ad si gnicandum hominem Romae et
ad signicandum hominem Oxoniae. . . . Sed aliud est aequivocum a consilio, quod
est signum ex impositione si gnicans aliqua primarie sub diversis actibus secundum
rationes dispares qua rum una tamen est analoga vel proportionalis ad aliam, sed non
correspondet illi signo adaequate aliqua una intentio vel conceptus ad omnem rem
system, what differentiates analogy from univocity is the way in which
a certain nature (or property) is shared by a set of things: analogous
things share it according to different degrees (secundum magis et minus, or
secundum prius et posterius), univocal things share it all in the same manner
and to the same degree.
32
Alyngton argues that being (ens) is not a sort
of genus in relation to the ten categories, since it does not manifest
their essence, nor is it predicated univocally of them; being is analo-
gous in relation to the ten categories. It is a sort of basic stuff of the
metaphysical structure of each reality, which possesses it in accordance
with its own nature and level in the hierarchy of essences.
33
sic signi catam. Et isto modo iste terminus canis signicat caeleste sydus, marinam
beluam et animal latrabile, et forte propter proportionales proprietates re pertas in
omnibus istis rebus, puta acute mordendi vel aliud huiusmodi. . . . Et isto modo dicitur
sanum aequivoce de dieta, urina, medico, subiecto sano. Et ita de aliis. . . . Sed tertio
modo dicitur signum aequivocum analogum cuius primarium si gnicatum est analogum.
Et hoc contingit quando illud signicatum participa tur a pluribus secundum prius et
posterius quo ad ordinem intelligendi, vel secundum maius et minus; modo quo ens
participatur a substantia et accidente. Non enim contingit intelligere accidens, cum sit
modus substantiae, nisi praeintelligendo substantiam; immo omnis substantia est magis
ens quam accidens. Vel dicitur, secundum aliquos, ens a parte rei analogum eo quod
est par ticipatum a generibus diversis, quorum unum genus causat omne quod est per
se in alio genere et secundum quidlibet sui causat aliquid alterius generis; sicut ens
communicatur substantiae et accidenti. Nam omnis substantia causat aliquod accidens
et omne accidens causatur ab aliqua substantiased de isto posterius. . . . Sed quarto
modo dicitur signum aequivocum generale si signicet res di versarum perfectionum
essentialium. Et sic omne genus est aequivocum; propter quod dicit Aristoteles in VII
Physicorum quod in generibus multae latent ae quivocationes.
32
Cf. Alyngton, In Cat., cap. de univocis, fol. 71v: Notandum quod tripliciter est
aliquod signum aut eius primarium si gnicatum univocum. Primo modo dicitur omne
signum univocum quod sub uno conceptu plura adaequate signicat, sive praedicetur
de illis aequa liter, sive secundum maius et minus, sive secundum prius et posterius. Et
isto modo transcendentia sunt univoca. Sed secundo <modo> magis proprie dicitur
signum univocum si plura sub uno conceptu adaequate signicet quibus competit aliqua
differentialis pro prietas aeque primo. Sic omne genus praedicabile est univocum. Et
voco diffe rentialem proprietatem proprietatem essentialem non participatam secundum
maius et minus, per quam distingui tur ens unius praedicamenti vel generis a re alterius
praedicamenti vel gene ris. Et dico aeque primo propter analoga, quae secundum
maius et minus sive secundum prius et posterius competunt suis contentis, ut substantia
est tam maius ens quam prius ens quam accidens. Sed tertio modo strictissime dicitur
signum univocum quod solum res eiu sdem perfectionis essentialis sub eadem denitione
signicat. Sic species specialissima solum est univoca.
33
Universals and Predication
Among the many kinds of entia that Alyngton admits, the most important
set is that consisting of universal items. Universals are among the most
disputed topics in Medieval philosophical literature.
34
Like Wyclif and
the other Oxford Realists, Alyngton claims that universals fully exist
outside the minds and are really identical-to and formally distinct-from
the individuals which instantiate them.
Alyngton speaks of universals, individuals, and predication mainly
in the chapter on substance, but interesting remarks can be found also
in the chapter on subject and predicate (de subiecto and praedicato). Like
Albert the Great,
35
whom he quotes by name, Alyngton recognizes
three main kinds of universals: (1) ante rem, or ideal universalsthat
is, the ideas in God, the archetypes of all that is; (2) in re, or formal
universalsthat is, the common natures present in the individual
items as constitutive parts of their whole reality; and (3) post rem, or
intentional universalsthat is, the mental signs which signify formal
universals. The ideas in God are the causes of formal universals, and
formal universals are the causes of intentional universals.
36
Furthermore,
like Burley and Wyclif , Alyngton holds that formal universals actually
exist (in actu) outside our minds, and not potentially only (in potentia),
as moderate realists thoughteven if, unlike Burley, he maintains they
are really identical with their individuals, for otherwise it would be
impossible to explain, against the Nominalists, why and how individual
substances show different and more or less close kinds of similarity
among themselves.
37
Like Wyclif , Alyngton supports the thesis that formal universals are
common natures in virtue of which the individuals that share them
are exactly what they are, just as humanity is the form by which every
man formally is a man. As natures, they are prior and indifferent to
any division into universals and individuals. Although universality is
not a constitutive mark of the nature itself, it is its unique, inseparable
property. As a consequence, formal universals can be conceived of in
34
For a brief account of the problem of universals (and predication) in the (Later)
Middle Ages see Conti , Categories and Universals in this volume.
35
Cf. Albert the Great, De quinque universalibus, tr. de universalibus in communi, capp. 3
and 5, ed. Col., vol. 1.1A, pp. 2425 and 3132.
36
Cf. Alyngton, In Cat., cap. de substantia, pp. 276279.
37
Cf. Alyngton, In Cat., cap. de substantia, pp. 267268. See also p. 290.
two different ways: by themselves, as rst intentions, or in union with
and from the point of view of their inseparable property, that is, the
universalitas, and therefore as second intentions. In the rst case, they are
natures of a certain kind and are really identical with their own indi-
viduals. For example, man is the same thing as Socrates . In the second
case, they are properly universals (that is, something that can be present
in many things at once qua constitutive element of their essence), and
distinct from their own individuals, considered qua individuals, because
of the opposite constitutive principles: universalitas and particularitas.
38

Hence, universals are really (realiter) identical to, but formally ( formaliter)
distinct from, their individuals. In fact, universals are like formal causes
in relation to their own individuals, while individuals are like material
causes in relation to their universals.
39
Thus three different kinds of
items can be qualied as formal universals: (1) the common natures
(or essences) instantiated by individualscommon natures which are
things of rst intention; (2) the form itself of universality that belongs
to a certain common nature when seen in its relation to the individu-
alsform of universality which is a thing of second intention; (3) the
intelligibility proper to the common nature, by which it is a possible
object of our mindin other words, the real principle which connects
formal universals with mental universals.
40
Since Alyngton accepted the core of the traditional, realist account
of the relationship between formal universals and individuals, he, like
Wyclif , had to dene its logical structure more accurately, in order to
avoid the inconsistencies stressed by Ockham
41
and his followers. Thus
he restates that (1) a universal in the category of substance can directly
receive only the predications of substantial forms more common than
it; and (2) the accidental forms inhering in individual substances can
be predicated of the universal substantial form that those individu-
als instantiate only indirectly (essentialiter) in virtue of the individuals
themselves having that substantial form.
42
For this reason, Alyngtons
38
Cf. Alyngton, In Cat., cap. de substantia, p. 268.
39
40
Cf. Alyngton, In Cat., cap. de substantia, p. 277. This partition of the formal
universal is very similar to that propounded by Wyclif in his Tractatus de universalibus,
cap. 2, p. 64.
41
See Conti s article, Categories and Universals.
42
Cf. Alyngton, In Cat., cap. de regulis praedicationis, pp. 246248; cap. de substantia,
pp. 288289.
description of the logical structure of the relationship between universals
and individuals demanded a redenition of predication.
Alyngton presents his own theory of predication when commenting
on Categories 5, 3a45, where Aristotle seems to admit the possibility that
an accidental form can inhere in a universal substance, as he afrms
that the fact that an individual man is a grammarian implies the fact
that also the man and the animal are such.
43
In order to explain and
justify this afrmation, Alyngton introduces a new interpretation and
partition of the predication, different from both the standard one
and Wyclif s. Indeed, he was the rst one to ameliorate Wyclif s theory
(1) by cutting off Wyclif s habitudinal predication, and (2) by dividing
predication into formal predication ( praedicatio formalis) and remote inher-
ence (inhaerentia remota), or predication by essence ( praedicatio secundum
essentiam). Remote inherence is grounded on a partial identity between
subject and predicate, which share some, but not all, metaphysical
constituents, and does not demand that the form signied by the
predicate-term be directly present in the entity signied by the subject-
term. On the contrary, such a direct presence is required by formal
predication. Man is an animal and Socrates is white are instances of
formal predication; (What is) singular is (what is) common (singulare est
commune) and Humanity is (something) running (humanitas est currens)
are instances of remote inherence, since, according to Alyngton, the
property of running is imputable to the form of humanity, if at least one
man is running. He is careful, however, to use a substantival adjective
in its neuter form as a predicate-term, because only in this way can it
appear that the form signied by the predicate-term is not directly pres-
ent in the subject, but is indirectly attributed to it through its individu-
als.
44
Formal predication itself is in turn divided into formal substantial
and formal accidental predication, since formal predication necessarily
demands the direct presence of a form in a substrate, and, according
to Alyngton, this can occur in two different ways: either as one of the
inner constitutive element of the substrate (substantially), or as one of
its subsidiary properties (accidentally). Formal accidental predication
is then further divided into secundum motum and secundum habitudinem.
45

43
On this passage from Aristotles Categories and its signicance to the Aristotelian
theory of predication see James Duerlinger , Predication and Inherence in Aristotles
Categories, Phronesis 15 (1970), 179203.
44
45
The basic idea of this last division seems to be that modes of being and
natures of the accidental forms determine the set of substances which
can play the role of their substrate. Alyngton distinguishes between
those accidental forms that require a substance capable of undergoing
change ( per se mobile) as their own direct substrate of inherence, and
those ones which do not need a substrate with such a characteristic.
Forms like quantity, whiteness, risibilitas, alteration, diminution and so
on belong to the rst group, while relations of reason and respectus,
like causation, difference, dilectio and so on, fall under the second one.
The forms of the rst group bring about formal accidental predication
secundum motum, while the forms of the second group bring about for-
mal accidental predication secundum habitudinem. The former necessarily
entail singular substances as their substrates, since singulars alone can
undergo change, while only the latter can directly inhere in universal
natures ( possunt inesse denominative universalibus).
46
Substance
These remarks on the relations between the accidental forms and
substances bring us to the core of Alyngtons ontology: the doctrine
of substance, developed in the fth chapter of his commentary on the
Categories. Alyngtons discussion can be divided into two main parts. (1)
The rst attempts to clarify what characterizes substance, and therefore
(1.1) what falls by itself into this category; (2) the second is concerned
with the distinction between primary and secondary substances.
Alyngton lists seven opinions about the nature and mode of being of
substance, the last of which he supports. (1) According to the rst one,
proper to grammarians, substance is what the term substance refers to
when utilized in a broad sense, that is, the quiddity (quidditas) or essence
(essentia) of anything. In this case, substance is not a category, since the
items which fulll this description do not share any common nature.
47

(2) The second opinion is that of Avicenna , who afrms that any entity
which does not inhere in something else is a substance.
48
According to
46
Cf. Alyngton, In Cat., cap. de subiecto et praedicato, p. 245. See also cap. de substantia,
p. 274.
47
48
Cf. Avicenna , Liber de philosophia prima, tr. 8, cap. 4, ed. Simone Van Riet, 2 vols.
(Louvain, 197780), vol. 2, pp. 403404.
this view, God, substantial differences, and negative truths can be said
to be substances, even though only in an analogical way.
49
(3) A third
meaning of the term substance can be drawn from the use (of that
term) proper to common people and theologians: everything which
plays the role of foundation ( fundamentum) in relation to something else
is a substance. In this sense, the surface is the foundation (and therefore
the substance) of the whiteness.
50
(4) The fourth opinion seems to be
the same as the anonymous one discussed and partially criticized by
Burley in his last commentary on the Categories.
51
Substance would be
(i) a positive being, which (ii) does not inhere in something else, and
(iii) is naturally apt to play the role of subject (subiectum) in relation to
absolute accidents (that is, quantities and qualities). According to this
view, matter, form, the composite made up of matter and form, and
the angelic intelligences are substances, whereas substantial differences
and negative truths are not, since the former do not satisfy the third
requisite, neither the latter the rst one.
52
(5) The fth opinion is that
of Boethius ,
53
according to whom substance is (i) a positive being,
which (ii) does not inhere in something else, and (iii) is a compound of
matter and form.
54
(6) The sixth opinion is that of Burley,
55
to whom
Alyngton refers by the expression moderni logici. According to Bur-
ley, (i) not being in a subject, (ii) having an essence, (iii) autonomy and
independent existence, and (iv) the capacity of underlying accidental
forms are the main aspect of substances. This means that primary
substances alone are substances properly speaking, while matter and
form, and substantial differences are not.
56
(7) The last opinion is that
of Wyclif ,
57
quoted extensively and almost verbatim. Alyngton claims
that it is superior to the preceding ones (septima est expositio metaphysica
et altior ad intelligendum quam praenominatae). According to this view, the
constitutive principle of the substance is not the capacity of underly-
ing absolute accidents, but it is the capacity of underlying potency
and act, which are its inner foundationsthe capacity of underlying
49
50
51
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de substantia, fol. 22rbva.
52
53
Cf. Boethius , In Categorias Aristotelis libri quattuor, 184Ab.
54
55
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de substantia, fol. 24ra.
56
57
accidents being only a derivative property.
58
Wyclif s position about the
nature of substance implies that the distinction between potency and
act is, from the point of view of metaphysics, the most fundamental
distinction, of which that between form and matter is but one example.
As (1) prime matter is pure potentiality, while form is act, and (2) the
distinction between potency and act is wider than that between matter
and form, the latter is a particular case of the former. In fact, accord-
ing to Wyclif, the distinction between potency and act runs though
the whole of creation, since it applies also to any kinds of spiritual
creatures, whereas the distinction between matter and form is found
only in the corporeal creatures. On the contrary, Alyngton seems to
interpret the distinction between potency and act as a particular case of
the distinction between matter and form, since he constantly explains
the meaning of the opposition potency-act in terms of the opposition
between matter and form in a crucial passage that he quotes from
Wyclif s De ente praedicamentali:
59
According to this view, Aristotle says that primary substance exists in
a proper way, because it is an absolute thing which does not inhere in
something else, as accidents do (since their own being consists in inher-
ing). And primary substance is said to exist in an eminent way in relation
to potency and act (i.e., matter and form), which <i> are not complete
beings, nor <ii> exist in an eminent way, as they exist in virtue of the
composite. And primary substance is said to exist at the highest level in
relation to those three, <that is: accidents, potency, and act>. Hence,
primary substance is said to be at the highest level also in the chain of
our knowledge, since nobody can understand accidents if he has not
understood substance. Nor can anyone understand potency and act (i.e.,
58
Cf. Alyngton, In Cat., cap. de substantia, p. 267: Prima ratio substantiae est substare
potentiae et actui sicut suis intrinsecis fundamentis, et non substare accidenti absoluto,
cum hoc sit passio posterior substantiae.
59
Alyngton, In Cat., cap. de substantia, p. 267 (= Wyclif , De ente praedicamentali, cap. 5,
pp. 3839): Et secun dum istam sententiam dicit Aristoteles quod substantia prima pro-
prie est, cum sit res absoluta non inhaerens sicut accidentia, quorum esse est inhaerere.
Et principaliter dicitur in comparatione ad actum et potentiam (id est: ad mate riam et
formam), quae non sunt completa entia nec principaliter, cum sint propter compositum.
Et maxime dicitur in comparatione ad haec tria. Unde et maxime dicitur esse in notitia
hominum, cum nemo cognoscit accidens nisi praecognoscendo substantiam; nec ali quis
cognoscit potentiam vel actum (id est: materiam vel formam) nisi praeco gnoscendo per se
existens et transmutationem de uno esse ad aliud. Alyngtons additions are in italics.
Wyclif and Alyngton misunderstand the meaning of Aristotles statement about primary
substance (Categories 5, 2a1113), since the Stagirite is afrming that substance, in the
truest and primary and most denite sense of the word, is what is neither predicable
of a subject nor present in a subject.
matter and form) if he has not understood <i> what exists by itself, and
<ii> the change from one being to another.
The result is a sort of universal hylomorphism, since in this way mat-
ter and form serve as principles in the order of being as well as in the
order of becoming. All the more so because Alyngton seems to accept
the thesis that (angelic) intelligences are not pure forms existing by
themselves, but formal principles necessarily joined to the matter of
heaven in such a way as to make up living beings.
60
A second consequence of this approach to the problem of the nature
of substance is that, within his system, primary substances alone are sub-
stances properly speaking. This conclusion is conrmed by his analysis
of the distinctive mark ( proprium) of substance: while remaining numeri-
cally one and the same, being capable of admitting contrary properties,
the modication taking place through a change in the subject itself of
the motion at issue. Alyngton appears to think that this description is
satised only by primary (that is, individual) substances.
61
Secondary
substances therefore are per se in the category of substance only insofar
as they are constitutive parts of primary substances. Thus, secondary
substances belong to the category of substance by virtue of the indi-
vidual substances that instantiate them, since they are not formally
substances. In fact, unlike primary substances, secondary substances
are forms, and consequently incomplete entities with an imperfect
and dependent mode of existence. They require composite substances
in order to exist properly. No form as such, not even the substantial
60
61
Cf. Alyngton, In Cat., cap. de substantia, fol. 104v: Maxime autem substantiae
videtur esse proprium quod cum sit unum et idem numero, est susceptivum contrario-
rum. Haec est sexta proprietas. Pro qua notandum est quod duplex solet dici subiec-
tum transmutationis, scilicet per se subiectum et per accidens subiectum. Voco per
se subiectum transmuta tionis substantiam compositam ex materia et forma, cui inest
primo denominative transmutatio. Et dicitur per se subiectum transmutationis quia
est subiectum non subiectatum ulterius, quia primo sibi inest transmutatio. Sed per
accidens subiectum potest esse tripli citer. Aut quia non inest sibi transmutatio primo
denominative, sed secundum quid vel secundum partem, ut si dicatur hominem trans-
mutari quia pars eius transmutatur. Et tale potest esse per se subiectum transmutationis.
Sed secun do per accidens dicitur esse subiectum transmutationis omne accidens quod
po test recipere contrarias denominationes propter transmutationem subiecti per quod
extenditur vel in quo subiectatur. Et tale numquam potest esse per se subiectum trans-
mutationis, cum non potest non inhaerere. Et tertio omne contentum in alio, sive per
modum formae actuantis sive per modum locati moti ad motum sui contentis, dicitur
per accidens su biectum transmutationis; ut homo in navi, anima in corpore. Et talium
aliqua possunt esse per se subiecta transmutationum et aliqua non.
ones, is formally a substance, since no form as such has the capacity
of underlying potency and act. Secondary substances are related to
primary (or individual) substances as formal principles of the latter. It
is in this way that humanity and (say) Socrates are linked together. For
this reason no secondary substances as such are totally identical with
primary substances.
62
As a consequence, while man is animal (homo
est animal ) is a sentence to which a formal predication corresponds in
the world, a predication by essence matches humanity is animality
(humanitas est animalitas). Any (individual) man is an animal because
of the form of humanity present in him qua its essential constituent,
albeit the form of humanity as such is not the principle of animality.
Therefore, humanity is not formally animality nor rationality, even
though it is animality plus rationality.
63
Primary substances are the substrate of existence of any other kinds
of categorial being, as nothing exists in addition to primary substances
but secondary substances and accidents, which both are forms present
in individual substances. Like Aristotle (Categories 5, 2b56), Alyngton
can therefore afrm that primary substances are the necessary condition
of existence for any other items of the world: nothing could exist, if
primary substances stopped existing.
64
This does not mean, however,
that it would be possible to nd in the world a primary substance (1)
that would not belong to a certain species, and (2) without any accident
inhering in it. It means that, from the point of view of full existence,
accidents and secondary substances always presuppose primary sub-
stances, as to be a primary substance is to be an autonomous singular
existing item (hoc aliquid), whilst (1) to be a secondary substance is to be
an inner and essential determination (or form) of a primary substance
(quale quid), and (2) to be an accident is to be an outer determination
or aspect of a primary substance.
65
62
Cf. Alyngton, In Cat., cap. de substantia, p. 280. See also pp. 281 and 282283.
63
Cf. Alyngton, In Cat., cap. de substantia, pp. 283284. See also cap. de quantitate,
fols. 106v107r.
64
65
Cf. Alyngton, In Cat., cap. de substantia, fol. 101v: Ubi notandum est quod hoc
aliquid est individuum de cuius essentia non est, ut huiusmodi, cum alio componere
qualitative aliquid. Et sic nec materia prima nec forma substantialis materialis est
proprie hoc aliquid. Et solum ta le dicitur proprie ponere in numerum quod (qui ms.)
est quantitas discretaut patebit posterius. Ideo solum tale communiter vocatur unum
numero. Et sic non est universale formaliter unum numero, quamvis large loquendo de
uno numero, prout dicit ens quod cum alio constituit multitudinem, posset universale
dici unum numero. Sic ergo patet quomodo substantia prima repraesentat hoc aliquid.
Accidents
As we have just seen, Alyngton thinks of primary substances as the
ultimate substrates of existence in relation to anything else. As a con-
sequence, for him, the only way (1) to safeguard the reality of accidents
as well as their distinction from substance and from each other, while
at the same time, (2) to afrm their dependence on substance in exis-
tence, was to conceive of them as forms of the substance itself, and
therefore as something existentially incomplete. Accordingly, he insists
that quantity, quality, and relations, considered as abstract accidents,
are forms inherent in primary substances, whereas, if considered from
the point of view of their actual existence as concrete items, they are
not really distinct from the substance in which they are present, but
only formally, as they are modes of substances. So, the chief features
of Alyngtons treatment of accidents are (1) his twofold consideration
of them as abstract forms and as concrete properties as well as (2) his
commitment to their objective reality, since, in his opinion, they are
mind-independent items of the world in both cases. Hence, the main
goals of his reading of the chapters 68 of the Aristotelian treatise
are (1) showing the ordered internal structure of the chief categories
of accidents, and (2) reasserting their reality and real distinction from
the category of substance, against those thinkers, like Ockham and his
followers, who had attempted to reduce quantity and relations to mere
aspects of material substances.
Secunda vero substantia videtur sub gura appellationis repraesentare principaliter
hoc aliquid, ut cum hoc dixerit hominem vel animal. Sed non est verum. Sed magis
signicat quale quid, id est: substantialem qualitatem. Neque enim verum est quod
subiectum est, neque substantia secunda est unum numero quod subicitur, modo quo
primae substantiae subiciuntur; sed de pluribus dicitur secunda substantia, ut homo et
animal. Et quia aliquis posset credere quod substantia secunda signicat qualitatem,
modo quo albedo aut aliud accidens signicat qualitatem, removet hoc Aristoteles
dicens quod secunda substantia non signicat simpliciter qua le, quemadmodum album
et albedo, quia album (id est: albedo) nihil aliud signicat a subiecto quam qualita-
tem acci dentalem (id est: non signicat aliquod quid vel aliquam partem quidditatis
substantiae, sed solum qualitatem). Genera autem et species quae sunt secundae
substantiae determinant quamdam qualitatem circa substan tiam (id est: signicant vel
determinant qualitatem substantialem quae est pars quidditatis substantiae). Sic enim
vocatur differentia substantialis qualitas substantialis, V Metaphysicae, commento 19.
See also fol. 102rv.
1. According to the standard realist interpretation of the Aristotelian
doctrine of the categories, followed also by Alyngton,
66
among the nine
genera of accidents, quantity is the most important one, as it is the
basis of all further accidents, since quantity orders material substances
for receiving quality and the other accidental forms. On the contrary,
Ockham had claimed that it was superuous to posit quantitative forms
really distinct from substance and quality, since quantity presupposes
what it is intended to explain, that is, the extension of material sub-
stances and their having parts outside parts. As an accident, quantity
needs substance as its substrate of inherence.
67
Like Burley and Wyclif ,
Alyngton denies that material substance can be actually extended with-
out the presence of the general form of quantity in it, thereby afrming
its necessity. Hence, he tries to confute Ockhams argumentation. He
thinks that the existence of quantity always implies that of substance,
but he also believes that the actual existence of parts in a substance
necessarily implies the presence of a general form of quantity in it,
(1) really distinct from the substance (say Socrates ) in which it inheres,
and (2) formally distinct from the fact, grounded on the substance at
issue, that this same substance is a quantied thing. For Alyngton, what
characterizes quantity and differentiates it from the other accidental
forms, and in particular from quality, are the following features: (1)
being the appropriate measure of anything, and (2) being an absolute
entity which makes it possible that material substances actually have
parts outside parts.
68
66
Cf. Alyngton, In Cat., cap. de quantitate, fol. 106r: Quantitatis aliud continuum aliud
discretum. Hoc est secundum capitulum secundae partis huius libri, in quo tractatur
de secundo praedicamento, quod est quantitas. Sicut enim materia prima est prima
in via generationis substantiae materialis, ita quantitas, quae consequitur materiam
seu substantiam materialem, videtur esse prior in ordine procedendi quam fuit alia
accidentia consequentia substantiam materialem. Ideo ponitur a multis tamquam basis
aliorum accidentium.
67
Cf. Ockham, Expositio in librum Praedicamentorum Aristotelis, cap. 10.4, in Opera
philosophica, vol. 2, pp. 205224.
68
Cf. Alyngton, In Cat., cap. de quantitate, fols. 106v: Patet ergo quod quantitas
eo quod absolutum sufcienter distinguitur a generibus respectivis, et in hoc quod de
se habet partes extra partes distin guitur a qualitate; 107r: Omnis autem quantitas
habet rationem mensurae; 114v: Sed ultra notandum quod est mensurare secun-
dum replicationem et mensura re secundum adaequationem. Dicitur ergo quod omne
mensurare secundum adae quationem reperitur aeque primo in quantitate continua
sicut discreta. Nam subiectum quantum mensuratur adaequate per suam quantitatem
quo ad esse quantum, et aggregatum quo ad esse multum mensuratur adaequate per
numerum. Sed mensurare secundum replicationem inveni tur principalius in numeris,
If the summum genus of the category of quantity is a form, the seven
species Aristotle enumerates (line, surface, solid, time, place, number,
and speech) clearly are not. Alyngton tries to meet this difculty (1)
by reformulating the notion of quantied-thing (quantum), and (2) by
proposing a method for deducing the seven species of quantity from
the highest genus (a sort of sufcientia quantitatum). He considers the
seven species Aristotle lists not as quantitative forms, but as the most
proper and primary bearers of the quantitative nature, revealed by
the highest genus of the category. In fact, encouraged by the Aristo-
telian distinction between strict and derivative quantities (Categories 6,
5a38b10), like Burley ,
69
he distinguishes two different ways of being
quantied: by itself and per accidens. Only the seven species of quantity
would be quanta by themselves, while any other quantum would be such
per accidens, indirectly, because of its connection to one (or more) of the
seven quanta per se.
70
In Alyngtons view, these seven species of quantity
correspond to the seven possible ways of measuring the being (esse) of
the material substance. In fact, substance has two main kinds of esse:
permanent and in succession. And both of them can be either discrete
or continuous. In turn, the esse permanent and continuous of material
substance can be measured either from inside or from outside. If from
inside, then in three different modes: according to one, two or all the
three dimensions proper to material substances. In the rst case, the
measure is line, in the second surface, and in the third solid. If it is
measured from outside, then it is place. The being of material substance
that is permanent and discrete is measured by numbers. The being
that is in succession and continuous is measured by time. And nally,
cum ex numeris sciatur principaliter quotiens u num mensuratum contineat mensurans
secundum replicationem.
69
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de quantitate, fol. 29va.
70
Cf. Alyngton, In Cat., cap. de quantitate, fol. 116r: Notandum quod dupliciter
dicuntur aliqua quanta. Primo per se, sicut quantitates absolutae superius numera-
tae. Secundo modo per accidensquod po test esse tripliciter. Primo dicitur aliquid
quantum per accidens quod est su biectum formaliter quantum per suam quantitatem
quam subiectat. Et sic sola substantia materialis est quanta per quantitatemsicut e
contra quantitas est quanta per substantiam talem causaliter, sed non formaliter. Sed
se cundo modo dicitur quantum per accidens quod est accidens extensum per subiec-
tum quantum. Et tertio modo dicitur aliquid quantum per accidens per respectum ad
aliquid per se quantum, sicut angulus dicitur per accidens quantus in comparatione ad
basim, et actio quanta in comparatio ne ad tempus. Et sic motus dicitur habere duplicem
magnitudinem, permanentem scilicet et successivam. Et ita de aliis.
the being of substance that is in succession and discrete is measured
by the quantity called speech (oratio).
71
Alyngtons derivation of the seven species of quantity from a unique
principle common to them is unconvincing, but extremely interesting,
nonetheless, as it clearly shows that he wants (1) to stress the unity of
the category of quantity, which at rst appears heterogeneous, and
(2) to trace the problem of reality and real distinction of quantities
back to that of the nature and status of its distinctive mark.
2. Aristotles treatment of ad aliquid in the Categories is opaque and
incomplete, since (1) he does not have any notion of relation, as he
speaks of relatives and conceives of them as those entities to which
non-absolute terms of our language refer; (2) he does not discuss the
question of the reality of relatives; (3) he does not clarify the connec-
tion between the two denitions of relatives he proposes in the seventh
chapters of the Categories; (4) he does not give any effective criterion for
distinguishing relatives from some items belonging to other categories.
72

Because of these facts, in the late antiquity and in the Middle Ages
many authors tried to reformulate the doctrine of relatives.
The most successful and interesting attempt was that of the Neopla-
tonic commentators of the sixth century, such as Olympiodorus and
Simplicius . Unlike Aristotle, they were able to elaborate a notion of
relation (schesis) almost equivalent to our modern notion of two-place
predicates, as they conceived of relations as abstract forms whose dis-
tinctive feature was the property of being present-in and joining two
71
Cf. Alyngton, In Cat., cap. de quantitate, fol. 107rv: Sed ulterius pro divisione
quantitatis est notandum quod substantia, quae est basis et fundamentum omnium alio-
rum generum, habet esse permanens et habet esse successivum. Esse autem permanens
substantiae aut est continuum esse aut esse discretum. Et similiter esse successivum aut
est continuum aut esse discretum. Omnis autem quantitas habet rationem mensurae.
Aliqua ergo quantitas mensurat esse substantiae permanens et continuum: et hoc
secundum u nam dimensionem tantum, et sic est linea; aut secundum duas, et sic est
super cies; aut secundum tres, et sic quantitas corporea, quae dicitur corpus de genere
quantitatis. Potest enim corpus sumi vel pro substantia corporea vel pro quantitate molis,
permanente, longa, lata et profunda, de qua nunc loquitur. Omnes tamen istae tres
maneries quantitatum habent rationem mensurandi intrinsece . . . Secundum istos tres
modos potest substantia quanta mensurari ab extrinseco, et sic est locus eius mensu ra,
ut alius est locus linealis, et alius supercialis et a lius corporeus. Sed esse substantiae
permanens et discretum mensuratur per nu merum. Et esse eius successivum et con-
tinuum mensuratur per tempus. Et esse eius successivum discretum per quantitatem
mensuratur quae vocatur oratio. Ita quod in permanentibus discretis est numerus, in
successivis di scretis oratio. Et sic patet sufciens distinctio specierum quantitatis.
72
Cf. Ackrill , Aristotles Categories and De interpretatione, pp. 98103; and Julia Annas ,
Aristotles Metaphysics: Books M and N (Oxford, 1976), p. 198.
different substances at once.
73
This view was rejected by Latin medi-
eval authors. According to Boethius relation (respectus or habitudo) is an
accidental form which is in a substance (its substrate of inherence) and
simply entails a reference to another, without inhering in that other.
74

Albert the Great explicitly denied that a relation could inhere (in the
technical sense of the word) in two substances at once.
75
Some years
later, the same theses were held by Simon of Faversham.
76
Also Walter
Burley shared this approach, since it appeared to him to be the only
thing consistent with one of the basic principle of medieval metaphys-
ics: the equivalence and correspondence between accidental forms and
their substrates of inherence, so that no accidental form could inhere
at the same time and in full in two (or more) substances.
77
On the
contrary, Wyclif seems to support a different opinion, similar to that
of the Neoplatonists , as he maintained that relation (1) is different from
quality and quantity, since it presupposes them, and (2) qua such is a
sort of link between two things.
78
Alyngtons theory of ad aliquid is worthy of note, as he was the
only late medieval author who followed and developed Wyclif s ideas
on that topic. He conceived of relation (relatio) as an accidental form
which is present in both the relatives at onceeven though in differ-
ent ways, since it names only one of them. Consequently his relation
73
For a more detailed discussion of Neoplatonic theory of relations see Franoise
Caujolle-Zaslawsky , Les relatifs dans les Catgories, in Concepts et Catgories dans la
pense antique, ed. Pierre Aubenque (Paris, 1980), pp. 167195; Alessandro D. Conti ,
La teoria della relazione nei commentatori neoplatonici delle Categorie di Aristotele,
Rivista critica di storia della losoa 38 (1983), 259283; Concetta Luna , La relation chez
Simplicius, in Simplicius: sa vie, son oeuvre, sa survie, ed. Ilsetraut Hadot (Berlin, 1987),
pp. 113147.
74
For a short analysis of Boethius theory of ad aliquid see Alessandro D. Conti , La
teoria degli ad aliquid di Boezio: osservazioni sulla ter minologia, in Atti del Congresso
Internazionale di Storia della Logica. San Gimignano, 48 dicembre 1982, eds. V. M. Abrusci
E. Casari M. Mugnai (Bologna, 1983), pp. 247262.
75
Cf. Albert the Great, Liber de praedicamentis, tr. 4, cap. 10, pp. 240241.
76
Cf. Simon of Faversham, Quaestiones in librum Praedicamentorum, q. 43, pp. 137
138.
77
Cf. Burley , Expositio super librum Sex principiorum, cap. de habitu, in Expositio super Artem
Veterem Porphyrii et Aristotelis cit., fol. 63ra: Intelligendum quod nullum accidens unum
numero est simul secundum se totum in diversis subiective. Aliquod tamen accidens,
ut numerus, est in diversis subiective secundum suas partes. On Burleys theory see
Conti , Ontology in Walter Burley s Last Commentary, pp. 165170.
78
Cf. Wyclif , De ente praedicamentali, cap. 7, p. 61. On Wyclif s theory see Conti ,
Wyclif s Logic and Metaphysics, pp. 110113.
can be considered as a sort of ontological counterpart of our modern
functions with two variables, or two-place predicates.
79
According to Alyngton, whose account partially differs from those
of Burley and Wyclif , in the act of relating one substance to another
four distinct constitutive elements can be singled out: (1) the relation
itselffor instance, the form of paternity; (2) the subject of the rela-
tion, that is, the substance that denominatively receives the name of the
relationfor instance, the (substance that is the) father; (3) the object
of the relation, that is, the substance which the subject of the relation
is connected withfor instance, the (substance that is the) son; and (4)
the foundation ( fundamentum) of the relation, that is, the absolute entity
in virtue of which the relation inheres in the subject and in the object.
80

