Burridan - Treatises

JEAN BURIDAN'S LOGIC
THE TREATISE ON SUPPOSITION
THE TREATISE ON CONSEQUENCES

SYNTHESE HISTORICAL LIBRARY
TEXTS AND STUDIES IN THE HISTORY OF
LOGIC AND PHILOSOPHY
Editors:
N. KRETZMANN, Cornell University
G. NUCHELMANS, UniversityofLeyden
L. M. DE RIJK, University of Leyden
Editorial Board:
J. BERG, Munich Institute of Technology
F. DEL PUN T A, Linacre College, Oxford

D. P. HENRY, University of Manchester
J. HINTIKKA, Florida State University, Tallahassee
B. MATE S, University of California, Berkeley
J. E. MURDOCH, Harvard University

G. P A TZ I G, University of Gottingen
VOLUME 27
JEAN BURIDAN'S LOGIC
THE TREATISE ON SUPPOSITION

THE TREATISE ON CONSEQUENCES
Translated, With a Philosophical Introduction

by
PETER KING
University of Pittsburgh
D. REIDEL PUBLISHING COMPANY

A MEMBER OFTHE KLUWER ACADEMIC PUBLISHERS GROUP
DORDRECHT / BOSTON / LANCASTER /TOKYO

Library of Congress Cataloging in Publication Data
Buridan, Jean, 1300-1358.
Jean Buridan's Logic.
(Synthese historical library ; v. 27)

Bibliography: p.
Includes indexes.
I. Logic-Early works to 1800. I. King,Peter,1955-
II. Buridan, Jean 1300-1358. Tractatus de suppositionibus. English.
1985. III. Buridan, Jean, 1300 -135 8. Tractatus de consequentiis.
English. 1985. IV. Title. V. Title: Logic. VI. Series.
BC60.B875 1985 160 85-24383
BC60.B875 1985
Published by D. Reidel Publishing Company,

P.O. Box 17,3300 AA Dordrecht, Holland
Sold and distributed in the U.S.A. and Canada

by Kluwer Academic Publishers,
190 Old Derby Street, Hingham, MA 02043, U.S.A.
In all othcr countries, sold and distributed

by Kluwer Academic Publishers Group,
P.O. Box 322, 330() AH Dordrecht, Holland.
All Rights Reserved

© 1985 by D. Reidel Publishing Company, Dordrecht, Holland
No part of the material protected by this copyright notice may be reproduced or
utilized in any form or by any means, electronic or mechanical
including photocopying, recording or by any information storage and
retrieval system, without written permission from the copyright owner
ISBN -13:978-94-010-8836-7 e-ISBN -13:978-94-009-5289-8
DOl: 10,1007/978-94-009-5289-8
Softcover reprint of the hardcover 1st edition 1985

Et credo quod tanta fuit orta controversia
inter opinantes ex defectu logicae ...
- Buridan, MS B.N.lat.l6128 fo1.232 VO

TABLE OF CONTENTS
Preface Xl
INTRODUCTION. BURIDAN'S PHILOSOPHY OF LOGIC
Section 1. John Buridan: Life and Times 3

Section 2. The Treatises 5
Section 3. Meaning and Mental Language 7
3.1. Levels of Language 7
3.2. Nominalist Semantics and Equiformity 8
3.3. Mental as an Ideal Language 10
Section 4. The Properties of Terms 15
4.1. Syncategorematic Terms 15
4.2. Absolute and Appellative Terms 17
4.3. Intentional Verbs 22
Section 5. Sentences 25
5.1. Sentences as Assertions 25
5.2. Categorical Sentences
5.3. Hypothetical Sentences 29
5.4. Principles of Sentential Logic 30
5.5. Truth and Sentential Signification 32
Section 6. The Theory of Supposition 35
6.1. Supposition and the Theory of Reference 35
6.2. Personal and Material Supposition 37
6.3. Discrete and Common Supposition 40
6.4. Absolute and Relative Supposition 41
6.5. Natural and Accidental Supposition 43
6.6. Determinate and Confused Supposition 45
6.7. Distributive and Non-Distributive Supposition 48
6.8. Ampliation: Time and Modality 51
6.9. Truth-Conditions 57
vii
Vlll TABLE OF CONTENTS
Section 7. Consequences 59
7.1. Conditionals, Inferences, and Consequences 59
7.2. The Definition and Division of Consequences 61
7.3. Assertoric Consequences 63
7.4. Divided Modal Consequences 65
7.5. Composite Modal Consequences 66
Section 8. The Syllogism 69
8.1. Definition of the Syllogism 69
8.2. Syllogistic Semantic Principles 70
8.3. Reduction and Proof-Procedure 73
8.4. Assertoric Syllogistic 75
8.5. Composite Modal Syllogistic 79
8.6. Divided Modal Syllogistic 80
TRANSLATION. THE TREATISE ON SUPPOSITION

Introduction 85
Chapter 1. Signification, Supposition, Verification, Appellation 86
1.1. The Aim of Chapter 1 86
1..... Signification and Supposition 86
1.J. Supposition and Verification 88
1.4. Supposition and Appellation 90
Chapter 2. Kinds of Significative Words 93
2.1. Complex and Incomplex 93
2.2. Subject and Predicate 93
2.3. Categorematic and Syncategorematic Terms 96
2.4. Complex and Incomplex Concepts 98
2.5. Perfect and Imperfect Expressions 99
2.6. Rules for Supposition 100
Chapter 3. The Kinds of Supposition 117
3.1. Proper and Improper Supposition 117
3.2. Material and Personal Supposition 118
3.3. Common and Discrete Personal Supposition 125
3.4. Natural and Accidental Supposition 125
3.5. Confused and Determinate Supposition 129
3.6. Distributive and Merely Confused Supposition 130
3.7. Rules for Distributive Confusion 131
3.8. Rules for Non-Distributive Confusion 140
TABLE OF CONTENTS lX
Chapter 4. The Supposition of Relative Terms 148

4.1. The Meaning of 'Relative Term' 148
4.2. The Kinds of Relative Terms 149
4.3. Reference to the Antecedent 150
4.4. Supposition of the Antecedent 150
4.5. 'This' and 'That' 152
4.6. 'Himself' 153
4.7. 'His Own' 154
4.8. 'Such' and 'So Much' 155
4.9. Differentiating Relative-Terms 155
Chapter 5. Appellation 159
5.1. Review 159
5.2. The Appellation of Subject and Predicate 159
5.3. Appellation and Signification 162
5.4. Appellation of Particular Terms 166
Chapter 6. Ampliation and Restriction 168
6.1. The Status of a Term 168
6.2. Ampliation 168
6.3. Restriction 170
6.4. Alienation 171
6.5. The Alienation of Supposition 172
6.6. Cancelled Supposition 172
TRANSLATION. THE TREATISE ON CONSEQUENCES
Book I. Consequences in General and Among Assertoric

Sentences 177
1.1. The Truth and Falsity of Sentences 177
1.2. The Causes of the Truth and Falsity of Sentences 179
1.3. The Definition of 'Consequence' 181
1.4. The Kinds of Consequences 184
1.5. The Supposition of Terms 186
1.6. The Ampliation of Terms 188
1.7. The Matter and Form of Sentences 194
1.8. Theorem About Assertoric Consequences 195
Book II. Consequences Among Modal Sentences 228
2.1. Modal Sentences 228
x TABLE OF CONTENTS
2.2. Composite and Divided Modal Sentences 228

2.3. Affirmative arid Negative Divided Modals 229
2.4. Ampliated Terms in Divided Modal Sentences 231
2.5. Equipollent Modal Sentences 233
2.6. Theorems About Divided Modals 234
2.7. Theorems About Composite Modals 245
Book III. Syllogisms With Assertoric Sentences 258
3.1. The Kinds of Consequences 258
3.2. The Syllogism 260
3.3. Finite and Infinite Terms 262
3.4. Theorems 263
3.5. Sentences With Oblique Terms 278
3.6. Syllogistic Extremes and Middle 280
3.7. Theorems 281
Book IV. Syllogisms with Modal Sentences 295
4.1. Syllogisms With Composite Modal Sentences 295
4.2. Syllogisms With Divided Modal Sentences 298
4.3. Divided Contingent Modals 315
4.4. Syllogisms With Reduplicative Sentences 320
NOTES
Notes. Buridan's Philosophy of Logic 327

Notes. Treatise on Supposition 336
Notes. Treatise on Consequences 343
Book I. Notes 343
Book II. Notes 347
Book III. Notes 349
Book IV. Notes 354
Bibliography 367
Indexes 369
Index of Names 369
Index of Subjects 371
Index of Rules and Theorems 377
PREFACE
Buridan was a brilliant logician in an age of brilliant logicians, sensitive

to formal and philosophical considerations. There is a need for critical
editions and accurate translations of his works, for his philosophical
voice speaks directly across the ages to problems of concern to analytic
philosophers today. But his idiom is unfamiliar, so editions and trans-
lations alone will not bridge the gap of centuries. I have tried to make
Buridan accessible to philosophers and logicians today by the introduc-
tory essay, in which I survey Buridan's philosophy of logic. Several
problems which Buridan touches on only marginally in the works trans-
lated herein are developed and discussed, citing other works of Buridan;
some topics which he treats at length in the translated works, such as the
semantic theory of oblique terms, I have touched on lightly or not at all.
Such distortions are inevitable, and I hope that the idiosyncracies of my
choice of philosophically relevant topics will not blind the reader to
other topics of value Buridan considers.
My goal in translating has been to produce an accurate renaering of
the Latin. Often Buridan will couch a logical rule in terms of the
grammatical form of a sentence, and I have endeavored to keep the
translation consistent. Some strained phrases result, such as "A man I
know" having a different logic from "I know a man." This awkwardness
cannot always be avoided, and I beg the reader's indulgence. All of the
translations here are my own.
I have deviated from faithful accuracy in two inconsequential ways. In
place of the punctuation given by the editors I have changed the punc-
tuation where it seemed to me the rules of English, good sense, or good
logic demanded. Since the punctuation in mediaeval manuscripts is
virtually nonexistent, I have no hesitation in altering previous editorial
decisions. Second, I have tried to replace Buridan's clumsy system of
imprecise cross-references with an exact numerical system. I cannot
believe Buridan would not have used this system were it available to
him, and I make no apologies for exploiting it fully in my translation.
My debts are many. I would like to thank Nadia Gerstenkorn, on
whose unfamiliar French typewriter keyboard I first began translating
Xl
xu PREFACE
Buridan's treatises, Jim van Aken and Bob Brandom discussed condi-
tionals at great length with me, lowe a special debt of gratitude to
Calvin Normore, Paul Spade, and Anna Greco, each of whom read this
work in manuscript and made many valuable suggestions, I also want to
thank the students in my graduate seminar on Mediaeval Logic, given in
Fall 1983 at the University of Pittsburgh, who stimulated my thinking
about many of the philosophical issues explored in the succeeding
pages,
INTRODUCTION:
BURIDAN'S PHILOSOPHY OF LOGIC

1. JOHN BURIDAN: LIFE AND TIMES
Buridan is best-known to philosophers for the example of "Buridan's

Ass," starving to death between two equidistant and equally tempting
bales of hay, who appears in Spinoza, Ethica II, scholium to prop. 49.
But this poor fragment of Buridan's great reputation is as apocryphal
as his supposed amorous adventures with the Queen of France,
famous from Franc;ois Villon's poem "La testament," or his founding of
the University of Vienna: Buridan's ass is not to be found in Buridan,
though his examples are studded with asses. l
Our knowledge of Buridan's life is sketchy." We know that he was
French, but little else about his background; various examples in his
writings suggest a man familiar with Picardy. Just as we do not know
where Buridan was born, we do not know when he has born. He must
have been born by 1300, but this is the only reliable inference we
can make.
Buridan is first glimpsed in the pages of history in 1328, the rector
of the University of Paris, vir venerabilis et discretus, presiding over a
debate which took place on February 9. The next year, on 30 August
1329, he received a benefice from Pope John XXII; on 2 November
1330 he received another benefice from the same Pope, who addressed
him each time as a Master of Arts. We then lose sight of him until 25
September 1339, when Buridan was a signatory to a condemnation of
certain doctrines (supposedly including those of William of Ockham);
during this period he received an expectation of a prebend from Pope
Benedict XII. In 1340 he was again rector of the University of Paris.
The last time he graces the pages of history is as a figure signing a
border treaty under the authority of the University, on 12 July 1358.
From several remarks (e.g. TC 3.4.14) we know Buridan spent
his life as a career Master in the Faculty of Arts - a rarity, for the
Faculty of Arts was generally made up of students who were going on to
advanced study in theology, and there was a fast turnover.
These few facts are all that we know of Buridan's life. Yet we
possess his works (in large measure), and in them there is a wealth of
material for the philosopher. Buridan's influence and reputation were
3
4 BURIDAN'S PHILOSOPHY OF LOGIC
immense, both during his life and for centuries afterwards. He was
known for his contributions to ethics, physics, and, perhaps most
important, for his philosophy of logic. It is the latter to which this
introduction and the translations are devoted. Buridan's mediaeval
voice speaks directly to modern concerns: the attempt to create
a genuinely nominalist semantics; paradoxes of self-reference; the
nature of inferential connections; canonical language; meaning and
reference; the theory of valid argument. It is to be hoped that Buridan
can reclaim his lost reputation among contemporary philosophers for
his penetrating and incisive views on these and other matters.
2. THE TREATISES
The "Treatise on Supposition" [TS] is the fourth treatise of a much

longer work known as the Summulae de dialectica, the contents of
which Buridan himself describes in the beginning of the first chapter of
the first treatise:
We divide this work into nine treatises, of which the first

will be about sentences and their parts and passions;
the second about the predicables; the third about the
categories; the fourth about supposition; the fifth about the
syllogism; the sixth about dialectical loci; and the seventh
about fallacies. An eighth <treatise> about division, definition,
and demonstration is added, which our author did not deal
with in his book; the ninth will be about the practice of
sophisms - but in my lectures I shall not follow the other
lectures with this last treatise.
"Our author" is Peter of Spain,3 and "his book" is the Summulae

logicales. Buridan adopted this work as the basis of his lectures on logic,
for the reason he states in the last sentence of the Proemium to the
Summulae de dialectica:
Wishing to say certain general things about the whole of

logic without excessively painstaking investigation, I shall
particularly rely upon the brief treatise of logic which
the venerable Doctor, Master Peter of Spain, has already
composed, analyzing and supplementing what he wrote
and said in another way when at times it seems to me
opportune.
The Summulae de dialectica is written as a commentary on Peter of

Spain. But there were topics Peter of Spain said little or nothing about;
and sometimes what he did say Buridan regarded as hopeless. In the
former case Buridan wrote an independent treatise, as he remarks for the
5
eighth and ninth treatises listed. In the latter case Buridan simply jettisons
the text written by Peter of Spain and substitutes his own text, comment-
ing upon it. Such a case is the fourth treatise, the treatise on supposition.
The Summulae de dialectica is one of two or three major logical
works we have of Buridan's: the ninth treatise seems to have been
considered an independent work, called the Sophismata. 4 The other
work is the Treatise on Consequences [TC], translated here, an advanced
independent investigation in logic. If TS was the textbook for Buridan's
introductory course on logic, TC is a handbook to the logic graduate
seminar. The rest of Buridan's works on logic are quaestiones on the
standard logical corpus: Porphyry's Isagoge, and Aristotle's Categories,
De interpretatione, Prior Analytics, Posterior Analytics, Topics, and
Rhetoric. 5
For TS I have used the edition of the text given by Maria Elena
Reina in "Giovanni Buridano: Tractatus de suppositionibus," Rivista
critica di storia della filosofia (1959), pp. 175-208 and pp. 323-352.
For TC I have used the edition of the text given by Hubert Hubien,
Iohannis Buridani tractatus de consequentiis: Edition critique, in the
series Philosophes medievaux Vol. XVI, Universite de Louvain 1976.
Divergences from these texts are noted in the translation where they
occur.
Abbreviations
TS = Treatise on Supposition
TC = Treatise on Consequences
QM = Questions on the Metaphysics
QCM = Questions on the De caelo et mundo
QSP = Questions on the Physics
QNE = Questions on the Nicomachean Ethics
3. MEANING AND MENTAL LANGUAGE
3.1. LEVELS OF LANGUAGE
Buridan and other logicians of the fourteenth century were inspired by a

remark Aristotle made in De interpretatione cap. 1 16a3-8°:
Therefore, the things that are in speech are the marks
(notae) of the passions that are in the soul, and the things
that are written are <the marks> of those that are in speech.
And just as letters are not the same for all people, so the
utterances are not the same. But the first things of which
these utterances are marks are passions of the soul, the same
for all people, and the things of which the latter are
likenesses (similitudines) - real things - are also the same
<for all people>.
There are three distinct levels of language: Written, Spoken, and
Mental. Each of these is a fully developed language in its own right,
with vocabulary, syntax, formation-rules, and the like. These languages
are hierarchically ordered; elements of Written Language (inscriptions)
are the "marks" of Spoken Language, elements of Spoken Language
(utterances) are the "marks" of Mental Language. The elements of
Mental Language are "likenesses" of real things. Buridan will call these
likenesses concepts, which are mental particulars, "acts of the soul" (OM
V q. 9 fol. 33rb). The relationship of one language to another is not
holistic, but piecemeal. A particular inscription will be related to a
particular utterance, a particular utterance to a particular concept.
Buridan calls the relations which obtain among these languages
signification. A written word immediately signifies a spoken expression;
a spoken word immediately signifies a concept or concepts of the mind;
the concepts of the mind signify those things of which they are the
natural likenesses. The written and the spoken term have an "ultimate"
signification, namely what is conceived by the concept (TS 3.2.8,
Sophismata I Theorem 2). Clearly, a change in the signification of a
term of one level will change the (ultimate) signification of the lower-
7
level terms which signify it. The relation among the various levels, then,
is rather like encoding.? Immediate signification is conventional, but the
signification of concepts is natural, and indeed universal; concepts are
"the same for all."
3.2. NOMINALIST SEMANTICS AND EQUIFORMITY
Logic, for Buridan, will include Aristotelian psychology as a component

(QM VI q. 12 fol. 41 vb): the constituent elements of Mental are
concepts, and Buridan holds substantive doctrines about (i) the causal
theory of concept acquisition, and (ii) the activity of the mind in
combining concepts. Logic, of course, is not simply psychology, for logic
includes a normative dimension.
The central theme dominating Buridan's logic, and indeed his
approach to philosophy in general, is nominalism. Mediaeval nominal-
ism took many different forms, but at root is the denial of universal or
metaphysically shared entities. H This has several forms in Buridan: he
will deny the existence of universals; he will minimize the number of
categories; he will reject various 'abstract' entities, such as propositions.
This last is important, and will be discussed in more detail in Section
5.5; what is relevant to our purposes here is Buridan's attempt to create
a nominalist semantics, a theory of logic and language that is based on
particular inscriptions, utterances, and thoughts. The crucial notion in
this enterprise is equiformity.Y
Buridan rejects any abstract notion of 'proposition' to serve as truth-
bearer and fundamental constituent of semantic analysis, and replaces it
with a theory of logic and language designed to apply to the individual
inscription. to Of course, all inscriptions are particular individuals,
discrete and different from one another. Buridan is thus faced with a
familiar dilemma: the laws of logic are general, and govern classes or
types or sets of inscriptions, not individual inscription-tokens; 'class;
'type; and 'set' are abstract entities. Now whether mediaeval nominalism
ought to countenance such entities is not clear: they are not obviously
the sort of metaphysically shared item it proscribes. But they do most of
the same work, and they are abstract, and suspiciously so; mediaeval
nominalists should look on them with a jaundiced eye. But how is logic
possible without recourse to such abstract notions?
There are many familiar attempts to escape the horns of this
3. MEANING AND MENTAL LANGUAGE 9
dilemma. Quine, for example, endorses sets, but argues that as abstract
entities go they are pretty well-behaved (so well-behaved that he insists
on calling himself a 'nominalist'). Mediaeval nominalism might take the
same route, although then it must account for the ontological status of
such abstract entities. Others have tried to give reductivist accounts of
'set' or 'class' or 'type,' suggesting that they are no more than reducible
equivalence classes based on resemblance or similarity. This is Buridan's
approach.
But according to the latter, is not 'resemblance' equally an abstract
entity, a universal of the sort proscribed? Buridan avoids reifying
'resemblance' by interpreting it pragmatically. Each inscription-token is
unique. Logical laws are stated for inscriptions which are similar
(similis), that is, sufficiently resemble one another in the relevant
respects. But there is no saying what respects are the relevant ones, or
which degrees of resemblance are sufficient; these factors depend on the
context, on our interests and aims. Inscriptions which are treated as the
same in a given context are called "equiform," but there is no such thing
as equiformity tout court, and so Buridan's nominalism is not compro-
mised. Logical principles are therefore restricted to equiformity-classes
of inscriptions. In some cases we may be forced to be very restrictive in
our choice of equiformity-class:
The sentence printed below this one is false.
The sentence printed below this one is false.
2+2=17
The first two inscriptions cannot be equiform with respect to truth-
value, for the first is false and the second true (QM V q. 1 fol. 26vb).
Equally, when discussing the Liar, for example, we have to carefully
distinguish individual inscription-tokens (Sophismata VIII, Hughes
[1982]8.4.3,13.5-6,15.8.2).
Buridan seems to have kept an open mind on how arbitrary an
equiformity-c1ass may be. There may be an eventual limit, imposed by
the causal theory of concept-formation and the requirement that
concepts be natural likenesses of the things of which they are concepts,
but Buridan does not say so explicitly. For most cases a handy practical
criterion will serve to demarcate equiformity-classes; Buridan frequently
uses the rule that terms supposit and appellate in equiform sentences
just as in the original sentence (or part of a sentence: this will be
important for the theory of consequences): see TC 3.7.41.
Buridan's pragmatic interpretation also allows him to deal with

classes of possible inscriptions, In general the laws of logic are not stated
for actual sequences of inscriptions but rather specify what sequences of
inscriptions would be acceptable, if formed, I I This is, of course, an
aspect of the normative element of logic, but for nominalists the point is
deeper: logical laws cover not only actual inscriptions but possible
inscriptions as welL There is no other way to describe basic principles of
logic. For instance, any sentence can have a contradictory, but in actual
fact such a contradictory may never have been inscribed; how then can
we say that a non-actual inscription must be true if its actual contradic-
tory is false? Buridan sensibly states his principles for actual and
possible inscriptions, and, in some cases (such as the distinction of the
possible and the possibly-true), they must be carefully distinguished,
shedding new light on philosophically unexplored terrain.
A final point. The term "equiform' might be read as "having the same
(or similar) logical form: which is narrower than the usage I have given
it. To be sure, Buridan does sometimes use it in this way, but then only
as a special case of the more general sense: the relevant respects of
similarity in question are the series of syncategorematic terms. Unless
specifically noted, I shall follow Buridan in his practice of taking
"equiform' to mean any similarity-class of inscriptions, where the
respects in which the inscriptions are similar are defined contextually.
3,3. MENTAL AS AN IDEAL LANGUAGE
Mental Language is of the utmost importance for Buridan's logic.

For while Mental is a natural language, in a way in which Spoken or
Written are merely conventional, we should not understand this the way
today "natural languages" such as English or French contrast with
symbolic language in rigor; rather, Mental is a natural language
perspicuous in rigor. The basic claim, common to Buridan and many
other fourteenth-century logicians, is that Mental is a canonical
language, an ideal or logically perfect language. 12 This claim involves
five theses:
[11 Mental is a universal language,
[21 Mental is adequate in expressive power.
[31 Mental is disambiguated,
[4] Mental is nonredundant.

[5] Mental sentences display their logical form.
Each thesis calls for further comment.
Ad [1]. The universality of Mental is a matter of its structure, not its

content. That is, we do not necessarily all have the same stock of
concepts; I may completely lack the concept of lion, which you
possess, due to our different past interaction with the world. But the
structure is the same for all, which means roughly that we all have
similar mental abilities: we can all combine simple concepts into
complex concepts, for example. Moreover if two people each have the
same term of Mental, then their concepts differ only numerically: your
concept of a lion and mine may have been acquired through the
experience of different lions, but the concepts are equiform, that is,
for most practical purposes they may be regarded as the same. The
universality of Mental will provide the foundation for logic to be a
full-fledged (Aristotelian) science.
Ad [2]. The adequacy of a language is a matter of its resources; a
language is expressively adequate if it has the resources to express
whatever can be expressed. But Mental is literally the language of
thought: we think in Mental. If we were telepaths we would speak to one
another in Mental; angels, who are telepaths, do SO.I.1 Therefore Mental
is expressively adequate, for whatever is thought must be expressed in
Mental.
This approach will have certain difficulties: it is not immediately
obvious how to handle non-declarative uses of language, performatives,
and certain indexical terms. But the general aim of the project is clear
even if not all of the details are; we shall leave them to one side.
Before we discuss [3] and [4] we need to clarify the role of
translation-rules. A translation-rule correlates expressions of ordinary
language with their perspicuous canonical representations. The best-
known modern example of a translation-rule is Russell's rule about
definite descriptions. where a sentence such as
The present King of France is bald
is correctly "'translated" in first-order logic as
[3x] [Kx & (y) (Ky ~ (x = y»]
For Buridan translation-rules are principles saying which elements of
Mental an inscription or utterance corresponds to. We may call

this "correspondence" subordination. l4 An ambiguous utterance or
inscription is ambiguous because it is subordinated to more than one
concept or complex of concepts in Mental; utterances or inscriptions are
synonymous because they are subordinated to the same concept or
complex of concepts in MentaL The utterance or inscription 'bank'
is ambiguous because it could signify the concept of the side of a
river, or it could signify the concept of an institution dealing in
money. The utterances Tully' and 'Cicero' are synonymous, because
they are subordinated to the same concept, the concept of a particular
individuaL Buridan takes ambiguity and synonymy to be features only of
Spoken and Written. The exact nature of the translation-rules he
proposes is extremely important, for it is a key element of his nominalist
approach. We shall discuss some principles when we discuss the
distinction between absolute and appellative terms in Section 4.2. If the
proper explanation of ambiguity and of synonymy is found in the
translation-rules, then clearly Mental cannot have ambiguous or synony-
mous terms, for there is nothing further to which concepts of Mental are
subordinated.
Ad [31. The terms of Mental are concepts, which have a natural like-
ness to their objects. An ambiguous term in Mental, then, might be a
concept which truly applied to two distinct kinds or groups of things.
But then such a concept will have a natural likeness to each group of
things. We may then have a broader concept than we originally thought,
but not an ambiguous concept. More likely, though, we have a
disjunctive concept, a concept which has internal logical structure. Thus
no term in Mental is ambiguous.
What of amphiboly, that is, ambiguous sentences, in which no single
term is ambiguous, such as "Flying planes can be dangerous"? Buridan
does not want to allow amphiboly in Mental, and equally does not want
to compromise the principle that the translation-rules correlate terms
with concepts. He avoids this problem gracefully by a general rule that
the subjects of Mental sentences always stand for what they signify.l) He
justifies this rule by his realism about Mental as the "language of
thought": to have a concept in mind is to be thinking about what the
concept is a natural likeness of; to have a concept which does not stand
for that of which it is the natural likeness would be for a person to not
think about what he is thinking about - which is impossible.
Ad [41, The nonredundancy of Mental is a matter of its containing
no synonyms. Suppose that 'vixen' and 'female fox' are in fact

synonymous; they then differ only by their internal make-up, which
is a matter of orthography for inscriptions and phonology for
utterances. The analogy for Mental, then, would be to have two
concepts which differ only in their "internal make-up." What could be
analogous to orthography or phonology for concepts? Nothing at all, if
we only consider simple concepts. II> But if we allow complex concepts, it
seems as though Mental must have synonyms, since the concepts
corresponding to 'vixen' and to 'female fox' differ in their logical
structure: the former appears simple, while the latter seems to combine
the two concepts corresponding to 'female' and to 'fox.' Will not logical
structure play the role in Mental played by orthography in Written and
phonetics in Spoken?
Buridan's answer is that it does not, at least not so as to produce
redundancy. The concept corresponding to 'female fox' does have a
complex logical structure, but its signification differs from that of the
concept corresponding to 'vixen.' For the signification of complex
expressions Buridan endorses the Additive Principle (TS 2.3.10,
Sophismata II Theorem 5, OM IV q. 14 fol. 23vb):
The signification of a complex expression is the sum of
the signification of its (categorematic) parts.
Hence the concept corresponding to 'female fox' signifies all females,
foxes or not, and all foxes, female or not. This is quite reasonable if
we think of signification as everything a term brings to mind -
perhaps the original meaning of 'signification.' Of course, 'female fox'
will only stand for vixens; that is not signification, but supposition,
discussed in Section 6.l.
Ad [5]. A sentence of Mental perspicuously displays its logical form.
To see how this is so, let us return to a problem postponed in the
discussion of [4], the problem of complex expressions with the same
logical components, such as "Socrates is taller than Plato" and "Plato is
taller than Socrates." In both expressions we have the same components:
a concept corresponding to Socrates, a concept corresponding to Plato,
and a concept corresponding to 'is taller than.' What distinguishes them?
The utterances are distinguished by the temporal order in which the
phonemes are uttered, and hence by the medium of time; the inscrip-
tions are distinguished by the spatial sequence of the letters, and hence
by the medium of space.
Is there such a discriminatory medium for Mental? Buridan thinks

there is no need for one: such a medium would distinguish "Fred or
Barney" from "Barney or Fred," and this is too fine-grained for
Mental, for the latter are genuine synonyms. The point of such
examples is that certain logical operations are sensitive to order while
others are not. Logical operations in Mental are thus like functions,
taking concepts into expressions, some of which are sensitive to order
(i.e. nonsymmetric) and others not. These 'functions' are themselves
concepts: Buridan calls them complexive concepts (TS 2.3.12-13,
2.4.2- 3), and they act to combine simple or complex concepts into
complex concepts. They correspond to logical operators, that is,
syncategoremata, discussed in Section 4.1. The complexive concepts
which are the most important correspond to the two forms of the
copula, to 'is' and 'is not,' which combine concepts to produce a
Mental sentence. l 7
The complexive Mental concept corresponding to the affirmative
copula will be nonsymmetric: that is what allows Buridan to construct
different theoretical roles for the subject and predicate. The concept
corresponding to 'is taller than' will be asymmetric, so TALLER
(Socrates, Plato) implies NOT-TALLER (Plato, Socrates). The com-
plexive concept corresponding to conjunction is symmetric: AND (Fred,
Barney) = AND (Barney, Fred).
Thus expressions of Mental are individuated by their semantic
properties alone; Mental therefore behaves like the semantics of
Spoken or Written. The logical form of a Mental sentence, then, is
displayed in its essential composition. Note that this need not be
taken as a literal composition: the thought of a complex need not be a
complex of thoughts. But the thought must have an internal logical
articulation as we have described it.
Mental is thus a canonical language in virtue of [11-[51. Hence it is
the primary concern of the logician. The surface structure of Mental is
identical with its logical structure, and reflects the logical form of
Spoken or Written.
4. THE PROPERTIES OF TERMS
4.1. SYNCATEGOREMATIC TERMS
To have a theory of logical form, we have to distinguish logical from

non-logical constants; this is roughly the distinction between syn-
categorematic and categorematic terms. We take such a difference as
primitive: logical constants are listed separately and they appear in the
syntactical rules in special ways. Buridan draws the distinction in terms
of signification. Purely syncategorematic terms lack an ultimate signifi-
cation, but this does not mean they lack all signification; every word
which can be put into a sentence is imposed for some signification (TS
2.3.7). Syncategoremata have only an immediate signification, and only
have an ultimate signification in combination with categorematic terms
(TS 2.3.6).
What do syncategoremata immediately signify? Purely syncategore-
matic terms signify simple concepts which are complexive: the semantic
functors described at the end of Section 3.3 (TS 2.3.11-143). Since
syncategoremata do not have ultimate signification, they do not change
the ultimate signification of a sentence; "Some men are sexists" and "No
men are sexists" signify the same, as indeed do "God is God," "God is
not God" and the term 'God.' The logical form of a sentence can be
described as the exact series of syncategoremata (TC 1.7.3-4).
Some syncategorematic terms operate on terms to produce terms;
others operate on sentences to produce sentences (TS 2.4.4). Whichever
we examine, at the heart of all logical constants is the notion of their
scope, that is, which terms in a sentence they affect. Buridan usually
gives his scope-rules syntactically, and often simply explains the action
of a syncategorematic term in a sentence by describing its effect on
terms which come before or after the syncategorematic term. Such
relations will correspond to order-relations in Mental, as we have
seen.IX
There are two forms of the copula, each of which corresponds to a
complexive concept in Mental: 'is' and 'is not' (TS 2.3.12). Buridan flirts
with the suggestion that the Mental copula is tenseless (TS 3.4.8), but he
15
does not explore it in any detail. The copula produces sentences from
terms or expressions, according to very complicated rules about what
can act as subject or predicate; a partial list of such rules is given in TS
2.6.1.
A negative particle (such as non) may act in two ways (TS 2.2.7). It
may be a sentential-functor taking sentences to sentences; Buridan calls
such a negation "negating," anu it acts upon the copula. Or it may be a
term-functor taking terms to terms; Buridan calls such a negation "infin-
itizing," and it acts upon the individual term, producing what is called an
"infinite" term (TS 3.7.35). Buridan's logic will therefore distinguish
sentences not only by their quantity but also by the character of the
negation occurring within them, their quality.
Signs of quantity, that is, quantifiers, are more complex. Buridan
argues that such signs of quantity, no matter where they occur in the
sentence, not only affect the term immediately following them but also
act as "conditions of the whole sentence" (TS 2.2.15), a conclusion
requiring careful argument (TS 2.2.8-14). Buridan characterizes such
signs of quantity as either affirmative or negative, depending on whether
they involve a negation, and as either distributive (that is, "universal") or
as particular (that is, "existential"). There are several kinds of quantifi-
ers, too; there is at least one for each category (TS 3.7.4), and each has a
corresponding identificatory relative-term (TS 4.2.3, TC 3.7.19-27).
For example, a universal affirmative sign in the category of Quality is
"however," and its corresponding identificatory relative-term is "such,"
as in the sentence "However Socrates is, such is Plato," i.e. every quality
which Socrates has Plato also has (Rule RT-6 in TS 4.8.1, or TS 3.7.27).
Moreover some signs distribute parts of an integal whole; others a uni-
versal whole.
Conjunction and disjunction are equally term-forming functors ap-
plied to terms and sentence-forming functors applied to sentences (TS
2.3.13), obeying the usual logical rules, except for certain uses such as
forming 'conjunctive terms; e.g. the term 'Peter and Paul' in "Peter and
Paul lifted a table" (TS 3.2.3). Conjunctive terms are discussed in TS
2.6.64-67 and 2.6.77. Note that the conjunction of terms, their 'collec-
tive' sense, is distinct from term-combination as in e.g. adjective-noun
expressions.
Terms such as 'if and illative particles such as 'therefore' and 'hence'
are sentential functors, producing consequences, just as the copula pro-
duces categorical sentences (TS 2.3.13, TC 1.3.2).
4. THE PROPERTIES OF TERMS 17
In addition to such typical syncategoremata, there are also exceptives

(terms such as 'but' or 'except'); delimitives Conly,' 'at most: 'at least');
and many others, causing Buridan to exclaim in the first treatise of the
Summulae de dialectica that syncategoremata "are the source of vir-
tually all the confusions which plague logic." Such syncategoremata are
often mixed; in TS 2.4.4 Buridan suggests analyzing 'only' as it appears
in "Only a man is running" as "A man is running and nothing other than
a man is running," so that the concept corresponding to 'only' includes
an affirmative copula and a negative distribution.
Buridan uses the analysis of syncategoremata to say what pertains to
the logical form of a given sentence (TS 1.7.3): the copula, negations,
signs of quantity, the relative-terms, the number of constituent elements,
connectives, and, most of all, the order in which these occur. Examples
of each are given in TC 1.7.4.
Some terms are neither purely syncategorematic nor purely categore-
matic but are "mixed" or "mediate," such as 'somewhere: which involves
the purely syncategorematic term 'some' and implies the categorematic
term 'place,' for it restricts the quantification to places (TS 2.3.1); such
terms are present only in Spoken or Written, being analyzed into their
components in Mental.
4.2. ABSOLUTE AND APPELLATIVE TERMS
Purely categorematic terms have ultimate signification and do not imply

any syncategoremata (TS 2.3.1). Some categorematic terms have simple
concepts corresponding to them, and others complex concepts (TS
2.4.5). The grammatical form of an inscription or utterance is not in
general a good guide to the complexity or simplicity of the correspond-
ing concept; since words signify by convention we can treat the word 'A'
as equivalent to the sentence "Some man is running" (TS 2.4.5), and
then 'A' immediately signifies a complex concept. How can we tell the
difference?
The key distinction here is between simple (or incomplex) concepts
and complex concepts. Buridan suggests that we can distinguish terms
which correspond to a simple concept, which he calls absolute terms,
from those which do not, by the theory of definition.
Definition, as it occurs in Spoken and Written, is the analogue of
complexity in Mental; 'vixen' is definable as 'female fox,' and if a person
possesses the concepts 'female' and 'fox' then he can form the complex
concept 'female fox' to which the inscription or utterance 'vixen' is sub-

ordinated. Thus the composition of concepts in Mental is reflected by
the process of definition.
Again, if some Mental terms are literally composed of others then we
impose a hierarchy on Mental terms: the primitive terms are the inc om-
plex concepts in Mental, which we call absolute; others are produced
through logical composition with the syncategoremata. Buridan argues
that there must be such simple concepts: 19 "if anyone were to say that
complex concepts exist, then they are composed of simples, for there
can be no regress to infinity in the resolution of concepts." In QSP I q.4
Buridan merely argues for the existence of such simple or incomplex
concepts; in TS 2.4.5 he explicitly says that 'man,' 'whiteness,' and
'white' correspond to such simple concepts. 20 Equally, purely syncate-
gorematic terms correspond to simple, though complexive, concepts (TS
2.4.3).
There were two competing requirements on definitions in mediaeval
philosophy: (i) the definiens was to be synonymous with the definien-
dum; (ii) the definiens was to express the real nature or essence of the
definiendum. Definitions satisfying (i) were called nominal, expressing
the quid nominis, because they did not specify the nature of the defi-
niens and only gave information about how the term is applied (and
hence are about the "name"); those satisfying (ii) were called real or
quidditative. 21 With this technical machinery in place, we can begin
sorting out absolute and non-absolute terms. Let us carefully set out
Buridan's exact claims:
(1) A term corresponds to a complex concept if and only if the
term has a nominal definition.
In TS 2.4.1 Buridan says that terms which correspond to complex con-
cepts have nominal definitions; in QM IV q.14 fol.23va and QSP I q.4
fol.5rb he says that terms with nominal definitions correspond to com-
plex concepts. Together these yield the equivalence stated in (1), which
is explicitly endorsed in Rule Sup-12 (TS 2.6.1.). From (1) we may easily
derive the next thesis (TS 2.4.1):
(2) A term corresponds to an incomplex concept if and only if
the term has no nominal definition.
The motivation for (1)-(2) is obvious; if a term is subordinated to a
complex concept, then by definition it is synonymous with tht. ~xpres-
sion stating how the relelvant concepts are combined. This is why
Buridan suggests that (i) indefinable substantial terms correspond to
simple concepts (OM IV q.14 fo1.23va and OSP I q.4 fo1.5vb); (ii) pure-
ly syncategorematic terms correspond to simple complexive concepts
(TS 2.4.3). Thus we may view all non-absolute terms as mere abbrevi-
ations for their nominal definitions. Mental, for obvious reasons, need
only contain absolute terms and purely syncategorematic terms; com-
plex concepts may be logically constructed, by complexive syncategore-
mata, from simple concepts.
The thesis complementary to (1)-(2) would be that a term corre-
sponds to a simple concept if and only if the term has a real definition.
But here we must introduce another distinction among categorematic
terms: some terms are appellative, and others are not. 22 Appellative
terms are more complicated than absolute terms. We shall investigate
their nature carefully.
Buridan characterizes some appellative terms by the Remainder
Principle: 23
If a term signifies something which it does not stand for,

then the term is appellative.
By 'stand for' Buridan means what a term refers to or supposits for (dis-
cussed in Section 6). Appellative terms falling under the Remainder
Principle have no real definition (OM VII q.5 fo1.44va); presumably
they have a nominal definition and so by (1) correspond to complex
concepts. Buridan lists several examples of appellative terms: (i) every
term in an oblique case;2.J (ii) nondenoting terms, which may be impos-
sibilia such as 'round square' or Jigmenta such as 'centaur' (TS 1.4.7);
(iii) concrete terms in categories other than Substance;25 (iv) transcen-
dental terms convertible with 'being; such as 'thing; 'one; and the like
(OM IV q.5 fo1.15vb); (v) the term 'potency' (OM IX q.6 fo1.59ra); (vi)
most combinations of terms, so that complex subjects and predicates are
connotative. What is more, terms in oblique cases combined with a sub-
stantive connote the relation between the subject and what the term
would stand for in the nominative case. In this case, or when an attribu-
tive adjective is combined with a substantive, the term appellates the
adjacence of the associated property with the subject.
What, exactly, is appellation? Buridan explains it thus (TS 1.4.1,
5.1.1,5.2.5; Sophismata IV Remark 2):
A term appellates that which it connotes as in some way ad-

jacent or non-adjacent to that for which it stands or is apt to
stand.
'Connotation' is a semantic relation, a kind of indirect or oblique. form

of significationY' The concrete accidental term 'white' stands for white
things, connotes whiteness, and appellates whiteness insofar as it inheres
in those things (TS 104.8). Equally the term 'wealthy' stands for a man,
connotes his riches, and appellates those riches as adjacent to him as a
possessor. 'Adjacence' is Buridan's general term for the metaphysical re-
lation between two items, e.g. inherence; non-adjacence its opposite.
Thus 'blind' appellates vision privatively, as non-adjacent to its subject.
In most cases, a term connotes something other than that for which it
stands, and so is appellative by the Remainder Principle. But there are
exceptions. In OM VII qA fol.44ra Buridan says that the appellative
term 'creative' connotes a power which is not really distinct from the
possessor of the power; it has no 'external' connotation. Buridan men-
tions 'extraneous' and 'distinct (alienae), connotation in OM IV q.l
foLl3ra, and in TS 1.4.6 Buridan says that 'rational' in the combination
'rational animal' connotes no accident, for there is no appellation of a
distinct (alienae) disposition: here it is a constitutive differentia. 27 These
issues immediately embroil us in metaphysics, for Buridan holds that
such modes of adjacence are all Aristotle's categories really amount to.
That is, the primary metaphysical types of adjacence simply are the
Aristotelian categories, which are therefore not classifications of (dif-
ferent kinds of) beings. Corresponding to these types of adjacence will
be various modes of predication, each involving a different inherence-
relation. Here is how Buridan puts the matter in OM IV q.6 foLl 7va:
There are ten categories or generalis sima because con-
crete accidental terms are connotative ... the categories
should be distinguished by the distinct modes of predication
(modos praedicandi) <something> of primary substances - or,
<that is>, of singular terms contained under the genus quid or
a liquid ...
The same point is stated more forcefully in Sophismata IV Remark 3:
Thirdly, it should be remarked that the different modes of
predication, such as in quale, in quantum, in quando, in ubi,
how one thing is related to another, and so forth, are taken

from (proveniunt) the different modes of adjacence of the
things appellated to the things for which the term supposit.
The different categories are taken (sumuntur) from these dif-
ferent modes of predication.
Buridan discusses such modes of adjacence in OM V q.8; he states the
key problem, Bradley's regress, in fol.31 rb-va, and in fol.31 vb asserts
that there is a "mode of relation" which he calls an 'inseparable disposi-
tion; which is the inherence of an accident in a subject. These disposi-
tions are inseparable; to destroy them is to destroy the inherence, and
conversely; and "they are accidents which are inseparably related to
their subjects in this manner."2X These 'dispositions' are qualities of
qualities which are inseparable, but they are not a new kind of entity;
they are unusual entities of an old kind, namely qualities.
The semantic counterpart of these ontological issues is now clear. A
term like 'white' (i) stands for (supposits for) an individual person; (ii)
connotes the quality whiteness; and (iii) "appellates the quality insofar as
it is adjacent to what it stands for," that is, appel/ates the special disposi-
tion of inherence (TS 1.4.8), which itself is a quality (indeed a quality of
a quality), inseparable from the white thing itself without destroying the
white thing. Of course, "to signify the added disposition is not to signify
that the disposition is added" (OM V q.7 foI.31ra). A term is therefore
appellative if it is apt to satisfy (i)-(iii).29 Obviously, we may treat the
semantic relation of 'appellation' as a kind of naming, denoting the
inseparable disposition which is the appropriate form of adjacence. For
'wealthy' we may treat the mode of adjacence as various mental qualities
of the people involved which constitute the recognition of ownership,
for example.
The Mental correlate of an appellative term is a primitive complexive
concept: roughly, it is the functor 'thing-having-x' which is applied to the
absolute concepts of abstract qualities, e.g. 'thing-having-whiteness.' The
primitiveness of the functor means that there need be no nominal defini-
tion of the term in question, though of course there may well be. The
functor must be complexive as well, since Buridan allows accidents like
whiteness to exist without inhering in a particular subject,30 and the
inseparable disposition must be involved.
The theory of definition was our initial guide to whether a term was
absolute or appellative. We can now see that the usefulness of the theory
of definition was in that definition provides a generally reliable guide to

whether a term in Mental is complex or not But in certain cases the
theory of definition is not sufficient, and we had to investigate deeper
metaphysical issues, The net result is that appellative terms are those
which have strllcture in Mental, not merely those which definition; such
structure can be introduced by the primitive complexive functor de-
scribed in the preceding paragraph,
4,3. INTENTIONAL VERBS
One of the key uses of the doctrine of appellation is to analyze the

behavior of terms in sentential contexts with intentional verbs. 1l His
remarks equally apply to the participles and nouns derived from them.
Such verbs (and the terms derived from them - henceforth I shall omit
this clause) differ from other verbs in that the "verbal action" each spec-
ifies "goes over" to their object not directly, into the things for which the
terms supposit, but indirectly, by means of "certain mediating concepts
indicated by those terms" (TS 1.6.12-14, TC 3.7.3). In particular, such
verbs cause the terms with which they are construed to appellare suas
rationes, that is, to appellate the concept or ratio by which the terms
were imposed to signify (TS 3.8.25, TS 5.3.1 Rules App-l and App-
2, TC 3.7.5).
For Buridan, we are only mediately in touch with the things the
concepts are about, by means of mediating concepts: perhaps a single
concept, perhaps a Mental sentence. Such concepts give the ratio signif-
icandi to an inscription or an utterance; knowledge, in this sense, is not
direct but requires an instrument (OM XII q.8 foI.70vb). As Buridan
says in TS 3.7.10, the logical analysis of a sentence such as
Socrates knows A
is given by
Socrates knows A according to the ratio or concept by
which the term 'A' is applied, that is, according to the con-
cept-of-A
Buridan is therefore a "descriptionalist": there is no pure immediate
knowing; all knowledge is mediated by some concept(s) specifying a de-
scription under which we are in touch with an object
Intentional verbs behave in an odd way: when a term follows the
verb 32 we have an opaque context: Coriscus may not know the one
approaching is his father and he knows his father; the syllogism is
prevented by the appellation of a ratio, and we can only conclude that
Coriscus does not know his father as the concept 'the-one-approaching'
applies to him, which is perfectly acceptable. Buridan describes this
appellation as similar to material supposition (TS 3.8.30, TC 3.7.6), for
substitutivity is prevented. On the other hand, when the term precedes
the verb, it is said to appellate all its rationes indifferently (Rule App-5
in TS 5.3.1 and the discussion; TC 3.7.7): in this case substitutivity is
preserved and we have a transparent reading: 33 the sentence "The one
approaching is someone Coriscus does not know" is false, for the ratio
'the-father-of-Coriscus' equally applies to the one approaching.
Buridan's analysis permits the inference a parte priori to a parte post
for some ratio, which we shall call the Entailment Principle. The
converse entailment a parte post to a parte priori, generally fails, as the
nature of opacity suggests. But in certain cases the latter inference does
hold, and in particular for the verb 'know' (scire); we shall call this the
Converse-Entailment Principle (Sophismata IV Remark 8, TS 3.8.27).
The objectual version allows us to infer from "Socrates knows A" the
sentence "[There is an] A [which] Socrates knows," and the sentential
version (sophism 13) allows us to infer from "Socrates knows A to be F"
the sentence "[There is an] A [which] Socrates knows to be F." Buridan's
key argument for the Converse-Entailment Principle is that we should
otherwise have to deny that we have knowleldge of items in the world. 34
The Converse-Entailment Principle runs into two difficulties: coun-
terintuitive substitution-instances, and the lack of existential import (i.e.
when no A exists the sentence should be false).
Buridan takes up the first difficulty in Sophismata IV sophism 14:
Socrates, who has been studying astronomy, has been imprisoned and
cannot see the sky. We are permitted in this case to pass from "Socrates
knows that some stars are above the horizon" (by his astronomical
studies) to "[There are some] stars [which] Socrates knows to be above
the horizon." Which stars? Those which are in fact above the horizon,
which in the posited case is the constellation Aries. But surely this
seems false, for Socrates cannot see the sky.
Buridan's reply is to insist on the different reading a parte priori and
a parte post. The constellation Aries is indeed what Socrates knows, but
only under the complex ratio 'some-stars-above-the-horizon,' by the
Entailment Principle. This ratio will of course latch onto some actual
stars, though Socrates does not know which, The Converse-Entailment