The foundation is the main component, since it (1) joins the relation to
the underlying substances, (2) lets the relation link the substrate to the
object, and (3) transmits some of its properties to the relation.
81
Like
Wyclif,
82
Alyngton afrms that only qualities and quantities can be the
foundation of a relation.
83
Some consequences about the nature and status of relations and rela-
tives derive from these premises: (1) relation is a categorial item whose
reality is feebler than that of any other accidental form, as it depends
upon the simultaneous existence of three different items: the subject,
the object, and the foundation.
84
(2) A relation can start inhering in
a substance without any change in the latter, but simply because of a
change in another substance. For example: given two things, one white
and the other black, if the black thing becomes white, then, because of
this change, a new accident, that is, a relation of similarity, will inhere
also in the rst thing, apart from any other change in it. (3) All the true
relatives (relativa secundum esse) are simultaneous by nature, since the real
cause of being a relative is relation, which at the same time inheres in
two substances, thereby making both ones relatives.
85
On this basis Alyngton can divide relations into transcendental
and categorial relations,
86
and, what is more, among the latter he can
79
Cf. Alyngton, In Cat., cap. de relativis, pp. 295296.
80
81
82
Cf. Wyclif , De ente praedicamentali, cap. 7, p. 67.
83
84
85
86
Cf. Alyngton, In Cat., cap. de relativis, pp. 290291.
contrast real relatives (relativa secundum esse) with relatives of reason
(relativa rationis) without utilizing references to our mental activities nor
to semantic principles. In fact, on the one hand, Alyngton describes
real relatives as those aggregates (1) made up of a primary substance,
(2) an absolute accidental form (quantity or quality), and (3) a relation
which correlates the substance at issue to another substance exist-
ing in actu. On the other hand, he dene relatives of reason as those
aggregates characterized by the occurrence of at least one of these
negative conditions: (1) either the relations subject of inherence or its
object is not a substance; (2) the object is not an actual entity; (3) the
foundation of the relation is not an absolute accident.
87
The strategy
which supports this choice is evident: Alyngton wanted to substitute
references to mental activity with references to the external world, thus
using only objective criteria, based on the framework of reality itself
in order to classifying things.
3. The chapter on quality is the least complex and interesting part
of the whole commentary, since Alyngton is faithful to Aristotles text
and doctrine, and sometimes even offers rather unproblematic analyses
and elucidations. The main general topic he deals with is the internal
structure of the category.
In the rst lines of the eighth chapter of the Categories (8a2526)
Aristotle observes that quality is among those things that are spoken
of in a number of waysan afrmation which seems to imply that
quality is not a summum genus, as, according to Aristotle himself, what
is spoken of in a number of ways always gathers in several different
natures. Furthermore, the Stagirite speaks of four species of quality
(habits and dispositions, natural capacities and incapacities, affective
qualities and affections, gures and shapes), without explaining how they
are related to one another and to the highest genus of the category.
No Aristotelian commentator had ever thought that quality was spoken
of in many ways purely equivocally. Therefore no Aristotelian com-
mentator had ever presumed that the term quality could have several
different (but connected) meanings. On the contrary, they unanimously
took for granted that it had a unique ratio, common to all the items
belonging to the category. They disagreed, however, about the status
and hierarchical order of the four species mentioned by Aristotle. For
example, Albert the Great held that quality at once and directly splits
87
up into the four species, which would all be equally far from the high-
est genus.
88
Duns Scotus ,
89
Ockham,
90
and Walter Burley
91
maintained
that the so called species of the quality were not properly species (or
intermediate genera), but modes of quality, since many singular quali-
ties would belong to the rst three species at the same time, as, unlike
species, modes are not constituted by opposite properties. Alyngton
rejected both opinions. The latter because it compromises the actual
goal of a correct categorial theory,
92
and the former because it does not
t in with the standard infracatagorial structure described by Porphyry
in his Isagoge. Consequently, he inserts an intermediate level between
the highest genus of the category and the four species by claiming that
quality is rst of all divided into perceptible (sensibilis) and non-percep-
tible (insensibilis) qualities. Affective qualities and affections, gures and
shapes stem from the former kind of quality, while habits and disposi-
tions, natural capacities and incapacities derive from the latter. In fact,
(1) gures and shapes are those perceptible qualities which inhere in
substances because of the mutual position of its quantitative parts,
while affective qualities and affections inhere in substances because
of the form itself of the substantial composite. (2) Natural capacities
and incapacities are inborn non-perceptible qualities, while habits and
dispositions are due to the activity, both physical and, if it is the case,
intellectual, of the substance in which they inhere.
93
88
Cf. Albert the Great, Liber de praedicamentis, tr. 5, cap. 2, pp. 245248.
89
Cf. Duns Scotus , Quaestiones super Praedicamenta, q. 36, pp. 497499.
90
Cf. Ockham, Expositio in librum Praedicamentorum Aristotelis, cap. 14.4, p. 271.
91
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de qualitate, fol. 38vb: Illa
quae hic ponuntur species qualitatis non sunt species ex opposito distinctae, quomodo
distinguuntur homo et asinus, sed species accipitur in proposito pro modo. Unde
quattuor sunt species qualitatis, id est: quattuor sunt modi qualitatis, qui modi non
sunt ex opposito distincti. Nam idem numero non continetur sub speciebus ex opposito
distinctis; sed idem numero est in prima specie qualitatis, in secunda et in tertia; ergo
illae species non sunt oppositae.
92
Cf. Alyngton, In Cat., cap. de qualitate, fol. 130r: Sed melius et consonantius est
dicere quod istae species distinguuntur ex opposito, ita quod nihil quod est per se in
una specie est per se in alia specie.
93
Cf. Alyngton, In Cat., cap. de qualitate, fol. 130r: Pro quo est notandum quod omnis
qualitas aut est active sensibilis aut insensibilis. Insensibilium autem qualitatum aliquae
sunt maturaliter acquisi biles vel naturaliter inexistentes, et omnes ta les et solum tales
dicuntur esse in secunda specie qualitatis; sed aliae sunt insensibiles praeternaturaliter
solum acquisibiles immediate per actionem, cuiusmodi sunt habitus corporei, ut sanitas
et inrmitas, et habitus spirituales, ut virtutes morales vel intellectuales, quae omnes
sunt insensibiles, quamvis occasionaliter sensu possunt percipi. Sunt etiam ex arte et
usu acquisibiles, quamvis aliqua sit sanitias connaturata animali. Nec generantur istae
qualitates a sibi similibus immediate, sed per transmuta tionem in qualitatibus primis
Concluding Remarks
As the preceding analyses show, Alyngtons theory of categories is an
interesting example of that partial dissolution of the traditional doctrine
which took place in between the end of the fourteenth and the begin-
ning of the fteenth centuries. Within Alyngtons metaphysics, (1) the
relationships between primary (or individual) substances and secondary
(or universal) substances, and (2) between substances and accidents (both
abstract and concrete) as well as (3) the inner natures of essential and
accidental predications are so different from their Aristotelian originals
that the general meaning of the categorial doctrine is deeply modi-
ed. According to Alyngton, the formal-and-essential predication and
the formal-and-accidental predication would correspond to Aristotles
essential and accidental predications respectively. But he regards remote
inherence as more general than formal predication. Therefore, in his
system formal predication is a sort of sub-type of the remote inherence.
This means that he recognises a single ontological pattern, founded
on a partial identity, as the basis of every kind of predicative relation.
Thus, the praedicatio formalis essentialis and the praedicatio formalis acciden-
talis are very different from their Aristotelian models, as they express
degrees in identity as well as the remote inherence. As a consequence,
the relationships between substance and accidents and between indi-
viduals and universals (and hence the ontological status of universal
substances and that of accidents) are completely changed. In Alyngtons
view, both concrete accidents, qua modes of individual substances,
and universal substances, qua the main components of the natures of
individual substances, are really identical-to and formally distinct-from
primary substances. Moreover, in the Categories Aristotle (1) character-
izes primary substances as those beings which are neither present in a
subject nor predicable of a subject, and (2) considers the capacity of
underlying accidents as the constitutive principle of substance, while
Alyngton (1) denes primary substance as what (i) is apt to underlie
potency and act, and (ii) has matter and form as its inner foundations,
and (2) explicitly afrms that underlying accidents is only a derivative
property of substance. Finally, because of his strong propensity towards
vel aliter per actiones animae. Et ista sunt in prima specie qualitatis. Si autem sint
qualitates sensibiles vel consequuntur positionem quantitatis, et sic sunt in quarta
specie, vel insunt absolute ra tione formae et non ratione positionis quantitatis, et sic
sunt in tertia spe cie.
hypostatization (as we have seen, Alyngton methodically replaces logical
and epistemological rules with ontological criteria and references), he
interprets Aristotles theory of homonymy, synonymy, and paronymy
as an ontological theory about real items and not as a semantic one
about the relations between terms and things.
In conclusion, Alyngton conates the real and logical (so to say)
levels into one: like Burley and Wyclif , he considers logic as (1) the
theory of the discourse on being (ens), (2) turning on structural forms
and relations, which exist in the world and are totally independent of
the mental acts by which they are grasped. It is through these structural
forms and relations that the network connecting the basic items of
reality (individuals and universals, substances and accidents) is clearly
disclosed. Yet, because of his peculiar ideas on substance and predica-
tion, his world is different from those of Burley and of Wyclif.
On Burley s view, macro-objects (i.e., what is signied by a proper
name or by an individual expression, such as Socrates or this particular
horse) are the basic components of the world.
94
They are aggregates
made up of really different items: primary substances and substantial
and accidental forms existing in them. Primary substances and sub-
stantial and accidental forms are like simple (or atomic) objects, each
possessing a unique, well-dened nature. Although they are simple,
some of these components are in a sense composite because they are
reducible to something elsefor example, primary substance is com-
posed of a particular form and matter. Primary substance differs from
the other components of a macro-object because of its peculiar mode
of being as an autonomous and independently existing objectin
contrast with the other categorial items, which necessarily presuppose
it for their existence. Primary substances are therefore substrates of
existence in relation to everything else. The distinction between sub-
stantial and accidental forms derives from their different relations to
primary substances: substantial, universal forms disclose the natures of
primary substances; by contrast, those forms that simply affect primary
substances without being actually joined to their natures are accidental
forms. As a result, the macro-object is not simply a primary substance
but an orderly collection of categorial items, so that primary substance,
94
On Burley s ontology of the macro-objects see Conti , Ontology in Walter
Burley s Last Commentary on the Ars Vetus, pp. 121176; and also Alessandro D.
Conti, Signicato e verit in Walter Burley, Documenti e studi sulla tradizione losoca
medievale 11 (2000), 317350.
even though it is the most important element, does not contain the
whole being of the macro-object.
On Wyclif s view, whatever is is a proposition (pan-propositionalism).
95

The constituti ve property of any kind of being is the capacity of being
the object of a complex act of signifying (omne ens est primarie signicabile
per complexum). This choice implies a revolution in the standard medieval
theory of transcendentals, since Wyclif actually replaces being with true.
According to the common belief, verum was nothing but being itself con-
sidered in relation to an intellect, no matter whether divine or human.
According to Wyclif, being is no more the main transcendental and its
notion is not the rst and simplest, but there is something more basic
to which being can be brought back: the truth (veritas) (or trueverum).
Only what can be signied by a complex expression is a being, and
whatever is the proper object of an act of signifying is a truth. Truth
is therefore the true name of being itself. From the ontological point
of view this entails the uniqueness in type of the signicata of every
class of categorematic expressions. Within Wyclif s world it is the same
(kind of ) object which both concrete terms and propositions refer to,
as primary substances have to be regarded as (atomic) states of affairs.
According to him, from the metaphysical point of view a singular man
(iste homo) is nothing but a real proposition ( propositio realis), where the
actual existence in time as an individual (ista persona) plays the role of
subject, the common nature, i.e., human nature (natura humana), plays
the role of predicate, and the singular essence (essentia istius hominis),
that is what by means of which this individual is a man, plays the role
of the copula. The result is that Wyclif s world consists of molecular
objects, which are not simple, but composite, because they are reducible
to something else, belonging to a different rank of reality, and unable
to exist by itself: being and essence, potency and act, matter and form,
abstract genera, species and differences. For that reason, everything one
can speak about or think of is both a thing (or molecular object) and
a sort of atomic state of affairs, while every true proposition expresses
either an atomic or a molecular state of affairs, that is the union (if
the proposition is afrmative) or the separation (if the proposition is
negative) of two (or more) molecular objects.
95
On Wyclif s pan-propositionalism see Laurent Cesalli , Le pan-propositionna-
lisme de Jean Wyclif, Vivarium 43 (2005), 124155; and Conti , Wyclif s Logic and
Metaphysics, pp. 7886.
On the contrary, Alyngtons world consists of atomic objects whose
constitutive elements are (1) abstract forms (or essences), both substantial
and accidental, (2) potency and act, and (3) matter. In fact, according
to him, unlike Burley , (1) what is signied by a proper name or by an
individual expression is a primary substance, and, unlike Wyclif , (2)
simple and complex expressions have different signicata.
96
Moreover, in
his view, Socrates cannot be regarded as an aggregate, since the beings
of the substantial universal forms predicated of him and those of the
concrete accidental forms inhering in him coincide with the being of that
primary substance Socrates himself is. Thus, if we consider Socrates
from the point of view of his being, Socrates is simply an atomic object,
a primary substance. If we consider him from the point of view of
the essences that he contains in himself, then he is a compound of
really different forms, which can exist only in it, as its components,
and through its being. This is the inner sense (1) of the formula really
identical and formally distinct that Alyngton employs for explaining
the relation between universals and individuals as well as the relation
between substance and concrete accidents; and (2) of his description
of the nature and peculiar mode of being of the primary substance:
to be a primary substance is to be the being of whatever can be.
96
Cf. Alyngton, In Cat., cap. de complexo et incomplexo, pp. 248250.
THOMAS MAULEVELTS DENIAL OF SUBSTANCE
Robert Andrews
With this paper
1
I hope to stimulate interest in an unjustly neglected
medieval author, Thomas Maulevelt. I would like to do so by draw-
ing attention to his ontology, one which entertains a hypothesis more
radical than any other of the Middle Ages, and one unparalleled until
the time of Hume .
Little is known about Maulevelts life. He was born in England,
but spent most of his active academic life in Paris in the 1320s and
1330s.
2
If he is famous at all, it is chiey for his works in the parva
logicalia, which were popular in the fteenth century, including at the
newly-established University of Uppsalasome of the manuscripts
of his work remain here.
3
It is even difcult to trace him through his
name; besides being called Thomas Anglicus, his name is also spelled
in multiple ways: Maulvelt, Maulfelt, Manlevelt, Mauleld, Manlefelt,
Maulefelt, and even Manlove. The text with which I shall here be
concerned is his commentary on Aristotles Categories. It is one of the
lengthiest of the Middle Ages, surviving in a single, densely-written
manuscript of over a hundred folios. In modern times it has hitherto
never been transcribed or even discussed.
1
The rst version of this paper was presented at a conference, Skepticism in Medieval
and Renaissance Thought, in Uppsala, Sweden, on May 8, 2005 (An ESF Workshop within
the program From Natural Philosophy to Science). I would like to thank the organizer of
that conference, Henrik Lagerlund, as well as the participants in the conference, for
their useful comments and suggestions.
2
The most extensive discussion of Maulevelts life is in Snke Lorenz , Thomas
Manlefelt (Maulefelt), Zu Leben und Werk, in Schule und Schler im Mittelalter. Beitrge
zur europischen Bildungsgeschichte des 9. bis 15. Jahrhunderts, eds. Martin Kintzinger, Snke
Lorenz, and Michael Walter (Beihefte zum Archiv fr Kulturgeschichte 42) (Kln, 1996), pp.
145164.
3
The inuence of Maulevelt is witnessed in Anders Piltz , Studium Upsalense. Speci-
mens of the oldest lecture notes taken in the medieval University of Uppsala (Uppsala, 1977). A
Maulevelt manuscript is also discussed in my Resoluble, Exponible, and Ofciable
Terms in the Sophistria of Petrus Olai, ms. Uppsala C 599, in Sophisms in Medieval
Logic and Grammar: Acts of the Ninth European Symposium for Medieval Logic and Semantics,
held at St. Andrews , June 1990, ed. Stephen Read (Dordrecht, 1993).
348 robert andrews
It is clear from the work that Thomas Maulevelt was following in
the footsteps of William Ockham, another Englishman writing a bit
earlier in the beginning of the 1300s.
4
Maulevelt adopts as central
the principle of Ockhams Razor (actually deriving from Aristotle)
that entities should not be multiplied beyond necessity, with which
Ockham sliced away at the undergrowth of medieval realism. Ockham
was responsible for an enormous conceptual shift within philosophy,
and has been accorded credit variously for the Scientic Revolution,
5

the Protestant Reformation,
6
and the modern world view.
7
As against his other targets, Ockham applied his reductive principle
to Aristotles ten categories, writing like Maulevelt a commentary on the
Categories of Aristotle.
8
The Categories in medieval ontology comprised
a division of the world into ten categories, in a division that sounds
familiar, if quaint, to us today: it consists of Substance, plus nine acci-
dents: Quality, Quantity, Relation, Action, Undergoing, Space, Time,
Situation, and Having.
9
Already before Ockham this list had begun to
appear problematic. It seemed to be arbitrary and improportionate:
should the nal categories of being in a reclining situation or hav-
ing shoes on be of the same status as quality or quantity? Problems
with relation, too, led to ontological embarrassment: if a relation may
itself be related, then between any two objects there must be an innite
number of relationships. Ockhams solution to this proliferation of
categories was to wield his razor. Entities not necessary to explanation
must be excised. So with one stroke Ockham sliced away eight of the
categories; in his ontology, only substance and quality remain as distinct
entities. All of the other categories are merely aspects of, or ways of
talking about, substances or qualities.
4
Another early Ockhamist is edited in Defensorium Ockham: An Edition, Cahiers de
linstitut du moyen-Age grec et latin 71 (2000), 189273, and discussed in The Defensorium
Ockham, in R. Andrews , ed. Essays in Honor of Fr. Gedeon Gl, O.F.M. on his Eightieth
Birthday: Franciscan Philosophy and Theology II. Franciscan Studies 54 (1995), 100121.
5
P. Duhem Le Systme du monde, 10 vols. (Paris, 191359).
6
Heiko Oberman , The Harvest of Medieval Theology: Gabriel Biel and Late Medieval
Theology (Durham, 1983) 45.
7
Gordon Leff , William of Ockham: The Metamorphosis of Scholastic Discourse (Manchester,
1975), p. xiii.
8
Guillelmus de Ockham, Expositio in librum Praedicamentorum Aristotelis, in Guillelmi
de Ockham Opera philosophica et theologica, Opera philosophica II, ed. Gedeon Gl (St.
Bonaventure, N.Y., 1978).
9
The nal nine categories are termed accidents, but may also be referred to as
properties, or somewhat misleadingly, qualities; here I shall use these three terms
interchangeably.
thomas maulevelts denial of substance 349
Maulevelt, like Ockham, was engaged in a radical reduction in the
number of the categories. One by one all of the lesser categories fall,
each shown to be merely a way of speaking. Maulevelt, however, is will-
ing to go a step farther than Ockham. Maulevelt entertains a unique,
radical hypothesis: we have no need to posit that substance exists! The
only category needed to describe the things of the world is that of
quality. Lets investigate how this remarkable proposal came about.
Maulevelts ideas are presented within the framework of a Categories
commentary, and to be able to understand them one has to understand
the conventions of the medieval philosophical question-commentary.
A medieval question commentary never presents an authors opinion
immediately, but instead rst delivers it at the middle or near the end
in a determinatio. Therefore one rst has to wade through the questions
preliminary arguments, which by convention argue for the position
ultimately to be rejected. Such arguments were framed as counterfac-
tualand since in a preliminary argument an author could openly
challenge dogma or the tenets of Aristotle, these arguments provide
some of the most intriguing reasoning to be found in the Middle Ages.
Such is certainly the case in Maulevelt.
But rst it should be remarked that the very title of Maulevelts ques-
tion regarding substance is unusual. Titles of commentary questions in
the Middle Ages are signicant, for they indicate that a specic prob-
lem was thought worthy of discussion. Some titles are indeed routine
exercises which everyone was expected to ask and answer, while others
reect the particular interests of an author or philosophical school.
10

But among all the traditions and debates of Categories commentaries,
no one before or after ever posed the question Maulevelt does: Does
substance exist?
11
The reason for this, I suspect, is that it was held as
obvious and self-evident that, of course, substance exists. Certainly no
one ever asked the question in order to answer it in the negative.
10
See my discussion in Question Commentaries on the Categories in the Thirteenth
Century, Medioevo: Rivista di storie della losoa medievale 26 (2000), 141183.
11
See infra, Maulevelt, Quaestiones super Praedicamenta, q. 16, 1.
350 robert andrews
The Preliminary Arguments
Of the six preliminary arguments in Question 16, I intend to discuss
only two. The rst couple of preliminary arguments may be dismissed
as implausible semantic quibbles; arguments ve and six are based upon
theological problems. Notable, however, in the second argument is that
Maulevelt cites approvingly Ockhams razor: Entities should not be
multiplied beyond necessity.
12
It is in the third preliminary argument that we are presented with
the paradigm evidence for Maulevelts skepticism about substance: the
Eucharist. Here is his argument:
We do not experience, nor are we convinced by any argument, that
there is any substance, therefore we should not suppose that there is
any substance. The argument is clear . . ., because naturally speaking we
dont experience, nor have any reason to believe, that there is any more
substance in an unconsecrated communion wafer than in a consecrated
one (according to theologians); for the same reason we cannot experience
nor prove that there is any substance in an unconsecrated communion
wafer, and likewise for any other thing whatsoever.
13
The background of this argument is, of course, the church doctrine of
the Eucharist, the bread and wine which at the moment of a priests
blessing is turned into body and blood. In this consecration, none of the
external appearances of the bread or wine is thereby altered; their quali-
ties remain the same, and only their substance has been altered, from
mundane to divine. We have no awareness of the substantial change,
and by implication, we would have no awareness of any other such
substantial change that might take place in other circumstances. The
conclusion is that we have no awareness of substance whatsoever.
The same example is reinforced in the fourth preliminary argument.
14

If no substances are supposed, still all the appearances can be saved.
This phrase, saving the appearances, is important for Maulevelts
argument in a way that it later comes to be important for the scientic
revolution of the 17th century. Copernicus can defend his heliocentric
12
Ibid., 5, Multiplicity ought not be supposed without necessity.
13
Ibid., 7.
14
Ibid., 8: All the appearances can be saved without supposing any substance,
therefore we ought not suppose any substance. The argument seems clear. The ante-
cedent is shown again by the consecrated and unconsecrated communion wafer: the
unconsecrated communion wafer has color, size, edibility, a good taste, and all its other
appearances, just like the consecrated wafer or communion wine.
model by claiming that it saves the appearances (that is, accounts for all
phenomena) as well as the geocentricbut the heliocentric system is to
be preferred for its simplicity. Likewise, the example of the Eucharist
for Maulevelt shows that external properties are all that we witness in
a case of transubstantiation; whatever internal status there might be
remains inaccessible to us. And so we need never posit an underlying
substance to account for the appearance of any object whatsoever.
If these preliminary arguments were all there are to Maulevelts case,
they could be dismissed as sterile exercises, nothing more than intrigu-
ing arguments for the sake of argument. However, it is what follows
in the main body of Maulevelts treatment that is truly unusual. In his
determinatio, his considered opinion on the matter, Maulevelts conclusions
are not what is expected according to the traditional structure of the
medieval commentary. In two of four conclusions, the author accepts
as at least probable the initial proposal: no substance exists.
Conclusion One: No Substance in the Terrestrial World
In his rst two conclusions, with responses to various objections, Mau-
levelt claims that it is a defensible position that there is no substance
anywhere in the world, whether terrestrial or celestial. He does this not
only negatively, by arguing that substance is unnecessary for explanatory
purposes, but also positively, by explaining how natural phenomena
may be understood in the absence of substance.
Maulevelt begins by limiting his case against substance to earthly
things (in istis inferioribus). His argument is, as above, based on saving
the appearancesnothing is lost in explanatory power if substance
is sliced away using Ockhams razor. And, again, his example is the
consecrated Eucharist. He appeals to various aspects of change (using
Aristotelian terminology) in the following account:
The rst conclusion is this, which physically speaking can be maintained as
probable, that there is no substance whatsoever among earthly thingstak-
ing substance to mean the composite of matter and form, or any part
of such a composition. This is shown thusly: any theory whose opposite
cannot be proven with evidence (evidenter) should be held as probable;
but [this conclusion] is like this; therefore [it should be maintained].
The minor premise is proven, because all of the appearances can be
saved without positing any substance; therefore it cannot be proven with
evidence that there is any substance among earthly things. . . . The ante-
cedent is furthermore declared, because supposing that there exist only
352 robert andrews
accidents which support and adhere to each other, all of the following
can be saved: generation and corruption; increase and decrease; change
and local motionall of this is evident in the case of a consecrated
communion wafer.
15
Here also Maulevelt gives us a rst hint of the mechanism he shall
propose to explain the natural world in the absence of substance: his
alternative explanation is that accidents . . . support and adhere to
each other. Thus the picture is of a world in which physical objects
are adhering aggregates of their properties: an apple is red because it
smells pleasant, is cold because it is solid, and tastes sweet because it
is round. All of its properties adhere to one another, and form each
others substrate, without the need for an intangible and inaccessible
substance.
Let me pause to aver that this radical reformulation of nature is
completely unknown otherwise in the Middle Ages. It does, however,
sound surprisingly similar to the deductions of the British Empiricists
400 years subsequently. David Hume was to write in 1739:
I would fain ask those philosophers, who found so much of their reason-
ings on the distinction of substance and accident, and imagine we have
clear ideas of each, whether the idea of substance be derived from the
impressions of sensation or of reection? If it be conveyed to us by our
senses, I ask, which of them; and after what manner? If it be perceived
by the eyes, it must be a colour; if by the ears, a sound; if by the palate,
a taste; and so of the other senses. But I believe none will assert, that
substance is either a colour, or sound, or a taste. The idea, of substance
must therefore be derived from an impression of reection, if it really
exist. But the impressions of reection resolve themselves into our pas-
sions and emotions: none of which can possibly represent a substance.
We have therefore no idea of substance, distinct from that of a collection
of particular qualities, nor have we any other meaning when we either
talk or reason concerning it.
16
There is no evidence to suggest that Maulevelt the Englishman could
have inuenced his skeptical countryman of 400 years later. Indeed, it
is almost certain that Maulevelts commentary on the Categories, which
survives today only in a single manuscript, had little or no direct impact
on subsequent generations. Rather, Maulevelt must have made his own
15
Ibid., 21.
16
David Hume , A Treatise of Human Nature, Sect. VI, Of Modes and Substances,
(New York, 1992).
extrapolations from the reductionism of Ockham, and proposed a radi-
cal parsimony that was to be unequalled until an age unshackled from
the medieval world-view.
To return to Maulevelts argument, in the text he next faces a series
of self-penned objections to his proposala not at all unusual pro-
cedure in a medieval commentary, where interpolated doubts could
be addressed in order to clarify the authors position. In this case, the
doubts deal with various aspects of Aristotelian physics which seem
compromised by Maulevelts stance.
First, according to Aristotle, any generation of a new entity is dened
as change to a substantial form.
17
Obviously, Maulevelts new concep-
tion does not agree with this denition. Accordingly, Maulevelt rejects
it: generation is the perfect production of some thing into being, and
corruption is its perfect destruction.
18
Thus Maulevelt is willing and
prepared to modify the details of Aristotelian science.
Next, Maulevelt confronts the weightiest objection to his theory:
what is the underlying subject when an object undergoes a supercial
or accidental change? Maulevelt here makes explicit what he intimated
earlier: the subject for any accident is always another accident: And
it might be asked, what might be the subject in such generation? The
response is that one accident is the subject for another accident in cases
of generation and corruption. The theologians must necessarily hold
this when they say that the Word is generated out of a communion
wafer being consecrated, or when some color such as white or red is
introduced onto a consecrated communion wafer.
19
In two further objections,
20
Maulevelt counters difculties arising
from a case of radical transformation, such as when water changes
to rethat is, when one basic element changes into its opposite. On
the one hand, it seems that in such a drastic change, there are no
properties remaining which can serve as an underlying foundation for
the new properties. Is this, then, a case of an entirely new creation
out of nothing, ex nihilo? Aristotle would deny that there can be any
creation ex nihilo.
Maulevelt provides two possible ways of answering this crucial
objection. One is an ad hoc solution: perhaps there never is such a
17
Maulevelt, Quaestiones super Praedicamenta, q. 16, 22.
18
Ibid., 23.
19
Ibid., 245.
20
Ibid., 268.
354 robert andrews
case of transformation of an element into its direct opposite; perhaps
such changes must always pass through an intermediate form, so that
water must rst change into air before becoming re.
21
Maulevelts
other solution, which he calls more probable, is to claim that any
natural change comes about because of a stronger force. If water
were to change into re, this would have to be because of a powerful
inuenceand so the possibility for change does not have to lie within
the water, which is the passive part; instead the possibility would lie
in the more powerful force which is responsible for the change in any
aggregate of qualities.
22
Conclusion Two: No Substance in the Celestial World
In his second important conclusion, Maulevelt moves to a discussion
of the celestial world. All appearances may be saved, he says, even if
there is no substance posited among the heavenly bodies. The celestial
bodies may be regarded as composed out of various accidents perpetu-
ally adhering to each other. In a surprising analogy, Maulevelt says
that the mover of any celestial orb [may be regarded as] an accident
inhering in that orb, just like gravity ( gravitas) inheres in a rock, mak-
ing it fall. This comparison is astoundingly prescientto suggest that
what moves a planet is like gravity in a rockbut there is no reason
to suppose anything other than sheer coincidence in the association to
Newton s correct description of gravity as the mover of planets. Rather,
Maulevelts basis for his comparison is rmly medieval: A consecrated
communion wafer, in which there is no substance, falls just like an
unconsecrated one.
23
In what immediately follows, Maulevelt provides further interest-
ing arguments, mostly dealing with minutiae of medieval theology.
But I shall omit those arguments here, to move to his dramatic nal
conclusions.
21
Ibid., 29.
22
Ibid., 31.
23
Ibid., 33.
Conclusions Three and Four: The Acceptance of Substance
In his conclusions three and four, Maulevelt shows the other side of
the coin: it is equally probable that substance exists. Maulevelt does not
provide much argument for this viewpoint, since it is what everyone
commonly holds. He merely lists all the types of substance that can be
said to exist: God, angels, celestial bodies, the four elements earth, air,
re, and water, matter and form, the human soul, and ordinary bod-
ies.
24
(In his fourth and nal conclusion he also reafrms the nominalist
position that substance as a term of second intention, such as genus
and species, also exists.)
There is a striking symmetry in his four conclusionstwo in favor of
substance, and two againstwhich shows that the rejection of substance
is regarded as equally plausible as its acceptance. Maulevelt did not
pose his doubts about substance as merely contentious preliminary argu-
ments; rather he weighed his claim in balanced opposition to all con-
ventional acceptance of substance. His arguments against the existence
of substance are necessarily more extensive because this is the radically
unfamiliar position. Nevertheless, that he should evaluate the evidence
as equally probable for both sides is a provocative agnosticism against
the monolithic Aristotelian certainty regarding substance.
Why did Maulevelt not pursue a denitive denial of substance, as
Ockham did earlier with his rejection of eight accidental categories?
Certainly Maulevelt would have faced alarm and hostility from tradi-
tionalists confronted with a total rejection of substance. It was certainly
no small matter to challenge the authority of Aristotle, as even Ockham
was hesitant to do; it was even more serious to have to reckon with all
24
Ibid., 43: The third conclusion, which can be held as probable, since everyone
holds it, is that there is one rst substance which depends on nothing else, and that is
God himself. Also, there are other individual and incorruptible separate substances,
which indeed depend for the being upon the First Substance, but which subsist per se,
and are neither accidents nor essential parts of anything else. Also, other substances
are incorruptible bodies, such as celestial bodies, which depend simply upon the First
Substance but subsist per se in the way that has been described. Other substances are
composites of matter and form, but not out of other parts, and are homogenous; but
these are corruptible, such as the four elements. There are other corporeal substances
not composed out or prime matter and form, but out of homogenous parts like elements;
these are called mixed or elemental substances. Finally, there are matter and form,
which are the parts of corruptible substances. Some of these are incorruptible, such
as the intellective soul, and matter, which is the foundation and subject of generation
and corruption. Others are corruptible, like other substantial forms. Since everyone
holds this position, it need not be explained further.
356 robert andrews
the theological difculties which would have arisen from a conceptual
shift that dispensed with the fundamental notion of substance.
I suggest that Maulevelt likely did not have the intellectual resources
at his disposal to be able to systematically revise the substance/quality
distinction. The strength of Maulevelts challenge to substance lies in
its ability to save the appearancesthat is, it is able to account for
our observations of the natural world. Yet original observation was
something new at this time, and it is with Ockhams innovative emphasis
upon the individual that the controlled observation and experimenta-
tion of the scientic method is in its infancy. Maulevelts explanations
might have been able to save the appearances, but they could not have
saved the theorization out of which so many medieval systems were
constructed. To have attempted to apply Maulevelts insights into more
complex areas of physics and natural science would have been an enor-
mous undertaking, and would surely have run afoul of the watchdogs
of orthodoxy and institution.
Having said this, I would yet assess Maulevelts hypothesis as insight-
ful and courageous. It was indeed bold in the early 1300s to accept
Ockhams radical reduction of categorical entities, and to attempt to
extend Ockhams reduction even further. And it was also perspicuous
to see that there are reasonable grounds to question the cogency of the
idea of substancesince it is something of which we have no direct
experience, as Berkeley, Hume , and even Kant later were to recognize;
and to question its indispensabilitysince the functions of substance
could conceivably be replaced by an aggregate of qualities.
Furthermore, I think it is important to note that Maulevelts ideas
are not essentially theological. True, his inspiration for questioning
the existence of substance is rooted in the miraculous transformation
of the Eucharist; but the Eucharist merely provides the occasion for
observing that substance is hidden from our observation. Maulevelt is
able to readily transfer this claim to any object in the world because of
his willingness to pursue Ockhams reductionism to a logical extreme,
and because of his insightful distinction between what is known by
observation and what is presupposed by theorization.
Maulevelts insight built upon an implicit weakness in Aristotles sys-
tem, one made apparent by Ockhams radical reductionism. Maulevelt
hypothesized a step further than Ockhammoving from Ockhams
ontology of substances and qualities, to one of qualities alone. Whether
he would have been able to sustain this position indenitely, when faced
with indignant attacks by his contemporaries, we shall never be able
to know. His presentation of the hypothesis as merely plausible, and
as equally saving the appearances, made him retreat from committing
to such a radical experiment. But it was an experiment which was to
recur, with a vengeance, in the era of Early Modern Philosophy and
the Scientic Revolution.
Maulevelts speculation remains a footnote to the history of phi-
losophy, since his work was uninuential in his lifetime, and remained
unread until our own. But the discovery of such an inspired innovator
makes a strong case for the value of recovering the treasures still buried
in medieval manuscripts.
QUAESTIONES SUPER VETERI ARTE FF. 1RA145VB
QUAESTIONES SUPER PRAEDICAMENTA
ERFURT SB AMPL. Q 288 (XIV) FF. 43RB145VB
Thomas Maulevelt
f. 55va
<Quaestio 16>
Consequenter quaeritur utrum aliqua substantia sit.
1. Et arguitur primo quod non. Nihil est per se subsistens; igitur
nihil est substantia. Consequentia videtur ex hoc evidens, quia illa est
differentia inter substantiam et accidens, videlicet quod substantia est
illud quod per se existit, et accidens quod alteri inexistit. Antecedens
patet, quia omne quod existit cum alio existit, sicut prima substantia
cum secunda, et secunda cum prima.
2. Item, si aliqua substantia esset, vel illa esset divisibilis vel indivisi-
bilis. Non divisibilis, quia omne divisibile est divisibile in aliquas partes;
et omnis pars est relatio, quia alicuius totius pars; et omnis relatio est
accidens, quia adest et abest praeter subiecti corruptionem;
1
igitur
omnis pars est accidens. Sed nihil compositum ex accidentibus [non]
est substantia, quia ex non-substantiis non potest eri substantia,
secundum Aristotelem primo Physicorum;
2
igitur nullum compositum
est substantia.
Item, cuiuscumque compositi partes sunt quantitas continua vel
discreta; igitur non sunt substantiae. Consequentia videtur evidens. Et
antecedens est de se notum.
Et quod non sit aliqua substantia indivisibilis probatur, quia nihil
tale experimur; nec est aliquid ratio cogens ad hoc ponendum; igitur
nullum tale a nobis |55vb| est ponendum. Consequentia patet per
1
Porphyrius Isagoge, 20.78: Accidens vero est quod adest et abest praeter subiecti
corruptionem.
2
Cf. Aristoteles Physica I,7 190a345.
1
2
3
4
5
thomas maulevelt: QUAESTIO 16 359
Aristotelem primo Physicorum,
3
ubi vult quod pluralitas non est
ponenda sine necessitate.
Item, si aliqua substantia est indivisibilis, illa esset maxime prima
causa; sed illa, secundum theologos, contineat in se tres personas dis-
tinctas; igitur non est indivisibilis. Et voco ad praesens indivisibile illud
quod non habet distinctionem aliquorum rerum in se.
3. Item, non experimur nec ratione convincimus aliquam substantiam
esse; igitur substantia non est a nobis ponenda. Consequentia patet,
ut prius. Antecedens declaratur, quia non habemus aliquem experien-
tiam vel rationem naturaliter loquendo quod magis substantia sit in
hostia non consecrata quam in hostia consecrata, et per nihil possimus
experiri vel ratione probari substantiam esse in hostia consecrata, sicut
patet per theologos; igitur pari ratione per nihil possimus experiri vel
probare substantiam esse in hostia non consecrata; et pari ratione nec
in aliqua alia re.
4. Item, omnes apparentiae possunt salvari, non ponendo aliquam
substantiam; igitur nulla substantia est ponenda. Consequentia vide-
tur evidens. Antecedens declaratur adhuc per hostiam consecratam
et per hostiam non consecratam, quia \sicut/ hostia non consecrata
est colorata, quanta, apta ad intereundum, bonum habens saporem
gustis, et sic de quibuscumque apparentiis, ita etiam hostia consecrata
vel vinum consecratum.
5. Item, si aliqua substantia esset, hoc maxime videtur de prima
substantia, utpote de prima causa; sed ista non est; igitur nec aliqua
alia. Assumptum probatur, quia si prima causa esset, tunc esset illud
quo maius excogitari non potest, sicut patet per theologos dantes
signicationem huius vocabuli deus; et sicut satis expresse dicitur a
Boethio libro tertio De consolatione philosophiae;
4
sed nihil est tale
quo maius excogitari non potest, quia quocumque tali posito, posset
excogitari adhuc quod essent duo talia, et per consequens duplum ad
unum istorum potest excogitari, et triplum, et sic in innitum.
3
Aristoteles Physica I,4 188a178; Hamesse , Auctoritates Aristotelis, p. 141 (26): . . . pec-
catum est eri per plura quod potest eri per pauciora; cf. Guillelmus de Ockham
Scriptum in librum primum SententiarumOrdinatio I prol. q. 1 (OTh I 74); cf. Gualterus
de Chatton Reportatio II d. 2 q. 1 (cod. Paris. bibl. nat. lat., 1587) f. 115va: paucitas
ponenda ubi pluralitas potest vitari per connotationes.
4
Boethius, De consolatione philosophiae III (1973): quo uero, inquit, habitet, ita con-
sidera. deum, rerum omnium principem, bonum esse communis humanorum conceptio
probat animorum; nam cum nihil deo melius excogitari queat, id quo melius nihil est
bonum esse quis dubitet?
6
7
8
9
360 robert andrews
6. Item, si aliqua substantia est composita ex materia et forma, tunc
materia esset; et tunc omnis talis materia esset deus, quia sibi competent
omnes proprietates dei, utpote incorruptibilitas, immutabilitas, et inni-
tas. Et quod incorruptibilitas patet per Aristotelem primo et secundo
Physicorum. Et quod immutabilitas declaratur, quia non videtur mutari
nisi quia aut habet aliquid quod prius non habuit, vel non habet quod
prius habuit; sed propter hoc non debet dici mutabilis, cum ipsa in se
praecise eadem remanet, quia deus ipse [non] habet quod aliquando
non habuit, utpote materiam humanam et naturam humanam, et tamen
propter hoc non dicitur mutabilis. Et quod innitas |56ra| probatur,
quia materia est potens ad recipiendum non tot formas quin plures;
5