Principle allows us to infer "[There are] some stars [which] Socrates
knows to be above the horizon" and, since substitutivity works a parte
priori, we may infer from the fact that the constellation Aries is above
the horizon "The stars of Aries Socrates knows to be above the hori-
zon," But the Entailment Principle only licenses us to pass back again to
"Socrates knows stars (under some ratio) to be above the horizon," and
the ratio in question is 'some-stars-or-other: This, Buridan holds, is not
counterintuitive at all but the natural view of the matter.
But this answer might seem to be a cheat. 3 ) For "it surely trades on
the peculiar characteristic [of scire] in that what you know must be so."
Yet this is exactly what Buridan has been emphasizing all along, and is
the very reason why the Converse-Entailment Principle holds only for
scire.
The second difficulty mt:ntioned above was that it seems I can know
that thunder is a sound in the clouds even in the absence of any thunder,
but by the Converse-Entailment Principle then "Any thunder I know to
be a sound in the clouds" should be true, and in the posited case the
subject-term is empty; affirmative sentences with empty subject-terms
are automatically false (TC 2.3.3, QM IV q.14 f01.23va).36
Buridan's answer involves his theories of ampliation and natural sup-
position, which we shall discuss below in Section 6.5.
5. SENTENCES
5.1. SENTENCES AS ASSERTIONS
Sentences express assertions: an inscription or utterance is a sentence

only through subordination to a Mental sentence, which is an act of
thinking. Inscriptions which are not being read, utterances which are
not heard (as a tape recorder playing in the wilderness), are sentences in
only a derivative fashion. To construe sentences as assertions, as
Buridan does, has three important consequences. First, only a Mental
sentence is true or false; inscriptions and utterances have truth-value
only derivatively, as we call a furry winter coat 'warm' (TS 1.3.3-11).
Strictly, only the act of thinking which is the Mental sentence is the
act of assertion. Second, no part of an assertion may itself be an
assertion, as Frege also maintained; hence no part of a sentence is
itself a sentence (Rule Sup-6 in TS 2.6.1, Hughes [1982] 13.4.1.1.1).
However, a sentence may have constituent parts which are equiform to
or sup posit for sentences. 37 Third, the reason for the existential
import of universal affirmatives is clear: a sentence like "All swans
are white" is not to be understood as the disguised conditional
(x) (Sx ...... Wx), from which (3x) (Sx & Wx) does not follow, but rather
as something like "Consider the swans: each is white" or "About the
swans: they are all white," where 'consider the swans' or 'about the
swans' captures the assertive force of a sentence. The existential claim
"Some swans are white" is then "Consider the swans: some are white"
which in fact does seem to follow from the universal sentence. 38 This is
why Buridan takes quantification to be a condition of the whole sent-
ence, as discussed in Section 4.1.
Buridan says that every sentence is either categorical or hypothetical
(TC 1.3.2), which is an exclusive and exhaustive division. 39 These should
be regarded as sentence-frames;4o the expressions which may be
substituted in each frame may be logically complex. The categorical
sentence-frame functionally defines its constituents as subject and
predicate (TS 1.3.8); its general form is therefore:
(subject)-[copula]-(predicate)
25
There seem to be three types of hypothetical sentences: conjunctive

sentences, disjunctive sentences, and consequences. The last func-
tionally defines its constituents as antecedent and consequent; its
general form is therefore:
(antecedent)-[illation]-(consequent)
where the illation is usually signalled by 'therefore' or its equivalent, or
by 'if preposed to the antecedent or its equivalent; Buridan favors the
former method. Disjunctive or conjunctive sentences are sentences
which are disjunctions or conjunctions of parts equiform to sentences;
Buridan does not investigate their general form; neither shall we, since
their behavior is familiar.
5.::!. CATEGORICAL SENTENCES
The general form of the categorical sentence discussed above, subject-

copula-predicate, is the actual form of the Mental sentence; in Spoken
or Written not all of these components will be apparent (TC 1.8.72).
For example, existential statements have misleading surface grammar;
the Spoken or Written expression "S is" is represented in Mental as "S is
a being" or "S are beings" (TS 1.3.8 and TC 1.8.84). Equally, subjects
paired with intransitive verbs are resolved into the copula with a
participle, so "S runs" becomes "S is running"; and the like. Buridan
tentatively suggests that the copula in Mental is timeless (TS 3.4.8),
but does not explore the implications. We have remarked above the
complexive nature of the copula. Strictly speaking there are two forms
of the copula: 'is' and 'is not,' which correspond to distinct functions in
Mental (TS 2.3.12).
It is harder to describe the nature of the subject and the predicate,
which Buridan discusses in Chapters II-2 and U-6 of TS. A complex
series of rules specify some restrictions on which expressions can serve
as subject or predicate. In general, any noun phrase can be subject or
predicate; this includes (i) any substantive nominative term, whether
discrete or common, and any adjective taken substantively (Rules Sup-l
and Sup-5); (ii) all referring expressions (Rule Sup-2); (iii) infinitive
verbs (Rules Sup-4 and Sup-7); (iv) grammatical combinations of nouns
and adjectives, or, more generally, nouns and oblique terms (Rule
Sup-8); (v) certain combinations of categorematic and syncategorematic
terms (Rules Sup-9 and Sup-l 0). Conversely, taken per se the following
cannot serve as subject or predicate: (i) pure syncategoremata (Rule
5. SENTENCES 27
Sup-2); (ii) oblique terms not combined with a substantive (Rule Sup-3);
(iii) finite verbs (Rules Sup-4 and Sup-7); (iv) sentences (Rule Sup-6).
Categorical sentences may be distinguished with respect to quantity,
quality, or mode. Regarding quantity, first, signs of quantity are properly
not parts of the subject or predicate, but are "conditions of the whole
sentence" (TS 2.2.15), the same way a negating negation (though not an
infinitizing negation) is a condition of the copula. Second, the quantity
of a sentence is determined by the syncategorematic terms present in
the sentence. Distributive and particular signs of quantity are the usual
syncategorematic terms which specify the quantity of a sentence, but
other syncategoremata may also affect the quantity - negations, excep-
tives, and the like (see Section 4.1). Finally, the quantity of a sentence is,
in general, given by the semantic generality of the subject of the
sentence, and not by the items actually denoted; "All unicorns have
horns" is distributive, and so has more quantity than "Some men are
sexists," despite the fact that 'unicorn' is an empty term and 'man' is not.
Modern logicians recognize differences of quantity and of mode:
sentences are universal or particular, assertoric or modal. Differences of
quality - whether a sentence is affirmative or negative - are not usually
taken into account. Yet Buridan takes two forms of the copula, 'is'
and 'is not; as primitive; hence there are two fundamental types of
sentences, affirmative and negative. Is this defensible? The answer turns
on the significance attached to the distinction between predicate nega-
tion and sentence negation, which cannot be clearly drawn in modern
logic: there is no difference between belonging to the extension of the
complement of a predicate and not belonging to the extension of the
predicate. Buridan, however, insists on the difference: the truth of
"Socrates is non-blue" suggests that Socrates has a property, being non-
blue, or at least presupposes that he exists, whereas the denial "It is not
the case that Socrates is blue" is not to claim that he has any property,
or even to presuppose that he exists:.\! Thus for Buridan the truth-
conditions for sentences differing in quality may be quite different.
For example, since negatives are denials which do not presuppose
the existence of their subjects, negatives with empty subject-terms are
automatically true: of course unicorns fail to have a given property;
they fail to have every property, since they do not exist. Conversely,
affirmative sentences with empty subject-terms must be false, even if the
predicate is infinite.
Categorical sentences may also be distinguished with respect to
28 BURlDAN'S PHILOSOPHY OF LOGIC
mode: they are either assertoric or modal (TC 2.1.3). Buridan distin-
guishes them in the following way: although every sentence is either
possible, necessary, or contingent, only those sentences which explicitly
include a "mode" are counted as modal; the terms 'possible; 'necessary;
'contingent; 'true,' 'false,' 'known,' and so forth are modes. Buridan thus
treats 'modal logic' as much wider than the logic of possibility and
necessity. Such sentences also include a dictum, which Buridan defines
as "that whole found in the sentence aside from the mode and the
copula and the negations and the quantifiers, or other determinations of
the mode or the copula" (TC 2.3.3).42 Examples are "That a man runs is
possible" or "A man can possibly run." The former modal sentence is
composite, the latter divided: a modal sentence is composite when the
mode is put as the subject and the dictum is predicated, or conversely
(TC 2.2.2); a modal sentence is divided when part of the dictum is put
as the subject and the other part as the predicate, and the mode is taken
as a determination of the copula (TC 2.2.5).
Divided modal sentences ampliate their subjects and can be treated
easily as a case of sentences with a special ampliative copula (see
Section 6.8). Composite modal sentences, on the other hand, are a
peculiar type of assertoric categorical: we may understand
It is possible that p
as equivalent to
A possible <sentence> is that-p
where the analyzed sentence has an ordinary copula and the predicate is
'that-p' (TC 2.7.1-5). Such sentences are indefinite and may thus be
quantified:
Some possible <sentence> is that-p
Every possible <sentence> is that-p
To understand this we need to know what it means to call something a
'possible' or a 'possible <sentence>.' Buridan implicitly endorses the
following analysis:
"Some possible <sentence> is that-p" is true if and only if
for a sentence p* equiform to p, how p* (if formed) signifies
to be can be the case.
There are two points to note. First, the clause 'a sentence p* equiform to
5. SENTENCES 29
p' must be included because the predicate of the original sentence is

'that-p,' which is not itself a sentence.43 Indeed, no part of a sentence is a
sentence. Second, the clause 'if formed' is not within the scope of 'can be
the case': that is precisely Buridan's distinction between the possibly-
true and the possible. 44 The analysis given above is true of a possible
sentence; roughly, a sentence is 'possible' when it describes a possible
situation, and 'possibly-true' when it is true of and in a possible situa-
tion. Thus we may also give the analysis:
"Some possibly-true <sentence> is that-p" is true if and
only if for a sentence p* equiform to p, how p* signifies to be
can be the case, if formed.
A sentence is only possible-true if it can be a part of the possible
situation it describes. The standard example Buridan uses to distinguish
the possible and the possibly-true is the sentence "No sentence is
negative": it describes a possible situation, for it is surely possible that
all utterances have been affirmative - but the sentence itself can never
be part of such a situation, since it is not affirmative in form.
5.3. HYPOTHETICAL SENTENCES
In additional to categorical sentences there are hypothetical sentences,

that is, sentences whose main connective is not the copula. There are
three types of hypothetical sentences: conjunctive sentences, disjunctive
sentences, and consequences (QM VI q. 10 fol. 40rb). The truth-
conditions for the first two are straightforward and given in terms of
the truth-conditions of sentences equiform to their constituent parts.
Consequences, however, are more complex.
In particular, we may specify truth-conditions for each kind of hypo-
thetical sentences. Hypothetical disjunctive sentences are true just in
case one of the sentences equiform to its disjuncts is true; hypothetical
conjunctive sentences are true just in case each of the sentences
equiform to its conjuncts is true:~5 Standard laws of the propositional
calculus are appropriately represented in Buridan, e.g. DeMorgan's
Laws (TS 3.7.37).
The truth-conditions for consequences are more complicated, for
they are not simply truth-functional: their truth or falsity depends on the
semantic analysis of the sentences equiform to the protasis and apodosis
(discussed in Section 7.2).
The truth of hypothetical sentences, then, depends on our under-

standing the truth of categorical sentences: it is not a simple recursive
dependence, in which the truth-value of the complex is truth-func-
tionally determined by recursive rules, The acceptability of a con-
sequence depends on the modal connection of the truth-values of the
antecedent and consequent: it must be impossible for things to be as the
antecedent says with things not being as the consequent says. Therefore,
despite the fact that not all hypothetical sentences can have their
truth-value specified recursively and truth-functionally, the analysis of
the truth of categorical sentences is the key ingredient in determining
the truth or acceptability of a hypothetical sentence. After considering
some basic principles of sentential logic, we shall take up the problem of
determining the truth-value of categorical sentences (Section 5.5).
Obviously, hypothetical sentences are similar to the contemporary
logician's 'molecular formulae: But there are important differences: the
parts of hypothetical sentences are not sentences, though parts of
well-formed formulae may be formulae; equally, the truth-value of a
hypothetical sentence is not a simple recursive matter - indeed, con-
sequences are properly called acceptable or not, not true or false. These
differences stem from a more basic difference, namely Buridan's con-
strual of sentences as assertions, discussed in Section 5.1. Conjunctive
and disjunctive hypothetical sentences may be straightforwardly treated
as assertions: a conjunctive assertion is closely related to a conjunction
of assertions; a disjunctive assertion is closely related to a disjunction
of assertions. (See TS 2.6.59, in which Buridan describes this as the
"natural condition" of disjunction.) In these cases, although the main
connective is not the copula, it is closely related to it so no problems
need arise. 46 But, as is notorious among modern logicians, the notion of
'conditional assertion' is not so amenable to direct treatment. Here it
seems we may be forced to distinguish between the predicative force of
copUlation and assertive force. And indeed, that is what Buridan does,
though he takes a different route than Frege in classifying consequences
as acceptable rather than as possessing truth-value, This is a deep point
about Buridan's system which we shall cover thoroughly in Section 7.
5.4. PRINCIPLES OF SENTENTIAL LOGIC
The core of sentential logic has several elements: (i) opposition in all its
forms, such as contradiction, contrareity, and subcontrareity; (ii) rules of
5. SENTENCES 31
inference or entailment, including equipollence and conversion; (iii)

principles of well-formedness, both as criteria for the acceptability
of sentences and rules for compounding sentences. The theory of
consequences is crucial for (ii), and will be discussed in detail in
Section 7. We have generally canvassed (iii) in the preceding sections.
The present section will explore (i).
In TC 1.S.1 Buridan sets forth a general principle:
For any contradiction, one of the contradictories is true
and the other false, and it is impossible that both are true
together or false together; again, every sentence is true or
false, and it is impossible for the same sentence to be true
and false at the same time.
Several principles governing opposition are endorsed here: bivalence,
i.e. "every sentence is true or false"; the Law of Non-Contradiction, i.e.
"it is impossible for the same sentence to be true and false at the same
time"; and what we may call the 'Law of Contradictories; i.e. contradic-
tories must have opposite truth-value. (See also TC 1.1.7 for bivalence
and the Law of Non-Contradiction.) Buridan never discusses bivalence
at any length; like most mediaeval logicians, he simply assumes its
truth:H On the other hand, the Law of Non-Contradiction and the Law
of Contradictories are carefully investigated in OM IV qq. 11-15.
Buridan begins his discussion in q. 11 by asking whether contradic-
tion is the greatest form of opposition, and argues first that, strictly
speaking, only sentences are contradictories, such as "Socrates is
running" and "Socrates is not running." However, there are also
contraries; sometimes terms themselves, or their referents, are called
contraries; most properly, though, only sentences can be contraries.
There are two kinds of contraries: (i) sentences which are contraries due
to their terms, such as "Socrates is white" and "Socrates is black"; (ii)
sentences which are contraries due to their logical form, such as "Every
man is running" and "No man is running" (fol. 21ra). Technically, we
can distinguish two additional principles: the 'Law of Contraries,' i.e.
such a pair of sentences can both be false but cannot both be true
(discussed by Buridan in TS 2.6.39), and the 'Law of Subcontraries,' i.e.
such a pair of sentences can both be true but cannot both be false.
In q. 13 Buridan asks "whether the sentence 'it is impossible that the
same be present (inesse) and not present to the same at once, with
respect to the same, and so forth for other conditions' is the first
complex principle" (fol. 22ra), that is, the Law of Non-Contradiction.

His answer is complex. Sentences may be classified according to quality,
quantity, mode, time, and the like; Buridan argues that there is no
general way to state truth-conditions applicable to all these different
types of sentence. 4X For example, it is possible that the same be possibly
present and possibly not present to the same at once. In particular, there
will be no general definition of 'contradictory,' 'contrary,' or 'subcon-
trary,' but particular definitions for each type of sentence. Hence the
formulations given above are only schematic. there is a form of the Law
of Non-Contradiction for each type of sentence; "all of the aforemen-
tioned principles are indemonstrable, although some are more evident
and simple <than others>" (fol. 22va).49 Equally, there will be particular
forms of the Laws of Contradictories, Contraries, and Subcontraries.
The traditional Square of Opposition (Fig. 1) expresses the logical
relations among a limited class of sentences: they are present-time
assertoric categorical sentences, the terms of which are logically simple,
where the predicate-terms are common nouns and have an implicit
particular quantification,'O indefinite sentences are treated as implicitly
quantified (TS 2.2.12), and singular sentences added. Buridan's rules,
however, are stated generally, so the Square of Opposition only
represents a limited amount of theory.
AIISisP contraries
------------------------- No SisP
(A-form) (E-form)
Some S is P Some S is not P

(I-form) subcontraries (O-form)
Fig. I. The traditional square of opposition
5.5. TRUTH AND SENTENTIAL SIGNIFICATION
What are the truth-conditions for categorical sentences? Buridan inves-

tigates this question in Sophismata I-II, TC 1.1-2, and OM VI q. 8. He
5. SENTENCES 33
begins by citing the standard definition of truth: "A (categorical) sen-

tence is true if howsoever it signifies so it is (qualitercumque significat
ita est)." To understand the definition we need the notion of the
signification of a sentence. Buridan considers and rejects four possible
answers.
First, let us suppose that the definition refers to the immediate signifi-
cation of all of its terms. But then we have the unfortunate conclusion
that every Spoken sentence is true, because its immediate significate
is the Mental sentence, which does exist (Sophismata I sophism 1);
indeed, an inscription or utterance is only a sentence in virtue of the
corresponding Mental sentence. The definition cannot mean the imme-
diate signification of the terms.
Second,51 suppose the definition concerns the ultimate signification,
of the terms in the sentence. Since syncategoremata by definition
lack ultimate signification, they are excluded. Yet this will have two
unfortunate results. First, a sentence and its contradictory may differ
only in their syncategoremata, as do "Socrates is a sexist" and "Socrates
is not a sexist"; they ultimately signify the same. Hence the proposed de-
finition of truth will not assign opposite truth-values to a sentence and
its contradictory, which is absurd. Second, by the Additive Principle, the
ultimate signification of a sentence such as "A man is an ass" is the sum
of the ultimate significations of its categorematic parts, namely men and
asses. But there surely are men, and there surely are asses; does this
mean that the sentence "A man is an ass" is true?
Third, several philosophers thought the significate of a sentence was
an abstract entity called a complexe sign ifica bile, rather similar to the
modern notion of a proposition. Buridan rejects this view for a variety
of reasons: (i) even proponents of the complexe significabile, such as
Gregory of Rimini and Adam Wodeham, admitted that it was not to be
found in any of the Aristotelian categories; Buridan draws from this the
conclusion that it is nothing, and it is impossible to see how a sentence is
true or false in virtue of what is nothing at all. Yet (ii) suppose it were
not nothing, but something; then it either exists or fails to exist, and
presumably in the former case the sentence which signifies it is true and
in the latter case the sentence is false. But then one is maintaining that a
false sentence literally has no signification, which seems absurd: such
sentences are false, not meaningless. Finally, (iii) suppose an existent
complexe significabile is the significate of true or of false sentences in
some fashion. But suppose things changed in the world. The abstract
entity which is the complexe significabile is unchanged, not being part

of the world, But then a sentence is true or false not in virtue of
the world at all, which seems obviously ridiculous. Hence the complexe
significabile is not sufficient to account for truth-value.
Fourth, suppose that the significate of a sentence is the true or the
false, in the Fregean manner. But then the proposed definition is both
circular and uninformative.
The problem will all of these suggestions, Buridan thinks, is that they
fail to notice that sentences are far more than the terms which compose
them; they are the way in which we talk about the world. Buridan
modifies the proposed definition by adding a single clause:
A categorical sentence is true if and only if howsoever it
signifies so it is, in the thing(s) signified.
To understand what is 'in' the thing(s) signified we must move beyond
the simple signification of terms: we have introduced the notion of refer-
ence, which for Buridan is part of the theory of supposition. Hence we
need to understand the nature of supposition before we can account for
the truth or falsity of sentences.
6. THE THEORY OF SUPPOSITION
6.1. SUPPOSITION AND THE THEORY OF REFERENCE
Supposition is a semantic relation, holding between term(s) and thing(s).

The relation of signification, however, is also a relation of term(s) and
thing(s). Yet it is one matter to assign certain terms to certain things, so
that a language may be set up in the first place; this is the contribution of
signification. It is quite another matter to actually use that language to
talk about things; this is explained by supposition, which accounts for
the referential use of (significative) terms. Hence there are two major
differences between supposition and signification: first, terms have signi-
fication wherever they are found, inside or outside a sentence, but it is
only in a sentence that we use terms referentially, that is, actually talk
about things and say something about them. Therefore:
(1) A term only has supposition in a sentential context.
Second, we do not always use terms to talk about what those terms
signify; we use them in other ways as well. Therefore:
(2) The kind of supposition a term has depends on its sentential
context.
These principles govern the theory of supposition, and provide a con-
text for the various divisions of supposition.
The theory of supposition should not be assimilated to formal logic,
but to the philosophy of logic; it is the mediaeval theory of reference.
The claim that supposition, or any relation R, is in fact a reference-
relation is justified if the R-relation is sufficiently similar to our uses of
'reference.' There are two main uses in contemporary philosophy of lan-
guage. The first, associated with Davidson, takes the R-relation to figure
in a compositional theory of how sentences acquire their meaning and
truth-value, especially if in this theory the R-values are the referents. As
we shall see in Section 6.8, the theory of supposition does provide an
account of the truth-value of sentences, although not their meaning.
Hence this first way is inconclusive.
35
The second way, associated with Quine, is definitive. On this ap-

proach we take the variables of quantification to be the paradigm case
of the sort of term which has an R-value. Now this approach is usually
rejected out of hand because there are no variables of quantification in
supposition theory: terms are bound, not variables. Yet such rejection is
facile; there may be no overt variables of quantification, but we can look
at the mediaeval treatment of variable-binding operators in the lan-
guage. In particular, consider the case of anaphoric pronouns whose
antecedents are quantified terms, such as 'he' in "Some man has a
daughter and he loves her." This is a case of 'relative supposition'
(discussed in detail in Section 6.4) in which the supposition of 'he' is
determined by the supposition of 'some man.' Equally there are various
indexing devices available; in the example, gender disambiguates the
particular antecendent to each pronoun. Such antecedents can be multi-
plied and indexed at will in supposition theory. Hence supposition has
the key characteristics of a reference-relation, and may be treated as
such.
Supposition theory is a theory of reference. It is a unified theory,
which has as its goal to specify what a term is used to talk about in a
given sentence. 52 The various divisions of supposition illustrate the ways
in which a term may supposit for something. Buridan's division of sup-
position is shown in Figure 2. Of course, we must specify the scope of
the theory; this is the point of the first division, the distinction between
proper and improper supposition. A term has improper supposition
when it is used rhetorically in a sentence and not literally. The theory of
supposition is only developed for the literal uses of terms, as indeed are
modern theories of reference, and both are equally far from explaining
the meaning of 'literal' or understanding non-literal uses of language. 53
Personal and material supposition illustrate the kind of thing a term
may refer to in a sentence, namely its ultimate significate or itself,
whether as inscription, utterance, or concept. The various divisions of
personal supposition specify how many of its ultimate significates a term
may stand for: exactly one (discrete); at least one (determinate); several
(non-distributive confused): all present instances (distributive); all past,
present, and future instances (natural). The status of a term, as defined
in TS 6.1.1, requires this interpretation. The theory of supposition,
of course, does not say exactly which things a term stands for; like any
semantic theory, that is left to the facts of the matter to decide - terms
may fail to refer, the reference will change with changes in the world,
6. THE THEORY OF SUPPOSITION 37
. - / SUPPOSITION ~
Proper Improper
/ ~
Material / ' Personal ~
Discrete / ' Common ~
/ Absolute"" Relative
/ Accidental~ Natural
Determinate / Confused ~
Distributive Non-Distributive
Fig. 2. Buridan's Division of Supposition.
and so on. But it does specify the semantic function a term may have
in different sentential contexts. This point is important: supposition
theory will explain the semantic role of terms in a sentential context -
evaluating the truth of the sentence is another matter. Hence it is a
mistake to suppose that supposition theory will say exactly which
things a term in fact sup posits for. Rather, supposition theory will
specify what things a term semantically supposits for, and then it is
a separate question whether the supposition is successful. Failing to
appreciate this point can only lead to confusion.
6.2. PERSONAL AND MATERIAL SUPPOSITION

The general motive for distinguishing material and personal supposi-
tion is clear: a sign may be used to refer to things (its ultimate signifi-
cates), or it may be used to refer to other signs. In the former case a
term has personal supposition; in the latter, material supposition. 54 For
example, in the sentence "Every man is running" the term 'man' has
personal supposition; we are asserting something about individual
men, who are the ultimate significates of the term 'man.' But in the
sentences "Man has three letters," "Man has one syllable," "Man is a
concept," the term 'man' is used to talk about not individual men, but
rather an inscription, an utterance, and a concept respectively, each of
which is itself a sign. Note that the subject-terms of all these sentences
are the same. 55
Buridan defines personal supposition as follows (TS 3.2.1):

Supposition is called 'personal' when a term supposits
for its ultimate significate(s).
Thus the term 'man' has personal supposition in the sentence "A man
is running". The divisions of personal supposition, discussed in Sec-
tions 6.3-7, describes which of its ultimate significates a term may sup-
posit for in particular. This same question is also at the heart of the
theory of ampliation and restriction, discussed in Section 6.8, which
analyzes how the referential domain of a term in personal supposition
may be either widened or narrowed respectively. We can generalize
Buridan's account in TS 1.2.9 to arrive at the following characteriza-
tion of personal supposition:
A term t has personal supposition in a sentence if and
only if either (i) some sentence of the form "This is f' is
true, or (ii) some clause of the form ' ... and that is f can
be added to an existential sentence or a sentence presup-
posing an existential sentence to produce a true sentence.
The demonstrative or relative pronoun and the copula of (i)
and (ii) should be taken in the appropriate number, tense,
and mode.
The motivation for adding (i) in TS 1.2.9 is to circumvent worries
about whether we can point to God, who is not open to direct inspec-
tion (in this life), which might seem to be required for ostension. In
fact (ii) is more general: it will serve to sidestep worries about "point-
ing to" past, future, or possible existents as well, which are also not
open to direct inspection. Thus in the sentence "Some man will lecture
tomorrow" the term 'man' has personal supposition, since some sent-
ence of the form "This is a man" is true.
A term is in material supposition, on the other hand, if it does not

sup positfor what it ultimately signifies (TS 3.2.1):
But supposition is called 'material' when an utterance
supposits for itself or something similar to itself or for its
immediate significate, which is the concept according to
which it is imposed to signify, as the term 'man' in the sent-
ence 'Man is a species:
Logicians preceding Buridan posited another kind of SupposItion,

simple supposition, in which a term was taken to stand for a universal
or common nature. William of Ockham had retained simple supposi-
tion as a category, but his nominalist scruples forced him to say that in
simple supposition a term supposits for the concept it immediately
signifies, which, in the case of common terms, simply is the universal.
It was left to Buridan to see this extra division of supposition as a relic
of realism (TS 3.2.4-5): since a concept is just as much a sign as an
utterance or an inscription, albeit naturally and not conventionally
significative, Buridan drew the conclusion that supposition for a con-
cept as the immediate significate was just another form of material
supposition (TS 3.2.6).
Buridan argues that material supposition is proper only to Spoken
or Written, and that there is no material supposition in Mental: "no
term in a Mental sentence supposits materially, but always personally"
(Summulae dialectica VlI.iii.4). Hence Buridan is not even faintly
interested in divisions of material supposition: such a series of divi-
sions would only be of limited interest to the logician, whose concern
is with Mental and not with Spoken or Written. Mental therefore has a
complete apparatus to accomplish all of the functions of material sup-
position. For example, the sentence "Man is a species" is properly repre-
sented in Mental in the following way (Summulae de dialectica VII.iii.4):
I say that the Mental sentence corresponding to the
sentence "Man is a species" (taken as it is true) is not a
sentence in which the specific concept of man is put as the
subject, but rather is a sentence in which the concept by
which the specific concept of man is conceived is put for
the specific concept of man, from which it is clear that the
aforementioned paralogism is a fallacy in dictione by the
change of supposition.
The "aforementioned paralogism" is the argument "Man is a species,
and Socrates is (a) man; therefore, Socrates is a species." It fails be-
cause there is a change in supposition, perspicuously seen in Mental:
the sentence "Man is a species" is subordinated to the sentence "The
concept by which the specific concept of man is conceived is a spe-
cies," which is true: genera and species are concepts of concepts, not
individual general concepts themselves.
Material supposition acts like quotation in many ways, serving to
distinguish the use of a term from its mention, but the differences
should not be overlooked; material supposition is much wider than
our use-mention distinction, The similarities are obvious: in sentences
such as "Man has three letters" or "Man is a monosyllable" the term
'man' has material supposition; we should render these by using the
quote-functor, giving '''Man' has three letters" or '''Man' is a monosyll-
able," However, there are differences even in such cases, First, applica-
tion of a quote-functor produces a new term, one which names the
term inside the quotes, but for Buridan a term is the same whether in
personal or material supposition, Indeed, Buridan was familiar with
the device of naming expressions, but he never proposes it as an
account of material supposition,56 Equally quotation does not seem to
require a sentential context the way material supposition does, Second,
the quote-functor may be iterated, 57 but material supposition cannot;
there is no mediaeval analogue of the sentence ""Socrates" names
'Socrates' names Socrates, who was Greek.",5~ Material supposition
could be no more than a first-order fragment of quotation theory,
Third, the substitution classes differ; a term can materially supposit for
what is only similar to it, so that accusative-infinitive phrases supposit
for sentences, and changes in case and gender are permitted, In the
composite modal sentence "It is possible for Socrates to be a bishop"
the expression 'for Socrates to be a bishop' materially supposits for the
sentence "Socrates is a bishop" (or a similar sentence),
Material supposition, then, is more inclusive than the distinction of
use and mention, even when we restrict ourselves to the material
supposition of utterances or inscriptions, Allowing material supposi-
tion of concepts merely points up the relevant differences,
When is a term in personal supposition or in material supposition?
Unlike Ockham, Buridan does not try to give precise rules, but rather
trusts to good sense and good logic (TS 3,2,15-22), usually taking
context to decide, Since there is no material supposition in Mental,
Buridan is not at fault for not providing precise rules; the vagaries of
Spoken or Written are met individually,
6.3. DISCRETE AND COMMON SUPPOSITION
We now properly begin the divisions of personal supposition, which

were drawn in an effort to clarify which of its significates a term is
used to refer to. By 'discrete term' Buridan means a singular referring
expression - that is, an expression which is only predicable of a single

item. Hence it is obvious when using a discrete term that one is talking
about the very thing the term refers to: there are no other choices.
Hence personal supposition is divided into discrete and common (TS
3.3.1). As examples of discrete terms Buridan offers us 'Socrates' and
'this man.' Presumably all proper names are discrete, despite the fact
that two people may have the same name; the 'name' is not really
common in this case. Demonstratives combined with common terms
are also discrete. Buridan never says so, but pure demonstratives
should also be discrete terms. Any singular referring expression is, as a
matter of semantics, a discrete term: if the Scholastics had articles,
they would surely have treated definite descriptions as discrete
terms.5'1 However, not every term which refers only to one thing is a
discrete term; in a world with only one man, the common term 'man'
has only one referent, but it is nonetheless common, since it is predic-
able of many. Only those expressions which can only be predicated of
one referent are discrete terms as a matter of semantics, although
other terms may as a matter of fact be used to talk about only one ref-
erent. Criteria for distinguishing the kinds of common supposition a
term can have will depend on the relation of the sentence in which the
common term appears to sentences with disscrete terms. Common
terms, of course, will be general referring expressions.
6.4. ABSOLUTE AND RELATIVE SUPPOSITION
As instances of relative supposition Buridan examines 60 what is now

called anaphoric reference: a 'relative-term' is not a term in the
category of relation, but a reference indicator, something "relative to a
thing said before or recalled" (TS 4.1.3). Anaphoric terms include
what Buridan calls identificatory relative-terms (TS 4.2.1): the personal,
reflexive, and possessive pronouns (TS 4.3-4, TS 4.6-7); relative pro-
nouns (TS 4.5); terms paired with categorial quantifiers, as 'such' is
paired with the universal affirmative sign 'however' of the category of
Quality (TS 4.8); they also include differentiating relative-terms, such
as 'another; 'other; 'different; and like (TS 4.9).
There are two rules governing identificatory relative-terms terms: (i)
the relative-term supposits only for those things for which its antece-
dent is verified (Rule RT-1 in TS 4.3.1); (ii) the supposition of the
relative-term is the same as the supposition of its antecedent (Rule RT-2

in TS 4.4,1).61
The motivation for Rule RT -\ is apparent in "Some men are sexists
and they are disgusting": the relative-term 'they' supposits not simply
for men, nor simply for some men, but for those men who are sexists.
They' are not simply men, for we are not claiming that men are dis-
gusting; nor are they simply 'some men: for in that case the sentence
could be true if there are non-sexist disgusting men, which is certainly
not what we mean. Equally, in the sentence "Some men are not sexists
and they are liberated" the relative-term 'they' supposits for those men
who are not sexists; if there are none, the sentence is false. Therefore
such cases of anaphoric reference act as restricted quantifiers, refer-
ring to those items the antecedent supposits for which the predicate
also supposits for. This explains why Buridan treats issues of pronom-
ial reference together with identificatory relative-terms paired with the
categorial quantifiers. h2 Now Buridan treats all such cases on a par,
and so 'pronouns of laziness' will also have a quantificational aspect: in
"Socrates is disputing and he is ugly" the relative-term 'he' does not
simply refer to Socrates, but to Socrates as disputing - and if Socrates
is not disputing, then 'he is ugly' will be false due to its empty subject-
term (TS 1.2.9). There are two reasons which motivate Buridan here:
(i) recall that no part of a sentence is a sentence, and so "Socrates is
disputing and he is ugly" is a conjunctive sentence making a single
assertion; it is no wonder that the semantics for 'he is ugly' depend on
the semantics for 'Socrates is disputing: What is more, (ii) if Socrates
is not disputing then it is in fact questionable what 'Socrates is dis-
puting' refers to, and hence what 'he is ugly' refers to; it is a
philosophical question whether such a sentence refers to Socrates and
makes a false claim about him or simply fails to refer, which Buridan
points up by asking for the reference of 'he: While the first alternative
is customary today, Buridan opts for the second alternative as more
fruitful philosophically.
Reflexives must be in the same sentence as their antecedent term and
are construed with them (Rule RT -4 in TS 4.6.1): from "Every man
amuses himself" we correctly infer "Socrates amuses Socrates, and Plato
amuses Plato, and so on." Term like 'such' and 'so-much' supposit for
what their antecedents do, where the antecedent is governed by the
appropriate kind of quantifier (Rule RT -6 in TS 4.8.1).
Differentiating relative-terms are also paired with antecedent terms

but serve to distinguish the terms which follow them; the rules
governing them are complex (Rule RT-7 in TS 4.9.1). For example,
"Socrates is other than Plato" should be analyzed as "Socrates is a
being and Plato is a being and Socrates is not Plato," by Rule RT-
7(b). Strictly speaking this is not anaphoric reference, but the logical
analysis of sentences containing certain terms.
6.5. NATURAL AND ACCIDENTAL SUPPOSITION
Natural supposition is a form of ampliation: the present tense of the

copula does not accurately indicate the temporal range of the terms
occurring in the sentence. Just as a term such as 'man' signifies all its
significates, men, for all times, so too in certain sentential contexts it
sup posits for all of its significates, not merely present existents.(,] Such
supposition is paradigmatic ally common, for the term is used to refer
to everything which it may stand for. Buridan defines it as follows (TS
3.4.1 ):
Supposition is called 'natural' when a term indifferently
supposits for everything for which it can supposit, past and
future as well as present; this is the sort of supposition we
use in demonstrative science.
This definition is repeated in QNE IV q. 664 (p. 501):
According to the older logicians, the common supposi-
tion of a term is twofold: natural and accidental ... it is
<natural> when it supposits indifferently for all of its
supposits, whether past or future; demonstrative science
uses this supposition.
We may say, then, that a term has natural supposition if it supposits
for everything (past, present, future) which it signifies. As Buridan
indicates, he is reviving an old usage, for the logicians of the thirteenth
century had discussed natural supposition.
Buridan cites four reasons why we should admit natural supposi-
tion: (i) epistemic verbs require it (TS 3.4.3); (ii) the disjunctive
subject-terms of sentences such as "What was or is or will be A is
running" have it (TS 3.4.4); (iii) natural supposition is produced by

terms such as 'perpetually; 'eternally,' and the like (TS 3.4.5); (iv)
demonstrative science uses it (TS 3.4.6-7 and ONE VI q.6 pp.
501-502). Buridan argues that we can and do construct concepts
indifferent to time (TS 3.4.8-12 and ONE VI q.6 p. 501): our con-
cepts do not have any temporal reference, for it is possible to conceive
a thing "without understanding along with it a determinate time" (ONE
VI q.6 p. 501). Indeed, this is the only way our concepts can signify
things of all times.
Natural supposition leaves us with a problem: cases like (i)-(iii)
quite obviously require omnitemporal reference, but in discussing (iv)
Buridan cites sentences which do not have any clear indication of their
natural supposition, such as "Thunder is a sound in the clouds."
Buridan says that such an utterance or inscription is properly subordi-
nated to the Mental sentence "Any thunder, whenever it was or is or
will be, is or was or will be a sound in the clouds" (TS 3.4.7); in parti-
cular he calls the former sentence an abbreviation (ad breviloquium)
of the latter. 65 The temporal quantifier 'whenever' has 'it was or is or
will be' in its scope, and restricts the supposition of 'thunder' to times
when thunder exists, for each case of which it either was or is or will
be a sound in the clouds (matching the times to the tense of the
copula). Buridan specifically rejects the conditional reading of such
sentences (p. 502).
Hence whether a sentence such as "Thunder is a sound in the
clouds" has natural supposition will depend on our understanding of
the sentence, that is, what the sentence corresponds to in Mental: the
same Spoken or Written sentence (or equiform sentences) may be
ignored, doubted, opined, and demonstratively known; to hold other-
wise would be to make all knowledge strictly a matter of semantics.
Sentences of natural science are thus to be understood as follows:
Socrates knows "Thunder is a sound in the clouds" (a

particular utterance or inscription).
Socrates knows that any thunder, whenever it was or is
or will be, was or is or will be a sound in the clouds.
The latter expresses the Mental sentence to which the utterance or
inscription is subordinated. The Converse-Entailment Principle dis-
cussed in Section 4.3 then licenses:
Thunder (whenever it was or is or will be) Socrates

knows to have been or to be or to be going to be a sound
in the clouds.
Therefore for any instance of thunder Socrates knows it at least under
the ratio 'sound-in-the-clouds' (matching tenses appropriately). This
removes the difficulty about empty subject-terms, for the subject of
this last sentence is nonempty; indeed, it is ampliated to all and only
those times at which there is thunder. 06
Supposition is accidental, on the other hand, when the supposition
of the term is not indifferent to time (TS 3.4.1), that is, refers to items
existing only at some determinate time or other (ONE VI q.6 p.501).
The remaining divisions of supposition apply only to accidental sup-
position, for the restriction to a given time excludes some excludes
some items from the (putative) reference-class of the term.
6.6. DETERMINATE AND CONFUSED SUPPOSITION
A term is said to have determinate sUPPQsition in a sentence when a

sufficient condition for the truth of the sentence is that it be true for
some determinate singular falling under the common term (TS 3.5.1),
where 'some' here means for at least one such singular (TS 3.5.3).
Buridan gives two conditions which must be met for a term to have
determinate supposition in a given sentence; these conditions are not
given by the semantics of the sentence; rather, they are conditions
which apply in virtue of inferential connections which obtain between
the original sentence and related sentences:
(1) From any given singular falling under the common term the
sentence with the common term follows, all else remainting
unchanged (TS 3.5.5).
(2) All of the singulars can be inferred disjunctively in a dis-
junctive sentence (TS 3.5.6).
Roughly, a common term will have determinate supposition if it is in
the scope of a particular quantifier which is not in the scope of
another logical sign. Thus the term 'man' in "Some man is not a sexist"
has determinate supposition, because (i) "Socrates is not a sexist;
therefore, some man is not a sexist" is an acceptable consequence; (ii)
"Some man is not a sexist; therefore, Socrates is not a sexist or Plato is
not a sexist or Aristotle is not a sexist et sic de singulis" is an accept-

able consequence. The conditions for determinate supposition, then,
roughly correspond to existential generalization and instantiation.
A term in determinate supposition is used to talk about at least one
of the things it signifies. To what, then, does it refer? The answer is
clear: 'some man' in such sentences refers to some man. Geach has
noted a problem here: 67 if 'supposition' is a relation of reference, then
what does 'some man' in the sentence "Some man is F' refer to when
the sentence is false? If 'some man' refers to some man, then we may
ask which man. But the sentence is false, and so it cannot refer to
those men who are in fact F, for there are none. Perhaps it refers to
every man, being false in each particular case? But then there would be
no difference between 'some' and 'every; since 'every man' refers to
every man. 6H Thus the very idea that a term can refer to only part of
its extension is incoherent.69
Geach himself (unwittingly) offers a solution ([19621 p. 7):
Suppose Smith says, as it happens truly: "Some man has
been on top of Mount Everest." If we now ask Smith
"Which man?" we may mean "Which man has been on top
of Mount Everest?" or "Which man were you, Smith,
referring to?" Either question is in order ...
Without realizing it, Geach is pointing out that 'some' can have a refer-
entialor an attributive use. 70 In its referential use, 'some' picks out a
particular man or men, so that we may immediately assess the sent-
ence's truth-value. in its attributive use, though, 'some' commits us to
the sentence's truth-value, so that we may determine the reference
(extension) of the term. The referential use roughly corresponds to
r
'some (i.e. this or these and the attributive use to 'some (i.e. some-or-
other).'
To return to Buridan: a term used referentially has determinate sup-
position, and a term used attributively has non-distributive confused
supposition. The particular sign of quantity 'some' normally has deter-
minate supposition, that is, is used referentially; the inference-rules
given above guarantee this. In TS 3.5.1 Buridan says that a term has
determinate supposition when it is true for some determinate supposit,
that is, referentially. The very point of the first rule is to insist on the
referential reading. Further support is found in the uncommon idiom
for negatives, which is equivalent to branching particular quantifica-
tion, each taken referentially though not necessarily for the same
item(s). But 'some' in certain sentential contexts can, in combination
with other logical terms, have non-distributive confused supposition,
that is, be used attributively. Such uses will be discussed in the next
section. The rules of supposition are primary; it is a mistake to think
that a term always has the same reference, no matter what the senten-
tial context. That is one of the central points of supposition theory,
noted in Section 6.1.
Hence 'some man' in the false sentence "Some man is F" refers to
some man, i.e. to some particular man. The sentence is false because
that particular man is not F. Even if other men are F, the sentence will
be false, for they are not being referred to. Modern logicians are
accustomed to read 'some' attributively, and so to take the F-ness of
other men to suffice for the truth of the sentence. Yet nothing forces
us to accept the attributive reading; the referential reading is certainly
a possible interpretation, and in fact seems to accord with ordinary
language and linguistic intuitions far better than the attributive reading.
It might be objected that on the referential reading there is no dif-
ference between discrete and determinate supposition. But this is an
artifact of our examples; we have taken 'some' to stand for a single
individual. In such a case there is no difference. But 'some' may also
stand for a determinate number of individuals: I may assert "Some
men are bald," referring to six friends of mine whom I know or
believe to be bald. Taken referentially, if any is bald, the sentence is
true; if not, not. The semantic understanding of determinate supposi-
tion should now be evident. A term has determinate supposition if it is
taken referentially for at least one determinate individual it signifies.
Thus what Geach took to be the great sin of mediaeval logic turns out
to be its great virtue. There is nothing incoherent in the notion that a
term may only stand for some of its significates on a particular occa-
sion of its use in a sentence: Geach's worries stem from not taking
seriously the pragmatic dimesnion of semantics, and the contribution
of the logical grammar of a given sentence to determining the refer-
ence of a term.
A final point. The same term may appear in different sentences,
and in each case have determinate supposition. The theory of supposi-
tion has nothing to say about when such occurrences are coreferential.
Nor should it; supposition theory specifies the reference of a term in a
given sentential context. Coreferentiality involves more than a single
sentence;7l additional semantic principles are needed for the wider

context of several sentences. Clearly such principles are vital for the
theory of inference: they are provided in the general theory of the syl-
logism. In particular, the dictum de omni et nullo, discussed in Section
8.2, fixes coreferentiality.
A common term has confused supposition in a sentence if it is not
sufficient for the truth of that sentence that it be true for a singular
term falling under the common term (TS 3.5.1). This is equivalent to
saying that conditions (1)-(2) for determinate supposition are not met
(TS 3.5.7).
6.7. DISTRIBUTIVE AND NON-DISTRIBUTIVE SUPPOSITION
There are two forms of confused supposition; Buridan gives a condi-

tion for each, and then a series of syntactical rules. A term in distribu-
tive confused supposition is used to talk about any or all of the things
it signifies; a term in non-distributive confused supposition (sometimes
called "merely confused supposition") is used to talk about several of
the things it signifies indifferently.
A common term has distributive confused supposition in a sentence
when any of the singulars falling under the common term can be
inferred individually, or all inferred conjunctively in a conjunctive
sentence (TS 3.6.1). Hence the term 'man' in "Every man is running"
has distributive confused supposition, because (i) "Every man is runn-
ing; therefore, Socrates is running" is an acceptable consequence; (ii)
"Every man is running; therefore, Socrates is running and Plato is
running and Aristotle is running et sic de singulis' is an acceptable
consequence. Distributive confused supposition, then, is similar to uni-
versal quantification. The semantic relations involved in distributive
confused supposition are clear: reference is made to everything (pres-
ently existing) which the term signifies; it is "distributed" over each
individual.
Buridan gives five rules for when a common term in a sentence has
distributive confused supposition; they syntactically indicate, by means
of scope conventions governing negation and quantification (and com-
parison), when a common term has distributive confused supposition.
First, a universal affirmative sign distributes the common term it gov-
erns (Rule DC-l in TS 3.7.1); Buridan has a long analysis of what
counts as such a universal affirmative sign (TS 3.7.2-3.7.33). Second,
negations affect distribution: an infinitizing negation distributes terms

in its scope (Rule DC-3 in TS 3.7.42), while a negating negation dis-
tributes every term in its scope which would otherwise not be distri-
buted (Rule DC-2 in TS 3.7.34) - and terms which imply negative
syncategoremata have similar effects (Rule NDC-5 in TS 3.7.50).
Third, comparative contexts produce distribution, that is, the use of a
comparison, an adjective of degree or a superlative (Rule DC-4 in TS
3.7.45).
On the other hand, a common term in a sentence has non-dis-
tributive confused supposition when neither (i) any of the singulars
individually follow, nor (ii) do the singulars follow disjunctively in a
disjunctive sentence, although sometimes a sentence with a disjunctive
extreme follows (TS 3.6.1). The term 'animal' in "Every man is an ani-
mal" is in non-distributive confused supposition, because (i) "Every
man is this animal" does not follow; (ii) "Every man is this animal or
every man is that animal or ... " does not follow - although in this
case, "Every man is this animal or that animal or ... " does follow.
Semantically, a term has non-distributive confused supposition
when it is used attributively of its extension. This explains why a sent-
ence containing such a term does not entail any of the singulars indi-
vidually or in a disjunctive sentence - this characterizes referential
uses - but why a sentence with a disjunctive extreme does follow. The
attributive use of a term will apply to a definite number of individuals
in its extension (depending on the facts of the matter), but it does so
indifferently, applying to each for exactly the same reasons. Such sen-
tences presuppose their truth to determine the (actual) reference of the
term, i.e. the subclass of the signification which actually have the prop-
erty in question. In this sense, non-distributive confused supposition is
close to our use of existential quantification.
Thus distributive confused supposition corresponds to universal
quantification, determinate supposition to existential quantification
used referentially, non-distributive confused supposition to existential
quantification used attributively, and discrete supposition to naming or
denotation. It is clear that this scheme is complete: there is no room
left for other sorts of reference. Mediaeval semantic theory, therefore,
codifies reference-relations by their use and by their extension.
The syntactic specification of non-distributive confused supposition
is more motley than distributive confused supposition; two rules deal
with the effect of quantification, and two with special terms. First, a
universal affirmative sign causes non-distributive confused supposition

in a term not in its scope which is construed with a term in its scope
(Rule NDC-1 in TS 3,8,1 and 3,8.4), and a common term has non-
distributive confused supposition where it is in the scope of two
universal signs, either of which would distribute the rest were the
other not present (Rule NDC-2 in TS 3.8.13). Third, temporal and
locative locutions "often" produce non-distributive confusion (Rule
NDC-3 in TS 3.8.19), as do terms involving knowing, owing, and
desiring (Rule NDC-4 in TS 3.8.24), where the latter have certain
special characteristics.
Buridan's syntactic rules accomplish more than merely telling us
which kind of confused supposition a term has; they permit the exten-
sion of supposition-theory beyond the narrow realm of simple categor-
ical sentences, to deal with complex cases of multiple quantification
such as "Every woman gives some food to every cat." Nevertheless, if
we restrict ourselves to sentences in which the subject and predicate
are logically simple, then we can briefly summarize the kinds of sup-
position possessed by the subject-term and the predicate-term in the
Square of Opposition:
(1) In universal affirmatives such as "All S is P" [A-form], S has

distributive confused supposition and P has non-distributive
confused supposition.
(2) In universal negatives such as "No S is P" [E-form], Sand P
each have distributive confused supposition.
(3) In particular affirmatives such as "Some S is P" [I-form], S
and P each have determinate supposition.
(4) In particular negatives such as "Some S is not P" [a-form],
S has determinate supposition and P has distributive con-
fused supposition.
If we augment the Square of Opposition by adding sentences with sin-

gular subject-terms, then in singular affirmatives and negatives the pre-
dicate has the same supposition as the corresponding particular form,
and the subject has discrete supposition. Do not overlook the fact that
Buridan's theory is much more general than (1 )-( 4) suggest.
Once the supposition of a term has been determined, the sentence
may be inferentially related to other sentences. The most interesting
such sentences are those in which terms which do not have discrete
supposition are replaced by terms which do have discrete supposition;
depending on the inferential direction, the relations are relations of
ascent or descent.72 For example, the I-form sentence "Some dog is
healthy" entails "Some dog is this healthy thing or that healthy thing or
..." where 'some dog' can be replaced by the reference-class in ques-
tion, since it has a referential use.
Modern mediaeval scholarship finds a fatal flaw in the theory of
supposition at this point, known as the "problem of the O-form." The
difficulty is as follows. Suppose that only Socrates and Plato exist,
each of whom is Greek, and consider the (false) O-form sentence
"Some man is not Greek." If we descend under the subject-term, then
(assuming that the reference is to Socrates and Plato) we get "Socrates
is not Greek or Plato is not Greek." Each of the disjuncts is false, and
so the sentence is false. However, if we descend under the predicate-
term, we get "Some man is not this Greek and some man is not that
Greek." If 'this Greek' and 'that Greek' refer to Socrates and Plato
respectively, then the first conjunction is true, taking 'some man' to
be Plato, and the second conjunct is true, taking 'some man' to be
Socrates; both conjuncts are true, and so the conjunction is true,
though the original sentence is false.
The interpretation of supposition-theory offered above suggests a
solution. In the latter case in which we descend first under the predic-
ate-term, a crucial step in the argument was that 'some man' in each
conjunct be taken not referentially, but attributively. Thus each occur-
rence of 'some man' in "Some man is not this Greek and some man is
not that Greek" should be read as 'some man or other (indifferently).'
Yet this is to assign non-distributive confused supposition to the sub-
ject-term; in the example, the subject-term was said to have determin-
ate supposition. To put the point a different way: if the subject-term
has determinate supposition then an equivalent descent can only be
made taking the subject-term first. The assignment of determinate sup-
position to the subject-term thus corresponds to a 'priority of analysis'
rule, and the so-called 'problem' of the O-form is resolved.
6.8. AMPLIATION: TIME AND MODALITY
Natural supposition, as we have seen, ampliates a term to stand for all

of the items it signifies, whether past, present, or future (Rule Amp-5
in TS 6.2.1). Accidental personal supposition, though, may be con-

cerned with the actual or the possible, with the present or past or
future, BuriGan, like most mediaeval logicians, takes tensed sentences
to be complete: they are not to be replaced by tenseless sentences with
a time-index, or expanded to a canonical form; their truth-value can
be assessed immediately, Now personal supposition as described
above strictly applies to present-time assertoric discourse; the exten-
sion of the theory into tensed and modal contexts is the province of
the theory of ampliation (TS 6,1.1): a term is said to be ampliated if
the scope of its reference is widened, made more "ample" (TC 1.6.4),
Naturally, this suggests that ampliation is only possible for terms which
are able to be predicated of many, that is, terms which are semanti-
cally general; singular referring expressions cannot be ampliated and
remain singular. This is what Buridan maintains: in Sophismata V soph-
ism 3 he asserts that discrete terms cannot be ampliated. Hence the
theory of ampliation will only deal with common terms.
Buridan suggests in TS 3.4,8-10 that the copula of a Mental sent-
ence is timeless, but whether this is so will not make any appreciable
difference to his theory; he still has to explain sentences about times
other than the present Buridan admits two forms of temporal amplia-
tion: (i) the copula itself is modified, either adverbially or by tensing
the verb; (ii) certain terms in themselves imply a difference of time. In
either case a term is ampliated from its original status (TS 6.2,1),
which is the present supposition it possesses.
Let us begin with the case in which the copula is tensed; by TC
1.6.3 we know that a term retains its supposition for present items,
and so Buridan offers the following paradigms for tensed sentences:
S was P = What was or is S was P [Rule Amp-l in TS
6.2.1; TC 1.6.4].
Swill-be P = What is or will-be Swill-be P [Rule
Amp-2 in TS 6.2.1; TC 1.6.4].
The supposition of the subject is thus ampliated to include past or
future existents as well as present existents, analyzed as a hypothetical
sentence with a disjunctive subject This sentence is not equivalent to a
disjunction of two sentences. 73 These paradigms have the peculiar con-
sequence that "An old man will be a young man" may be true: this
sentence is equivalent to "What is or will be an old man will be a
young man," and this sentence may be true of a bouncing baby boy,
who will eventually be an old man, will (first) be a young man (Sophis-
mata IV sophism 4). On the other hand, Buridan's analysis avoids
nasty ontological puzzles; there is no way to move from the sentences
"There was a statue" and "That statue no longer exists" to the conclu-
sion "There is a statue which does not exist." However, a version of the
"what is ..." locution can temporally fix the extension of the subject-
term: the subject-term of "What was S was/is/will-be P" refers only to
past existents, of "What is S was/is/will-be P" refers only to present
existents, of "What will-be S was/is/will-be P" refers only to future
existents.
The predicate-term of sentences in which the copula is tensed is
carried along to supposit for items at the time of the verb, provided
that the term itself does not imply a different time (TC 1.6.1 and
1.6.20). Thus in a sentence of the form "S will be P" the subject-term
is ampliated to supposit for what is or will be S, while the predicate-
term sup posits for what will be P.
In the second case individual terms produce temporal ampliation;
the most obvious instance is of tensed participles, such as 'going-to-
dispute' (disputurus, TS 5.2.10-12); when such a term appears in pre-
dicate position it takes us to a time relative to the time of the verb: in
"Aristotle was going-to-dispute" the predicate-term 'going-to-dispute'
supposits for the future relative to the time of the verb, which is some
point in Aristotle's life (TS 5.2.11, 6.2.7; TC 1.6.20). This is quite like
the systematization of tense-logic given by Reichenbach [1947]; there is
no reason these relations cannot be iterated.
Other terms also produce temporal ampliation; we can divide them
into two categories (RuJe NDC-3 in TS 3.8.19): (i) terms which imply
in themselves a difference of time, such as 'dead'; (ii) terms which
determine, relatively or absolutely, a different time, such as 'tomorrow'
or 'afterwards' (relative) or a dating of the sentence as '12 July 1358'
or 'eternally' (absolute).
Buridan is a temporal divisibilist: all intervals are divisible into
intervals, and intervals only, infinitely; in particular, Buridan rejects
an ontology which includes instants or moments of time. Two argu-
ments for temporal divisibilism are offered in Sophismata VII: first,
there is no single instant at which a spoken sentence exists, since the
subject-term is uttered before the rest of the sentence; but it is part
and parcel of Buridan's nominalism that a sentence, whether utterance
or inscription, only has a truth-value when it exists; hence truth-value

is relative to a given temporal interval (sophism 1). Second, time is
divided into past, present, and future; but the past and the future
admittedly have duration. To suppose that the present has no duration
would introduce an asymmetry into the account of time. Hence the
present is not an instant, but an interval (sophism 4). Therefore tensed
sentences are evaluated for truth-value with respect to a given divisible
interval. The interval with respect to which a given sentence is evalu-
ated is determined by context, and may be as long or as short as you
like; events within a given interval are treated as co-present. In order
to preserve bivalence, Buridan relies on the asymmetry between
affirmative and negative sentences:
An affirmative sentence is true with respect to a given
interval if and only if it is true in some subinterval.
A negative sentence is true with respect to a given inter-
val if and only if it is true in every subinterval.
Hence (i) "Socrates is asleep" and "Socrates is awake" can both be
true with respect to the same temporal interval, and (ii) the consequ-
ence "Socrates is asleep; therefore, Socrates is non-awake" is accept-
able for the same interval. 74 But (i) and (ii) are compatible; Buridan
rejects the consequence "Socrates is non-awake; therefore, Socrates is
not awake," which would deny bivalence: it is illegitimate to pass from
an affirmative sentence with an infinite predicate to a negative sent-
ence with a finite predicate.
Of course, the subject and predicate of tensed sentences may be
quantified; this will not affect the analysis unless the quantifier is tem-
poral, e.g. "whenever" or "sometime," in which case it will affect the
supposition in an obvious way.
If there is no temporal ampliation, that is, in a sentence with a pres-
ent-time verb and a non-ampliative predicate such as "S is P," Buridan
allows the consequence "S is P; therefore, S is or exists" (OM II q.4).
More generally, Buridan explores the logic of ampliated sentences in
TC III, in Theorem III-9 through Theorem III-12, where he argues
(TS 3.4.72) that their logic wholly derives from the logic of non-ampli-
ated sentences.
Ampliation is only relevant to divided modal sentences, since com-
posite modals may be treated as ordinary quantification across sen-
tences; divided modals behave in much the same way as tensed

sentences. In fact, they are so similar it is not clear why Buridan
doesn't give a uniform treatment of both. For example, he could dis-
tinguish composite and divided temporal sentences, such as "It will be
the case that Smith is the mayor" and "Smith will be the mayor," treat-
ing such sentences like composite and divided modals. The copula of
present-time assertoric discourse seems doubly indexical, for presen-
tiality and actuality; I do not know why Buridan did not exploit this
apparent parallel.
Divided modals in which 'possibly' or 'necessarily' adverbially mod-
ify the copula, or where the verb is 'can' or 'must; ampliate the subject
to supposit for possibles (TC 2.4.1):
S can be P = What is or can be S can be P [Rule Amp-3

in TS 6.2.11.
S must be P = What is or can be S must be P [Rule
Amp-4 in TS 6.2.1, Theorem II-2 in TC 2.6.81.
In both cases the subject-term is ampliated to stand for possibles; this

has the interesting corollary that when we know the truth of a modal
sentence we know an infinite, the possibles (QSP I q.ll). In fact, there
is a close link between time and modality: Buridan says that possibility
is "directed toward the future." But note that he does not say that
every possibility in actualized at some time; he gives the example
"Many things are possible which neither were nor are nor will be" (TC
1.6.9), and in stating Rule Amp-4 (TS 6.2.1) he suggests that a term
may supposit for things which never exist, the "merely possible."
What are the ontological commitments of temporal and modal dis-
course? Contemporary philosophers, following Quine, take the idiom
of ontological commitment to be determined by quantity alone: exis-
tentially-quantified sentences are the bearers of ontological commit-
ment. Buridan, however, takes the idiom of ontological commitment to
be determined by time, quality, and mode: present-time affirmative
assertoric sentences are the bearers of ontological commitment. This
alternative view has several important consequences. First, quantity is
irrelevant; universal as well as existential present-time affirmative
assertoric sentences carry ontological commitment - the so-called
"existential import" of universal sentences. Thus "All unicorns have
horns" commits one to the existence of unicorns as surely as "Some
unicorns have horns" does. Second, negative sentences do not carry

any ontological commitment; their truth-value is determined by the
truth-value of their contradictory affirmative. Thus "Some unicorns do
not have horns" does not carry any commitment to the existence of
unicorns.1 5 Third, Buridan insists that time and mode are relevant to
ontological commitment. Time is not normally taken into account by
contemporary logicians; Quine, for example, takes "eternal sentences"
to be the logical form of sentences in his canonical language. Buridan,
though, allows tensed sentences in Mental, and so his insistence that
time is relevant to ontological commitment is a substantive disagree-
ment with the modern tradition. Thus "Some man will be born in
exactly ten years" does not carry any ontological commitment for
Buridan. Equally, mode is relevant; "Some man can be standing in the
doorway" does not carry any ontological commitment for Buridan.
Let us explore this difference further. Modal logicians today usually
reduce de re to de dicta modality, and cash the latter out in terms of
what actually happens in peculiar conditions: modal sentences are
reduced to ordinary assertoric sentences with the ontology expanded
to include possible worlds. Thus "Socrates can be bald" is canonically
represented as "There is a possible world in which Socrates is bald".7n
Mediaeval logicians, on the other hand, reduce de dicta to de re
modals, or countenance both sorts. In particular, Buridan refuses to
reduce modal discourse to any other form: rather than being a dis-
guised form of assertoric discourse under peculiar conditions, modal
discourse is a primitive form of discourse in normal conditions. Sim-
ilarly, tensed discourse is not reducible to present-time or tenseless
discourse with an expanded ontology including past and future exist-
ents; it is a primitive form of discourse in normal conditions. We
might capture the difference by saying that mediaeval logicians sever
quantification (and hence supposition or reference) from ontological
commitment. The truth of modal or tensed sentences does not require
us to expand our ontology. Buridan is a nominalist: he does not admit
any abstract entities. But it is not at all clear that he must likewise
refuse to quantify across past or future items, or possible items, so
long as they are not abstract. Indeed, since supposition is a theory of
reference, it seems clear that can can talk about such items; what is
important is that we not take them to exist. And that is the moral of
Buridan's alternative conception of ontological commitment.
6.9. TRUTH-CONDITIONS
Armed with the theory of supposition, we can return to the truth-con-

ditions of categorical sentencs. First, Buridan takes the proposed
definition of truth "howsoever a sentence signifies to be, so it is, in
the thing(s) signified" - to apply directly to a present-time affirmative
assertoric sentence. But there are other kinds of sentence. Past-time
sentences are true if howsoever the sentence signifies to have been, so
it was, in the thing(s) signified; future-time sentences are true if howso-
ever the sentence signifies to be going to be, so it will be, in the
thing(s) signified; and modal sentences are true if e.g. howsoever the
sentence signifies to be able to be, so can be, in the thing(s) signified
(TC 1.1.7-11, OM VI q.8). As remarked at the end of Section 5.5,
the application of this definition requires the theory of supposition,
explaining what is meant by the clause "in the thing(s) signified." If we
confine ourselves to the simple sentences of the traditional Square of
Opposition, then we may state their correspondence truth-conditions 77
as follows:
(1) If an A-form sentence such as "Every S is P" is true, then

everything the subject supposits for the predicate supposits
for.
(2) If an I-form sentence such as "Some S is P" is true, then
something the subject sup posits for the predicate supposits
for.
(3) The truth-value of the E-form (e.g. "No S is P") and
a-form (e.g. "Some S is not P") sentences is determined by
(1) and (2).
Now (1) and (2) are rather similar to giving truth-conditions in terms
of set inclusion among the extensions of the terms. But note that they
are stated as necessary conditions for truth, not as sufficient conditions.
That is because of problems with Liar-sentences (TC 1.5.5-1.5.7;
Sophismata II Theorem 12 and VIII sophisms 7 and 11 especially).
Recall that Buridan takes sentences to be assertions, and hence the
sentential form requires certain contextual prerequisites be met for the
sentence to count as a sentence. An affirmative assertion (sentence)
indicates that its terms supposit for the same, according to the require-
ments of the given sentence; a negative sentence indicates the opposite
(TC 1.5.1-2). But this is not all, for Buridan also holds what Hughes
has felicitously named the Principle of Truth-Entailment ([1982]) p.
110): A sentence 'p' and a sentence of the form 'A exists: where 'A'
names p, together entail a sentence of the form 'A is true: (Naturally
'p' as it occurs in the latter sentences is in material supposition and
only equiform to the original sentence.) This principle is quite intuitive
if we remember that sentences are assertions: an utterance which actu-
ally counts as an assertion contextually presupposes that the assertion
is true. Buridan sometimes expresses this loosely, by saying that the
truth of a sentence is nothing other than that very sentence itself (OM
II q.l fo1.8vb). He is thinking of Mental sentences, of course, but the
point is that 'truth' is eliminable: we do not need a truth-predicate.
What are the causes of the truth or falsity of a sentence? Buridan
discusses this question in TC Chapter 1-2 and OM VI q.8. We have
seen (in Section 5.5) how Buridan rejects various suggestions about
what a sentence signifies, especially the complexe significabile. But
what then is the cause of the truth or falsity of a sentence? Some fact
or event? But then it seems as though we have to admit negative facts,
or perhaps future facts, and the like. Modem logicians sidestep some
of these difficulties by taking negation as a sentential operator, so that
negated sentences are molecular; their truth-value is then derived from
the truth-values of atomic sentences. Buridan, however, admits as basic
two forms of the copula ('is' and 'is not,), and so cannot treat the
truth-value of negative sentences as derivative. Rather, he takes a more
radical line: there is no cause of the truth of a negative, just as there is
no cause of the falsity of an affirmative. The causes of the truth of an
affirmative are equally causes of the falsity of the corresponding nega-
tive: the way things are. And that is the end of the story. Further sup-
port for this view can be found in the fact that Buridan only states
correspondence truth-conditions for A-form and I-form sentences,
since there is no cause of the truth of the E-form and O-form sen-
tences.
Buridan also has several remarks about the "number" of causes of
truth, which are as one would expect: the sentence "All men are sex-
ists" has more causes of its truth than "Some men are sexists." These
causes of the truth of the sentences are related by set-inclusion, it
should be noted. Such principles are crucial for proving equipollence
and conversions, as we shall see in Section 7.3.
7. CONSEQUENCES
7.1. CONDITIONALS, INFERENCES, AND CONSEQUENCES
Buridan's theory of consequences covers material which modern logic

treats under the separate headings of a theory of conditionals and the
rules of inference. These are not distinguished by the logicians of the
fourteenth century; 7H this need not be an error - we need to recall the
philosophical motivations for drawing the distinction initially.
Consequences in modern logic are distinguished as conditionals
and rules of inference; they are not merely equivalent: to show that it is
possible to pass from one to the other a Deduction Theorem is needed,
which is not a trivial matter. It is not obtainable in incomplete systems.
Each is specified syntactically, but conditionality is represented by a
sign in the language (as primitive or a defined abbreviation) appearing in
formulae, whose behavior is given by axioms, while rules of inference
are neither stated in the language nor in the syntactic recursive
grammar, but metalinguistically govern the production of a formula
from other formulae. Inference rules are tied to deducibility and
provability, and hence to validity by the notion of syntactic con-
sequence; conditionals are loosely tied to truth and interpretation.
Buridan, as noted, does not distinguish an object-language from a
metalanguage, so it would be difficult for him to arrive at precisely our
distinction between conditionals and rules of inference. 79 Yet aside from
the incompleteness of logical systems, a worry it would be anachronistic
to ascribe to Buridan, there are two standard philosophical reasons
given for distinguishing conditionals and rules of inference. The first is
classically expressed by Lewis Carroll in "What the Tortoise Said to
Achilles," and takes the form of an infinite regress. Briefly, the Tortoise
asserts "p" and "if p then q" but rejects Achilles' claim that he must
perforce assert q: that only follows, maintains the Tortoise, if he asserts
"if p and if p then q, then q." Yet even if he accepts this principle, he
continues, he need not assert q, for that does not follow unless he also
asserts "if if p and if p then q, then q," and so on ad infinitum. But this
argument will not serve its purpose. First, if correct, it shows that any
59
system containing a Deduction Theorem includes this infinite regress,

though it be carefully disguised in an axiom-schema or a schematic
rule of inference. Second, the argument suggests the seeds of its own de-
struction: do we not need justification for the rule of inference, and will
not this justification stand in need of justification? The distinction does
not prevent the regress.
The second argument for distinguishing conditionals from rules of
inference is based on philosophical analysis of ordinary language: it
captures in a formal way the difference between assertions which do or
do not require a commitment to the truth of the first statement. One is
so committed when using the inferential form, one is not so committed
when using the conditional form. Perhaps these considerations about
acceptance and commitment are what motivate the infinite regress given
by the Tortoise - who, after all, is dialectically arguing with Achilles,
and so seeks persuasion.
The second argument is on the right track, but then there is no
overwhelming reason to distinguish the cases as being of different kinds
rather than as species of a single genus, the mediaeval notion of con-
sequence. Hence we are not forced to distinguish them.Ho
Consequences are similar to inference-rules in three principal
ways: first, we only have a consequence when it is impossible that the
antecedent obtain with the consequent failing to obtain; otherwise we do
not have a consequence at all. Conditionals are identified syntactially,
though, and are no less conditional for being false in an interpretation,
Consequences thus seem more similar to inference-rules, for there is
no inference when the first formula obtains and the second does not.
Second, consequences are not specified syntactically, but are defined
by the relations obtaining between what the antecedent and consequent
assert (to speak loosely). Third, consequences are called accceptable,
not 'true' or 'false.'
Consequences are similar to conditionals in two principal ways: first,
the sentences which appear as antecedent and consequent are used, not
mentioned, despite the fact that technically they are not sentences (no
part of a sentence is a sentence). Second, the definition of consequence
is remarkably similar to Lewis-Hacking's notion of strict implication, in
the modal characteristics of the definition.
7. CONSEQUENCES 61
7.2. DEFINITION AND DIVISION OF CONSEQUENCES
In TC 1.3.4-12 Buridan considers the proper definition of "con-

sequence," eventually proposing the following (see especially TC
1.3.11):
A sentence of the form "'if p then if' or "p; therefore, q" or
an equivalent form is a consequence if and only if a sentence
p* equiform to p and a sentence if' equiform to q are so
related that it is impossible that both (i) it is the case as p*
signifies to be, and (ii) it is not case as if' signifies it to
be, provided that they are put forth together, and mutatis
mutandis for each class of sentences.
This definition calls for some comment. First, the constituent parts of a
sentence are not themselves sentences, which is why we must specify
sentences equiform to the parts of a grammatically consequential
sentence, i.e. the protasis and apodosis of the sentence; Buridan
carefully calls these the 'first part' or the 'second part' of a grammatically
consequential sentence. Hl When we have a consequence, we may call
these the antecedent and consequent, but not otherwise. Second, the
clause 'provided they are put forth together' is meant to rule out cases
where either part is not asserted (TC 1.3.8). Third, the final clause
'mutatis mutandis for each class of sentences' is meant to remind us that
the tense and mood of the verbs in (i)-(ii) have to be taken seriously,
since Buridan denies a general formula for truth and gives criteria for
each class of sentence: assertoric or modal, past-time or present-time or
future-time, affirmative or negative, and so forth. Finally, note that the
definition is semantic, because it defines 'consequence' in terms of the
relation holding between what p* and if' assert or say is the case. H2
Buridan almost never calls a consequent 'true' and 'false' in this
treatise. Rather, he calls them acceptable (bona) or not, though
technically an unacceptable consequence is not a consequence (this is an
instance of the laxity noted above). A consequence is acceptable if it
satisfies the strict definition for consequences as given above. In that
respect consequences are indeed like inferences, which are valid or
invalid, not true or false. H3
Buridan's division of consequences is found in TC I.iv (see Figure 3).
The first division is between formal and material consequences. A
consequence is formal if and only if it satisfies the Uniform Substitution
Formal consequences Material consequences
/
Simple consequences Ut nunc consequences
Fig. 3. Buridan's Division of Consequences
Principle (TC 1.4.2-3): it is acceptable for any uniform substitution for

any of its categorematic terms; otherwise the consequence is material.
Formal consequences are in this respect like tautologies, and so like
strict implication: they remain acceptable (true) for any uniform sub-
stitution of categorematic terms (non-logical constants).
Material consequences fail this test, but may yet be necessary; such
material consequences are called simple. As an example Buridan points
out that the sentence "A man runs; therefore, an animal runs" is not
formal, for a substitution-instance is "A horse walks; therefore, wood
walks" (TC 1.4.3.). The genus-species relation between 'man' and
'animal' need not be preserved under substitution. On the other hand, it
is clearly a necessary consequence. K4 Material consequences come in
two forms: simple and ut nunc. A simple material consequence is a
consequence which is not formal but satisfies the strict definition of
consequence given above. They are acceptable only through reduction
to a formal consequence, namely "by the addition of some necessary
sentence or sentences which, when assumed with the antecedent, render
the consequence formal" (TC 1.4.4). In the example above the necessary
sentence is "All men are animals," so that "A man runs and all men are
animals; therefore, an animal runs" is a formal consequence. Simple
material consequences are thus treated as enthymematic (TC 1.4.5-6).
The final kind, ut nunc consequences, are strictly speaking not neces-
sarily consequences: they are acceptable if we relace the 'impossible' in
the definition of consequence with 'it is not the case.' An ut nunc
consequence, then, is a sentence such that it is not the case that the
antecedent obtains and the consequent fails to obtain. But here again we
7. CONSEQUENCES 63
may take the tense of the verb in 'it is not the case' seriously, and so
discuss consequences which hold as a matter of fact at other times - as
Buridan says, ut nunc, ut tunc, or ut nunc pro tunc. Thus Buridan's
theory of tense-logic will be confined to part of the theory of ut nunc
consequences. Note that the ut nunc consequence may behave just like
the material conditional; all that is required for it to be acceptable is the
factual lack of the antecedent obtaining with the consequent failing to
obtain.
Consequences are further examined in the first seven theorems in
TC 1. Theorem 1-1, Theorem 1-5, and Theorem 1-7 state the key
characteristics of consequences: they can never lead from truth to
falsity, nor from possibility to impossibility, nor from the necessary to
the non-necessary; equally, the necessary follows from anything, and
from the impossible (such as the conjunction of contradictories)
anything follows. In Theorem 1-3 the law of contraposition is stated for
consequences, and in Theorem 1-4 the law of transitivity. Together, all
of these characteristics define the nature of an acceptable consequence.
7.3. ASSERTORIC CONSEQUENCES
The rest of the theorems in TC I are devoted to particular assertoric

consequences. These are of two forms: equipollence, that is, sentences
which follow from each other as consequences, and conversion, in
which the subject-term and the predicate-term of a sentence are the
same, but their positions are reversed or the syncategoremata are
altered (or both). In TC 1.8.47, Buridan says that all equipollences
and conversions are contained in Theorem 1-8 (given in 1.8.40 but as
revised in 1.8.43):
[Theorem 1-8 (revised)] (a) Any two sentences of which
neither can have some cause of its truth which is not a
cause of the truth of the other sentence follow from the
same sentences; (b) any two sentences of which one has, or
can have, more causes of its truth than the other sentence,
although every cause of the truth of the latter is a cause of
the truth of the former, are so related that the sentence
with more causes of its truth follows from the sentence with
fewer, but (c) not conversely.
Though difficult to state precisely, Theorem 1-8 is extremely import-
ant: sentences will be shown to be equipollent by having the same

causes of their truth, and so by (b) mutually follow from each other.
This result Buridan deemed sufficiently important to repeat it in a
separate theorem, Theorem 1-9. Equally, one sentence will be shown
to be the conversion of another by having the same or more causes of
its truth than the original sentence, that is, by Theorem 1-8(a) or (b).
These two theorems are the basis of not only assertoric equipollences
and conversions, but also of modal equipollences and conversions (TC
1.8.50), though these are explored later. These theorems are perfectly
general; Buridan points out (TC 1.8.51) that the sentences "Of-any-man
no ass is running" and "Of-no-man an ass is running" are equipollent.
Hence his theory is applicable far beyond the simple sentences which
appear in the Square of Opposition.
Buridan only gives some examples of equipollences in TC 1.8.51;
they are the equivalences of signs of quantity and negations, so that
e.g. 'Every ... is not - - -' is equipollent to 'No ... is - - -'. The
equipollences are standard, but note that Buridan does not take them
as definatory of the signs of quantity, as modem logicians define one
quantifier by another; their equipollence must be established, which is
a corollary of Buridan's semantic approach to quantification.
Conversions are treated at greater length, in Theorems 1-10 through
1-17. Buridan begins in Theorem 1-10 with consequences by subalter-
nation and conversions of universals into particulars; the theorem
is stated for terms with distributive confused supposition, and
generalized to terms with non-distributive confused supposition in
Theorem I-II. The next pair of theorems, Theorem 1-12 and Theorem
1-13, deal with conversions of sentences containing ampliation or non-
ampliating restrictions (such as the "what is ..." locution).
Theorem 1-14 is a turning-point in Buridan's discussion, specifying
rules of simple and accidental conversion: H5
[Theorem /-14] (a) From any universal or particular
affirmative assertoric sentence there follows a particular
affirmative by conversion of the terms; (b) from any
universal negative there follows a universal and a particular
negative; and (c) no sentence follows from a particular
negative by its form alone.
Buridan discusses this theorem in TC 1.8.70-95. The extended
analysis is required because of the generality of Theorem 1-14: it
7. CONSEQUENCES 65
applies to sentences in which the logical constituents (subject-copula-

predicate) may be implicit rather than explicit, such as the one-word
sentence "Ambulo"; the copula of such sentences, whether explicit
or implicit, may be present-time, past-time, or future-time; the
predicate-term or the subject-term, or both, may be ampliative; the
'what is ... ' locution may be used; the sentence may be existential,
and so (apparently) lack a predicate-term; the rules of grammar may
be violated in reversing the subject-term and predicate-term; the
subject-term or the predicate-term may be logically complex, contain-
ing syncategoremata; they may contain or consist in terms in oblique
cases; certain syncategoremata may require special analysis. Buridan
takes each of these possibilities into account, and his discussion is a
marvel of complex precision.
A singular sentence follows from another singular sentence by
conversion of the terms, as Buridan notes in Theorem 1-15, paying
attention to ampliation and the form of the singular term.
In the next pair of theorems Buridan considers the results when the
subject-term or predicate-term is changed "according to finite and
infinite," i.e. the effects of term-negation (infinitizing negation) on
conversions. This may be called 'conversion' in a broad sense, Lven
though the terms themselves are altered (TC 1.8.101). Theorem 1-16
and Theorem 1-17 set forth such relations of conversion; it is often
necessary to suppose that the varied term is non-empty, the assump-
tion ofthe constantia terminorum. H6
It is a travesty to present mediaeval logic as though only sentences
appearing on the Square of Opposition were investigated. Buridan's
discussion of assertoric consequences is perfectly general, and ade-
quate for his philosophy of logic and language.
7.4. DIVIDED MODAL CONSEQUENCES
In TC 2.2.5 Buridan discusses the equipollence of the various modes

in combination with negations, and in Theorem II-I he establishes
such equipollences as a formal result. There are no surprises;
'necessary' is equipollent to 'not possibly not,' and so forth.x7 Yet there
is an important methodological corollary of such equipollences which
Buridan states in TC 2.6.7: theorems for divided modals need only be
stated for the case in which the mode is affirmed.
Conversions of divided modals are treated in Theorem 11-5 and
theorem II-6. The results are straightforward: affirmative divided

modals de possibili convert both simply and accidentally (to
affirmative divided modals de possibili, of course); universal negative
divided modals de necessario convert simply. No other conversions
are explored.
Buridan does, however, explore 'mixing' theorems, that is, theorems
about which consequences obtain between sentences of different
modes. In Theorem II-3 the relation between assertoric sentences and
divided modals de necessario is stated: the only consequence which
obtains is that from a universal negative de necessario to an assertoric
universal negative. In Theorem II-4 the relation between assertoric
sentences and divided modals de possibili is stated: the only
consequence which obtains is that from an affirmative assertoric to a
particular affirmative de possibili. Buridan specifically remarks the lack
of consequential connection between divided universals de possibili
and their corresponding assertorics (TC 2.7.32), and between divided
particulars de possibili and their corresponding assertorics (TC
2.7.33).
In'TC 2.6.33 Buridan defines the mode 'contingent': it is equipol-
lent to 'possibly and possibly-not.' This mode may also figure in
divided or composite modals, and Buridan explores its behavior in
Theorem II-7, Theorem II-8, Theorem II-14, and Theorem II-19. His
motivation for so doing is not clear; by equipollence, theorems about
composite or divided de possibili sentences will settle questjons about
contingents as well. The formula is simple: replace the contingent
sentence with a conjunction of de possibili sentences. Because this
theoretical simplication is available, we shall not discuss Buridan's
treatment of contingent sentences, either as consequences or as
syllogistic.
7.5. COMPOSITE MODAL CONSEQUENCES
Buridan begins his discussion of composite modal consequences with

Theorem II-9, which states that the following consequence-scheme is
acceptable:
Some [dictum] is [mode]; therefore, every [dictum] is
[mode].
Thus "Some <sentence> The sentence written on the wall is false' is
7. CONSEQUENCES 67
possible; therefore, every <sentence equiform to> 'The sentence written

on the wall is false' is possible" is an acceptable consequence. In par-
ticular the sentence which is the sole inscription on the wall 'The
sentence written on the wall is false' is possible. This view does not
land us in paradox; recall that to say a sentence is "possible" is to say
that it describes a possible situation. The particular inscription on the
wall does just that: for example, it describes the possible situation in
which only '2 + 2 = 1T is written on the wall. Sentences need not
describe their own actual situation to be possible. This is the point of
Theorem 11-9: if an inscription or utterance is possible, then it
describes a possible situation, and all equiform inscriptions or
utterances are equally possible since they may be taken to refer to that
situation. xx The inscription on the wall 'The sentence written on the
wall is false' is possible, but never possibly-true. With this point noted,
Buridan's theorems about composite modal equipollences and
conversions do not pose any special problem.
Conversions of composite modals are straightforward. If the dictum
is the subject, then all composite modals convert simply with the excep-
tion of the universal affirmative composite modal, which is converted
accidentally (Theorem II-I0). If the mode is the subject, then all
composite modals convert simply with the exception of the particular
negative composite modal, which is not converted (Theorem II-II).
Buridan also discusses "conversions with respect to the dictum," in
which the composite modal sentence itself is not converted, but the
dictum of the composite modal is converted. Such conversions with
respect to the dictum are discussed in Theorem U-12, Theorem 11-13,
and Theorem 11-14, and pose no special problems.
As with divided modals, Buridan also offers mixing theorems for
composite modals in relation to assertoric sentences - in particular,
the dictum of the composite modal. We may simplify Theorem 11-15 as
follows:
[Theorem II-15 (revised)1 (i) From any composite
affirmative modal de vero there follows its dictum, and
conversely; (ii) from any composite modal de necessario
there follows its dictum; (iii) from any sentence there
follows a composite modal de possibili of which it is the dictum.
These claims have a modern appearance: (i) is a version of the Tarski
biconditional; (ii) is like the law "Tp' -+ p"; (iii) is like the law "p -+
68 BURlDAN'S PHILOSOPHY OF LOGIC
Tp'." Equally, Theorem 11-16 can be seen as expressing the distribu-

tive laws "( 0 p & (p --+ q» --+ 0 q" and "( <> P & (p --+ q» --+ <> q."
Of course, Buridan expresses these theorems as consequences, which
are not simply conditionals or rules of inference; nor is it strictly
speaking correct to formalize his logic with the propositional calculus,
since no part of a sentence (such as ' 0 p') is a sentence (such as 'p'),
but only contains a part equiform to a sentence. These differences
noted, Buridan's claims are still quite similar to the modern theses.
Finally, Buridan gives some mixing theorems which 'mix' composite
modals and divided modals. Theorem 11-17 says that a particular
affirmative divided modal follows from an affirmative composite
modal (with the dictum affirmed); Theorem II-I8 says that a universal
negative divided modal follows from a universal composite modal
(with the dictum denied). These are the only consequential relations
which hold between composite and divided modals.
8. THE SYLLOGISM
S.l. DEFINITION OF THE SYLLOGISM
When we turn to syllogistic, it is necessary to recall that Buridan is not

working in an artificial language and stating a symbolic calculus: he is
working in a fragment of a natural language, although a highly stylized
fragment (scholastic Latin), and he possessed neither our modern
array of metalinguistic equipment nor our interest in formal metalogic.
It is easy to lose sight of these obvious facts, for Buridan's syllogistic
has an astonishing degree of rigor.
In Buridan's hands syllogistic is a logical instrument of great
flexibility and power, not the rigid and sterile doctrine it later became;
it is directly based on his philosophical semantics. Buridan seems
almost uninterested in the part of assertoric syllogistic which most
people have traditionally identified as 'syllogistic,' quickly moving on
to temporal syllogisms, oblique syllogisms, variation syllogisms, and
modal syllogisms (discussed in Section 8.4).
The initial definition of 'syllogism' in TC 3.2.1 is given only after a
long description of what kind of consequence a syllogism is (TC
3.1.1-17). The syllogism is a formal consequence, and in that sense
syllogistic is merely a branch of the theory of consequences; it is
distinguished by having a conjunctive antecedent and a single conse-
quent, made up with three terms. This paradigm is generalized when
Buridan turns to the oblique syllogism, and allows more than three
terms, provided that the terms connected in the conclusion are parts
of the terms of the premisses (TC 3.6.2-3), a result required for
variation syllogisms as well: more precisely, multiple-term syllogisms
are permitted under strict assumptions about the relations of the
terms.
We can be more exact by introducing some of Buridan's technical
terminology:
(1) The major sentence is the first premiss in the conjunctive
antecedent, and the minor sentence the second premiss in
the conjunctive antecedent (TC 3.2.3).
69
(2) The extremes of a sentence are its subject and predicate;

the syllogistic extremes are the extremes of the conclusion
(TC 3.6.2),
(3) The syllogistic middle is a term or part of a term common
to the premisses; the major extreme is the extreme in the
major premiss which is not the syllogistic middle, and the
minor extreme is the extreme in the minor premiss which is
not the syllogistic middle. The major and minor extremes,
or parts thereof, are connected in the conclusion (TC 3.2.3
and 3.6.2).
When the syllogistic extremes are the major and minor extremes, and
the middle is the same term in both premisses, we have the additional
paradigm of the syllogism; the deviations above are introduced to
allow for development of the syllogistic.
We need three further technical notions to develop syllogistic. First,
we may define a syllogistic figure as the ordering of the syllogistic
middle to major and minor extreme in the premisses (TC 3.2.4);
second, we call a conclusion direct or indirect if the major extreme is
predicated of the minor extreme in the conclusion, or conversely;
finally, we call a particular pair of premisses only characterized with
respect to quantity and quality a conjugation. From these definitions it
is clear that there are exactly four figures (TC 3.2.4-8): (1) the first
figure, in which the syllogistic middle is subject in the major sentence
and predicate in the minor; (2) the second figure, in which the syllogis-
tic middle is predicate in major and minor; (3) the third figure, in
which the syllogistic middle is subject in major in minor; (4) the fourth
figure, in which the syllogistic middle is predicate in the major
sentence and subject in the minor. But if we transpose the premisses
of the fourth figure we have the first figure, changing the conclusion
from direct to indirect or conversely. Buridan correctly notes this
point and does not bother to treat the fourth figure independently (TC
3.2.9). He is quite correct; given the way he has defined the syllogistic
figures, they are not really distinct; it is only on other definitions of
the syllogistic figures that there can be a genuine 'fourth figure'
controversy. H9
8.2. SYLLOGISTIC SEMANTIC PRINCIPLES
A syllogism is a formal consequence with a conjunctive antecedent

8. THE SYLLOGISM 71
and a single sentence as consequent, containing three or more terms.

To understand why a syllogism satisfies the Uniform Substitution
Principle is to understand the semantic framework of a syllogism, and
this requires us to examine how terms are conjoined - not merely in a
single sentence but across two sentences, the conjunctive antecedent.
Buridan sets out this semantic framework in two rules for the accepta-
bility of syllogisms, which require the notion of distribution.
Buridan's two rules are given as his version of the traditional
dictum de omni et nul/a, where one rule applies to syllogisms with
affirmative conclusions and the other to syllogisms with negative
conclusions. His statement is given in TC 3.4.5-7, but Buridan's
distinctions along the lines of discrete and common terms are unneces-
sary. In their full generality they are as follows:
[Rule 11 Any two terms which are called the same as a

third term by reason of the same thing for which that third
term supposits, not collectively, are correctly called the
same as each other.
[Rule 2] Any two terms, of which one is called the same
as some third term of which the other is called not the
same by reason of the same thing for which that third term
supposits, are correctly called not the same as each other.
Two points must be noted: first, the 'not collectively' clause in Rule 1
excludes cases in which two terms are collectively or conjunctively
called the same as a third term, as when "matter and form are said to
be the same as one and the same composite, and the matter is not the
same as the form (TC 3.4.3)."90 Second, the careful use of the phrases
'called the same' and 'called not the same' indicate that Buridan is here
talking in what we should call the formal mode: they characterize
sentences which, if formed, are respectively affirmative or negative in
qualityYl This is the key to Buridan's semantic characterization of the
dictum de omni et nullo: his talk of supposition is entirely in the
formal mode. Syllogistic is a branch of the theory of formal con-
sequence, and so the test for the acceptability of a syllogism is whether
it satisfies the Uniform Substitution Principle, but the principle applies
only if certain relations among the supposition of the terms obtain. In
a particular sentence the relations are made explicit by the nature of
the syncategorematic terms present, and Rule 1 and Rule 2 state how
the terms must be semantically related, i.e. in terms of their

supposition.
The clause 'by reason of the SaI!le thing for which that third term
sup posits' is the central semantic contribution of the dictum de omni
et nullo, because it makes explicit the coreferentiality required for the
same term appearing in different sentences. Coreferentiality is the
underpinning of the theory of inf~rence. Now there are many ways of
fixing reference; how can we tell if the clause 'by reason of the same
thing for which that third term supposits' is satisfied? That is: when
are different occurrences of a middle term coreferential? There are
three cases. First, the third term in question may be a singular
referring expression, that is, a discrete term, and here there is no
trouble, for a discrete term can only supposit for a single thing, as a
matter of semantics. A syllogism with a discrete term as syllogistic
middle is called an 'expository syllogism' (Theorem III-5 in TC
3.4.23-25). Second, if the middle is a common term, then we may
simply stipulate that the syllogistic extremes are called the same
or not the same for the same thing(s) for which the middle term
supposits; this stipulation must be visible in the sentences, since the
syllogism is a formal consequence, and consists in adding an identifica-
tory relative-term to the minor (Theorem III-6 in TC 3.4.29-30).
The third case also involves a common middle term, but where
there is no such stipulation of suppositional sameness. We do not
simply want co extension here; we want coextension as a matter of the
semantics, for a syllogism is a formal consequence - indeeed, a matter
of the semantics obvious from the syntactic form. The answer is given
by the theory of distribution.
Buridan does not specifically address distribution while discussing
the syllogism, but he does not need to: it is covered in his account of
common personal supposition. A term is distributed if in a sentence
it is taken to supposit for all it signifies, that is, if it is used to talk
about or refer to everything it signifies. The most obvious case of
distribution is where a term is joined with a distributive sign (a
universal quantifier), not in the scope of a negation. Buridan's rules
for distributive and non-distributive supposition, discussed above,
explicitly state when a term is said to be distributed and when it is not.
The theory of distribution, and hence of syllogistic, is therefore of
widespread applicability, but if we confine ourselves to simple
sentences as on the Square of Opposition, we may give some rules
8. THE SYLLOGISM 73
for distribution: universals distribute subjects, negatives distribute

predicates, and no other terms are distributed. Hence in an A-form
sentence only the subject is distributed, in an E-form sentence the
subject and the predicate are distributed, in an I-form neither subejct
nor predicate is distributed, and in an O-form sentence only the
predicate is distributed. 92
The motivation for the theory of distribution is clear; we want to
avoid the case in which one extreme is called the same or not the
same as part of what the middle term supposits for, while the other
extreme is called the same or not the same as the other part of what
the middle supposits for. In that case there is no connection between
the extremes through the middle, and no inference will hold by the
semantics alone. Distribution is a way of making sure that the
foregoing case does not occur, by talking about everything the middle
term supposits for. Indeed, it obviously follows from the theory of
distribution that no syllogism made up with two negative premisses
is acceptable (Theorem 111-2 in TC 3.4.15), as Buridan notes. The
principles governing the doctrine of distribution are given in Theorem
111-7 and Theorem 111-8 (TC 3.4.31-36).
8.3. REDUCTION AND PROOF-PROCEDURE
Buridan follows tradition in taking the first four moods of the first
figure to be evident or 'perfect': Barbara, Celarent, Darii, Ferio. Their
acceptability is shown directly by the preceding principles; what I
have called Buridan's "proof-procedure" for syllogisms is his way of
reducing all other syllogisms to these. 93 In each of the various forms of
assertoric and modal syllogistic Buridan takes up, he first points to the
evidentness and perfection of those moods corresponding to the first-
figure moods listed; no further justification is givenY4 Hence it is
incorrect to think that Buridan (or any other mediaeval logician) has a
formal metatheory of deductive systems. Buridan offers two methods
for showing the acceptability of a syllogism: the Reductio-Method
and the Method of Reduction. Both are founded on Aristotle, and
supplement the principles listed in the preceding section.
The Reductio-Method, traditionally only required to show the
acceptability of Baroco and Bocardo, is stated as a general principle
governing consequences with a conjunctive antecedent (Theorem IlI-3
in TC 3.4.17-18):
A grammatically consequential sentence of the form "p

& q; therefore, T' is a (syllogistic) consequence if and only
if for the sentence - r contradictory to r (i) the sentence "p
& - r, therefore, - q" is acceptable for the sentence - q
contradictory to q; (ii) the sentence "q & - r, therefore -
p" is acceptable for the sentence - p contradictory to p.
Since the antecedent of such a sentence is itself equiform to a conjunc-
tive sentence, the theorem holds by consequential contraposition
(Theorem 1-3), De Morgan's Laws, and consequential importation.
The Reductio-Method consists in assuming that a syllogistic con-
consequence fails to hold and showing that clauses (i) and (ii) are
satisfied. We can easily illustrate this method by proving the
acceptability of Bocardo: "Some M is not P, and all M is S; therefore,
some S is not P." Taking the contradictory of the conclusion "All S is
P" and the minor premiss we can syllogize in Barbara "All S is P, and
all M is S; therefore, all M is P," and this conclusion contradicts
the original major premiss; equally, taking the contradictory of the
conclusion and the major premiss we can syllogize in Baroco "All
S is p, and some M is not P; therefore, some M is not S," and this
conclusion contradicts the original minor premiss. Hence Bocardo is
acceptable.
The other way to show the acceptability of a syllogistic consequence
is the Method of Reduction, stated in Theorem III-4 (TC 3.4.20-22):
If the sentence "s; therefore q" is a consequence, then a

sentence of the form "p & q; therefore, T' is a consequence
if and only if a sentence of the form "p & s; therefore, T' is
a consequence.
This theorem holds by Theorem 1-4; we may illustrate the Method of

Reduction to show the acceptability of Datisi, which has the form "All
M is P, and some M is S; therefore, some S is P." We know from the
preceding sections which conversions hold, and in particular that we
immediately infer an I-form from an I-form by transposition of the
terms (obversion); thus the sentence "Some M is S; therefore, some S
is M" is a consequence. Hence we may replace the minor of Datisi
with "Some S is M" and syllogize in Darii "All M is P, and some
S is M; therefore, some S is P," and since this is an acceptable
consequence so is Datisi.
8. THE SYLLOGISM 75
8.4. ASSERTORIC SYLLOGISTIC
Buridan investigates four main forms of assertoric syllogistic: the

traditional syllogistic; the temporal syllogism, in which the terms or
copula are temporally ampliated; the oblique syllogism, in which the
sentences contain oblique terms; and the variation syllogism, in which
the middle term is finite in one premiss and infinite in the other.
Buridan states which theorems in particular are acceptable in
various theorems. His theorems are a model of rigor.'15 It will be useful
to investigate certain general points systematically. First, let us review
Buridan's procedure for the traditional syllogism - a distinctly non-
traditional approach.
In TC 3.4.37 Buridan notes that the semantic principles for the
syllogism discussed above specify the traditional syllogistic, and
in 3.4.38 he approaches the problem in a combinatorial way. A
conjugation is a pair of premisses; if we simply list them as permuta-
tions of quantity and quality, there are sixteen possibilities (numbered
for further references): (1) affirmative universal and affirmative
universal [AA]; (2) affirmative universal and affirmative particular [AI];
(3) affirmative universal and negative universal [AE]; (4) affirmative
universal and negative particular [AO]; (5) negative universal and
affirmative universal [EA]; (6) negative universal and affirmative
particular [EI]; (7) negative universal and negative universal [EE];
(8) negative universal and negative particular [EO]; (9) affirmative
particular and affirmative universal [IA]; (10) affirmative particular and
affirmative particular [II]; (11) affirmative particular and negative
universal [IE]; (12) affirmative particular and negative particular [10];
(13) negative particular and affirmative universal fOAl; (14) negative
particular and affirmative particular [01]; (15) negative particular
and negative universal rOE]; (16) negative particular and negative
particular [00].
The principles regarding distribution allow us to reject some con-
jugations immediately, no matter what the figure, since they have an
undistributed middle. By Theorem III-2 we reject all made up of a
pair of negatives, namely (7)-(8) and (15)-(16); since the middle is
not distributed at all in the case of two particular affirmatives, we
reject (10) as well (TC 3.4.39-40). Now we must move to particular
figures.
In the first figure, (9) and (13)-(14) have an undistributed middle,
and so they are to be rejected (3.4.41-42). Buridan accepts the

remaining eight as "useful" (utile).96 However, there are some
surprises. (1) can conclude directly, in which case we have l-AAA
(Barbara), or indirectly, in which case we have I-AAI (Baralipton); (5)
may also conclude directly or indirectly, and we have respectively
l-EAE (Celarent) and I-EAE (Celantes); (2) may also conclude
directly or indirectly, and we have respectively I-AIl (Daril) or I-AIl
(Dabitis). Then we have (3), which can only conclude indirectly as
l-AEO (Fapesmo), and (11), which can only conclude indirectly as
I-lEO (Frisesomorllm). But then the surprises start.
We need to understand what Buridan calls the "uncommon idiom
for negatives." The common idioms for negation, of course, are where
the negation precedes either the copula or the predicate. In the
"uncommon idiom" the predicate-term precedes the negation, so we
have a sentence such as "Some A B not is (Quaedam A B non est)."
Now Buridan first mentions this "uncommon idiom" in TC 1.8.70, but
he is nowhere very explicit about its logical behavior. From his
remarks, though, we may say that the "uncommon idiom" for negatives
is equivlent to a sentence in which the subject and predicate are both
particularly quantified. That is, a sentence such as "Some A B not is"
is to be read as though it were "Some A is not some B," where the
quantification is branching: this sentence is true if and only if there is
some A, say A" which is not the same as some B, say Br This holds
no matter what the rest of the A and the B are like. The reason
Buridan does not simply write this as I have suggested, as "Some A is
not some B," is that this violates his rules for distribution and scope.
Now take (6): when it concludes directly, we have l-EIO (Ferio).
But it may also conclude indirectly, so we have l-EI with a conclusion
not in the common idiom for negatives, as "No M is P, and some S is
P; therefore, some P S not is [i.e. some P is not some Sl." This is not
normally part of the traditional syllogistic, but Buridan is perfectly
correct to say that it is an acceptable syllogism. Similarly, (4) and (12)
conclude directly, though not in the common idiom for negatives, so
we have respectively l-AO "All M is P, and some S is not P; there-
fore, some S P not is [i.e. some S is not some Pl," and 1-10 "some M
is P, and some S is not M; therefore, some S P not is [i.e. some S is
not some Pl."
In the second figure, we reject (1 )-(2) and (9), because the middle
8. THE SYLLOGISM 77
is not distributed (TC 3.4.49); there are eight remaining useful moods.
They are as follows (note that the moods which conclude both directly
and indirectly are not given different traditional names): (5) concludes
directly and indirectly, so we have 2-EAE (Cesare); (3) concludes
directly and indirectly, so we have 2-AEE (Camestres); (6) concludes
only directly as 2-EIO (Festino); (4) concludes only directly as 2-AOO
(Baroco). These are all of the acceptable syllogisms recognized by
traditional syllogistic, but Buridan finds another four acceptable
syllogisms. There is (11), which only concludes indirectly as 2-IEO;
this is the same as Festino with the premisses transposed (making the
conclusion indirect), and so Buridan names this mood "Tifesno." There
is also (13), which only concludes indirectly as 2-0AO; this is the same
as Baroco with the premisses transposed (making the conclusion
indirect), and so Buridan names this "Robaco." Buridan also finds (12)
acceptable, as 2-10 with the conclusion not in the common idiom for
negatives, and therefore having the form "Some P is M, and some S is
not M; therefore, some S P not is [i.e. some S is not some P]." Finally,
(14) is acceptable, as 2-01 with the conclusion not in the common
idiom for negatives, and therefore having the form "Some P is not
M, and some S is M; therefore, some S P not is [i.e. some S is not
someP]."
The third figure requires us to reject (12) and (14), since the
middle is not distributed (TC 3.4.51); the remaining nine moods are
acceptable, and they are all in the common idiom for negatives. The
traditionally accepted moods are as follows: (13) concludes indirectly
as 3-0AO (Bocardo); (1) concludes both directly and indirectly as
3-AAI (Darapti, no separate names); (9) concludes both directly and
indirectly as 3-IAI (Disamis, no separate names); (2) concludes both
directly and indirectly as 3-AII (Datisi, no separate names); (5)
concludes directly as3-EAO (Felapton); and (6) concludes directly
as 3-EIO (Ferison). Buridan adds three acceptable syllogisms, each
of which has the premisses of a traditional mood transposed: (4)
concludes indirectly as 3-AOO, which Buridan names "Carbodo,"
since the premisses are transposed from Bocardo; (3) concludes indi-
rectly as 3-AEO, which Buridan names "Lapfeton," since the pre-
misses are transposed from Felapton; and (11) concludes indirectly as
3-IEO, which Buridan names "Rifeson," since the premisses are trans-
posed from Ferison.
Buridan is correct when he embraces his non-traditional moods:

they are in fact acceptable. Thus even the most traditional part of
assertoric syllogistic turns out to have surprises when Buridan turns
his attention to it. Buridan then considers which moods are acceptable
if the terms or copula is temporally ampliated: the temporal syllogism,
discussed in TC 3.5.54-72. We may summarize his results briefly,
since they are the basis for Buridan's tense-logic.
If the middle term is ampliative (we may regard an ampliative
copula in the minor as an ampliative middle), then if no other term is
ampliative the following moods alone are acceptable: (i) in the first
figure, both Celarent and Celantes; (ii) in the second figure, both Ce-
sare and Camestres, each concluding directly and indirectly; (iii) in
the third figure, all of the standard moods are acceptable, namely
Bocardo, Ferison, Felapton, Carbodo, Rifeson, Lapfeton, and Darapti
and Disamis concluding directly and indirectly. On the other hand, if
there is an ampliative major extreme (we may regard an ampliative
copula in the major as an ampliative major extreme), then if no other
term is ampliative the following moods alone are acceptable: (i) in the
first figure, Darii, Ferio, Baralipton, Celantes, Dabitis, Fapesmo, Frise-
somorum; (ii) in the second figure, Festino, Baroco, Tifesno, Robaco;
(iii) in the third figure, all of the standard moods are acceptable,
namely Bocardo, Ferison, Felapton, Carbodo, Rifeson, Lapfeton, and
Darapti and Disamis concluding directly and indirectly.
These two accounts may be put together to obtain a complete
account of temporal syllogistic (TC 3.4.70-72), in the obvious way: if
the major extreme and middle term ampliate to the same time, the
traditional syllogistic applies; if they ampliate to different times, then
the acceptable syllogisms are those each account finds acceptable
(namely the acceptable third-figure syllogisms). All cases reduce to
these two.
The oblique syllogism is Buridan's next subject, which he treats in
Theorem III-13 through Theorem III-17 (TC 3.6.1-3.7.30). These
theorems largely regulate the behavior of oblique terms in distributive
contexts, which is at the heart of syllogistic consequence. Buridan
treats the oblique syllogism in a way unlike all other syllogistic forms
he investigates: his procedure maintains the high level of rigor, but the
theorems are rather more like recipes - they indicate how to build an
acceptable syllogism from a sentence with certain characteristics. For
example, if the oblique term appears at the very beginning of the
8. THE SYLLOGISM 79
sentence, then if we treat it as the subject and the rest of the sentence
as the predicate all of the standard moods apply. We shall not con-
sider the oblique syllogism in any detail here.
The last form of assertoric syllogism Buridan considers is the
variation syllogism, that is, a syllogism in which the middle term is
finite in one premiss and infinite in the other97 (i.e. the term is "varied"
from one to the other), in Theorem III-18 and Theorem III-19 (TC
3.7.31-45). We may summarize his claims: in every figure, there is an
acceptable pair of variation syllogisms for the conjugations (1 )-(2),
(7)-(10), and (15)-(16); in the first figure, there is also an acceptable
pair of variation syllogisms for the conjugations (3)-(4), respectively
l-AE and l-AO, and for the conjugations (11)-(12), respectively
I-IE and 1-10; there are no additional variation syllogisms in the
second figure; in the third figure, there is an acceptable pair
of variation syllogisms for the conjugations (3)-(6), respectively
3-AE, 3-AO, 3-EA, 3-EI, and the conjugations (11)-(14), respec-
tively 3-IE, 3-10, 3-0A, 3-01. These are the only acceptable variation
syllogisms.98
8.5. COMPOSITE MODAL SYLLOGISTIC
Buridan investigates three forms of modal syllogistic: syllogisms whose

premisses are purely composite modals; syllogisms whose premisses
are purely divided modals; and syllogisms whose premisses and con-
clusions are mixed - either a mix of composite and divided modals,
or of assertoric and modal sentences.
Composite modals are the easiest case. If we recall from above that
in a composite modal the mode is subject and the dictum predicate, or
conversely, and that Buridan allows such sentences to be quantified,
then composite modal logic is simply a branch of the standard
assertoric syllogistic. For example, "No possible <sentence> is that snow
is white, and some utterance is that snow is white; therefore, some ut-
terance is not a possible <sentence>" is a straightforward syllogism in
Festino.
But there is a more interesting aspect to composite modal syllogis-
tic, namely if we consider what we may call "syllogisms with respect to
the dictum," on analogy with Buridan's conversions with respect to
the dictum, discussed in TC 2.7.18-24. By Theorem IV-l (TC
4.1.10-12) Buridan asserts we may prefix the indefinite mode
'necessary' or 'true' to the premisses and conclusion of a standard
assertoric syllogism and get an acceptable syllogism, which is never the

case for the modes 'possible' or 'false: For example, if we have an
assertoric syllogism in Darapti of the form "All M is P, and all M is S;
therefore, some S is P" we may form the acceptable composite modal
syllogism (with respect to the dictum) "It is necessary all M be P, and
it is necessary all M be S; therefore, it is necessary all S be P." Given
such an acceptable syllogism, by equipollence we can convert the
premiss or conclusion: since 'it is necessary that p' and 'it is impossible
that not-p' are equipollent, we may equally express the above
syllogism as (for example) "It is impossible some M be not P, and it is
necessary all M be S; therefore, it is impossible no S be P" (Theorem
IV-2 in TC 4.1.13-14). Obviously, the acceptability of the composite
modal syllogism depends on the mode in question; Buridan makes a
brief negative venture into epistemic-doxastic logic by pointing out
that no such syllogism is acceptable if the mode is 'known,' 'opined,'
'doubted,' or the like (Theorem IV-3 in TC 4.1.15-16).
8.6. DIVIDED MODAL SYLLOGISTIC
Divided modal syllogistic is far more interesting, and Buridan devotes