igitur est innite potentiae receptiva.
Ad oppositum est Aristoteles in textu,
6
et communis opinio, tam phi-
losophorum quam theolologorum; et principio alicuius inquisitionis vel
opinionis vel philosophiae.
<Distinctio prima> <Prima> distinctio est ista, quod iste terminus
substantia potest accipi dupliciter. Uno modo secundum quod est
terminus primae intentionis. Alio modo secundum quod est terminus
secundae intentionis.
Secundum quod est terminus primae intentionis, accipitur multipli-
citer. Uno modo proprie et strictissime, scilicet pro illo quod per se sub-
sistit, ita quod non pars alicuius per se unius; et simul cum hoc a nullo
dependet. Et isto modo accipitur tantummodo pro prima causa.
Alio modo accipitur quodammodo largius, scilicet pro illo quod per
se subsistit, ita quod non est pars alicuius per se unius, sive ab alio
dependeat sive non. Et isto modo extendit se ad substantias separatas
et ad formas compositas ex quibus aliae substantiae non componuntur,
sicut ad angelos et ad omnes alias substantias per se existentes.
Tertio modo accipitur adhuc largius, secundum quod se extendit ad
illud quod per se subsistit, et quod est pars essentialis per se naturaliter
5
Cf. Petrus Hispanus Tractatus XII,38 (1972) 2312; Richardus Sophista, Abstrac-
tiones (cod. Oxon. Bodl. Digby 24, f. 79va): Et eodem modo dicendum est ad istam
probationem: non duo quin plura sunt nita etc.; Gualterus Burlaeus De puritate artis
logicae tractatus brevior (1955) 260; Ioannes Duns Scotus Quaestiones super libros Metaphysi-
corum II q. 46 n. 32 (OPh III) 248: Sed non essent innitae actu, quia non sunt tot
quin plura possunt esse.
6
Aristoteles Cat. 5 2a124b19.
10
11
12
13
14
15
subsistens. Et isto modo materia et forma ex quibus componitur sub-
stantia sunt substantia.
Quarto modo accipitur largissime, scilicet pro cuiuscumque rei essen-
tia. Et isto modo convertitur cum ente, sicut accipitur in illa denitione
grammaticali nomen signicat substantiam cum qualitate, proprio vel
communi.
7
Secundum vero quod est secundae intentionis, accipitur pro signis
ipsas substantias primas tribus modis acceptas absolute vel proprie vel
appellatione signicantibus, cuiusmodi sunt tales terminos iste homo,
Sortes, homo, animal, corpus, substantia, et consimiles.
<Distinctio secunda> Secunda distinctio est ista, quod per se existere
accipitur multipliciter. Uno modo scilicet pro illo quod non est pars
alicuius per se unius, et a nullo dependet vel sustinetur. Et isto modo
solus deus per se subsistit.
Alio modo accipitur pro illo quod non est pars alicuius per se unius,
nec est in aliquo tamquam in subiecto. Et isto modo substantiae sepa-
ratae et substantiae compositae perfectae per se subsistunt.
Tertio modo dicitur per se subsistere omne illud quod non est in
aliquo tamquam accidens in subiecto, sive sit pars alicuius per se unius
sive non. Et isto modo non solum substantia composita perfecta per
se subsistet, sed etiam quaelibet cuius pars. Et voco per se unum in
proposito omne illud quod est unum indivisibiliter, sicut deus et angeli;
vel essentialiter sicut ignis et aer et cetera composita substantialia ex
eorum partibus essentialibus; vel continuatione, sicut lignum, lapis, et
cetera huiusmodi, cuius partes ad invicem continuetur.
<Conclusio prima> Prima conclusio est ista, quod probabiliter pos-
set susteniri physice loquendo, nullam penitus substantiam esse in istis
inferioribus, |56rb| accipiendo substantiam pro composito ex materia
et forma, vel pro aliqua parte talis compositi. Ista declaratur: omne
illud potest probabiliter teneri cuius oppositum non potest evidenter
probari; sed illud est huiusmodi; igitur etc. Maior videtur evidens. Et
minor declaratur, quia omnes apparentiae possunt evidenter salvari,
non posita aliqua tali substantia; igitur non potest evidenter probari
aliquam talem substantiam esse in istis inferioribus. Consequentia est
7
Cp. Priscianus Institutiones grammaticae II c. 3 n. 18, ed. Hertz (Leipzig, 1961), (GL
II) I 55: Proprium est nominis substantiam et qualitatem signicare.
16
17
18
19
20
21
362 robert andrews
satis evidens. Antecedens adhuc declaratur, quia positis tantummodo
accidentibus sibi invicem subsistentibus et adhaerentibus, salvatur gene-
ratio et corruptio, augmentatio et diminutio, alteratio et loci mutatio,
sicut patet in hostia consecrata.
<Obiectio> Sed contra istam conclusionem instatur: si non esset aliqua
substantia in istis generabilibus et corruptibilibus, tunc non esset genera-
tio; quia generatio est motus sive mutatio ad formam substantialem.
8
Ad illud breviter dicitur quod ponentes nullam substantiam in
istis inferioribus esse, negarent talem descriptionem generationis. Sed
dicerent quod generatio est perfecta productio alicuius rei in esse, et
corruptio perfecta destructio.
<Obiectio> Et si quaeratur tunc, quid est subiectum in tali genera-
tione?
Dicitur quod unum accidens est subiectum respectu generationis et
corruptionis alterius accidentis, sicut oportet necessario theologos dicere
quando Verbum generatur ex hostia consecrata, vel quando alicuius
color generatur, utpote pallor vel rubedo in hostia consecrata.
<Obiectiones> Et si dicatur quod tunc [non] posset assignari subiectum
in generatione, contra:
Pura aqua generatur ex puro igne; quia secundum istam positionem,
pura aqua nihil aliud est nisi compositum ex duabus qualitatibus primis,
utpote ex frigiditate et humiditate; et similiter purus ignis non est nisi
compositum ex caliditate et siccitate. Et in tali generatione, totus ignis
secundum utramque suam qualitatem corrumpitur, (corrumppitur var.
ms.) et aqua secundum utramque suam qualitatem generatur. Et per
consequens nihil est ibi quod manet in utroque termino generationis;
igitur nihil est ibi subiectum.
Item, sequitur quod talis generatio esset creatio, quia ex nihilo; quod
est contra principium omnium philosophorum, qui, secundum Phi-
losophum primo Physicorum,
9
concordant quod ex nihilo nihil t.
8
Cf. Aristoteles Phys. V,1 225a127; Hamesse , Auctoritates Aristotelis, p. 152 (152):
Generatio est mutatio de non esse ad esse.
9
Aristoteles Physica I,4 187a279; Hamesse , Auctoritates Aristotelis, p. 141 (16): Secun-
dum omnes philosophos, ex nihilo nihil t.
22
23
24
25
26
27
28
Ad primum istorum dicitur dupliciter. Uno modo quod talis generatio
est impossibilis, quia dicunt aliqui quod mediata elementa non possunt
ex se mutuo immediate generari, sed tantummodo mediate; utpote si
ignis debeat generari ex aqua, generatur mediante aere vel terra. Et
secundum istos tunc satis faciliter dicitur quod qualitas symbola
10
est
subiectum talis generationis.
Et secundum istam viam, secundum argumentum non habet locum.
Aliter vero dicitur, et satis probabiliter, quod in tali generatione vel
corruptione non requiritur aliquod subiectum aliud a re generanda vel
corrumpenda. Sed generaliter quando generatum approximatur suo
generanti quod est maioris virtutis, ex suo generato generatur sibi simile;
non tamquam ex |56va| subiecto, sed tamquam ex termino.
Et ulterius dicitur ad secundum argumentum negando consequen-
tiam, quod talis generatio esset creatio, quia creatio nihil praesupponit
ex parti passi, scilicet nec subiectum nec terminum a quo; sed omnis
generatio praesupponit aliquod passum tamquam terminum a quo,
utpote generatum, in quo tale generans(?) agat.
<Conclusio secunda> Secunda conclusio est ista, quod probabiliter
posset sustineri, nullam substantiam esse in rerum natura. Et probatur
ista conclusio sicut praecedens, quia omnes apparentiae aeque evidenter
possunt salvari, si nulla substantia ponatur, (salvatur ms.) sicut si aliqua
ponatur. Igitur probabiliter posset sustineri nullam substantiam esse.
Consequentia videtur evidens. Antecedens declaratur, quia nullum
inconveniens, ut videtur, sequitur si ponatur corpora supracaelestia esse
composita ex diversis accidentibus perpetuue (existentibus perpetuum
ms.) sibi invicem adhaerentibus, etsi ponatur cuiuslibet orbis motorem
esse unum accidens adhaerens orbi, et ipsum movenssicut gravitas
adhaeret lapidi, faciens ipsum descendere. Nec ad hoc requiritur aliqua
substantia; quia ita naturaliter descenditur hostia consecrata, in qua
non ponitur substantia, sicut hostia non consecrata.
<Obiectiones> Et si dicatur quod omnia non ita uniformiter con-
cernent(?) unas(?) suas modi generandi <et> corrumpendi nisi guber-
10
Cf. Aristoteles De Gen. et cor. II,4 331a245: Quaecumque quidem enim habent
symbolum ad invicem, velox horum transmutatio, quaecumque autem non habent,
tarda . . . Cf. Thomas Maulevelot, Quaestiones super Praedicamenta, q. 9, conclusio prima.
29
30
31
32
33
34
364 robert andrews
netur ab aliquo uno principio, quod quidem principium vocatur deus,
sicut deducit Boethius tertio De consolatione,
11
prosa ultima
Item, Aristoteles quinto et octavo Physicorum
12
deducit ratione
naturali unum primum principium esse, quia aliter esset processus in
innitum in mobilibus et motoribus
Ad primum igitur istorum dicitur quod, si cui placeret, istam opi-
nionem tenens (tenere ms.) faciliter diceret regnum totius universi
uniformiter et naturaliter eri ab accidentibus, sicut aeque uniformiter
caliditas sine igne substantiali califaceret sicut cum igne substantiali.
Ad secundum dicitur quod tenens istam opinionem diceret argumen-
tum Aristotelis nullo modo cogere. Primo quia non esset inconveniens
in mobilibus et motoribus in innitum procedere, sicut patet in motione
ligni, quod si percutiatur in una extremitate, movetur reliqua extremitas,
et non nisi quia una pars movet aliam, et ita adhuc aliam, et sic in
innitam. Sicut etiam patet in causationibus rerum quarum non fuit
prima causa, quia quamcumque causationem aliqua alia (prima(?) ms.)
causa praecessit, secundum ponentes
13
mundum ab aeternum fuisse.
Secundo, supposito quod impossibile est in mobilibus et motoribus
in innitum procedere, adhuc non videtur argumentum Aristotelis con-
cludere unum esse primum immobile, quia ipsum posset moveri ab illo
quod ipsum movet; quia non videtur potius inconveniens ponere aliqua
duo sibi invicem moventia et mobilia quam ponere aliqua duo sibi
invicem esse causas, quod ponitur ab Aristotele et eius Commentatore
secundo Physicorum, sicut homo existens in navi est causa motus navi,
et navis est causa motus hominis; et sic se mutuo movent.
<Obiectio> Et |56vb| si dicatur quod omnem talem motorem per-
petuum esse innitae virtutis, quia movit perpetue
Dico breviter quod nulla est evasio(?) ad ponendum, quia anima poni-
tur esse perpetua, et tamen non ponitur esse innitae virtutis intensive.
Et pari ratione non oportet ponere motorem caeli esse innitae virtutis
intensive (intencive var. ms.) propter motum caeli perpetuum praecise,
11
Boethius De consolatione philosophiae III (1973): quibus, ait illa, gubernaculis mundus
regatur.memini, inquam, me inscitiam meam fuisse confessum, sed quid afferas, licet
iam prospiciam, planius tamen ex te audire desidero.mundum, inquit, hunc deo regi
paulo ante minime dubitandum putabas.
12
Cf. Aristoteles Physica VII,1 242a1520; Hamesse , Auctoritates Aristotelis, p. 155
(184): In moventibus et motis non est ire in innitum, et ergo necesse est devenire ad
primum motorem.
13
E.g., ipsemet Aristoteles; cf. Aristoteles De Gen. et cor. II,10 336a15336b9.
35
36
37
38
39
40
cum motus motu nulla habet resistentia. Et ideo, cum (non ms.) posset
imaginari ita parva potentia activa in corruptibilibus quae non posset
perpetue caelum movere, quam perpetuis, ex tali motione nullo modo
concluditur innitas intensiva in aliquo motore.
<Obiectio> Et si dicatur quod tunc nihil esset ita perfectum quin
perfectius posset imaginari
Huic dicitur concedendo consequens, saltem ad istum intellectum
quod non est aliquid ita perfectum quin posset haberi actus qui esset
imaginatio perfectioris, si illud perfectius esset.
<Conclusio tertia> Tertia conclusio est ista, quod probabiliter posset
susteniri tamquam illud quod ab omnibus communiter ponitur, quod est
una prima substantia quae a nullo dependet; quae scilicet est ipsemet
deus. Et quod sunt aliae substantiae separatae individuales et incor-
ruptibiles, quae quidem dependeant in esse a prima substantia, sunt
tamen per se subsistentes, sic videlicet quod non sunt accidentia nec
partes essentiales alicuius per se unius. Et quod sunt aliae substantiae
corporales incorruptibiles, utpote corpora supercaelestia, quae sim-
pliciter dependeant a prima substantia; sunt per se tamen subsistentia
modo iam dicto. Et quod sunt aliae substantiae compositae ex materia
et forma, et non ex aliis partibus et homogeneiis; sunt tamen corrupti-
biles, utpote quattuor elementa. Et quod sunt adhuc aliae substantiae
corporales, non tamen compositae ex materia prima et forma, videlicet
ex aliis partibus homogeniis, utpote \ex/ elementis; et istae substan-
tiae vocentur substantiae mixtae sive elementae. Et quod sunt aliae,
scilicet partes istarum substantiarum corruptibilium, utpote materia
et forma, quae vocantur partes essentiales, quarum quidem partium
quaedam sunt incorruptibiles, sicut materiaquae ponitur tamquam
fundamentum, et subiectum generationis et corruptioniset anima
intellectiva; et quaedam vero ponuntur corruptibiles, sicut aliae formae
substantiales. Et ista conclusio, quia ab omnibus ponitur, amplius ad
praesens non declaratur.
<Conclusio quarta> Quarta conclusio est ista, quod accipiendo sub-
stantiam secundo modo, scilicet prout est nomen secundae intentionis,
multae sunt substantiae, utpote species et genera absolute substantias
signicantia, sicut in sequentibus clarius apparebit.
41
42
43
44
366 robert andrews
1. Ad primum igitur principale dicitur, negando antecedens. Et ulte-
rius dicitur quod quamvis omne quod existit cum alio existit, ita quod
ipsummet non tantum existit, cum hoc tamen stat quod aliquod per
se subsistat., quod requiritur ad hoc quod aliquid sit substantia, sicut
declaratum in secunda |57ra| distinctione.
2. Ad secundum dicitur quod aliqua est substantia divisibilis, et
aliqua indivisibilis, sicut patet per primam conclusionem. Et ulterius
dicitur, negando istam: omnis pars est relatio, nisi subiectum sup-
ponat (subponat var. ms.) materialiter vel simpliciter; quia non pono ad
praesens relationem aliud esse nisi signum quod ad aliud refertur sub
habitudine alicuius casus virtute suae signicationis.
Et ulterius dicitur, negando quod omnis relatio sit accidens, acci-
piendo accidens realiter et non praedicatione; quia multa sunt signa
relativa scripta quae non sunt accidentia.
Et ulterius conceditur quod omnis pars, saltem divisibilis, est quanti-
tas continua vel discreta; praecipue in istis corruptibilibus. Propter hoc
tamen non sequitur quod sit accidens, quia teneo ad praesens quod
non omnis quantitas sit accidens.
Et ulterius dicitur quod, quamvis non experiamur aliquod esse indi-
visibile, nec etiam habemus rationem ita necessario cogentem quin
posset evadi; habemus tamen omnes evidentias ponendi tales substantias
indivisibiles, utpote communem opinionem philosophorum tales sub-
stantias ponentium, quae satis est evidentia sufciens, ubi nihil fortius
absistit, quia non est dicendum contra opinionem omnium et maxime
sapientium nisi propter rationem cogentem.
3. Ad tertium argumentum dicitur, negando consequentiam; quia
multa ponimus esse quae non experimur esse; nec etiam quae possumus
aliqua ratione convincenti (convicente var. ms.) probare; quia sufcit sicut
praedictum aliqua evidentia probabilis.
Et ulterius dicitur quod, sicut Magister vult in quarto Sententiarum,
non est articulus dei credere quod in hostia consecrata sint accidentia
sine subiecto; sed sufcit credere quod in illa hostia est unum corpus
Christi, sive sit sine substantia panis sive cum substantia panis. Et
ideo ponendo ibi substantiam panis esse nulla difcultas emergit ad
salvandum apparentias. Si vero ponatur substantiam ibi non esse, hoc
t miraculose, et per consequens omnia quae sequuntur quodammodo
sunt miraculosa.
4. Ad quartum dicitur quod, quamvis omnes apparentiae possunt
salvari non ponenda aliquam substantiam, praecipue in istis inferiori-
bus, non tamen aeque decenter; et aeque placeret communi opinioni
45
46
47
48
49
50
51
52
philosophorum et etiam omnium aliorum; et ideo potius ponanda est
substantia quam non ponenda.
5. Ad quintum dicitur, concedendo consequentiam. Et ulterius conce-
ditur quod deus est illud quo maius excogitatur non potest. Et ulterius
dicitur, concedendo quod cogitari possit duas primas substantias esse.
Et ulterius negatur consequentia secunda, duplum ad unam istarum
possit excogitari esse; quia prima causa est innita, et nullum innitum
habet duplum.
<Obiectio> Et si dicatur quod omnis dualitas est dupla ad suam uni-
tatem, quia dualitas est numerus par, et potest dividi in duas medietates
aequales
Ad illud breviter dicitur, negando |57rb| istam: omnis dualitas est
dupla ad suam unitatem, quia in dualitate innita est illa propositio
falsa.
Et ulterius dicitur quod ista propositio omnis numerus par potest
dividi in duas medietates aequales; intelligitur de numero praeinnito
et de divisione nominali, non reali.
Ad istum intellectum quod iste terminus unitas vel unum qui sig-
nicat vel est unum membrum divisionis, non est expressius ex virtute
signicationis maioris quantitatis quam iste terminus unum vel uni-
tas qui est aliud membrum divisionis; quia clarum est quod realiter
quaedam dualitas composita ex maxime monte et musca non habet
unitates aequales discretas; sed, quantum ad propositum est, sufcit
quod nulla dualitas inferiorum est dupla ad suam unitatem. Et ideo
dualitas patris et lii in divinis non est dupla ad patrem, quamvis iste
terminus dualitas vel iste terminus duo sit duplae denominationis ad
istum terminum unum vel unus.
6. Ad sextum dicitur, concedendo quod materia prima est ponenda,
et quod ista est incorruptibilis; sed non immutabilis.
Et ulterius dicitur quod ipsa mutatur in sua essentia intrinseca per
rarefactionem et condensationem, sicut per formarum substantialium
receptionem in sua essentia; quod non est in deo, quia quamvis deus
habeat extrinsece aliquid nunc quam prius, non tamen intrinsece. Et
ideo non mutatur.
Et ulterius conceditur quod materia sit innitae potentiae receptivae,
et passivae; sed hoc non arguit perfectionem, sed potius inperfectionem.
Et ideo negaretur ista consequentia est innitae potentiae receptive
et passivae, igitur est innita. Sed deus est innitae potentiae activae;
53
54
55
56
57
58
59
60
61
368 robert andrews
et hoc arguit perfectionem innitam. Et ideo similitudo inter deum et
materiam primum non sufcit ut materia sit deus.
Ad argumentum in oppositum, posset responderi, si cui placeret, per
superius dicta.
62
CATEGORIES AND UNIVERSALS IN THE LATER
MIDDLE AGES
Alessandro D. Conti
In the (later) Middle Ages categories and universals were closely linked
topics. Categorial doctrine concerned (1) the existence, inner natures,
and the mutual relationships of the basic, metaphysical items of the
world (individual and universal substances, individual and universal
accidents), and (2) the connections of such basic items to language.
Late medieval theories of universals
1
dealt with the problems of
(1) real existence of universals (or common natures), both substantial
and accidental ones, and (2) the relationship between them and the
(perceptible or otherwise intelligible) individuals. Hence, in one way, later
medieval theories of universals investigated more thoroughly some of
the many related questions which categorial doctrines went into. This is
not surprising, since (1) textually, medieval discussions on the problem
of universals derived from a well-known passage of Porphyry s Isagoge, a
work which was intended to be an introduction to Aristotles Categories,
and (2) philosophically, the medieval problem of universals is one of
the various aspects of the problem of meaning, which in its turn is one
of the two main subjects of any later medieval categorial doctrine. In
Isagoge 1,1316 Porphyry raises his famous series of questions, about
the ontological status of universals and their relation to individuals,
which medieval philosophers faced up to in their commentaries on the
Isagoge and treatises on universals: (1) whether genera and species exist in
themselves or are nothing but mere concepts; (2) whether, if they have
an extramental form of existence, they are corporeal or incorporeal;
(3) and whether they exist apart from perceptible objects or in and by
virtue of them. In another way, all late medieval theories of universals
respond to an implicit semantic question: is there something in re which
corresponds to the common nouns of our language in the same way
1
A comprehensive survey of the problem of universals from Antiquity to late
Middle Ages is provided in Alain de Libera , La querelle des universeaux: De Platon la n
du Moyen Age, Paris 1996.
as individuals correspond to proper names? The answer of Realists
was afrmative, negative that of Nominalists; while, within each group,
authors disagreed about the peculiar modes of being of universals and
the nature of their relation to individuals.
In what follows, I will draw the outline of the problem of univer-
sals in late Middle Ages both from a systematic and from a historical
point of view, trying to indicate the connections with the doctrine(s)
of categories. Accordingly, rst of all I will give a short account of
the standard theories of universals worked out between the thirteenth
and fourteenth centuries. Second, I will summarise Ockhams criticism
of the traditional view. Finally, in the third and fourth sections, I will
consider the main positions about universals elaborated from 1330 ca.
to 1430 ca. in some detail, and show their increasing relevance for the
categorial doctrines. From this and the concluding remarksI hopeit
shall emerge (1) why and how the debate over the status and nature
of universals evolved during the very last period of the Middle Ages,
and (2) the progressive subordination of the categorial view to the
conception of universals.
The Moderate Realist View
Since Robert Kilwardbys formulation of the problem of universals in
his commentaries on the Ars Vetus (Isagoge, Categories, Liber sex principiorum,
De interpretatione) the semantic origin of the so called moderate realist
view on universals (endorsed by authors such as Robert Kilwardby
himself, Albert the Great, Thomas Aquinas , Henry of Ghent , Simon
of Faversham, John Duns Scotus , Thomas Sutton , Giles of Rome, and
Walter Burley before 1324), is quite evident. What might be described
as the orthodox view (at least until it was challenged by Ockham and
the other Nominalists) was that universals (or common natures) are the
real signicata of general nouns, such as man, animal, and white-
ness.
2
As a consequence, moderate Realists conceived of universals
2
In his middle commentary on De interpretatione (Commentarius in librum Perihermeneias,
before 1310) in commenting on the starting lines of chapter seven (17a38b7), Walter
Burley claims that a linguistic expression is a general noun (nomen appellativum) if and
only if it signies a universal, that is an entity apt to be common to many individual
itemssee Stephen F. Brown , Walter Burley s Middle Commentary on Aristotles
Perihermeneias, Franciscan Studies 33 (1973), 42134, p. 85.
categories and universals in the later middle ages 371
as metaphysical entities, existing independently of our minds, which
are necessary conditions for our language to be signicant. Common
nouns would be meaningless if they did not signify something (1) that
exists somehow in reality, and (2) that has the peculiar feature of being
common to (namely, present in) many individual items. Moreover, they
investigated the metaphysical composition of such common natures
by the doctrine of categories, from a point of view that we can call
intensional and in a way similar to that exploited by our modern
componential analysis. Only by associating general nouns with such
entities as their proper signicatum did they think the fact could be
explained that a general noun can be used predicatively to ascribe a
given property (say, being a man or an animal) to many individuals at
the same time. According to them, a general noun stands for (supponere)
and labels (appellare) a certain set of individual items only by way of
the common nature (the universal) that (1) it directly signies, and (2) is
present in that set of individuals as their own intelligible essence. Since
common natures (1) connect general nouns up with their extensions
by determining the classes of the things to which they are correctly
applied, and (2) are what general nouns stand for when they have simple
supposition,
3
they are the intensions of common nouns; or better, the
hypostatisations of these intensions, inasmuch as they were conceived
of as entities existing independently of our minds.
This comes out quite clearly from the standard reading of Categories
5, 3b 1015, where Aristotle maintains that a primary substance signi-
es a single item (hoc aliquid according to the Latin translation) whilst
a secondary substance signies a qualifying (and therefore common or
universal) item (quale quid according to the Latin translation), notwith-
standing it seems to signify a single item. Thirteenth century authors
identied the secondary substance with the quale quid and the primary
substance with the hoc aliquid, and therefore secondary substances
(namely, the universals of the category of substance) with the signi-
cata of general nouns of that category (such as man) and primary
substances (namely, the individuals of the category of substance) with
the signicata of individual expressions of that category (such as this
man, which refers to a single human individual only). Furthermore,
3
Cf. e.g., Walter Burley , De suppositionibus (A.D. 1302), in Stephen F. Brown , Walter
Burleighs Treatise De suppositionibus and its Inuence on William of Ockham, Franciscan
Studies 32 (1972), 1564, pp. 3536.
they assumed that common nouns of the category of substance, when
used predicatively, specify which kind of substance a certain individual
substance is.
4
As a consequence, they thought of universals and indi-
viduals as linked together by a sort of relation of instantiation. In other
words, they conceived of primary substances as the tokens of secondary
substances (and, more in general, individuals as tokens of universals),
and secondary substances as the types of primary substances (and,
more in general, universals as types of individuals). In fact, according
to them, (1) individual substances are unique physical entities, located at
a particular place in space and time, and universal substances are their
specic or generic formsthat is, their intelligible natures, immanent
in them, having no independent existence, and apt to be common to
many different individuals at the same time. (2) Any individual substance
can be recognized as a member of a certain natural species by virtue
of its conformity to the universal substance that it instantiates, and by
virtue of its likeness to other individual substances.
From what has been said it is manifest that the crucial question of
the medieval realist approach to the problem of universals was not
that of the ontological status of the universals (as it was for Boethius
and the other Neoplatonic commentators of Aristotle); it was that of
their relation with the individuals. Since moderate Realists, agreeing
with Aristotle (Categories 5, 2a352b6), maintained that, if primary sub-
stances did not exist, it would be impossible for anything else to exist,
as everything else depends on them for its own being,
5
the question
of the status of universals necessarily became the question of their
relation to individual substances. In fact, according to the moderate
4
Cf. Robert Kilwardby, Notulae super librum Praedicamentorum, lectio 7 (Cambridge,
Peterhouse), ms. 206, fol. 47rab; Albert the Great, Liber de praedicamentis, tr. 2, cap. 8,
in Opera omnia, ed. A. Borgnet (Paris, 1890), vol. 1, pp. 181183; Thomas Sutton ,
Expositio super librum Praedicamentorum, cap. de substantia (Oxford, Merton College), ms.
289, fol. 11rbva (transcription in Alessandro D. Conti , Thomas Sutton s Commen-
tary on the Categories according to the Ms Oxford, Merton College 289, in The Rise
of British Logic, ed. P. O. Lewry (Toronto, 1985), pp. 173213, pp. 203204); Walter
Burley , Tractatus super librum Praedicamentorum, cap. de substantia (Cambridge, Peterhouse),
ms. 184, fol. 178rab.
5
Cf. Kilwardby, Notulae super librum Porphyrii, lectio 2 (Cambridge, Peterhouse), ms.
206, fol. 34vb; Albert the Great, Liber de preadicamentis, tr. 2, cap. 4, pp. 172174; Simon
of Faversham, Quaestiones super librum Praedicamentorum, q. 7, in Opera omnia, vol. 1, ed.
Pasquale Mazzarella (Padua, 1957), p. 77; Thomas Sutton , Expositio super librum Praedi-
camentorum, cap. de substantia, fol. 8rb; John Duns Scotus , Quaestiones super Praedicamenta,
q. 13, in Opera philosophica, vol. 1, pp. 369372 and 377; Burley , Tractatus super librum
Praedicamentorum, cap. de substantia, fol. 177va.
realist view, universals are not self-subsistent entities, but exist only in
individual items, as universals have no being (esse) outside the being
of their instantiations. As a result, moderate Realists thought that
universals could be said to be everlasting because of the succession
of their individuals, not because of a peculiar kind of being of their
own.
6
But whereas according to the most common opinion universals
existed in potentia outside the mind and in actu within the mind, on Duns
Scotus s and Burley s accounts they exist in actu outside the mind, since
their being is exactly the same as the being of individuals, which is
actual. Indeed, for Duns Scotus and Burley the necessary and sufcient
condition for a universal to be in actu is the existence of at least one
individual instantiating it. Therefore our mind does not give actuality
to universals, but a separate mode of existence only.
Like the Neoplatonic commentators of Aristotle, such as Ammonius
and Simplicius , medieval moderate Realists also argued that univer-
sals are threefold: (1) ante rem, or ideal universals, that is the ideas in
God, archetypes of all that there is; (2) in re, or formal universals, that
is the common natures shared by individual things; and (3) post rem,
or intentional universals, that is mental signs by which we refer to
the universals in re. On the other hand, differing from Neoplatonists ,
they were convinced that common natures really have the property
of being universal by themselves. More precisely, like Avicenna , they
believed that, properly speaking, common natures qua such are prior,
and so indifferent, to any division into universals and individuals.
However, universality is as it were their inseparable characteristic.
7
As
a consequence they thought that three different kinds of entities can
be qualied as real universals (universalia in re): (1) the common natures
instantiated by individualswhich are res of rst intention; (2) the form
itself of universality which belongs to a certain common nature when
seen in its relation to the individualswhich is a second intention,
such as being-a-genus, being-a-species; (3) the entities which are made
6
Cf. e.g., Simon of Faversham, Quaestiones super libro Porphyrii, q. 5, p. 27; Burley ,
Tractatus super librum Praedicamentorum, cap. de substantia, fol. 177va.
7
Cf. Kilwardby, Notulae super librum Porphyrii, lectio 2, fol. 34va; Albert the Great,
De quinque universalibus, tr. de universalibus in communi, capp. 3 and 5, ed. Col., vol. 1.1A,
pp. 2425 and 3132; Simon of Faversham, Quaestiones super libro Porphyrii, qq. 4 and 5,
in Opera omnia, vol. 1, pp. 2324 and 2627; Giles of Rome, Super librum I Sententiarum,
d. 19, p. 2, q. 1, ed. (Ve netiis, 1521), fol. 110va; Duns Scotus , Quaestiones in librum
Porphyrii Isagoge, q. 11, in Opera philosophica, vol. 1, pp. 5051; Lectura II, d. 3, p. 1,
q. 1, ed. Vaticana, vol. 18, p. 237.
up by the union of a common nature with one of these properties of
second intention. Moreover, they conceived of formal universals in two
different manners: as rst intentions or as second intentions. In the rst
case, they considered universals as natures of a certain kind, identical
with their own individuals (for example, man would be the same thing
as Socrates ). In the second case, formal universals were regarded as
properly universals and distinct from their own individuals, considered
qua individuals, because of the opposite constitutive principles: commu-
nicabilitas for universals and incommunicabilitas for individuals.
8
Hence,
moderate Realists thought of universals as formal causes in relation to
their own individuals, and individuals as material causes in relation to
their universals. Furthermore, agreeing with what Aristotle states in the
third (1b1015) and fth (2b23) chapters of the Categories, they main-
tained that (1) a universal could directly receive only the predications
of those forms more common than itself (i.e., those forms which are
put on a higher level in the linea praedicamentalis); and (2) the accidental
forms inhering in substantial individuals could be predicated of the
substantial form itself that those individuals instantiate only indirectly,
through and in virtue of the individuals of that substantial form.
9
Thus,
their position on the question of the relationship between universals
and individuals necessarily entailed a soft attitude towards the problem
of dening and classifying the types of identity and distinction (or dif-
ference), since it is evident that universals had to be considered at the
same time not totally identicalwith and not totally differentfrom
their own individuals.
Indeed, at the end of the thirteenth century two main attempts
were made to revise the common notions of identity and distinction
8
Cf. Albert the Great, De quinque universalibus, tr. de universalibus in communi, cap. 3,
p. 26, cap. 5, pp. 3233; Simon of Faversham, Quaestiones super libro Porphyrii, q. 4, pp.
23 and 25; Duns Scotus , Quaestiones in librum Porphyrii Isagoge, q. 3, pp. 1920.
9
For instance, if Socrates is white, then man (homo) is white, but the form of white-
ness cannot be directly attributed-to (or predicated-of ) the form of humanity itself:
humanity is not whiteness, nor white. Cf. Kilwardby, Notulae super librum Praedicamentorum,
lectio 4, fol. 44va; Albert the Great, Liber de praedicamentis, tr. 1, cap. 6, pp. 161162;
tr. 2, cap. 4, pp. 172173; Thomas Sutton , Expositio super librum Praedicamentorum, cap.
de ordine praedicati ad subiectum, fol. 6ra; cap. de substantia, fol. 7vb (transcription in Conti ,
Thomas Sutton s Commentary on the Categories, pp. 195196, and 197); Simon of
Faversham, Quaestiones super librum Praedicamentorum, q. 3, p. 76; q. 21, pp. 9596; Duns
Scotus , Quaestiones super Praedicamenta, q. 9, pp. 327332; q. 13, pp. 365377; Burley ,
Tractatus super librum Praedicamentorum, cap. de regulis praedicationis, fol. 174va; cap. de
substantia, fol. 177va.
by introducing a third kind of difference in between the real (realis) and
the notional (secundum rationem) ones, but neither of them was speci-
cally intended to offer an answer to the problem of the relationship
between universals and individuals. Henry of Ghent spoke of intentional
(secundum intentionem) difference,
10
which he characterised in the follow-
ing way: two entities intentionally differ one from the other if and only
if (1) both of them are constitutive elements of the same thing, but
(2) the denition of neither of them is part of the denition of the
other, so that (2.1) each of them can be understood even together with
the negation of the other. Duns Scotus spoke of formal distinction.
11

He gave two different denitions of it. In the Lectura and in the Ordinatio
he described it as a symmetrical relation between two entities which
cannot exist separately: two entities are formally distinct one from the
other if and only if (1) both of them are constitutive elements of the
same reality, but (2) neither of them can exist by itself, (3) nor is one
part of the denite description of the other.
12
In the Reportata Parisiensia
he dened it as an asymmetrical relation between a whole reality and
one of its constitutive elements: an entity x is not formally identical
with another entity y if and only if (1) y is not part of the denite
description of x, but (2) x and y are one and the same thing in reality.
13

He utilised these two rather different notions of formal distinction in
order to illustrate respectively (1) how the genus and the specic differ-
ence, and the specic nature and the individual difference are linked
to gether, and (2) the relations which hold between the divine nature
10
Cf. Henry of Ghent , Quodlibet X, q. 7, ed. R. Macken (Leiden, 1981), pp. 164166.
On Henry of Ghents doctrine of intentional distinction see John F. Wippel , God-
frey of Fontaines and Henry of Ghents Theory of Intentional Distinction Between
Essence and Existence, in Sapientiae procerum amore, ed. T. W. Koehler (Rome, 1974),
pp. 289321.
11
On Scotus s theory of formal distinction see Marilyn McCord Adams , Ockham
on Identity and Distinction, Franciscan Studies 36 (1976), 574, pp. 2543; Peter King ,
Duns Scotus on the Common Nature and Individual Difference, Philosophical Topics 20
(1992), 5176; Stephen D. Dumont , Duns Scotuss Parisian Question on the Formal
Distinction, Vivarium 43 (2005), 762. On further developments of Scotuss theory
in Oxford at the end of the fourteenth century see Alessandro D. Conti , Sviluppi e
applicazioni della distinzione formale scotista ad Oxford sul nire del XIV secolo, in
Via Scoti, Methodologica ad mentem Joannis Duns Scoti, ed. Leonardo Sileo, 2 vols. (Rome,
1995), vol. 1, pp. 319336.
12
Cf. Duns Scotus , Lectura I, d. 2, p. 2, qq. 14, ed. Vaticana, vol. 16, p. 216; Ordi-
natio I, d. 2, p. 2, qq. 14, ed. Vaticana, vol. 2, pp. 356357; II, d. 3, p. 1, q. 6, ed.
Vaticana, vol. 7, pp. 483484.
13
Cf. Duns Scotus , Reportata Parisiensia I, d. 33, qq. 23, and d. 34, q. 1, ed. Vivs,
vol. 22, pp. 402408, and 410.
and its three Persons, and between the human soul and its faculties.
On the other side, in his Quaestiones in librum Perihermeneias (A.D. 1301),
Walter Burley used the intentional difference for clarifying the relations
between genus and difference, and between essence and being.
14
He
claimed that there was not a real distinction between essence and being
(as Aquinas and Giles of Rome had taught), but they were really the
same and only intentionally distinct.
15
By means of these new kinds of
distinctions moderate Realists were trying to explain how it is possible
to distinguish many different real aspects internal to the same individual
thing, without breaking its unity.
In conclusion, the legacy to the late Middle Ages left by the debate
on universals which took place in between the thirteenth and the
fourteenth centuries can be summed up as follows: (1) universals exist
in a threefold way, as ideas in the mind of God, as common natures
in re, and as general concepts in our mind. (2) Real universals are forms
naturally apt to be present in many individual items as their main
metaphysical components. (3) Real universals have no being outside
the being of their individuals. (4) Real universals are partially identi-
cal-with and partially different-from their own individuals. (5.1) Real
universals exist in potentia only outside our minds, or, (5.2) according to
Duns Scotus s and Burley s views, they exist in actu outside our minds.
16

As far as the logical machinery was concerned, besides the real and
notional distinctions there were three other types of distinction which
could be employed in order to account suitably for the inner composi-
tion of beings: Henry of Ghent s intentional distinction, and the two
kinds of formal distinction drawn up by Duns Scotus.
Ockhams Critique
As is well known, in the rst decades of the fourteenth century, in his
commentary on the rst book of the Sentences, in his Expositio aurea, and in
14
See Alessandro D. Conti , Essenza ed essere nel pensiero della tarda scolastica
(Burley , Wyclif , Paolo Veneto), Medioevo 15 (1989), 235267.
15
Cf. Burley , Quaestiones in librum Perihermeneias, q. 4, in Stephen F. Brown , Wal-
ter Burley s Quaestiones in librum Perihermeneias, Franciscan Studies 34 (1974), 200295,
p. 273.
16
Cf. e.g., Burley , Expositio libri De Anima (1316 ca.) lib. I, q. 3: utrum universale
habeat esse extra animam, Civitas Vaticana, Bibl. Apostolica Vaticana, ms. Vat. Lat.
2151, fol. 10ravb.
the rst part of his Summa logicae, Ockham contended that (T1) the pre-
supposition of a relation of identity between universals and individuals
was inconsistent with the standard denition of real identity, according
to which two things a and b are identical if and only if, for all P, it is
the case that P is predicated of a if and only if it is predicated of b;
and (T2), from an ontological point of view, the only kind of distinc-
tion which could hold between two creatures was the real one, as (in
his opinion) any form of distinction between two creatures necessarily
implies a real distinction between them. From these two theses and the
assumption of the absolute truth of the Aristotelian claim that (T3)
there cannot be universal forms apart from their individuals, he derived
(T4) a rejection of any type of extramental existence for universals.
17