TC 4.3.1-74 to the subject. The first series of theorems, Theorem
IV-4 through Theorem IV-6, concentrate on syllogisms whose pre-
misses are de necessario or de possibili; the second series of theorems,
Theorem IV-7 through Theorem IV-9, concentrate on which syllo-
gisms are acceptable if the restrictive 'what is' locution is added to the
subject of one of the premisses (which effectively restricts the
referential domain of the subject to present or actual existent). The
remaining theorems, Theorem IV-lO through Theorem IV-20, are
mixing theorems, stating which mixtures of assertoric and modal
premisses produce valid syllogisms.
The pure divided modal syllogistic, as given in Theorems IV-4
through IV-6, is in many ways the most interesting; from the modern
point of view we may see Buridan as settling matters about the itera-
tion of modal operators. Take an example: the following divided
modal syllogism is acceptable (corresponding to (2) with a direct
conclusion):
All M is possibly P.
Some S is necessarily M
THEREFORE: Some S is possibly P.
8. THE SYLLOGISM 81
The way to validate the syllogistic conclusion is as follows. From the

premisses we can clearly infer that some S is necessarily something
which is possibly P; we may abbreviate this by writing "Some S is
necessarily possibly P." Here we have the modes iterated. Buridan's
claim about the syllogistic form, then, may be seen as a claim about
equivalences among such iterations; in the given example, Buridan is
holding that 'necesarily possibly' entails 'possibly' - which it certainly
does. Since some S is necessarily possibly P, then, that means that
some S is (simply) possibly P, which is the divided modal syllogism.
We may extend this point: if we take part of Buridan's divided
modal syllogistic as specifying how iterated modal terms are to be
treated - what single modal term corresponds to an iterated pair of
modal terms - then which divided modal syllogisms Buridan finds
acceptable, and what iteration-reductions they embrace, will tell us
(roughly) what system of modal logic Buridan is using,'!9 It seems to
be S5: 'necessarily necessarily' is replaced by 'necessarily'; 'possibly
possibly' is replaced by 'possibly'; 'necessarily possibly' is replaced by
'possibly'; and 'possibly necessarily' is replaced 100 by 'necessarily.' The
validity of this claim rests on our interpretation of certain modal
syllogisms Buridan accepts; since Buridan states only the general
theorem, it is not always obvious what form he is endorsing. I have
noted the cases requiring strong modal assumptions in the notes to the
translation with an asterisk.
This characterization, of course, simply takes S5 and other modal
systems to be specified syntactically. We may semantically characterize
S5 in the well-known way of taking the accessibility-relation among a
system of possible worlds to be an equivalence relation (reflexive,
symmetric, and transitive). The quantifiers in divided modal premisses,
then, would quantify across possible worlds, the modal terms state
which worlds are accessible, and Buridan's replacement of iterated
modalities reflects the fact that the accessibility-relation is an
equivalence relation. On this construal, it is easy to see the point of
introducing the 'what is' locution and the several mixing theorems:
these are devices which index some of the claims made to the actual
world.
Buridan simply specifies the modal character of the premisses in a
figured conjugated, and asserts that the conclusion will have a certain
modal character. In my notes to each theorem I have tried to list
the instances of the theorem. However, while Buridan's approach to
assertoric syllogistic is relatively unproblematic, there are certain

problems with his modal syllogistic. First, it is sometimes difficult to
give the modal character of the conclusion in exactly the general form
Buridan describes. In order for his claims about the modal character
of the conclusion to hold in each case, it is sometimes necessary to use
a tortuous combination of negations, where the conclusion is much
more readily expressed by using the equipollent modal term. Again,
sometimes even this device does not work; in the case of Theorem IV-
13(b) for mixed premisses in the third figure for (3) and (4), and also
in the case of Theorem IV-16 with the remark in TC 4.2.54 for mixed
premisses in the first figure for (5) and (6), the conclusions derivable
cannot be fit into Buridan's specifications.
But there is a more serious problem with Theorem IV-5, for modal
premisses in the second figure in (3) and (6): no conclusion at all
seems to be entailed by the premisses. The troublesome pair is as
follows: (i) "All P is possibly M, and no S is necessarily M"; (ii) "No P
is necessarily M, and some S is possibly not M." Try as I may I cannot
find any conclusion which follows acceptably from (i) or (ii). Perhaps
my interpretation of his strictures is incorrect; perhaps he has simply
made an error. If the latter, then the wonder is that there are not more
errors, given the abstract level at which theorems are stated and
proved. For all that, Buridan's modal syllogistic is an astonishingly
rigorous and precise construction, worthy of our admiration. It shows
the I11ediaevallogical mind at its best.
JEAN BURIDAN
SUMMULAE DE DIALECTICA
TRACTATUS IV
TREATISE ON SUPPOSITION
INTRODUCTION
[0.0.1]
Treatise IV will be about supposition and related matters.
And first, let us state the difference between supposition and
signification. I do not intend to speak here of the natural
signification of utterances (voces), I since we do not form sen-
tences from such utterances; I mean only to deal with utter-
ances which signify by convention (ad placitum).2 Nor do I
intend in this chapter to speak of material supposition, for
any inscribed utterance can supposit in this way, even if it is
not imposed to signify something. For now, I intend to speak
only of supposition in that an utterance mayor may not sup-
posit according to the signification imposed on it.
[0.0.2] Since there is no supposition of a sentence but rather of a term
which is either subject or predicate, Treatise IV, which is about supposi-
tion and related matters, should come right after the treatise on predic-
able terms. 3
[0.0.3] Treatise IV has six chapters: Chapter 1 is about the difference

between signification, supposition, verification, and appelation; Chapter
2 is about the several kinds of significative utterances (vocum
signijicativarum), in order to clarify which can supposit and which can-
not; Chapter 3 is about the general modes and divisions of supposition;
Chapter 4 deals in particular with the supposition of <what I shall call
there> 'relative terms'; Chapter 5 is about appellation; and Chapter 6 is
about ampliation.
85
CHAPTER 1
SIGNIFICATION, SUPPOSITION, VERIFICATION,

APPELLA TION
This first chapter has four sections: Section 1.1 sets forth the aim <of the
chapter>, Section l.2 gives the difference between signification and sup-
position, Section 1.3 the difference between supposition and verifica-
tion, Section 1.4 the difference between supposition and appellation.
SECTION 1.1. THE AIM OF THE CHAPTER
[1.l.11 The first part of [0.0.11 is clear enough: it puts forward in brief
the aim of Treatise IV: "suppoosition and related matters," that is, about
signification, verification, appellation, and ampliation or restriction. It
also immediately says what questions ought to be settled in Chapter 1
about the difference between signification and supposition, and about
which <kind> of signification is meant in Chapter 1, and similarly what
supposition is and what it is not. These matters are plain in the text of
[0.0.1].
[1.1.2] But it also be added that we do not deal with supposition as 'sup-
position' is understood by the grammarians," for then the term 'chimae-
ra' could supposit just as well as the terms 'man' and 'horse,' since the
expression (oralio) that a chimaera runs is as grammatical (congrua) as
"A man is reading," and so is not supposition as the logician takes it.
SECTION 1.2. SIGNIFICATION AND SUPPOSITION
[1.2.1]
Therefore, taking signification and supposition as spelled
out in [0.0.1], signification and supposition differ because any
given word (dictio) which, not taken materially,S is a part of
speech,6 signifies and gives rise to some concept to the one
hearing it, according to the institution conventionally given
to it. But not every such word has supposition. All and only
those terms supposit which, when something is pointed out
86
1. SIGNIFICATION, SUPPOSITION, VERIFICATION 87
(demonstrato) by the pronoun 'this' or several things by the

pronoun 'these,' can truly be affirmed of that pronoun. Thus
the term 'chimaera' cannot supposit since, no matter what is
pointed out, it is false to say "This is a chimaera." Similarly,
no matter which things are pointed out, it is false to say
"These are chimaeras."
[1.2.2] This passage makes two points, the first about signification, the
other about supposition.
[1.2.3] With respect to the first point of [1.2.2], it should be noted that
the word (sermo), or the power of speech, was ultimately given to us in
order that we be able to signify a concept to others, and hearing was
ultimately given to us in order that concepts of a speaker could be
signified to us. Aristotle says this in De anima [Ill.xiii 453b24-26]: "the
animal has hearing in order that something be signified to it; it moreover
has a tongue in order that it signify something to another." By 'the
tongue' Aristotle means here the power of speech, whatever that may
be.
[1.2.4] And so it is clear that a significant utterance should signify the

concept of the speaker to the hearer, and should give rise to a concept'
in the hearer similar to the concept of the speaker, unless prevented or
spoken deceptively, or perhaps only in the proper usage (which is the
very sound itself in the hearer who sings while alone), or for another
reason. Nevertheless, it is clear that all who speak and dispute intend
their words to be taken in the aforesaid two ways <namely, to give rise to
a concept in the hearer and have concepts from the speaker>.7
[1.2.5] It should also be mentioned that naturally significant utterances

are not used in sentences, since by themselves (without forming any
sentence) they are sufficient to produce <their> complete sense in the one
hearing them.
[1.2.6] Too, an utterance which neither signifies naturally nor by some

imposition conventionally given to it <is not used in a sentence>, unless
perhaps as it <signifies> itself or the concept of it. For example, 'bu ba' is
neither noun nor verb nor syncategoremata, and so has no place in a
sentence unless it is taken materially, as "I say 'bu: " Nevertheless, this is
88 TREATISE ON SUPPOSITION
not a sentence, since the utterance 'bu' has no signification, nor a man-
ner of signifying (modus significandi) which could provide a subject for
the verb. However, if it were taken materially, then the sentence holding
that the term 'bu' is running would be acceptable. K
[1.2.7] The second point of [1.2.2], which is about supposition,9 seems

clear to me, except that a student might doubt it for this reason: the term
'God' can supposit, but no matter what is pointed out by the pronoun
<'this; the term 'God'> would not be verified of the pronoun, since it
would require that God be pointed out and we cannot point out <God>.
[1.2.8] Some respond <to this doubt> that although God cannot be
pointed out to sense He nevertheless can be pointed out to the intellect.
[1.2.9] Alternatively, we can say that it is sufficient for a term to suppos-

it that it can be truly affirmed either of such a pronoun or of a relative-
term lO referring to some prior term. For example, we can say truly
"There is a first cause and it is God" or even "The omnipotent being is
not evil and He is God." Moreover, this <new criterion> does not hold
for 'chimaera,' since even if we were to say "A chimaera does not exist
and it is a chimaera" <the claim> "... and it is a chimaera" would be false,
for such a relative pronoun can equally be taken in the sentence for a de-
monstrative pronoun. 1 1
SECTION 1.3. SUPPOSITION AND VERIFICATION
[1.3.1]
Verification is different from supposition, for verification
is properly of a sentence and not of a term, but supposition is
of a term and not a sentence. 12 Again, it is possible that there
be supposition of the terms in a sentence without the verifi-
cation of the sentence, in negatives as well as in affirmatives,
as for example "A man is an ass" or "Man is not an animal."
Too, in negatives there can be verification without supposi-
tion, as in "A chimaera is not a goat-stag." But in true affirm-
atives there must be verification of the sentence by some
thing or things for which the term suppa sits. And if the term
is distributed, there must be verification by all those things
1. SIGNIFICATION, SUPPOSITION, VERIFICATION 89
for which it supposits if the sentence is to be true; if it is not

distributed, it is sufficient for some thing or things.
[1.3.2] In this third section there are four points.
[1.3.3] The first point is that verification is of a sentence and not of a

term, for being true or false is the proper characteristic of a sentence.
[1.3.4] Any sentence is true or false and anything true or false is a

sentence, as logicians use the terms 'true' and 'false.' So it is customarily
said, and it is true if we are speaking of mental sentences and of truth
and falsity said principally and not of truth or falsity by attribution to
another.
[1.3.5] Indeed, any mental sentence is true or false, and anytiung simply
true or false is a mental sentence, as the logician speaks of true and
false.
[1.3.6] If a spoken sentence is true or false, it is so only insofar as it

signifies a true mental sentence of the speaker or gives rise to a true
mental sentence in the mind of the hearer. And so a <spoken sentence> is
true attributively, just as urine is called healthy in that it signifies the
health of the animal, or food which brings about the health of an animal
<is called> healthy.
[1.3.7) But note as it seems to me that the spoken sentences which are
called true or false are significative only in that they represent true or
false mental <sentences>, and every mental sentence is true or false -
and so, therefore, any spoken sentence represents the true or the false.
Thus any spoken <sentence>, taken significatively or representatively, is
true or false.
[1.3.8) Nevertheless, I do not believe that any utterance which is signifi-

cantly true or false is a spoken sentence, since the impersonal verbs
'there-is-reading (legitur),' 'there-is-Ioving (amatur),' 'it-is-raining (p/uit),'
'it-is-thundering (tonal),' 'it-is-hailing (grandinat), seem to be counterex-
amples. The true or the false is represented by any of these,13 yet still
none of them is a sentence, for every sentence should have a subject and
predicate as its principal parts, which is not the case here. 14 And since
every sentence is an expression (oralio) - for a sentence is described as

an expression signifying the true or the false (dealing with a spoken
sentence) - and none of those <impersonal> verbs is an expression, since
they do not satisfy the definition of an expression (namely <that> whose
parts are significative separately), therefore those spoken <impersonal>
verbs are significative of the true or the false, because they represent
true or false mental <sentences>, and yet they are neither sentences nor
expressions, for they are not composed of several parts individually sig-
nificative.
[1.3.9] Further, it is necessary that a mental sentence correspond to it.

For those disputing can agree that they understand no more and no less
by 'B' then by "A man is running," inasmuch as signification is conven-
tional, and then if it is true to say "A man is running" then it is true to
say 'B.' Still, 'B' is neither an expression nor a sentence, since it does not
have parts which signify of themselves.
[1.3.10] Therefore, then, when it is said in the first point of [1.3.1] that
there is verification of a sentence and not of a term, this should be
understood as <verification of> a mental sentence, and we should under-
stand the foregoing for spoken sentences.
[1.3.11] There is no verification of a term, i.e. no term is true of false,

taking 'term' as that which taken significatively can be the subject or
predicate of a sentence. But if we take 'term' as a word equivalent to a
sentence, then such a term can be called true or false significatively, just
as a spoken sentence <can be called true or false>. Hence when it is in
the same clause in [1.3.19] that there is supposition of a term and not of
a sentence, this is true of 'sentence' taken significatively and of 'term' as
the subject or predicate of a sentence is called a term.
SECTION 1.4. SUPPOSITION AND APPELLATION
[1.4.1]
Appellation differs from supposition because there are
terms which supposit and do not appellate, for example
nominative terms in the category of Substance such as
'animal,' 'plant; 'gold'; there are terms which appellate but
do not supposit, for example 'chimaera; 'vacuum; and the
l. SIGNIFICATION, SUPPOSITION, VERIFICATION 91
whole 'man able-to-whinny'; and there are terms which

<both> supposit and appellate, such as 'white,' 'father,' 'sitting,'
or the phrase 'white man' (which appellates one thing and
supposits for another, as 'white' supposits for a thing which is
white and appellates whiteness). Indeed, any term supposit-
ing for something and signifying another thing for which it
does not supposit appellates everything which it signifies
other than what it supposits for as adjacent to what it sup-
posits for.I 5
[1.4.2] This fourth passage makes four points.
[1.4.3] The first point deals with terms which supposit and do not
appellate, which in the passage are nominative terms in the category of
Substance, and this is because they do not connote some disposition in
the substance which they signify and for which they supposit. This is
also so for many abstract terms in the category of Quality, for example
'whiteness,' 'hotness,' 'humanity.'
[1.4.4] But doubts may arise whether oblique terms in the category of
Substance supposit or appellate. I assert that, strictly speaking, they
never supposit, for (as I have assumed) strictly speaking no term sup-
posits unless it is the subject or predicate of a sentence. Whence it is
clear that it is not true to say of anything pointed out "This is of-a-stone"
or 'at-a-stone' except in this sense or a similar one, that this is of a stone
(i.e. this is part of a stone).16 But then 'of-a-stone' does not supposit, but
rather the whole 'part of a stone' supposits. Therefore an oblique <term>
cannot of itself be the subject or predicate of a sentence, whence it is
obvious that it cannot supposit.
[1.4.5] Nevertheless, I say that an oblique <term> can properly appellate,

for if I say "The ass of-Socrates is running" the phrase 'ass of-Socrates'
sup posits for the ass and not for anything but the ass, and nevertheless
the term 'of-Socrates' appellates Socrates as adjacent to the ass, i.e. as
possessing <the ass>, not as inhering <in it>.
[1.4.6] If you were to ask whether in saying "A rational animal is run-
ning" there is some appellation on the part of the subject here, I say that
there is not, assuming that 'animal' is truly a term in the category of
Substance and that 'rational' is truly the specific difference of animal,

connoting no accident. I 7 Here there is only a contraction of the supposi-
tion without the appelation of a distinct (alienae) disposition, as is also
the case in saying "The animal man runs" or "The animal which is a man
is running" - or even "The animal which is not a man is running." It
seems that to speak negatively in this way is nothing but to except or
remove man from the supposition of the term 'animal.' If anyone wants
to call this 'appellation' he is welcome, although the nominative substan-
tial term taken per se for substances more properly supposits and does
not appellate anything.
[1.4.7] The second point in [1.4.1] is about <terms> appellating and not
suppositing, as for example 'chimaera,' 'vacuum,' 'man able-to-whinny:
This is clear, because no matter what is pointed out it is not true to say
"This is a chimaera," "This is a vacuum," "This is a man able-to-whinny."
But in the sentence "A man able-to-whinny is running" the phrase 'able-
to-whinny' appellates the power or capacity of whinnying, not as in that
for which the whole subject supposits, but as in that for which some part
(namely 'man') would supposit if it were taken per se. Since what is
appellated cannot be adjacent to man, the term 'able-to-whinny' not
only contracts the supposition of 'man,' but completely removes it. As
for the term 'vacuum,' the answer is obvious if we take the nominal
definition for the term taken per se in the sentence, namely 'a place not
filled with body: The term 'place; if taken per se in the sentence, may
well supposit, but its supposition is removed by the appellation of the
tern,s added to it, <namely 'not filled with body'>. And the same seems to
be true of 'chimaera' if the nominal definition is given.
[1.4.8] The third point in [1.4.1] deals with terms which <both> supposit
and appellate. It seems that 'white' supposits for substance and appel-
lates whiteness insofar as it inheres in that substance.
[1.4.9] The fourth point in [1.4.1] is a rule which is clear enough in the
passage. But it should be added that when the subject of a sentence does
not supposit for anything, as with the words given in the second point,
there can still be appellation in the way discussed in [1.4.7].
CHAPTER 2
KINDS OF SIGNIFICATIVE WORDS
The second chapter is about the many divisions of significative words,

and has six sections, each dealing with one of the six divisions of signif-
icative words.
SECTION 2.1. COMPLEX AND IN COMPLEX
[2.1.1)
Now we should examine which words and expressions can
supposit and which cannot. And to see this we should put
forward some divisions of words significative by convention,
since we don't care about the rest. And the first division is
that some are complex and some are incomplex; the latter
are called words, the former expressions.
[2.1.2) This first passage makes two points, the first of which sets out
the aim of the chapter and is perfectly clear.
[2.1.3) The second <point> divides significative utterances into complex

and incomplex and is also perfectly clear, unless there is a doubt about
'it-is-raining,' 'it-is-thundering,' 'there-is-Ioving.' But insofar as 'there-is-
loving' is an impersonal verb, we have discussed this matter in [1.3.8).
SECTION 2.2. SUBJECT AND PREDICATE
[2.2.1)
The second division is that some incomplex utterances
taken per se l8 may be the predicate or the subject, for exam-
ple 'man,' 'stone,' 'plant,' 'gold,' 'white (album)'; others are
neither able to be the predicate nor the subject, such as
'whichever,' 'any,' 'nobody,' 'nothing,' 'none,' 'white (a/bus),'
'black (niger).' But none are able to be the subject unless they
can be the predicate.
93
[2.2.2] This division has three parts and the rejection of a fourth, and
the first and second parts <of the divisiofll are sufficiently clear.
[2.2.3] But the third part <of the divisiofll and the fourth clause are
doubtful - namely, why anything able to be the subject can be the pre-
dicate, but not conversely.
[2.2.4] This <doubt> can be answered by noting that in Chapter 2 we do

not intend to deal with words taken materially being the subject or the
predicate, but only words taken significatively.
[2.2.5] First of all,lY I say that according to grammarians non-substan-

tive adjectives of a neuter gender cannot provide a subject for a verb,
and hence cannot be subjects of sentences, for there is no grammatical
agreement. co But when the verb has a subject, adjectives can be added
to indicate what falls under the substantive term 2l or the subject and for
what it suppa sits; nevertheless, it is true that a substantive adjective of
neuter gender can be the subject <of a sentence>, since it can be resolved
into a subject and an adjective, as for example 'white' is <resolved into>
'white thing.'
[2.2.6] The terms 'whatever,' 'any,' 'nobody,' 'nothing,' and the like are
more difficult.c:
[2.2.7] Still, it should be said of 'nobody' and 'nothing' that a negation

following the copula of a sentence cannot be negating; hence it only
makes the term to which it is added into an infinite term, so that "Stone
is non-man" is affirmative with an infinite predicate. Moreover an infini-
tizing negation is a part of the subject or the predicate. But a negation
preceding the copula can be negating, if it is understood to fall under the
copula, and then it is taken to be part of the copula. Hence it is neither
part of the predicate nor part of the subject, but rather <part of> the
copula. Implicit negations, such as 'nothing: 'nobody,' 'never,' and so
forth are imposed for negating if they are found at the beginning of the
sentence. Hence if I say "Nobody is a stone," the sentence is negative
and the subject is the term 'man' implicit in 'nobody.'
2. KINDS OF SIGNIFICATIVE WORDS 95
[2.2.8] Affirmative signs, such as 'every' and 'whichever,' since they

cannot thus correctly be parts of the predicates as well as the subjects,
are more difficult.
[2.2.9] It can be said that a distributive sign taken with the predicate
does not act on the copula or the subject; hence it pertains wholly to the
predicate. But taken with the subject it acts on the copula and the pre-
dicate by confusing <them> (confundendo), though not distributively.23
Hence it is not taken with the subject or with the predicate, but is taken
as a certain condition of the whole sentence.
[2.2.10] But then there seems to be a greater difficulty with the particu-
lar sign: it does not act on the predicate or on the copula if it is taken
with the subject, according to which it seems as though it should be tak-
en as a part of the subject (if it is taken with the subject), just as it would
be a part of the predicate if it were taken with the predicate.
[2.2.11] Many answers can be given to this.
[2.2.12] The first <answer> is that a particular sign found in a sentence,

whether taken with the subject or with the predicate, is (if not pre-
vented) governed by the rule that any indefinite and particular are for-
mally equipollent, since absolutely nothing at all is changed in the above
by the addition or subtraction of the sign. Hence it should not be con-
sidered as part of the subject, nor as part of the predicate, nor as some
condition of the sentence, unless its addition is such that it does not act.
[2.2.13] I answer 24 by showing that the particular sign taken with the
predicate is not a part of the predicate, since these are formally equipol-
lent: "B is A" and "B is some A." Hence whatever contradicts one
contradicts the other; it happens that "No B is A" contradicts the first,
and so similarly contradicts the second. But it would not contradict <the
second> if 'some' were part of the predicate, for they would not have the
same subject and the same predicate, which is what is required for a true
formal contradiction. Hence it seems to me that such a word, whether
taken with the subject or the predicate, should not be called a part of the
subject or a part of the predicate, or if it is taken as part of the subject it
should thus be taken as part of the predicate, but only a part added to it
which has no effect, since, with it taken away, nothing is changed in the
above.
[2.2.14] But in another way the particular sign is taken in a sentence as

a sign able to determine an indefinite <sentence>, in that the universal
sign indicates that the predicate is verified of the subject for everything
it supposits for, and the particular sign indicates that it is verified of
some and not of all, or at least that the sentence is true for some and not
for all, or that it is known for some and not known for all. The indefinite
sentence is related to this indifferently, and so, with the particular sign
so taken, the readers can correctly give the difference between the par-
ticular and the indefinite sentence, and also more often understand the
universal and not the particular by the indefinite.
[2.2.15] In this way the universal and the particular sign do not fall
under the subject, but indicate the quantity of a true sentence, when
taken with the subject. 25
SECTION 2.3. CATEGOREMATIC AND SYNCATEGOREMATIC

TERMS
[2.3.1]
Some incomplex utterances are called 'categorematic,'
others 'syncategorematic,' others 'mediate' or 'mixed: Those
which signify nothing besides the concepts which they imme-
diately signify, except perhaps what the terms to which they
are adjoined signify are called 'purely syncategorematic,' as
for example the words 'not,' 'or,' 'therefore,' 'also,' 'because,'
'according-to-which,' and the like. Those which not only
signify the concepts they immediately signify but also the
things conceived by those concepts, and they may be subject
or predicate per se, and do not include any syncategoremata
are called 'purely categorematic,' as for example 'man,'
'stone,' 'whiteness,' 'white,' 'ass,' 'blackness,' and the like. But
those which either (i) signify, aside from the concepts they
immediately signify, the things conceived by the concepts,
but not able to be the subject or predicate per se, or (ii) imply
categorematic and syncategorematic terms, are called
'mediate' or 'mixed,' as for instance 'perhaps,' 'today,' 'some-
where,' 'nobody,' 'nothing,' 'with-itself,' and many others.
[2.3.2] This passage makes four claims.
[2.3.3] The first claim is a division with three parts, which is clarified by
the other points, and other three points declare the parts of the said
division.
[2.3.4] Note that a term is called categorematic either by predication or

signification. If it is so-called by predication, then any word which is
able to be subject and predicate per se and which includes no syncate-
goremata is called purely categorematic. There are many words able to
be a predicate which are not purely categorematic, since they imply
negations; for example, 'nobody,' 'nothing,' and privative terms.26
[2.3.5] But if a term is called categorematic by signification, then a cate-

gorematic term is significative (as it were) per se.
[2.3.6] Syncategorematic terms are then not significative per se, as it

were, but <only> significative with another. 27 This is why their prefix is
'syn-' i.e. 'with,' and so <'syn-categoremata' means> significative with
another (as it were).
[2.3.7] Nevertheless, it is necessary to know that every word which can

be put into a sentence is imposed per se for some signification, namely
to signify some concept of the mind, as was said in [2.1.1]. As Aristotle
says in De interpretatione [cap. i 16a2-4]: "those things which are
spoken are marks (notae) of those passions in the soul," i.e. significative
<of those passions in the soul>; whence there is no difference with
respect to this sort of signification but that every utterance found in a
sentence has a per se signification imposed on it by convention.
[2.3.8] Furthermore, some utterances imposed to signify per se signify

aside from those concepts the things which are conceived by those
concepts. For example, the term 'ass' <signifies> an ass, the term 'white'
whiteness, the terms 'yesterday' and 'today' some time. Such words
are called purely categorematic if they are able to be a predicate per
se and include nothing purely syncategorematic. They are not purely
categorematic if they are mediate or mixed or include something
syncategorematic.
[2.3.9] There are other significative utterances which, aside from those
concepts, do not signify per se the things which those concepts conceive,
but which are copulas and negations and distributive signs. 2x
[2.3.10] Hence these sentences "God is God" and "God is not God"
signify nothing more or nothing less external than is signified by the
term 'God: provided that this word <'is'> is taken precisely as the copula
(as will be discussed in [2.4.3]). Nor does "Every man is an animal"
signify anything more or less, aside from a concept of the soul, than "No
man is an animal."24
[2.3.11] Thus such a sign signifies only how the mental terms, and the
spoken terms corresponding to them, supposit, and nothing else by
their signification. 30
[2.3.12] The different 3l copulas 'is' and 'is not' <signify> different ways of
combining (complectendl) mental terms in forming mental sentences,
and those <ways> of combining are complexive concepts relevant to the
second operation of the intellect as it is added to the first operation. 32
[2.3.13] And so these words 'and,' 'or: 'if: 'therefore: and the like in-
dicate complexive concepts of many sentences together, or of terms
taken singly in the mind and not external <to it>, and such expressions
are called purely syncategorematic, since they are not significative of
anything external, unless <they are taken in combination> with other
<terms>, in the sense that the whole aggregate made of categorematic and
syncategorematic terms signifies things beyond the concepts. But <they
do so> by reason of the categorematic terms.
SECTION 2.4. COMPLEX AND INCOMPLEX CONCEPTS
[2.4.1]
Again some incomplex utterances correspond to complex
concepts and some correspond to incomplex concepts, and
those which correspond to complex concepts can and should
be analyzed with respect to the nominal definition by expres-
sions equivalent to them in signification. However, those
which correspond to incomplex concepts do not have precise
definitions explaining the nominal definition per se. 33
[2.4.2] Note that some words and expressions to which there corres-
pond incomplex concepts are categorematic, and others syncategore-
matic.
[2.4.3] I believe that the concepts which the words 'and,' 'or,' and 'if fall
under, although they are concepts complexive of many sentences or
words or terms, nevertheless are not complex or of several things but
are simple. So too is the concept of the verb 'is,' taken precisely as the
copula, and so it too is a purely syncategorematic word; nevertheless,
perhaps insofar as it connotes a certain time its simplicity is compro-
mised and is not purely syncategorematic, as other people say.
[2.4.4] It is true that there are some syncategorematic terms which

properly correspond to complex concepts, as for example 'only' or
'alone' and implicit syncategoremata such as "there begins ..." and so
forth. Hence sentences in which they are found need to be analyzed by
many sentences, as for instance "only a man is running" is analyzed as
"A man is running and nothing other than a man is running." Hence it is
clear that the word 'only' implies in the concept corresponding to it an
affirmative copula <and a negative copula> and a negative distribution.
[2.4.5] Some categorematic terms have simple corresponding concepts,

as for example 'man,' 'whiteness,' and 'white';34 others <have> complex
concepts <corresponding to them>, as for example 'chimaera' and
'vacuum'; the term 'vacuum' signifies the same, no more and no less,
than the expression 'place not filled with body.' Rather, since we impose
names to signify by convention, we can impose an equivalent word on
any given expression so that we understand the same by the word 'A' as
by the sentence "A man is running." Then it is clear that a complex
concept corresponds to the word 'A.'
SECTION 2.5. PERFECT AND IMPERFECT EXPRESSIONS
[2.5.1]
A complex utterance is called an 'expression'; some are
perfect and others are imperfect. Some perfect <expressions>
are sentences, as "A man is running"; others are not sen-
tences, as for example "Peter, read!" or "Slave, make a fire!"
as was said in [2.1.1).35 Moreover, (1) some imperfect

complex <expressions> are made up of a verb and a deter-
mination of the verb without a nominative term supplying a
subject for the verb, as e.g. "run swiftly" or "see a man"; (2)
others <are made up of> a nominative categorematic substan-
tive term and another <term> added to it as a determination,
as 'white man' or 'Robert's ass' or 'the horse which is walking';
(3) still others <are made up of> many substantives conjoined
by some conjunction, as 'man and horse' or 'horse or ass'; (4)
others from only a categorematic and a syncategorematic
term, such as 'non-man' 'every man' 'only Socrates.'
[2.5.2) All of the claims made in this passage are obvious in themselves
and give all the divisions of significative words needed for discussing
which of them can sup posit and which not.
[2.5.3) Since no utterance, strictly, can sup posit unless it can be subject
or predicate per se, we must therefore see which of the expressions in
[2.5.1) can be the subject or predicate and which cannot.
[2.5.4] Although in [2.5.1) I have only given four kinds of imperfect

expressions, I do not mean by this to rule out other kinds of imperfect
expressions. I have only given these four kinds so they may be consid-
ered at length; after considering them, it will easier to consider others.
SECTION 2.0. RULES FOR SUPPOSITION
[2.6.1)
We should now distinguish which utterances can be sub-
ject and predicate and which cannot, and also which can
supposit and which cannot. To do so, we give some rules. 36
[Rule Sup-I) As mentioned in [2.2.1), every term able

to supposit can be the subject or predicate of a categorical
sentence, but not conversely.
[Rule Sup-2) No purely syncategorematic word can be
a subject or a predicate per se.
[Rule Sup-3] No oblique term can be a subject or a

predicate per se.
[Rule Sup-4] No verb can be a subject or predicate per
se unless it is an infinitive, which can be a subject or pre-
dicate, as for example in "To cut is to act."
[Rule Sup-5] Any substantive nominative term, or a
nominative adjective of neuter gender, can be a subject or
predicate.
[Rule Sup-6] No perfect expression can be a subject or
predicate.
[Rule Sup-7] No expression imperfect like (1) <in
[2.5.1]> can be a subject or a predicate unless it is a verb in
the infinitive.
[Rule Sup-8] Any expression imperfect like (2) <Ill
[2.5.1]> can be a subject or a predicate.
[Rule Sup-9] An expression imperfect like (3) <Ill
[2.5.1]> can be subject or a predicate, though perhaps not
every such <expression>.
[Rule Sup-lO] An expression imperfect like (4) <in
[2.5.1]> can be a predicate, but not every such expression
can be a subject, unless the syncategorematic term is an
infinitizing negation, not a negating negation.
[Rule Sup-II] No expression imperfect like (2) <in
[2.5.1]> can supposit if what is given as the determination
is incompatible with the substantive term, but if it is not
incompatible then the expression may supposit.
[Rule Sup-12] Every word to which there corresponds
a complex concept should be treated just as the expres-
sion equivalent to it, namely the expression which states
the nominal definition of that word.
[Rule Sup-I3] Every nominative substantive term
which is purely categorematic to which there corresponds
a simple or incomplex concept is able to supposit.
[Rule Sllp-14] Every infinite name which can be a sub-

ject or predicate can supposit.
Many other rules can be given, but these are sufficient, for they seem the
most important ones.
[2.6.2] This long passage obviously makes fifteen points. The first
merely gives the aim of Section 2.6, and the other fourteen are four-
teen rules, some of which are obvious and others of which are rather
dubious.
[2.6.3] Rule Sup-l IS obvious enough from the discussion in Section

2.2.
[2.6.4] Rule Sup-2 is generally conceded. Thus in saying "Everything is

something" or "Nothing is a chimaera," do not think that 'everything'
and 'nothing' are purely syncategorematic; rather, they are substan-
tives. Hence a purely categorematic word is implied in a syncate-
gorematic distributive <term>, which is the 'subject of the sentence.
"Everything is something" must be resolved into "Every thing is
something" and, similarly, "Nothing is a chimaera" as "No thing is a
chimaera."
[2.6.5] Rule Sup-3 is very questionable,37 for most people hold that an
oblique term can be the subject of a sentence, namely when it is found
at the beginning of the sentence, either before the copula or before
some nominative term. For example, in the sentence "Of-a-man an ass
is running," they say that this word 'of-a-man' is the whole subject and
they take the rest as part of the predicate, or as the predicate and the
copula. Similarly, in "Of-a-man there is an ass" they take 'of-a-man' to
be the subject and the rest to be the predicate.
[2.6.6] But they all rightly concede that if a nominative term precedes
an oblique term, then the the oblique term is not the subject, but
rather the preceding nominative term is the subject of the sentence.
[2.6.7] Those who say this have many arguments, which I shall briefly
recapitulate now.
[2.6.8] The first <argument> is that this <sentence> is universal: "Of-

every-man the ass is running"; therefore, 'man' is the subject here and
not 'ass,' nor even the whole 'ass of a man.' The consequence is clear,
because the subject of a universal sentence must be distributed, and
only 'man' is distributed here. But it follows firstly because its contra-
dictory is particular, namely "Of-some-man the ass is not running,"
and a particular only contradicts a universal. Secondly, because univer-
sal affirmatives and universal negatives are contraries, and so "Of-
every-man the ass is running" and "Of-no-man the ass is running are
contraries, for they satisfy the Law of Contraries - they can both false
together but they cannot both be true together.
[2.6.9) <The second argument>: I assume,! from Aristotle's remarks in

the Posterior Anaiytics, that this is an acceptable syllogism with ob-
lique terms: "Of-every-man the ass is running, and Socrates is a man;
therefore, the ass of-Socrates is running." I also assume that this syl-
logism is in the first figure, since Aristotle does not want it to be in
any of the other figures.
[2.6.10) From the assumptions given in [2.6.9) it now follows that "Of-
every-man the ass is running" is universal, as said in [2.6.8), because, if
the major were particular or indefinite, the syllogism in the first figure
would not be valid.
[2.6.11) It secondly follows <from the assumptions of [2.6.9» that 'man'

is the subject, since it is the middle in the given syllogism, and in the first
figure only the middle should be the subject of the major and the predic-
ate of the minor.
[2.6.12) <The third argument>: let us syllogize as follows: 38 "Every

horse a man is seeing, and the horse of-the-king is a horse; therefore,
the horse of-the-king a man is seeing." I then ask what the subject
of the major is, whether it is the word 'horse' (in which case we have
what was to be proved) or the word 'man' or the aggregate of the
two.
[2.6.13) I now show that <the subject of the major> is neither 'man' nor
the aggregate of <'man' and 'horse'>.
[2.6.14) I claim that 'man' is not the subject because, as was said, the
major would not then be universal but rather indefinite, and the minor
would also be indefinite. And there is no valid syllogism with two in-
definite <premisses>.
[2.6.15] Again, if 'man' were the subject here then the term 'horse'
must be the predicate, which seems impossible, because neither the
predicate nor some part of the predicate ought to precede the copula.
Rather, the subject precedes the whole predicate.
[2.6.16] Again, if 'man' were the subject, then "The horse of-a-man
every man is seeing" and 'The horse of-a-man no man is seeing"
would be universals, and so would be contraries, which is not true,
since in a possible case they can both be true.
[2.6.17] Next I show that the whole 'horse a man' <in "Every horse a
man is seeing"> should not be called the subject, for those two terms
<'horse' and 'man'> have no connection in the conclusion, in which they
should be one term or one extreme of the sentence. And this is con-
firmed: in any syllogism there should be only three terms, namely the
middle and the two extremes. Nevertheless, in the aforesaid syllogism
there are exceedingly many terms, for there would be two terms in the
major, namely 'horse a man' and 'seeing'; in the minor there would be
two other terms, one 'the horse of-the-king' and the other 'horse'; and
there would be five terms added in the conclusion, namely the whole
'horse of-the-king a man; and these claims are ridiculous.
[2.6.18] <The fourth argument>: consider the case of equivalent sen-

tences, one of which is the analysis of the other: the subject in them
should be the same or equivalent, and either the same predicate or its
equivalent; but in such sentences where an oblique term is found first
are analyzed by equivalence into sentences where the oblique term is
obviously changed into a nominative term and put as the subject. For
example, "Of-a-man the ass is running" <is analyzed as> "There is a man
whose ass is running," and similarly "A horse a man is seeing" <is
analyzed by> "There is a horse which a man sees," and so forth for the
rest.
[2.6.19] Still, the arguments of [2.6.8J-[2.6.18] notwithstanding, I

believe the opposite, as I prove in Question 6 of <my work on> the first
part of the De interpretatione. 39 I now repeat some of the arguments

which I gave there.
[2.6.20] One argument is by the difference of the subject and the

predicate: the subject is that of which something is said and the pre-
dicate is that which is said of another. But by saying "Of-a-man the ass is
running" it is clearly apparent that the term 'running' is not only said of
'man' alone (by means of the verb 'is'), but of the whole 'of-a-man the
ass' - although more principally of 'ass' rather than of 'man,' according
to the grammarian (whom no science can contradict, since to be true is
<to be> compatible with all other truth, and every science is true). There-
fore, according to the <given> description of subject and predicate it is
more fitting to say that 'running' is the predicate and 'of-a-man the ass' is
the subject, and that 'ass' is the more principal part of the subject, than
to say otherwise.
[2.6.21] Again, if I say "Brunellus is the ass of-the-king" everyone

concedes that the whole 'ass of-the-king' is the predicate. And it is not
apparent to me that the predicate of this sentence can become the sub-
ject and the subject the predicate by conversion unless there is some
incongruity (according to the grammarian) and there is some falsity
(according to the logician).
[2.6.22] Again, in "The ass of-a-man is running" everyone concedes that

the whole 'ass of-a-man' is here the subject, and according to grammati-
cal agreement and logical truth or falsity, "Of-a-man the ass is running"
and "The ass of-a-man is running" are completely equivalent. Thus it
should be admitted that there is no difference between them, but that
the subject is the same.
[2.6.23] Again, <everyone> concedes that in "Of-a-man every ass is run-

ning" the whole 'ass of-a-man' is the subject. If you take away the syncate-
gorematic term 'every,' which is neither the subject nor the predicate nor
part of the subject or predicate, then why do we not have left the same
subject as before? No argument, but mere fantasy, can explain this.
[2.6.24] These arguments appear to me to be probable and more per-

suasive than the arguments for the opposite <conclusion>, but it is more
difficult to destroy their foundations - namely, that such sentences must
be resolved as described in [2.6.12]-[2.6.18]. For if this is taken away

then they cannot explain how the copula distinguishes between the
subject and the predicate, nor how such sentences are converted, and so
on for many others, which ought to be preserved by a resolution of this
sort.40
[2.6.25] Hence we argue against a resolution of this sort firstly that the
sentence "Of-Socrates the ass is" seems to use 'is' existentially (secunda
adiacens), but the resolutions use 'is' predicatively (tertia adiacens).
[2.6.26] Again, "Of-Aristotle the horse is dead" is true and its resolution
is false.
[2.6.27] Again, "Of-the-king the ass is running" is more clear with

respect to truth and falsity, and with respect to grammatical construc-
tion, than "There is a king whose ass is running" is. And we should not
say that the more clear needs to be resolved into the more obscure.
[2.6.28] Again, we have a complicated argument that if we must resolve

in the <proposed> way then it would follow that in the sentence "Of-a-
man the ass runs" there is no copula<tion> of the predicate to the sub-
ject, neither explicit nor implicitly sufficient, which is ridiculous: the
copula should formally be the <verb> of a sentence. The consequence is
clear, since no copula is implicit in a name, for the copula should be a
verb and signify with time, at least in actual supposition (as will be clear
later), and no name signifies with time. Nor is it implicit in the verb
'runs,' since although the verb 'runs' implies 'is running,' nevertheless
this whole would be part of the predicate, and another verb 'is' would be
necessary to be the copula for the whole sentence - and 'runs' does not
imply two 'is': therefore a sentence such as "Of-a-man the ass runs"
woulld not be well-formed or perfect, which is false.
[2.6.29] Again, it is necessary to resolve the rest <of the sentence>, since,
by whatever argument "Of-a-man the ass runs" must be resolved so too
by the same argument "There is a man whose ass runs" must be resolved
with respect to the part 'whose ass runs; and so on to infinity.
[2.6.30] Again, even if such a resolution could be maintained for af-

firmatives, still it cannot <be maintained> for negatives, since "Brunellus
Socrates does not see" is true,41 and positing that Brunellus and
Socrates are both dead, the resolution is nevertheless false, namely
"Brunellus is that which Socrates does not see." Similarly, "A chimaera
Socrates does not see" is true, but the resolution is false, i.e. "A
chimaera is what Socrates does not see."
[2.6.31] But some people cavil by saying that a negative <sentence>

should not be analyzed in this way, but as follows: "Brunellus Socrates
does not see" <is analyzed as> "Brunellus is not what Socrates sees."
[2.6.32] Against this reply: I put forth the sentence "Brunellus a man
does not see," positing that Socrates does not see him and Plato does
not see him, and then the given sentence is true and its analysis is false,
namely "Brunellus is not what a man sees." The contradictory is true,
and hence this seems to me to be a good and difficult argument.
[2.6.33] Therefore, it appears to me to be better to say that in a sen-

tence such as "Of-a-man the ass runs" the whole 'of-a-man the ass' is the
subject, just as it is the predicate in the converted sentence, i.e. "The
one running is the ass of-a-man." Thus 'of-a-man' acts as a kind of deter-
mination of the subject 'ass,' and there is no difference if <'of-a-man'> is
put before or after <'ass'>, with respect to grammar or to truth and falsity,
just as 'white man' and 'man <who is> white' do not differ. The sentence
still signifies the same with names and verbs transposed, if the syncate-
gorematic terms do not prevent it, as Aristotle says in De interpretatione
[cap.x 20bll-12].
[2.6.34] Hence it seems reasonable to me that in the sentence "Of-a-

man the ass runs" we should separate 'of-a-man' and 'ass' more than in
<the sentence> "The ass of-a-man runs."
[2.6.35] And it is easy to reply to the arguments for the opposed view.
[2.6.36] To the first argument in [2.6.8]: I say that the sentence "Of-
every-man the ass is running" is neither universal nor indefinite, but it
is indefinite for the nominative and universal for the oblique, just as was
said <in Summulae de dialectica Tractatus v> of modals that sometimes
they are universal de dicto as well as de modo and sometimes they are
universal de dicto and particular de modo and sometimes the converse.42
The quantity of the sentence should be denominated more from the sub-
stantive nominative <term> than from the determination: thus I may say
that "Of-every-man the ass is running" ought to be called indefinite,
though with a distributed oblique term; this sentence, though, is simply
universal: "Every ass of-a-man is running."
[2.6.371 But if you were to argue that the other <sentence> is universal
because its contradictory is particular, I deny this, for the contradictory
was not taken correctly. Rather, both can in some case be true, namely if
any man has two asses and one of them is running and the other is not.
The contrary is rather "Of-some-man no ass is running," and this sen-
tence should be called universal, although with an indefinite oblique
term.
[2.6.381 But when it is further said that "Of-every-man the ass is run-
ning" and "Of-some-man the ass is not running" are contradictories, I
deny this, for <these are the correct contradictories: "Of-every-man the
ass is running" and> "Of-some-man no ass is running."
[2.6.39J But when it is then argued that "Of-every-man the ass is

running" and "Of-no-man the ass is running" are contraries, I state that
they are not perfectly contrary, but "Of-every-man the ass is running"
and "Of-no-man no ass is running" are perfectly contrary (although the
previous pair are close to being contraries), in that in the negative the
nominative and the oblique terms are each distributed, and so they obey
the Law of Contraries, just as was said of some modals in the chapter on
modals [Summulae de dialectica Tractatus YJ.
[2.6.40] To the second argument in [2.6.9J: I concede that it is an

acceptable syllogism with oblique terms, since if we descend under
either <of the oblique terms> the conclusion may be inferred, whether the
term is the subject or the predicate, or part of the subject or part of the
predicate. The syllogism "Socrates is seeing Brunellus" is acceptable.
[2.6.41J Such syllogisms need not be strictly in any of the three figures
<of the syllogism> which Aristotle distinguishes in the Prior Analytics,
for there he only distinguished the figures and moods of syllogisms in
which the middle is the whole subject or predicate in each premiss. 43
[2.6.42] Hence 1 deny that the given syllogism is strictly in one of the
three figures. We say that it is in the first, since it has a greater similarity
to the <first figure> than to the other <figures>. Hence it may be correctly
concluded that the major is universal. Nor is it incorrect to syllogize in
this way from indefinites, that is, by taking something falling under a
distributed term, as for example: "Seeing every horse is a man, a beast
of-the-king is a horse; therefore, seeing a beast of-the-king is a man."
[2.6.43] Furthermore, about the other syllogism in [2.6.12] (namely

"Every horse a man is seeing, and the horse of-the-king is a horse; there-
fore, a horse of-the-king a man is seeing"), 1 state that it is an acceptable
syllogism. If you ask what is the subject of the major, 1 say that the word
'man' is there the subject and the predicate is 'seeing a horse,' since a
determination should be united with its determinable with which it is
construed, and the word 'horse' is construed with the word 'seeing' as a
determination with its determinable; hence 'seeing' is part of the pre-
dicate in the major, and 'horse' is also part of the predicate. 1 note that
according to grammarians it is sometimes permitted for the predicate to
be placed before the subject, and hence it is not surprising if sometimes
part of the predicate is put before the subject, since the word is put as a
determination and hence is appropriately construed with the substantive
or the subject. It follows, if it is possible, that it is construed with the
predicate <in the converted sentence>.
[2.6.44] 1 give an example which grammatically requires some agree-

ment: if 1 say "A white <man> am I" the subject of the sentence is not the
term 'white <man>,' for then the sentence is ungrammatical, but the term
'I.' But if you were to say "A white <thing> is I," then 'white <thing>' would
be the subject and not 'I,' since the sentence would be ungrammatical if
'I' were understood as the subject. Thus we may correlatively conclude
that I believe this is true, and an acceptable conversion: "A white <man>
am I; therefore, a white <thing> is I." This truly makes a subject of the
predicate and a predicate of the subject.
[2.6.45] From these remarks it is clear how all counterexamples are to

be resolved, except for that which says that these are universals: "A
horse of-a-man every man sees" and "A horse of-a-man no man sees."
They would consequently be contraries, which is false, since they can
both be true in a possible case.
[2.6.46J To the counterexample in [2.6.45J I reply that they are not con-
traries, since contraries require that anything pertaining to the subject is
distributed in the negative, so that it is incompatible with the affirmative.
The reason for this is that whatever pertains to the subject is distributed
in each, and whatever pertains to the subject and also the predicate is
distributed in the negative, and it is not so in the proposed sentence.
[2.6.4 7J To the last argument in [2.6.18J: the minor ought to be denied.