His nal argumentation, expressed in general terms, was that, if univer-
sals are something existing in re, really identical with their individuals,
when considered as rst intentions, then whatever is predicated of the
individuals must be predicated of their universals too, and so a unique
universal entity (say, the human nature) would possess contrary attributes
simultaneously (because of the different accidental forms inhering in
the various individuals really identical with that common nature at a
given time). A conclusion clearly unacceptable.
18
The crucial point of Ockhams attack on the traditional realist view
on universals is the demonstration of the thesis (T2), since (T3) was a
sort of undisputed dogma in his times, and (T1) was recongnised to
be somehow true by moderate Realists too. As a matter of fact, they
had tried to avoid that internal contradiction by introducing some
form of distinction between universals and individuals considered as
second intentionsas we have already seen. It was a common topic
in the explanation of Categories 3, 1b1015, that one cannot infer from
Socrates is a man and man is a species that Socrates is a species,
notwithstanding the identity between homo and Socrates.
19
On the other
17
On Ockhams theory of universals see Marilyn McCord Adams , William Ockham
(Indiana, 1987), 2 vols, vol. 1, pp. 369; and Alessandro D. Conti , Studio storico-
critico, in Johannes Sharpe, Quaestio super universalia, ed. Alessandro D. Conti (Florence,
1990), pp. 257294.
18
Cf. Ockham, Expositio in librum Praedicamentorum Aristotelis, cap. 8.1, in Opera philo-
sophica, vol. 2, pp. 164168; Summa logicae, pars I, cap. 15, in Opera philosophica, vol. 1,
pp. 5051.
19
Cf. Kilwardby, Notulae super librum Praedicamentorum, lectio 4, fol. 44va; Albert the
Great, Liber de praedicamentis, tr. 1, cap. 6, pp. 161162; Thomas Sutton , Expositio super
librum Praedicamentorum, prologus, fol. 2rb (transcription in Conti , Thomas Sutton s
Commentary on the Categories, p. 187); Simon of Faversham, Quaestiones super librum
hand, in their opinion, the thesis of the identity of universals and indi-
viduals was necessary for safeguarding (1) the division of predication into
essential and accidental, and (2) the difference between substantial and
accidental forms, stated by Aristotle in the second (1a201b5) and fth
(2a1833, 3b1721) chapters of the Categories.
20
Consequently, moderate
Realists had been forced to speak of identity between universals and
individuals, but to weaken that same identity by limiting the transitivity
of predications, since, according to them, not all that was predicated
of individuals could be predicated of universals and vice versa. Besides
differentiating common natures conceived of as rst intentions and as
second intentions, a further device moderate Realists had made use of
was to distinguish between two points of view from which one could
consider universal forms: (1) in abstracto (intensionally) and (2) in concreto
(extensionally). (1) Intensionally regarded a common nature was nothing
but the sum of essential properties which constituted a categorial item,
without any reference to the existence of individuals which, if that was
the case, instantiated it. (2) Extensionally regarded a common nature
was that same form conceived of as actually instantiated by at least
one individual. For instance, the human nature intensionally considered
was humanitas, extensionally considered was homo. Both terms humanitas
and homo referred to the same nature, but respectively viewed (1) as
simply a form, existentially incomplete and dependent, and (2) as a real
type, concretely instantiated by at least one individual. While it was not
possible to ascribe a property of one of its individual to the universal
form regarded in abstracto (obviously, humanity is neither white nor
running nor sick), it was possible to attribute it to the universal form
regarded in concreto (man is white and black and running etc.) without
any contradiction.
21
Praedicamentorum, q. 3, p. 76; Duns Scotus , Quaestiones super Praedicamenta, q. 9, pp.
327332; Burley , Tractatus super librum Praedicamentorum, cap. de regulis praedicationis, fol.
174va.
20
Cf. Kilwardby, Notulae super librum Praedicamentorum, lectio 3, fol. 44rab; lectio 6,
fol. 45rbva; lectio 7, fol. 47rb; Thomas Sutton , Expositio super librum Praedicamentorum,
cap. de subiecto et praedicato, fol. 5rbvb; cap. de substantia, fol. 11ra (transcription in Conti ,
Thomas Sutton s Commentary on the Categories, pp. 194195 and 201202); Burley ,
Tractatus super librum Praedicamentorum, cap. de complexo et incomplexo, fols. 173vb174ra;
cap. de substantia, fol. 177va.
21
Cf. Kilwardby, Notulae super librum Porphyrii, lectio 5, fol. 37ra; Notulae super librum
Praedicamentorum, lectio 1, fol. 43vab; lectio 8, fol. 49vb; Albert the Great, De quinque
universalibus, tr. de universalibus in communi, cap. 1, pp. 1719, and cap. 8, pp. 3738;
According to Ockham, (1) resorting to those expedients, and (2) posit-
ing a third kind of distinction in order to solve the problem of clarifying
the relation holding between universals and individuals amounted simply
to looking for ad hoc solutions. In his opinion, there was no room for any
further distinction in addition to the real one (traditionally viewed), since
any other possible kind of distinction necessarily implied identity, and
identity was a transitive, symmetrical, and reexive relation.
22
Moreover,
he apparently subscribed to both the Identity of Indiscernibles (for all
x and y, if for all P, P is predicated of x if and only if it is predicated
of y, then x is identical with y) and the Indiscernibility of Identicals (for
all x, y, and P, if x is identical with y, then P is predicated of x if and
only if it is predicated of y) laws.
23
As a consequence, he concluded
that it was impossible that contradictory properties (such as communica-
bilitas and incommunicabilitas) were truly asserted of the same res unless
the bearers of those contradictory properties were really distinct and
independent beings.
24
But given (T3), universals could not in any way
be real things (res); they necessarily were mental entities (entia rationis), as
no other alternative was possible.
25
So, for Ockham, the only universal
beings it made sense to talk about were universal concepts, and derivative
from them, universal terms in spoken and written languagetaking
for granted that such universal concepts, if regarded simply as beings,
were individual like all others, as they were universal only in the sense
of being the mental signs of a multiplicity of individual things. So that
for Ockham universality consisted simply in the universality of the
representative function of mental, spoken, and written terms.
26
Burley , Tractatus de abstractis, Civitas Vaticana, Bibl. Apostolica Vaticana, ms. Vat. Lat.
2146, fol. 249rb.
22
Cf. Ockham, Ordinatio I, d. 2, q. 6, in Opera theologica, vol. 2, p. 186; d. 33, q. 1,
in Opera theologica, vol. 4, pp. 416421. On real sameness and distinction in Ockham
see Marilyn McCord Adams , Ockham on Identity and distinction, especially pp.
612 and 4450.
23
Cf. Ockham, Ordinatio I, d. 2, q. 1, pp. 14 and 16; q. 6, pp. 173174; Summa
logicae, pars I, cap. 16, p. 56.
24
Cf. Ockham, Ordinatio I, d. 2, q. 1, p. 14: Impossibile est contradictoria vericari
de quibuscumque, nisi illa, vel illa pro quibus supponunt, sint distinctae res, vel dis-
tinctare rationes sive entia rationis, vel res et ratio; and q. 11, p. 363: Contradictoria
non possunt vericari nisi propter aliquam distinctionem vel non-identitatem; see also
Summa logicae, pars I, cap. 16, pp. 5455.
25
Cf. Ockham, Ordinatio I, d. 2, q. 1, pp. 1415.
26
Cf. Ockham, Ordinatio I, d. 2, q. 4, pp. 122124; Expositio in librum Porphyrii, prooem.,
in Opera philosophica, vol. 2, pp. 11 and 1416; Summa logicae, pars I, cap. 14, pp. 4849;
and cap. 15, pp. 5354.
As is well known, over the course of his career, Ockham modied
his belief on the status of universal concepts.
27
At the very beginning,
following the ideas of Henry of Harclay and Peter Auriole, he was of
the opinion that universals were purely intentional objects ( cta), in the
sense that they did not exist in our minds as in their own subjects of
inherence (subiective), but they were the objects (namely, the semantic
contents) apprehended by our minds through the acts of understanding
(obiective). Such semantic contents would be what we grasp by means
of common nouns and compare with reality in order to establish the
truth or falsity of a sentence.
28
Eventually, he changed his mind and in
his later Quaestiones super libros Physicorum, Quodlibet IV, and Summa logicae
adopted the so called intellectio-theory. The turnabout was caused by
Walter Chattons critique, that the Venerabilis Inceptor partially accepted.
Ockham maintained that universal concepts were those singular acts of
understanding by means of which our minds think of several individuals
at oncea choice that was consistent with his theories of supposition
and of meaning.
29
Universals, identity, and predication from Ockham to Wyclif
With the noteworthy exception of Ockhams followers,
30
such as John
Buridan , Albert of Saxony, Henricus Totting de Oyta, and Marsilius of
27
See Marilyn McCord Adams , Universals in the Early Fourteenth Century, in
Norman Kretzmann et al., eds. The Cambridge History of Later Medieval Philosophy. From the
Rediscovery of Aristotle to the Disintegration of Scholasticism 11001600 (Cambridge, 1982),
pp. 411439; Adams, Ockham, pp. 71107; Robert Pasnau , Theories of Cognition in the
Later Middle Ages (Cambridge, 1997), pp. 277289; Claude Panaccio , Semantics and
Mental Language, in The Cambridge Companion to Ockham, ed. Paul V. Spade (Cam-
bridge, 1999), pp. 5375.
28
Cf. Ockham, Ordinatio I, d. 2, q. 8, pp. 271281.
29
Cf. Ockham, Summa logicae, pars I, c. 12, pp. 4243: Dicendum quod circa istum
articulum diversae sunt opiniones. Aliqui dicunt quod non est nisi quoddam ctum
per animam. Alii, quod est quaedam qualitas subiective existens in anima. Alii dicunt
quod est actus intelligendi. Et pro istis est ratio ista, quia frustra t per plura quod
potest eri per pauciora. Omnia autem quae salvantur ponendo aliquid distinctum
ab actu intelligendi possunt salvari sine tali distincto, eo quod supponere pro alio et
signicare aliud ita potest competere actui intelligendi sicut alii signo. Igitur praeter
actum intelligendi non oportet aliquid aliud ponere. See also cap. 15, p. 53. On
Ockhams semantics see Adams , Ockham, pp. 327435; Claude Panaccio , Les Mots,
les Concepts et les Choses. La smantique de Guillaume dOccam et le nominalisme daujourdhui
(Montral-Paris, 1991).
30
I do not take their theories into consideration, since they did not add any remark-
able feature to Ockhams approach and doctrine.
Inghen, later medieval authors acknowledged that Ockhams critique
showed that the traditional realist description of the relation between
universals and individuals was untenable, but they were convinced that
realism as a whole still was defensible. Therefore, they tried to remove
the aporetic points of the traditional realist theory of universals by suit-
able strategies. Fundamentally these were two: (1) the afrmation of a
real distinction between universals and individuals; (2) the elaboration
of new notions of identity and distinction as means to interpret the
relation between universals and individuals, and thereby the nature
of predication. The rst strategy was that of Walter Burley , who after
1324 in his writings always maintained that universals, conceived of as
general forms, fully exist outside the mind and are really distinct from the
individuals in which they are present and of which they are predicated.
The second strategy was the most common in all of Europe during the
later Middle Ages. Burley was persuaded that Ockhams arguments for
proving the theses (T1) and (T2) were valid, and therefore renounced
his support for thesis (T3) in order to escape from the inconsistencies
stressed by the Venerabilis Inceptor. On the other hand, the other late
medieval authors stuck by (T3), recognised that Ockhams demonstra-
tion of (T1) was effective, and unanimously judged Ockhams reasons
for (T2) to be insufcient for proving it. As a consequence, they had to
revise the notions of identity and distinction so that they could utilise
them for expounding the peculiar relation of partial identity and partial
distinction holding between universals and individuals without falling
prey to the contradictions pointed out by Ockham.
There were two different forms of this second strategy. The rst one
was that of some Italian Dominican masters (such as Franciscus de
Prato and Stephanus de Reate), who, in order to avoid the conclusion
that a universal and its individuals were considered to be the same
thing, worked out new denitions for identity and distinction inspired
by Herveus Nataliss notion of conformity (conformitas).
31
They regarded
identity as an intersection of classes of forms (or properties), so that it
31
Cf. Hervaeus Natalis, Quodlibet I, q. 9, ed. Venetiis 1513, fol. 19ravb. On Her-
vaeuss theory see Jan Pinborg , Zum Begriff der Intentio Secunda: Radulphus Brito ,
Hervaeus Natalis und Petrus Aureoli in Diskussion, Cahiers de linstitut du moyen-Age
grec et latin 13 (1974), 4959; Dominik Perler , Peter Aureol vs. Hervaeus Natalis on
Intentionality. A Text Edition With Introductory Remarks, Archives dhistoire doctrinale
et littraire du Moyen Age 61 (1994), 227262; Dominik Perler, Theorien der Intentionalitt im
Mittelalter (Frankfurt am Main, 2002), pp. 258317; Fabrizio Amerini , What is Real.
A Reply to Ockhams Ontological Program, Vivarium 43 (2005), 187212; Fabrizio
Amerini, Realism and Intentionality: Hervaeus Natalis, Peter Auriol, and William
was possible to afrm (1) that two entities were really identical without
entailing that they were the same thing, and (2) that two entities were
not really identical without entailing that they were really different. The
second line of development was that of the most important school of
later medieval Realists, started up by John Wyclif in the second half
of the fourteenth century, the so called Oxford Realists: besides
Wyclif himself, the Englishmen Robert Alyngton ( 1398), William
Milverley, William Penbygull ( 1420), Roger Whelpdale ( 1423), and
John Tarteys, as well as the German Johannes Sharpe ( after 1415),
and the Italian Paul of Venice (13691429). According to the Oxford
Realists (1) universals and individuals are really identical but formally
distinct; (2) the two notions of difference (or distinction) and real identity
are logically compatible, and (3) predication is not a mental relation
between two things, but a real one. In particular, Wyclif revised Duns
Scotus s twofold notion of formal distinction and developed a form
of intensional logic where the basic relation between things is that of
formal distinction, intended by him as the measure of the coincidence
of the metaphysical components of two res. Improving to some extant
Wyclif s ideas, his followers (1) proposed new determinations of the
twin notions of identity and distinction, and (2) deeply modied the
standard medieval analysis of predication.
Chronologically, the rst solution we meet is that of Walter Burley . As
we have already seen, at the beginning of his academic career Burley
was a supporter of the moderate realism, but after 1324, because of
Ockhams criticism, he changed his opinions and evolved an original
form of platonic realism. In the prologue of his last commentary on
the Physics (132434), in the Quaestiones super Porphyrium, in the last
commentary on the Ars Vetus (A.D. 1337), and in the later Tractatus
de universalibus (after 1337), he expounds his new ontology of macro-
objects, based on a threefold real distinction:
32
(1) between universals
Ockham in Discussion, in Philosophical Debates at the University of Paris in the First Quarter of
the Fourteenth Century, eds. Theo KobuschStephen F. Brown (Louvain, forthcoming).
32
On Burley s ontology of macro-objects see Alessandro D. Conti , Ontology in
Walter Burley s Last Commentary on the Ars Vetus, Franciscan Studies 50 (1990), 121176;
on his semantic theory see Alessandro D. Conti, Signicato e verit in Walter Bur-
ley, Documenti e studi sulla tradizione losoca medievale 11 (2000), 317350; and Laurent
Cesalli , Le ralisme propositionnel de Walter Burley, Archives dhistoire doctrinale et lit-
traire du Moyen Age 68 (2001), 155221; on his theory of universals and individuals set
in its epistemological context see Alessandro D. Conti, La conoscenza del singolare
in Walter Burley, Documenti e studi sulla tradizione losoca medievale 15 (2004), 517540;
on his theory of universals set in its metaphysical and semantic contexts see Marta
and individuals;
33
(2) between categorial items (or simple objectsincom-
plexa) and real propositions (or states of affairspropositiones in re);
34
and
(3) among the ten categories.
35
Burley s most accurate and analytical treatments of the problem of
universals are those set in the treatise on universals and in the chapter
on substance of his last commentary on the Categories, but interesting
remarks can be found in his last commentaries on the Isagoge and the
Liber sex principiorum, and in his Quaestiones super Porphyrium. Like Albert the
Great, whom he quotes by name, Burley admits the typical thirteenth
century division of universals into ante rem, in re, and post rem;
36
however,
he follows also Auriole and the early Ockham in positing, besides the act
of understanding (the standard post rem, conceptual universal), another
mental universal, distinct from the former, and existing in the mind
only as its object (habens esse obiectivum in intellectu).
37
By introducing a
second mental universal, existing obiective in the mind, Burley hoped to
account for the fact that we can grasp the meaning of a general noun
even though we have never seen any individual among those for which
the noun can stand for in a predicative sentence, and therefore without
properly knowing the universal it directly signies.
In order to avoid the inconsistencies of the traditional moderate realist
view on universals pointed out by Ockham, he claims that universals
fully exist outside the mind and are really distinct from the individu-
als they are-in and are predicated of. According to him, if universals
are no longer constitutive parts of their own individuals, then the
inconsistencies stressed by Ockham vanish, as universals cannot take
Vittorini , Predicabili e categorie nellultimo commento di Walter Burley all Isagoge di Porrio,
Ph.D. diss., University of Salerno, academic year 200405.
33
Cf. Burley , In Physicam Aristotelis Expositio et Quaestiones, prooem., ed. Venetiis 1501,
fol. 9rb; Expositio super Praedicamenta Aristotelis, cap. de subiecto et praedicato and cap. de
substantia, in Expositio super Artem Veterem Porphyrii et Aristotelis, ed. Venetiis 1509, fols. 20rb,
23rbvb, and 24va; Expositio super librum Perihermeneias, cap. de oppositione enuntiationum,
in Expositio super Artem Veterem Porphyrii et Aristotelis, fol. 74rbva; Tractatus de universalibus,
ed. Hans-Ulrich Whler, Leipzig 1999, pars III, pp. 1440.
34
Cf. Burley , Expositio super Praedicamenta Aristotelis, prooem., fols. 17vb18va; cap. de
subiecto et praedicato, fol. 20rb; cap. de priori, fol. 47va; Expositio super librum Perihermeneias,
prooem., fol. 66rab.
35
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de numero et sufcientia praedi-
camentorum, fol. 21rab.
36
Cf. Burley , Expositio super librum Sex principiorum, cap. de forma, in Expositio super Artem
Veterem Porphyrii et Aristotelis, fol. 53rb.
37
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de priori, fol. 48vb; and Tractatus
de universalibus, pars VI, pp. 6066.
the (opposite) properties of the latter. Furthermore, the principle holds
that causes must be proportionate to the effects they produce: but the
causes of an individual, which is a singular effect, must be individual,
while obviously those of a common nature must be universal. There-
fore, individual substance cannot be composed of anything but singular
form and matter; whereas universals must be compounded by genus
and specic difference, and by any other universal form ordered over
the genus. Consequently, the lowest species is not a constitutive part
of the individuals it is in and is predicated of, but only a form coming
together with their individual essences, and making their metaphysical
structure known, since it is the species or genus (namely, the type) which
individuals belong to (namely, instantiate).
38
Hence, Burley sharply dis-
tinguishes between two main kinds of substantial forms: one singular
( forma perciens materiam) and the other universal ( forma declarans quid-
ditatem). The former affects a particular piece of matter and, together
with it, brings the substantial composite (or hoc aliquid) about. The latter,
the lowest species, discloses the nature of the individual substances in
which it is present and of which it is predicated, but it is not one of
their constitutive parts.
39
This distinction between a forma perciens mate-
riam and a forma declarans quidditatem is very like the one, very common
among the moderate Realists of the thirteenth century, between forma
partis (the singular form which in union with a clump of matter brings
the substantial composite about) and forma totius (the universal form or
38
Cf. Burley , In Physicam Aristotelis expositio et quaestiones, prooem., fols. 8rb9vb.
39
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de substantia, fol. 23rbva:
Ad primum in contrarium dicendum est quod substantia singularis non componitur
ex universalibus, sed solum ex singularibus, quia Sortes non componitur nisi ex hac
materia et hac forma, et non componitur ex genere et differentia quae praedicantur
de eo in quid. Sed species de genere substantiae componitur ex genere et differentia,
et ex omnibus superioribus ad ipsum. Et huius ratio est quia effectus particularis sunt
causae particulares et effectus universalis sunt cause universales; . . . sed individuum est
effectus particularis et species est effectus universalis. . . . Ad illud quando probatur quod
species est pars individui, quia est quidditas individui, dicendum quod quidditas et forma
unum sunt. Et ideo, sicut forma est duplex, scilicet forma declarans quidditatem et
forma perciens materiam, sic quidditas est duplex: quia quaedam est quidditas quae
est forma perciens materiam et quaedam est forma declarans quidditatem. Quidditas
quae est forma perciens materiam est pars individui cuius est quidditas; sed quidditas
declarans quidditatem non est pars individui cuius est quidditas, nec est de essentia
talis individui, sed est essentialiter concomitans essentiam eius. . . . Ad quintum princi-
pale, cum dicitur an haec species, homo, sit eadem omnino res in Sorte et Platone an
alia et alia, dicendum quod haec species, homo, est eadem in Sorte et Platone. This
same thesis is supported with new arguments based on the denition of identity in the
Tractatus de universalibus, pars III, pp. 2228.
essence which is the type that the substantial composite instantiates).
40

But, because of the real distinction between universals and individuals,
Burley was able to draw from it a further conclusion that moderate Real-
ists could not draw: the dissolution of the problem of individuation.
For him, individual substances are really distinct from their own species
and from one another in and of themselves. A primary substance really
differs (1) from its species because the latter is not a part of its essence,
but a form present in it and dependent on it for actual existence, and
(2) from the other primary substances belonging to the same species
because of its own singular form and matter.
41
Such a dissolution of the problem of individuation is not the only
point of agreement between the later Burley and Ockham: like the
Venerabilis Inceptor, Burley rejects any kind of distinction in addition to
the real one (and therefore even that form of intentional distinction
he had employed in his early works). As it has been already said, he
admits Ockhams thesis (T2). In his last commentary on the Categories
42

and in the Tractatus de universalibus
43
he considers (1) identity a transitive,
symmetrical, and reexive relation; and (2) identity and difference (or dis-
tinction) two mutually incompatible concepts. In fact, he denes identity
and distinction as follows: a is identical with b if and only if for all x,
it is the case that x is predicated of a if and only if it is predicated of
b; a differs from b if and only if (1) there is at least one x such that a is
predicated of x and b is not, or vice versa, or (2) there is at least one y
such that y is predicated of a and not of b, or vice versa. The other two
main theses of Burelys system (the existence of a real proposition and
a real distinction among the ten categories) depend on his new position
40
Cf. e.g., Albert the Great, De quinque universalibus, tr. de universalibus in communi, cap. 8,
pp. 3738; Metaphysica VII, tr. 1, cap. 1; VIII, tr. 1, cap. 3, ed. Col., vol. 16.2, pp.
316317, and 391.
41
Cf. Burley , Expositio super Universalia Porphyrii, cap. de specie, in Expositio super Artem
Veterem Porphyrii et Aristotelis, fol. 10va: Intelligendum est quod unum individuum
substantiae non distinguitur ab alio solum per huiusmodi proprietates accidentis, sed
formaliter per suam formam et materialiter per suam materiam.
42
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de oppositione, fol. 44rb: Nota
quod ex isto loco sumitur doctrina bona ad cognoscendum identitatem vel diversitatem
aliquorum ad invicem. Et est: si unum praedicatur de aliquo de quo non praedicatur
reliquum, illa non sunt eadem, sed diversa; et si aliquid praedicatur de uno quod non
praedicatur de reliquo, illa non sunt idem. Et e contrario: si quicquid vere praedicatur
de uno vere praedicatur de reliquo, illa sunt eadem.
43
Cf. Burley , Tractatus de universalibus, pars III, p. 22.
on identity and distinction, and consequently on what he thought was
necessary in order to defend a realist view of universals.
Because of his giving up of the intentional difference, Burley
was compelled to make the ontological status of propositiones in re
much stronger than it was before. While in the Quaestiones in librum
Perihermeneias and middle commentary on the De interpretatione he had
clearly stated that mental propositions exist in our minds as in their
own subjects of inherence (habent esse subiectivum in intellectu) and real
propositions ( propositiones in re) exist in our minds as their intentional
objects (habent esse obiectivum in intellectu solum),
44
in his last commentary
on the Ars Vetus he afrms that a real proposition is an ens copulatum
formed by the entities for which the subject and the predicate of the
corresponding mental proposition stand, together with an identity-
relation, if the proposition is afrmative, or a non-identity-relation, if
the proposition is negative.
45
Moreover, as far as the problem of the
ontological value of the Aristotelian categories is concerned, whereas
in his middle commentary on the Categories
46
Burley (1) judged only the
absolute categories (substance, quantity, and quality) to be really things,
and (2) considered the remaining ones as a sort of real aspects (respectus
reales) of the absolute ones, in his last commentary on the Categories he
claims that (1) the division into categories is rst of all a division of
res existing outside the mind, and only secondarily of the mental con-
cepts and spoken or written terms which signify them; and (2) things
in one category are really distinct from those in others.
47
What is more,
(3) he polemicizes against Ockhams strong reductionist position, argu-
ing that it compromises the actual goal of a correct categorial theory,
namely, the classifying and putting in hierarchical order all the items
according to their nature, inner metaphysical structure, and peculiar
modes of being.
48
Notwithstanding the assumption of a real distinction between uni-
versals and individuals, and the different evaluation of the categorial
44
Cf. Burley , Quaestiones in librum Perihermeneias, q. 3, pp. 248249; Commentarius in
librum Perihermeneias, pp. 6162.
45
Cf. Burley , Expositio super librum Perihermeneias, prooem., fol. 66rab.
46
Cf. Burley , Tractatus super librum Praedicamentorum, cap. de numero et sufcientia praedi-
camentorum, fols. 175rb176rb.
47
camentorum, fol. 21rab.
48
camentorum, fols. 21vab, and 22ra.
table in relation to his early works, Burley kept on supporting, without
introducing any restrictive clause, the Aristotelian principle (stated in
Categories 5, 2b56) that primary substances are the necessary condi-
tion of existence for any other categorial items (universal substances
included).
49
This was still possible since he held that universals were
forms, and therefore entities existentially incomplete and dependent
which require the existence of at least one individual for being. Ockham
had interpreted in an original way that same principle: he had claimed
that what Aristotle meant was that the truth of all the propositions of
the form Sortes is not necessarily entails the truth of the following
proposition no man is (nullus homo est),
50
so translating a metaphysical
principle into a logical rule. Paradoxically, the divergent interpretation
of Categories 2b56 is the only remarkable difference between the logical
structure of Burleys theory of universals and that of Ockham. Burley
himself stresses this point: in commenting on that passage, he observes
that it goes against Ockhams position on universals, since, if universal
substances were concepts, the destruction of all the members of a cer-
tain species could not imply the disappearance of the corresponding
universal.
51
Indeed, both Ockham and Burley subscribe to (T1) and
(T2), and do not accept (T3), even though Ockham only accepted it
in the sense that universal concepts (say, the concept of man) do exist
apart from and independently of their corresponding real individuals
(say, Socrates or Plato ). What prevents Burley from totally agreeing with
Ockham is precisely the opposite reading of Aristotles afrmation that
if primary substances did not exist, it would be impossible for anything
else to exist. Being faithful to Aristotle on this point, Burley had to
build up a sort of mixed theory of universal, where some principles
of Aristotelian ontology went alongside some principles of Platonic
ontology. In contrast, detaching himself from the real Aristotelian
intention, Ockham could construct a theory consistent with Aristotles
claim in Metaphysics Z 13 (1038b89, 1038b341039a3) that universals
are not substances.
49
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de substantia, fol. 24va.
50
Cf. Ockham, Expositio in librum Praedicamentorum Aristotelis, cap. 8.5, pp. 175176.
51
Cf. Burley , Expositio super Praedicamenta Aristotelis, cap. de substantia, fol. 24va: Hic
patet evidenter quod secundae substantiae non sunt conceptus in anima, quia, si secun-
dae substantia essent conceptus animae, tunc destructis primis substantiis possibile esset
alia remanere. Nam destructis omnibus rosis adhuc potest conceptus rosae remanere
in anima. Et ita species potest manere destructis omnibus individuis suisquod est
contra Philosophum hic.
Two main difculties arise from Burley s (new) solution to the prob-
lem of universals: (1) within his ontology, it was difcult to distinguish
between essential and accidental predication, since secondary substances
(namely, the universal forms of the category of substance) necessarily
presupposed primary substances for their existence in the same way as
accidental forms did. Thereby, their relation to primary substances was
(almost) the same as that peculiar to accidental forms, a sort of inher-
ence. (2) Universals had a being different from that of their individu-
alsa conclusion dangerously close to Plato s thesis on the subject.
As a consequence, many late medieval Realists tried other ways of
replying to Ockhams charges.
When Ockhamist logic arrived in Italy in the 1330s, some Ital-
ian Dominican masters brought into question Ockhams ontological
reductionist program, at the same time attempting to escape from the
exaggeration of Burley s version of realism. According to Francis-
cus de Prato and Stephanus de Reate,
52
in order to defend moderate
realism, it was necessary (1) to clarify the relation that holds between
universals and individuals, and (2) to rethink the twin notions of identity
and distinction in such a way to avoid both the inconsistencies pointed
out by Ockham, and Burleys real distinction between individuals and
universals. To preserve a real foundation of universal concepts and to
defend the entire extension of the table of the categories were the two
main steps of their philosophical strategy, aimed to restore the prin-
ciples of a Thomistic view of the world. In order to achieve this goal
they developed some of Hervaeus Nataliss chief logico-metaphysical
intuitions.
53
52
On their lives, works, and logico-metaphysical theories see Fabrizio Amerini , La
quaestio Utrum subiectum in logica sit ens rationis e la sua attribuzione a Francesco da
Prato. Note sulla vita e gli scritti del domenicano Francesco da Prato (XIV secolo),
Memorie Domenicane, n.s. 30 (1999), pp. 147217; Fabrizio Amerini, La dottrina della
signicatio di Francesco da Prato, O.P. (XIV secolo). Una critica tomista a Guglielmo di
Ockham, Documenti e studi sulla tradizione losoca medievale 11 (2000), 375408; Fabrizio
Amerini, I trattati De universalibus di Francesco da Prato e Stefano da Rieti (secolo XIV),
Centro Italiano di studi sullalto Medioevo (Spoleto, 2003), pp. 156 (at pp. 57132 the
critical edition of Fransciscuss treatise on universals, and at pp. 133145 the edition of
Stephanuss tract); Christian Rode , Franciscus de Prato (Stuttgart, 2004); Fabrizio Ame-
rini, La logica di Francesco da Prato. Con ledizione critica della Loyca e del Tractatus de voce
univoca (Florence, 2005), pp. 1248 (at pp. 249506 the critical edition of Franciscuss
Loyca, and at pp. 507597 the edition of his Tractatus de voce univoca); Amerini, What
is Real, pp. 187212.
53
Amerini , What is Real, pp. 200201, and 212.
Like Hervaeus , Franciscus and Stephanus rejected any kind of distinc-
tion that was midway between a real distinction and one of reason.
54

Their basic ideas were that (1) universal forms have no being outside the
being of their individuals;
55
(2) real identity may be more or less close,
and (2.1) the limit of that real identity is the entirely real identity (esse
idem convertibiliter or totaliter).
56
This means that they recongnised degrees
in real identity. Moreover, Franciscus explicitly assumes that (1) from the
fact that two items are not entirely really identical we cannot conclude
the they are really different.
57
According to Franciscus, a universal and
one of its individuals are really identical, but they are not entirely iden-
tical. There are properties that can be predicated of a universal form
and not of one of its individual, or vice versa, but this is not equivalent
to proving that they are really non-identical, so that they can be really
separate. For instance, Peter and man (homo) are really identical if they
are compared to each other, but they are not entirely identical if they
are compared to another individual man, say Martin. What is more,
according to Franciscus, saying that if two items are not identical in
any respect, then they are really different, amounts to formulating an
54
Ibid., p. 201.
55
Cf. e.g., Franciscus de Prato, Tractatus de universalibus (between 1341 and 1343), a. 5,
p. 110: Cuius ratio est quia albedo universalis, in tali casu, non habet in rerum natura
aliam entitatem, sive aliam albedinem, quam albedinem Petri; p. 121: Cuius ratio
est quia homo universalis realiter et essentialiter praedicatur de Petro et de Martino,
et est de essentia Petri et Martini, et est homo universalis unum realiter cum Petro et
idem est unum realiter cum Martino.
56
Cf. e.g., Franciscus de Prato, Tractatus de universalibus, a. 4, pp. 99100: Ex his quae
tacta sunt in ista conclusione ego elicio quattuor correlaria. Primum est quod universale
et singulare, ad invicem comparata, sunt idem realiter et essentialiter et differunt sola
ratione. . . . Secundum correlarium est quod universale et singulare, in comparatione
ad tertium, idest in comparatione ad aliud singulare, distinguuntur realiter. Cuius
ratio est quia universale identicat sibi aliud singulare nec differt ab eo realiter; unum
autem singulare non identitat sibi aliud singulare, sed differt realiter ab eo. Unde Petrus
et homo differunt realiter in comparatione ad Martinum, pro quanto homo includit
identice Martinum, qui differt realiter a Petro. Tertium correlarium est quod universale
et singulare non sunt idem convertibiliter. Cuius ratio est quia ex quo universale plura
includit et in plus se habet quam singulare sequitur quod non sunt idem convertibiliter.
Quartum correlarium est quod ex eo quod universale est idem realiter cum suis sin-
gularibus sequitur quod illud universale est subiective in intellectu cuius singulare vel
singularia sunt subiective in intellectu, et illud universale est extra intellectum subiective
cuius singulare vel singularia sunt extra intellectum subiective.
57
Cf. Franciscus de Prato, Tractatus de universalibus, a. 5, p. 121: Cum ulterius dicitur
si homo et Petrus non sunt idem omnibus modis realiter, ergo differunt aliquo modo realiter, dico quod
non sequitur . . . sicut etiam non sequitur Petrus non est omnis homo, ergo Petrus est
aliquid non-homo, vel sicut non sequitur Petrus non est omnibus modis identitatis et
omnibus modis unitatis homo, ergo Petrus aliquo modo est non-homo.
invalid consequence (in his opinion it is a case of fallacia consequentis).
58

Franciscus argues that the following consequences are valid: if man
and Peter are identical in any respect, then they are identical, if man
and Peter are not identical, then they are not identical in any respect;
while the following consequences are invalid: if man and Peter are
identical, then they are identical in any respect, if man and Peter are
not identical in any respect, then they are not identical.
59
In fact, he
distinguishes between being-not-identical (esse non idem), that he equates
with being-different (differre), and not-being-identical (non esse idem), as
the former relation implies the latter, but not vice versa.
60
Franciscus s position on universals and his explanation of the rela-
tionship between universals and individuals ultimately depend on the
acceptance of the Thomistic thesis of a real composition of essence
and being (esse) in creatures. Pursuing Aquinas s line of thought, he
assumed that essence and being are distinct from each other, but
related one to the other just as potency (essence) and act (being). All
the individuals belonging to a certain natural species share the same
essence (or nature), in the sense that each of them derives from one and
the same essence. On the other hand, that same common essence qua
the essence of a given individual is, from an ontological point of view,
singular and not universal, since it is contracted by the act of being
peculiar to that given individual.
61
Hence, each individual belonging
58
Cf. Franciscus de Prato, Tractatus de universalibus, a. 4, p. 100; a. 5, pp. 113; and
119121.
59
Cf. Franciscus de Prato, Tractatus de universalibus, a. 5, pp. 119120. See Amerini ,
What is Real, p. 205.
60
Cf. Franciscus de Prato, Tractatus de universalibus, a. 5, pp. 121122: Hic etiam nota
quod aliud est dicere quod aliqua duo aliquo modo non sunt idem et aliud est dicere
quod aliqua duo aliquo modo sunt non idem: nam prima propositio est de praedicato
negato et secunda propositio est de praedicato innito. Et quia ad propositionem de
praedicato negato non sequitur propositio de praedicato innito, e converso autem sic,
ut patet in II Perihermeneias, ideo ad hanc propositionem: homo et Petrus aliquo modo
(non) sunt idem, quae est de praedicato negato, non sequitur ista conclusio: homo et
Petrus aliquo modo sunt non idem, quae est de praedicato innito. Et ideo dato quod
possit concedi quod homo et Petrus (inter se) aliquo modo non sint idem realiter, quia
non sunt idem convertibiliter, tamen non potest concedi quod homo et Petrus inter
se aliquo modo sint non idem (realiter); et per consequens non potest concedi quod
inter se aliquo (modo) differant realiter, quia numquam de aliquibus potest vericari
quod aliquo modo differant realiter inter se, nisi possit vericari de eis quod aliquo
modo sint non idem realiter (inter se), quia differre et esse non idem uniformiter et
convertibiliter se habent.
61
Cf. e.g., Franciscus de Prato, Tractatus de sex transcendentibus, ed. Burkhard Mojsisch,
in Bochumer Philosophisches Jahrbuch fr Antike und Mittelalter 5 (2000), 177217, a. 1, pp.
to a certain natural species is really different from any other belonging
to that same natural species because of its own (act of ) being, and not
because of its essence. As a consequence, according to Franciscus (and
Stephanus), (1) there is real identity between two items a and b if and
only if they share at least one act of being, and (2) two items a and
b are entirely identical if and only if, for all the acts of being , it is
the case that is an act of being of a if and only if it is an act of
being of b. From this, it necessarily follows that (1) a universal essence
is really identical with each of its individuals but not entirely identical,
as it shares (or may share, in the case of universal essences instantiated
by one individual only) many other beings. For instance, humanity qua
instantiated by a given individual, say Peter, is really identical with
him, but qua instantiated by another one, say Martin, it is not entirely
identical with Peter. (2) Real identity is not a transitive relation. From
the fact that a certain universal essence, say humanity, is really identical
with each of its individuals (Peter, Martin, and so on) does not follow
that each of them is really identical with any other.
62
Franciscus s (and Stephanuss) theory of universals testies to the per-
manent inuence of Thomas Aquinas s approach within the Dominican
order in the middle of the fourteenth century. They dealt with universals
and categories from the point of view of a metaphysics of being, and
not of essences. The rules they explicitly laid down or implicitly followed
are ineffectual or contradictory if they are set in a different metaphysical
context. What Franciscus stated about identity and distinction makes
sense only if interpreted as concerning acts of being. The same is true
for his defence of the reality (and real distinction) of each category.
He observes that all the ten Aristotelian categories are things (res), but
according to two different meanings of the term res. In fact, it signi-
es (1) a real essence, or (2) the mode of being of a real essence. The
three absolute categories (substance, quantity, and quality) are things in
182 and 185; a. 2, pp. 187188; a. 6, p. 216. See Amerini , La logica di Francesco da
Prato, pp. 151157.
62
Cf. e.g., Franciscus de Prato, Tractatus de universalibus, a. 5, p. 119: Ad quadrage-
simum quartum dico quod dato quod homo universalis sit existens in rerum natura,
in (quo) conveniunt Petrus et Martinus, non tamen sequitur quod Petrus et Martinus
sint idem inter se. Et quando dicitur quod (quae) uni et eidem sunt eadem inter se sunt eadem,
dico quod verum est quod aliqua sint eadem realiter in aliquo quod sit unum realiter;
sed Petrus et Martinus sunt eadem realiter in entitate hominis, quae est unum non
realiter (immo est plures realiter), sed unum est tantum secundum rationem. In qua
unitate Petrus et Martinus sunt unum secundum rationem et non secundum rem, ex
quo remanent Petrus et Martinus distincta realiter.
the former sense (of the term); while the latter is the sense according
to which the other seven categories are said to be things.
63
The basic
pattern of his arguments aimed to prove, against Ockham, that the ten
categories are really different from each other is an example of reductio
ad absurdum: if a certain category a was really identical with another
category b, (namely, if a and b shared at least one act of being), then
one of them would be over-ordinated to the other, and therefore it
would transmit all its essence to the other, so that it would be possible
to nd all the essential properties of a among the essential properties
of b, or vice versa. But, for instance, none among the essential proper-
ties of substance is one of the essential properties of quantity, nor vice
versa, since they are described in totally different ways. And therefore
substance and quantity are really different categories.
64
Franciscus and Stephanus were inuenced by Thomas Aquinas s and
Hervaeus Nataliss metaphysical views, and especially by the theory
of a real composition of essence and being, but the most inuential
metaphysics of the later Middle Ages was Duns Scotus s. In fact, many
among the main philosophical convictions of Wyclif , who pointed to
the strategy (almost) all the subsequent Realists were to adopt, were
an original version of the most fundamental theses of Duns Scotuss
system, where much more stress was put on the ontological presup-
positions and entailments of the doctrines. In particular, no one
maintained a real distinction between essence and being: since later
medieval Realists considered being (ens) as a sort of stuff that the ten
categories modulated according to their own natures, they thought
that any categorial item (universal substances and accidents included)
was immediately something which is, so that the essences of creatures
do not precede their beings, not even causally, as every thing is (really
identical with) its essence.
65
63
Cf. Franciscus de Prato, Logica, pars I, tr. 5, a. 1, p. 381; Tractatus de sex transcen-
dentibus, a. 6, p. 215.
64
Cf. Franciscus de Prato, Logica, pars I, tr. 5, a. 7, pp. 415424.
65
Cf. e.g., John Wyclif , Tractatus de universalibus (1374 ca.), ed. Ian J. Mueller (Oxford,
1985), cap. 6, pp. 120123. In view of his position on the problem of being, Wyclif
maintains no real distinction between essence and being. According to him, the being
of a thing is brought into existence by God at the same instant as its essence, since
essence without being and being without essence would be two self-contradictory
states of affairs. In fact, essence without being would imply that an individual could be
something of a given type without being real in any way, and being without essence
would imply that there could be the existence of a thing without the thing itself. As a
consequence, the pars destruens of his theory on being and essence is a strong refutation
As just indicated, the starting point of the latest medieval realist
speculations on universals, predication, and categories were the theories
on these subjects worked out by Wyclif .
66
Like Burley , Wyclif reacted
to Ockhams attack to the traditional realist view on universals and
categories, but, unlike him, he (1) stuck by the thesis of the real identity
of universals and individuals, and (2) revised the theory of predication
and the notions of identity and distinction. He (1) explicitly presents
his opinion on universals as intermediate between those of Aquinas
and Giles of Rome, on the one hand, and of Burley, on the other
hand,
67
and (2) many times in his works expresses the deepest hostil-
ity to Ockhams approach to philosophy. Ockham and his followers
discriminated between things as they exist in the extra-mental world
and the concepts and schemata by means of which we can grasp and
signify them. As we have seen, according to the Venerabilis Inceptor, in
the world there are only individual substances and qualities; on the
contrary, the signs by which they are understood and signied are both
individual and universal, and of ten different types (the ten Aristotelian
categories). Nor do the relations through which we connect our concepts
in a proposition correspond to the real links which connect individual
items in a state of affairs. Thus, our knowledge does not reproduce
the world and its items, but merely concerns them. Wyclif maintained
that such an approach to philosophical questions was misleading and
deleterious. Like Burley, he thought that only on the ground of a close
isomorphism between mental language and the world could the signify-
ing function of terms and propositions, the possibility of denitions,
and nally the validity and universality of our knowledge be accounted
of the twin opinions of Aquinas and Giles of Rome. On Wyclif s theory of essence
and being see Alessandro D. Conti , Wyclif s Logic and Metaphysics, in A Companion
to John Wyclif, ed. Ian C. Levy (Leiden, 2006), pp. 8995.
66
On Wyclif s main logico-metaphysical theories see John A. Robson , Wyclif and
the Oxford Schools (Cambridge, 1961); Paul V. Spade , Introduction, in John Wyclif,
On Universals, trans. Anthony Kenny (Oxford, 1985), pp. viixlvii; Anthony Kenny,
Wyclif (Oxford, 1985), pp. 130; Anthony Kenny, The Realism of De Universalibus,
in Wyclif in his Times, ed. Anthony Kenny (Oxford, 1986), pp. 1729; Alessandro D.
Conti , Logica intensionale e metasica dellessenza in John Wy cli f, Bollettino dellIstituto
Storico Italiano per il Medioevo e Archivio muratoriano 99.1 (1993), 159219; Alessandro D.
Conti, Analogy and Formal Distinction: on the Logical Basis of Wyclif s Metaphysics,
Medieval Philosophy and Theology 6.2 (1997), 133165; Paul V. Spade, The Problem of
Universals and Wyclif s Alleged Ultrarealism, Vivarium 43 (2005), 111123; Laurent
Cesalli , Le pan-propositionnalisme de Jean Wyclif, Vivarium 43 (2005), 124155;
Conti, Wyclif s Logic and Metaphysics, pp. 67125.
67
Cf. Wyclif , Tractatus de universalibus, cap. 4, pp. 8687.
for and ensured. He rmly believed that mental language was an
ordered collection of signs, each referring to one of the metaphysical
constituents of the world (indivi duals and universals, substances and
accidents, concrete properties, like being-white, and abstract forms,
like whiteness), and that true propositions were like pictures of the
inner structure and mutual relationships of such items (or essences,
according to his terminology). So the main characteristics of his own
form of realism, which were to inspire the strategy and doctrines of
the (other) Oxford Realists, were the trust in the scheme object-label as
the fundamental interpretative key of semantic problems, and a strong
propensity towards hypostatization.
Like moderate Realists of the thirteenth century, Wyclif recognizes
three main kinds of universals: (1) ante rem, or ideal universals; (2) in re,
or formal universals; and (3) post rem, or intentional universals. On
the other hand, just like Burley , Wyclif holds that formal universals
exist in actu outside our minds, and not in potentia, as moderate Real-
ists thoughteven if, unlike Burley, he maintains that they are really
identical with their own indivi duals.
68
In his view, (1) universals and
individuals share the same empirical reality, which is that of individuals,
but (2) have opposite constituent principles, when properly considered as
universals and individuals. According to his terminology, they are really
the same but formally distinct.
69
This formulation is only another way
of saying that universals and individuals are the same identical things
if conceived as rst intentions, and differ from each other if conceived
as second intentions. So at last Wyclif accepts the very core of the
traditional realistic account of the relationship between universals and
individuals; but he tries to improve it by dening more accurately its
predicative structure.
According to Wyclif , because of the formal distinction, not everything
predicable of individuals can be directly predicated of universals and
vice versa, although an indirect predication is always possible. As a
consequence, Wyclif distinguished three main non mutually exclusive
68
Cf. Wyclif , Tractatus de universalibus, cap. 2, p. 69. In addition to this partition
of universals, standard in the Middle Ages, Wyclif adopts another one, which was
very successful among his follo wers, based on the different functions that universal
es sences perform (Tractatus de univer salibus, cap. 1, pp. 1516): he divides universals into
(1) universals by causality (causatione), (2) universals by community (communi catione), and
(3) universals by representation (repraesentatio ne).
69
Cf. Wyclif , Tractatus de universalibus, cap. 2, pp. 6263; cap. 4, pp. 8687; cap. 10,
pp. 208213.
types of predication (that he conceives as a real relation which holds
between metaphysical entities),
70
each more general than the preced-
ing one (or ones). In the Tractatus de universalibus they are the following:
formal predication ( praedicatio formalis), predication by essence ( praedi-
catio secundum essentiam), and habitudinal predication ( praedicatio secundum
habitudinem). (1) Formal predication is that in which the form signied
by the predicate-term is directly present in the entity designated by the
subject-term. This happens whenever an item in the categorial line is
predicated of its inferior or an accident of its subject of inherence.
(2) It is sufcient for predication by essence that the same empirical
reality is both the real subject and predicate, even though the formal
principle connoted by the predicate-term differs from that connoted
by the subject-term. God is man and (What is) universal is (what is)
singular (universale est singulare) are instances of this kind of predica-
tion. (3) There is habitudinal predication when the form connoted by
the predicate-term does not inhere, directly or indirectly, in the thing
signied by the subject-term but simply implies a relation to it, so that
the same real predicate may be at different times said truly or falsely
of its real subject without any change in the subject itself. According
to Wyclif, we use this kind of predication mainly when we want to
express theological truths, such as: that God is known and loved by
many creatures, and brings about, as efcient, exemplar, and nal
cause, many good effects.
71
Habitudinal predication does not require any kind of identity between
the entity signied by the subject-term and the entity signied by the
predicate-term, but formal predication and essential predication do.
Thus, the ontological presuppositions of the most general type of
predication, implied by the other types, are completely different from
those of the other two. In any case, it is clear that Wyclif s efforts were
directed towards unifying the various kinds of predication through a
unique basic relation of partial identity: the formal distinction. In the
Tractatus de universalibus it is described as the difference by which things
differ from each other even though they are constitutive elements of the
same single essence or supposit. According to Wyclif, this is the case for:
(1) the concrete accidents inhering in the same substance, as they coin-
cide in the same particular subject, but differ from each other because
70
71
Cf. Wyclif , Tractatus de universalibus, cap. 1, pp. 2830 and 34.
of their own natures; (2) the matter and substantial form of the same
individual substance; (3) what is more common in relation to what is
less common, like (3.1) the divine nature and the three Persons, (3.2) the
world and this world; and, (3.3) among the categorial items belonging
to the same category, a superior item and one of its inferiors.
72
As it is
evident, the rst sub-type of the formal distinction is a slightly differ-
ent version of the Scotistic formal distinction as dened in the Lectura
and in the Ordinatio, while the third sub-type is a reformulation of the
Scotistic formal distinction as described in the Reportata Parisiensia.
Formal distinction is one of the three main kinds of difference that
Wyclif recognises, as he speaks of real-and-essential and real-but-not-
essential differences also.
73
Still, formal distinction is the main kind of
transcendental relation holding among (Wyclif s) items, as it is the tool
by means of which the dialectic one-many internal to the individual
substances is regulated. It is intended to explain why one and the same
individual substance (say, Socrates ) is one thing, even if it contains in
itself a lot of simpler entities, and how many different entities can
constitute just one thing. Moreover, formal distinction sets the relations
(1) between a concrete accident, such as being-white (album), and its sub-
stance, namely, the substance in which the corresponding abstract form
(say, whiteness) inheres; (2) among all the concrete accidents belonging
to one and the same substance. Consequently, formal distinction is a
central notion for the categorial doctrine too.
Wyclif held that the extramental world is divided into ten genera
of beings, or categories, none of which can be reduced to another
one.
74
Since, like Aristotle (and the moderate Realists), he thought of
substance as the ultimate substrate of existence and subject of predica-
tion in relation to anything else, the only way to demonstrate the reality
of the items belonging to other categories was to conceive of them
as forms and attributes of substance. Accordingly, he insisted that the
items falling into the accidental categories, considered by themselves,
are forms inherent in the composite substances.
75
In this way, Wyclif
wanted to safeguard the reality of accidents as well as their distinction
72
73
Cf. Wyclif , Tractatus de universalibus, cap. 4, pp. 9091. On Wyclif s notions of dif-
ference see Conti , Analogy and Formal Distinction, pp. 158163; Conti, Wyclif s
Logic and Metaphysics, pp. 7278.
74
Cf. Wyclif , De ente praedicamentali (1369 ca.), cap. 4, ed. Rudolf Beer (London,
1891), pp. 3032.
75
Cf. Wyclif , De ente praedicamentali, cap. 6, p. 48.
from substance and from one another, while at the same time afrm-
ing their dependence on substance in existenceso following some of
the main principles stated by Aristotle in the Categories. And in fact, in
the De ente praedicamentali (1) he clearly states that quantity, quality, and
relation are entities with well determined natures, and really distinct
from substance;
76
and, (2) like all the other medieval Realists, he also
subscribes to the Aristotelian thesis that primary substances are the
necessary condition of existence for any other categorial items.
77
All
the more so because, as what he says about the rst sub-type of formal
distinction makes evident, accidents considered in an absolute way,
according to their own natures, are abstract forms, really distinct from
substance; yet, if considered from the point of view of their existence
as concrete accidents, they are only formally, but not really distinct from
the substance in which they are present and that they affect, since, in
this case, they are modes of that substance.
78
The nal result of Wyclif s approach to the problems of universals
and categories was a system of intensional logic where (1) the copula
of any standard philosophical proposition (like Socrates is white, or
man is an animal) had to be interpreted as meaning degrees in identity
between the things signied by the subject-term and the predicate-term;
and (2) individuals and universals, considered qua beings, appear to be a
sort of hypostatisation of intensions, since they are what is signied by
proper and common nouns respectively. Only in virtue of renouncing
an extensional approach to the matter was Wyclif able to give a (par-
tially) satisfactory solution of the problem of the relationship between
universals and individuals, which had always been the most difcult
issue for any medieval form of realism.
The Oxford Realists
Wyclif s logico-metaphysical system, however rigorous in its general
design, containsas we have seensome unclear and aporetic points,
76
Cf. Wyclif , De ente praedicamentali, cap. 6, pp. 48 and 50; cap. 7, pp. 6162.
77
78
Cf. Wyclif , De actibus animae (1365 ca.), pars 2, cap. 4, in Johannis Wyclif miscellanea
philosophica, ed. Michael H. Dziewicki, 2 vols. (London, 1902), [vol. 1, pp. 1160], pp.
12223, and 127. On Wyclif s categorial doctrine see Conti , Logica intensionale e
metasica dellessenza, pp. 197209; Conti, Wyclif s Logic and Metaphysics, pp.
103113.
which needed to be removed. All the same, Wyclif s philosophy exer-
cised an enormous inuence on the forms of later medieval realism,
since his intuitions concerning universals, predication, and categories
played a large role both in logic and metaphysics of many authors,
and especially of the Oxford Realists.
79
According to these authors
(who in their works show the closest familiarity with Wyclif s writings)
(1) universals and individuals are really identical but formally distinct;
(2) predication is a real relation between things; (3) the ten Aristotelian
categories are ten really distinct kinds of beings. Partially modifying
Wyclif s doctrine, they (1) introduced a new type of predication, called
predication by essence (secundum essentiam), based on a partial identity
between the entities for which the subject and predicate stood; and
(2) redened the traditional kinds of predication (the essential and
79
Besides Wyclif himself, Robert Alyngton, William Milverley, William Penbygull,
Roger Whelpdale, John Tarteys, Johannes Sharpe, and Paul of Venice. All those phi-
losophers studied and/or taught in Oxford: Alyngton at Queens College, Penbygull
at Exeter College, Whelpdale at Balliol and Queens Colleges, Tarteys at Balliol Col-
lege; Paul of Venice at the Augustinian studium from 1390 to 1393. On their lives and
works see A. B. Emden , A Biographical Register of the University of Oxford to A.D. 1500
(Oxford, 195759), 3 vols., sub nominibus. On Paul of Venices life and works see Alan
R. Perreiah , Paul of Venice: a Bibliographical Guide (Ohio, 1986). Sharpes Quaestio super
universalia has been edited in Sharpe, Quaestio super universalia, pp. 1145; Penbygulls
De universalibus has been edited in Alessandro D. Conti , Teoria degli univer sali e
teoria della predicazione nel trattato De universalibus di William Pen bygull: discussione
e difesa della posizione di Wyclif, Medioevo 8 (1982), 137203, pp. 178203. Excerpta
from Alyngtons main work, a commentary on the Categories, in Alessandro D. Conti,
Linguaggio e realt nel commento alle Categorie di Robert Alyngton, Do cumenti e studi
sulla tradizione losoca medievale 4 (1993), 179306, pp. 242306; excerpta from Milverleys
Compendium de quinque universalibus, Whelpdales Tractatus de universalibus, Tarteys Problema
correspondens libello Porphyrii, and Paul of Venices Quaestio de universalibus, in Sharpe, Quaestio
super universalia, Appendices II, III, IV, and V, pp. 159164, 165187, 189197, and
199207 respectively. For analyses of their main works and doctrines and information
on Wyclif s inuence see: Conti, Teoria degli universali e teoria della predicazione,
pp. 137166; Conti, Studio storico-critico, in Sharpe, Quaestio super universalia, pp.
211238, and 295336; Alain de Libera , Questions de ralisme. Sur deux arguments
antiockhamistes de John Sharpe, Revue de metaphysique et de morale 97 (1992), 83110;
Alessandro D. Conti, Il problema della conoscibilit del singolare nella gnoseologia
di Paolo Veneto, Bollettino dellIstituto Storico Italiano per il Medio Evo e Archivio muratoriano
98 (1992), 323382; Conti, Linguaggio e realt, pp. 179241; Alessandro D. Conti,
Esistenza e verit: Forme e strutture del reale in Paolo Veneto e nel pensiero losoco del tardo medioevo
(Rome, 1996); Alain de Libera, La querelle des universaux. De Platon la n du Moyen Age
(Paris, 1996), pp. 402428; Alessandro D. Conti, Paul of Venice on Individuation,
Recherches de Thologie et Philosophie mdivales 65 (1998), 107132; Alessandro D. Conti,
Paul of Venices Theory of Divine Ideas and its Sources, Documenti e studi sulla tra-
dizione losoca medieva le 14 (2003), 407448; Alessandro D. Conti, Johannes Sharpes
Ontology and Semantics: Oxford Realism Revisited, Vivarium 43 (2005), 156186;
Conti, Wyclif s Logic and Metaphysics, pp. 118125.
accidental predications) in terms of this partial identity. Furthermore,
some of them, starting from Wyclif s characterization of identity and
distinction, formulated new denitions of these transcendental relations,
which were the main tools that they utilised in building up their own
philosophical systems.
Like Wyclif , Oxford Realists accepted the core of the moderate realist
approach to the problem of universals. Their accounts (1) of the kinds
and status of universals, and (2) of the relationship between universals
and individuals are very similar to those worked out by moderate realist
authors of the second half of the thirteenth century, such as Albert the
Great and Duns Scotus . According to Oxford Realists, we can count
the following entities universal: (1) those causes that have a multiplicity
of effects (universalia causalia, or causatione), such as God and the angelic
intelligences; (2) the ideas in God (universalia idealia, or exemplatione);
(3) universal forms, or real universals (universalia realia, or formalia, or
communicatione, or praedicatione); and (4) universal concepts (universalia
repraesentatione, or cognitione).
80
In their view, formal universals are common
natures in virtue of which the individuals that share them are exactly
what they are, just as humanity is the form by which every man formally
is a man. Like Wyclif, all of them thought that common natures exist
in actu in the external world and are really (realiter) identical-with and
formally ( formaliter) distinct-from their own individuals.
81
Since, like Wyclif , Oxford Realists accepted the essence of the tradi-
tional, realist account of the relationship between (formal) universals and
individuals, they too had to dene its logical structure more accurately,
but at the same time trying to avoid Wyclif s aporias. As a consequence
80
Cf. Alyngton, Litteralis sententia super Praedicamenta Aristotelis, cap. de substantia, pp. 279
and 287; Penbygull, De universalibus, p. 178; Milverley, Compendium de quinque universalibus,
p. 159; Tarteys, Problema correspondens libello Porphyrii, p. 171; Whelpdale, Tractatus de uni-
versalibus, p. 189; Paul of Venice, Lectura super libros Metaphysicorum, prooem, cap. 2, Pavia,
Biblioteca Universitaria, fondo Aldini, ms. 324, fol. 9vab. Tarteys and Whelpdale add
a fth kind of universal to that list: the universal by reception (receptione), namely, place
and primary matter. Sharpe (Quaestio super universalia, pp. 4950) adds two other kinds:
the universal quantiers (such as omnis and nullus), and universal propositions, both
afrmative and negative.
81
Cf. Alyngton, Litteralis sententia super Praedicamenta Aristotelis, cap. de substantia,
p. 268; Penbygull, De universalibus, pp. 181 and 189; Milverley, Compendium de quinque
universalibus, p. 163; Tarteys, Problema correspondens libello Porphyrii, pp. 178179; Whelp-
dale, Tractatus de universalibus, pp. 193194; Sharpe, Quaestio super universalia, pp. 9192;
Paul of Venice, Quaestio de universalibus, p. 199; Expositio super Universalia Porphyrii, cap.
de genere, in Expositio super Universalia Porphyrii et Artem Veterem Aristotelis (A.D. 1428), ed.
(Venetiis, 1494), fol. 14vb.
they had to redene predication, and through it (implicitly or explicitly)
also the notions of identity and distinction (or difference).
In particular, Alyngton, and some years later Sharpe, Milverley, and
Tarteys, divided predication into formal predication and predication
by essence (secundum essentiam), that Alyngton calls also remote inher-
ence (inhaerentia remota). Predication secundum essentiam (1) shows a partial
identity between the real subject and predicate, which share some, but
not all, metaphysical component parts, and (2) does not require that
the form connoted by the predicate-term is directly present in the
item signied by the subject-term. Formal predication, on the contrary,
requires such a direct presence. Man is an animal and Socrates is
white are instances of formal predication. Unlike Wyclif , who applied
predication by essence to second intentions only, these later philoso-
phers thought that it held also when applied to rst intentions. So they
afrmed that it was possible to predicate of the universal-man (homo in
communi ), or of the abstract form of humanity (humanitas), the property
of being white, if at least one of its individuals was white. They were
careful, however, to use a substantival adjective in its neuter form as a
predicate-term, because only in this way could it appear that the form
signied by the predicate-term is not directly present in the subject, but
is indirectly attributed to it through its individuals. Therefore, according
to them (what is) singular is (what is) common (singulare est commune),
universal-man is (something) white (homo in communi est album) and
humanity is (something) running (humanitas est currens) are instances
of predication by essence (or remote inherence).
82
Formal predication
itself is in turn divided by Alyngton and Milverley into formal sub-
stantial and formal accidental predication, since (1) formal predication
necessarily demands the direct presence of a form in a substrate, and,
(2) according to them, this can occur in two different ways: either as
one of the inner constitutive element of the substrate (substantially),
or as one of its subsidiary properties (accidentally).
83
According to Alyngton, Milverley, and Tarteys the formal-and-
essential predication and the formal-and-accidental predication would
82
Cf. Alyngton, Litteralis sententia super Praedicamenta Aristotelis, cap. de substantia,
p. 289; Milverley, Compendium de quinque universalibus, p. 160; Tarteys, Problema correspondens
libello Porphyrii, London, Lambeth Palace, ms. 393, fols. 204(235)rv, and 209(240)rv;
Sharpe, Quaestio de universalibus, pp. 8991.
83
Cf. Alyngton, Litteralis sententia super Praedicamenta Aristotelis, cap. de subiecto et
praedicato, p. 245; cap. de substantia, pp. 274 and 289; Milverley, Compendium de quinque
universalibus, pp. 160161.
correspond to Aristotles essential and accidental predication respec-
tively. But, as a matter of fact, they agreed with Wyclif in regarding
predication by essence as more general than formal predication. As a
consequence, in their theories formal predication is a particular type of
the predication by essence. This means that they implicitly recognized
a single ontological pattern, founded on a partial identity, as the basis
of every kind of predicational statement. But in this way, the praedicatio
formalis essentialis and the praedicatio formalis accidentalis are very different
from their Aristotelian models, as they express degrees in identity as
well as the predication by essence.
In contrast, Penbygull and Whelpdale, who possibly belong to the
same generation as Sharpe, were closer to Wyclif s teaching as mani-
fested in the Purgans errores circa universalia in communi (136668), since
they divided predication into formal ( praedicatio formalis), by essence
(secundum essentiam), and causal (secundum causam). They describe formal
and by essence predications in the same way as the other Oxford Real-
ists, and causal predication in the same way as Wyclif.
84
According to
Penbygull and Whelpdale, there is causal predication when the item
signied by the predicate-term is not present in any way in the item
signied by the subject-term, but the real subject has been caused by
the real predicatedies est latio solis super terram is an example of this
kind of predication.
85
These two interpretative schemes of the nature and kinds of predica-
tion (that worked out by Alyngton, Milverley, Tarteys, and Sharpe, and
that adopted by Penbygull and Whelpdale) are ultimately grounded on
various notions of identity which are different from that common in the
thirteenth century, and xed by Walter Burley in his last commentary
on the Categories and treatise on universals. So it is not surprising that
Penbygull, Milverley, and Sharpe put forward new criteria for identity
and distinction.
84
Cf. Wyclif , Purgans errores circa universalia in communi, cap. 2, in Samuel H. Thomson,
A Lost Chapter of Wyclif s Summa de ente, Speculum 4 (1929), 339346, p. 342. As
a matter of fact the ms. Cambridge, Trinity College, B.16.2, used by Dziewicki for
his edition of the work (London 1909), lacks the second chapter and the rst section
of the third chapter. Samuel H. Thomson integrated the text on the basis of the ms.
Wien, sterreichische Nationalbibliothek, 4307.
85
Cf. Penbygull, De universalibus, pp. 186188; Whelpdale, Tractatus de universalibus,
pp. 190192.
Penbygull and Milverley
86
(1) distinguish between the notion of non-
identity and that of difference (or distinction); (2) deny that the notion
of difference implies that of non-identity; (3) afrm that the two notions
of difference and real identity are logically compatible, thus admitting
that (3.1) there are degrees in distinction, and (3.2) that the degrees of
distinction between two things can be read as the inverse measure of
their (partial) identity. In his turn, Sharpe seems to combine the theo-
ries of Duns Scotus , Wyclif , and Penbygull and Milverley themselves.
87