Neither the analysis nor the equipollence is universally appropriate, as is
claimed, although sometimes there is an equipollence. But again, it is
not always true with respect to the major that there is the same subject
in the truth or falsity of all equipollents, since "A certain man runs"
and "A certain runner is a man" are equipollent, as are also "Of-a-man
the ass runs" and 'The ass of-a-man runs." Nevertheless, in these there
need not be the same subject, even if (as I think) they are false; but still
it does not follow from this that they support Rule Sup-3.
[2.6.48J Thus, these things are probable for Rule Sup-3, even if they do
not appear to be demonstrated to everyone.
[2.6.49] Rule Sup-4 is not open to doubt; all grammarians say that "He
loves is a man" is ill-formed unless 'he loves' is taken materially, as for
example when we say "He loves is a verb," "He loves (amat) is a two-
syllable word," "He loves is a significative utterance," and the like. But
Aristotle forms sentences with infinitives, as given in the example <for
Rule Sup-4>.
[2.6.50J Rule Sup-5 has no counterexample for the first part. But for the
second part, the non-substantive adjective cannot be the subject of a
sentence per se (namely without a substantive), since it would be ill-
formed unless understood with a substantive - it is doubtful whether it
can be put as a predicate unless a substantive is understood, since
conversion should make the predicate a subject and the subject the
predicate, and an adjective cannot become a subject per se unless it is
substantive. On this matter we go back to the grammarians, if the truth
cannot be determined in another way.
[2.6.51 J Rule Sup-6 is also conceded, since the phrase would be ill-
formed to any grammarian.
[2.6.52] Rule Sup-7 is true in the same way as Rule Sup-4 and Rule
Sup-6.
[2.6.53] Rule Sup-8 is similarly found never to have a counterexample.
[2.6.54] There is a problem with Rule SUp-9,44 just as with Rule Sup-3;
many people hold that in such conjunctive or disjunctive categorical
sentences that the whole conjunction or disjunction should not be the
subject, but only the first part. For example, if I say "A man and an ass
are running," <they hold that> 'man' should be taken as the subject and
the rest taken as the predicate. Thus the analysis of "A man and an ass
are running" is "A man exists who, with an ass, is running" and similarly
<the analysis of> "Every man or ass is risible" is "Every man exists who,
or an ass, is risible."
[2.6.551 They hold that view because it seems to them that otherwise a
quantity cannot correctly be given to the sentences "Every man or ass is
risible," "Every man or ass run," "Socrates and an ass run," "Brunellus
and Socrates run." The first of these cannot be called singular, for the
subject is predicable of many (namely of Socrates and also of Plato).
Similarly, the quantity of "Every man or ass runs" cannot be given, since
it appears to be universal, but it cannot be universal since one cannot
give a singular which follows from it. (Although "<Every man or ass
runs>; therefore, Socrates or Brunellus runs" follows the latter is argu-
ably not singular.)
[2.6.56] Every indefinite ought to be assigned to be either universal or

singular, but then they ask whether "A man or an ass is running" or even
"A and Bare C" are universal or singular.
[2.6.571 Again, the syllogism in the first figure "Every man or ass is
running, and Socrates is a man; therefore, Socrates or an ass is running"
should like any syllogism of the first figure have the middle be the
subject in the major and be the predicate in the minor, and this is the
term 'man.'
[2.6.58] However, these things notwithstanding, I believe the contrary,

and I can adduce many arguments for it similar to those given for Rule
Sup-3.
[2.6.59) But for now I argue that if I say "A chimaera or an ass is an
animal" the sentence is well-formed and the whole disjunction 'a
chimaera or an ass' is the subject of the verb 'is' so that the verb with the
following predicate is true for one of the disjuncts by the natural con-
dition of disjunction. The reason is that the sentence is true and the
<proposed> resolution would be false, namely "A chimaera is what, or an
ass, is an animal."
[2.6.60) Again, when we dealt with questions about hypotheticals, it was

said that if these 45 are hypothetical: "Socrates is where Plato is," "Soc-
rates lectures when Plato listens," then by a similar argument these are
hypothetical: "Socrates is how Plato is," "Socrates is as much as Plato
is," and "Socrates is what Plato is." Therefore, "There is a man who, or
an ass, is running" is similarly hypothetical, the first categorical of which
is ''There is a man" and the second categorical is "who, or an ass, is
running," in which the whole disjunction, 'who or an ass' is the subject.
Unless you allow an infinite regress, we have what was to be proved.
[2.6.61) Again, it is obvious that "The matter and the form are the
composite" is simpler and more clear than "There is matter which, with
the form, is the composite," and so the latter should be resolved into the
former <and not the other was around>.
[2.6.62) Everybody permits a disjunctive predicate, for example

"Socrates is a man or an ass," and it is unreasonable that such a pre-
dicate could not become a subject by conversion.
[2.6.63) Hence it does not seem to me that Rule Sup-9 should be

denied on account of the difficulties <raised in [2.6.55)-[2.6.57J>.
[2.6.64) It is true that a conjunctive subject is typically taken in <one of>

two ways. In one way it is taken collectively, as in "The matter and the
form are the composite," or "Brunelluls and Favellus weigh ten pounds"
(positing that one of them weighs six and the other four), and then
such a sentence should not be resolved into a conjunction, i.e. as "The
matter is the composite and the form is the composite" or "Brunell us
weighs ten pounds and Favellus weighs ten pounds."
[2.6.65) Thus if you ask what the quantity of "The matter and the form
are the composite" is, I say that it is indefinite, and its contradictory is
"No matter and form are the composite." If you ask what the singular
<sentences corresponding to such a sentence> are, I say that the affirma-
tive singulars are "This matter and this form are the composite," and so
on for the rest, <in each case> pointing out a single matter and single
form.
[2.6.66J But if you ask what the quantity of "Brunellus and Favellus
weigh ten pounds" is, I say that it is singular, and its contrary is "Brunel-
luIs and Favellus do not weigh ten pounds."
[2.6.67J If you ask what the universal of such a singular <sentence as

given in [2.6.66J> is, I say that one must take a common term which will
be the conjunctive <subject> 'ass and ass,' and then make a distribution
like so: "Every ass and an ass weigh ten pounds .. "
[2.6.68J But in the other way <from [2.6.64J>, a conjunctive subject is

taken dividedly, as 'dividedly' is opposed to 'collectively.' Then the sen-
tence is resolvable into a conjunctive <sentence>, as "Socrates and Plato
run" <is resolved into> "Socrates runs and Plato runs." The contraries
should be given by preposing a negation to the whole sentence, and by
the negative of the disjunctive subject, i.e. "Socrates or Plato do not run"
or even according to the resolution "Socrates does not run or Plato does
not run."
[2.6.69J Similarly, if I say "Every man and ass is running," then it

follows by the resolultion into a conjunctive sentence that every man is
running and an ass is running. If you ask for the quantity of that sent-
ence, then I say that it is neither purely universal nor purely indefinite,
but universal with respect to the first part of the subject and indefinite
with respect to the second part of the subject. Its contradictory <is given>
either by preposing a negation, or something similar, as said in [2.6.68J.
[2.6.70J But then something should also be said about the disjunctive
subject, as if I say for example "Socrates or Plato run" or "Brunellus or
Favellus weigh ten pounds."
[2.6.71J First, note that the collective sense does not have a place here,
since the names 'collective' and 'disjunctive' are in a certain way incom-
patible.
[2.6.72] Further, note that when a distributive sign is put at the be-
ginning a sentence may be distinguished with respect to composition
and division.
[2.6.73] For example, if I say "Every man or ass is running" then the
subject can be understood to distribute only the first term, and then the
sentence is not wholly universal but partially universal and partially
indefinite and is the same as "Every man is running or an ass is runn-
ing." It is contradicted by "Neither every man nor an ass is running," or
by a negation preceding the whole sentence and falling under the copula
or the whole sentence, since it is a general rule, or by a conjunctive
sentence contradicting the equivalent disjunctive sentence.
[2.6.74] In the other way, the sign can be understood to distribute the
whole subject, and then the sentence is purely and simply universal, and
is the same as "Everything which is a man or an ass is running." It is then
contradicted in that a man or an ass is not running, and the singulars
which lead up to such a universal are "Socrates is running," "Brunellus is
running," and so on for all the other men and asses. This is not difficult
to see, for even old women speak in this fashion.
[2.6.75] But then there is a doubt whether "Socrates or Brunellus is

running" is singular. I say that it is not, as was correctly argued in
[2.6.55], but rather has more of the character of an indefinite. Indefinites
indeed have disjunctive force: "A man runs" is the same as "Socrates
runs," or "Plato runs," and so on; hence the singulars of that indefinite
sentence would be "Socrates runs," "Brunell us runs," "Plato runs,"
"Favell us runs," and so on for the rest of the men and the asses.
[2.6.76] Similarly, I say that "Every man or ass is running" is not strictly
or simply universal, since a universal should have a conjunctive charac-
ter; "Every man or ass is running" is the same as "Socrates is running,
Plato is running," and so on for all the men; hence the universal of the
indefinite "A man or an ass is running" is simply "Every man or every
ass is running" or "Every man or every ass are running."
[2.6.77] Note also that in a sentence with a conjunctive subject, if the

verb is singular, as in "Every man and ass is running," then for the
sentence to be well-formed the addition 'and ass' must be analyzed as a

determination of the preceding name, understanding the sense as
'everything which is a man and an ass' or understanding the sense as
'every man who is an ass.' Therefore, strictly speaking the sentence does
not have a conjunctive subject, for a conjunctive subject is equivalent to
a numerically plural subject, and so as the subject of the verb should
have the verb be plural.
[2.6.78] And these are the principal points which some people never-
theless manage to doubt, due to their poor instruction in grammar or in
logic.
[2.6.79] It should also be noted that someone might ask whether the
composite or the divided sense is more in line with the properties of the
terms <involved> in "Every man or ass is running," namely whether only
the term 'man' is distributed or that the whole subject is distributed. I
say that the answer should be different for the different manners of
speaking and writing: if immediately after 'man' there appears a sign of
division, namely a pause or a period, then the sentence will be divided
and only 'man' will be distributed, and if not, then it will be called
composite, and the whole subject will be distributed.
[2.6.80] These matters are laid down for Rule Sup-9.
[2.6.81] What is added <to Rule Sup-9>, namely 'not every such sent-
ence; was added for conjunction or disjunction from or with oblique
terms, as in "Socrates's and Plato's" or "Socrates's or Plato's," since such
a complex cannot be subject or predicate, as neither can an oblique
nameperse.
[2.6.82] Rule Sup-lOis obvious, since if you say "Every man is running"
or "No man is running," or "Not: a man runs (Non homo currit)" (so that
the negation is negating), we say that only the term 'man' is the subject.
[2.6.83] If however I say "Non-man runs (Non homo currit)" (so that
the negation is infinitizing), then we say that the whole 'non-man' is the
subject 46 •
[2.6.84] But if we were to say "A stone is no man" or "Socrates is every

man" we say that the whole 'no man' or 'every man' is the predicate, as
was said before.
[2.6.85) Rule Sup-II is clear enough, since if I say "A man able-to-
whinny is running" the word 'able-to-whinny' prevents the supposition
of men and the word 'man' prevents the supposition of things that can
whinny, and so the whole does not supposit for anything. The same
would be the case if we were to say "A non-risible man is running" or
"A man who is a horse is running" or "A man who is not an animal is
running."
[2.6.86) Rule Sup-12 is known per se, since the word 'vacuum' does not
sup positfor anything, just as neither does the expression 'place not filled
with body.' Nor does any word which is imposed to be equivalent to
such a sentence supposit, or is subject or predicate.
[2.6.87) Rule Sup-13 is also clear, since there is no cause in the subject
preventing the supposition, nor is any implied in some determination in-
consistent with its determinable.
[2.6.88) Rule Sup-14, the last rule, seems true to me, because among
infinite terms it seems that the term 'non-being' can not supposit most of
all, which is false since, although it does not sup posit for anything,
nevertheless supposits for many things collectively or conjunctively. For
we say truly that an army is not a being <since an army is composed of
many beings> and Socrates and Plato are not a being, for they are, and
they are not a being47. Therefore, if there were a counterexample to
Rule Sup-14 with such a term 'non-being which is not beings,' for indeed
Socrates and Plato are called a non-being, and nevertheless they are a
non-being for they are beings.
CHAPTER 3
THE KINDS OF SUPPOSITION
SECTION 3.1. PROPER AND IMPROPER SUPPOSITION
[3.1.1]
We should now speak of the common modes and divi-
sions of supposition. The first division of supposition is
that some <supposition> is proper and some is improper or
tranferred. <Supposition> is called proper when an utterance
supposits according to its nature or according to the signifi-
cation commonly instituted for it. It is improper when an
utterance supposits according to the signification of another
wllJrd from transferrence according to likeness or some other
cause of the sort, as if we were to say that a meadow IS
cheerful, or to a naughty boy "You are a good boy."
[3.1.2] Chapter 3.1 is about the common modes of supposition 4H , and

has eight sections.
[3.1.3] Section 3.1 is about the division of supposition into proper and
improper supposition; Section 3.2 is about the division into personal
and into impersonal and material; Section 3.3 is about the division of
personal <supposition> into common and discrete, and into absolute and
relative; Section 3.4 is the division of common supposition into natural
and accidental; Section 3.5 is the division into confused and into
determinate <supposition>; Section 3.6 is the division of confused
<supposition> into distributive and non-distributive; Section 3.7 is about
the causes 49 and kinds of distributive confusion; Section 3.8 is about the
causes and kinds of non-distributive confusion.
[3.1.4] The first part of [3.1.1] is clear: sometimes we call some man an
ass, if he is rude, and the ass seems to be a rude animal. Thus the term
'ass' in not taken in the sentence to supposit properly, since <it does not
117
supposit> for itself or for an ass (which it was imposed to signify), but for
those things which the term 'rude' supposits for as if it were to be taken
in the place of the term 'ass' in the sentence.
[3.1.5] Thus we call a cruel man a lion, or God a pious lamb or the lion
of Judea, and so we say that a meadow is cheerful, and that someone is
plowing the beach (if someone is doing something useless). Similarly,
but ironically, a father says to his son "You are a very good boy,"
intending that the predicate 'good boy' sup posit not for a good boy but
for a bad one.
[3.1.61 Improper SuppOSItIon is irrelevant to the speculative sciences

even though it is relevant to poetry, and so I pass over it for now.
SECTION 3.2. MATERIAL AND PERSONAL SUPPOSITION
[3.2.1]
Supposition is divided in one way into personal and in
another way into material <suppposition>. Supposition is
called 'personal' when the subject or the predicate of
the sentence supposits for its ultimate significates or its
ultimate significate, as the term 'man' supposits for men in
the sentence "Man is running." But supposition is called
'material' when an utterance supposits for itself or something
similar to itself or for its immediate significate, which is the
concept according to which it is imposed to signify, as the
term 'man' in the sentence "Man is a species, animal is a
genus."
[3.2.2] This passage makes three points.
[3.2.3] The first point is the division of supposition50 , and the other two
state the parts of the division.
[3.2.41 It should be known about the first point, <the division of

3. THE KINDS OF SUPPOSITION 119
supposition into material and personal>, that some people posit a third
part of the division, which they call simple supposition. They believed
natures to be universals distinct from singulars outside of the soul.
Hence they said a common term supposits personally when it supposits
for those singulars, and <it supposits> simply when <it supposits> for the
universal nature, and <it supposits> materially when <it supposits> for it-
self.
[3.2.5] But I believe that Aristotle correctly refuted this posItIOn In

Metaphysics VII, and hence simple supposition should be dispensed
with, at least according to the analysis <mentioned in [3.2.4]>.
[3.2.6] Similarly, others call supposition 'simple' when an utterance sup-

posits for the concept according to which it is imposed, and 'material'
when it sup posits for itself or for something similar to it. This may be
permitted, but I do not care, since I call each material supposition.
[3.2.7] Note also that supposition is not called 'personal' because it

supposits for a person, nor 'material' because it supposits for matter.
Hence the names <'personal' and 'material'> are used improperly in the
senses which the other two points explain. We agree to use those names
in this way, which is permissable, since names signify by convention.
[3.2.8] For the other points recall what was said in Section 2.3, namely
that categorematic words apt to supposit signify some things through
mediating concepts, according to which concepts (or similar ones) they
are imposed to signify. Thus we call those things conceived by those
concepts the 'ultimate significates' of a sentence, and we call those
concepts the 'immediate significates.' The second point is clear from this
<terminology>.
[3.2.9] But note for the third point that since universals, in the predica-
tion principally given do not exist outside of the soul sl - they are only
concepts of the soul by which the soul indifferently conceives many
things. For example, since <the soul> conceives all men indifferently by
some concept, the name 'man' is imposed. The same for the concept of
animal with respect to animals. Hence since species and genera are
universals according to predication, it is apparent that species and
genera are such concepts of the soul; it is true that spoken terms are
consequently called genera and species (for example the utterances

'man' and 'anima\,), but they are not called universals or genera and
species principally, but rather significatively, according to their attribu-
tion to those universal <concepts>, in that according to them they are
imposed to signify other things, just as urine is called 'healthy' in that it
signifies the health of an animal. Consequently even written terms are
called genera and species, or singulars, or subjects and predicates, and
the like, by virtue of the mediating utterances.
[3.2.101 Therefore the spoken term 'man,' when it supposits for that
concept according to which it is imposed to signify men, as by saying
"Man is a universal" or that man is a species, i.e. that mental term to
which the term 'man' is subordinated, then this is called material sup-
position. Similarly, when an utterance were to supposit for itself or
something similar to it, or for an inscription corresponding to it, as for
example by saying that 'man' is a monosyllable or a name with three
letters and is written in my book with beautiful handwriting: any such
supposition is called material supposition.
[3.2.111 Note about material supposition that if some utterance sup-

posits materially for an utterance, it does not nevertheless have to sup-
posit for itself, but often supposits for something else which is similar. 52
For example, if I say "That man is an animal is true" or 'That man is a
stone is false," the subjects of the sentences, which are infinitive phrases,
supposit materially for sentences, since I assume that only a sentence is
true or false.
[3.2.12] Hence if "That man is a stone is false" is true, then 'that man is
a stone' must sup posit for some sentence, and therefore not for itself,
since it is not a sentence." It supposits for some sentence in the indica-
tive, such as the sentence "Man is a stone" or something similar.
[3.2.13] We thus agree to use such a sentence in the indicative, even

though it can supposit personally, as for instance "That Socrates acts
well is a good thing" or "Cutting is acting" or "A man being white is a
man being colored" or "God's being is God."
[3.2.14] But there are several doubts which can be raised about the
aforesaid sentences. 54
[3.2.15] The first doubt is: if a term or a spoken expression can supposit
materially and personally in some sentence, which supposition is there-
fore most appropriate de virtute sermonisr s That is, are sentences such
as "Man is a species," "Animal is a genus," "To love is a verb," and so
on, true or false de virtute sermonis, or should we draw some distinc-
tions?
[3.2.16] There are three opinions on this question.
[3.2.17] Some people say that "Man is a species" is false de virtute ser-
monis, since the principal supposition is personal. Utterances are im-
posed to supposit for their (ultimate) significates, since we cannot carry
the ultimate significates to disputations, and they were not imposed to
signify themselves, since utterances not imposed to signify could sup-
posit in this fashion - as, for example, "Bu has two letters" or "Buba has
two syllables." They then say that truth and falsity should de virtute ser-
monis be given with respect to the principal given supposition, and
hence <in this case> to personal supposition. Thus "Man is a species" is
false, and hence it is simply false de virtute sermon is.
[3.2.18] Another view is the contrary <of the view expressed in [3.2.17]>,
namely that by the meaning of the words "Man is a species" is true,
because a word (sermo) signifying and suppositing significatively has
only the powers given to it by imposition and usage, and we can only
know what was imposed by the use of an authority; authorities and phi-
losophers use "Man is a species" as true, and this is only true by material
supposition; thus we account this sentence to have material supposition
by the meaning of its words.
[3.2.19] There is a third view, with which I agree, that an utterance in

signifying or suppositing has only the proper powers we give to it. And
so by an agreement of those disputing, as in Obligationes,56 we may
impose a new signification on it and not use the common signification,
just as we may speak by transferrence or ironically with another signifi-
cation.
[3.2.20] Nevertheless, we call a locution 'proper' when we use an utter-

ance in its common signification instituted principally, and we call a lo-
cution 'improper' if we use it otherwise, although it is permissable to use
it otherwise. It is absurd to say simply that a sentence of an authority is

false if he gives it according to an improper locution by which it is true.
We should rather say that it is true according to the sense given to it, in
conformity with Aristotle's remark in the Nicomachean Ethics [I. iii
1094b4] that words are received and understood according to their sub-
ject-matter. Hence we should correctly say that a sentence would not be
true if it were given and received as a proper locution.
[3.2.21] Therefore, if we want to conventionally call the properties of

words their 'powers' (virtus), then I would say that "Man is a species" is
true, since it is laid down by authority, although it is not true de virtute
sermonis, i.e. it would not be true if it were given as a proper locution,
since primarily and principally utterances were imposed to signify in
order that they might stand for (starent pro) their significates and for
themselves, as was said in [3.2.171.
[3.2.22] It therefore seems to me completely that when it is apparent

that an authority has put forward a sentence in some true sense, al-
though not as a proper locution, to simply deny the sentence would be
contentious and mere cavil. But to avoid any error it should be said that
the sentence is not true in its proper sense or de virtute sermon is, and
the sense in which it is true should be made explicit.
[3.2.23] Some other minor doubts can be raised.
[3.2.24] <For example>, how does 'runs' supposit in the sentence "Runs
cannot be the subject of a sentence"? If you say that it sup posits materi-
ally for itself, then the sentence would be false, for as a matter of fact it
is put as the subject of the given sentence.
[3.2.25] I reply that it supposits materially for any such term, but the
sentence needs analysis, namely as "A term such as 'runs' can <not> be
the subject of a sentence taken significatively, but nevertheless it can
taken materially."
[3.2.26] Another doubt is how the subject of "A name is an expression

significant by convention" supposits; it seems that it supposits material-
ly, since it is taken for itself and for many other terms.
[3.2.27] I say that it supposits personally, since it does not supposit for
itself by the ratio by which the term 'name' is such an utterance, but by
the ratio by which the term 'name' and many other terms are the ultimate
significates of the word 'name: 57 Thus in the spoken sentence "Every
utterance is a sound" the terms supposit personally, although they sup-
posit for themselves, since they supposit for themselves only by the ratio
that they are, with other utterances and sounds, the ultimate significates
of the terms 'utterance' and 'sound:
[3.2.28] There is another doubt about the sentence "Genus is a species

of universal."
[3.2.29] I say that this sentence can well be true, taking the whole
significatively and personally, and even taking the terms 'genus' and
'universal' materially. It is true personally since the terms 'animal;
'plant; 'substance; 'color; and the like are about their ultimate signifi-
cates, the genus of which terms I call 'universal.'5H The sentence is true
significatively, since the term 'animal; which is a genus, is a species of
the term 'substance; which is a universal. <The sentence> is true materi-
ally, since the term 'genus' is a species of the term 'universal:
[3.2.30] There is another doubt about the sentences "Socrates is a

primary substance" and "Man is a secondary substance."
[3.2.31] I say that the terms 'Socrates' and 'man' supposit materially,
insofar as such terms are found in the category of Substance; as for the
term 'substance; it should be said that it is used in logic and in philoso-
phy to supposit for terms of the category of Substance, which we call a
category of Substance. And it is apparent to me that this is material sup-
position, and this is <the sort of supposition> in the given sentences, since
they are true and they are laid down by authorities. But when the name
'substance' supposits for subsisting things signified per se by substantial
terms, then there is personal supposition. And the same should be said
for the term 'quality' and the term 'accident; and so on for many others;
they supposit for terms in the category of Quality and for the category
of Accident, and so <they supposit> materially. Sometimes they too <sup-
posit> for things signified by those terms, and then <they supposit>
personally. Thus if we say "White is an accident" the terms supposit
materially, and if we say "A man is white" or "A substance is white" then
the terms suppa sit personally - I say this insofar as such sentences are
true.
[3.2.32] There is also a doubt about this sort of sentence: "Color is the
primary object of vision," "Sound is the primary object of hearing," "It
happens to a man to be white," "It is appropriate per se and primarily
for a triangle to have three angles equal to two right angles," and so on.
[3.2.33] I say that sentences such as "Color is the primary object of

vision" and "Sound is the primary object of hearing" can be analyzed by
personal supposition. For some color is <the primary> object of vision,
and sound is the primary object of hearing since some sound is the
primary object of hearing; and <such a claim> can be true if among all
the things which are principally objects of vision there is some color
which is the object of vision and none of the others is a prior object of
vision. 59
[3.2.34] But Aristotle, holding the eternity of the world, thus denied
"Color is the primary object of vision," since no color is an object of
vision where nothing else is a prior object of vision, if we posit that every
color is generable and corruptible. And the same can be said (with the
relevant changes made) for sound and hearing.
[3.2.35] But perhaps authorities put such sentences forward in the sense
of material supposition, since a sentence such as "Color or the colored
is visible" is per se and primarily true, and one must take 'color' widely
as it is extended to light. So too with "Sound is called the primary object
of hearing," i.e. the sentence "Sound or what is sounded is audible" is
true per se and primarily.
[3.2.36] (Here indeed I say 'per se and primarily' as they are said of the
sentence "Man is risible.")60
[3.2.37] However, in the sentence "A man happens to be white," 'white

or the whole should be taken in a material sense, so that such a sentence
is true per accidens. Similarly for the triangle and the isoceles triangle:
that is, they should be taken <in the material> sense, so that a sentence
such as "A triangle has three <angles equal to two right angles>" is true
per se and primarily, and the sentence "An isoceles triangle <has three
angles equal to two right angles>" is true per se, but not primarily, since
it does not convert.
[3.2.38] Simiarly, if I say "It is false that God does not exist" or "It is
true that man is an animal," the subjects supposit materially for sen-
tences such as "God does not exist" and "Man is an animal." But if I say
truly "That man is an ass is nothing" then all the terms are taken signifi-
catively and the subject does not supposit. However, if I say "God's be-
ing is God" I say something true and both terms supposit personally.
SECTION 3.3. COMMON AND DISCRETE PERSONAL SUPPOSITION
[3.3.1]
Personal supposition is divided into common and discrete:
a common term supposits with common supposition, as for
example 'man,' and a discrete term <supposits with> discrete
supposition, as for example 'Socrates' or 'this man.' Again,
personal supposition is divided into absolute, which comes
about through an absolute term,!') and relative, which comes
about through a relative term.
[3.3.2] This passage presents two divisions, which are clear in them-
selves.62
[3.3.3] Some particular points about relative supposition will be made in

Chapter IV.
SECTION 3.4. NATURAL AND ACCIDENTAL SUPPOSITION
[3.4.1]
Common supposition is customarily divided into natural

supposition and accidental suppositon. Supposition is called
'natural' when a term indifferently supposits for everything
for which it can supposit, past and future as well as pre-
sent; this is the sort of supposition we use in demonstrative
science. Supposition is called 'accidental' when a term sup-
posits only for the present, or only for the present and past,
or only for the present and future, according to the require-
ments of the verb and the predicate, as will be explained
later. We use this sort of supposition most of all in historical

writing, and Sophists also use a great deal.
[3.4.2) Many modern writers deny <that there is> natural supposition in
the sense given in [3.4.1), but but without going through their views I
show that we often do use it. 63
[3.4.3) First of all, the verbs 'to understand,' 'to know,' 'to comprehend,'
'to opine,' 'to signify,' and many others make the terms construed with
them supposit or be held for the past and the future as much as for the
present. Hence "A man who was is understood to be Aristotle; there-
fore, a man is understood" follows, as does similarly "A man who will be
is understood to be the Antichrist; therefore, a man is understood," as I
have explained in my Sophismata cap.v.
[3.4.4) Secondly, <natural supposition> is apparent in the sentence "What

was or is or will be B is running." The subject supposits for the past and
present and future, whether the sentence is true or false.
[3.4.5) Thirdly, natural supposition is apparent where a term is con-

fused by these words (or similar ones): 'perpetually,' 'eternally,' and
'always: Such words are imposed, as seems apparent from the usage of
authorities, to be distributed for every time, past, present, and future.
However, this does not explain the usage of the word 'always' when we
say that Socrates always runs, in that he runs only for this whole hour,
since we only use it for the present. Thus in the given case we concede
"Socrates runs at every time," but even so all would repudiate the claim
that Socrates eternally runs, or that Socrates perpetually runs, or that
Socrates always runs, but Aristotle allows such of him in that he is
moved. Thus it is said that "Man is always animal" is not equivalent to
"Man at every time is an animal," but rather to "At every past, present,
and future time man was, is, and will be an animal." However, it seems
to me that neither this nor its opposite can be demonstrated, since it de-
pends on the conventional signification of words.
[3.4.6) Next, fourthly, demonstrative science uses this sort of supposi-

tion. If we say in the Meteorologica that all thunder is a sound in the
clouds, or that any rainbow is a reflection or refraction of light, we do
not mean to say this for the present only. Indeed, even if there is no
thunder or rainbow at the present time, nevertheless we say that it is so;

and if the geometer has a demonstration that every triangle has three
angles equal to two right angles, it should not be thought he has this
knowledge only for the triangle which exist now. Rather, if that know-
ledge is habitual, remaining for three years, and in the interim many tri-
angles come into being, he has knowledge of all of them, without any
new demonstration. And this is clearly what Aristotle says in Posterior
Analytics [I. iv 73a30-35]: "for all these things, I say it is not in one
thing thus and not thus in another, nor is it sometimes thus and some-
times not." And he states this by an example, saying "as animal of every
man, if it is true to call something a man it is true to call it an animal,
and if true now of one then it is true now of the other."
[3.4.7] Nevertheless, it can rightly be said in this case that such a locu-
tion with this meaning is not de virtute sermonis, but an abbreviation (ad
breviloquium). When we want to speak demonstratively, we say that all
thunder is a sound in the clouds, and that every eclipse of the Moon is
due to the Earth occupying a position between the Sun and the Moon;
these sentences are not true de virtute sermonis, since the verb 'is' is li-
terally only imposed to signify the present time, and it may be that there
is no thunder or eclipse of the Moon at present. Such sentences are ab-
breviations put in place of "Any thunder, whenever it is, or was, or will
be, is or was or will be a sound in the clouds" and "Every eclipse of the
Moon, whenever it is, was, or will be, is or was or will be because of the
position of the Earth <between the Sun and the Moon>." Nor should
such sentences be denied, since they are true as put forth, though they
would be false if put forward or received in their strict sense. And since
sophists want to receive sentences in only their strict sense, it is hence
not customary to use them.
[3.4.8] Again, just as the intellect can conceive man and animal without
any difference of time by those concepts by which the terms 'man' and
'animal' are imposed, it can thus similarly form a complexive concept of
them without a difference of time,64 and then there will be a mental
sentence indifferent to any past or present or future time. Hence the
terms supposit indiferently for all those times. But we do not have an
expression strictly imposed to signify such a mental <timeless> copula,
and so we can conventionally use the verb 'is' to signify such a copula
through which we signify the present time no more than the past or the
future, but rather no <time at all>. This will be the natural supposition of
the terms.
[3.4.9] Perhaps we may even show that we form such mental sentences,
because God can conserve all things in rest without motion, that is,
wholly without motion. 65 Let us posit that He does so. In that case there
will be no time, if all time is motion, as Aristotle showed in Physics
[IV.xi 218b- 219a]. Nevertheless, the souls of the blessed would know
and understand that God is good and that they stand with Him; by
means of the mental sentences and other mental sentences conjoined
with them, they would know how they understand time and that they
know no time to exist, and so they would know themselves and God not
to exist at the present time, nor existing at the present time.
[3.4.10] It seems to me that the spoken copula imposed precisely to

signify such a complexive concept would be purely syncategorematic;
verbs which connote a certain time would then be like categorematic
terms, since aside from the concept they signify a thing beyond the con-
cept, and something other than these, that the subject and the predicate
signify time.
[3.4.11] One can argue for natural supposition in this way: just as I can
conceive all animals (e.g. men, horse, etc.) indifferently by a concept
from which the name 'animal' is derived, completely free from the fact
that I can conceive men or horses or asses by specific conepts from
which the terms 'man' and 'ass' are derived,!'!> so too I can conceive a
common concept indifferent to all times, from which the name 'time' is
derived, completely free from the fact that I can conceive the present or
the past or the future by specific concepts.
[3.4.12] Therefore, assuming that the copula in a categorical sentence

must signify with a time, nevertheless the corresponding concepts need
not consignify determinately the present time or determinately the past
or the future. Rather, it can consignify time with no difference of past or
present or future by a concept common with respect to time.
[3.4.13] These matters are put forth about the supposition which some,
using names conventionally, call 'natural supposition:
SECTION 3.5 CONFUSED AND DETERMINATE SUPPOSITION
[3.5.1]
Common personal supposition is divided into confused sup-
position and determinate supposition. The supposition of
some term is called 'determinate' when it is necessary for the
truth of the sentence (or an equiform sentence) that it is true
for some determinate supposit. For example, if "A man is
white" is true, then it is true for this man and this white, or
for that man and that white, and so forth. But the supposition
of some term is called 'confused' when the sentence in which
it is found, or an equiform sentence, can be true without
being true for some <one> determinate thing falling under the
term. For example, the sentence "Every man is an animal" is
true, and nevertheless it is not sufficient for its truth that it is
true for Socrates, nor <that it is true> for Plato, and so on.
[3.5.2] This passage makes three points, namely a twofold division and
an explanation of each part of the division, and the analysis of each part
is sufficiently clear in the passage. n7
[3.5.3] Nevertheless, I say that for determinate supposition it is not ne-

cessary that the truth be for one supposit only, but sometimes it is true
for any <supposit>. Of course, this is necessary and sufficient for it to be
true for a single <supposit>.
[3.5.4J Note immediately, then, that there are two conditions for the de-
terminate supposition of some common term.
[3.5.5] The first <conditiOn> is:
[Condition 1J From any given supposit of a term the com-

mon term can be inferred with the other <terms> unchanged
in the given sentence.
For example, in "A man is running" the term 'man' supposits determi-
nately, and so "Socrates is running; therefore, a man is running" follows,
or "Plato is running; therefore, a man is running" follows, and so on for
any singular contained under <the term> 'man:
[3.5.6] The second condition is:

[Condition 2] From a common term suppositing this w~y,
<i.e. with determinate supposition>, all the singulars can be m-
ferred disjunctively in a disjunctive sentence.
For example, this follows: "A man is running; therefore, Socrates is
running or Plato is running or John is running ..." and so on for the
rest.
[3.5.7] However, in all confused supposition neither of these conditions

are met.
[3.5.8] If you ask how I know the supposition is determinate, I say that
you can know this if you see that these two conditions are met, or if you
see that there is no reason for confusion. These reasons are discussed in
Section 3.7 and Section 3.8.
SECTION 3.6. DISTRIBUTIVE AND MERELY CONFUSED

SU PPOSITION
[3.6.11
Confused supposition is divided into distributive and non-

distributive <supposition>, which is customarily called 'merely
confused <supposition>,' <Supposition is called> 'distributive'
according to which from a common term there can be in-
ferred any of its supposits individually or even all conjoined
together in a conjunctive sentence. For example, it follows
from "Every man is running" that therefore "Socrates is
running," or "Plato is running," and so for the rest. But sup-
position is <called> 'merely confused' when there does not
follow any of its singulars individually, keeping the other
terms in the given sentence the same. Neither do the singu-
lars follow disjunctively in a disjunctive sentence, although
perhaps a sentence with a disjunctive extreme follows.
[3.6.2] This passage, just like [3.5.11, makes three points, namely a two-
fold division and an explanation of each part <of the division>.()~
[3.6.3] It is clear that distributive supposition differs from other <types

of> supposition, because according to it one can infer from a common
term any of its singulars, which cannot be done in the other <kinds of
supposition>. Therefore,
[Condition 3] If a sentence is true <with distributive supposi-
tion>, it must be true for any supposit, which is not required
in the other kinds of supposition.
[3.6.4] Merely confused supposition differs from determinate supposi-

tion in that with confused supposition singulars are not inferred from a
common term in a disjunctive sentence, which correctly takes place in
determinate supposition. For example, in the sentence "Every man is an
animal" the term 'animal' has merely confused supposition; "Every man
is an animal; therefore every man is this animal or that animal or ..."
does not follow, since the first is true and the latter are all false.
[3.6.5] Hence these kinds of supposition are related to each other such
that from distributive supposition there follows determinate 69 or merely
confused supposition, but from neither of the latter does distributive
supposition follow as a formal consequence.
[3.6.6] However, in the common manner of confusion there does follow

from determinate supposition confused <supposition>; indeed, from "An-
imal is every man; therefore, every man is animal" does rightly follow.
But determinate supposition does not follow from confused supposition
unless the reason for the confusion is removed. Hence "Every man is
animal; therefore, animal is every man" does not follow, but when the
reason for the confusion, the distributive sign, is removed, it does
follow; for instance, "Every man is animal; therefore, animal is man."
Why I have said 'in the common manner of confusion' will be clear in
RuleNDC-1.
SECTION 3.7. RULES FOR DISTRIBUTIVE CONFUSION
[3.7.1]
There are many reasons for the distributive confusion of a
common term.
[Rule DC-II A universal affirmative sign distributes the term

immediately following it with which it is construed. For ex-
ample, in the sentence "Every man is animal" <the term>
'man' is distributed.
[3.7.21 Section 3.7 deals with the reasons for distributive confusion, and
we give below five rules. 71l <The other parts of the passage will be given
in the course of this section.>
[3.7.31 Rule DC -1 <given in [3.7.11 above> is about the universal affirma-

tive sign.71
[3.7.41 There are many and varied universal affirmative signs;some are
substantial and others pertain to an accidental category
[3.7.51 Some substantial <affirmative distributive signs> distribute part of

the subject, as for example 'every,' 'whichever,' 'whatever,' 'whomso-
ever,' or 'each'; others distribute an integral part, as for example 'whole.'
[3.7.61 Universal signs pertaining to accidental categories are 'when-

ever,' 'in-what ever-way,' 'however,' 'howsoever,' 'however-many,' 'how-
ever-much,' 'whenever,' 'howsomany,' 'always,' 'perpetually,' 'eternally,'
and even 'any-given-amount,' 'any-given-way,' 'any-given-place,' and so
on. Something brief must be said about all these. (The signs 'every' and
'whichever' are taken to be equivalent.)
[3.7.71 There is a doubt whether in "Not every man is running" the term
'man' is distributed. It is said that it is not, since the rule is that for a ne-
gating negation, if there is found after it some terms which would be dis-
tributed with the negation removed, then the distribution is removed,
and if there is some common term which would not be distributed with
the negation removed, then it is distributed <Rule NDC-2 in [3.8.131
below>.
[3.7.81 It is then asked whether in "Every white man is running" if

'man' or 'white' is distributed. I say that neither 'man' nor 'white' is
distributed, but rather the whole 'white man' is distributed, since the rule
is that the complex made up of an substantive and a adjective, as a
determinable and a determinate, is taken in a sentence as a single term
such that the whole is distributed together and not partially, whether the
substantive 72 precedes the adjective or vice-versa.
[3.7.9] But then it is asked of "Every man is white" whether the whole
'white man' is distributed. I say that it is not, but only 'man' <is dis-
tributed>, since there is no complex made up of a determinable and a
determinate, but rather <we have here> a subject and predicate mediated
by the copula.
[3.7.10] It is asked what is distributed in "Any ass of-a-man is running,"

and similarly "Any of-a-man ass is running," and also of the pair "Of-a-
man any ass is running" and "Of-any-man an ass is running."
[3.7.11] I say of the first pair,73 and all such in which the distributive
sign precedes the nominative and oblique terms, that the sign goes with
the nominative term and never with the oblique term; I say that the
whole complex of nominative and oblique term is distributed together
and neither <is distributed> individually, as said about a substantive and
an adjective in [3.7.8]. The oblique term is related to the nominative
term as a determination to a determinable, as with the adjective and the
substantive, unless a syncategorematic term intervenes or precedes it to
make a difference, which is discussed in [3.7.121.
[3.7.12] I say of the third sentence in [3.7.10], "Of-a-man any ass is

running," that the word 'man' is not distributed in any way and retains
only its accepted determination, but that the word 'ass' is distributed,
although not simply but as contracted by the word <'man'> taken deter-
minately. Hence if we want to take something falling under that distribu-
tion when we syllogize, we must syllogize as follows: "Of-a-man any ass
is running, and Brunellus is an ass of-that-man; therefore, Brunellus is
running."
[3.7.13] I say of the fourth sentence in [3.7.10], "Of-any-man the ass is

running," that the nominative term is not distributed, but rather only
<the oblique term> 'man,' in that the sign is not nominative but oblique.
[3.7. 141 Next, it seems to me that the distributive <signs> 'whoever,'

'whichever,' 'whatever,' and 'whatsoever' are equivalent to one another,
and are usually taken as equivalent to the signs 'every' or 'any,' and so
no more need be said about them.
[3.7.15] But sometimes they are taken antecedently to their relative-

term, either expressly or understood, as when I say "Whatever runs is
moved"; when they are so taken, there is a doubt about what is dis-
tributed.
[3.7.16] I say that they should be resolved <in such cases> into a substan-
tive adjective, so that 'whatever runs' <becomes> 'whatever thing runs,'
or "Whatever thing runs that thing is moved." Then it is clear that the
word 'thing' is distributed, but only as restricted by 'runs,' so that it is
equivalent to say "Whatever [or 'whichever'] thing runs that thing is
moved" and to say "Every thing which runs is moved," and here the
whole 'thing which runs' is distributed. I say that these are equivalent
with respect to distribution; in either, the term 'thing' is only distributed
for the supposits to which the predicate 'runs' is appropriate. But they
differ with respect to ampliation, according to which it was said in
Sophismata <cap.v sophism 8> that "Every thing which will be is" should
be allowed, since 'thing' is construed as restricted with 'is: But "What-
ever thing will be, that thing is" should be denied, since 'thing' here is
ampliated and construed with 'will be:
[3.7.17] That is the mode of distribution of the name 'whatever' or

'whichever' when taken antecedently to its relative-term. But if we say
simply that whichever man is an animal, then 'whichever' and 'any' are
equivalent.
[3.7.18] There is a doubt about the term 'each:
[3.7.19] It is said that it distributes the term following with each of the
things pointed out, as in 'each of these' of 'each of those men,' or by a
demonstrative relation, as in "A horse and an ass are white and each of
them exists."
[3.7.20] There is also a double about what is distributed by the term

'whole' when one says "The whole <of> Socrates is less than Socrates" or
"The whole man is in the house."
[3.7.21] It should be said that we often used the word 'whole' as

categorematically equivalent to 'having parts,' which is not relevant to
the proposed; otherwise, it is taken syncategorematically, and then it is
the same as saying 'the whole' or 'any part.' Thus in the given sentences
the word 'whole' is not purely syncategorematic, but implies in itself the
categorematic term 'part.' So, when I say "The whole of Socrates is less
than Socrates," it must be analyzed or resolved so as to express what is
distributed; I say that the complex 'part of a man' is distributed.
[3.7.22] It should also be known 74 that 'whole man' is analyzed in a

different way from 'man whole,' since the former is 'whatever part of a
man' and the latter 'of-a-man a certain part.' These are quite different,
since if Socrates is in a house and Plato is outside, "Some part of-a-man
is in the house" is false and "Of-a-man some part is in the house" is true.
[3.7.23] There are many other ways of taking the term 'whole,' concern-
ing which there will be more in the Sophismata.
[3.7.24] We should now discuss the distributives 'anywhere' and

'always' in the same way. I say that these words imply distributive
syncategoremata and distributed categoremata: 'anywhere' is the same
as 'every place' and 'always' is the same as 'any time,' or perhaps even
the same as 'the whole in the past, present, and future,' as said in the
same way for 'eternally' and 'perpetually' in [3.4.5].
[3.7.25] The same should correspondingly be said for 'whenever,' and

for 'whenever' as said for 'whichever': if we take 'whenever' for 'always'
and 'wherever' for 'anywhere,' then the matter has been settled for them.
[3.7.26) But those names seem more to be imposed to be antecedent to

the relative-terms 'there' and 'then.' For example, "Whenever there is an
eclipse of the Moon, then the Earth is placed between the Sun and the
Moon" and "Wherever there is a man, there there is an anima!." There is
a doubt as to what is distributed in these cases; I say that it is apparent
by the resolution of 'whenever' <as 'in any time' and 'wherever' as> 'in
any place'; hence the terms 'time' and 'place' are distributed, not simply,
but restricted by what they are added to. In the above examples, time is
only distributed as the time at which there is a lunar eclipse, and place
as only the place in which there is a man.
[3.7.27] 'However' and 'however-much' should be treated similarly: they

distribute the terms 'quality' and 'quantity' as restricted by what they
are added to, and are antecedent to their relative-terms 'such' and 'so
much.' For example, "However Socrates is, such is Plato": the sense is
that whatever quality by which Socrates is such, Plato is also such by
such a quality; thus 'quality' is only distributed across the qualities by
which Socrates is such. In "However-much is Socrates, so much is Plato"
'quantity' is distributed across those quantities by which Socrates is so
much, and the sense is that by whatever quantities Socrates is so much,
by so much is Plato so much.
[3.7.28] Again, if I say "However-many men there are, there are so

many asses" here the term 'number' is distributed as restricted to those
numbers by which species are some amount, and is the same as "In
whatever number there are men, in that same number there are so many
asses."
[3.7.29] Thus if I say "However-long Socrates sleeps, so long he

dreams" the term 'a time' is distributed, as restricted to the times at
which Socrates sleeps; it is the same as "At whichever times Socrates
sleeps, so many times does he dream." Since the term 'a time' signifies
the number of times his sleep is interrupted, the whole term is dis-
tributed, namely 'the number of interrupted periods of sleep in which
Socrates sleeps.' Thus it is the same as "By whatever number of inter-
rupted periods of sleep Socrates sleeps, in the same number of inter-
rupted dreams does he dream."
[3.7.30] Thus if I say "In-whatever-way Socrates acts, so does Plato act"

or "In-whatever-way Socrates is related to Plato, so is John to Albert"
the term 'way' is distributed, as restricted to the ways in which Socrates
acts or the ways in which he is somehow related to Plato, and the sense
is "In whatever what Socrates acts, so does Plato act."
[3.7.31] The term 'howsoever' seems to me to be equivalent to the term

'in-whatever-way.'
[3.7.32] It should be noted that sometimes we take the term 'who-

ever' without any relation to the relative-term following it, and so
perhaps we may do the same with 'however,' 'however-much,' 'when-
ever,' etc. And if Socrates is called 'however' or 'however-much,' then

this signifies that he is such by every quality, or so much by every quant-
ity.1t is obvious what is distributed here, viz. quality and quantity.
[3.7.34)
[Rule DC-2) A negating negation distributes every common
term following it which, with that negation removed, would
not be distributed; and it does not distribute anything pre-
ceding it.
[3.7.35) (I say here a 'negating negation' to distinguish it from an infini-

tizing negation, which only distributes the term it makes infinite. Thus if
we say "Not: a man is white," taking the negation negatively, then the
terms 'man' and 'white' are distributed, as is the verb 'is' with respect to
the time it consignifies.)
[3.7.36) Note that the distribution of a universal negative sign is re-

duced to this mode of distribution, since a universal negative sign is
nothing other than a word implying in itself a negation with a particular
affirmative sign. 'None' is the same as 'not any' or 'not some'; 'neither' as
'not either'; 'never' as 'not ever.' Hence such a sign in virtue of that nega-
tion distributes every term following it which would not be distributed
with that negation removed, with an implicit particular sign.
[3.7.37) Note also that if a sentence is hypothetical then a negation

found in the first categorical does not act on the second categorical
unless it is preposed to the whole sentence so that it is understood to act
on the whole hypothetical. For example, in saying "A man does not run
and a horse is white" the terms 'horse' and 'white' are not distributed.
But if 1 were to say "Not: a man is running and a horse is white" (with
the negation covering the entire hypothetical), the sense would be that it
is not the case that a man is running and a horse is white, and so would
distribute the whole, contradicting the conjunctive sentence "A man is
running and a horse is white." Hence it is the same as the disjunctive
sentence "No man is running or no horse is white."
[3.7.38) But then a doubt arises about the sentence "Socrates is not
where a man is": what is distributed?
[3.7.39] I say that this sentence can be stated as a categorical, whose

sense is: "Socrates is not a being where a man is," and the whole 'being
where a man is' is a predicate composed from a determinable and its
determination; 'being' is determinable and 'where a man is' is its deter-
mination, and the sentence is converted as "No being where there is a
man is Socrates." But it can also be put as a hypothetical which is the
same as a single categorical, as was said; it would then be the same as
"Socrates is not anywhere and there is a man here." In this case, 'man' is
not distributed.
[3.7.40] Next, there is a doubt about whether 'substance' is distributed

in "An animal which is not a man is a substance."
[3.7.41] It should be said that 'substance' is not here distributed, since

the negation does not negate the whole sentence; the predicate of the
sentence is 'substance' and the subject is the whole 'animal which is not
a man,' and the copula is the verb 'is' put between them. But 'man' is
properly distributed here, since part of the sentence is 'which is not a
man.' Even though this is not a sentence (since it is part of a sentence) it
nevertheless has a likeness to a sentence with respect to the distribution
and supposition of the terms; such terms supposit and appellate in an
expression which is part of a sentence and which taken of itself is a
sentence as they would in a sentence taken per se.
[3.7.42]
[Rule DC-3] A common term is distributed by an infinitizing
negation, as for example if I say "Animal is non-man," since
it follows therefore that "Animal is non-Socrates," "Animal is
non-Plato," and so on for the rest. Similarly, "Non-man is
animal; therefore, non-Socrates is an animal" follows, even
though the negation is infinitizing.
[3.7.43] There is a doubt about Rule DC-3, namely how 'man' is treated
in "No <non>-man runs" and "Every non-man runs."
[3.7.44] I say that in the first sentence 'man' is no longer distributed,

since the negation preceding it falls under the distribution of the infini-
tizing negation. Hence "No non-man runs; therefore, no non-Socrates
runs" does not follow. Thus it should be said that the whole infinite term
(namely 'non-man') is distributed, and so, pointing out anything which is

not a man, "Any non-man runs; therefore, this non-man runs" follows.
For "Every non-man runs" it should similarly be said that 'man' is no
longer distributed, since there is a traditional rule put forward: "what
moves the immobile does not move the mobile," i.e. what naturally
distributes an undistributed term can remove the distribution of a dis-
tributed term.75 Now it is a fact that in saying "Non-man runs" the word
'man' is distributed and 'non-man' is not distributed. Hence, if we add a
universal sign which naturally distributes, 'man' is no longer distributed
and 'non-man' is distributed. Thus "Every non-man <runs>; therefore,
every non-Socrates runs" does not follow.
[3.7.45]
[Rule DC-4] A common term is distributed by comparison
following a comparative term, such that the term of the com-
parative grade distibutes the ablative term following it and
ends the comparison. For example, in saying 'greater-than
man' the term 'man' is distributed. Similarly, tre same term
with a mediating 'than' can distribute the nominative follow-
ing it, ending the comparison, as in 'greater than man' the
term 'man' is distributed. Similarly, in a comparison with a
term of a positive grade, with the word 'as' mediating and
taken comparatively, the comparative following it is distri-
buted, ending the comparison. For example, in "Socrates is
as great as some horse" the term 'horse' or 'some horse' is
distributed. Again, a superlative grade has a special way of
distributing: it distributes the complex made up of its sub-
stantive and the differentiating relative-term added to it,
which is clear by its analysis. Something is called the First
Being either because it is prior to any other being or because
no other being is prior to it, and a man is the strongest either
because he is stonger than any other man or because no
other man is stronger than him.
[3.7.46] Note briefly about this passage that if the ablative term pre-
cedes the comparative it is not distributed, as when I say "Socrates a
man is stronger than," 'man' is not distributed. 76
[3.7.47) But there is a doubt whether 'man' is distributed in "The horse

is stronger than any man."
[3.7.48) I say that 'man' is not distributed, according to the rule men-
tioned in [3.7.44). Here there is a double distributive on 'man,' namely
'stronger than' and 'any: Hence one takes away what the other pro-
duces, and so it is equivalent to "The horse is stronger than man."
Similarly, these are equivalent: "The horse is stronger than any man"
and "The horse is stronger than man."
[3.7.49) If any other things should be said about this passage, they will
be said in the Sophismata.
[3.7.50)
[Rule DC-5) Distribution comes about by many words im-
plying negations in themselves or in their analysis, as for
example in privative words or differentiating relative-terms,
as the verbs 'to differ; 'to begin; 'to cease' and the words
'without; 'beyond,' 'only,' and many others which will be
treated in particular in the Sophismata.
[3.7.51) I only note about this passage that if some sentence in which a
distributive is found requires analysis, due to its obscurities, into
another sentence or sentences, a term can be said to have the supposi-
tion <or terms> to have the supposition which is found in the resulting
sentences.
SECTION 3.8. RULES FOR NON-DISTRIBUTIVE CONFUSION
[3.8.1)
Similarly, there are many reasons for non-distributive
confusion.
[Rule NDC-Ij A universal affirmative sign non-distributively
confuses a common term not following it immediately.
For example, in the sentence "Every man is animal" the term
'animal' supposits confusedly and non-distributively, which is
clear since "Every man is animal; therefore, every man is
Socrates" does not follow. Hence it does not supposit dis-

tributively. Similarly, the disjunctive sentence "<Every man is
animal>; therefore, every man is this animal or every man is
that animal (and so on for all the animals)" does not follow,
for the antecedent is true and the consequent is false. Hence
it does not supposit determinately and so supposits in the
third way, namely with merely confused supposition.
[3.8.2) Section 3.8 deals with the reasons for non-distributive confusion,
and like Section 3.7 has several short passages given below with several
rules. 77
[3.8.3) There is a doubt about Rule NDC-1 given here in [3.8.1), namely
how 'animal' supposits in the sentence "Seeing any ass is an animal":7x it
seems that it has merely confused supposition according to the given
rule.
[3.8.4) Here it need only be said that 'animal' supposits determinately

and not confusedly, and hence we should add to Rule NDC-1 that the
common term not immediately following the sign, if it is to be confused
by the sign, is construed with some term confused by that sign. For
example, "Of-every-man an ass is running" has the term 'man' first con-
fused by the term 'every' distributively, and then the verb 'is' is confused
non-distributively with respect to the consignified time (since it is taken
with the term 'ass'), and finally the term 'running' is construed with 'ass'
and with 'is.'
[3.8.5) But when we say "Seeing every ass is an animal," the term
'animal' is not construed with some prior confused term, since it is not
construed with 'ass' which is the term primarily confused; thus if I say
"The ass of-every-man is running" neither the verb 'is' nor the predicate
'running' is confused, since neither has a connection in the general
construction with the term 'man' which is primarily confused.
[3.8.6) But if you were to ask whether I say therefore that 'ass' is con-
fused in "Every man an ass is seeing," I say that 'seeing' is construed
with 'man,' which is primarily confused, and for the rest 'ass' and the
verb 'is' are construed with 'seeing' and so are distributed, according to
the universal rule that if a universal sign precedes the entire remainder
of the sentence then it distributes something, and the remainder is

connected with what is distributed.
[3.8.71 There is also a doubt about "Any animal of-the-king is a horse,"

positing that no king has any animal other than a horse; then the
sentence is true and we ask how 'animal,' 'king,' and 'horse' stand.
[3.8.81 I say that the whole 'animal of-the-king' is distributed, as was

said in [2.6.331, but that neither part is distributed (i.e. neither 'animal'
nor 'king'), but rather each is held confusedly and non-distributively. We
cannot descend to a determinate supposit, for if we descend under
'animal' to an ass the sentence is false, and if we descend under 'king' to
a king who lost all of his animals, the sentence is false; thus it is clear
that neither part of the distributed complex is distributed, but is merely
confused.
[3.8.91 There is also a doubt about "All matter and form are a com-
posite."
[3.8.101 I say that 'form' has merely confused SuppOSItion, since we

cannot descend to some supposit, nor to all <supposits> in a disjunctive
sentence. And such would be the case for any such sentence, namely
where a conjunctive subject is taken collectively. But if it is taken
divide diy, as in saying "Every man and ass are running," then 'ass' is
held determinately in that such a sentence, as said in [2.6.681, is resolved
into a hypothetical, namely "Every man is running and every ass is run-
ning," where 'ass' supposits deterrninately.
[3.8.111 But it was said in [2.6.69] that "Every man and ass is running"
can be understood in one way so that the distribution only falls under
'man,' and then the rest should be resolved in the hypothetical "Every
man is running or an ass is running," where 'ass' supposits determinately.
Hence it supposits in that way in the first <sentence before the resolu-
tion>.
[3.8.12] But in another way the distribution can be understood to fall

under the whole subject, and then the whole 'man or ass' is distributed
for everything which is a man or an ass. And since every man is a man
or an ass, and every ass is a man or an ass, it is thus distributed for every
man and for every ass; hence the whole is taken distributively.
[3.8.13]
[Rule NDC-2] A common term is non-distributively con-
fused by a double distributive antecedent to it, where either
<distributive> would distribute the rest without the other.
For example, if I say "Every non-man is running" or "No
non-man is running," then the term 'man' is non-distribu-
tively confused. Thus if I say "No man is no man" or "Every
man is every man," the term 'man' found in the predicate is
merely confused.
[3.8.14] From Rule NDC-2 the issue is clear for differentiating relative-
terms and for the verb 'to differ,' which is taken to distribute the ablative
terms following it. For example, in "B differs from a man" or "B is other
than, or diverse from, man" the term 'man' is distributed. If a negating
negation were put in front, then 'man' would be merely confused, as in
saying "B does not differ from a man" or "B is not other than, or diverse
from, man." Thus if we were to say "B differs from every man" the term
'man' is merely confused there.
[3.8.15] Again, in "No man sees every ass" the term 'ass' is merely
confused, but in "Every man sees every ass" then 'ass' is distributed
since, although two distributive signs precede it, Rule NDC-2 does not
apply here in that the first sign would not distribute 'ass' without the
second sign.
[3.8.16] But it seems to me that distribution by comparison should be

an exception to Rule NDC-2. For example, if I say that the horse is
greater than man, then 'man' is distributed, and if I add a sign by saying
that the horse is greater than any man, it is true that the distribution is
removed but the confusion remains, or, rather, that 'man' stands deter-
minateiy. I prove this: we can descend to the supposits distributively in a
disjunctive sentence. For instance, if I say "The horse is greater than any
man" then this horse is greater than the least man and then, specifying
that least man as Socrates, it will be true that the horse is greater than
Socrates or Plato and so on for the rest.
[3.8.17] Therefore, if you ask why the change - why Rule NDC-2 holds
more in other cases than in this - I say that surpassing by some amount,
or even by one, must surpass their minimum, and if it surpasses all it
must surpass the minimum. But if every man is an animal, there need
not be any minimum of them and maximum of them, but this is one and
that the other.
[3.8.18] Again, we ought to make an exception to Rule NDC-2 for two

negations taken together related in a similar way to what follows. They
only cancel each other out, and hence the terms supposit as they would
if the negations were taken away. For example, "Not: no man runs" or
"Socrates does not see no man" have 'man' suppositing determinately.
[3.8.19]
[Rule NDC-3] Certain special terms confuse without distrib-
ution the common terms following them, namely terms which
are given in answer to the question "When?" such as '3:00' or
'4:00: etc. as in "I ate bread yesterday at 3:00." Again, terms
from the categories of Time and Place have the ability to
confuse in this way, as for example "Yesterday and today 1
drank wine" or "Bread is sold in Paris and in Rome."
[3.8.20] The confused supposition <which these terms produce>, it

seems to me, comes from the fact that such sentences are resolved into
conjunctive sentences whose categoricals have the same term in the
predicate, where nevertheless one categorical is verified of that term in
one way and the other in another. For example, "At 3:00 1 ate bread" is
true because at one time I ate bread and at another time 1 ate bread and
at a third I ate bread, although these are different times.
[3.8.21] Similarly for "Yesterday and today I drank wine," which is

verified for different times, and so too for "Bread is sold at Paris and at
Rome," since bread is sold at Paris and bread is sold at Rome, although
at different places. Hence if other modes similar to this are found, they
will be conceded to have a similar supposition.
[3.8.22] Some people want to say that in exceptive sentences common

terms have merely confused supposition, as for example in "Every man
except Socrates runs" the term 'man' is thought to have merely confused
supposition, since we cannot descend to some supposit. We cannot,
after all, say "therefore, Plato except Socrates runs," since that would be
ungrammatical and ill-formed.
[3.8.23] But I say that 'man' is distributed in that sentence, although not
simply, but by the exception restricted to other men than Socrates, so
that the whole 'other than Socrates' is distributed, which is clear, since
the sentence is resolved into the hypothetical "Socrates does not run and
every man other than Socrates runs," and then we can obviously
descend.
[3.8.24]
[Rule NDC-4] The verbs 'to know,' 'to comprehend,' 'to
understand,' and many others (discussed more fully in the
Sophismata),79 and their participles, confuse without distrib-
ution the terms following them which terminate their action.
For example, if I say "I know a triangle" it does not follow
that therefore I know an isoceles triangle or I know an equil-
ateral triangle, and so on. So too "lowe you a horse" does
not entail that therefore I owe you Brunellus, or lowe you
Favellus, and so on.
[3.8.25] It seems to me that this kind of confusion is quite different from

the preceding modes, although they are each called 'confusion without
distribution: In the preceding modes, although we could not descend to
the supposits by a disjunctive sentence, we nevertheless could <descend>
by a categorical sentence with a disjunctive extreme. For example, al-
though "Any man is an animal; therefore, any man is this animal or any
man is that animal or ..." etc. does not follow, nevertheless "Any man is
an animal; therefore, any man is this animal or that animal or ..." etc.
does follow. This consequence is true, since the antecedent is clear by
induction, but in this kind of confusion it is not permissable to descend
to the supposits either by a disjunctive sentence or by a categorical with
a disjunctive extreme, since such verbs make the terms following them
appellate their rationes, namely those according to which they were
imposed to signify.
[3.8.26] For example, Aristotle clearly spelled out the difference

between "The one approaching 1 know" and "I know the one approach-
ing" [De sophistici elenchi cap.xxiv 179b2-lO]: I posit the case that you
know your father perfectly well under the ratio according to which he is
called your father, but you do not know him under the ratio according
to which he is said to be approaching since, although he is approaching,
you do not know anything of the one approaching. Then in this case
"You know the one approaching" would be false, since it signifies
through appellation that you know the one approaching according to the
concept under which he is said to be approaching, which is false. Thus
"You know your father and your father is the one approaching; there-
fore, you know the one approaching" does not follow. But "The one
approaching you know" is true, since it follows by an expository syllog-
ism. Thus under whatever ratio you know Socrates, if Socrates is a man
or an animal and is approaching, it follows that <there is> a man you
know and an animal and a father approaching <whom> you know, since
<when the term> precedes the verb it does not appellate its ratio as it
does when it follows the verb.
[3.8.27] From this, it is clear that there is another great difference

between this kind of confusion and the preceding kinds, since in this
kind of confusion no consequence obtains from the confused term to
the determinate term (unless the reason for the confusion is taken
away), as for example "Every man is an animal; therefore, an animal is
every man" is not valid. But in this kind of confusion, "I know a triangle;
therefore, <there is> a triangle I know" is valid, since if I know a triangle,
that thing I know under the ratio according to which it is called a tri-
angle, and hence <there must be> a triangle 1 know. Indeed, although
there is a triangle I know and conversely, in this mode of non-distribu-
tive supposition the confused term does not follow from the deteminate
term as in the other cases; "A triangle 1 know; therefore, I know a
triangle" does not follow, but "Animal is every man; therefore, every
<man> is an animal" does indeed follow.
[3.8.28] There is still another important difference between this and the
other kinds of non-distributive confusion, namely in this mode of confu-
sion the distributed term follows from the confused term, changing the
order from before to after <the verb>, which is not so in the other modes.
"I know a triangle; therefore, any triangle 1 know" follows, since if 1
know a triangle, I then have the concept by which the name 'triangle'
was imposed. By means of that concept I conceive a triangle indif-
ferently, since 1 conceive anyone triangle no more than any other
triangle, and so every triangle 1 know. However the converse <conse-
quence> from the distributed term to the non-distributed term does not
follow; "Every triangle 1 know; therefore, I know a triangle" does not
follow, since it is possible that 1 have knowledge of every triangle under
the common ratio of figure or quality or being, not having the proper
ratio of triangle; hence 1 do not know a triangle.
[3.8.29] Still, it is apparent that the accusative objects of these verbs, if

they have a universal affirmative sign immediately preceding them, are
not taken distributively so that we can descend. Although an isoceles
triangle is a triangle, "I know every triangle; therefore, I know an
isoceles triangle" does not follow, since I might have a general non-
distributive XO concept of a triangle in my mind and not have the specific
or individual concepts. Then I would know every triangle and neverthe-
less I would hot know an isoceles triangle or an equilateral triangle.
[3.8.30] These matters, then, are laid down about the special condition
or construction of those verbs, and in particular about the mode of
confusion in their following accusative-objects. From the discussion we
can see how these accusative objects seem to partake of material suppo-
sition, since they appellate their concepts, although they do not supposit
for them; hence we cannot descend to the names of other concepts.
They appellate their concepts this way because we understand a thing
according to those concepts, but fire does not heat water by a concept,
nor a stone touch the ground <by a concept>.XI
[3.8.31] We should say the same for nominative terms following passive
verbs as we have said for accusative objects following these passive
verbs. For example, if 1 say "Any triangle is understood by Socrates" it
does not therefore follow that he understands an isoceles triangle, etc.
CHAPTER 4
THE SUPPOSITION OF RELATIVE TERMS
SECTION 4.1. THE MEANING OF 'RELATIVE TERM'
[4.1.11
Now that we have generally considered supposition and how
terms are taken in sentences, there remains a great difficulty
with the supposition and how relative terms are taken. Thus
we shall treat them in particular now.
[4.1.21 Chapter 4 deals with the supposition of relative terms and how
they should be taken in particular <cases>.X2
[4.1.31 Note well that here we do not mean by 'relative term' those
terms which we say fall under the category of Relation, such as 'father;
'son; 'master; 'slave; but rather what the grammarians call a 'relative
term; namely relative to a thing said before or recalled, such as the
pronouns 'other,' 'same,' 'such,' 'so much:
[4.1.41 Chapter 4 has nine sections: Section 4.1 gives the aim of the
chapter; Section 4.2 distinguishes and enumerates the kinds of relative
terms; Section 4.3 gives a general rule by which a relative refers to its
antecedent; Section 4.4 gives another rule, that a relative sup posits as its
antecedent does; Section 4.5 deals with the relatives 'this' and 'that';
Section 4.6 deals with the relative 'himself'; Section 4.7 deals with the
reflexive 'his own'; Section 4.8 deals with the relatives 'such,' 'so much,'
and the like; Section 4.9 is about the relatives 'other' and 'different:
[4.1.51 The first part <of [4.1.11> is sufficiently clear, unless it is necessary
to note what is said here about supposition and the way a term is taken:
here we mean more by the 'taking of a term' for something in a sentence
than the supposition of a term for something. It was said in Section 1.4
that strictly only a subject and a pr~dicate are said to supposit, accord-
ing to which an oblique term does not supposit, as 'man' in "Socrates
sees a man," since nothing is 'man' (hominem), although 'man' in this
148
4. THE SUPPOSITION OF RELATIVE TERMS 149
sentence is rightly taken for a man (homo) whom Socrates sees, and can
in this way be confused without distribution, or taken determinately, or
even taken personally and materially. It is true that we sometimes use
the name 'supposition' widely as synonymous with how a term is taken,
and we spoke widely rather than strictly in Chapter 3, as we shall do in
Chapter 4.
SECTION 4.2. THE KINDS OF RELATIVE TERMS
[4.2.1]
Relative terms are exceedingly diverse. Some are called 'sub-
stantial relatives; and others are called the relatives of
another category, as for example 'such' and 'so much' and so
on. Some substantial relatives are called 'identificatory rela-
tive-terms,' others 'differentiating <relative-terms>,' as for
instance 'other' and 'different: Some identificatory relative-
terms are said reciprocally, as 'itself or 'its own' and the like;
others are not reciprocal, as 'that thing' or 'the very thing'
and so on.
[4.2.2] Section 4.2 gives three divisions of relative-terms.
[4.2.3] With respect to the first, note that it is not necessary for a
substantial relative-term to have its antecedent be a term in the category
of Substance; rather, its antecedent can be of any category. For example,
"The wall is white and it is seen," "Whiteness is in the wall and it is a
color," "Something six-feet-tall sees itself," "My Master rides his horse."
We do not call something a substantial relative-term when it signifies
that for which it is taken or for what it refers to to be somehow or some-
where or some amount, and so on with other accidents, but rather only
that it is something. Accidental relative-terms signify <something> to be
somehow or some amount, etc.
[4.2.4] It is clear in the second division that an identificatory relative-

term is taken for that for which its antecedent is taken, and a differen-
tiating relative-term is not, but for another.
[4.2.5] Concerning the third division, it is clear that if an act signifies a

ISO TREATISE ON SUPPOSITION
transition, or a reversion into that from which <it comes>, or a reversion

into that by which it is, for example if we say that Socrates loves himself,
or is loved by himself, or is a lover of himself, or acquires <this> for
himself, this signifies that the act of love or of acquisition is Socrates's
and is of him insofar as he loves or acquires, or even insofar as it is of
him who loves or what he acquires; but it is not so for the other relative-
terms.
[4.2.6) Other divisions <of relative terms> can be given; for example,
sometimes they are given in the nominative and sometimes as oblique
terms, and they differ in these cases, as will be clear in Rule RT-S.
SECTION 4.3. REFERENCE TO THE ANTECEDENT
[4.3.1)
[Rule RT-l] An identificatory relative-term need not sup-
posit or stand in a sentence for everything for which its
antecedent suppa sits or stands; rather, the relative term only
refers to its antecedent for those of its supposits for which
the categorical in which the antecedent is given is verified.
Hence this is false: "Animal is man and it is an ass."
[4.3.2) It is immediately apparent from Rule RT-1 that a relative term

cannot be ampliated any more than its antecedent is ampliated. Thus if I
say "A man is running and he is disputing," the relative-term 'he' does
not supposit only for the present, just as 'man' does not <supposit only
for the present>, but <the relative term> can also be more restricted, as
when I say "A man is running and he is white": the antecedent 'man'
supposits for all men indifferently, although indefinitely, but the term
'he' does not thus supposit for all, but only for those for which the
sentence "A man is running" is true. If the sentence is not true for any
men, then the relative term will suppa sit for nothing, and so the
conjunctive sentence given above will be false (since each conjunct is
false), even if all men were white.
SECTION 4.4. SUPPOSITION OF THE ANTECEDENT
[4.4.1)
[Rule RT-2) An identificatory relative-term supposits or is
taken in a sentence as its antecedent <supposits or is taken>,

namely materially if materially, personally if personally, dis-
tributively if distributively, determinately if determinately,
merely confused if merely confused, with the exception of
the conditions given in Rule RT -1.
[4.4.21 The exceptions are clear, since perhaps a relative-term supposits

or is taken for nothing in a sentence, although the antecedent supposits
for many things - that is, because the sentence with the antecedent is
true for none of them, as for example when I say "A man is a stone and
he runs."
[4.4.31 It is asked whether 'if supposits materially as 'mirror' does in

"Mirror is a noun and it has two syllables." It seems that it does not,
since it does not supposit for itself nor for something similar to it 111
voce. HJ
[4.4.41 I say that it supposits materially relatively, and so if it were to

supposit for itself or for something similar to it ill voce, then it would
rightly supposit for itself, but not relatively, but when it supposits for the
antecedent or for something similar to it ill voce, then it supposits for
something materially relatively. If it were to supposit personally, then it
would necessarily supposit for the ultimate significates of its antecedent.
[4.4.51 There is also a doubt, if we posit that only Socrates is running

and that he is white and we say "A man is running and he is white,"
whether this relative-term supposits commonly and indeterminately or
discretely and singularly; it seems that it supposits discretely, since it
does so for only one, namely Socrates. But the opposite seems plausible,
since in the former case it would not supposit as does the antecedent.
[4.4.61 I say that 'he' supposits indeterminately and not discretely, since
it is not found in the given sentence as a discrete term but as a common
term, and so apt to supposit for many, if many men were to run.
[4.4.71 Again, I posit that only two men run, namely Socrates and Plato,
and that Socrates is white and Plato is black, and we say "A man is
running and everyone such is white": then it is unclear how 'one'
supposits or whether the sentence is true.
[4.4.8] I say that it seems to me that it is silly and useless to add a

universal or a particular sign to a relative term taken relatively, since its
nature is that it is to be taken as is its antecedent. Hence the sign does
not help it to name its supposition, just as neither does a preceding
negation, as when I say "An animal is running and no man is that." Here
the relative-term 'that' is not distributed, but is as indeterminate as its
antecedent. Thus it appears to me to be neither grammatical nor helpful
to say' ... and everyone such is white: even though such signs can be
added to a differentiating relative-term or to an accidental relative-term
quite well, since they supposit for the same thing as their accidents, as
will be clear in Section 4.9.
[4.4.9] If anyone wanted to hold that a locution of this sort were gram-
matical and helpful, then they should say that 'one' is distributed and
that Rule RT-2 should be understood where a relative is taken
accidentally without the addition of a new sign.
SECTION 4.5. 'THIS' AND 'THAT'
[4.5.1]
Now let us consider different relative terms more closely. It
seems that the words 'this' and 'that' are not always taken
relatively, but are sometimes taken demonstratively, as when
in pointing at a man I say "That man is running." This is not
relevant to Section 4.5. Sometimes, though, it is taken an-
tecedently to the relative term 'who' or 'what,' as when I say
"That man is running who was sitting down before," and then
we have the following rule:
[Rule RT-3] 'That' <taken antecedently> is taken only as a
particular sign and 'who' is taken in the place of the
relative 'that' by marking out a conjunction.
Thus the sentence is resolved into the whole conjunctive
sentence "Some man is running and that <man> was sitting
down before."
[4.5.2] Note that the terms 'who' and 'he' or 'this,' and 'what' and 'him'
or 'that,' are typically used in the same sentence, providing subjects for
the two verbs, one to one and the other to the other, and this cus-
tomarily happens in two ways: (i) each of those terms precedes the first
and the second verb, as in saying "He who sleeps dreams"; (ii) some-
times one of the terms precedes the first verb and the other precedes the
second verb, as in "He sleeps who dreams."
[4.5.3) There is this difference between the two sentences, that in (i) the
sentence is categorical, the subject of which is made up of the two terms
and the first verb, so that when I say "He who sleeps dreams" the whole
'he who sleeps' is the subject composed of a determinable and a deter-
mination, for the word 'he' is put in as a determinable, or at least as an
understood subject and expressly as when I say "That man who sleeps
dreams," and the residual 'who sleeps' is found as the determination.
The verb 'dreams' is put as the copula and predicate, and in such a
sentence the term 'he' is not taken relatively, but rather antecedently,
and means no more than the term 'someone.'
[4.5.4) But in (ii) the sentence is hypothetical, and equivalent to a

conjunctive sentence, as was said, and sometimes the term 'he' precedes
and 'who' follows, as in "He sleeps who dreams," and sometimes the
converse, as in "Who dreams, he sleeps," or even distributively, as in
"Whoever sleeps, he dreams," and they are equivalent sentences,
whether 'who' comes first or 'he' does; for whichever of them comes
first, it is not taken relatively, but is the same as the name 'someone,'
and whichever of them comes next is taken relatively.
[4.5.5) It should be noted that these terms 'this' and 'that,' when they are
taken relatively, are equivalent to 'he.' It is rare that the term 'he' is taken
only relatively. In this way we often use 'same' as synonymous with 'it' so
that it is equivalent to say "An animal runs and it is a man" as to say "An
animal runs and the same is a man."
SECTION 4.6. 'HIMSELF'
[4.6.1)
[Rule RT-4) We should remark that the relative term 'him-
self (a) is always found in the same categorical sentence as
its antecedent, for example, "Socrates amuses himself'; (b) if
taken distributively, we cannot descend to some determinate

supposit of it, with the rest remaining the same, but we must
descend singly. "Every man amuses himself; therefore, every
man amuses Socrates" does not follow, but "<Every man
amuses himself>; therefore, Socrates amuses Socrates and
Plato amuses Plato (and so on for the rest)" does follow.
[4.6.2] Against Rule RT-4(a) it is argued that if I say "Socrates runs and
he amuses himself' then 'Socrates' is the antecedent and is not found in
the same categorical as the relative term 'himself.'
[4.6.3] I reply that 'himself does not refer immediately to 'Socrates' but
only by the mediation of another relative 'he; which is immediately
antecedent to 'himself and found along with it in the same categorical.
[4.6.4] With respect to Rule RT-4(b), it seems that it is not strictly a

property of the relative-term 'himself but rather sometimes is shared by
the relative 'his own' or even the relative 'his' put in the place of 'his
own.' For example, if I say "Every man sees his own horse" or "Every
man sees his horse" it does not follow that therefore every man sees
Socrates's horse, but it does follow that Socrates sees Socrates's horse
and Plato sees Plato's and Albert Albert's, and so on for the rest.
SECTION 4.7. 'HIS OWN'
[4.7.11
[Rule RT-5] The relative-term 'his own' does not refer to
its antecedent by a resolution <of the antecedent> into the
nominative, but rather into the genitive case. For example, if
I say "Socrates drinks and his own horse eats" then it does
not therefore follow that Socrates eats, hut that the horse of
Socrates eats. Hence the relative 'his own' is the same as the
relative 'his; and it is the same to say "Socrates rides his own
horse" and "Socrates rides his horse."
[4.7.21 We can note that 'his own' is taken in many ways: (i) it marks our
possession, and then it refers to its antecedent as the possessor; (ii)
sometimes it notes an effect, as in 'Aristotle and his book,' and then it
refers to its antecedent as an efficient cause; (iii) when it is added to a
subject in the category of Relation, then it refers to its antecedent as a

relation in the converse disposition, as when I say "Socrates and his own
son" it is noted that Socrates is his father, and when I say "Socrates and
his slave" it is noted that Socrates is his master, and so on for the others.
SECTION 4.8. 'SUCH' AND 'SO MUCH'
[4.8.1]
[Rule RT-6] The relative terms 'such' and 'so much' require
antecedent terms in the category of Quality and Quantity, as
in "Socrates is somehow [or some amount] and Plato is such
[or so much]." Or, in particular, "Socrates is white [or a six-
foot-tall] and Plato is such [or so much]." And it is clear that
they do not need to supposit for the same things their antece-
dents supposit for, but it is sufficient that they supposit for
similar or equal things.
[4.8.2] Note about the relation of these terms to each other, 'such' and
'as' and 'so much' and 'as much,' that they are related just as 'this' and
'that' <were said to be related> in Section 4.2. Each is put antecedently to
the rest and the rest is relative to it, as in saying "As is Socrates, such is
Plato" or in saying "Socrates is such as is Plato," and each sentence is
the same as the conjunctive sentence "Socrates is somehow and Plato is
such." But sometimes 'as' and 'such' are found in the same categorical,
as in saying "Such as Plato is runs," and then the whole 'such as Plato is'
is the subject.
[4.8.3] Similar things should be said for 'so much' and 'as much,' 'when'
and 'then,' and 'where' and 'there,' and the like.
SECTION 4.9. DIFFERENTIATING RELATIVE-TERMS
[4.9.1]
'Other,' 'another,' 'different,' and 'diverse' are usually taken
to be differentiating relative-terms. But it is clear to me that
these are not relative terms as we understand 'relative-terms'
in this chapter, but they are relatives, i.e. terms of the cate-
gory of Relation, just as <the terms> 'similar' and 'dissimilar:

'equal' and 'unequal.' So too for the term 'same,' unless it is
taken in the place of the relative-term 'this'; anything which is
same is the same as something and anything which is diverse
or other is diverse from or other than something. Neverthe-
less, since names of this sort have special difficulties over and
above other names of the category of Relation, we want to
say something about them.
[Rule RT-7] (a) The foundation of the term 'same' and the
terminus which it is said ought to supposit or be taken for
the same. For example, if Socrates is the same as a man,
then the terms 'Socrates' and 'man' ought to suppa sit for
the same. But the foundation of the terms 'other' or
'different' or 'diverse' or 'another' should not stand for and
be verified of the same, but different things. (b) A sent-
ence in which a relative-term of this sort is found must be
analyzed by a conjunction of three categoricals, namely
two affirmative existentials with the foundation and the
terminus of the relative-term, and the third a negative in
which the terminus is denied of the foundation, or con-
versely. For example, if I say "B is other than A" this is
analyzed as "A exists, and B exists, and B is not A."
[4.9.2] There are three claims made in this passage.
[4.9.3] The claim is how the names 'other: 'another,' etc. should be
called relatives, and this is clear. Thus if I say "A man is running and
another is white" then it is not necessary that the term 'another' have
'man' as its antecedent or even as its foundation or its terminus, but
rather this term 'another' is a substantive adjective of neuter gender;
hence the sentence is equivalent to "A man is running and another thing
is white" and so the term 'thing' is the foundation of the term 'another.'
It is not necessary that here 'another' is said with respect to 'man,' unless
'and other from him' or 'other than man' is added, just as we might add
'other than ass' or '<other> than a hat.'
[4.9.41 Note that the terms 'other' and 'another' and 'different' and
'diverse' dialecticians and sophists often use as synonyms, and so indeed
do we speak of them, although they are sometimes restrictedly attri-

buted to diverse intentions, as is clear in the discussion of 'other' and
'another' in Porphyry [Isagoge cap.iii], and in Metaphysics X [cap.iii
1054b7-8] this is clear for 'different' and 'diverse; since the verb 'to
differ' is understood to be the same as '<it is> different:
[4.9.5] About Rule RT-7(a) it should be known that, since the aforesaid
terms are adjectives, they require substantives governing them; indeed, if
the term 'another' is taken as a substantive, it is till resolved into a
substantive and an adjective, as 'another' is 'another thing: I call the sub-
stantive the foundation in the sentence; I call the oblique term, whether
ablative or dative, which it rules such a relative term and the terminus.
For example, in "Man is the same as animal" or "the same with animal"
the term 'man' is put as the foundation and 'animal' as the terminus. If I
say "Man is other than stone," then 'man' is the foundation and 'stone' is
the terminus. Once these things are pointed out, Rule RT-7(a) is suffi-
ciently clear.
[4.9.6] By Rule RT-7(b) it is clear that the aforesaid differentiating

relative-terms imply in themselves some negations, according to which
the names following them as termini of those relatives are distributed.
Thus "It is other than man; therefore, it is other than Socrates" follows,
or "It is different from man; therefore, it is different from Socrates."
Nevertheless, if the aforesaid termini precede those relative-terms, then
they are not distributed by them, as when I say "From man Socrates dif-
fers" or "From man Socrates is other," 'man' is not distributed. How-
ever, "From Plato Socrates is other; therefore, from a man Socrates is
other" follows, since 'man' here stands determinately.
[4.9.7] But there is a doubt here about whether 'animal' in "Other than
man is animal" is distributed.
[4.9.8] I say that it is not, since the negation implicit in 'other' does not
negate in itself, but only in an infinitizing way, and so does not fall under
the copula of the whole sentence, nor under the predicate of the whole
sentence; hence the sentence is affirmative, although it implies a negative.
And so "A non-white man is an animal" is affirmative, which neverthe-
less implies a negation or a negative, namely that a man is not white.
[4.9.9] If you ask how the sentence in [4.9.7] should be analyzed, I say
that the first adjective should be resolved into a substantive, so that
"Other than man is animal" is "A thing other than a man is an animal,"
and then it is analyzed that there is a thing and that there is a man and
that a thing which is not a man is an animal. It is clear that 'animal' here
stands determinately. Even if I say "Another animal than man is run-
ning" the term 'animal' is not then confused, since such a relative does not
naturally confuse, unless it is the terminus and not its foundation. Hence
it is equivalent to say 'another animal than man' and 'an animal other
than man: These are similar to privative terms: if I say "A man is
deprived of vision" then 'vision,' but not 'man,' is distributed, since it is
analyzed as follows: i.e. "There is a man not having vision," and thus this
word is not taken negatively but indefinitely.
CHAPTER 5
APPELLA TION
SECTION 5.1. REVIEW
[5.1.1]
Now we turn to appellation. Some terms are appelative and
others are not. Substantial nominative terms or terms
connoting nothing beyond that for which they supposit are
not strictly appellative, but every term connoting something
other than that for which it supposits we call 'appellative,'
and it appellates that which it connotes as adjacent to that for
which it supposits. For example, 'white' appellates whiteness
as adjacent to a thing for which the term 'white' is naturally
apt to supposit. These and other points were mentioned in
Chapter l.
[5.1.2] Chapter 5 deals with appeliation,H4 and has four sections. Section
5.1 gives the intention of Chapter 5 and repeats some things said in
Section 1.4; Section 5.2 is about the differences in appellation on the
part of the subject and on the part of the predicate; Section 5.3 is about
the difference between appellation and the thing signified or consigni-
fied and the ratio according to which there is signification; Section 5.4 is
about the appellation of certain terms in particular.
[5.1.3] This first passage in [5.1.1] is clear if you carefully consider what
was said in Section 1.4 and the discussion there.
SECTION 5.2. THE APPELLATION OF SUBJECT AND PREDICATE
[5.2.1]
[Rule App-l] An appellative term always appellates its form,
whether it 1:-' found on the side of the subject or put before
the verb, or on the side of the predicate or put after the verb;
159
but they appellate it differently <depending on> whether they

are before or after <the verb>.
[Rule App-2] If it is put after the verb, then it appellates its
form precisely at the time of the verb.
[Rule App-3] If it is put before the verb, then (a) if the verb is
present-time and the predicate is not ampliative then it
appellates its form precisely for the present time; (b) if the
verb were of another time or the predicate were ampliative
then the term put before the verb would appellate its form
indifferently, quasi-disjunctively, for the present time and the
time of the verb or the time to which the predicate ampliates
(if it is ampliative).
[5.2.2] There are three rules in [5.2.1], and to understand any of them
we need to know what we should understand by 'the form of the
appellative term' in a sentence.
[5.2.3] Therefore, I say that we usually understand conventionally by

the 'matter' of a term that for which the term is apt to supposit. Thus just
as Aristotle says of matter in Metaphysics IX [cap.viii 1049b-6], where
he says that things whose matter is one are called 'numerically one,' that
is, the predicate 'numerically one' is verified of those terms whose
matter is one, i.e. they supposit for the same thing, we therefore say truly
that man and animal are numerically one.
[5.2.4] By the 'form' of a term we customarily understand whatever the

term appellates, whether it is substance or accident, or matter or form,
or a substance composed out of many. Now the term 'feudal-lord'
supposits for a man, and so 'man' is called its matter; it appellates the
house and fields and sheep and other things which he possesses as
adjacent to him as possessions, and so such things which are possessed
are called the 'form' of the term 'feudal-lord.'
[5.2.5] In this way, therefore, I say that an appellative term found in a

sentence appellates its form, i.e. those things which it connotes or of
which it is appellative, and it appellates them as adjacent to something
else, either past or present or future or possible, i.e. the thing for which
that term supposits or for which a substantive nominative term con-
5. APPELLATION 161
strued with it would supposit for if it were the subject or predicate of the
sentence.
[5.2.6] Hence "A just devil is running" is false, since the subject of the
sentence does not supposit for anything. Nevertheless the term 'just'
appellates justice as adjacent to the devil for which the term 'devil'
would supposit if it were the subject or the predicate of a sentence, and
so the sentence is false since the subject does not supposit for anything,
due to the nonxistence of the appellated adjacent thing.
[5.2.7] Rule App-2 is clear, since if we posit that in the present Socrates
is white and he will never be white again in the future, then the sentence
saying that Socrates will be white is false. The reason for this is that
'white' in the given sentence does not appellate whiteness at the present
time but only in the future, and we posited that whiteness will never be
adjacent to Socrates in the future, although it is adjacent to him at
present. It would be otherwise if we were to say "A white thing will be
Socrates," since the sentence would be true if whiteness were to be adja-
cent now to that thing which will be Socrates, even though <whiteness>
would never be adjacent to it in the future.
[5.2.8] Rule App-3(b) is clarified by this; if a term is placed before the

verb it appellates its form, with a disjunction of the present or future
time and the time ofthe verb.
[5.2.9] This is also so if the predicate were of past or future time even if
the verb is present-time, as for example when I say "A white man is
going-to-dispute": this is true, whether whiteness now is adjacent to a
man who is going to dispute and never afterwards will be adjacent to
him, or it will be so adjacent and is not now.
[5.2.10] An objection is raised against Rule App-2: if we were to say

that "Socrates is going-to-dispute," then the term 'going-to-dispute'
appellates the act of disputing at a future time and not at the present
time, although the verb is present-time.
[5.2.11] I reply that 85 the term 'going-to-dispute' appellates a future time

with the act of disputing by its manner of signifying (modus significandl);
nor can an appellation of this sort be removed by its connection with a
verb of another time, but when it thus appellates that future time, it
ought to be future with respect to the time connoted by the verb. It need
not be future with respect to the time at which we form the sentence,
which we treat in our locution as the present time, unless the verb is of
the present time. If the verb is present-time, then the appellated future
<time> must be future with respect to the present time.
[5.2.12] For example, if I say "Aristotle is going-to-dispute," then the

word 'going-to-dispute' appellates a time which in the present time is
future. If I say "Aristotle was going-to-be-generated" then the term
'going-to-be-generated' appellates a time which is not future, but which
was future at the time which is connoted by the verb 'was,' which time
did indeed precede the generation of Aristotle. And it is clear that the
terms 'going-to-dispute' and 'going-to-be-generated' still appellate their
forms, namely a future time in the given sentences for the time of the
verb, such that in the time of the verb that is or was future, and ought
not to be present in the time of the verb. Thus if I say "Antichrist will be
generated," <the term 'generated'> appellates a past time which is not now
past but which in the future time will be past, namely in the time
desgnated by the verb 'will be:
[5.2.13] Again, note that Rule App-2 and Rule App-3(b) should be
understood for verbs and copulas which are not ampliative, since if the
verb or the copula of the sentence were to be ampliative, as for example
the verb 'can' ur the copula in "The Antichrist is possibly white," then it
silould not h'~ said that 'white' appellates its form precisely at the present
time, but raLher that it appellates it indifferently at the time to which it is
ampliated.
SECTION 5.3. APPELLATION AND SIGNIFICATION
[5.3.1]
There is a great difference with respect to appellation
between verbs signifying an act of the intellective soul, as the
verbs 'to comprehend,' 'to understand,' 'to signify,' 'to sup-
pose,' 'to permit,' and this sort, and other verbs such as 'to
cut,' 'to move,' and the like.
[Rule App-4] With respect to such verbs not signifying this
5. APPELLATION 163
sort of act of the soul, terms do not appellate the rationes by

which they ultimately signify.
[Rule App-5J With respect to the verbs which do signify such
acts of the soul, (a) if the terms follow those verbs and are
construed with them as terminating their actions, they appel-
late the rationes according to which they signify what they
signify; (b) if those terms precede the verbs, then they do not
thus appellate the rationes.
[5.3.2J Note that this subject was treated carefully in Sophismata cap.iv,
and many things were said about it in the discussion of Rule NDC-4 in
[3.8.24J-[3.8.31 J.X6
[5.3.3J Hence this alone is required here, that 1 now briefly state that it
is very different to say "I know the one approaching" and "The one
approaching 1 know," since whenever you know your father under many
rationes, as those according to which he is called a man and an animal
and a father and great, etc. and you do not know him under the ratio
from which the term 'approaching' is taken, then (positing that he is in
fact approaching) "You know the one approaching" is false, since the
sense according to many appellations of the ratio is that you know him
under the ratio of the one approaching. Nevertheless, "The one
approaching you know" is true, since it is not necessary for its truth that
you know him under that ratio, but it is sufficient that you know him
under some <ratio>, whatever it may be.
[5.3.4J Hence we conclude that 1 do not infer one term from another
found after such verbs, no matter how much they may supposit for the
same, unless the terms are synonymous or the ratio of the following
terms is included in the ratio of the antecedent term in such a way that it
is impossible that the ratio of the antecedent term be in the soul without
the ratio of the following term. Thus "I know a triangle; therefore, 1 know
a;. isoceles triangle" does not follow, but "<1 know a triangle>; therefore,
I know a triangle" or "<I know a triangle>; therefore, 1 know every
triangle" do correctly follow, since the ratio corresponding to the
expression 'every triangle' cannot be <in the soul> without the ratio
corresponding to 'triangle:
[5.3.5J But one term can be inferred from another as found before such
verbs, assuming that they supposit for the same, as in "Your father you
know, and he is the one approaching; therefore, the one approaching
you know."
[5.3.6] In sentences with other verbs one term can be inferred from
another whether they are found before or after <the verb>, assuming
that they supposit for the same. This follows: "A stone falls on the one
approaching, and he is white or is your father; therefore, a stone falls on
something white or on your father."
[5.3.7] Yet a major doubt is raised against these claims, namely that if
the foregoing were true, then it would follow that it would be impossible
for you to see a man or a horse, which seems ridiculous. The consequ-
ence is clear, since the ratio or the concept according to which the term
'man' or 'horse' is imposed is a common or universal ratio, and hence
that you see a horse signifies that you see it according to a universal
concept, which is impossible, since the senses only sense what is
singular.
[5.3.8] Because of this doubt some people deny the aforementioned

appellations of rationes.
[5.3.9] Nevertheless this ought not to be denied, if Aristotle and others

who speak normally are to say things which are true.
[5.3.10] We say that Averroes did not believe the First Principle to be
three and one, and nevertheless he believed the First Principle to be
God; but as taken before the verb, we must concede "The three and one
Averroes believed to be God." This is clear by an expository syllogism,
since the First Principle he believed to be God and that First Principle is
three and one; therefore, the three and one he believed to be God.
[5.3.11] And it is also clear that such an argument with the term as
found after the verb would not be valid, that is, "The First Principle he
believed to be God, and God is three and one." It is clear that this does
not follow, since the premisses are true and the conclusion is false, and
nobody can explain why it should be valid when the term is before the
verb but not when it is after the verb except by the appellatiun of the
rationes in the succeeding and not the preceding.
5. APPELLATION 165
[5.3.12J Aristotle clearly speaks in a similar fashion in De sophistici

elenchi [cap.xxiv 179b2-lOJ; he denies the consequence "You know
Coriscus, and Coriscus is the one approaching; therefore, you know the
one approaching" and nevertheless when the term is before the verb a
consequence such as "Coriscus you know, and Coriscus is approaching;
therefore, the one approaching you know" cannot be denied, since it is
truly an expository syllogism. And nobody can rightly say why these are
related in this way except for the appellation of rationes in the latter
case and not in the former.
[5.3.13J Hence the doubt raised in [5.3.7J should be replied to as fol-

lows: "I see a man" is not a strict and true locution, but such a sentence
is often accepted as taking the place of a sentence such as "I see this
man" or in the place of "A man 1 see." Nevertheless, we agree to use the
improper sentence, since from the visual notion of the singular there
arises immediately an intellectual notion of the universal. Thus by seeing
this man we immediately understand man, and so we believe ourselves
to judge that man is an animal by sight, which is not true.
[5.3.14J Furthermore, note that aside from this doubt some people say
that the predicate appellates its form such that no sentence of the past
time or of the future is true unless it had or will have something of the
present which is true, if it is formed, keeping the whole predicate the
same or the whole thing being similarly disposed, which is not thus
necessary for the subject. Hence they do not call a de possibili sentence
true unless it has a corresponding true assertoric sentence which pre-
serves the entire predicate; thus they call these false and impossible:
"I shall lecture tomorrow," "I lectured yestereday," since "I am now
lecturing tomorrow" and "1 am now lecturing yesterday" are impossible.
[5.3.15J 1 have discussed these in Sophismata cap.iv, and the discussion

there should be read thoroughly as a complement to this chapter.
[5.3.16J Nevertheless, it seems to me true that what I said in the solu-

tion to Sophism 6 <of Sophismata cap.iv> should be corrected, namely
that 1 saw Peter and Robert, or even that 1 will see or can see any star;
for perhaps those sentences should not be conceded because they
appellate their rationes, in that 'see' signifies an act of the intellective
soul. Hence in the place of those sentences 1 put others, which do the
job for which these were put forward: in the place of "I saw Peter and
Robert" I put the sentence "Peter and Robert were in this house." I take
this sentence to be true if one were in it at one time and the other at
another, even though they were never both there together. Similarly, in
the place of the sentence "I see every star" or "I can see every star," I put
the sentence 'The Sun will be or can be in any of the twelve signs of the
Zodiac," which 1 think should be conceded.
[5.3.17] 1 do not think the discussion in that part <of the Sophismata>
needs to be altered, since I did not seek to verify the examples, as is said
in Prior Analytics [Il.xxiv 69aI4-17]. But now that I have corrected the
text, if you want to seek to verify the examples, do so as was done in the
Sophismata.
SECTION 5.4. APPELLATION OF PARTICULAR TERMS
[5.4.1]
We take up next the appellation of certain names in par-
ticular.
[Rule App-6] I say that an oblique substantive term ap-
pellates that which its nominative term would supposit as
adjacent to that which the nominative term governing it
supposits.
[Rule App-7J The terms 'mine' and 'yours' appellate me and
you as adjacent to those things for which their subjects
supposit.
[Rule App-8] Similarly, 'ours' and 'your' appelate us and you,
and the term 'his own' appellates that for which its antece-
dent supposits as adjacent to that for which the substantive
of the term 'his own' supposits.
[Rule App-9] The adjacence is changed in all the corre-
sponding ways as said for 'his own' in Section 4.7.
[5.4.2] This last passage in Chapter 5 gives four rules.
[5.4.3] The first three are about different terms in particular, and the
fourth is a general rule for all these terms.
5. APPELLATION 167
[5AAI Rule App-6 deals with oblique terms. Since they are neither
subjects nor predicates of sentences, they cannot strictly supposit.
Hence they appellate the things which they signify, as well as those
things which their nominative term would appellate, if the nominatives
were to appellate, as those things for which they sup posit.
[5A.51 For example, if I say 'the ass of-Socrates,' the term 'of-Socrates'
appellates Socrates as adjacent to his ass as a possessor. If I say "That is
the book of-Aristotle," the term 'of-Aristotle' appellates Aristotle as
adjacent to the book as a maker, namely as he who made it or he who
made the first copy from which all others have been made. If I say
"Socrates is the father of-Plato," then the term 'of-Plato' appellates Plato
as adjacent to his father Socrates as a son. And so forth for the rest.
[5A.6J In the same way, if I say "Socrates sees Plato," then the term
'Plato' appellates Plato as adjacent to Socrates's sight as seeing, and so
on for the rest.
[5 A. 7J I speak of Rule App- 7 in [5 A.l J in the same way: if I say 'my ass,'
the term 'my' appellates me as adjacent to my ass as a possessor; and if I
say 'my father,' the term 'my' appellates me as his son. And so in the
same way for 'your,' and 'our,' and this can be seen in Rule App-8,
insofar as it can be seen to support Rule App-9.
CHAPTER 6
AMPLIATION AND RESTRICTION
SECTION 6.1. THE STATUS OF A TERM
[6.1.1]
Finally, something should be said about the ampliation and
restriction of supposition and appellation. Thus it is neces-
sary to examine first the status according to which a term is
called neither ampliated nor restricted, in respect of which
status it can sometimes be ampliated and sometimes re-
stricted. That status can be given when the term precisely
supposits for and appellates all its significates of the present
time. For example, when I say "A man is running" or "Every
man is running," 'man' sup posits indifferently for all the men
who exist at the present time, and for nothing else.
[6.1.2] Chapter 6 deals with ampliation and restriction and has six
sections. K7 Section 6.1 gives the status with respect to which ampliation
and restriction are said; Section 6.2 deals with the restriction of that
status; Section 6.4 deals with the alienation of the status; Section 6.S
with the alienation of supposition or appellation; and Section 6.6 with
the cancellation of the supposition.
[6.1.3] The passage in [6.1.1] makes two points, the first of which gives
the aim of Chapter 6 and begins it, which is clear.
[6.1.4] The second gives the status <according to which a term is called
neither ampliated nor restricted>; this is clear enough, and I do not care
if others wish to specify a different status.
SECTION 6.2. AMPLIATION