(1) Like Penbygull and Milverley, he considers identity and distinction
(or difference) as the two possible inverse measures of the coincidence
of the metaphysical components of two given entities. Moreover,
(2) he speaks of formal and real (or essential) identity, formal and real (or
essential) distinction (or difference), and (2.1) states that formal identity
is stronger than real (or essential) identity, since the former entails the
latter, while real difference is stronger than formal distinction, since the
latter is entailed by the former. (3) Finally, he admits degrees in formal
distinction (distinctio formalis consistit in gradibus), as he recognizes two
different types, the rst of which comes very close to that proposed by
Scotus in his Ordinatio, while the second is drawn from Wyclif s Tractatus
de universalibus (third sub-type).
Yet, among the Oxford Realists, the most original was Paul of
Venice. His philosophical theories (culminating in a metaphysics of
essences which states the ontological and epistemological primacy
of universals over any other kind of beings) are the nal and highest
result of the preceding realistic tradition of thought, since he (1) fully
developed the new form of realism started up by Wyclif ; (2) renewed
Burley s attacks against nominalistic views; (3) was open to inuences
from many other directions, as he (3.1) held in due consideration also
the positions of authors such as Albert the Great, Thomas Aquinas ,
and Giles of Rome, and (3.2) critically discussed the doctrines of the
main Nominalists of the fourteenth century, William Ockham, John
Buridan , and Marsilius of Inghen.
Paul of Venice divides predication into identical and formal, and
denes them in a slightly different way in relation to his sources. To
86
Cf. Penbygull, De universalibus, pp. 190191; Milverley, Compendium de quinque uni-
versalibus, p. 163.
87
Cf. Sharpe, Quaestio super universalia, pp. 9192; Quaestio super libros De Anima, q. 2:
utrum potentia intellectiva distinguatur ab essentia animae, Oxford, New College, ms.
238, fol. 236rv (transcription in Conti , Studio storico-critico, p. 216, note 18).
speak of identical predication it is sufcient that the form signied by
the subject-term of a (true) proposition and the form signied by the
predicate-term share at least one of their substrates of existence. This is
the case of propositions like man is (an) animal and the universal-man
is something white (homo in communi est album). One speaks of formal
predication in two cases: (1) when for the truth of the proposition it is
necessary that the form signied by the predicate-term is present in all
the substrates of existence of the form signied by the subject-term in
virtue of a formal principle (declared in the proposition itself ) which
is in its turn directly present in all the substrates of existence of the
form signied by the subject-term. This is the case of propositions like
man is formally (an) animal, Socrates qua man is an animal. (2) Or
when the predicate of the proposition is a term of second intention,
like species or genus. This is the case of propositions like man is a
species, animal is a genus.
88
As is evident, identical predication is extensionally dened, whereas
formal predication is intensionally dened, since formal predication
entails a relation modally determined between the subject-thing and the
predicate-thing. In fact, the formal predication presupposes that there
is a necessary connection between the subject-thing and the predicate-
thing of the given proposition. For this reason Paul denies that sentences
like (what is) singular is (what is) universal (singulare est universale), that
Wyclif and the other Oxford Realists acknowledged as true ones, are
true propositions. For Wyclif (and the other Oxford Realists) the sen-
tence at issue is an example of predication by essence, but for Paul it
is an example of formal predication. As a consequence, Paul rewrites
the preceding sentence in this form: (what is) singular is this universal
(singulare est hoc universale), where the presence of the demonstrative this
modies the kind of predication from formal to identical. So corrected
the sentence is true, since it signies that a certain item, in itself singular,
is the substrate of existence of an universal essence.
89
As a result, Paul builds up a mixed system, where the copula of the
standard philosophical sentences which he deals with can have a three-
fold value: it means (1) a partial identity between the real subject and
the real predicate, in the case of identical predication; (2) a necessary
88
Cf. Paul of Venice, Quaestio de universalibus, pp. 201202.
89
Cf. Paul of Venice, Quaestio de universalibus, pp. 206207. See Alessandro D. Conti ,
Il sosma di Paolo Veneto: Sortes in quantum homo est animal, in Sophisms in Medieval
Logic and Grammar, ed. Stephen Read (Dordrecht, 1993), pp. 304318.
link between forms, in the case of the rst type of formal predication;
and (3) that the real subject in virtue of itself necessarily is a member of
a given class of objects, labelled and referred-to by the predicate-term
of the proposition at issue, in the case of the second type of formal
predication, namely, when the predicate is a term of second intention.
Paul therefore inverts the terms of the question in relation to Wyclif s
approach, since Paul is attempting to reduce multiplicity to unity (the
passage is from many to one). What he wants to account for is the
way in which many different entities of a certain kind (namely, of an
incomplete and dependent mode of existence) can constitute one and
the same substance in number.
As the previous analysis has shown, Paul of Venices world consists
of nite beings (namely, things such as men and horses), which are
aggregates made up by a primary (or individual) substance and a host
of formal items (common natures and accidental forms, both universal
and singular) existing in it and by it. The components of the nite beings
are nothing but the categorial items themselves, taken together with
their own modes of being. All these items are real, namely, they are
beings (entia), in the sense that they are mind-independent. Yet, primary
substances only are existent (in other words, primary substances only
are actual beingsentia in actu). Individuation is what causes this passage
from being (esse) to existence, and from specic natures (universals) to
individuals. What is more, Paul of Venices metaphysics celebrates the
centrality of specic natures, which correspond to the ideas in the mind
of God, and in relation to which the actual exi stence of individuals
is functional, as they exist only as material substrates ( partes subiectivae)
of the natures themselves. Hence, within his system, notwithstanding
the real identity with individuals, common natures have a form of
being their own, which is independent of individuals. So that, if all
the individuals belonging to a certain natural species were annihilated,
their corresponding nature would keep on being, even though only
potentially, as a mere metaphysical possibility (esse indeterminatum et in
potentia).
90
In commenting on Categories 5, 2b56 (as we have seen, one
90
Expositio super Universalia Porphyrii, prooem., fol. 8va: Licet repugnet universali
habere determinatum esse absque suo singulari, non tamen repugnat habere indeter-
minatum esse absque illo; et consequenter non repugnat illi separari a suo singulari
tam realiter quam intentionaliter. Unde nulla rosa existente adhuc est aliquo modo
quidditas rosae, alioquin ea corrupta non amplius haberetur scientia de rosa, ex quo
scientia non habetur de non ente . . . Et si allegatur illud Aristotelis in Praedicamentis:
Destructis primis substantiis impossibile est aliquid horum remanere, demonstratis
of the most important passages of the work), Paul restates that same
thesis, and adds that a certain common nature would be annihilated if
(and only if ) all the individuals belonging to the corresponding natural
species were destroyed not only in relation to their actual existence, but
also in relation to their potential being.
91
Still, the potential being of
individuals is nothing but the essential being proper to universals, and
therefore the destruction of the individuals in relation to their potential
being is the destruction of universals themselves.
Concluding Remarks
As is well known, in the Categories Aristotle sketches the main features
of his metaphysics and semantics (which in turn are the ground of
these medieval disciplines), since he (1) sets out the basic elements of
the world (individual and universal substances, individual and universal
accidents) in their mutual relationships, and (2) shows their links to lan-
guage. As a consequence, that treatise is not a homogeneous text, but
a compound one. There are in it three main doctrinal bodies, which
can be split up into a few minor theories:
(1) The categorial doctrine properly so-called concerns the table of
categories (ch. 4), and the standard, internal structure of the (ten)
categorial elds (chs. 2, 3, and 5). Two other theories complete it:
(1.1) a semantic one, about homonymy, synonymy, and paronymy
(ch. 1), and (1.2) a metaphysical one, on substance, where the Stag-
irite states the epistemological and ontological primacy of individual
substances over any other kind of beings (ch. 5). This rst group of
universalibus, illud est sic intelligendum: destructis primis substantiis impossibile est
aliquod universale eorum remanere secundum esse determinatum. Cum quo stat quod
possint remanere secundum esse indeterminatum. Destructis ergo omnibus singularibus
alicuius universalis illud universale remanet, non actu sed potentia, non determinatum
sed indeterminatum; remanet quidem in potentia activa primi motoris, et in potentia
passiva primae materiae, et in potentia cognitiva animae. See also Quaestio de univer-
salibus, fol. 127rab.
91
Cf. Paul of Venice, Expositio super Praedicamenta Aristotelis, cap. de substantia, in Expositio
super Universalia Porphyrii et Artem Veterem Aristotelis, fol. 57vab: Destructo quolibet homine
coniunctim ita quod non sit aliquod individuum speciei humanae, non manet homo,
sed manet essentia hominis in suis causis. . . . Si tamen decerent omnes homines et actu
et potentia, non maneret species humana secundum esse neque secundum essentiam.
On this point see Conti , Esistenza e verit, pp. 119123, and 190192.
doctrines culminates in a threefold distinction: (1.3) between sub-
stance and accidents; (1.4) between individuals and universals; and
(1.5) between two different types of predication: (1.5.1) the being-
in-something-as-a-subject (or accidental predication) and (1.5.2)
the being-said-of-something-as-a-subject (or essential predication)
(chs. 2 and 5).
(2) The regional ontologies, about the foremost categorial elds after
substance (chs. 69). This second group of doctrines consists of phys-
ical and metaphysical theories of quantity (ch. 6), relation (ch. 7),
quality (ch. 8), action and passion (ch. 9). There Aristotle (2.1)
roughly describes (2.1.1) the items falling into these categorial elds,
(2.1.2) their own modes of being, and (2.1.3) their main properties,
and (2.2) supplies preliminary lists of them.
(3) The Postpraedicamenta (according to the terminology proper to the
Latin tradition), a bundle of several loosely related topics con-
cerning opposition, priority and posteriority, change, and having
(chs. 1015).
Medieval commentators usually did not ascribe much signicance to
this nal section of the tract, but chiey concentrated on the rst two
parts, and on the general questions of (1) the subject, (2) purpose, and
(3) importance of the work. The Categories is not only a composite, but
also a rather ambiguous text, as Aristotles Categories can be considered
as both a classication of things and a classication of the signs signi-
fying those things. Therefore, from late Antiquity onwards there were
many disputes about those three questions. Depending on the general
evaluation of the treatise, whether it primarily deals with world things
or their signs, it is customary to classify medieval authors as being
Realists or Nominalists. Furthermore, Nominalists (such as Peter of
John Olivi , Ockham, Buridan ) maintained that (1) in the world there
are two (substance and quality), or three (substance, quantity, and qual-
ity), supreme genera of beings only, but (2) we grasp and signify the
items falling into those two (or three) real categories by means of ten
semantically different kinds of terms. On the contrary, Realists (such
as Robert Kilwardby, Albert the Great, Thomas Aquinas , Simon of
Faversham, Henry of Ghent , Duns Scotus , Burley , Franciscus de Prato,
Jacobus de Placentia , Wyclif , Alyngton, Paul of Venice) held that the
ten Aristotelian categories are the supreme genera of beings, irreduc-
ible to one anothereven though there were some signicant differ-
ences among them in establishing the nature and ontological status of
those ten categories. Some of them, such as Robert Kilwardby, Henry
of Ghent, Simon of Faversham, the rst Burley, and Franciscus de
Prato judged only the items falling into the three absolute categories to
be things (res), and considered the remaining ones real aspects (respectus
reales) proper to the former. Some, such as Duns Scotus, the second
Burley, Wyclif, Alyngton, and Paul of Venice, claimed that all the ten
Aristotelian categories are distinct kinds of things. Some of them, such
as Albert the Great, Simon of Faversham, Burley, Alyngton, thought that
the ten categories are characterized by their own modes of being (modi
essendi ); some, such as Thomas Aquinas, believed that their constitu-
tive and distinctive principles are their own modes of being-predicated
(modi praedicandi ); and some, such as Duns Scotus and Paul of Venice,
afrmed that the ten real categories differ from each other in virtue of
their own essences.
Originally, the problem of universals was only one of the main ques-
tion internal to the theory of categories considered in the most proper
sense. As a consequence, even though from a purely logical point of
view the opinions on categories and universals are independent of each
other (one might support the thesis that categories chiey are really
distinct genera of things while rejecting the thesis that universals have
some existence of their own outside the mind, or vice versa), historically,
in the (later) Middle Ages, realism concerning categories was always
matched by a realist conception of universals, whereas nominalism on
the question of categories was always paralleled by a nominalist posi-
tion about universals.
If we consider the moderate realist view of universals, it is easy to
see that it is totally determined by (1) the general evaluation of the
categories, and (2) the main principles and theses stated by Aristotle in
the rst ve chapters of the book. When moderate Realists interpreted
the relation between universals and individuals in terms of identity, they
were trying to save the ontological primacy of individual substances
at the same time reading in a realist way (1) the nature and division
of predication, and (2) the twofold partition (into substantial and acci-
dental, individual and universal items) set in the second chapter of the
Categories. On the one hand, they assumed that being a universal was
equivalent to being-said-of-something-as-a-subject, on the other hand,
they considered that relation as a real relation between two different
kinds of beings. As a result, they were compelled to postulate a form of
identity between universals and individuals: universals are (metaphysi-
cal) parts of their individuals. Otherwise it would have been impossible
to distinguish the relation of being-said-of-something-as-a-subject (or
essential predication) from the relation of being-in-something-as-a-
subject (or accidental predication, or inherence). Both universal sub-
stances and accidents are somehow present in individual substances
and neither can exist apart from individual substances, but universals
are parts of individuals and accidents are not.
92
On the other side,
universals and individuals cannot be entirely identical, since there is
not a complete transitivity in predication between them. To elaborate
new notions of identity and distinction was therefore a necessity for
the Realists of the fourteenth century, as the logical machinery they
had at their disposal was insufcient for their purposes. Because of
Ockhams critique (1) of the moderate realist view, and (2) formal dis-
tinction, almost all the Realists of the later Middle Ages became fairly
unsatised with Henry of Ghent s and Duns Scotus s formulations of
distinctions midway between the real distinction and the distinction of
reason. So they tried to improve the theory of universals by modify-
ing (1) the standard Aristotelian analysis of predication, and (2) the
notion of formal distinction. In fact, the only other possible way of
overcoming Ockhams argumentations against realism was to assume,
like Burley did, that universals and individuals are really distinct. A
choice which entailed a change from an Aristotelian metaphysics to a
Platonic oneas we have seen. Indeed, this second way also led to
a paradoxical result: the partial dissolution of the traditional doctrine
of categories. Within the new metaphysical system that Oxford Realists
built up, universals and individuals, essential and accidental predication
are far too different from their Aristotelian patterns. As Paul of Venices
explicit denial of the existential primacy of individual substances shows,
the metaphysics proper to the Oxford Realists substantially is a Platonic
metaphysics, where universal essences, and not individual substances,
are the main kind of being. According to the moderate Realists of the
second half of the thirteenth century, the actual existence of at least one
individual was necessary in order to guarantee the existence in potentia
of the corresponding universal. In Paul of Venices view, the existence
of a universal essence is the necessary condition for the existence of
92
Cf. Aristotle, Categories 2, 1a2425, where the Stagirite posits that something is
in something as a subject if, not being in anything as a part, it cannot exist apart from
what it is in.
individuals. In other words, for moderate Realists universals were types
and individuals were the tokens which instantiated those types; about
one hundred and fty years later, for Paul of Venice universals were
projects and individuals were their possible fullments. There cannot
be types without tokens, but no project, as such, needs to be fullled.
BIBLIOGRAPHY
Abed, Shukri , Language, in History of Islamic Philosophy, eds. S. Nasr and O. Leaman ,
London, 1996, pp. 90413.
Ackrill , J. L., Aristotles Categories and De interpretatione, trans. J. L. Ackrill, Oxford,
1963.
Adams , Marilyn McCord, Ockham on Identity and Distinction, Franciscan Studies
36 (1976), 574.
, Universals in the Early Fourteenth Century, in The Cambridge History of Later
Medieval Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism;
11001600, eds. Norman Kretzmann , Anthony Kenny , and Jan Pinborg, Cambridge,
1982, pp. 41139.
, William Ockham, Indiana, 1987.
Albertus Magnus, Analytica posteriora, in Opera omnia, ed. Auguste Borgnet, Paris, 1890.
, Analytica priora, in Opera omnia, ed. Auguste Borgnet, Paris, 1890.
, De anima, ed. Clemens Stroick, Mnster, 1968.
, De bono, eds. Heinrich Khle et al., Mnster, 1951.
, De causis et processu universitatis, ed. Winfried Fauser, Mnster, 1993.
, De natura et origine animae, ed. Bernhard Geyer, Mnster, 1955.
, De praedicamentis, eds. Carlos Steel, Silvia Donati and Manuel Santos Noya
(Mnster, 2008).
, Ethica, in Opera omnia, ed. Auguste Borgnet, Paris, 1891.
, Liber de Praedicabilibus, in Opera omnia, ed. A. Borgnet, Paris, 1890.
, Liber de praedicamentis, in Opera omnia, ed. A. Borgnet, Paris, 1890.
, Metaphysica, ed. Bernhard Geyer, Mnster, 1960.
, Peri hermeneias, in Opera omnia, ed. Auguste Borgnet, Paris, 1890.
, Super Dionysium De divinis nominibus, ed. Paul Simon, Mnster, 1972.
, Super libros sententiarum, in Opera omnia, ed. Auguste Borgnet, Paris, 1893.
, Super Porphyrium De V universalibus, ed. Manuel Santos Noya , Mnster, 2004.
al-Frb, Al-Maniq inda l-Frab, ed. R. Al-am, vol. 3 (Beirut, 1986); trans. as
al-Frabs Paraphrase of The Categories of Aristotle, Islamic Quarterly (1957),
15897.
, Book of the concordance in view of the two sages, the divine Plato and Aristotle, eds. &
French transl. F. W. Najjar & D. Mallet, Damascus, 1999.
, Greater Commentary on De interpretatione, eds. W. Kutsch and S. Morrow, Beirut,
1960, trans. F. W. Zimmermann, London, 1981.
, Kitb al-Alfz al-Mustamala f l-mantiq, ed. M. Mahdi, Beirut, 1968.
, Kitb al-urf, ed. M. Mahdi, Beirut, 1970.
Alyngton, Robert , Litteralis sententia super Praedicamenta Aristotelis, in Alessandro D. Conti ,
Linguaggio e realt nel commento alle Categorie di Robert Alyngton, Documenti e
studi sulla tradizione losoca medievale 4 (1993), 179306.
Amerini , Fabrizio, La dottrina della signicatio di Francesco da Prato, O.P. XIV secolo.
Una critica tomista a Guglielmo di Ockham, Documenti e studi sulla tradizione losoca
medievale 11 (2000), 375408.
, La quaestio Utrum subiectum in logica sit ens rationis e la sua attribuzione a
Francesco da Prato. Note sulla vita e gli scritti del domenicano Francesco da Prato
XIV secolo, Memorie Domenicane n.s. 30 (1999), 147217.
, Realism and Intentionality: Hervaeus Natalis, Peter Auriol, and William Ockham
in Discussion, in Philosophical Debates at the University of Paris in the First Quarter of the
Fourteenth Century, eds. Theo Kobusch and Stephen F. Brown , Paris, forthcoming.
412 bibliography
, What is Real. A Reply to Ockhams Ontological Program Vivarium 43 (2005),
187212.
, La logica di Francesco da Prato. Con ledizione critica della Loyca e del Tractatus de
voce univoca, Florence, 2005.
Andrews , Robert, Peter of Auvergne on Denomination, in Meaning and Inference in
Medieval Philosophy: Studies in Memory of Jan Pinborg , ed. Norman Kretzmann, Dor-
drecht, 1988.
, Petrus de Alvernia, Quaestiones super Praedicamentis: An Edition, Cahiers de lInstitut
du Moyen ge grec et latin 55 (1987), 384.
, Question Commentaries on the Categories in the Thirteenth Century, Medioevo
26 (2001), 265326.
, Resoluble, Exponible, and Ofciable Terms in the Sophistria of Petrus Olai,
ms. Uppsala C 599, in Sophisms in Medieval Logic and Grammar: Acts of the Ninth European
Symposium for Medieval Logic and Semantics, held at St. Andrews , June 1990, ed. Stephen
Read, Dordrecht, 1993.
, Thomas of Erfurt on the Categories in Philosophy, in Was ist Philosophie im Mit-
telalter? Akten des X. Internationalen Kongresses fr mittelalterliche Philosophie der
Socit Internationale pour lEtude de la Philosophie Mdivale, 25. bis 30. August
1997 in Erfurt, eds. Jan A. Aertsen and Andreas Speer (Miscellanea Mediaevalia
26) Berlin, 1998, pp. 8018.
, ed. The Defensorium Ockham, Essays in Honor of Fr. Gedeon Gl, O.F.M. on his Eightieth
Birthday: Franciscan Philosophy and Theology II, Franciscan Studies 54 (1995), 10021.
, Peter of Auvergne s Commentary on Aristotles Categories: Edition, Translation, and Analysis,
Dissertation: Cornell University, 1988.
Annas , Julia, Aristotles Metaphysics: Books M and N, Oxford, 1976.
Anonymus Escorial, Quaestiones super Metaphysicam, partim ed. P. Porro , in Le Quaestiones
super Metaphysicam attribuite a Enrico di Gand.
Anonymus Matritensis, Quaestiones super librum Praedicamentorum, ed. R. Andrews, in
Cahiers de lInstitut du moyen-ge grec et latin 56 (1988), 11792.
Anton , J. P., The Aristotelian Doctrine of Homonymy in the Categories and its Platonic
Antecedents, Journal of the History of Philosophy 6 (1968), 315326.
Aristoteles, Analytica Priora et Posteriora, eds. W. D. Ross and Laurent Minio-Paluello,
Oxford, 1964.
, Categoriae, ed. Laurent Minio-Paluello, Oxford, 1949.
, De anima, ed. W. D. Ross, Oxford, 1956.
Aristotle, Aristotles Categories and De Interpretatione, trans. J. L. Ackrill , Oxford, 1963.
, Metaphysica, ed. W. Jaeger, Oxford, 1924.
, On Interpretation, ed. Laurent Minio-Paluello, Oxford, 1949.
, Physics 1.1, ed. Henri Carteron, Paris, 1961.
, Prior Analytics, ed. W. D. Ross, Oxford, 1964.
Ashworth, E. J. and Paul V. Spade , Logic in Late Medieval Oxford, in The History
of the University of Oxford, eds. Jeremy I. Catto and Ralph Evans, vol. 2, Oxford,
1992, pp. 5062.
, Lanalogie de ltre et les homonymes: Catgories, 1 dans le Guide de ltudiant,
in Lenseignement de la philosophie au XIII
e
sicle. Autour du Guide de ltudiant du ms. Ripoll
109, d. par Claude Laeur and Joanne Carrier, Turnhout, 1997, pp. 28195.
, Language and Logic, in The Cambridge Companion to Medieval Philosophy, New
York, 2003, pp. 7396.
Asztalos , Monika, Boethius as a Transmitter of Greek Logic to the Latin West,
Harvard Studies in Classical Philology 95 (1993), 367407.
Augustine , Confessions, ed. Pierre de Labriolle, Paris, 1950.
Averroes, In Aristotelis librum II Metaphysicorum commentarius, Darms, 1966.
, Quaestiones in libros logicae, in Aristotelis Opera cum Averrois Commentariis, Venice
156274; repr. Frankfurt am Main, 1962.
bibliography 413
, Tahfut al-Tahfut, trans. Simon van den Bergh, London, 1954.
Avicenna, Al-Maqlt, eds. G. Anawati et al., Cairo, 1959.
, A-ir, ed. A. Badaw, Cairo, 1966.
, Fi Nafs, ed. G. Anawati, Cairo, 1962.
, Liber de philosophia prima sive scientia divina, ed. S. Van Riet, 3 vols., Leuven,
19771983.
, Logyca, Venice, 1508.
Bck , Allan, Aristotles Discovery of First Principles, in From Puzzles to Principles, ed.
May Sim, Maryland, 1999, pp. 16382.
, Avicenna s Ontological Pentagon, The Journal of Neoplatonic Studies 7 (1999),
87109.
, Review of Averroes Middle Commentary on Aristotles Poetics, trans. Charles But-
terworth, Canadian Philosophical Reviews 10 (1990), 92101.
, The Ordinary Language Approach in Traditional Logic, in Argumentationstheorie,
ed. K. Jacobi, Leiden, 1993, pp. 50730.
, The Triplex Status and its Justication, Studies in the History of Logic, eds.
I. Angelelli & M. Cerrezo, Berlin, 1996.
Badaw, A., New Philosophical Texts Lost in Greek and Preserved in Islamic Transla-
tions, in Islamic Philosophical Theology, ed. P. Morewedge, Albany, 1979, pp. 313.
, La transmission de la philosophie grecque au monde arabe, Paris, 1968.
Bakker , Paul J. H. M., Aristotelian Metaphysics and Eucharistic Theology: John
Buridan and Marsilius of Inghen on the Ontological Status of Accidental Being.
In The Metaphysics and Natural Philosophy of John Buridan, eds. J. M. M. H. Thijssen
and Jack Zupko , Leiden, 2001, pp. 24764.
Beuchot , Mauricio, La naturaleza de la lgica y su conexin con la ontologa en
Alberto Magno, Dinoia 33 (1987), 23546.
Black , Deborah L., al-Frb, in S. H. Nasr and O. Leaman , eds. History of Islamic
philosophy, London, 1996, pp. 17897.
Bloch, Marc , The Historians Craft, trans. P. Putnam, New York, 1953.
Bods , R., Introduction, in Aristote, Catgories. Texte tabli et traduit par R. Bods,
Paris, 2001, pp. XXIVCX.
, Linuence historique du stocisme sur linterprtation de luvre philosophique
dAristote, Revue des Sciences philosophiques et thologiques 79 (1995), 553586.
Boethius , De divisione liber, ed. J. Magee, Leiden, 1998.
, In Categorias Aristotelis libri quatuor, ed. J.-P. Migne, PL 64, Paris, 1847; reprint
Turnhout, 1979.
, In Isagogen Porphyrii Commenta, eds. G. Schepss and S. Brandt, Lipsiae, 1906.
Bonaventura , Commentarius in quattuor libros Sententiarum Petri Lombardi, in Opera omnia IIV,
Collegium S. Bonaventurae, 18821902.
Bonin , Thrse, Creation as Emanation. The Origin of Diversity in Albertus Magnuss On the
Causes and the Procession of the Universe, Notre Dame, 2001.
Bos, E. P. and A. C. van der Helm , The Division of Being over the Categories
According to Albertus Magnus, Thomas Aquinas and Duns Scotus , in John Duns
Scotus: Renewal of Philosophy: Acts of the Third Symposium Organized by the Dutch Society for
Medieval Philosophy Medium Aevum May 23 and 24, 1996, ed. E. P. Bos, ELEMENTA:
Schriften zur Philosophie und ihrer Problemgeschichte 72 Amsterdam, Rodopi,
1998, pp. 18396.
Bos, E. P., John Versors Albertism in his Commentaries on Porphyry and the Categories,
in Chemins de la pense mdivale. tudes offertes Znon Kaluza, ed. Paul J. J. M. Bakker ,
Textes et tudes du Moyen ge, 20 Turnhout, 2002, pp. 4778.
Boulnois , O., tre, luire et concevoir. Note sur la gense et la structure de la concep-
tion scotiste de lesse obiective. Collectanea franciscana 60 (1990), 11735.
Brabant, Siger, crits de logique, de morale et de physique, ed. Bernardo Bazn, Philosophes
Mdivaux xiv, Louvain-la-neuve, 1974, pp. 78.
414 bibliography
Breton , Stanislas, La dduction thomiste des catgories, Revue philosophique de Louvain
60 (1962), 532.
Brito, Radulphus , Quaestiones super librum Porphyrii, in Quaestiones subtilissimae magistri
Rodulphi Britonis super arte veteri, Venetiis, 1499.
, Quaestiones super Praedicamenta, in Quaestiones subtilissimae magistri Rodulphi Britonis
super arte veteri, Venetiis, 1499.
Brown , Stephen F., Walter Burley s Middle Commentary on Aristotles Perihermeneias,
Franciscan Studies 33 (1973), 42134.
Brunschwig , J., Aristote de Stagire. LOrganon: tradition grecque, in Dictionnaire des
Philosophes Antiques, vol. 1, Paris, 1989, pp. 482502.
Buridan , John, Jan Buridan, Kommentarz do Isagogi Porriusza [Introduction in Polish
and Edition of the Questiones in Isagogen Porphyrii by Ryszard Tatarzyski], Przeglad
Tomistyczny 2 (1986), 11195.
, In Metaphysicen Aristotelis Questiones argutissimae, Paris, 1588 actually 1518; reprint,
Kommentar zur Aristotelischen Metaphysik, Frankfurt am Main, 1964.
, Johannes Buridanus, Summulae: In Praedicamenta, Introduction, critical edition and
notes by E. P. Bos, Artistarium 103, Nijmegen, 1994.
, John Buridan: Sumulae de Dialectica: An Annotated Translation with a Philo-
sophical Introduction by Gyula Klima, (Yale Library of Medieval Philosophy)
Connecticut, 2001.
, Quaestiones in Praedicamenta, ed. J. Schneider, Mnchen, 1983.
Burnet , J., Early Greek Philosophy, 3rd ed, London, 1920.
Burnyeat , M., A Map of Metaphysics Zeta, Pittsburgh, 2001.
Caujolle-Zaslawsky , Franoise, Les relatifs dans les Catgories, in Concepts et Catgories
dans la pense antique, ed. Pierre Aubenque, Paris, 1980, pp. 16795.
Cesalli , Laurent, Le pan-propositionnalisme de Jean Wyclif , Vivarium 43 (2005),
12455.
, Le ralisme propositionnel de Walter Burley , Archives dhistoire doctrinale et littraire
du Moyen Age 68 (2001), 155221.
Chadwick , Henry, Boethius. The Consolations of Music, Logic, Theology and Philosophy,
Oxford, 1981.
Charles , David, Aristotle on Names and their Signication, in Language, ed. S. Ever-
son, Cambridge 1994.
Chase , Michael, Did Porphyry write a commentary on Aristotles Posterior Analytics?
Albertus Magnus, Frb and Porphyry on per se predication, in P. Adamson, ed.
Classical Arabic Philosophy: Sources and Reception, London, forthcoming.
, ed. & trans., Simplicius : On Aristotles Categories 14, Ithaca, 2003.
, tudes sur le commentaire de Porphyre sur les Catgories dAristote adress Gdalios. Diss.
cole pratique des Hautes tudes, Paris, 2000.
Chvarri , Eladio, El orden de los escritos lgicos de Aristteles segn San Alberto
Magno, Estudios loscos 9 (1960), 97134.
Chenu , M.-D., La thologie comme science au XII
e
sicle, 3rd ed. Paris, 1969.
Cohen , Marc S., Aristotles Metaphysics, The Stanford Encyclopedia of Philosophy Winter
2003 Edition, ed. Edward N. Zalta, URL = http://plato.stanford.edu/archives/
win2003/entries/aristotle-metaphysics.
Conti , Alessandro D., Analogy and Formal Distinction: on the Logical Basis of Wyclif s
Metaphysics, Medieval Philosophy and Theology 6 (1997), 13365.
, Boezio commentatore e interprete delle Categorie di Aristotele, in Scritti in
onore di Girolamo Arnaldi offerti dalla Scuola Nazionale di Studi Medievali, eds. A. Degrandi
et al., Rome, 2001, pp. 77102.
, Essenza ed essere nel pensiero della tarda scolastica Burley , Wyclif , Paolo
Veneto, Medioevo 15 (1989), 23567.
, Il commento di Giacomo da Piacenza allIsagoge e alle Categorie, in L insegna-
mento della logica a Bologna nel XIV secolo, eds. Dino Buzzetti, Maurizio Ferriani, and
Andrea Tabarroni, Bologna, 1992, pp. 44160.
bibliography 415
, Il problema della conoscibilit del singolare nella gnoseologia di Paolo Veneto,
Bollettino dellIstituto Storico Italiano per il Medio Evo e Archivio muratoriano 98 (1992),
32382.
, Il sosma di Paolo Veneto: Sortes in quantum homo est animal, in Sophisms in Medieval
Logic and Grammar, ed. Stephen Read, Dordrecht, 1993, pp. 30418.
, Johannes Sharpes Ontology and Semantics: Oxford, Realism Revisited,
Vivarium 43 (2005), 15686.
, La conoscenza del singolare in Walter Burley , Documenti e studi sulla tradizione
losoca medievale 15 (2004), 51740.
, La teoria degli ad aliquid di Boezio: osservazioni sulla ter minologia, in Atti
del Congresso Internazionale di Storia della Logica. San Gimignano, 48 dicembre 1982, eds.
V. M. Abrusci, E. Casari, and M. Mugnai, Bologna, 1983, pp. 247262.
, La teoria della relazione nei commentatori neoplatonici delle Categorie di
Aristotele, Rivista critica di storia della losoa 38 (1983), 25983.
, Linguaggio e realt nel commento alle Categorie di Robert Alyngton, Do cumenti
e studi sulla tradizione losoca medievale 4 (1993), 179306.
, Logica intensionale e metasica dell essenza in John Wy cli f, Bollettino dellIstituto
Storico Italiano per il Medioevo e Archivio muratoriano 99 (1993), 159219.
, Ontology in Walter Burley s Last Commentary on the Ars Vetus, Franciscan
Studies 50 (1990), 12176.
, Paul of Venice on Individuation, Recherches de Thologie et Philosophie mdivales
65 (1998), 10732.
, Paul of Venices Theory of Divine Ideas and its Sources, Documenti e studi sulla
tradizione losoca medieva le 14 (2003), 40748.
, Second Intentions in the Late Middle Ages, in Medieval Analyses in Language and
Cognition, eds. Sten Ebbesen, Russell L. Friedman, Copenhagen, 1999, pp. 45370.
, Signicato e verit in Walter Burley , Documenti e studi sulla tradizione losoca
medievale 11 (2000), 31750.
, Studio storico-critico, in Johannes Sharpe, Quaestio super universalia, ed. Ales-
sandro D. Conti, Florence, 1990, pp. 21138, and 295336.
, Sviluppi e applicazioni della distinzione formale scotista ad Oxford, sul nire
del XIV secolo, in Via Scoti, Methodologica ad mentem Joannis Duns Scoti, ed. Leonardo
Sileo, 2 vols., Rome, 1995, vol. 1, pp. 31936.
, Teoria degli univer sali e teoria della predicazione nel trattato De universalibus
di William Pen bygull: discussione e difesa della posizione di Wyclif , Medioevo 8
(1982), 137203.
, Thomas Suttons Commentary on the Categories according to MS Oxford,
Merton College 289, in The Rise of British Logic, ed. P. Osmund Lewry, Toronto,
1983, pp. 173213.
, Wyclif s Logic and Metaphysics, in A Companion to John Wyclif, ed. Ian C. Levy,
Leiden, 2006, pp. 67125.
, Esistenza e verit: Forme e strutture del reale in Paolo Veneto e nel pensiero losoco del
tardo medioevo, Rome 1996.
Copleston, Frederick , A History of Medieval Philosophy New York, 1972.
Courcelle , Pierre, Le Connais-toi toi-mme chez les no-platoniciens grecs, in Le
noplatonisme. Actes du Colloque de Royaumont, 913 juin 1969, Paris, 1971, pp.
15366.
Cunningham , Francis A., Speculative Grammar in St. Thomas Aquinas, Laval
Thologique et Philosophique 17 (1961), 7686.
DAncona , Cristina, Greek into Arabic: Neoplatonism in translation, in P. Adamson
and R. C. Taylor, eds. The Cambridge Companion to Arabic Philosophy, Cambridge, 2005,
pp. 1031.
Daiber , H., Review of Peters Aristoteles Arabus, Gnomon 42 (1970).
de Courtraco, Siger, Expositio in librum Aristotelis Perihermeneias, in Zeger van Kortrijk, Com-
mentator van Perihermeneias, ed. C. Verhaak, Brussels, 1964.
416 bibliography
, Summa modorum signicandi, Sophismata, New edition by Jan Pinborg , Sophisma IV:
Album potest esse nigrum Amsterdam Studies in the Theory and History of Linguistic
Science. Series III, Studies in the History of Linguistics 14, Amsterdam, 1977.
de Libera , Alain, Questions de ralisme. Sur deux arguments antiockhamistes de John
Sharpe, Revue de metaphysique et de morale 97 (1992), 83110.
, Thorie des universaux et ralisme logique chez Albert le Grand, Revue des
sciences philosophiques et thologiques 65 (1981), 5573.
, Albert le Grand et la philosophie, Paris, 1990.
, La querelle des universaux. De Platon la n du Moyen ge Paris, 1996.
de Mottoni , Barbara Faes, Il commento di Pietro di S. Amore agli Analitici Posteriori
ritrovato?, Studi Medievali 27 (1986), 383405.
de Rijk, L. M., ed. Peter of Spain: Tractatus called afterwards Summulae Logicales, Assen,
1972.
, Logica modernorum: A Contribution to the History of Early Terminist Logic, vol. 2, part
I, Assen, 1967.
Decorte , J., Avicennas Ontology of Relation. A Source of Inspiration to Henry of
Ghent , in Avicenna and his Heritage. Acts of the International Colloquium, Leuven Louvain-
la-Neuve, September 8September 11, 1999, eds. J. Janssens and D. De Smet, Leuven,
2002, pp. 197224.
, Giles of Rome and Henri de Gand on the Reality of a Real Relation, Documenti
e studi sulla tradizione losoca medievale 7 (1996), 183211.
, Relatio as Modus Essendi: The Origins of Henry of Ghent s Denition of Rela-
tion, International Journal of Philosophical Studies 10 (2002), 30936.
Defensorium Ockham: An Edition, Cahiers de linstitut du moyen-ge grec et latin 71 (2000),
189273.
Dekker , Dirk-Jan, John Buridan s Treatise De dependentiis, diversitatibus et convenientiis:
An Edition, Vivarium 42 (2004), 10949.
Delisle , L., Manuscrits divers acquis par la Bibliothque Nationale en 1876, 1877 et
1878, in Mlanges de Palographie et de Bibliographie, Paris, 1880, pp. 46971.
Denie , Heinrich and Emile Chatelain, eds. Chartularium Universitatis Pariensis I, Paris,
18891897.
Dillon , John, Iamblichus Noera theria of Aristotles Categories, Syllecta Classica 8 (1997)
6577.
Donati , Silvia, Per lo studio dei commenti alla Fisica del XIII secolo. I: Commenti di
probabile origine inglese degli anni 12501270 ca., Documenti e Studi sulla Tradizione
Filosoca Medievale II 2 (1991), 3845.
Druart, Thrse-Ann , Averroes on Gods Knowledge of Being qua Being, in Studies
in Thomistic Theology, ed. P. Lockey, Houston, 1996.
Duerlinger , James, Predication and Inherence in Aristotles Categories, Phronesis 15
(1970), 179203.
Duhem , P., Le Systme du monde, 10 vols., Paris, 191359.
Dumont , Stephen D., Duns Scotuss Parisian Question on the Formal Distinction,
Vivarium 43 (2005), 762.
Dunlop , D. M., Averroes on the Modality of Propositions, Islamic Studies, 1 (1962).
Ebbesen, Sten, Albert the Greats Companion to the Organon, in Albert der Grosse. Seine
Zeit, seine Werk, seine Wirkung, ed. Albert Zimmermann, Berlin, 1981, pp. 89103.
, Ancient Scholastic Logic as the Source of Medieval Scholastic Logic. in
The Cambridge History of Later Medieval Philosophy: From the Rediscovery of Aristotle to the
Disintegration of Scholasticism; 11001600, eds. Norman Kretzmann , Anthony Kenny ,
and Jan Pinborg, Cambridge, 1982, pp. 10127.
, Anonymous DOrvillensis Commentary on Aristotles Categories, Cahiers de
lInstitut du Moyen Age grec et latin 70 (1999), 229423.
bibliography 417
, Boethius as an Aristotelian Commentator, in Aristoteles: Werk und Wirkung, ed.
Jrgen Wiesner, 2 vols., Berlin, 1987, pp. 286311.
, Les Catgories au moyen ge et au dbut de la modernit, in Les Catgories et
leur histoire, eds. O. Bruun and L. Corti, Paris, 2005, pp. 24574.
Emden , A. B., A Biographical Register of the University of Oxford, to A.D. 1500, 3 vols.,
Oxford, 195759.
Endress, Gerhard , Die wissenschaftliche Litteratur, in Grundrisse der Arabischen Philologie,
ed. H. Gtje , vol. 2, Wiesbaden, 1987, pp. 40234.
Euclid , The Thirteen Books of the Elements, trans. Thomas L. Heath, 2nd ed. New York,
1956.
Eustratus , In primum Aristotelis moralium ad Nicomachum, eds. H. Paul and F. Mercken,
Leiden, 1973.
Evans, J. D. G. , Aristotles Concept of Dialectic, Cambridge, 1977.
Fischer , Wolfdietrich, A Grammar of Classical Arabic, 3rd ed., trans. J. Rodgers, New
Haven, 2002.
Franceschini , Ezio, Giovanni Pago: Le sue Rationes super Praedicamenta Aristotelis e la loro
posizione nel movimento aristotelico del secolo XIII, Sophia 2 (1934), 17282.
Franciscus de Prato, Logica, in Fabrizio Amerini , La logica di Francesco da Prato. Con
ledizione critica della Loyca e del Tractatus de voce univoca, Florence, 2005.
, Tractatus de sex transcendentibus, ed. Burkhard Mojsisch, in Bochumer Philosophisches
Jahrbuch fr Antike und Mittelalter, 5 (2000), 177217.
Frede, Michael, Categories in Aristotle, Studies in Aristotle, ed. Dominic J. OMeara
Washington, D.C., 1981, pp. 124.
, The Title, Unity, and Authenticity of the Aristotelian Categories, in Essays in
Ancient Philosophy, ed. M. Frede, Oxford, 1987, pp. 1128.
Friedrich , Paul, Polytropy, in Beyond Metaphor, ed. J. Fernandez, Stanford, 1991, pp.
1955.
Galluzzo, G., Met. Z 13 in the Contemporary Debate and in Aquinas s Interpretation,
Documenti e studi sulla tradizione losoca medievale, 14 (2003), 159226.
Gaskin, Richard , trans. & comm., Simplicius , On Aristotles Categories 915, Ithaca, 2000.
Gtje , Helmut, Simplikios in der arabischen berlieferung, Der Islam 59 (1982),
631.
Gellius , Noctes Atticae 20,5,112, in Aristoteles, Privatorum Scriptorum Fragmenta, ed. M. Plezia,
Leipsig, 1977.
Giles of Rome, Expositio super libros Elenchorum Aristotelis, Venetiis, 1500; reprint Frankfurt
am Main, 1968.
, Super librum I Sententiarum, Ve netiis, 1521.
Gilson , tienne, Les sources grco-arabes de lAugustinisme avicennisant, Archives
dhistoire doctrinale et littraire du Moyen Age 4 (192930); reprint, Paris, 1986.
, Quasi Denitio Substantiae, in St. Thomas Aquinas 12741974. Commemorative
Studies, eds. A. Maurer et al., Toronto, 1974, vol. 1, pp. 111129.
, Being and Some Philosophers, Toronto, 1952.
, History of Christian Philosophy in the Middle Ages, New York, 1955.
Glorieux , P., Pierre de Saint-Amour. Pierre Bellison. Neveu de Guillaume de Saint-
Amour. Ds 1250 chanoine de Saint-Vincent de Mcon. Recteur de lUniversit en
1281. Mort le 1er juin 1295; in La Facult des artes et ses matres au XIII
e
sicle, ed.
P. Glorieux, Paris, 1971.
Gohlman , W., ed. and trans., The Life of Ibn Sina, Albany, 1974.
Gorman , Michael and Jonathan J. Sanford, eds. Categories: Historical and Systematic Essays,
Washington, DC, 2004.
Gracia , J. E. and Lloyd Newton , Medieval Theories of Categories, Stanford Encyclopedia
of Philosophy, 14 April 2006, <http://plato.stanford.edu/entries/medieval-categories>
8 August 2006.
418 bibliography
Greimas , J., Dictionnaire de lancien franais jusquau milieu du XIV
e
sicle, Paris, 1987.
Grosseteste, Robertus, De veritate propositionis, ed. L. Baur, Aschendorff, 1912.
Grotz, Stephan, trans., De Modis Signicandis, Thomas von Erfurt, Abhandlung ber die bedeut-
samen Verhaltensweisen der Sprache Tractatus De modis signicandi, Amsterdam, 1998.
Gutas, Dimitri, Avicenna and the Aristotelian Tradition, Leiden, 1988.
, Greek Thought, Arabic Culture, London, 1998.
Guthrie, W. K. C. , A History of Greek Philosophy, vol. 1, Cambridge, 1962.
, The Presocratic Philosophers, 2nd ed. Cambridge, 1983.
Hadot , Ilsetraut, Deuxime point du premier schma introductif: la division nopla-
tonicienne des crits dAristote, in Simplicius , Commentaire sur les Catgories. Traduc-
tion commente sous la direction de Ilsetraut Hadot, fascicule I, Leiden, 1990, pp.
6393.
, La vie et luvre de Simplicius daprs des sources grecques et arabes, in
I. Hadot, ed. 1987, 339.
, ed. Simplicius , sa vie, son uvre, sa survie; actes du colloque international de Paris 28 sept.
1 oct. 1985, Berlin, 1987.
, Simplicius , Commentaire sur les Catgories: traduction commente, I: Introduction. Premire
partie: pp. 19, 3 Kalbeisch, sous la dir. de Ilsetraut Hadot; trad. par Philippe
Hoffmann ; avec la collab. de Ilsetraut Hadot & Pierre Hadot; commentaires et
notes la trad. I. Hadot; avec des appendices de P. Hadot & Jean-Pierre Mah,
Leiden, 1990.
, Simplicius , Commentaire sur les Catgories: traduction commente, III: Prambule aux
Catgories. Commentaire au premier chapitre des Catgories: pp. 2140 Kalbeisch,
sous la dir. de Ilsetraut Hadot; trad. de Philippe Hoffmann ; avec la collab. de Ilse-
traut Hadot, Pierre Hadot & Concetta Luna ; commentaires & notes la trad. par
C. Luna, Leiden, 1990.
Hadot, Pierre, Philosophy as a way of life: spiritual exercises from Socrates to Foucault., ed.
Arnold Davidson, trans. John Michael Chase, Oxford, 1995.
, Porphyre et Victorinus , 2 vols., Paris, 1968.
, What is ancient philosophy?, trans. John Michael Chase Cambridge, Mass, 2002.
Hamesse , Jacqueline, Auctoritates Aristotelis, Senecae, Boethii, Platonis, Apulei et quorundam
aliorum, Louvain, 1974.
Hankey , Wayne, Aquinas and the Platonists, in The Platonic Tradition in the Middle
Ages: A Doxographic Approach, eds. Stephen Gersh and Maarten J. F. M. Hoenen, with
the assistance of Pieter Th. van Wingerden, Berlin, 2002, pp. 279324.
, Philosophy as a Way of Life for Christians ? Iamblichan and Porphyrian Reec-
tions on Religion, Virtue, and Philosophy in Thomas Aquinas , Laval thologique et
philosophique (2003), 193224.
, Review of: Simplicius : On Aristotles Physics 5, translated by J. O. Urmson, notes
by Peter Lautner. The Ancient Commentators on Aristotle. Ithaca, New York: Cornell
University Press, 1997, Bryn Mawr Classical Review, 9.6 (1998), 5858.
, Thomas Neoplatonic Histories: His following of Simplicius, Dionysius 20
(2002), 15378.
Hasse , Dag, Avicenna s De Anima in the Latin West London, 2000.
Henle , R. J. Saint Thomas and Platonism, Hague, 1956.
Henninger , Mark, Relations: Medieval Theories 12501325, Oxford, 1989.
Henry of Ghent , Quodlibeta, ed. J. Badius Ascensius, Paris, 1518, reprint Leuven,
1961.
, Quodlibet VII, in Opera omnia 11, ed. G. A. Wilson, Leuven, 1991.
, Quodlibet IX, in Opera omnia 13, ed. R. Macken, Leuven, 1983.
, Quodlibet X, ed. R. Macken, Leuven, 1981.
bibliography 419
, Summa quaestionum ordinariarum, ed. J. Badius Ascensius, Paris, 1520, reprint St.
Bonaventure, N.Y., 1953.
, Summa quaestionum ordinariarum, art. XXXIXXXIV, in Opera omnia 27, ed. R. Macken,
Leuven, 1991.
Henry, D. P ., Medieval Logic and Metaphysics, London, 1972.
Hervaeus Natalis, Quodlibeta, Venetiis, 1513.
Hintikka , Jaakko, The Development of Aristotles Ideas of Scientic Method, in
Aristotles Philosophical Development, ed. W. Wians, Lanham, Md., 1996.
Hoffmann , Philippe, Catgories et langage selon Simplicius . La question du skopos du
trait aristotlicien des Catgories, in Ilsetraut Hadot , ed. 1987, pp. 6190.
Honoratus, Maurus Servius, in Vergilii carmina commentarii, eds. G. Thilo et H. Hagen
Leipzig, 1881.
Hourani , George, Ibn Sina on Necessary and Possible Existence, The Philosophical
Forum 14 (1974), 7486.
, Averroes on the Harmony of Philosophy and Religion Leiden, 1959.
Hugonnard-Roche , Henri, Remarques sur la tradition arabe de lOrganon daprs
le manuscrit Paris, Bibliothque nationale, ar. 2346, in Glosses and commentaries on
Aristotles logical texts, ed. C. Burnett, London, 1993, pp. 1928.
Hume , David, A Treatise of Human Nature, New York, 1992.
Ivry, Alfred , Towards a Unied View of Averroes Philosophy, The Philosophical Forum
14 (1974), 87113.
Jacobus de Placentia , Scriptum super librum Praedicamentorum, Venice, Biblioteca Marciana,
ms. lat. VI.97 2594, fol. 30rbva.
Jean le Page ( Johannes Pagus), Rationes super Predicamenta Aristotelis, in ed. E. France-
schini , Giovanni Pago: le sue Rationes super Predicamenta Aristotelis e la loro
posizione nel movimento aristotelico del secolo XIII, Sophia 2 (1934), 17282,
32950, 47686.
John Duns Scotus, Ordinatio, in Opera omnia, Vaticana, 1950.
, God and Creatures: The Quodlibetal Questions, trans. Felix Alluntis and Allan B.
Wolter , Princeton, N. J., 1975.
, Lectura, in Opera omnia, ed. Vaticana, 1950.
, Quaestiones in duos libros Perihermenias Aristotelis, in Ioannis Duns Scoti Opera philo-
sophica II, eds. R. Andrews , G. Etzkorn, G. Gl, R. Green, T. Noone, and R. Wood ,
St. Bonaventure, 2004.
, Quaestiones in Isagogen Porphyrii et Quaestiones super Praedicamenta Aristotelis, in Opera
philosophica I, eds. R. Andrews, G. Etzkorn, G. Gl, R. Green, T. Noone, and
R. Wood , St. Bonaventure, 1999.
, Quaestiones super libros Metaphysicorum Aristotelis, Libri IIX, in Opera philosophica,
vols. 3 and 4, eds. R. Andrews, G. Etzkorn, G. Gl, R. Green, F. Kelley, G. Marcil,
T. Noone, and R. Wood , St. Bonaventure, 19971998.
, Questions on the Metaphysics of Aristotle, trans. Girard J. Etzkorn and Allan B. Wolter ,
2 vols. St. Bonaventure, 1997.
, Reportata Parisiensia, vol. 22, ed. Vivs, 1891.
Jordan , Mark D., Rewritten TheologyAquinas After His Readers, Oxford, 2006.
Kenny , Anthony, The Realism of De Universalibus, in Wyclif in his Times, ed. Anthony
Kenny, Oxford, 1986, pp. 1729.
, Wyclif , Oxford, 1985.
Kilwardby, Robert, Notulae super librum Praedicamentorum, prooem., and lectio 5, Cambridge,
Peterhouse, ms. 206, fols. 41ra, and 44vb45ra.
, Notulae super Praedicamenta, ed. P. O. Lewry in Robert Kilwardbys Writings on the
Logica Vetus, Studied with Regard to their Teaching and Method, D.Phil. Thesis, University
of Oxford, 1978.
420 bibliography
King , Peter, Between Logic and Psychology: Jean Buridan on Mental Language. Paper
presented at John Buridan and Beyond, Copenhagen, September, 2001. Accessed at
http://individual.utoronto.ca/pking/presentations/Buridan_on_Mental.pdf.
, Duns Scotus on the Common Nature and Individual Difference, Philosophical
Topics 20 (1992), 5176.
, Jean Buridan . In Individuation in Scholasticism: The Later Middle Ages and the Counter-
Reformation; 11501650, ed. Jorge J. E. Gracia , New York, 1994, pp. 397430.
, John Buridan s Solution to the Problem of Universals. In The Metaphysics and
Natural Philosophy of John Buridan, eds. J. M. M. H. Thijssen and Jack Zupko , Leiden,
2001, pp. 128.
, Scotus on Metaphysics, in The Cambridge Companion to Duns Scotus, ed. Thomas
Williams, Cambridge, 2003, pp. 2838.
Klima , Gyula, Aquinas Theory of the Copula, Logical Analysis and History of Phi-
losophy 5 (2002).
, Buridan s Logic and the Ontology of Modes, in Medieval Analyses in Language
and Cognition, eds. Sten Ebbesen and Russell L. Friedman, Copenhagen, 1999, pp.
47395.
, Buridan s Theory of Denitions in his Scientic Practice, in The Metaphysics
and Natural Philosophy of John Buridan, eds. J. M. M. H. Thijssen and Jack Zupko ,
Leiden, 2001, pp. 2948.
, Ockhams Semantics and Ontology of the Categories, in The Cambridge Companion
to Ockham, ed. P. V. Spade , Cambridge, 1999, pp. 11842.
, The Semantic Principles Underlying Saint Thomas Aquinass Metaphysics of
Being, Medieval Philosophy and Theology 5 (1996), 87141.
Knudsen , Christian, Intentions and Impositions, in The Cambridge History of Later
Medieval Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism;
11001600, eds. Norman Kretzmann , Anthony Kenny , and Jan Pinborg, Cambridge,
1982, pp. 47995.
Kogan , Barry, Averroes and the Metaphysics of Causation, Albany, 1985.
Kretzmann , Norman and Eleonore Stump , eds. The Cambridge Companion to Aquinas,
New York, 1993.
Kretzmann , Norman, History of Semantics, in The Encyclopedia of Philosophy, ed. Paul
Edwards, New York, 1967, vol. 7, p. 371.
, Anthony Kenny and Jan Pinborg, eds. The Cambridge History of Later Medieval
Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism; 11001600,
Cambridge, 1982.
, William of Sherwoods Introduction to Logic, St. Paul, 1966.
Laarmann , M., Deus, primum cognitum: Die Lehre von Gott als dem Ersterkannten des menschlichen
Intellekts bei Heinrich von Gent 1293, Mnster, 1999.
Laeur, C. , Le Guide de ltudiant dun matre anonyme de la Facult des arts de Paris au XIII
e