[6.2.1]
Therefore it should be known that a term IS ampliated

beyond that status in many ways.
168
6. AMPLIATION AND RESTRICTION 169
[Rule Amp-I) If it is found before a past-time verb in its

categorical, then it is ampliated to stand for past things,
whether in supposition or in appellation, along with things of
the present time. For example, if I say "A man was running,"
the term 'man' supposits indifferently for all the men who are
and who were.
[Rule Amp-2) In the same way (with the appropriate
changes), a term found before a future-time verb is ampliated
to stand for future things.
[Rule Amp-3) A term found before the verb 'can' or before
the copula of a divided sentence de possibili is ampliated to
stand for possible things, though they do not and did not
exist. Thus "A golden mountain can be just as large as Mont
Ventoux" is true.
[Rule Amp-4) The same should be said for a term found
before the copula of a divided sentence de necessario, since
such is equivalent to one de possibili.
[Rule Amp-5) A term is ampliated to the past, the future, and
the possible if it is construed with a verb signifying an act of
the intellective soul, whether the term is found before or
after the verb. Thus "A man I understand" is true, if I under-
stand Aristotle or the Antichrist; and "I understand a rose" is
true even if no roses exist.
[Rule Amp-6) If verbal names or participles become the
predicates of sentences, they act just like the aforementioned
verbs to ampliate their subjects. Hence something is gener-
able or corruptible, or generated or corrupted, opinable or
inopinable, opined or unopined, which does not exist.
[6.2.2) This passage contains six rules, and they are all sufficiently clear.
[6.2.3) It follows from [6.2.1) that such sentences are analyzed by hypo-
theticals with a disjunctive subject. For example, "B will run" as "What is
B or will be B will run"; similarly, "B is dead" as "What is or was B is
dead"; similarly, "A can run" as "What is or can be A can run"; similarly,
"The one creating is of necessity God" as "What is or can be creating is
of necessity God"; and similarly "A is understood" as "What is or was or

can be A is understood."
[6.2.5] Hence such sentences should be converted into sentences with

disjunctive predicates, as for example "A man is dead; therefore, a dead
thing is or was man"; similarly, "Socrates is going to dispute; therefore,
something going to dispute is or will be Socrates"; similarly, "Socrates
can be white; therefore, something white is or can be Socrates" (and
here it suffices to say "therefore, something white can be Socrates");
similarly, "Everything which is or can be creating of necessity is God;
therefore, what of necessity is God is or can be creating," or even
"therefore, God can be creating."
[6.2.6] Just as a universal affirmative is not converted into a universal

but into a particular, a de necessaria is not converted into a de neces-
saria but into a de passibili, since 'necessary' and 'possible' are just like
'universal' and 'particular: Similarly, "A golden mountain is understood;
therefore, what is understood is or was or can be a golden mountain."
[6.2.7] Similarly too it is said that the aforementioned names or partici-

ples, if they are taken as adjectives, ampliate in their own way the
substantive to which they are added, and the subjects of these sentences
are conjoined from the substantive and the adjective, as said in [3.7.8].
Thus "Ari~totle is a dead man" should be conceded, even though
Aristotle is not a man <since he is dead>; so too should "A horse is a
dead horse," but not "A dead horse is a horse."
SECTION 6.3. RESTRICTION
[6.3.1]
A term is further restricted in that status in many ways.
[Rule Res-I] By a determination appropriate for it but not
for all its supposits, as for example in "A white man is runn-
ing" 'man' is not taken for all present men but only the white
ones; or even in "The ass of-Socrates is running" 'ass' is not
taken for all asses but only for those which are Socrates's.
[Rule Res-2] An adjective is restricted by its substantive if it is
appropriate to other things to which the substantive is not

appropriate. For example, in "A white man is running" the
term 'white' is not taken for all white things, but only for the
white things which are men.
[6.3.2] This passage gives two obvious rules, about which it should be
noted that if the supposition of a term were to be ampliated to the future
or the past or the possible, nevertheless both rules hold for restricting
the ampliation with respect to the supposits. For example, if I say
"Every white man will be good," then the term 'man' is held for present
as well as future things, but still not for all of them, but only for those
which are or will be white. Nor does the term 'white' supposit for all
white things, <but only for those which are or will be men>.
SECTION 6.4. ALIENATION
[6.4.1]
Sometimes a term does not keep the aforementioned status,
and it is neither ampliated nor restricted, but alienated, such
that the term can no longer sup posit or be held for the
present things for which it supposits or holds for in its given
status, but only for future or past things, unless it happens
that those future or past things are also present; this is how
terms following past-time or future-time verbs supposit or
appellate.
[6.4.2] The passage will be clear if we give an example. Now if I say "A
horse was white," 'white' supposits for only what was white in the past.
Let us posit that all horses are white at the present time, and that no
horses were ever white in the past. Then the term 'white' does not
supposit for any horse, or rather, it does not supposit for anything,
unless there was something <else> white in the past, even if all things are
white in the present. Nevertheless, it is surely true that if a horse HH which
is present existed in the past and was white, then the predicate 'white'
would supposit for it, but not by the reason or for the reason that it is
white, but by and for the reason that it was white in the past. And so in a
similar manner for the future.
[6.4.3] Similarly, in the given sentence the term 'white' appellates white-
ness only as adjacent to that which was <white>; if it is still adjacent now
to what it was adjacent to then, it will be adjacent with respect to the
verification of the sentence. Thus if Socrates was white yesterday and is
still white today, the sentence "Socrates was white" is true, but not
because whiteness is now adjacent to him but rather only because it was
adjacent to him. This would not be so for the subject, if we were to say
"A white thing will be Socrates," positing that Socrates will be; the
sentence is true if whiteness is now adjacent to Socrates, even'if it will
never again be adjacent to him.
SECTION 6.5. THE ALIENATION OF SUPPOSITION
[6.5.1]
Supposition or appellation is sometimes alienated, so that an
ampliated term is not held for its significates but for others.
This comes about by an infinitizing negation, as when I say
"Man is a non-ass": the term 'ass' is prevented from standing
for asses and is forced by the negation to stand for every-
thing else indifferently. Thus if I say "A non-white man is
running" the term 'white' does not appellate whiteness as
adjacent to men, but rather as it is not adjacent to them.
[6.5.2] This passage is sufficiently explained; this is what it is for a nega-

tion to be infinitizing, namely that it make a term hold for all the things
for which it would not hold without it and does not hold for the things
for which it would hold.
SECTION 6.6. CANCELLED SUPPOSITION

[6.6.1]
Sometimes the entire supposition of a term is cancelled, and
this happens in two ways. (1) By the addition of a positive
term attributively (indistans) to a term in the subject or pre-
dicate where one of the terms is inappropriate to the other.
For example, in saying "A man able-to-whinny is running,"
the subject does not suppa sit for anything. (2) By a privative
combination of terms, one of which neces~.,rily is appropriate
to the other. For example, in saying "A non-risible man is
running," the subject does not suppa sit for anything.
[6.6.2] This last passage is clear. Nevertheless, it should be known that

in the sentence "A man able-to-whinny is running" that the names 'man'
and 'able-to-whinny' are taken significatively, since each establishes in
us a concept of their ultimate significates, such that each, with the other
removed, would supposit for its ultimate significates. But 'able-to-
whinny' cancels the supposition for men, since no men are able to
whinny, and 'man' cancels the supposition for things able to whinny,
since nothing able to whinny is a man.
JEAN BURIDAN
TRACTATUS DE CONSEQUENTIIS
TREATISE ON CONSEQUENCES
BOOKI
CONSEQUENCES IN GENERAL AND AMONG

ASSERTORIC SENTENCES
CHAPTER 1-1. THE TRUTH AND FALSITY OF SENTENCES
[1.1.1] In Book I, I would like to deal with consequences by discuss-

ing, so far as I may, their causes - about which many things have been
sufficiently proven by other people before, although perhaps not
reduced to the first causes through which they are said to hold.
[1.1.2] Now it will be necessary to layout some assumptions; in

Chapter 1-1 I shall state why a true sentence is called true, a false
<sentence> false, a possible one possible, an impossible one impossible,
and a necessary one necessary.
[1.1.3] Some people claim that a true sentence is true because howso-
ever it signifies is the case (qualitercumque ipsa significat ita est), that
is, in the thing or things signified.
[1.1.4] I do not think [1.1.3] is true, taken literally.! For if Calvin's

horse, which centered well, is dead, then
(1) Calvin's horse cantered well
is true, but it is not the case in the thing <signified>, for the "thing" <in
question, namely Calvin's horse> is <now dead and is> decomposing.
Or suppose that <the horse> were annihilated: then there is nothing at
all in the thing signified, hence there is nothing in it in some way, and
hence <a fortiori> there is <nothing in it> in either one way or another.
[1.1.5] Still, (1) is properly true, for it was the case as the sentence
signifies it to have been. Thus
(2) Antichrist will preach
is true, not because it is the case as (2) signifies, but because it will be
the case as (2) signifies will be. Similarly,
(3) Something which never will be can be
177
178 TREATISE ON CONSEQUENCES
is true not because it is the case as (3) signifies but rather because it
can be the case as (3) signifies to be able to be.
[1.1.6] Thus it is clear that it is necessary to assign to the different

types of sentences the causes of their truth in different ways.
[1.1.7] Careful readers can now set forth from the preceding <discus-
sion> how the <causes of the truth> of affirmative sentences should be
assigned. The causes of their falsity can be known similarly. Since,
indeed, the same sentence cannot be both true and false,2 and never-
theless if it is formed it must be either true or false, 3 one must assign
the cause of the truth and the cause of the falsity of the same sentence
in the manner of contradictories. The cause of the falsity of a present-
time assertoric sentence will therefore be because howsoever it
signifies to be is not the case; of a past-time <assertoric sentence>,
because howsoever it signifies to have been was not the case; of a
possible <sentence>, because it cannot be as it signifies it to be able to
be the case (or to be able to have been, or to be able to be, etc.); and
so on for other modes and circumstances.
[1.1.8] It should be remarked about negative <sentences> that any

negative has, or can have, some affirmative <which is> contradictory to
it. 4 Whatsoever is the cause of the truth of the one sentence is the
cause of the falsity of the other, and conversely. Hence it is impossible
that they both be simultaneously true or simultaneously false by any
power,s
[1.1.9] Therefore, by [1.1.7] and [1.1.8], it is true for any negative

present-time assertoric sentence that it is not howsoever the affirma-
tive <sentence> contradictory to it would (if propounded) signify to be
the case. The same is true for possible or necessary, past-time or
future-time <negative sentences>, making the obvious changes.
[1.1.1 0] By [1.1.9], then, any false negative present-time assertoric

sentence is false because it is howsoever the affirmative <sentence>
contradictory to it would (if it were propounded) signify to be the
case; and so for other times and modalities too, each in their own way.
[1.1.11] Next, a possible affirmative present-time assertoric sentence is

I. GENERAL AND ASSERTORIC CONSEQUENCES 179
called possible because howsoever it signifies to be can be the case. If

it were instead past-time, then it is possible because howsoever it
signifies to have been is possible to have been the case; and in like
manner for the others, each in their own way.
[1.1.12] 1 do not argue this; nor do 1 state anything more or specify;

rather, 1 lay it down. Yet since names signify by convention - <that is>,
just as some people agree to speak - I declare that for the sake of
brevity 1 might use the idiom "howsoever it signifies is the case" for
any cause of the truth of whatever kind of sentence. The reader should
take the meaning according to my remarks above, and not according
to its literal meaning. 6
CHAPTER 1-2. THE CAUSES OF THE TRUTH AND FALSITY

OF SENTENCES
[1.2.1] Since for any sentence there can be numerous causes by which it
is made true or false, in Chapter 1-2 1 lay down some things about the
number (large or small) of this kind of cause, according to the supposi-
tion of its terms.
[1.2.2] First, 1 state that any sentence with some common term not
distributed, or an equiform <sentence>,? has or can have more causes
of its truth than a sentence with the same common term distributed,
the rest <of the sentence> remaining the same. 8
[1.2.3] (By "the cause of the truth" of some sentence 1 mean those
sentences anyone of which is sufficient for the truth of the given
sentence.)
[1.2.4] The claim in [1.2.2] does not require a proof, but a statement
like this one: A sentence with an undistributed term, or an equiform
<sentence>, has or can have with respect to that term as many causes of
its truth as that term supposits for. For example, if 1 say
(4) A man is running
then (4) is true if Socrates alone is running, or if Plato alone <is
running>, and so forth with the rest <of the men> - and it is no less
true if all <are running> at once. But the sentence with the term distri-
buted can have with respect to that term only one cause of its truth:
namely that it is so for all, not merely for one or two.
[1.2.5] Now I say in [1.2.2] "that <sentence> or an equiform

<sentence>" because
(5) A man is an ass
cannot have any cause of its truth, for it is not able to be true, though
another <sentence> equiform to it might have, such as
(6) A horse is an animal
or
(7) Water is hot
have.
[1.2.6] Thus <to restate [1.2.2]> it is not inconceivable from the form
of the sentence 9 that any given sentence with an undistributed term
can have more causes of truth than that <same sentence> with the term
distributed, other things being equal; but it is inconceivable from the
form of the sentence that the one with the term distributed have more
<causes of its truth than the same sentence with the term not distributed>.
And this is what I mean in the rest <of the treatise> by 'a sentence
having more or fewer causes of its truth.'
[1.2.7] Next, in a similar fashion I say of undistributed terms that a

sentence with a common term suppositing confusedly has more causes
of its truth than the sentence with the term suppositing determinately
with respect to the term by whose distribution <the supposition> is
confused. 1o That is, it is not inconceivable from the form of the
sentence that it have many <causes of its truth>, though the contrary is
inconceivable.
[1.2.8] For example, let us examine the difference between these two
sentences:
(8) Every B is A
(9) A is every B
It happens that (9) has as many causes of its truth with respect to A as
there are A, for (9) would be true if this A were B, and similarly «9)
would be true> if that A were B, and so on. But (8) also has the same
number of causes of its truth <as (9», for it would be true if every B
were this A, and if every B were that A, and so on. But there is another
cause of the truth of (8), namely that this B is this A, and that B is
that <other> A, and another B is yet another A, and so on. In fact,
(10) Man is an animal
is true in this way, and this <last> cause of its truth is inconceivable
<for a sentence with a term> in determinate supposition <as in (9)
above>.
[1.2.9] Finally, 1 say that if <in some sentence> both terms are undistri-
buted, but supposit determinately, then there are more causes of its
truth than if one <term> were distributed and the other <were to
supposit> confusedly without distribution. This is clear, since any cause
sufficient for the truth of "Every B is A" is sufficient for "B is A" to
be true, but not conversely.
[1.2.10] Thus <we may conclude from the above discussion that> a
sentence with both terms undistributed would have the most causes of
its truth; <a sentence> with one term distributed and the other term
confused without distribution would have fewer <causes of its truth>; a
sentence with one term distributed and the other put determinately
without distribution would have still fewer <causes of its truth>; and a
sentence with both terms distributed would have the fewest <causes of
its truth>.
CHAPTER 1-3. THE DEFINITION OF 'CONSEQUENCE'
[1.3.1] In Chapter 1-3 we clarify what is meant by 'consequence,' and

what by 'antecedent' and 'consequent.'
[1.3.2] In all branches of knowledge one must first know the nominal
definition. To this end, I say that t';c- are two types of sentence:
categorical and hypothetical. Furthermore, a consequence is a
hypothetical sentence, composed of many sentences conjoined by the
word ( dictio) "if' or the word "therefore" or their equivalent."
[1.3.3] These words <"if' and "therefore"> indicate (designant) that of

the sentences conjoined by them one follows from the other (una
sequatur ad aliam), and they differ in that "if' indicates that the sent-
ence immediately following it is the antecedent and the other the
consequent, while "therefore" indicates the converse.
[1.3.4] Some people say that any such hypothetical sentence (that is,
one conjoining many sentences by "if" or "therefore") is a consequence,
and that there are then two types of consequences, those which are
true 12 and those which are false.
[1.3.5] Others say that <such a sentence> should not be called a conse-
quence if it is false, but only if it is true.
[1.3.6] But between II .3.4]-[1.3.5] there is no real disagreement here.

For in this treatise - whether or not it may be true - words signify
by convention; I mean to understand by "consequence" a true conse-
quence, and by "antecedent" and "consequent" sentences one of which
follows from the other by an acceptable (bona) consequence.
[1.3.7] Now "antecedent" and "consequent" are said relative to one

another and so should be described with reference to each other. Thus
many people say of two sentences that one is antecedent to the other
which cannot be true without the other's being true, and that <that
sentence> is consequent to the rest which cannot fail to be true with
the rest being true. The result is that any sentence is antecedent to any
other sentence which cannot be so without the other <sentence> being
true.
[1.3.8] But the description <of "consequence" given in [1.3.7]> is defec-

tive or incomplete, because
(11) Any man runs; therefore, some man runs
is a correct consequence, and yet it is possible that the first <sentence
of (11» be true with the second <sentence of (11» not being true, or
rather, with the second <sentence> not existing. 13
[1.3.9] Hence some say that the description given in [1.3. J should be
I
supplemented like so:

That sentence is antecedent to another sentence which

cannot fail to be true with the other <sentence> being true,
when both <sentence> are formed together.
[1.3.10] But I say that the description <proposed in [1.3.9]» IS no

good, for
(12) No sentence is negative; therefore, no ass is running
is not an acceptable consequence, though according to the description
in [1.3.9] it would be necessary to concede it to be acceptable. There-
fore <this description is not adequate either>. I prove my first premiss:
the opposite of the antecedent does not follow from the opposite of
the consequent, that is,
(12*) A certain ass is running; therefore, some sentence is
negative
does not follow. My second premiss is obvious: the first <sentence>
(namely that which is designated as the antecedent) cannot be true; it
is therefore impossible that it be true with the other not being true. 14
[1.3.11] For this reason, other people' define <"antecedent" and

"consequent" >in another way, asserting that:
One sentence is antecedent to another <sentence> which is

so related to it that it is impossible howsoever the one
signifies to be the case that it is not howsoever the other
signifies to be the case, when they are put forth together.
Now this description is not literally true, for it assumes that any true
sentence is true because howsoever it signifies is the case, and that
claim was denied previously <in [1.1.4]>. Nevertheless, it has also been
said that we would use this idiom for the sense given <in [1.1.12]>; and
in this way we allow the description. I 5
[1.3.12] In fact, we shall often also use the idiom of the first
description in [1.3.7], clearly disproved in [1.3.8], since there are
counterexamples for only a few consequencs. Yet whatever manner of
speaking we adopt, we intend the sense touched upon in [1.3.11].
[1.3.12] Moreover, a consequence can be described in this way: A

consequence is a hypothetical sentence composed of antecedent and
consequent indicating that the antecedent is antecedent and the
consequent is consequent. This occurs by "if' or "therefore" or their
equivalent, as said in [1.3.2].
CHAPTER 1-4. THE KINDS OF CONSEQUENCES
[1.4.1] In Chapter 1-4 we put forward a division of consequences into

material and formal <consequences>.
[1.4.2] A consequence which is acceptable 10 any terms IS called

formal, keeping the form the same. IO Or, if <the sentence> explicitly
deals with the meaning (vis) of its words, <then> a consequence is
formal for which any equiform sentence which might be formed would
be an acceptable consequence; for example,
(13) What is A is B; therefore, what is B is A
[1.4.3] A consequence for which not every equiform sentence would
be an acceptable consequence is called material - or, as is commonly
said, <a consequence is called material> which does not hold in all
terms (keeping the form the same). For example,
(14) A man runs; therefore, an animal runs
<is a material consequence> because it is not acceptable in these terms:
(15) A horse walks; therefore, wood walks 17
[1.4.4] It seems to me that no material consequence is evident in
inferring <the consequent from the antecedent> except by its reduction
to a formal <consequence>. Moreover <a material consquence> is
reduced to a form <conseuqence> by the addition of some necessary
sentence or sentences which, assumed with the antecedent, render the
consequence formal.
[1.4.5] Thus if I assert (14), I prove the consequence by the fact that
all men are animals, for if any man is an animal and a man runs, it
follows as a formal consequence that an animal is running. All who
argue using enthymemes must prove their consequences, if they are
not formal, in this way.
[1.4.6] Finally, since I have started to discuss the kinds of con-

sequence, I should say that certain material consequences are called
simple consequences because they are acceptable consequences simply
speaking, for it is not possible that the antecedent be true with the
consequent being false, or being the case, and so on.1X
[1.4.7] The rest are called consequences ut nunc, which are not
acceptable simply speaking, for it is possible that the antecedent be
true without the consequent <being true>, yet they are acceptable for
now (ut nunc), since it is completely impossible with things related
as they are now related for the antecedent to be true without the
consequent <being true as well>.I,)
[1.4.8] Common people often use these consequences, so that if

we were to say "Cardinal White has been elected Pope" we would
conclude "Therefore a Master of Theology has been elected Pope,"
and if I say "I see one such man" you conclude "Therefore you
certainly see a deceitful man."20 This consequence is reduced to a
formal <consequence> by the addition of a sentence true but not
necessary, or by <the addition of> some sentences true but not
necessary, as in [1.4.7]. In the preceding example, <we add the truths
that> Cardinal White is a Master of Theology, and a man of this sort is
a deceitful man.
[1.4.9] In this <following> example there is an acceptable consequence

by [1.4.7], positing that Socrates and Plato and Robert are the only
men:
(16) Socrates is running, Plato is running, Robert is running;

therefore, every man is running
for the consequence is perfected by this truth: "Any man is Socrates or

Plato or Robert."
[1.4.10] We should point out that promissory-consequences are this

kind of consequence ut nunc. For example, when Plato says to
Socrates
(17) If you come to me I shall give you a horse
this sentence might be true and a consequence, or it might be false

and so not a consequence.
[1.4.11] If the antecedent were impossible - that is, if Socrates could

not come to Plato - the consequence is simply true, for anything
follows from the impossible, as will be discussed in Theorem 1-1. But
if the antecedent is false, though not impossible, the consequecne is
acceptable ut nunc. <The reason for this is that> any consequence ut
nunc follows from any falsehood, as will be discussed in [1.8.69],
provided that we extend 'consequence ut nunc' to consequences ut
tunc, whether they are past, future, or determined to any other time
<than the present>.cl
[1.4.12] If the antecedent of (17) were true, namely that Socrates will
come to Plato, then it may be said that even now the consequence is
acceptable, since a formal <consequence> can come about through
truths added to (17), in this way: Whatever Plato wills to do in the
future (i) he will be able to do by holding to that volition (and holding
in any circumstances to what he wills); and (ii) when he is not
prevented he does that thing when and how he wills. (Modify (i)-(ii)
until they are true by what Aristotle says in Metaphysics IX [cap. v
1048a11-16].) But Plato wills to give a horse to Socrates, who will
come to him, when he will come to him. Therefore Plato will give
Socrates a horse. If, then, these claims about Plato's will and abilities
are true, Plato asserts a true consequence ut nunc to Socrates. But if
they are false, then Plato told Socrates a falsehood <which is> not a
consequence - and if Plato believed these added claims (or similar
ones) to be true, then Plato lied to Socrates.
CHAPTER 1-5. THE SUPPOSITION OF TERMS
[1.5.1] In Chapter 1-5 I state that an affirmative sentence indicates

(designat) that its terms supposit for what is the same, or was or will
be or can be the same, according to the requirements of the <given>
sentence. If I say "A is B" I indicate that A is the same as B; if I say
"A was B" I indicate that A was the same as B; and so on for the rest.
[1.5.2] A negative sentence indicates the opposite, that is, that what
1. GENERAL AND ASSERTORIC CONSEQUENCES 187
the subject supposits for and what the predicate supposits for are not
the same, or were not or will not be the same, and so on.
[1.5.3] The whole <negative sentence> is analyzed according to its

requirements. Because of the words <involved>, one must say one thing
about universal <negatives> and another about particular <negatives>, so
that <for the universal negative> nothing for which the subject
supposits is the same as something for which the predicate supposits,
and for the particular negative that something for which the subject
supposits is not the same as something for which the predicate
supposits. 22 And this is why
(18) A chimaera is a chimaera
is not calkj true, namely since the subject does not supposit for any-
thing, and hence does not supposit for anything the same as what the
predicate supposits for. 23
[1.5.4] Hence I believe that no affirmative sentence, some term of

which does not supposit for anything past or present or future, at least
possibly, is true. Positing that no chimaera exist, then, these are false:
(19) A chimaera is able-to-be-understood.
(20) A chimaera is opinable.
(21) A chimaera is signified by 'chimaera'.
I do not intend now to resolve the doubts which may be raised about
this, for that would require a separate treatise.
[1.5.5] Even if these points are conceded, it does not follow that any
affirmative present-time assertoric sentence whose terms supposit for
the same is true, for in a sentence saying itself to be false it can
happen that the term supposit for the same and yet <the sentence> is
false 24 - e.g. if someone were to say
(22) The sentence which I am speaking is false.
The reason <a sentence such as (22) is not true> is because, despite the
fact that (22) indicates by its form that the terms supposit for the same
and is the case, nevertheless according to the signification of the
predicate (22) indicates that <what the terms supposit for> is not the
same. When we call a sentence false, we indicate that <the terms do
not supposit> for the same, <as pointed out in [l.5.2] >. This is why a
sentence like (22) says that it is the same and it is not the same - and
so, although how it signifies is the case, it is nevertheless not the case
howsoever it signifies; and so it is false.
[l.5.6] In the same way, it is not necessary that every negative

sentence whose terms do not supposit for the same is true, for a
counterexample is the <negative sentence> which denies itself to be
true, as when I say
(23) The sentence I am now speaking is not true.
[l.5.7] Sophisms called "insoluble" are based on these matters, about

which I shall say no more at present.
[l.5.8] Yet I think that any affirmative <sentence> would be true whose
terms were to supposit for the same, in the manner appropriate to its
form, provided that <the sentence> does not proclaim itself to be false,
formally or explicitly, consecutively or implicitly.2s And likewise for
negatives.
[l.5.9] These matters, then, are set forth in Chapter 1-5. I do not care
whether they, like those which have been set forth earlier and will
be set forth later, are not conceded by some people, for they or
something similar must be conceded, by conceding which what we
later derive about consequences follows.
CHAPTER 1-6. THE AMPLIATION OF TERMS
[l.6.1] Because the predicate always follows the condition of the verb
in suppositing for a time, in Chapter 1-6 I take up the difference
between the subject and the predicate in such sentences. For example,
ifl say
(24) A man will be white
then (24) would be false, positing that all existing men are white white
nevertheless none are white in the future. And so with the past as well.
[1.6.21 Similarly, if the time of the verb were determined, if I say

(25) A man will be white on Sunday
then (25) would be false even if all men are white now and all will be
white in the future after Sunday, <provided that no man is white on
Sunday>. If, however, no man is white and on Sunday a single white
man is born, (25) would be true. And so, in their own way, for past-
time and present-time <sentences>.
[1.6.31 Yet the subject of the sentence is not thus determined to the
time of the verb, but rather always keeps its supposition for things of
the present time.
[1.6.4] Furthermore, if the verb is of another time <than the present,

the subject> is ampliated with present things to those of the time of the
verb.
[1.6.5] This is the reason a sentence with a past-time or a future-

time verb has two causes of its truth, according to Aristotle [De
interpretatione cap. ix 18b21-24 (?)]. For example,
(26) He who works was healthy
can be true either (i) because he who is working was healthy or (ii)
because he who was working was healthy. And <the sentence>
(27) A will be B
can be true either (i) because what is A will be B or (ii) because what
will be A will be B.
[1.6.6] Thus I say that the present-time verb 'is,' which is the copula
of a sentence, strictly makes the predicate as well as the subject
suppa sit for present things, unless the predicate acts in another way
(about which more will be said in [1.6.101 and following). And these
verbs 'will-be' and 'was' and other similarly tensed verbs determine the
predicate to their time and ampliate the subject to <supposit for things
of> their time along with <suppositing for things of> the present time.
[1.6.7] Perhaps this is <what is meant> by "the predicate appellates its

form," or something of this sortY'
[1.6.8] Because possibility is <directed> towards the future and com-

pletely to possible things, the verb 'can' similarly ampliates the subject's
supposition to all those things which can be. Thus
(28) B can be A
has two senses, or two causes of its truth, namely (i) because what is B
can be A or (ii) because what can be B, and is not <now B>, can be A.
One must say the same for <the verb> 'cannot-be,' since in contradic-
tories the terms must supposit for the same. One must also say the
same for <the verbs> 'is-necessarily' and 'is-necessarily-not' since they
are equipollent to <verbs> of impossibility, and likewise for 'is-not-
necessarily' and 'is-not-necessarily-not,' since they are equipollent to
<verbs> of possibility.
[1.6.9] But for the rest, most or all people treat these sentences as
though they were <literally> true:
(29) Calvin's horse is dead.
(30) My grandfather [or the Pope, or the King, etc.] is dead.
(31) Many men are able-to-be-generated.
(32) Many men are going-to-be-generated.
(33) Although no rose exists, still a rose is able-to-be-understood.
(34) What does not exist can be remembered.
(35) Many things are possible which neither were nor are nor will
be.
[1.6.10] Since all languages use this manner of expression, it thus seems
to me that such past-<time> or future-time predicates <as in (29)-(35»
ampliate their subjects just like past-time or future-time verbs. similarly,
predicates involving possibility function just like the verb 'can,' as for
example in
(36) Whatever can be is possible.
(37) Whatever can be is able-to-be-generated.
[1.6.11] But subjects do not ampliate predicates; rather, they modify

(dimittunt) them according to the requirements of the words which
follow. Thus
(38) Some dead man is a man
is not true, although
(39) Some man is dead
is true.
[1.6.12] It should be noted that there are still certain verbs 27 <which>,
no matter what their time may be, go over into their accusative-objects
(transeunt in suos accusativos) they govern after themselves with
temporal indifference. Examples are the verbs 'I know,' 'I think,' 'I
understand,' so that something may be understood without any
temporal difference. For example, there is a difference between
sentences of this sort:
(40) 1 beat a horse.

(41) 1 set fire to a house.
(42) 1 boil water.
and sentences of this sort:
(43) 1 think-of a rose.

(44) 1 hope-for health.
(45) I desire a good wine.
because the actions of the verbs of (40)-(42) necessarily go over into

a present thing without a mediating concept, while the actions of the
verbs in (43)-(45) befall (cadit) things which are not concepts by
means of mediating concepts which may be indifferent to past, present,
and future.
[1.6.13] For example, if 1 understand man, it properly follows that 1

have present to me that concept <of man>, but it does not follow that
there is present to me that man which 1 understand, for the concept
may be about absent things (either past or future). Such verbs ampliate
their objects to past, future, and even possible things; indeed, God
understands some things which neither were nor are nor will be, but
yet they are possible. 2R
[1.6.14] Similarly, verbs indicating an act of desiring or prorrusmg

or owing ampliate through a mediating cognition according to their
individual requirements, as for example
(46) I promise (or owe) you a horse.

(47) I remember the past.
(48) I have-hope-for the future.
and similar sentences.
[1.6.15J Passive verbs ampliate their subjects III a similar fashion, as

for example
(49) A rose is understood.

(50) A horse is promised to you.
and the like.
[1.6.16J Even participles and nouns, if they become predicated by

descent. ampliate their subjects like the verbs mentioned above. Thus
it is true to say
(51) A rose is able-to-be-understand
even though no rose may exist, and <it is equally true to say>
(52) A generated horse is promised to you.
(53) A generated horse can-be-promised (promissahilis)
(54) Roses are promised to you.
(55) What does not yet exist can-be-promised.
<for the same reasons>.
[1.6.7J It is obvious <from the preceding discussion> that sentences

with ampliative verbs or predicates have more senses, or more causes
of their truth (if the subject is taken without distribution) than do
sentences with non-ampliative verbs or predicates. if the predicate is
not ampliative, then
(56) BisA
has as the only cause of its truth that B is A, but
(57) B will be A
has as causes of its truth (i) because what is B will be A or (ii) because
what will be B will be A. And even
(58) A man is dead
l. GENERAL AND ASSERTORIC CONSEQUENCES 193
has as causes of its truth either (i) because he who is a man is dead, or
(ii) because he who was a man is dead. Even
(59) A man is white
has causes of its truth similar to those given for (56). The <sentences>
(60) A can-be-understood.
(61) A is understoood.
have as causes of their truth what was or is or will be A is understand-

able.
[1.6.18] From [1.6.17] it is clear that an ampliated term is related to

<the same term> not ampliated as the more general to the less general,
and ampliated in a certain way it supposits for more than <the same
term> not ampliated, insofar as it is part of the form of the sentence.
[1.6.19] To remove any doubt, it should be noted that when I assert

(58) the predicate need not supposit for present things, but rather
supposits for past things in the past time-period.~~ But since the verb
in (58) is present-time, it must supposit for thse past things insofar as
they are <now> present.
[1.6.20] This is how the initial claim in [1.6.1], namely that the verb
takes the predicate to its own time, should be understood. If the
predicate is present-time or without <any reference to> time, it suppos-
its for things at the time of the verb. Thus if I say
(62) There were animals living in Noah's Ark
since the term 'living' is present-time it supposits for those things

which were then present objects. But if I say
(63) In Aristotle's time Averroes was going-to-be-born
since the term 'going-to-be-born' is future-time it supposits for those

things which were future at the time <of Aristotle>, although now they
are past <since Averroes has been born>. It is clear from this example
that a future-time participle can sup posit for things neither present nor
future, but already past and passed away. And similarly for the
converse, <that is, a past-time participle can supposit for things yet to
come>, as when I say
(64) Antichrist will-have-died.
CHAPTER 1-7. THE MATTER AND FORM OF SENTENCES
[1. 7.1] Since we have discussed the form of a sentence and the
distinction of consequences into materilll and formal, in Chapter 1-7
we must add what we take to be relevant to the form of a con-
sequence or sentence and <what we take tv oe relevant to> the matter
<of a consequence or sentence>.
[1.7.2] 1 say that in a sentence in which we speak of matter and form

we understand the matter of the consequence or sentence to be the
purely categorematic'O terms, namely the subject and the predicate.
The syncategorematic <terms> added to it, by which the <subject and
predicate> are connected or denied or distributed or taken to supposit
in a certain way are not included. And we say the entire remaining
part <of the sentence> pertains to the form. And thus we hold that (i)
the copulas of hyporthetical as well as categorical sentences pertain to
the form; (ii) negations and signs of quantity <pertain to the form>; (iii)
the number of sentences as well as the number of terms; (iv) the order
of the negations, copula, signs of quantity, number of terms and
sentences; (v) the relations of the relative terms; (vi) the modes of
signifying pertaining to the quantity of a sentence (such as discrete or
common) - and (vii) many other things which the diligent student can
recognize if they occur.
[1.7.3] For examples of the foregoing: (i*) the form <of sentences> with
copulas of different modes differ from <the form of> assertoric
sentences; (ii*) the forms of affirmatives are different from those of
negatives by the negations and <the form of> universals <differ from the
form of> particulars; (vi*) a singular sentence has a form different from
an indefinite <sentence> according to the commonness or discreteness
of terms; (iii*) the following sentences have different forms due to the
number of terms:
(65) A man is a man.
(66) A man is an ass.
as do these hypothetical sentences or consequences:

(67) Any man runs; therefore, some man runs.
(68) Any man runs; therefore, some ass walks.
Similarly, (iv*) the following differ according to the order <of the
terms>:
(69) Every man is an animal.
(70) An animal is every man
and these consequences <also have different forms due to the order of
the terms>:
(71) Any B is A; therefore, a certain B is A.
(72) Any B is A; therefore, a certain A is B
(v*) According to relation, this conjunctive sentence
(73) A man runs and a man does not run
has a different form from
(74) A man runs and he does not run
since (74) is impossible by its form and (73) is not. When other things
relevant to the form occur the diligent recognize them, as for example
(vii*) the <nominative or oblique> case of terms; the sentence
(75) An ass is a man.
(76) An ass is a man's
have different forms. And so on for the rest.
CHAPTER 1-8. THEOREM ABOUT ASSERTORIC CONSEQUENCES
[1.8.1] 1 add this principle 3 ] to those set forth above:

For any contradiction, one of the contradictories is true and
the other false, and it is impossible that both are true
together or false together; again, any sentence is true or
false and it impossible for the same sentence to be true and
false at the same time.
[1.8.2] Other theorems can be proven from these principles <and

Chapter 1-8 is devoted to this>.
[1.8.3] The first theorem is:
[Theorem I-I] From any impossible sentence any other

follows, and any necessary sentence follows from any other
sentence.
[1.8.4] Theorem I-I is obvious from the nominal definition of antece-

dent and consequent <given in [1.3.11] >, for it is impossible that some
impossible sentence be true, or even howsoever it signifies to be the
case. Therefore it is impossible that howsoever it signifies be the case
while not being howsoever any other <sentence> signifies.
[1.8.5J Similarly, it is impossible that it not be howsoever any neces-

sary sentence signifies <to be the case>. Therefore it is impossible that
howsoever it signifies to be does not obtain, no matter how <the other
sentence> signifies <to be the case>.
[1.8.6] It should also be noted that Theorem I-I should be put in an

appropriate way for consequences ut nunc, i.e.
[Theorem 1-1*1 From any false sentence any other <sen-
tence> follows as a consequence ut nunc, and also any true
<sentence> follows from any other as a consequence lit nunc
since it is impossible ut nunc that a sentence which is true be not true
as things are now related. Hence neither is it possible that it not be
true with any other true <sentence> obtaining.
[1.8.7J If there were a word for past-time or future-time <sentences>,

then the consequence would be called 'ut tUIlC' (or whatever name you
want to give it). For example here is a consequence lit nunc, or ut
tunc, or lit nunc pro tunc:
(74) If the Antichrist will not be generated, Aristotle never
existed.
Since although it is simply true that it is possible that Antichrist is not
going to be, it is nevertheless impossible, with things being related as
they have been <so> related, that he will not exist; indeed, he will exist,
and it is impossible that he will and will not exist.
[1.8.8] The second theorem is:

[Theorem 1-2] (a) From any sentence there follows any
other whose contradictory cannot obtain together with <the
first sentence>, and (b) from no sentence does there follow
another whose contradictory can obtain together with it.
[1.8.9] (I mean by a sentence "obtaining" the same thing as <the

sentence> being true, or the same thing as that howsoever it signifies
is the case. By "obtaining together" I mean the same thing as that
howsoever they signify is the case together.)
[1.8.10] Let us posit that it is impossible for sentence B to obtain with

sentence A. Then I say that the contradictory of B, namely not-B,
follows from A. Proof <of Theorem I-2(a»: either (i) A cannot obtain
and so is impossible, and then anything follows from it (by Theorem
1-1); or (ii) A can obtain, and then B or not-B must obtain while A
obtains, since one part of any given contradiction is true. But it is
impossible <ex hypothesi> that B obtain while A obtains, and con-
sequently it is impossible that not-B not obtain when A obtains. Thus
not-B follows from A.
[1.8.11] Theorem 1-2(b) is proved: if A and not-B obtain together,

then it happens that not-B obtains while A obtains; but not-B and B
cannot obtain together. Therefore it is possible that B fail to obtain
while A obtains, and thus B does not follow from A.
[1.8.121 The third theorem 32 is:

[Theorem 1-31 (a) For any acceptable consequence the con-
tradictory of the antecedent must follow from the
contradictory of the consequent, and (b) any sentence
formed as a consequence is an acceptable consequence if
the contrary of the antecedent of the indicated consequence
follows from the contradictory of the antecedent of the
indicated consequence.
[1.8.131 Theorem 1-3(b) is usually put this way: "Any consequence is

acceptable when the opposite of the antecedent follows from the
opposite of the cosnequent." But I do not put it this way because,
strictly speaking, it would be a petitio principii. It literally says
that there is already here a consequence - both antecedent and

consequent - and so <by the definition in [1.3.11] > an acceptable
consequence.
[1.8.14] Theorem I-3(a) is proved by positing that B follows from A;

we then say that not-A follows from not-B, because it is the case or it
is possible that A obtain together with not-B, by Theorem 1-2. But <ex
hypothesi> B must obtain while A obtains, and so Band not-B obtain
together, which is impossible.
[l.8.15] Theorem I-3(b) is proved by the same <reasoning>.
[1.8.16] The fourth theorem" is:

[Theorem /-4] For any acceptable consequence, (a) what-
ever follows from the consequent follows from the ante-
cedent; (b) whatever the antecedent follows from the
consequent <also> follows from. Similarly and negatively, (c)
whatever doesn't follow from the antecedent doesn't follow
from the consequent; (d) whatever the consequent does not
follow from the antecedent does not follow from.
[1.8.17] Theorem 1-4 has four parts; Theorem 1-4(a) is clear, because
if B were to follow from A and C <were to follow from> B, then it
would be impossible for A to obtain without B, and, necessarily, if B
were to obtain it obtains together with C. Thus it is impossible for A
to obtain without C, and so C follows <from A>.
[1.8.18] Now Theorem I-4(b) is equally clear, since if B were to

follow from A, which is antecedent to C, 1 say C must follow from A.
This is because this returns us to case described by Theorem 1-4(a),
namely that if B follows from A and C <follows> from B, then C
therefore <follows> from A.
[l.8.19] And Theorem 1-4(c) is also obvious, for if we assume that B

follows from A and C, and C does not follow from A, then I say that
C does not follow from B: C cannot obtain with A by Theorem 1-2,
and therefore <C cannot obtain> with B, since <ex hypothesi> B must
obtain when A obtains. Thus it follows by Theorem 1-2 that C does
not follow from B.
[1.8.20] Finally, Theorem 1-4(d) is also clear, since if we assume that

B is consequent to A and that B does not follow from C, then 1 say
that A does not follow from C: not-B obtains with C by Theorem 1-2
and not-A follows from not-B by Theorem 1-3. Therefore not-A
obtains with C, and so A does not follow from C.
[1.8.21] The fifth theorem is:

[Theorem 1-5] It is impossible that (a) a false <sentence>
follow from truths, or (b) an impossibility from possible
<sentences>, or (c) a non-necessary <sentence> from a neces-
sary <sentence>.34
[1.8.22] Proof <of Theorem 1-5(a»: if each part of a consequence (i.e.

the antecedent or the consequent), or both parts, were not formed,
there would not be any consequence. But with each formed in a
consequence it is impossible that it is as the antecedent signifies to be
the case with it not being as the consequent signifies; this is clear by
definition. Thus it is impossible, then, that the antecedent be true with
the consequent not being true. Therefore it is impossible that the
antecedent be true while the consequent is false.
[1.8.23] The same <reasoning> is clear for Theorem 1-5(b), since if the
antecedent is possible then it can be <the case> as it signifies. Assuming
this, it must be the case as the consequent signifies, and therefore the
consequent is possible, not impossible.
[1.8.24] The same <reasoning> is clear for Theorem 1-5(c), since it is

impossible for the antecedent to obtain without the consequent; there-
fore if it is always necessary that the antecedent obtain it will always
be necessary that the consequent obtain - and this is the consequent's
being necessary.
[1.8.25] We can equivalently deduce from Theorem 1-5 the following:
[Remark] When the antecedent and the consequent are

formed together, if the antecedent is true the consequent is
true, and if it is possible <the consequent> is possible, and
if it is necessary <the consequent> is necessary. Conversely,
if the consequent is false the antecedent is false; if the

consequent is impossible the antecedent is impossible; if the
consequent is not necessary the antecedent is not necessary.
[1.8.26) Nevertheless, [1.8.21)-[1.8.25) notwithstanding, it can happen

that the antecedent can be true and the consequent cannot be true, so
that
(78) Any sentence is affirmative
can be true and
(79) No sentence is negative
cannot be <true>, although (79) follows from (78). Thus it follows as a
corollary that the following are not quite the same (idem adequate):35
(80) A sentence is possible.
(81) A sentence can be true
or even that <these are quite the same>:
(82) A sentence is impossihl·'!.
(83) A sentence is not possIbly true
since (79) is true because it follows from a possible <sentence, namely
(78», and (79) is not possibly true.
[1.8.27) Yet despite the discussion of [1.8.21)-[1.8.25), the true and

the necessary can follow from the false or the impossible, since
anything follows from the impossible, as pointed out above <in
Theorem 1-1 >.
[1.8.28) There is a sophistical objection to Theorem 1-5, namely that

the following is a acceptable syllogistic consequence:
(84) Anything running is a horse, and every man IS running;

therefore, every man is a horse.
Although the conclusion of (84) is impossible each of the premisses is

possible. The solution <to this sophism> is that neither of the premisses
is the whole antecedent of the conclusion. 36 Rather. the antecedent is
the single conjunctive <sentence> composed of those two premisses,
namely "Anything running is a horse and every man is running," and

this conjunctive sentence is impossible, like the conclusion.
[1.8.29] The sixth theorem is:

[Theorem 1-6] If from any given sentence with some neces-
sary <sentence> added to it, or some necessary <sentences>
added to it, there follows some conclusion, <then> from the
same sentence alone, without the addition of the necessary
sentence or sentences the same conclusion follows.
[1.8.30] Proof: If the conclusion B follows from the sentence A with

several necessary sentences added, then I say that B follows from A
<alone>, because if B follows from those necessary <sentences> without
A, then either A is impossible, and so anything follows from it, or A is
possible, and then it is either impossible that B not obtain while A
obtains (and thence B follows from A), or it is possible that B not
obtain while A obtains. And if anyone objecting were to suppose that
the latter were the case, then since A can only obtain with all
necessities obtaining together, it follows that it is possible for B not to
obtain with all the necessities while A obtains, and therefore B does
not follow from A with certain necessities added to it, which violates
our initial assumption.
[1.8.31] In the same way <as Theorem 1-6 >:

[Theorem 1-6*] If from any given sentence with some truth
added to it, or several truths added to it, there follows
some conclusion, <then> from the same sentence alone there
follows the same conclusion as a consequence ut nunc,
though not as a simple consequence.
This claim would be proved in a manner like Theorem 1-6.
[1.8.32] The seventh theorem is:

[Theorem 1-7] From any conjunctive sentence composed
of two <sentences> contradictory to each other any other
<sentence> follows as a formal consequence.
[1.8.33] It is immediately clear from the statement <of Theorem 1-7>
that from any such <conjunctive sentence> - rather, from anything

even implying a contradiction - anything follows, since any such
<sentence> is impossible, <and we may apply Theorem 1-1>. But it is
necessary to see how this is a formal consequence. 37
[1.8.34) Therefore, I say that anything follows from the conjunctive

sentence
(85) Any B is A and a certain B is not A
and likewise in whatever terms such a consequence might be formed.
Proof: I assume for the sake of an example that this would follow:
(86) <Any B is A and a certain B is not A>; therefore, a stick
stands at an angle
Since (85) is posited then "Any B is A" also holds, since either
conjunct follows from a conjunctive sentence. Thus "Any B is A or a
stick is at an angle" follows, since from anything there follows itself in
disjunction with any other. Thus I argue from this and the latter part
of Theorem 1-1 as follows:
(87) Any B is A or a stick stands at an angle, and a certain B is
not A; therefore, a stick stands at an angle
This syllogism 3H holds through the locus a divisione, for if one of the
two things put in a disjunction were to be refuted then the remaining
disjunct would be inferred. Therefore, from beginning to end the
conclusion follows from the antecedent (85), since whatever follows
from the consequent follows from the antecedent <by the transitivity of
consequences, Theorem 1-4>.
[1.8.35) Some people believe that if the argument given 10 [1.8.34)

holds, then
(87) Only a father exists; therefore, a stick stands at an angle
would be a formal consequence. 39 But this is not so, because the given
form fails to hold in other terms, for example
(88) Only God exists; therefore, an ass is standing in the stable.
Indeed, these people say that the following are formal consequences:
(89) Only a father exists; therefore, a son exists.

(90) A son exists; therefore, not only a father exists
but this is not true, for neither (89) nor (90) are formal <con-
sequences>, since they fail in other terms:
(91) Only God exists; therefore, an ass exists.
(92) A being exists; therefore, not only one exits
and (91 )-(92) do not follow.
[1.8.36] It is apparent from Theorem 1-7 how any consequence with

an impossible antecedent is reduced to a formal consequence by the
addition of some necessity: if the antecedent is impossible, then its
contradictory is necessary, and by adding <the contradictory> to the
<antecedent> the result, no matter what the consequent, will be a
formal consequence, as has been remarked.
[1.8.37] What some people wonder about, then, namely how the
consequence
(93) Nothing exists; therefore, an ass is running
is consistent, is apparent from [1.8.36]. I say that it is consistent
because "something exists" is necessary,40 <and taken> together with
<the antecedent, namely "nothing exists">, (93) follows. And it is the
same with this consequence:
(94) An ass is running; therefore, an ass is moved.
Whence indeed does it follows? Since "Anything running is moved" is
necessary (or at least "If something is running then anything running is
moved" is necessary), then this added to the antecedent of (94) makes
the consequence evidently formal.
[1.8.38] It is also obvious that any consequence ut nunc with a false

antecedent is reduced to a formal consequence by the addition of
some truth (clearly, a truth contradicting that antecedent).
[1.8.39] Yet some people, despite the discussion,will yet wonder how
any consequence whose consequent is necessary is reduced to a formal
<consequence>, e.g.
(95) An ass is running; therefore, God is just.