sicle, Qubec, 1992.
Leaman , Oliver, Averroes and his Philosophy, Oxford, 1988.
Leff , Gordon, William of Ockham: The Metamorphosis of Scholastic Discourse, Manchester,
1975.
Lewry , P. O., Robertus Anglicus and the Italian Kilwardby, in English Logic in Italy
in the 14th and 15th Centuries, Acts of the 5th European Symposium on Medieval
Logic and Semantics, Rome, 1014 November 1980, ed. A. Maier, Naples, 1982,
pp. 345.
, The Liber sex principiorum, A Supposedly Porretanean Work. A Study in Ascrip-
tion, in Gilbert de Poitiers et ses contemporains. Aux origines de la Logica modernorum,
eds. Jean Jolivet and Alain de Libera, Naples, 1987, pp. 25178.
, Robert Kilwardbys Writings on the Logica Vetus, Studied With Regard to Their Teaching
doctoral thesis, Oxford, 1978.
bibliography 421
Lohr , Charles H., Medieval Latin Aristotle Commentaries. Authors: AF, Traditio
23 (1967), 313414.
, Medieval Latin Aristotle Commentaries. Authors GI, Traditio 24 (1968),
149246.
, Medieval Latin Aristotle Commentaries. Authors: JacobusJohannes, Traditio
26 (1970), 135216.
, Medieval Latin Aristotle Commentaries. Authors: Johannes de KanthiMyn-
godus, Traditio 27 (1971), 251351.
, Medieval Latin Aristotle Commentaries. Authors: NarcissusRichardus,
Traditio 28 (1972), 281396.
, Medieval Latin Aristotle Commentaries. Authors: RobertusWilgelmus,
Traditio 29 (1973), 93192.
Lorenz, Snke , Thomas Manlefelt Maulefelt, Zu Leben und Werk, in Schule und
Schler im Mittelalter. Beitrge zur europischen Bildungsgeschichte des 9. bis 15. Jahrhunderts,
eds. Martin Kintzinger, Snke Lorenz, and Michael Walter, Beihefte zum Archiv fr
Kulturgeschichte 42, Kln, 1996, pp. 14564.
, Studium Generale Erfordense. Zum Erfurter Schulleben im 13. und 14. Jahrhundert, Mono-
graphien zur Geschichte des Mittelalters, 34 Stuttgart, 1989.
Luna , Concetta, La relation chez Simplicius, in Simplicius: sa vie, son oeuvre, sa survie,
ed. Ilsetraut Hadot, Berlin, 1987, pp. 11347.
, Simplicius , Commentaire sur les Catgories dAristote, Chapitres 2 4; trad. par
Philippe Hoffmann ; avec la collab. de Ilsetraut Hadot et Pierre Hadot; commentaire
par Concetta Luna, Leiden, 2001.
MacDonald, Scott, Theory of Knowledge, in The Cambridge Companion to Aquinas, eds.
Noman Kretzmann and Eleonore Stump, Cambridge, 1993, pp. 160195.
Madkour, Ibrahim, Le trait des Catgories du Shifa, Mlanges de lInstitut dominicain
dtudes orientales du Caire 5 (1958), 25378.
, LOrganon dAristote dans le monde arabe, 2
e
d., Paris, 1969.
Majercik , Ruth, Porphyry and Gnosticism, Classical Quarterly 55 (2005), 27792.
Marenbon, J. , Vocalism, Nominalism and the Commentaries on the Categories from
the Earlier Twelfth Century, Vivarium 30 (1992), 5161.
Markowski , Mieczyslaw, Johannes Buridans Polemik gegen die Universalienlehre des Walter
Burleigh, Mediaevalia Philosophica Polonorum 26 (1982), 717.
, Repertorium commentariorum medii aevi in Aristotelem latinorum quae in bibliothecis Wiennae
asservantur. Opera philosophorum medii aevi, Textus et studia, Wroclaw, 1985.
Marmo, Costantino, A Pragmatic Approach to Language in Modism, in Sprachtheorien
in Sptantike und Mittelalter, ed. Sten Ebbesen, Tbingen, 1995, pp. 16983.
, Suspicio: A Key Word to the Signicance of Aristotles Rhetoric in Thirteenth Cen-
tury Scholasticism, Cahiers de lInstitut du moyen-ge grec et latin 60 (1990), 17075.
, The Semantics of the Modistae, in Medieval Analyses in Language and Cognition,
Acts of the Symposium, The Copenhagen School of Medieval Philosophy, Janu-
ary 1013, 1996, eds. Sten Ebbesen and Russell L. Friedman, Copenhagen, 1999,
pp. 83104.
Marmura , Michael E., Avicenna on the Division of the Sciences in the Isagoge of His
Shifa, Journal of the History of Arabic Science 4 (1980), 23951.
Martin of Dacia , Opera, ed. H. Roos, Corpus Philosophorum Danicorum Medii Aevi,
Copenhagen, 1961.
Matthen , Mohan, The Categories and Aristotles Ontology, Dialogue 17 (1978),
22841.
McGee , C. Douglas, Who Means What by Synonymy?, Inquiry 2 (1959), 199212.
McGrade , A. S., ed. The Cambridge Companion to Medieval Philosophy, New York, 2003.
McHardy , A. K., The Dissemination of Wyclif s Ideas, in From Ockham to Wyclif, eds.
Anne Hudson and Michael J. Wilks, Oxford, 1987, pp. 36168.
422 bibliography
McInerny , Ralph, Albert on Universals, Southwestern Journal of Philosophy 10 (1979),
318.
, Being and Predication, Being and Predication: Thomistic Intrepretations (Washington,
D.C., 1986), pp. 173228.
McMahon , William E., Albert the Great on the Semantics of the Categories of Sub-
stance, Quantity, and Quality, Historiographia Linguistica 7 (1980), 14557.
, Radulphus Brito on the Sufciency of the Categories, Cahiers de lInstitut du
Moyen Age grec et latin 39 (1981).
Meyer , Ruth, Eine neue Perspektive im Geistleben des 13. Jahrhunderts: Pldoyer fr
eine Wrdigung der Organon-Kommentierung Alberts des Groen, in Geistesleben
im 13. Jahrhundert, eds. Jan A. Aertsen and Andreas Speer, Miscellanea Mediaevalia
27, Berlin, 2000, pp. 189201.
Mill , John Stuart, A System of Logic: Ratiocinative and Inductive, in Collected Works VIIVIII,
eds. J. M. Robson and R. F. MacRae, Toronto, 1973.
Minio-Paluello , Laurentius, Les traductions et les commentaires aristotliciens de
Boce, Studia Patristica 64 (1957), 56784, reprint in Id., Opuscula, The Latin Aristotle,
Amsterdam, 1972, pp. 32835.
, Note sullAristotele latino medievale VI. Boezio, Giacomo Veneto, Guglielmo
di Moerbeke, Jacques Lefvre dEtaples e gli Elenchi Sostici, Rivista di Filosoa
Neo-Scolastica 44 (1952), 398411, reprint in Id., Opuscula. The Latin Aristotle Amster-
dam, 1972, pp. 16477.
Moody , E. A., Ockham, Buridan , and Nicholas of Autrecourt, in Inquiries into Medieval
Philosophy: A Collection in Honor of Francis P. Clarke, ed. James F. Ross, Westport, CT,
1971, pp. 275314.
Moraux, P. , Der Aristotelismus bei den Griechen. Von Andronikos bis Alexander Von Aphrodisias,
vol. 2 Berlin, 1984.
Moravcsik , J. M. E., Aristotles Theory of Categories, in Aristotle: A Collection of Critical
Essays, ed. J. M. E. Moravcsik, New York, 1967, pp. 12545.
Muckle , J. T., Isaac Israelis Denition of Truth, Archives dhistoire doctrinale et littraire
du Moyen Age 8 (1933), 58.
Narcy , Michel, Lhomonymie entre Aristote et ses commentateurs noplatoniciens,
Les tudes Philosophiques 1 (1981), 3552, p. 47.
Newton , Lloyd A., Duns Scotuss Questions on the Categories of Aristotle: A Translation of the
Whole, Together with a Philosophical Analysis and Commentary on Questions 18, Dissertation:
University of Dallas, 2003.
Normore , Calvin G., Buridan s Ontology, in How Things Are: Studies in Predication
and the History of Philosophy and Science, eds. James Bogen and James E. McGuire,
Dordrecht, 1981, pp. 189203.
, The Tradition of Medieval Nominalism, in Studies in Medieval Philosophy, ed.
John F. Wippel, Washington, D.C., 1981, pp. 20171.
Nuchelmans , G., Theories of the Proposition. Ancient and Medieval Conceptions of the Bearers
of Truth and Falsity, Amsterdam, 1973.
Nussbaum, Martha , The Fragility of Goodness, Cambridge, 1986.
OBrian , Joseph, Duns Scotuss Teaching on the Distinction Between Essence and
Existence, New Scholasticism 38 (1964), 6177.
OMeara , Dominic, Pythagoras revived. Mathematics and philosophy in Late Antiquity, Oxford,
1989.
Oberman , Heiko, The Harvest of Medieval Theology: Gabriel Biel and Late Medieval Theol-
ogy, Durham, 1983.
Ockham, William, Expositio in librum Praedicamentorum Aristotelis, in Guillelmi de Ockham Opera
philosophica et theologica, Opera philosophica II, ed. Gedeon Gl, St. Bonaventure, 1978.
, Ordinatio, in Opera theologica, vol. 2, ed. J. C. Wey, St. Bonaventure, 1980.
, Quodlibeta septem, in Guillelmi de Ockham Opera theologica IX, ed. J. C. Wey, St.
Bonaventure, N.Y., 1980.
bibliography 423
, Scriptum in librum primum Sententiarum ordinatio. Prologus et distinctio prima, in Guillelmi
de Ockham, Opera theologica I, ed. Gedeon Gl, St. Bonaventure, N.Y., 1967.
, Summa Logicae, in Guillelmi de Ockham, Opera philosophica et theologica I, eds. Gedeon
Gl, et al., St. Bonaventure, N.Y., 1974.
Ottaviano , Carmelo, Le Quaestiones super libro Praedicamentorum di Simone di
Faversham, Atti della R. Accademia Nazionale dei Lincei. Memorie della Classe di Scienze
morali, storiche e lologiche 3 (1930), 258351.
Ottman , Jennifer and Rega Wood , Walter Burley : His Life and Works, Vivarium 37
(1999), 123.
Owen, G. E. L. , Tithenai ta Phainomena, in Logic, Science, and Dialectic, ed. Martha
Nussbaum, Ithaca, 1986, pp. 23951.
Owens , Joseph, Aristotle on Categories, The Review of Metaphysics 14 (1960), 739.
, An Elementary Christian Metaphysics, Milwaukee, 1963.
Panaccio , Claude, Semantics and Mental Language, in The Cambridge Companion to
Ockham, ed. Paul V. Spade , Cambridge, 1999, pp. 5375.
, Connotative Concepts and Their Denitions in Ockhams Nominalism, in La tradition mdivale
des Catgories XII
e
XIV
e
sicles. Actes du XIII
e
Symposium europen de logique et de
smantique mdivales Avignon, 610 juin 2000, ds., J. Biard and I. Rosier-Catach,
Louvain, 2003.
, Les mots, les concepts et les choses. La smantique de Guillaume dOccam et le nominalisme
daujourdhui, Montral, 1991.
Pasnau, Robert and Christopher Shields , The Philosophy of Aquinas, Boulder, 2004.
Pasnau , Robert, Theories of Cognition in the Later Middle Ages, Cambridge 1997.
Pattin , A., Pour lhistoire du Commentaire sur les Catgories dAristote de Simplicius
au moyen ge, in Arts libraux et philosophie au Moyen Age: IV
e
Congrs international de
philosophie mdivale, Montral 1967, Paris, 1969, pp. 107378.
, ed. Simplicius , Commentaire sur les Catgories dAristote, Traduction de Guillaume de
Moerbeke, 2 vols., Louvain, 19711975.
Perler , Dominik, Peter Aureol vs. Hervaeus Natalis on Intentionality. A Text Edition
With Introductory Remarks, Archives dhistoire doctrinale et littraire du Moyen Age 61
(1994), 22762.
, Theorien der Intentionalitt im Mittelalter, Frankfurt am Main, 2002.
Perreiah , Alan R., Paul of Venice: a Bibliographical Guide, Ohio, 1986.
Perrod , M., Pierre de Saint-Amour, recteur de lUniversit de Paris, chanoine de Mcon, archidiacre
de Vauxrenard, Besanon, 1902.
Peter of Auvergne , Quaestiones super Porphyrium, in Le questioni su Porrio di Pierre
dAuvergne, ed. A. Tin, Archives dhistoire doctrinale et littraire du Moyen Age 64 (1997),
253333.
, Quaestiones super Praedicamentis, ed. Robert Andrews, Cahiers de lInstitut du moyen-ge
grec et latin 55 (1987), 384.
Peter of John Olivi , Quaestiones in secundum librum Sententiarum, ed. B. Jansen, I Ad Claras
Aquas, 1922.
Peters , F. E., Aristoteles Arabus, Leiden, 1968.
, Aristotle and the Arabs, New York, 1968,
Pickav , Martin, Heinrich von Gent ber Metaphysik als erste Wissenschaft. Studien zu einem
Metaphysikentwurf aus dem letzten Viertel des 13. Jahrhunderts, Leiden, 2007.
, Henry of Ghent on the Categories, (forthcoming).
Piltz , Anders, Studium Upsalense. Specimens of the oldest lecture notes taken in the medieval
University of Uppsala, Uppsala, 1977.
Pinborg , Jan, Die Logik der Modistae, Studia Mediewistyczne 16 (1975), 3997.
, Speculative Grammar, in The Cambridge History of Later Medieval Philosophy: From
the Rediscovery of Aristotle to the Disintegration of Scholasticism; 11001600, eds. Norman
Kretzmann , Anthony Kenny , and Jan Pinborg, Cambridge, 1982, pp. 25470.
424 bibliography
, Zum Begriff der Intentio Secunda: Radulphus Brito , Hervaeus Natalis und
Petrus Aureoli in Diskussion, Cahiers de lInstitut du Moyen Age Grec et Latin 13 (1974),
4959.
Pini, Giorgio, Absoluta consideratio naturae: Tommaso dAquino e la dottrina avicenniana
dellessenza, Documenti e studi sulla tradizione losoca medievale 15 (2004): 387438.
, Scotus on Assertion and the Copula: A Comparison with Aquinas, Medieval
Theories on Assertive and Non-Assertive Language. Acts of the 14th European Symposium on
Medieval Logic and Semantics, Rome, June 1115, 2002, eds. A. Maier and L. Valente
Firenze, Olschki, 2004, pp. 30731.
, Scotus on Deducing Aristotles Categories, in La tradition mdivale des Catgories
XII
e
XIV
e
sicles. Actes du XIII
e
Symposium europen de logique et de smantique
mdivales Avignon, 610 juin 2000, ds. J. Biard and I. Rosier-Catach, Louvain,
2003, pp. 2335.
, Scotuss Realist Conception of the Categories: His Legacy to Late Medieval
Debates, Vivarium 43 (2005), 63110.
, Categories and Logic in Duns Scotus: an Interpretation of Aristotles Categories in the Late
Thirteenth Century, Leiden, 2002.
Porphyry , In Cat., ed. A. Busse, CAG IV,1, Berlin, 1887.
, On Aristotles Categories, trans., S. K. Strange, London, 1992.
, Porphyrii Isagoge, ed. A. Busse, Berlin, 1887.
Porro , P., Le Quaestiones super Metaphysicam attribuite a Enrico di Gand: elementi per
un sondaggio dottrinale, Documenti e studi sulla tradizione losoca medievale 13 (2002),
507602.
Powers , Lawrence, Equivocation, in Fallacies, eds. H. Hansen & R. Pinto, University
Park, 1995, pp. 293301.
Priscianus , Institutiones grammaticae, ed. M. Hertz, Leipzig, rpt. Hildesheim, 1961.
Pseudo-Gilbert of Poitiers , Anonymi fragmentum vulgo vocatum Liber sex principiorum, ed.
Laurent Minio-Paluello, Bruges, 1966.
Rashed , Marwan, La classication des lignes simples selon Proclus et sa transmission
au monde islamique, in Aristotele e Alessandro di Afrodisia nella tradizione araba, eds.
C. DAncona and Giuseppe Serra, Padova, 2002, pp. 25779.
Reeve, C. D. C. , Practices of Reason, Oxford, 1992.
Rescher, Nicholas , Studies in the History of Arabic Logic, Liverpool, 1963.
Robertu Anglicus, Scriptum super libro Porphyrii, in Le problme des universaux la facult des
arts de Paris entre 1230 et 1260, ed. D. Pich, Paris, 2005.
Robson , John A., Wyclif and the Oxford Schools, Cambridge, 1961.
Rode , Christian, Franciscus de Prato, Stuttgart 2004.
Rosier, Irne, La grammaire spculative des modistes (Lille, 1983).
Ruland , H.-J., Zwei arabischen Fassungen der Abhandlung des Alexander von Aphro-
disias ber die universalia, Nachr. der Akademie der Wiss., Gttingen, 1979.
, Die arabischen Fassungen zweier Schriften des Alexander von Aphrodisias: ber die Vorsehung
und ber das liberum arbitrium, diss. Saarbrcken, 1976.
Sabra , A. I., Avicenna on the Subject Matter of Logic, The Journal of Philosophy 77
(1980), 74664.
Santos Noya , Manuel, Die Universalienlehre der Nominales in der Darstellung
Alberts des Groen, in Albertus Magnus. Zum Gedenken nach 800 Jahren: Neue Zugnge,
Aspekte und Perspektiven, ed. Walter Senner, Quellen und Forschungen zur Geschichte
des Dominikanerordens 10, Berlin, 2001.
Scheu , M. Marina, The Categories of Being in Aristotle and St. Thomas, Washington D.C.,
1944.
Shields , Christopher, Unity in Multiplicity, Oxford, 1999.
Schmidt , Robert W., The Domain of Logic According to Saint Thomas Aquinas, The Hague,
1966.
bibliography 425
Schnberger , R., Relation als Vergleich. Die Relationstheorie des Johannes Buridanus im Kontext
seines Denkens und der Scholastik, Leiden, 1994.
Senko , W., Repertorium commentariorum medii aevi in Aristotelem latinorum quae in Bibliothecis
publicis Parisiis asservantur, Opera philosophorum medii aevi, Textus et studia tom. 5 fasc. 12,
Warsaw, 1982.
Shiel , James, Boethius Commentaries on Aristotle, Mediaeval and Renaissance Studies
4 (1958), 21744.
Simon of Faversham, Notulae super Summulas Petri Hispani, ed. L. M. de Rijk, Vivarium
6 (1968), 69101.
, Quaestiones super libro Porphyrii, in Opera omnia, vol. 1, ed. P. Mazzarella, Padova,
1957.
, Quaestiones super librum Praedicamentorum in Opera omnia, vol. I, ed. P. Mazzarella,
Padova 1957.
, Quaestiones super libro Elenchorum, eds. S. Ebbesen et al., Toronto, 1984.
Simplicius , Simplicius: On Aristotles Categories 14, trans. Michael Chase, Ithaca,
2003.
Solmsen , F., Boethius and the History of the Organon, American Journal of Philology
65 (1944), 6974.
Spade , Paul V., Introduction, in John Wyclif , On Universals, trans., Anthony Kenny ,
Oxford, 1985.
, The Problem of Universals and Wyclif s Alleged Ultrarealism, Vivarium, 43
(2005), 111123.
Stehkmper , Hugo, ed. Albertus Magnus: Ausstellung zum 700. Todestag, Cologne, 1980
81.
Strange , Steven K., ed. Porphyry On Aristotles Categories, trans., Steven K. Strange,
London, 1992.
Stump , Eleonore, Aquinas : The Arguments of the Philosophers, New York, 2005.
Szlezk , T. A., Pseudo-Archytas ber die Kategorien, Berlin, 1972.
Tabarroni, Andrea , Lo pseudo Egidio Guglielmo Arnaldi e uninedita continuazione
del commento di Tommaso al Perihermenias, Medioevo 14 (1988), 391403.
Tardieu , Michel, biens coraniques et biens de arrn, Journal asiatique 274
(1986), 144.
, Recherches sur la formation de lApocalypse de Zostrien et les sources de Marius Victorinus ,
Pierre Hadot , Porphyre et Victorinus: questions et hypotheses, Leuven, 1996.
Taylor, A. E., Varia Socratica, Oxford, 1911.
Thomas Aquinas, Commentary on Aristotles Metaphysics: St. Thomas Aquinas, trans. John
P. Rowan, Notre Dame, 1995.
, Commentary on the Posterior Analytics of Aristotle, trans. F. R. Larcher, 2 New York,
1970.
, De ente et essentia, ed. Hyacinthe F. Dondaine, Rome, 1976.
, De potentia, in Quaestiones disputatae, vol. 2, Rome, 1953.
, De veritate, in Opera omnia XXIII, ed. Leonina, Rome, 1882.
, Expositio libri Peryermenias, in S. Thomae de Aquino, Opera omnia I*, 1, ed. Com-
missio Leonina, Rome, 1989.
, Expositio libri Posteriorum, in S. Thomae de Aquino, Opera omnia, I* 2, ed. Com-
missio Leonina Rome, 1989.
, In duodecim libros Metaphysicorum Aristotelis Expositio, eds. M.-R. Cathala and
R. M. Spiazzi, Rome, 1964.
, In octo libros Physicorum Aristotelis expositio, ed. P. M. Maggilo, Turin, 1954.
, Quaestiones quodlibetales, in Opera omnia 25, ed. Leonina, Rome 1996.
, Scriptum super libros sententiarum, ed. R. P. Mandonnet, Paris, 1929.
, Summa contra Gentiles, in Opera omnia 1315, ed. Leonina, Rome, 19181930.
, Summa Theologiae, in Opera omnia 412, ed. Leonina, Rome, 18881906.
426 bibliography
Thomas of Erfurt , Grammatica Speculativa. An Edition with Translation and Commentary, ed.
G. L. Bursill-Hall , London, 1972.
Thomas Sutton , Expositio super librum Praedicamentorum, Oxford, Merton College, ms.
289, fols. 3rbva and 7ra, transcription in Alessandro D. Conti , Thomas Sutton s
Commentary on the Categories according to the Ms Oxford, Merton College 289,
in The Rise of British Logic, ed. P. O. Lewry , Toronto, 1985, pp. 173213.
Tomarchio , John, Aquinass Division of Being According to Modes of Existing, Review
of Metaphysics 54 (2001), 585613.
Torrell , Jean-Pierre, Initiation Saint Thomas dAquin: Sa personne et son oeuvre, Paris,
2002.
Tremblay , Bruno, A First Glance at Albertus Magnuss Teachings on Analogy of
Words, Medieval Philosophy and Theology 5 (1996), 26592.
, Albert le Grand: De ce qui vient avant la logique, History and Philosophy of Logic
25 (2004), 165203.
, Ncessit, rle et nature de lart logique, daprs Albert le Grand, Bochumer
philosophisches Jahrbuch fr Antike und Mittelalter, forthcoming.
Urvoy, Dominque , Ibn Rushd, in History of Islamic Philosophy, eds. S. Nasr and
O. Leaman, London, 1996, pp. 33045.
Vallat , Philippe, Frb et lEcole dAlexandrie. Des prmisses de la connaissance la philosophie
politique, Paris, 2004.
Van Steenberghen , F., Aristote en occident, Louvain, 1946, trans. L. Johnston, as Aristotle
in the West, Louvain, 1970.
Verhaak , C., Zeger van Kortrijk, Commentator van Perihermeneias, Brussels, 1964.
Vittorini , Marta, Predicabili e categorie nellultimo commento di Walter Burley all Isagoge di
Porrio, Ph.D. diss., University of Salerno, academic year 20045.
von Perger , Mischa, Understanding the Categories by Division: Walter Burley vs. Wil-
liam of Ockham, in La tradition mdivale des Catgories XII
e
XV
e
sicles, eds. Joel Biard,
Irne Rosier-Catach, Louvain-la-Neuve, 2003, pp. 3752.
Wagner , Hans, ber das Aristotelische pollachos legetai to on, Kantstudien 53 (1962),
7591.
Wallace , William A., Albertus Magnuss Inventive Logic: His Exposition on the Topics
of Aristotle, American Catholic Philosophical Quarterly 70 (1996), 1139.
Wallerand , G., Les oeuvres de Siger de Courtrai Les Philosophes Belges, VIII, Louvain,
1913.
Walter Burley , De suppositionibus (A.D. 1302), in Stephen F. Brown , Walter Burleighs
Treatise De suppositionibus and its Inuence on William of Ockham, Franciscan Studies
32 (1972), 1564.
, Expositio libri De Anima (before 1316), Civitas Vaticana, Bibl. Apostolica Vaticana,
ms. Vat. Lat. 2151, fols. 1ra88rb.
, Expositio super librum Perihermeneias, in Expositio super Artem Veterem Porphyrii et Aris-
totelis cit., fols. 66ra93rb.
, Expositio super Praedicamenta Aristotelis, in Expositio super Artem Veterem Porphyrii et
Aristotelis cit., fols. 17va50va.
, Expositio super Universalia Porphyrii, in Expositio super Artem Veterem Porphyrii et Aris-
totelis (A.D. 1337), Octavianus Scotus per Bonetum Locatellum, Venetiis 1509, fols.
2ra17va.
, In Physicam Aristotelis Expositio et Quaestiones (after 1324), Simon de Luere, Venetiis
1501
, Quaestiones in librum Perihermeneias (A.D. 1301), in Stephen F. Brown , Walter
Burleys Quaestiones in librum Perihermeneias, Franciscan Studies 34 (1974), 20095.
, Tractatus de abstractis, Civitas Vaticana, Bibl. Apostolica Vaticana, ms. Vat. Lat.
2146, fol. 249ra249vb.
, Tractatus de universalibus (after 1337), ed. Hans-Ulrich Whler, Leipzig 1999.
bibliography 427
, Tractatus super librum Praedicamentorum (before 1310), Cambridge, Peterhouse, ms.
184, fols. 171rb189va.
Walton , William M, The Second Mode of Necessary or Per Se Propositions According
to St. Thomas Aquinas, The Modern Schoolman 29 (195152), 293306.
Walzer , Richard, Porphyry and the Arabic tradition, in Porphyre, Vanduvres, 1966,
Entretiens sur lAntiquit classique 12, pp. 27599.
Washell , Richard F., Logic, Language, and Albertus Magnus, Journal of the History
of Ideas 34 (1973), 44550.
Weinberg , Julius, Abstraction, Relation and Induction, Madison, 1965.
Wians , William, Saving Aristotle from Nussbaum s Phainomena, in Essays in Ancient
Greek Philosophy V, eds. A Preus and J. Anton , Albany, 1992.
Wieland , Georg, Untersuchungen zum Seinbergriff im Metaphysikkommentar Alberts des Groen,
Mnster, 1972.
Wippel, John F., Essence and Existence, in The Cambridge History of Later Medieval
Philosophy: From the Rediscovery of Aristotle to the Disintegration of Scholasticism; 11001600,
eds. Norman Kretzmann , Anthony Kenny , and Jan Pinborg, Cambridge, 1982, pp.
385410.
, Godfrey of Fontaines and Henry of Ghent s Theory of Intentional Distinc-
tion Between Essence and Existence, in Sapientiae procerum amore, ed. T. W. Koehler,
Rome 1974, pp. 289321.
, Thomas Aquinas s Derivation of the Aristotelian Categories, Journal of the
History of Philosophy 25 (1987), 1334.
, The Metaphysical Thought of Thomas Aquinas Washington, D.C., 2000.
Wlodek , Sophie, Albert le Grand et les Albertistes du XV
e
sicle: le problme des
universaux, in Albert der Grosse. Seine Zeit, seine Werk, seine Wirkung, Band 2, ed. Albert
Zimmermann, Miscellanea Mediaevalia 14, Berlin, 1981, pp. 193207.
Wolter , Allan, Transcendentals and their Function, Washington, D.C., 1946.
Wright , W., A Grammar of the Arabic Language, Cambridge, 1967.
Wyclif , John, De actibus animae 1365 ca., in Johannis Wyclif miscellanea philosophica, ed.
Michael H. Dziewicki, 2 vols., London, 1902.
, De ente in communi, in Summa de ente libri primi tractatus primus et secundus, ed. Samuel
H. Thomson, Oxford, 1930.
, De ente praedicamentali, ed. Rudolf Beer, London, 1891.
, Tractatus de universalibus 1374 ca., ed. Ian J. Mueller, Oxford, 1985.
Zedler , Beatrice, trans. & comm., Destructio Destructionum Philosophiae Algazels, Milwaukee,
1961.
Zimmermann, A., Ontologie oder Metaphysik? Diskussionen ber den Gegenstand der Metaphysik
im 13. und 14. Jahrhundert: Texte und Untersuchungen, 2nd ed. Leuven, 1998.
Zimmermann , F. W., al-Frb s commentary and short treatise on Aristotles De Interpretatione,
Oxford, 1981.
Zupko , Jack, John Buridan : Portrait of a Fourteenth-Century Arts Master, Notre Dame,
2003.
LIST OF CONTRIBUTORS
Robert Andrews study of Categories commentaries from the Middle
Ages arises out of a broader interest in the change and development
of ideas in the history of philosophy. After obtaining a doctorate from
Cornell University, he joined the Scotus Project of the Franciscan
Institute at St. Bonaventure University, editing the philosophical works
of John Duns Scotus. He has since taught in Sweden, Norway, and the
United States, and is currently teaching in Kristinehamn, Sweden.
Allan Bck received his B.A. in philosophy (Phi Beta Kappa) from
Reed College, and his Ph.D. in philosophy from The University of
Texas at Austin. He has published widely in such areas as the history
and philosophy of logic, ancient philosophy, medieval philosophy, and
comparative philosophy. His rst book (On Reduplication (Leiden: Brill,
1996)) received a prize from VGWort. In 1999 he received a For-
schungspreis from the Alexander von Humboldt Stiftung for his research,
and was in residence at the University of Freiburg (im Breisgau). He
has taken an American Philosophical Association Fellowship for the
Institute for Advanced Studies at Edinburgh University. He is a Professor
of Philosophy at Kutztown University in Pennsylvania. He has taught
at other schools, including Rice University and Emory University.
Todd Bates received his Ph.D. in 2003 from the University of Penn-
sylvania, writing a dissertation on Duns Scotus and the problem of
universals. In 1993, he received a B.A., summa cum laude, in Philosophy,
and a B.A., cum laude, in Classics from the University of Massachusetts
at Amherst. In 2004, he published an article on natural theology with
Jim Ross in The Cambridge Companion to Duns Scotus. Dr. Bates joined the
faculty of the University of Central Florida in 2004; before that, he
lectured at Stetson University and Rollins College.
Michael Chase (Ph.D. Ecole Pratique des Hautes Etudes, Paris [Sor-
bonne] 2000), is Ingnieur de Recherche at the French National Center
of Scientic Research (UPR 76, Villejuif-Paris), where he works as
Associate Editor at the periodical lAnne Philologique. He is the author
of a translation of Books 14 of Simplicius Commentary on the Categories
430 list of contributors
(London: Duckworth, 2003, in the series Ancient Commentators on Aristotle),
as well as of several articles on Late Antique, Islamic and Medieval
thought, several of which have appeared in R. Goulets Dictionnaire des
Philosophes Antiques. He has also translated a half-dozen works by the
French scholars Pierre and Ilsetraut Hadot, most recently Pierre Hadot,
The Veil of Isis. An Essay on the History of the Idea of Nature, Harvard
University Press, Autumn 2006.
Alessandro D. Conti is Associate Professor in the Faculty of Letters and
Philosophy at the University of LAquila (Italy). His research interests
include the tradition of commentaries on Porphyrys Isagoge and on
Aristotles Categories in late Antiquity and Middle Ages, and Realism in
late Middle Ages. He has published editions of texts by Thomas Sutton,
Robert Alyngton, Johannes Sharpe, William Penygull, William Russell,
and Paul of Venice. Among his books is Esistenza e verit: Forme e strut-
ture del reale in Paolo Veneto e nel pensiero losoco del tardo medioevo (Rome
1996). He has been the Guest Editor of Vivarium 43 (2005), dedicated
to Realism in the late Middle Ages.
Alexander Hall is an Assistant Professor of Philosophy at Clayton
State University. He is the author of Thomas Aquinas and John Duns Scotus:
Natural Theology in the High Middle Ages (Thoemmes Continuum), and
Logic and Analogy: Aquinass Five Ways, in The Logic and Semantics
of Thomas Aquinas: Essays in Honor of Professor E. J. Ashworth.
Lloyd A. Newton (University of Dallas, 2003) is Associate Profes-
sor of Philosophy at Benedictine College, in Atchison, Kansas. His
research interests include medieval logic, metaphysics and the phi-
losophy of nature. He has an article on Duns Scotuss understanding
of the categories in the American Catholic Philosophical Quarterly and has
co-authored with Jorge J. E. Gracia an article on Medieval Theories
of the Categories for the Stanford Encyclopedia of Philosophy.
Martin Pickav (University of Cologne, 2003) is Assistant Professor
of philosophy and medieval studies at the University of Toronto. His
research interests include medieval metaphysics and philosophy of mind
and especially the philosophy of Henry of Ghent . He is the author of
several articles and recently published Heinrich von Gent ber Metaphysik
als erste Wissenschaft. Studien zu einem Metaphysikentwurf aus dem letzten Viertel
list of contributors 431
des 13. Jahrhunderts (Leiden 2007). He also co-edited several volumes of
the Miscellanea Mediaevalia.
Giorgio Pini (Ph.D., Scuola Normale Superiore, 1997) is Associate
Professor at Fordham University, New York. His research interests
include late medieval logic, metaphysics, and theology, especially in Giles
of Rome, Henry of Ghent , and John Duns Scotus. He is the author
of several articles and two books, Categories and Logic in Duns Scotus. A
Thirteenth-Century Interpretation of Aristotles Categories (Brill 2002) and Scoto
e lanalogia. Logica e metasica nei commenti aristotelici (Pisa 2002).
Paul Symington (Ph.D., University at Buffalo in New York, 2007)
is an Assistant Professor of Philosophy at Franciscan University of
Steubenville. His dissertation, directed by Jorge J. E. Gracia , examined
the debate between Thomas Aquinas and John Duns Scotus on the
sufciency of Aristotles categories. His most recent article, Naming
and the Analogy of Being, is forthcoming in International Philosophical
Quarterly. He also has articles in American Catholic Philosophical Quarterly
and Philosophy Today. He presently lives in Buffalo with his wife Eliza
and three daughters.
Bruno Tremblay (Universit Laval) is Assistant Professor in the Philoso-
phy department of St. Jeromes University, Canada. He has published
articles on Boethius, Albertus Magnus, Thomas Aquinas, and Michel
de Montaigne, which have appeared in varied international journals:
Carmina Philosophiae, Angelicum, Medieval Philosophy and Theology, Freiburger
Zeitschrift fur Philosophie und Theologie, Estudios Filoscos, History and Phi-
losophy of Logic, and Montaigne Studies.
INDEX
Abdallh ibn al-ayyib, 12
Abed, 46 n. 57, 411
Abelard, 33, 43, 46
Ab Yaqb, 31, 65
Ab Zakary Yay ibn Ad, 12
Ab-l-asan al-Amir, 12
Ackrill, 171 n. 44, 300 nn. 1617, 302
n. 24, 305 n. 34, 317 n. 1, 338 n. 72,
411413
Adams, 163 n. 30, 297 n. 9, 375 n. 11,
377 n. 17, 379 n. 22, 380 nn. 27, 29,
413
Afnan, 45 n. 56
Albert, 10, 73, 73 nn. 2, 4, 74, 74 nn.
56, 75, 75 n. 10, 76, 76 n. 11, 77,
77 n. 15, 7879, 79 nn. 24, 2729,
80 nn. 30, 33, 81, 82 nn. 46, 48, 83,
83 n. 51, 84, 84 n. 56, 8588, 9094,
94 n. 101, 95, 95 n. 102, 102 n. 15,
9697, 97 n. 112, 111112, 113 n.
58, 114 n. 62, 115, 119 n. 4, 120,
121 n. 9, 243, 322, 322 n. 17, 327,
327 n. 35, 339, 339 n. 75, 341, 342
n. 88, 370, 372 nn. 45, 373 n. 7,
374 nn. 89, 377 n. 19, 378 n. 21,
380, 383, 385 n. 40, 399, 402, 406,
411, 422
Alexander, 17, 36 n. 22, 4142, 44 nn.
50, 52, 66, 6869, 170 n. 42, 422,
424
al-Frb, v, 9, 12, 12 n. 16, 17, 17
n. 36, 18 n. 38, 25, 28 n. 76, 38,
4041, 42 n. 43, 44, 44 nn. 49, 52,
4547, 47 n. 59, 53 n. 76, 54, 58
n. 94, 61, 62 n. 99, 6568, 154, 411,
413, 427
Al-Ghazl, 65, 67, 6970
al-asan ibn Suwr, 12
al-Jzjn, 40, 42 n. 45
al-Masd, 28
al-Nadm, 12, 12 n. 12
al-Qif, 12, 12 n. 13
al-Yasin, 19
Alyngton, 147, 162, 167168, 168 n. 40,
169, 169 n. 41, 171, 176, 178, 178
n. 54, 411
Amerini, 318 n. 4, 381 n. 31, 388 nn.
5253, 390 nn. 5961, 411413, 417
Ammonius, 9 n. 1, 25, 40 n. 35, 42, 373
Andrews, v
vi, 75 n. 10, 99, 99 n. 1,

104 n. 24, 105 nn. 29, 34, 111 n. 50,
119 n. 2, 155 n. 18, 158 n. 23, 160
n. 26, 183 n. 2, 185 nn. 34, 189
n. 14, 195 n. 32, 196 nn. 35, 3738,
197 n. 41, 201 n. 50, 214 n. 87, 215
nn. 88, 91, 217 n. 96, 233, 234 n. 44,
235 n. 47, 241 n. 61, 347, 347 n. 3,
348 n. 4, 413, 419
Andronicus, 149
Annas, 338 n. 72, 412
Anonymous of Madrid, 195, 197, 201,
214, 215 nn. 89, 91, 217218, 218
n. 99
Anonymus Escorial, 198 n. 43
Antiochus, 13
Anton, 34 n. 15, 52 n. 73, 414, 429
Aquinas, v, 9, 10 n. 3, 1113, 13 n. 17,
15, 15 n. 26, 16 nn. 2930, 17 n. 32,
20, 20 n. 50, 23, 24 n. 61, 2526,
71, 79, 81, 81 n. 44, 84 n. 56, 94
n. 101, 99, 101, 103, 107 n. 38, 116,
119, 119 nn. 2, 4, 120, 120 nn. 58,
121, 121 nn. 910, 122, 122 n. 12,
123, 123 n. 17, 124125, 125 n. 21,
127129, 129 n. 35, 130, 130 n. 38,
131, 131 n. 43, 132139, 139 n. 64,
140143, 147, 155, 155 n. 18, 157,
157 n. 22, 158, 166, 166 n. 34, 170
n. 43, 171, 171 n. 45, 172, 172 n. 46,
173, 186 n. 5, 193, 193 n. 26, 201,
201 n. 51, 202, 202 n. 56, 203, 203
nn. 5758, 204, 207, 207 n. 70, 216,
228, 228 n. 28, 229, 230 n. 33, 231,
231 nn. 35, 42, 239 n. 55, 251, 254,
254 n. 95, 321, 321 n. 15, 370, 376,
390, 392, 392 n. 65, 393, 393 n. 65,
402, 406, 413, 417, 420, 425, 427
Archytas, 44
Ashworth, 75 n. 10, 7677, 143 n. 74,
152 n. 15, 153 n. 15, 319 n. 6, 412
Asztalos, 76 n. 12, 412
Augustine, vii, 71, 74, 74 n. 8, 412
434 index
Averroes, 11 n. 9, 12, 31, 32 n. 6,
65, 65 nn. 115116, 66, 66 n. 118,
6768, 68 nn. 124, 128, 69, 69 nn.
129131, 133, 7071, 110 n. 4849,
111, 167, 412413, 416, 419420
Avicenna, v, 31, 31 nn. 1, 3, 32, 32 nn.
57, 33, 3738, 38 nn. 27, 29, 39,
39 nn. 30, 32, 40, 40 n. 33, 41, 41
nn. 3940, 42, 42 nn. 4243, 45, 43,
43 n. 47, 44, 44 nn. 48, 50, 52, 45,
45 n. 56, 46, 46 n. 58, 6568, 68
n. 125, 6971, 78, 78 nn. 18, 20, 23,
79 n. 24, 80, 85, 89, 89 n. 78, 91, 95
n. 102, 97, 97 n. 113, 107 n. 38, 151,
154, 161, 167, 169 n. 41, 205 n. 65,
206 n. 65, 215 n. 88, 233, 242, 330,
330 n. 48, 373, 413, 418
Bck, v, 31, 32 n. 5, 34 n. 14, 413
Bacon, 46
Badaw, 32 n. 6, 38 n. 27, 44 nn. 50,
52, 413
Bakker, 74 n. 7, 295 n. 1, 413
Bekker, 102 n. 15, 145
Beuchot, 77 n. 15, 413
Black, 413
Bloch, 37 n. 25, 413
Bods, 148 n. 2, 149 nn. 56, 151
n. 11, 170 n. 43, 171 n. 45, 413
Boethius, vii, 9 n. 1, 46, 73 n. 3, 76,
76 n. 12, 91, 91 n. 84, 92, 92 n. 91,
106107, 107 n. 40, 108, 108 n. 43,
110, 110 n. 49, 111, 111 n. 51, 112
n. 54, 115, 148 n. 4, 150, 150 nn.
79, 163, 167168, 173, 174 n. 48,
234, 242243, 302, 317, 317 n. 2,
331, 331 n. 53, 339, 339 n. 74,
359 n. 4, 364, 364 n. 11, 372, 413,
425
Bonaventura, 413
Bonin, 80 n. 33, 413
Bos, 74 n. 7, 84 n. 56, 94 n. 101, 119
n. 4, 121 n. 9, 129 n. 35, 302 n. 27,
303 nn. 2930, 413
Bouffartigue, 27 n. 71
Boulnois, 413
Breton, 129 n. 35, 414
Brito, 119 n. 2, 147, 155, 155 n. 18,
156, 159, 159 n. 25, 162, 232, 381
n. 31, 414, 424
Brown, 370 n. 2, 371 n. 3, 376 n. 15,
381 n. 31, 382 n. 31, 411, 414, 426
Brunschwig, 145 n. 1, 148 n. 3, 414
Buridan, vi, 67, 162, 163 n. 29, 295,
295 nn. 1, 4, 296, 296 nn. 67, 297,
297 nn. 78, 298, 298 nn. 1012,
299, 299 n. 13, 300 nn. 1415,
302303, 303 n. 30, 304, 304 n. 33,
305, 305 nn. 35, 3739, 306 nn.
3940, 307, 308 nn. 4345, 309,
309 n. 46, 310, 310 n. 48, 311314,
314 n. 50, 315, 318, 380, 402, 406,
413414, 416, 420, 422, 427
Burley, 147, 160, 162, 164, 167, 167 nn.
3638, 168, 168 n. 39, 171, 176, 177
n. 52, 297298, 298 n. 10, 318, 318
nn. 45, 320, 322, 322 n. 19, 323,
323 n. 21, 324325, 325 n. 29, 327,
331, 331 nn. 51, 55, 336337, 337
n. 69, 339, 339 n. 77, 340, 342,
342 n. 91, 344, 344 n. 94, 346, 370,
370 n. 2, 371 n. 3, 372 nn. 45,
373, 373 n. 6, 374 n. 9, 376, 376 nn.
1416, 377 n. 19, 378 nn. 1921, 379
n. 21, 381382, 382 n. 32, 383, 383
nn. 3337, 384, 384 nn. 3839, 385,
385 nn. 4143, 386, 386 nn. 4448,
387, 387 nn. 49, 51, 388, 393394,
401402, 406, 408, 414
Burnet, 34, 34 n. 11, 414
Burnyeat, 148 n. 3, 170 n. 42, 414
Campo, 82 n. 46
Caujolle-Zaslawsky, 339 n. 73, 414
Cesalli, 167 n. 38, 345 n. 95, 382 n. 32,
393 n. 66, 414
Chadwick, 317 n. 2, 414
Charles, 1 n. 1, 32 n. 6, 48 n. 64, 100
n. 6, 101 n. 12, 104 n. 20, 115, 115
n. 63, 413414, 421
Chase, v, 2, 9, 11 n. 7, 13 n. 19, 16
n. 27, 19 n. 45, 27 n. 70, 28 n. 76,
44 n. 50, 45 nn. 5354, 119 n. 1,
414, 418, 425, 429
Chvarri, 77 n. 15, 414
Chenu, 199 n. 46, 414
Chiaradonna, 21 n. 52
Cohen, 50 nn. 7071, 62 n. 99, 301
n. 19, 414
Confucius, 41
Conti, vi, 78, 76, 76 n. 13, 77, 155
n. 18, 156 nn. 15, 19, 162 n. 28, 167
n. 36, 168 n. 40, 169 n. 41, 178 nn.
5354, 183 n. 2, 317 n. 2, 318 nn. 3,
5, 319 nn. 59, 321 n. 16, 325 n. 30,
327 n. 34, 328 n. 41, 339 nn. 7374,
index 435
7778, 344 n. 94, 345 n. 95, 369,
372 n. 4, 374 n. 9, 375 n. 11, 376
n. 14, 377 nn. 17, 19, 378 n. 20,
382 n. 32, 392 n. 65, 393 nn. 6566,
396 n. 73, 397 n. 78, 398 n. 79, 402
n. 87, 403 n. 89, 405 n. 91, 411,
414415, 426, 430
Copleston, 31 n. 2, 415
Courcelle, 27 n. 73, 415
Cunningham, 415
DAncona, 11 n. 9, 24 n. 62, 415, 424
Daiber, 44 n. 51, 415
Damascius, 26 n. 67
de Haas, 21 n. 52
de Libera, 80 n. 33, 83 n. 51, 319
n. 9, 369 n. 1, 399 n. 79, 416
de Mottoni, 100, 416
de Rijk, 180 n. 55, 188 n. 11, 234
n. 45, 416
Decorte, 205 n. 65, 206 n. 65, 416
Dekker, 298 n. 11, 416
Delisle, 102, 102 n. 16, 416
Denie, 100 n. 2, 416
Dexippus, 9 n. 1
Dillon, 23 n. 56, 416
Donati, 99 n. 1, 416
Druart, 68 n. 128, 416
Duerlinger, 329 n. 43, 416
Duhem, 348 n. 5, 416
Dumont, 375 n. 11, 416
Dunlop, 68 n. 124, 416
Durand de St. Pourain, 11, 103
Ebbesen, 73 n. 4, 151 n. 13, 183 n. 1,
186 nn. 5, 7, 187, 187 n. 10, 209
n. 74, 213 n. 84, 218, 219 n. 103,
301 n. 19, 317 n. 2, 319 n. 8,
415417
Elias, 9 n. 1
Emden, 319 n. 6, 381 n. 79, 417
Endress, 44 nn. 5051, 417
Euclid, 11, 227 n. 25, 230 n. 32, 417
Eustratus, 80, 417
Evans, 34 n. 16, 319 n. 6, 412, 417
Fischer, 53 n. 76, 417
Franceschini, 75 n. 10, 154 n. 16, 417,
419
Franciscus, 318, 318 n. 4, 381, 388,
388 n. 52, 389, 389 nn. 5557, 390,
390 nn. 5861, 391, 391 n. 62, 392,
392 nn. 6364, 406, 417, 424
Frede, 120 n. 5, 145 n. 1, 148 n. 2, 149,
417
Friedrich, 64 n. 112, 417
Galluzzo, 171 n. 45, 417
Gaskin, 45 n. 55, 417
Gtje, 11 nn. 910, 12 nn. 12, 1415,
44 nn. 5051, 417
Gauthier, 101
Gellius, 41 n. 38, 417
Gerard of Nogent, 100, 103, 106107,
116, 233
Gilbert of Poitiers, 83 n. 51
Giles, 147, 158, 158 nn. 23, 370, 205,
206 n. 65, 373 n. 7, 376, 392394,
394 n. 65, 402, 417
Gilson, 28 nn. 7778, 31 n. 2, 120 n. 7,
203 n. 58, 417
Glorieux, 100 n. 3, 417
Godefroid, 11
Gohlman, 39 n. 31, 417
Gorman, 417
Gracia, viii, 119 n. 2, 144 n. 76, 300
n. 14, 417, 420, 430431
Greimas, 118 n. 68, 418
Gremper, 75 n. 9, 418
Grosseteste, 107 n. 38, 418
Grotz, 106 n. 35, 418
Guilelmus Dorotheus, 10
Gutas, 31 n. 1, 38, 38 nn. 2729, 39
nn. 30, 32, 40, 40 n. 33, 41 nn.
3940, 42 nn. 4243, 44 nn. 5152,
45, 45 n. 56, 46 n. 58, 418
Guthrie, 34, 34 n. 11, 418
Hadot, Ilsetraut, 18 n. 43, 27 n. 74, 148
n. 4, 339 n. 73, 418419, 421
Hadot, Pierre, 20 n. 50, 24, 24 n. 59,
27, 27 n. 70, 425
Hall, vi, 67, 105 n. 32, 295, 426, 430
Hamesse, 107 n. 39, 108 n. 42, 110
n. 46, 113 n. 57, 359 n. 3, 362 nn.
89, 364 n. 12, 418
Hankey, 11 n. 7, 13, 13 nn. 15, 26, 17,
17 n. 33, 20, 20 nn. 4951, 21, 23,
23 n. 58, 24 n. 61, 25 nn. 63, 66, 26,
26 n. 68, 418
Hasse, 46 n. 58, 418
Henle, 25 n. 65, 418
Henninger, 201 n. 51, 202 n. 52, 205,
206 n. 65, 210 n. 77, 418
Henry of Ghent, 5, 11, 198, 198 n. 43,
201, 201 n. 51, 202 nn. 52, 55, 203,
436 index
203 n. 57, 205206, 206 n. 65, 209,
210 n. 77, 212, 217, 239, 253, 318,
318 n. 4, 370, 375, 375 n. 10, 376,
406, 408, 416, 418419, 423, 427,
430431
Henry, D. P., 419
Hervaeus, 381 n. 31, 388389, 392,
411, 419, 423424
Hintikka, 70, 70 n. 134, 419
Hoffmann, 9 n. 1, 18 n. 43, 418419,
421
Honoratus, 114 n. 61, 419
Hourani, 65 n. 116, 66 n. 117, 68
n. 125, 69 n. 133, 419
Hugonnard-Roche, 12 n. 14, 419
Hume, 7, 347, 352, 352 n. 16, 356, 419
unayn Isq, 38, 43
Iamblichus, 14, 20, 20 n. 50, 2123, 23
n. 56, 26, 26 n. 67, 27, 27 nn. 69,
7273, 416
Ivry, 69 n. 131, 419
Jacobus de Placentia, 318 n. 5, 406, 419
Jacques de Thrines, 11
James of Viterbo, 11
Jean Quidort, 11
Johannes Pagus, 74, 153, 154 n. 16,
233, 419
John of Seccheville, 104
Jordan, 9 n. 2, 419
Kenny, 393 n. 66, 411, 416, 419420,
423, 425, 427
Kilwardby, 74
King, 298 n. 12, 299, 299 n. 13, 300
n. 14, 305 n. 37, 308, 318 n. 5,
375 n. 11, 420
Kirk, 34
Klima, 131 n. 43, 163 n. 30, 295, 295
n. 4, 296 n. 6, 296 n. 8, 299 n. 13,
303, 305, 305 nn. 3435, 306 n. 39,
308, 308 n. 45, 309 n. 46, 310 n. 47,
420
Knudsen, 261 n. 5, 262 n. 8, 420
Kogan, 68 n. 128, 420
Kretzmann, 78 n. 19, 95 n. 102, 111
n. 50, 121 n. 10, 210 n. 77, 261 nn.
56, 297 n. 9, 301, 380 n. 27,
411412, 416, 420, 423, 427
Laarmann, 209 n. 75, 420
Laeur, 75 n. 10, 152 n. 15, 420
Leaman, 12 n. 16, 31 n. 3, 46 n. 57, 65
n. 115, 66 n. 118, 411, 414, 420, 426
Leff, 348 n. 7, 420
Lewry, 75 n. 10, 76 n. 13, 83 n. 51,
96 n. 107, 97 n. 112, 104, 104 n. 23,
152 n. 15, 155 n. 18, 183 n. 2, 233,
234 n. 43, 318 n. 3, 372 n. 4,
419420, 426
Lohr, 100 n. 6, 101, 101 n. 12, 104
n. 20, 105 n. 30, 115, 115 n. 63, 421
Lombard, 78
Lorenz, 105 n. 33, 347 n. 2, 421
Luna, 9 n. 1, 45 n. 54, 49 n. 66, 51
n. 72, 52 n. 73, 53 n. 77, 54 n. 79,
339, 418, 421
Mabdi, 41 n. 39
Madkour, 40 n. 36, 97 n. 113, 151
n. 11, 421
Maimonides, 41, 65
Majercik, 24 nn. 5960, 27 n. 71, 421
Marenbon, 151 n. 10, 421
Markowski, 101 n. 8, 105 n. 28, 298 nn.
1011, 421
Marmo, 186 n. 5, 421
Marmura, 78 n. 20, 421
Martin of Dacia, 5, 214, 214 n. 85,
222, 232233, 233 n. 41, 234 n. 46,
235, 240, 252, 259, 421
Matthen, 300 n. 18, 421
Maximus the Confessor, 24 n. 62
McGee, 54 n. 79, 421
McGrade, 421
McHardy, 317 n. 6, 421
McInerny, 77 n. 15, 79 n. 27, 422
McMahon, 76 n. 11, 119 nn. 2, 4, 422
Meyer, 73 n. 4, 422
Mill, 181, 181 n. 56, 422
Milverley, 162
Minio-Paluello, 10 nn. 45, 83 n. 52, 90
n. 81, 150 n. 9, 422
Moody, 298 n. 10, 422
Moraux, 170 n. 42, 422
Moravcsik, 300, 301 n. 18, 422
Muckle, 107 n. 38, 108 n. 41, 422
Najjar, 17 n. 36, 19, 411
Narcy, 50 n. 70, 422
Neoplatonic, 9 n. 1, 11, 11 n. 7, 13,
13 n. 18, 15 n. 24, 16, 18, 19 n. 45,
2021, 2325, 27, 32 n. 5, 66, 146,
148149, 170, 170 n. 42, 180, 339
n. 73, 372373, 413, 418
index 437
Neoplatonist, 13, 15, 24 n. 61, 27, 31,
36, 38, 80, 92, 339, 373
Newton, v, 5, 73 n. 1, 119 n. 2, 144
n. 76, 221, 222 n. 4, 259, 277, 288
n. 11, 354, 417, 422, 430
Nicholas of Paris, 74, 233
Normore, 297 n. 9, 305 n. 37, 422
Noya, 74 n. 6, 81 n. 43
Nuchelmans, 166 n. 34, 422
Nussbaum, 34 n. 15, 35, 35 nn. 17, 20,
422423, 427
OBrian, 254 n. 95, 422
OMeara, 14 n. 22, 120 n. 5, 417, 422
Oberman, 348 n. 6, 422
Ockham, vii, 71, 147, 160, 162163,
163 n. 30, 164, 164 n. 31, 165, 165
nn. 3233, 166, 166 nn. 3435, 167,
171, 175176, 176 nn. 5051, 180,
295, 297 n. 9, 298, 298 n. 10, 305,
306 n. 39, 308 n. 45, 423
Olivi, 147, 160, 161 n. 27, 162, 406,
423
Olympiodorus, 9 n. 1, 338
Ottaviano, 75 n. 10, 423
Ottman, 298 n. 10, 423
Owen, G. E. L., 35, 35 n. 17, 423
Owens, Joseph, 137, 138 n. 59, 170
n. 43, 423
Panaccio, 163 n. 30, 380 nn. 27, 29,
423
Pasnau, 120 n. 7, 380 n. 27, 423
Pattin, 10 nn. 36, 11 n. 8, 423
Paul of Genezano, 10 n. 5
Pengybull, 162
Perler, 381 n. 31, 423
Perreiah, 398 n. 79, 423
Perrod, 100, 100 n. 3, 423
Peter of Auvergne, 11, 104 n. 24, 105
n. 29, 111 n. 50, 155, 155 n. 18, 183
n. 2, 189, 189 n. 14, 195, 196 n. 37,
197 n. 41, 198, 214, 214 n. 87, 215
nn. 8889, 91, 232, 234 n. 44, 235,
235 n. 47, 239, 241 n. 61, 252, 412,
423
Peter of Saint-Amour, 100103, 106,
110, 115116, 118
Peter of Spain, 180, 180 n. 55, 302,
303 n. 28
Peters, 12 n. 11, 43 n. 47, 44 n. 51,
423
Philoponus, 9 n. 1, 44 n. 52
Pickav, v, 5, 183, 221, 221 n. 2, 237
n. 50, 423, 430
Piltz, 347 n. 3, 423
Pinborg, 104 n. 27, 105, 105 n. 31, 111
n. 50, 223 n. 7, 261 n. 6, 262, 381,
412, 416, 423424
Pini, vvi, 76, 76 n. 14, 77 n. 15, 79
n. 25, 95 n. 102, 119 n. 3, 121 n. 9,
129 n. 35, 131 n. 43, 144 n. 76, 145,
152, 155 n. 17, 156 n. 19, 160 n. 26,
166 n. 34, 183 n. 1, 185, 189 n. 12,
192 n. 23, 194 n. 29, 197 n. 40, 215
n. 89, 216, 216 n. 93, 238 n. 54,
239 nn. 5556, 240 n. 60, 243, 243
n. 66, 244 n. 72, 247 n. 80, 248249,
249 n. 83, 250251, 251 n. 92, 253
n. 94, 254, 254 n. 98, 258 n. 105,
259261, 261 n. 4, 262, 262 nn.
1012, 263264, 264 n. 18, 265, 265
n. 20, 266268, 273275, 301302,
318 n. 5, 424
Plato, 1315, 15 n. 24, 1617, 17 nn.
34, 36, 20, 23, 2526, 27 n. 72, 28,
34, 36, 41, 43, 67, 6970, 83, 122,
194, 205, 207, 207 n. 71, 265, 299,
299 n. 13, 301, 301 n. 22, 309, 313,
387388, 411
Platonism, vii
Platonist, 25, 25 n. 63
Plotinus, 21, 21 n. 52, 2223, 25, 26
n. 67, 31, 36, 38, 42, 149
Porphyry, vii, 9 n. 1, 1315, 15 n. 23,
17 nn. 3435, 19 n. 45, 21, 21 n. 53,
2223, 23 n. 56, 24, 24 n. 59, 2526,
26 n. 67, 27, 27 n. 71, 28, 28 n. 76,
32, 40, 43 n. 47, 44 n. 50, 46, 66, 73
n. 3, 74 n. 7, 77, 86, 90, 94 n. 101,
9596, 96 n. 107, 100, 145, 149150,
167168, 170 n. 42, 178, 180, 213,
223 n. 6, 225, 229, 234, 242, 245,
248 n. 82, 249 n. 86, 300 n. 18, 301
n. 20, 302, 342, 369, 414, 421, 424,
427
Porro, 198 n. 43, 412, 424
Powers, 64, 64 n. 113, 424
Presocratics, 11
Priscianus, 361 n. 7, 424
Proclus, 14, 2324, 24 n. 62, 25, 25
n. 66, 26 n. 67, 27 nn. 7273, 31,
38, 424
Pseudo-Dionysius, 24, 25 n. 66
Pseudo-Gilbert of Poitiers, 424
Pythagorean, 16
438 index
Quine, 35
Rashed, 24 n. 62, 424
Raven, 34
Reeve, 35 n. 21, 424
Rescher, 68 nn. 125, 127, 424
Robert Kilwardby, 75 n. 10, 96 n. 107,
97 n. 112, 152, 152 n. 15, 419
Robertus Anglicus, 104 n. 23, 152, 420
Robson, 181 n. 56, 393 n. 66, 422, 424
Rode, 388 n. 52, 424
Ross, 34, 90 n. 81, 298 n. 10
Ruland, 44 n. 50, 424
Sabra, 79 n. 24, 97 n. 113, 424
Santos Noya, 411, 424
Scheu, 129 n. 35, 424
Schmidt, 186 n. 5, 223 nn. 67, 226,
226 n. 18, 227, 227 n. 22, 228 n. 27,
229, 229 n. 31, 233 n. 42, 424
Schoeld, 34
Schnberger, 206 n. 65, 425
Scotus, vi, viii, 10 n. 5, 11, 71, 76 n. 14,
77 n. 15, 79, 79 n. 25, 84 n. 56, 94
n. 101, 95 n. 102, 104 n. 24, 105,
105 n. 29, 119 nn. 34, 121 n. 9, 129
n. 35, 131 n. 43, 147, 152, 155, 155
n. 18, 156, 158, 158 n. 23, 159160,
160 n. 26, 162, 165, 166 n. 34, 171,
173174, 174 n. 49, 175, 183 n. 1,
184185, 189 n. 12, 192, 194, 197
n. 40, 215 n. 89, 216 n. 93, 221,
223, 223 n. 6, 225, 229, 231232,
238 n. 54, 239, 239 n. 55, 240 n. 60,
241243, 243 n. 66, 244, 244 n. 72,
245, 245 n. 75, 246247, 247 n. 80,
249, 249 nn. 83, 85, 250252, 253
n. 94, 254, 254 nn. 95, 98, 255257,
258 n. 105, 259260, 260 n. 2,
261264, 264 nn. 16, 19, 265, 265
n. 22, 266267, 267 n. 27, 268269,
269 n. 28, 270, 270 n. 30, 271, 271
n. 33, 272274, 297, 301 n. 20, 302,
318, 321, 342, 360, 370, 372373,
373 n. 7, 374, 374 n. 8, 375, 375 nn.
1113, 376, 378 n. 19, 382, 392, 399,
402, 406, 408, 413, 419, 424
Segonds, 27 nn. 7273
Senko, 102, 102 n. 17, 104 nn. 2122,
425
Sharpe, 162, 162 n. 28
Shiel, 317 n. 2, 425
Shields, 63, 63 nn. 106107, 64 n. 109,
120 n. 7, 423424
Siger of Brabant, 11, 413
Siger of Courtrai, 4, 11, 100, 103104,
106, 109110, 113, 115116, 415
Simon, v, 69 n. 129, 80 n. 34, 183185,
185 n. 3, 186 nn. 56, 187, 187
n. 10, 188, 188 n. 11, 189, 189 n. 12,
190, 190 n. 15, 191, 191 n. 17, 192,
192 n. 21, 193, 193 n. 26, 194, 194
nn. 27, 29, 195, 197, 197 n. 40, 198,
198 n. 42, 199, 199 n. 45, 200, 200
nn. 47, 49, 201, 206207, 207 n. 69,
208213, 213 n. 84, 215218, 218
n. 99, 219220, 232, 236237, 237
n. 50, 238, 238 n. 54, 239, 239 nn.
5556, 240, 240 n. 58, 252, 318, 318
n. 4, 322, 322 n. 18, 339, 339 n. 76,
370, 372373, 373 n. 7, 374 nn. 89,
377, 406, 413, 425
Simplicius, v, 2, 9, 9 n. 1, 10, 10 nn. 3,
5, 11, 11 nn. 79, 12, 12 n. 12, 13,
13 nn. 1720, 1415, 15 n. 26, 16,
16 nn. 2728, 17, 17 n. 31, 18, 18
n. 43, 1921, 23, 23 nn. 5657, 24
n. 61, 2526, 26 n. 67, 27, 27 n. 74,
29, 36 n. 22, 40 n. 36, 41 n. 38, 44,
44 nn. 4950, 45, 45 nn. 5355, 48,
48 n. 65, 4951, 51 n. 72, 5254, 54
n. 79, 56, 59, 60 n. 98, 66, 119, 119
n. 2, 148 n. 4, 167, 183, 221, 373,
414, 417419, 421, 423, 425, 429
Socrates, 27 n. 70, 86, 122, 124129,
129 n. 35, 131, 133134, 137138,
138 n. 60, 143, 155, 205, 207, 207
n. 71, 208, 208 n. 73, 225226, 228,
265266, 268, 296, 298, 299 n. 13,
304, 306, 306 n. 40, 307, 309311,
313, 315, 328329, 334, 336, 344,
346, 374, 374 n. 9, 377, 387,
396397, 400, 403, 418
Solmsen, 148 n. 4, 150 n. 7, 425
Sophonias, 48 n. 65
Sosigenes, 68
Spade, 163 n. 30, 269 n. 28, 319 n. 6,
380 n. 27, 393 n. 66, 412, 420, 423,
425
Stehkmper, 74 n. 6, 425
Stoic, 23, 66, 146, 148, 180
Strange, 21 n. 52, 45 n. 54, 150 n. 7,
301 n. 20, 302 n. 25, 424425
Stump, 120 n. 7, 121 n. 10, 420, 425
index 439
Sutton, 155, 155 n. 18, 318 n. 3, 321
n. 16, 372 n. 4, 374 n. 9, 377 n. 19,
378 n. 20, 426
Syrianus, 26 n. 67, 49
Szlezk, 44 n. 49, 425
Tabarroni, 100, 100 n. 5, 101, 101 nn.
9, 13, 102, 116 n. 64, 319 n. 5, 414,
425
Tardieu, 24 n. 59, 28, 425
Tarteys, 162, 382, 398 n. 79, 399 nn.
8081, 400, 400 n. 82, 401
Taylor, 34, 34 n. 11, 425
Themistius, 25, 42, 44 n. 52, 66
Theophrastus, 70
Thomas of Erfurt, 100, 103, 105, 105
nn. 32, 34, 106, 116, 412, 426
Thomas of Strasbourg, 11
Thomas Sutton, 11, 76 n. 13, 155, 155
nn. 1819, 183 n. 2, 318 n. 3, 321,
321 n. 16, 370, 372 nn. 45, 374
n. 9, 377 n. 19, 378 n. 20, 426
Tomarchio, 130 n. 38, 426
Torrell, 9 n. 2, 426
Tremblay, v, 3, 73, 73 n. 1, 74 n. 5, 80
n. 30, 246 n. 78, 426, 431
Urvoy, 31 n. 3, 426
Vallat, 12, 13 n. 18, 14 n. 21, 15 nn.
2425, 17, 17 nn. 3637, 19, 19 nn.
4648, 25 nn. 64, 66, 426
Van den Bergh, 67 nn. 119, 121, 68 nn.
122123, 69 nn. 129130, 132
van der Helm, 84 n. 56, 119 n. 4, 121
n. 9, 129 n. 35, 413
Van Steenberghen, 31 n. 2, 151 n. 12,
426
Verhaak, 104, 104 n. 26, 105 n. 30,
106, 106 n. 36, 109 n. 45, 415, 426
Versor, 74 n. 7, 94 n. 101
Victorinus, 24 n. 59, 418, 425
Vittorini, 319 n. 5, 385 n. 32, 426
von Perger, 322 n. 19, 426
Wagner, 49 n. 68, 426
Wallace, 73 n. 2, 426
Wallerand, 104 n. 25, 426
Walter Burley, 4, 147, 319 n. 5, 322
n. 19, 339 n. 77, 344 n. 94, 370 n. 2,
376 n. 15, 382 n. 32, 414415, 423,
426
Walton, 121 n. 10, 139 n. 65, 427
Walzer, 17 nn. 3435, 28 n. 76, 427
Washell, 77 n. 15, 427
Weinberg, 44 n. 50, 427
Whelpdale, 162, 382, 398 n. 79, 399
nn. 8081, 401
Wians, 34 n. 15, 70 n. 134, 419, 427
Wieland, 80 n. 33, 427
William of Moerbeke, 9
William of Saint-Amour, 100
Wippel, 120121, 121 n. 9, 122, 122
n. 12, 123, 123 n. 17, 124125, 125
n. 21, 127129, 129 n. 35, 203 n. 58,
205 n. 67, 210 n. 77, 297, 321 n. 15,
375 n. 10, 427
Wlodek, 80 n. 33, 82 n. 46, 427
Wolter, 249 n. 85, 254 n. 95, 419, 427
Wood, 104 n. 24, 105 n. 29, 267 n. 27,
298 n. 10, 419, 423
Wright, 53 n. 76, 427
Wyclif, 7, 319, 319 nn. 67, 321, 321
n. 13, 323, 323 n. 22, 324325, 325
n. 30, 327328, 328 n. 40, 329, 331,
331 n. 57, 332 n. 59, 336, 339, 339
n. 78, 340, 340 n. 82, 344345, 345
n. 95, 346, 376 n. 14, 380, 382, 392,
392 n. 65, 393, 393 nn. 6667, 394,
394 nn. 6869, 395, 395 nn. 7074,
396, 396 nn. 7275, 397, 397 nn.
7678, 398 n. 79, 399401, 401 n.
84, 402404, 406, 414415, 419, 421,
424425, 427
Zedler, 69 n. 129, 427
Zeno, 34
Zimmermann, 73 n. 4, 80 n. 33, 191
n. 17, 199 n. 46, 427
Zupko, 105 n. 33, 163 n. 29, 295, 295
n. 1, 299, 300 n. 15, 303 n. 30, 305
n. 38, 413, 420, 427
Brills Companions
to the
Christian Tradition
Volumes deal with persons, movements, schools and genres in medieval and early modern
Christian life, thought and practice. Written by the foremost specialists in the respective
fields, they aim to provide full balanced accounts at an advanced level, as well as synthesis
of debate and the state of scholarship in 8-15 substantial chapters. Volumes are in English
(contributions from continental scholars are translated). 2-3 volumes of 350-600 pages will
be published each year. The series is expected to comprise 20-30 volumes.
1. Schabel, C. (ed.). Theological Quodlibeta in the Middle Ages. The Thirteenth Century.
2006. ISBN 978 90 04 12333 5
2. Cox, V. & J.O. Ward (eds.). The Rhetoric of Cicero in Its Medieval and Early Renaissance
Commentary Tradition. 2006. ISBN 978 90 04 13177 4
3. McGuire, B.P. (ed.). A Companion to Jean Gerson. 2006. ISBN 978 90 04 15009 6
4. Levy, I.C. (ed.). A Companion to John Wyclif. Late Medieval Theologian. 2006.
ISBN 978 90 04 15007 2
5. Swanson, R.N. (ed.). Promissory Notes on the Treasury of Merits. Indulgences in Late
Medieval Europe. 2006. ISBN 978 90 04 15287 8
6. Stayer, J.M. & J.D. Roth (eds.). A Companion to Anabaptism and Spiritualism, 1521-1700.
2006. ISBN 978 90 04 15402 5
7. Schabel, C. (ed.). Theological Quodlibeta in the Middle Ages. The Fourteenth Century.
2007. ISBN 978 90 04 16288 4
8. Kirby, T. (ed.). A Companion to Richard Hooker. With a Foreword by R. Williams. 2008.
ISBN 978 90 04 16534 2
9. Rummel, E. (ed.). Biblical Humanism and Scholasticism in the Age of Erasmus. 2008.
ISBN 978 90 04 14573 3
10. Newton, L.A. (ed.). Medieval Commentaries on Aristotles Categories. 2008.
ISBN 978 90 04 16752 0
11. Kolb, R. (ed.). Lutheran Ecclesiastical Culture, 1550-1675. 2008.
ISBN 978 90 04 16641 7

Lloyd A. Newton Medieval Commentaries On Aristotles Categories Brills Companions To The Christian Tradition 2008

Uploaded by

Copyright:

Available Formats

Lloyd A. Newton Medieval Commentaries On Aristotles Categories Brills Companions To The Christian Tradition 2008

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Lloyd A. Newton Medieval Commentaries On Aristotles Categories Brills Companions To The Christian Tradition 2008

Uploaded by

Copyright:

Available Formats

What is the book about?

What topics are covered in the book?

Medieval Commentaries on

iii] For Avicenna these would suggest that

c), in combination with

vi, 75 n. 10, 99, 99 n. 1,

You might also like