I say that this consequence is reduced to a consequence with an
impossible antecedent as in Theorem 1-3: the sentence
(96) No God is just; therefore, no ass is running
follows, and therefore it follows <that (95) is a formal consequence>.
[1.8.40] The eighth theorem-lJ is:

[Theorem 1-81 (a) All sentences with the same number
of causes of their truth follow from the same <sentences>;
(b) all having more <causes of their truth> follow from any
having fewer or the same <number of causes of their truth>
by some <sentences> of those <having> more, but (c) not
conversely.
[1.8.41 J Theorem 1-8 is customarily stated: There is no acceptable

consequence from <a sentence> with more causes of its truth to <a
sentence with> a single <cause of its truth>, but there is an acceptable
consequence from a single <cause of truth> to many.
[1.8.42] Now Theorem 1-8 as given in [1.8.40J should be understood

not only to be about <sentences> actually having <more causes of their
truth> but about the possibility or the consistency of having <more
causes of truth>. If these <causes of truth> come from the form of the
sentence, the sentences are formal consequences, and if they do not
<come from the form of the sentence> but rather from the signification
of the terms, such as
(97) A man is running
or
(98) The risible <thing> is running
then they are not formal consequences.
[1.8.43J Therefore, speaking literally, the first part of Theorem I-8(a)

should be put like this: "Any two sentences of which neither can have
some cause of its truth which is not a cause of the truth of the other
<sentence> ... " Or rather, altering the later part as well: "Any two sen-
tences of which neither can have some cause of being how it signifies
which is not <a cause of being how the other signifies> ... " in the
sense previously determined. (Recall that we said in [1.1.12] that no
matter what idiom we use we mean the sense given there; we shall not
repeat this again in this work.)
[1.8.44] Theorem 1-8(a) is therefore proved by going back to the

principle <stated in [1.8.1]>: it is perfectly clear that it is impossible for
any two sentences to be how one signifies while not being how the
other signifies if every cause of being as one signifies is equally a cause
<of being as> the other signifies, and there is no other <cause>, or if
nothing is the cause of being as one signifies <which is not the cause of
being> as the other signifies.
[1.8.45] Theorem 1-8(b) is similarly clear, for if nothing can be a

cause of being as A signifies which is not also a cause of being as B
signifies (though the converse is possible), then it is obvious that A
cannot be true without B, or it cannot be as A signifies unless it is the
case as B signifies.
[1.8.46] Theorem 1-8(c) is also evident. If there is some cause of the

truth of A which is not a cause of the truth of B, then A can be true
without B, and so <B> does not follow from <A>.
[1.8.4 7] From Theorem 1-8 there follow all equipollences and conver-
sions. But we must specify what these are, and so we give them in the
following theorems.
[1.8.48] The ninth theorem is:
[Theorem /-9] All sentences of the same quality in the same

terms, sup positing in the same way and with the same
copula, mutually follow from one another.
[1.8.49] The reason for this theorem is that the causes of their truth
are numerically equal and are in fact the same, since the causes of the
truth of one sentence can only differ from the <causes of the truth of
the> other, either (i) because the terms differ; (ii) because the <terms>
supposit in different ways; (iii) because the copulas differ; or (iv)
because of the quality of the sentence - which can be reduced to (iii),

if one were to say that a negation's negating <action> is a determination
of the copula. And we confirm <Theorem 1-9> for the other cases,
since if it were not to hold this would seem to be by the transposition
<of the terms>. But one cannot say this, for an affirmative signifies this
thing to be the same as that one: either <the same as> something (in the
case of a particular sentence), or <the same as> whatever is supposited
for (in the case of a universal sentence). But if this thing is the same
as that one, then that thing must be the same as this one, and so the
conversion <of the sentence> cannot be prevented. Similarly, a negative
sentence signifies this thing not to be the same as that one, and if this
thing is not the same as that one then that thing must not be the same
as this one. 42
[1.8.501 All formal equipollences of categorical assertoric sentences

are contianed in Theorem 1-9, as are all equipollences of necessary
sentences to necessary sentences, of possible sentences to possible
sentences, and so on for other modes, except for the equipollences of
contingents by conversion to the opposite quality.
[1.8.511 By Theorem 1-9 these follow from each other mutually as

equipollences (taking the negations negatively):
(99) Socrates is not running
and
(99*) Not: Socrates is running.
Similarly, the following pair <are equipollent> by the first rule of
equipollence:
(100) Every man is not running.
(100*) No man is running.
And these two <are equipollent as well>:
(101) Not: No man is running.
(101 *) Some man is running.
Or even
(102) Not every man is running.
(102*) Some man is not running.
And these too:

(103) Not: Some man is running.
(103*) No man is running
<and these equipollences hold) by the second rule of equipollences.
And by the third rule of equipollences <the following are also
equipollent):
(104) Not: No man is not running.
(104*) Some man is not running.
In like manner, the following pairs are equipollent:
(105) Of-any-man no ass is running.
(105*) Of-no-man an ass is running.
(106) One part of any given contradiction is true.
(106*) Neither part of some contradiction is true.
And we may continue in this fashion for other <sentences). For
example, these are equipollent modal sentences:
(107) Every B is necessarily A.
(107*) No B is not necesssarily-A.
(108) No B is possibly A.
(108*) Every B is not possibly-A.
and so on for the rest.
[1.8.52] Simple conversions are also contained in Theorem 1-9, but I

put them in <separate) theorems to make their truth more evident.
[1.8.53] The tenth theorem is: 43
[Theorem 1-10] For any sentence with a distributed term

there follows a sentence with the same term not distributed,
all else remaining the same, as a formal consequence - but
never the converse.
[1.8.54] The reason for <Theorem 1-10) is found in Theorem 1-8 and
[1.2.6]: a sentence with an undistributed term has more causes of its
truth than <the same sentence with) the term distributed, other things
being equal.
[1.8.55] Consequences by subalternation and conversions of univer-

sals into particulars hold by Theorem 1-10, such as
(109) Of-no-man an ass is running; therefore, of-every-man some
ass is not running
and soon.
[1.8.56] The eleventh theorem is: 44
[Theorem I-II] From no sentence with some term

suppositing only confusedly by the distribution of the
preceding term does there follow a sentence with that term
suppositing determinately not free of distribution, but there
does follow <a sentence with the term> free of distribution.
[1.8.57] Theorem 1-11 is proved just like Theorem 1-10, and so

(110) Every B is A; therefore, A is every B
does not follow, but
(111) <Every B is A>; therefore, A is B
does follow, with the restrictions on supposition given <in Theorem
1-11 >. Likewise,
(112) Of-any-man an ass is running; therefore, the ass of-any-man
is running
(113) <Of-any-man an ass is running>; therefore, an ass of-some-
man is running
correctly follows.
[1.8.58] The twelfth theorem is:
[Theorem 1-12] (a) No such consequences are formal:

(114) B was A; therefore, what is B was A.
(115) B will be A; therefore, what is B will be A.
(116) B is A; therefore, what is B is A
but (b) the converses of (114)-(116) are formal con-
sequences; (c) if «114)-(116» were stated as universal

affirmatives, none of them would be formal consequences;
again, (d) none of these are formal consequences:
«117) Nothing which is B was A; therefore, no B was A>.
«118) Nothing which is B will be A; therefore, no B will be A>.
(119) Nothing which is B is A; therefore, no B is A
but (e) the converses of (117)-(119) are formal <con-
sequences>; (f) if «117)-(119» were stated as particular
negatives, none of them would be formal consequences.
[1.8.591 Theorem 1-12 makes six claims. Now Theorem I-12(a) is

clear, because the locution "B is A" permits the ampliation of the
subject if the predicate is ampliative, as in
(120) A man is dead
but the locution 'what is B is A' does not permit the ampliation of its
subject, namely B; indeed, it <contracts> and restricts <the supposition
of its subject> to present things by the present-time verb 'is' which
precedes it. Thus if the predicate is ampliative, "B is A" has more
causes of its truth than "what is B is A," and from more to less is not
an acceptable consequence <by Theorem 1-8>.
[1.8.60] Theorem I-12(b) is clear by the same <reasoning>. Note for

Theorem I-12(b) and Theorem I-12(e) that by "the converse" we do
not mean that the terms <of the sentence>, i.e. the subject and
predicate, are transposed, but a transposition of the locution "B is A"
and "what is B is A." Then Theorem I-12(b) is proved as follows:
there are no causes of the truth of "what is B is A" which are not
causes of <the truth of> "B is A," except for counterexamples involving
God-for
(121) What is the Father is the Son
is allowed in the case of God, but
(122) The Father is the Son
should not be allowed. But this is special to the Holy Trinity in that
it is together a trinity and an undivided unity, which non-Christian
philosophers do not generally admit.
[1.8.61] 1 state that names are conventional; 1 call consequences

'formal' which have no counterexamples except in the case of God,
even though strictly speaking they ought not to be called formal.
[1.8.62] Theorem 1-12(c) is clear because each sentence has some

cause of its truth which the rest do not have. Although the term is
ampliative,
(123) Everything which is B is A
is true only if it is the case for all <existing things>, and
(124) Every B is A
does not have this cause of its truth. For example, as soon as Noah
left the Ark it was true to say
(125) Everything which is a man was in Noah's Ark
because it follows that everything which was a man was in Noah's Ark.
Thus more things are distributed in "Every A is B" if the predicate is
ampliative than are <distributed> in "Everything which is A is B," and
the wider distribution does not follow from the narrower distribution
- for example,
(126) Every man ... ; therefore, every animal ...
<does not follow>. Similarly, the converse (124), "every B is A," can be
true although nothing is B, as for instance
(127) Every rose is-able-to-be-understood
<can be true> even when there are no roses, and then (123) is not true.
[1.8.63] Theorem 1-12(d)-(f) are proved by Theorem 1-12(a)-(c),

already demonstrated, along with Theorem 1-3.
[1.8.64] Note that a sentence with the subject ampliated by the

predicate should be analyzed as a disjunction in the subject of the
present time and the time(s) of the ampliation, as in the following
pairs:
(128) B will be A.
(128*) What is or will be B will be A.
(129) A man is dead.

(129*) What is or was a man is dead.
(130) The Antichrist can be a man.
(130*) What is or can be the Antichrist can be a man.
(131) A rose is understood.
(131 *) What was or is or will be a rose is understood.
[1.8.65J It is clear that there is no formal consequence from 'is' used

predicatively (tertium adiacens) to 'is' used existentially (secundum
adiacens). For instance, these do not follow:
(132) The Pope is dead; therefore, the Pope is
(133) A rose is able-to-be-understood; therefore, a rose is
but this does follow:
(134) The Pope is dead; therefore, the Pope was or is

and so for the others, each in their own way.
[1.8.66J The thirteenth theorem is:
[Theorem 1-13J For all sentences with no ampliative predic-

ate, (a) consequences such as "B is A; therefore, what is
B is A" are acceptable, as is the converse; (b) so too for
<consequences such as> "Every B· is A; therefore, everything
which is B is A" and its converse; (c) so too for negatives.
[1.8.67J <Proof>: no term supposits for anything else or in a different

way in each, and the copula is the same; <thus Theorem 1-9 applies>.
[1.8.68] Similarly, under the same assumptions <as Theorem 1-13> the
consequence from 'is' used predicatively to 'is' used existentially is
valid.
[1.8.69] The fourteenth theorem is:
[Theorem 1-14] (a) From any universal or particular

affirmative assertoric <sentence> there follows a particular
affirmative by conversion of the terms; (b) from any
universal negative there follows a universal and a particular

negative; and (c) no <sentence> follows from a particular
negative by its form <alone>.
[1.8.70] Theorem 1-14(b)-(c) should be understood about negatives

as negative <sentences> are commonly agreed to be formed. 45 1 call
the manner 'agreed-upon' when the negating negation precedes the
predicate, as for example
(135) No B is A.
(136) Some B is not A.
But 1 call the manner 'not agreed-upon' when the predicate precedes
the negation, and so is not distributed by it, as for example
(137) Every B A not is
(138) Some B A not is
Theorem 1-14 does not hold for negatives which are not in the agreed-
upon manner, for
(139) Every Sun a planetary body not is
is true and no universal follows by conversion, since the predicate is
not distributed in the antecedent and would be distributed in the con-
sequent when it becomes the subject <by conversion>; whence
(140) Every planetary body is not the Sun
is false. But from any particular negative <not formed in> the agreed-
upon manner there always follows another particular negative, since
if B A not is it follows that A B not is, with the same terms and
the same supposition of the terms, as in Theorem 1-9. This is why
Theorem 1-14 is stated to apply to negatives formed in the agreed-
upon manner, assuming that by "conversion of the terms" we under-
stand the transposition of the terms, that is, the subject becomes the
predicate and the predicate the subject.
[1.8.71] Even with these things out of the way, the proof <of Theorem
1-14> is long and difficult, due to the many and varied different kinds
of forms of assertoric sentences. 46 We shall now speak of these, and
put forward certain distinctions.
[1.8.72] Thus, 1 say that a categorical sentence ought to have a

subject, a predicate, and a copula.47 And these three <elements>
are sometimes implied in a single verb, as when 1 say "It-is-raining
(pluit)," "There-is-thunder (tonat)," or when 1 say impersonally
"There-is-running (curritur)," "There-is-reading (legitur)," "It-is-said-
(that) (loquitur)." Sometimes the subject is explicit and the predicate
and copula are implicit in the same verb, as "A man runs," "Man
exists," or "A man comes-to-be (fit)." And sometimes the predicate
is explicit and the subject and copula are implied together, as
"It-hap pens-that man runs (contingit hominem currere)," i.e. it is
contingent that a man is running, or in "It-is-required-that you study
(decet te studere)," i.e. you are required to study, or "Socrates sees
a man," i.e. Socrates is seeing a man. Yet sometimes all three are
explicit, as in "A man is white." Conversions are simplest for this last
case, and so let us begin with that.
[1.8.73] We say that the copula is either present-time, past-time, or

future-time. If it is present-time, then either both terms are ampliated
to some other time, or neither is ampliated, or only the predicate is
ampliated, or only the subject.
[1.8.74] If neither <the subject nor the predicate> is ampliated, then

the conversion of a universal negative and a particular affirmative is
obvious, since they are each in the same terms and the terms supposit
in the same way, and they have the same copula, and so mutually
follow from one another by Theorem 1-9. If a particular affirmative
follows from a particular affirmative, then <that particular affirmative>
follows from the universal, since whatever follows from the consequent
follows from the antecedent. But a universal does not follow from a
universal affirmative, since the predicate of the first is not distributed
in the first and it is distributed in the second. And likewise it is clear
that a particular negative is not converted, since its subject is not
distributed and in the <proposed conversion>, when it would be the
predicate, it would be distributed.
[1.8.75] Thence, continuing to assume that the copula is present-time,

if the predicate or the subject is ampliated, or both, then either the
sentence to be converted is formed with the locution "B is A," "B is
not A," and similarly for universals, or <the sentence to be converted
is formed> with the locution "what is B is A," "what is B is not A," and
similarly for universals. If, therefore, the term in the converted as well
as in the original <sentence> is formed with the "what is ... " <locution>,
then it is treated just as if it were not an ampliated term, for this locu-
tion prevents any ampliation of the subject by the predicate. And so
we infer that those conversions with the "what is ... " <locution> are
formal; they hold for ampliated terms as well as they hold for non-
ampliated <terms>.
[1.8.76] If, however, the sentences to be converted are formed without

the "what is ... " <locution>, the affirmatives must be converted by
forming a disjunction in the subject of the converted sentence of the
present time and the time(s) of the ampliation, as for example
(141 ) A man is dead; therefore, something dead was or is a man.
(142) A rose is able-to-be-understood; therefore, something able-
to-be-understood was or is or will be or can be a rose
and so on for the others. Otherwise we would move from more causes
of truth to less, or the first sentence would have some cause of its
truth which the second <sentence> lacks, and so would not be an
acceptable consequence. But the universal negative <sentence>, if the
predicate is ampliated, does not require this sort of disjunction,
provided that its subject is not ampliated. This is because the supposi-
tion of the predicate of the first <sentence> is not changed but only the
supposition of the subject of the first <sentence> is changed, from
ampliated to non-ampliated - in which manner, negatively and distri-
butively, the consequence is acceptable. Denying in the case of more is
also denying in the case of less, and denying a disjunctive <sentence>
denies each part. For example, if nothing which was or is B is A, it
follows that nothing which is B is A and also that nothing which was B
is A,
[1.8.77] Moreover, if the sentences are formed without the "what is

... " <locution> and the subject is ampliated and the predicate is not,
then the affirmatives are converted without a disjunction and without
<using> the "what is ... " <locution>. For this takes us from the non-
ampliated to the ampliated and undistributed, and so from fewer
causes of truth to more <causes of truth>, and thus is an acceptable
consequence. But the universal negative must be converted by using
l. GENERAL AND ASSERTORIC CONSEQUENCES 215
the "what is ... " <locution> in the subject of the converted <sentence>
to prevent ampliation; for example,
(143) Nothing dead is a man; therefore, nothing which is a man is
dead.
Otherwise we would go from the less ample distribution to the more
ample distribution, which would not be an acceptable consequence.
[1.8.78] If each term is ampliated, then (as before) affirmatives

must be converted by a disjunction and negatives by the "what is ... "
<locution>.
[1.8.79] Now if the copula is past-time or future-time, then the

sentence is formed either with or without the "what is ... " <locution>.
If <it is formed> with the "what is ... " <locution>, then we get a conver-
sion retaining the same phrasing, as in
(144) What is white was black; therefore, what was black is white.
(145) Nothing which was white is black; therefore, nothing which
is black was white.
(146) What was white will be black; therefore, what will be black
was white
because the terms remain the same and supposit for the same. Now
these conversions may be proved by an expository syllogism; however,
if the sentences are given without the "what is ... " <locution>, as in "A
white thing was black" or "Nothing white was black," then we must
add the word "what" <or "which"> to the subject of the converted
<sentence>, with the verb preceding <the subject> and putting 'is' as the
copula, using the grammatical disjunction in the predicate of the con-
verted <sentence>. For example:
(147) A white thing was black; therefore, what was black was or
is white.
(148) Nothing white was black; therefore, nothing which was
black was or is white.
The remaining terms supposit for the same things in the same way.
[1.8.80] If the subject is explicit and the predicate and copula are
implicit in the same verb, then, if possible, the predicate should be
made explicit by the copula, and then the conversion works as before.
For example,
(149) A man runs.
(149*) A man is running.
(150) A man will-run ( curret).
(150*) A man will-be running (erit currens).
[1.8.81) Sometimes, though - perhaps because of a defect of words
under the rules of grammar, perhaps because of the condition of their
signification - no such explicating sentence will serve, as with "A man
comes-to-be ifit)," "A man is-loved (amatur)." There is no present-
time participle for these verbs <'come-to-be' and 'being-loved'>, and so
there is no simple resolution <of such sentences into sentences where
all three elements are explicit>. For if I say "A man is made ifactus)"
or "A man is having-been-loved (amatus)," it is dubious whether they
are equivalent to the original <sentences>, since people might offer as
counterexamples to the case of love where a man is <now> loved but
still yet is not having-been-loved, and <for the case of coming-to-be>,
what comes-to-be does not yet exist, as Aristotle says, and so is not
made <already>.
[1.8.82) In such cases the sentence should be converted by the addi-
tion to the subject of "what" for the name, with the verb, and by the
addition of the copula 'is: And if the predicate is ampliated there
should be along with the foregoing a disjunction in the predicate of the
converted sentence, as was said before. For instance,
(151) A man is-loved; therefore, what is-loved is a man.
(152) A rose is-understood; therefore, what is-understood was or
is or will be or can be a rose.
[1.8.83) Because the verbs 'comes-to-be; 'is-generated,' and 'begins-
<to-be>' indicate that there is something which did not exist before, or
there will be <something> which does not <now> exist, and the verbs
'ceases-<to-be>' and 'is-corrupted' and others of this sort <indicate> the
converse, namely that something exists and afterwards will not exist, or
<something> did exist although it does not <now> exist, conversions of
these verbs take place in this way:
(153) A comes-to-be; therefore, what comes-to-be is or will be A.
(154) A is-corrupted; therefore, what is-corrupted is or was A.
If the predicate follows the verb 'comes-to-be: as in "A white thing

comes-to-be black" or "8 comes-to-be A," it is safest to convert this
way:
(155) <A white thing comes-to-be black>; therefore, what comes-
to-be black was or is white.
(156) <8 comes-to-be A>; therefore, what comes-to-be A was or
is B.
The verbs 'begins-to-be' and 'ceases-to-be' are treated the same way.
[1.8.84] But the verb 'is' taken existentially should, 1 believe, be

resolved into 'is' taken predicatively and the participle 'being' (single
number) and 'beings' (plural number) under a disjunction, if the
consequence is to be formal. For God is, and an army is, and God is a
being and not beings, and an army is beings not a being;4K hence
(157) A is; therefore, A is a being
does not follow as a formal consequence, but rather
(158) <A is>; therefore, A is a being or beings
<is a formal consequence>. Thus we may convert:
(159) A is; therefore, a being is or beings are A
and we may break up the predicate, referring each to the appropriate
verb-form, with the compound <sentence>
(160) <A is>; therefore, a being is A or beings are A.
Nevertheless, if in a particular case it seems that the subject sup posits
for no one thing but only collectively for many which are not some
one thing, <the sentence in question> should be converted as
(161) A is; therefore, beings are A
for example
(162) The poulace is; therefore, beings are the populace.
[1.8.85] We may conclude as a corollary <of [1.8.84]> that

(163) A is; therefore, A is a being
does not follow as a formal consequence, but only that <A> is a being
or beings. Similarly,
(164) A and B are; therefore A and B are beings
does not follow, because if <A and B> are synonymous names suppo-
siting only for the same simple things, as for example both names
imposed synonymously to signify Peter's soul, then A and B are, but
nevertheless they are not beings but a <single> being, as the Creator
and God are and are not some things but some thing.
[1.8.861 Since we have begun to discuss collective names, it should

be remarked that there is great doubt about how these should be
converted: 4Y
(165) An Army is round.

(166) An Army is square
because it is false to say that a square is an army and that squares are
an army. Grammarians, not allowing adjectives to supposit unless they
are used substantively in a neuter gender, do not even permit us to say
"A round is an army." I therefore say that because (165) is true only if
the army is a round army, and not because it is something round or
some round things, should be converted by supplying in the predicate
of the <sentence> to be converted the substantive <term> 'army,' and
then the conversion is obvious:
(167) <An army is round>; therefore, a round army is an army.
[1.8.871 Now if the predicate is explicit and the subject and copula are
implicit in the verb, then it is possible to make explicit what the verb
takes as its subject (quod sibi reddit suppositum); for example:
(168) It is required that you act correctly.
(168*) A requirement is your acting correctly.
(169) It happens that a man runs.
(169*) A contingent <fact> is <that> a man runs.
(170) Socrates sees a man.
(170*) Socrates is seeing a man.
And if there is no participle of the verb either imposed or in use, then
create a name, supply what needs to be supplied, and analyze what
needs analysis, until we explicitly have a subject and a predicate and a

copula, as for example:
(171) It is of interest to the king to do well.
(171 *)The king's interest is in doing well.
(172) My life disgusts my soul.
(172*) My life is disgusting to my soul [or: Something in my life <is
disgusting to my sou!>].
(173) It pains me to have been a man.
(173*) <What is> paining me is <my> being a man.
F or the business at hand we do not care whether sentences of this sort
are true according to the properties of the terms or not, but only how
they may be resolved so as to make the subject, predicate, and copula
apparent.
[1.8.88] For sentences in which the whole is implicit in a single verb, I

say that they should be resolved as in [1.8.72] so that the whole is
explicit, as for example:
(174) It-is-raining.
(174*) Rain falls from the sky.
(175) It-is-snowing (ninguit).
(175*) Snow falls from the sky.
(176) There-is-thunder.
(176*) Thunder is (tonitruum est) or, A sound is produced in the
sky.
or however you want to analyze the verbs. <We deal with verbs used
impersonally> in the same way, e.g.
(177) There-is-running.
(177*) Something is running.
(178) There-is-reading.
(178*) Some man or woman reads or some men or women read.
[1.8.89] One must be careful in some conversions lest by the change a

grammatical error is made. For example, this does not follow:
(179) No phoenix (nullus phoenix) is a woman; therefore, no
woman (nulla mulier) is a phoenix
because in the first sentence 'phoenix' was restricted to the masculine,

and it was not in the second. Similarly,
(180) No man (nullus homo) is a woman; therefore, no woman
(nulla femina) is a man
does not follow. Rather, so that a term may supposit in the antecedent
and the consequent for the same, something restricting the general
kind to the sort <of thing} which is dealt with in the antecedent should
be put in the consequent, by saying
(181) <No man is a woman}; therefore, no woman is some man
(nulla femina est aliquis homo).
(182) <No phoenix is a woman}; therefore, no woman is some
phoenix (nulla mulier est aliquis phoenix).
[1.8.90J Again, it is also necessary to pay attention to oblique terms,

about which there are different views. 50 If an oblique term alone
precedes the copula, many people say that it alone is the subject of the
sentence, as in
(183) Socrates sees a horse (Equum videt Socrates).
Indeed, even if a nominative (rectus) term precedes the verb with the
oblique term, and the oblique term comes before the nominative term,
they say that the oblique term alone is the subject and the whole
remainder the predicate. They say that <such sentences} should be
resolved so that the oblique term is changed into a nominative term,
the copula 'is' added, and the relative pronoun "what" <or "which"} put
into the predicate in the case the oblique term was in. For example,
(184) A man is riding a horse (equum equitans est homo).
(184*) There is a horse, riding which is a man (equus est quem
equitans est homo).
(185) Of-any-man an ass is running.
(185*) Any man is, of whom the ass is running
and so forth for other cases; and then it is clear how they should be
converted.
[1.8.91J I say that there is no doubt that a consequence is acceptable

after such a resolution, so long as care is taken first to make sure there
is no ampliation of the subject of the <sentence> the resolution must

include a disjunction. For
(186) Noah saw a man; therefore, there is a man whom Noah saw
(186*) <Noah saw a man>; therefore, there was or is a man whom
Noah saw
does follow. Neither does
(187) The preachings of the Antichrist are going-to-come; there-
fore, the Antichrist is
follow, but rather
(187*) <The preachings of the Antichrist are going-to-come>; there-
fore, the Antichrist is or will be
does follow.
[1.8.921 Nevertheless, I do not affirm that such an oblique term is the

subject of the sentence on this account; rather, I believe that a
nominative term, either expressly or as understood, is always to be put
as the subject, either by itself or with the determinations given in the
sentence. Thus in (185) I believe the subject to be the whole "ass of-a-
man," and it is converted as
(185**) <Of-any-man as ass is running>; therefore, the one running
is an ass of-a-man.
One must distribute according to how confused supposition is changed
into determinate <supposition>. Thus we convert:
(188) Of no contradiction is each part true.
(188*) Nothing true is each part of some contradiction.
(189) A man is seeing a horse (homo equum est videns).
(189*) The one seeing a horse is a man (equum videns est homo).
I believe that in (189) the word 'horse' pertained to the predicate even
though it was put before the copula. Thus even if I say
(190) Any horse a man is seeing (omnem equum homo est
videns).
I believe that 'horse' still pertains to the predicate as a determination

of it, and so (190) is converted as
(190*) A certain one seeing a horse is a man (Quoddam videns
equum est homo).
[1.8.93] Whether the view of [1.8.92] or that of [1.8.91] is true makes

no difference to the theory of consequences (ars de consequentiis). It is
sufficient that there are acceptable consequences in either of the ways
given.
[1.8.94] Finally, it should be noted that if certain special syncategore-

matic terms are found in sentences, so that the sentences need a
great deal of analysis to express their senses (as <for instance> excep-
tive, exclusive, or reduplicative sentences), then they should be rather
called hypothetical (hypotheticae) and not categorical; they are first to
be resolved according to analysis and thence converted, according to
the requirements of their analysis. For example,
(191) Every man except Socrates is running
<is correctly analyzed as>
(191 *) Socrates is not running and every <man> other than Soc-
rates is running
and is therefore converted as
(191 **) Socrates is not a runner and some man other than Socrates
is a runner.
But the <sentence>
(192) Every animal with the exception of a certain man is running
is only converted with respect to a single analysis, for a particular
negative is not converted. But <the sentence>
(193) Man is only an animal
<is analyzed as>
(193*) Man is an animal and man is nothing other than an animal
which is converted
(193**) Man is an animal and no man is <anything> other than an

animal.
And since the conjunctive sentence (193**) is equivalent to the univer-
sal affirmative
(193***) Every man is an animal
we say that (193) is converted as (193***).
[1.8.95) And these remarks about converting sentences formed with

common terms are sufficient <to prove Theorem 1-14>.
[1.8.96) The fifteenth theorem is:

[Theorem 1-15] From any singular sentence there follows
another sentence by conversion of the terms; for example,
"Socrates is running; therefore, <some> runner is Socrates"
"Plato is not running; therefore, no runner is Plato".
[1.8.97) The supposition of the terms remains the same. But you
should pay attention to ampliation, as has been pointed out. This does
not follow:
(194) The Antichrist is going-to-be-generated; therefore, <some-
thing> going-to-be-generated is the Antichrist.
But this does follow:
(194*) <The Antichrist is going-to-be-generated>; therefore, <some-
thing> going-to-be-generated is or will be the Antichrist
[1.8.98) <We should also note the following>:

[Remark) If a singular <sentence> has a common term with
a demonstrative pronoun added to it, then it can be con-
verted simply, that is, to another singular <sentence> with
the same term <subject to the> demonstrative.
For example,
(195) That man is running; therefore, that runner is a man.
(196) That man is not running; therefore, the runner is not this 51
man.
[1.8.99J Note, incidentally, that from any such singular <sentence>

there follows a particular or indefinite <sentence>, except in the case
where in a negative <sentence> something is pointed out for which the
common term may not supposit. For if one points to wood, then
(197) That man is not running; therefore, man is not running
does not follow.
[1.8.100J The sixteenth theorem is: 52
[Theorem 1-16J Conversion by contraposition (a) is not a

formal consequence, but (b) can be <a formal consequence>
under the assumption that all of the terms supposit for
something.
[1.8.10 1J Conversion by contraposition is not strictly conversion,

since the terms do not remain the same but are changed according to
finite and infinite.
[1.8.102J And this sort of conversion <namely by contraposition> of a

universal affirmative fails <to produce an acceptable consquence> when
a finite predicate supposits for all beings, since then the infintie term
opposite to it supposits for nothing, and thus a true affirmative cannot
be formed from it. For example,
(198) Every man is a being; therefore, every non-being is not a
man
does not follow, since the first <sentence of (198» is true and the
second is false.
[1.8.103J Similarly, the conversion <by contraposition> of a particular

negative fails when the finite subject supposits for nothing, since then
the infinite term opposed to it supposits for all beings, and so cannot
be truly denied of some term which supposits for something. For
example,
(199) A chimaera is not a man; therefore, a non-man IS not a
non-chimaera
does not follow, since the first <sentence of (199» is true and the
second false - for the contradictory of the second part, namely "every
non-man is a non-chimaera," is true. And so Theorem I-16(a) is clear.
[1.8.104] Theorem I-16(b) is shown by assuming that the negation

acts to product an infinite <term>, which is what keeps a term from
suppositing for that which it supposits when taken finitely and makes it
supposit for all other things than it supposits for <normally>; thus the
following are equipollent:
(200) Non-man <is not man>.
(200*) <Non-man is something other than man>.
(201) <Non-animal is not animal>.
(201) <Non-animal is something> other than animal
unless someone could cancel the given equipollence immediately by a
defect in the supposition of the predicate <in (200*) and (201*», in
that Aristotle says in Metaphysics IV [cap. iii 1005b 19ff.] that every
being is the same as or different from every other being, but being is
not the same as or different from being; thus although man is a non-
chimaera, nevertheless <man> is not <something> other than a chimaera.
[1.8.105] Assuming these things, then, it is clear that the opposite of

the antecedent follows from the opposite of the consequent. Indeed,
the first consequence is
(202) Every B is A; therefore, every non-A is not B
or <equivalently>
(202*) <Every B is A>; therefore, everything other than A is
<something> other than B.
The opposite of this consequence is
(203) Something other than A is not <something> other than B
and from (203) it follows that <something> other than A is the same as
B, in that every being is the same as or different from every being, and
we assume non-A as well as A to be beings, just as non-B and B.
Thence it follows that:
(204) <Something> other than A is the same as B; therefore, B is
<something> other than A.
226 TREA TISE ON CONSEQUENCES
(205) <Something other than A is the same as B>; therefore,

nothing other than A is B.
(206) <Something other than A is the same as B>; therefore, no B
is <something> other than A.
(207) <Something other than A is the same as B>; therefore, every
B is the same as A
in that every being is the same as or different from every being; and so
it follows that every B is A, which contradicts the first.
[1.8.106] It can also be shown that the conversion <by contraposi-

tion> of a particular negative <holds>: if we take the consequence
(208) Some B is not A; therefore, something other than A is not
<something> other than B
then the contradictory of the consequent is "anything other than A is
other than B," and further "nothing other than A is B," and further yet
"no B is other than A," and thus "every B is the same as A," in that
every being is the same as or different from every being; and so it
follows that every B is A, which is contradictory to (208).
[1.8.107] The seventeenth theorem is:

[Theorem 1-17] (a) From any affirmative there follows a
negative by changing the predicate according to finite and
infinite, keeping the rest the same, but (b) there is no
formal consequence from a negative to an affirmative,
although (c) there is a consequecne under the assumption
that all of the terms supposit for something.
[1.8.108] Theorem 1-17(a) is true because a term as finite and as

infinite do not supposit for anything the same; hence of whatever one
can be truly affirmed the other must be truly denied. This consequence
is shown to be evident because the opposite of the antecedent follows
from the opposite of the consequent. For example, if these are
consequences:
(209) Every B is A; therefore, no B is non-A.
(209*) <Every B is A>; therefore, no B is <something> other than A
then the opposite of the consequent is "some B is <something> other
J. GENERAL AND ASSERTORIC CONSEQUENCES 227
than A," from which it follows that this B is not the same as A, since
nothing is the same and <different from> one thing. Therefore, some B
is not A, which is contradictory to (209)-(209*). We can prove the
same for the consequence
(210) Every B is non-A; therefore no B is A
because the opposite of the consequent is "some B is A," and thus
some B is not <something> other than A, which contradicts (210),
which says that every B is <something> other than A.
[1.8.1091 Theorem I-17(b) is clear, since a consequence fails when the

subject does not supposit for anything; for example,
(211) A chimaera is not a man; therefore, a chimaera is a
non-man
does not follow.
[1.8.1101 Theorem I-l7(c) is also clear: if B exists, it must either be A

or <something> other than a, and so if it is not A it is <something>
other than A and if it is not <something> other than A then B is A.
[1.8.1111 And this finishes Book I, which is about consequences

in general and also particularly about formal consequences of one as-
sertoric categorical to another.
BOOK II
CONSEQUENCES AMONG MODAL SENTENCES
CHAPTER II-I. MODAL SENTENCES
[2.1.11 In Book II we deal with particular consequences, namely those

among modal sentences. Certain things should be laid down about
them.
[2.1.21 First, we assume that although there are many kinds of modal
sentences the principal <kinds>, which Aristotle and other Masters
treat, are modal sentences de possibili and de impossibili, de neces-
sario and de contingenti, and de vero and de fa/so - which is why I
intend to deal only with these sorts.
[2.1.31 But note that sentences are not called de necessario or de pos-
sib iii insofar as they are necessary or possible, but rather in that the
modes 'possible' or 'necessary' are found in them; similarly for the
other modes. For a sentence may well be necessary which is yet not de
necessario, or possible which is not de possibili, and conversely. For
example, according to Aristotle
(212) Man is an animal
is necessary; nevertheless, it is purely assertoric, while
(213) A man of necessity is running
is de necessario, even though it is not necessary but false and impossi-
ble; and
(214) It is possible that a man is an ass
is de possibili although impossible.
CHAPTER II-2. COMPOSITE AND DIVIDED MODAL SENTENCES
[2.2.11 In Chapter 11-2 we lay it down that modal sentences of this sort
are commonly put in one of two ways: some are called 'composite' and
some 'divided:
228
II. CONSEQUENCES AMONG MODAL SENTENCES 229
[2.2.2] <Modal sentences> are called composite when the mode is put
as the subject and the dictum is predicated, or conversely.
[2.2.3] (I call the terms 'possible,' 'necessary,' 'contingent,' and so on

the modes, and 1 call the dictum that whole found in the sentence
aside from the mode and copula and negations and quantifiers, or
other determinations of the mode or of the copula. I)
[2.2.4] For example, these are composite:

(215) That a man runs is possible
(216) It is of necessity that man is an animal
The subject of (215) is 'that a man runs' and the predicate 'possible';
the subject of (216) is 'it is of necessity' and the predicate 'that man is
an animal.'
[2.2.5] <Modal sentences> are called "divided" when part of the dictum
is put as the subject and the other part as the predicate. Moreover, the
mode is taken with the copula as a determination of it. For example,
(217) A man is capable of running
(218) A man is possibly running
(219) A man is of necessity running
<are divided modals >, and so forth. In (217)-(219) the subject is 'man'
and the predicate 'running.' The copula, though, is 'is capable of' or 'is
possibly' or 'is of necessity' or 'is necessarily.' The sentence (217) can
be analyzed as
(217*) A man can possibly be a runner
just as "A man is running" <is analyzed as> "A man is a runner."
CHAPTER 11-3. AFFIRMATIVE AND NEGATIVE DIVIDED

MODALS
[2.3.1] Divided <modal> sentences are simply affirmative when no

negation is found in them, as for example
(220) A man can-possibly-be white
(221) God must-necessarily-be just
[2.3.21 <Divided modal sentences> are negative in two ways. The first
way is when the negation occurs in the mode and so precedes it, as for
example in
(222) A man can-not-possibly-be an ass
(223) No man can-possibly-be an ass
The other kind <of divided modal negative> is when the negation does
not occur in the mode but follows it, as for example in
(224) A man can possibly not-be an ass
(225) God must necessarily not-be evil
[2.3.31 Now some people wonder whether (224)-(225) should be

called affirmatives or negatives, simply speaking. Whatever others may
say, I believe that they ought to be called negatives, because on the
one hand the sentence "B can not-be A" is equipollent to "B must not
necessarily-be A," which is obviously negative;2 and on the other hand,
because an affirmative sentence is not true if some term does not sup-
posit for anything, and nevertheless
(226) A chimaera must necessarily not-be an ass
is true, and consequently
(227) A chimaera can possibly not-be an ass
<is also true>.
[2.3.41 There are other <sentences> which have a double negation, one
in the mode and the other in the predicate, as for instance
(228) B can not possibly not-be A
(229) B must not necessarily not-be A
I believe that (228)-(229) are in fact affirmatives, since they are equi-
pollent to sentences which are obviously affirmatives. For (228) is
equipollent to
(228*) B must-necessarily-be A
which is purely affirmative, and
(230) No B must necessarily not-be A
is equipollent to
(230*) Every B can be A
which is purely affirmative.
(2.3.5] And so it is clear how certain divided modal sentences, af-

firmative as well as negative, have the mode denied and others have
the mode not denied.
CHAPTER II-4. AMPLIATED TERMS IN DIVIDED MODAL

SENTENCES
(2.4.1] In Chapter 11-4 we assume that a divided sentence de possibili

has a subject ampliated by the mode following it to suppa sit not only
for things which exist but also for those things which can exist,
although they do not <in fact exist>. In this way it is true that air can
come from water, although this is not true for any air which exists. 3
And so the sentence
(231) BcanbeA
is equivalent to
(231 *) What is or can be B can be A
(2.4.2] Some say that (231) is equivalent to a single disjunctive hypo-

thetical sentence, namely.
(231 **) What is B can be A or what can be B can be A
[2.4.3] But I don't believe that it is <equivalent>, for it is very different

to say (231*) and (231 **).4 For by holding <the equivalence suggested
by (231 *» the sentence
(232) God-while-creating can not-be God
is false since its contradictory is true, no matter what the circumstances.
Nothing which is or can be God-while-creating can not-be God. For
all and only God can be God-while-creating, and He cannot not-
be God. But if <we adopt the equivalence suggested by (231**», we
must allow (232) in the case where God is not now creating, since it
would be equivalent to the disjunctive <hypothetical sentence>
(232**) He who is God-while-creating can not-be God or He who
can be God-while-creating can not-be God
And (232**) should be allowed, because the first <disjunct> is true in
the posited case, since God-while-creating is nothing, and what is
nothing can not-be God, or God-while-creating is not God, and what
is not God can not-be God.
[2.4.4] Note that those two <proposed equivalences> agree in holding

that there are three causes of the truth of the sentence
(233) B can not-be A
The first <cause of the truth of (233» is when a certain thing is Band
that thing can not-be A; the second is when a certain thing can be B,
though it is not <now> B, and it can not-be A. The third cause <of the
truth of (233», according to the second <form of equivalence>, is that
nothing is B which could be A, although it could be - which is not
sufficient for the first <form of equivalence, which requires that> it is
necessary that nothing which is or could be B could be A.
[2.4.5] I say that <the same sort of equivalence> holds similarly for
past-time or future-time sentences. The sentence
(234) B will be A
is equivalent to
(234 *) What is or will be B will be A
and is not equivalent to the disjunctive <sentence>
(234 **) What is B will be A or what will be B will be A
These two <forms of equivalence> differ because if God IS not now
creating and tomorrow will create, then
(235) God will not be the one creating
is false, and would be true <by the second proposed equivalence>.
[2.4.6] The reason why I hold the first equivalence and not the second
is that if I say
(236) Any B can be A
there is here a unique predicate and a unique subject and a single sim-
ple categorical sentence, and the subject is distributed by a unique dis-
tribution. Thus it seems better that it is analyzed as a single categorical
sentence with one subject and one predicate, even if there is a disjunc-
tion of 'is' and 'can be' in the subject of the analyzed sentence, by the
ampliation of the subject. And the manner of analyzing universals and
particulars will be the same, namely with a disjunction of 'is' and 'can
be.' But according to the other <proposed form of equivalence> parti-
culars must be analyzed by disjunctions <of sentences> and universals
by conjunctions <of sentences>. I hold that
(237) Anything creating was God
is true even if nothing is now creating, and (237) would be false
according to the other equivalence.
[2.4.7] Nevertheless, I do not hold it capable of definite proof that

either of the two ways cannot be held, for names and expressions are
significant by convention. Still, in what follows I shall speak in the first
way.
CHAPTER II-S. EQUIPOLLENT <MODAL SENTENCES,
[2.5.1] In Chapter 11-5 I set forth what Aristotle and others assume,
namely that 'necessary' and 'impossible not to be' are equipollent, and
that 'necessary not to be' and 'impossible to be' are equipollent. It
seems obvious on the face of it that what is necessary is impossible
not to be and conversely, and what is necessary not is necessary not to
be is impossible to be.
[2.5.2] I take 'impossible' and 'not possible' to be equipollent, since

there is a negation implicit in 'impossible.' Thus these are equipollent:
(238) B is not possibly A
(238*) B is impossibly A
(239) B cannot be A
(239*) B is not possibly A
since 'can be' and 'possibly is' signify the same. Similarly, these are
equipollent:
(240) Every B is impossibly A
(240*) Every B is not possibly A
(241) No B is possibly A
(241 *) No B can beA
[2.5.3] I also claim that universal affirmatives contradict particular

negatives and universal negatives <contradict> particular affirmatives in
the same way, where the negation in the negative is put into the mode.
For example, these are contradictory pairs:
(242) Every B is possibly A
(242*) Some B is not possibly A
(243) NoB can beA
(243*) Some B can be A
and similarly for the other modes.
CHAPTER II-6. THEOREMS ABOUT DIVIDED MODALS
[2.6.1] We finally assume in Chapter II-6 that modal sentences can be

propounded without any restriction of the subject by "what is ..." or
"what was ... " or others of this sort, such as <the modal sentences>
(244) Every B can be A
(245) Every C is necessarily D
and so forth.
[2.6.2] Sometimes <modal sentences> can be propounded by adding

"what is ... or "what was ..." or "what can be ... " (or something of
the sort) to the subject. For example,
(246) What is B can be A
and then B is restricted to supposit only for those things which are B.
Similarly,
(247) What was B is necessarily A
where B is restricted to supposit only for the things which were B, and
likewise for future-time sentences. We can say the same for
(248) What can be B is necessarily A

(249) What is necessarily B happens to be A
These points are obvious.
[2.6.3] With these points covered, then, we first <give theorems> about
divided modals <in Chapter II-6> and then about composite modals <in
Chapter II-7 >, and first of all those in which the subject is taken with-
out any of the additions "what is ..." or "what can be ...", etc.

[Theorem II-i] From any sentence de possibili there follows
by equipollence another <sentence> de necessario, and from
any <sentence> de necessario <there follows> another <sent-
ence> de possibili, so related that if a negation were added
either to the mode or to the dictum (or to both) in one
<sentence> it is not added to it in the other <sentence>, and
if it were not added in the one it is added to the other, with
all else being the same. 5
[2.6.5] For example, these are equipollent:

(250) B is necessarily A
(250*) B is not possibly-not A
This is clear from Chapter II-5, for 'impossible' and 'not possibly' hold
for the same. By parity of reasoning these are also equipollent:
(251) Any B is necessarily A
(251 *) Any B is not possibly-not A
And consequently <equipollent to>
(251 **) No B is possibly not A
And thus it follows that the contradictories of (251 )-(251 **) are equi-
pollent, namely
(252) No B is necessarily A
(252*) Any B is possibly-not A
(253) Some B is not necessarily A
(253*) Some B is possibly not A
[2.6.6] Similarly, these two are equipollent:

(254) B is necessarily not A
(254*) B is not possibly A
for, as I have said <in Chapter 11-5>, 'not possible' and 'impossible'
hold for the same. By parity of reasoning, these are equipollent:
(255) Every B is necessarily not A
(255*) Every B is not possibly A
(255**) No B is possibly A
Therefore the contradictories of these are equipollent, namely:
(256) No B is necessarily not A
(256*) Every B is possibly A
(257) Some B is not necessarily not A
(257*) Some B is possibly A
[2.6.7] A corollaryl> may be inferred from Theorem II-I:
[Corollary] (a) If matters are settled for <sentences> de pos-
sibili having an affirmed mode, then they are settled suffi-
ciently for <sentences> de necessario having a denied mode,
for they are the same, other things being equal. In like man-
ner, (b) if matters are settled for <sentences> de necessario
having an affirmed mode matters are settled for sentences
de possibili having a denied mode, for they too are equipol-
lent, as has been said.
Thus we shall only deal with <sentences> de possibili and de necessario
which have an affirmed mode. And for the rest of Chapter 11-6, when
we speak of <sentences> de possibili or de necessario we shall always
mean those having an affirmed mode. Always keep this in mind in the
following theorems. Also, we shall not deal with <sentences> de impos-
sibili, since they are equivalent to <sentences> de possibili which have a
denied mode; 'not possible' and 'impossible' hold for the same, and so
<sentences> de impossibili are equivalent to those de necessario.7
[Theorem 11-2] In every divided sentence de necessario the
subject is ampliated to supposit for those things which can
be.
[2.6.9] Theorem U-2 is obvious, for otherwise <sentences> de neces-

sario would not be equipollent to those de possibili having a denied
mode, since in sentences de possibili the subject is evidently allowed to
be ampliated in this way.K Thus the sentence
(255) B is necessarily A
is analyzed as
(255*) What is or can be B is necessarily A
and the sentence
(256) Every B is necessarily A
is analyzed as
(256*) Everything which is or can be B is necessarily A
and negatives <are analyzed> similarly. This is manifestly clear if an
ampliation of this type is permitted in <sentences> de possibili, for
(257) B can be not A
contradicts
(257*) No B can be not A
which is equipollent to
(257**) Every B is necessarily A
The situation is the same for the analyzed sentences, since
(258) What is or can be B can be not A
contradicts
(258*) Nothing which is or can be B can be not A
and this is equipollent to
(258**) Anything which is or can be B is necessarily A
[2.6.10] From Theorem U-2 I conclude as a corollary that these sen-

tences are true:
(259) The one creating is of necessity God
(259*) The one creating is necessarily God
even in the posited case where God is not now creating. Similarly, the
universal, namely that everything creating is of necessity God, is true,
since it is equivalent to
(260) Everything which is or can be creating is of necessity God
and (260) is simply true and necessary. Nevertheless, if such sentences
with an ampliated subject are analyzed by sentences compounded by
disjunction or conjunction, that is, particulars by disjunctions and uni-
versals by conjunctions, as some claim,9 then
(261) Anything creating is necessarily God
is false and
(262) Something creating is not necessarily God
is true in the posited case.
[2.6.11] But it is objected that this is not so: what is not, is not neces-
sarily God, and in the posited case He is not creating; therefore <som-
ething> creating is not necessarily God. Again, anything which is
<creating> of necessity is God, but <in the posited case> nothing is
creating; thus the one creating is not of necessity God.
[2.6.12] These objections fail in the posited case:

(263) The one creating can be God
is conceded, even given that it is analyzed by a disjunction <of sen-
tences>.IO Therefore,
(264) Nothing creating can be God
should be denied, since it contradicts (263). Therefore, anything equi-
pollent to (264) should be denied, namely
(264 *) Anything creating is necessarily not God
And similarly
(264 **) Anything creating is possibly not God
should be denied. Thus the <sentences> contradictory to (264**)
should be conceded, namely either
(265) Something creating is of necessity God

(265*) Something creating is necessarily God
taking both (265) and (265*) as the same.
[2.6.13] Thus the objections are resolved by saying that anything which
is necessarily God - or rather anything which can be God - always
was and always will be. Thus, since the one creating is necessarily God,
the one creating always necessarily was and is and will be. And He
always was and is and will be, but He was not always creating, nor will
be creating, nor is now creating (in the posited case). With this state-
ment, then, the premisses of the objections are conceded and <their
conclusions> denied, by the syllogistic form, which will be clear in
what follows.
[2.6.14] The third theorem is:

[Thearem IJ-3] From no sentence de necessaria does there
follow an assertoric sentence, or conversely, unless an
assertoric universal negative follows from a universal nega-
tive de necessaria. I I
I speak always of those <sentences> with the mode affirmed, and also if
the subject is not restricted by a "what is ..." <locution>. I add this last
<qualification> because <the sentence> "What is B is necessarily A;
therefore, B is A" is an acceptable consequence.
[2.6.15] First of all, it is clear that a sentence de necessaria does not

follow from an assertoric sentence, since an assertoric sentence can
deal with contingent matters, which is not true of the sentence de
necessaria. For example, these do not follow:
(266) Every man is running; therefore every [or 'some'] man is
necessarily running
(267) No man is running; therefore, every [or 'some'] man is
necessarily not running
[2.6.16] Next, this does not follow:

(268) Anything creating is necessarily God; therefore, the one
creating is God
for the first <claim of (268» is true, positing that God does not create
now, and the second <claim of (268» is false. The cause of the falsity
of this consequence is that you deduce only one part of a disjunc-
tion. 12 Similarly, this does not follow:
(269) Some planetary body shining on our hemisphere is neces-
sarily not the Sun; therefore, some planetary body shining
on our hemisphere is not the Sun
bcause if we were to posit the case that only the Sun of all planetary
bodies is now shining on our hemisphere, then the first claim of (269)
would be true (as is clear by the analysis) and the second false. The
reason for this is that the subject in the sentence de necessario suppos-
its for things other than the Sun, due to its ampliation, and in the
assertoric sentence <the subject> does not supposit for other things.
[2.6.17) But that

(270) Every B is necessarily not A; therefore, no B is A
is an acceptable consequence is obvious. if B supposits in both parts
<of (270» for something, then in the first part B is distributed for all
those things for which it is distributed in the second <part>, and per-
haps (as may be) for more. Thus if the first <part of (270» is true for
all, then the second <part> will also be true for all. If, however, B does
not suppa sit for anything in neither <part of (270», then this will be
the cause of the truth of each, and so neither will be true without the
other. But if B supposits for something in the first <part> and for noth-
ing in the second <part>, then the second will be true whether the first
is true or false. And so no case can be given in which the first is true
with the second being false.
[2.6.18) The fourth theorm is:

[Theorem II-4j From no sentence de possibili does an
assertoric sentence follow, or conversely, except that a par-
ticular affirmative de possibili follows from any affirmative
assertoric sentence. 13
I speak, as always, <of sentences> with the mode affirmed.
[2.6.19] Theorem 11-4 is proved because particular affirmatives de

possibili contradict universal negatives de necessario, and universal

affirmatives <contradict> particular negatives, and similarly universal
negatives de possibili <contradict> particular affirmatives de necessario
and particular negatives <contradict> universal affirmatives. But there
are no consequences among assertoric and de necessario <sentences>,
except for universal negatives de necessario <by Theorem 11-3>. There-
fore, there will be <no consequences> among their contradictories
either, which are all assertoric <sentences> and all sentences de possib-
iii, except for the particular affirmative de possibili. The consequence
holds by Theorem 1-3.
[2.6.20] It is clear that a particular affirmative de possibili follows from

any assertoric affirmative <sentences>, since in Theorem II-3 the con-
tradictory of the antecedent followed from the contradictory of the
consequent <of this claim>.
[2.6.21] Theorem II-4 can be shown in another way, since it is clear

that if some B is A then that thing can be A. It is also clear that these
do not follow:
(271) Anything white can be black; therefore, <something> white
is black
(272) Anyone running can be not running; therefore, <someone>
running is not running
Nor yet does it follow that
(273) Someone creating is not God; therefore, someone creating
can be not God
because if God does not now create the first is true and the second is
false. The preceding may be confirmed by this conclusion, since they
are converted to one another by Theorem 1-3.
[2.6.22] It should also be known that

[Remark] A sentence de possibili follows from any sentence
de necessario, and not conversely.
This is true because what must necessarily be can possibly be, and
what necessarily is not is possibly not. (Since this is obvious I haven't
called it a "theorem.")
[2.6.23J The fifth theorem is:
[Theorem 1/-5] (a) From any affirmative de possibili there

follows by conversion of the terms a particular affirmative
de possibili, but not a universal,14 and (b) from no negative
de possibili does there follow by conversion of the terms
another sentence de necessario.
[2.6.241 Theorem II-5(a) is clear from an expository syllogism. If Bean

be A, then pick out that B <which can be A> and is this -C. Then: this
C is or can be B and the same can be A; therefore, what can be A is or
can be B - from which it follows that the very thing can be B, since
being able to be B follows from <actually> being B (by Theorem 11-4).
Thence it follows that:
(274) What can be A can be B; therefore, A can be B
Now the consequence (274) follows, because A supposits more widely
in the second part of (274) than in the first part, and so they are equi-
pollent.
[2.6.25J Thus it is immediately clear that a particular affirmative fol-

lows from a universal affirmative by conversion of the terms, since a
particular follows from a universal by subaIternation and a particular
follows from a particular by conversion and hence follows from the
universal, for whatever follows from the consequent follows from the
antecedent <by Theorem 1-4>. That a universal does not follow from a
universal is proved as in the case of assertoric <sentences>.
[2.6.26J Theorem 1I-5(b) is proved: the <sentence>

(275) Any God can not be creating; therefore, one creating can
not-be God
does not follow, for the first is true and the second false. The reason
for this is a certain mode of restriction which the verb 'is' exercises in
the predicate which is not exercised in the subject. Whence one can
say not only of the verb but even of the predicate "B can be A" and
"B could have been [or 'can be going to be'J A"

[Theorem II-6] (a) From no sentence de necessario does
there follow another sentence de necessario by conversion of
the terms, except (b) a universal negative follows from a uni-
versal negative. I 5
[2.6.28] Theorem II-6(a) is clear, since <the sentence>

(276) Anything creating is necessarily God; therefore, God is
necessarily creating
does not follow, for the first is true and the second false. Similarly for a
particular negative:
(277) Some animal is necessarily not an ass; therefore, some ass
is necessarily not an animal
which does not follow.
[2.6.29] Theorem II-6(b) is proved: the contradictory of the antecedent

follows from the contradictory of the consequent <by Theorem 1-3>, and
so
(278) Some A can be B; therefore, some B can be A
follows by Theorem II-4. 16
[2.6.30] Nevertheless, this should be known:

[Remark] Any affirmative de necessario can be converted
by resolving the conversion through the "what is ..."
<locution>P
For example, the <sentence>
(279) The one creating is necessarily God; therefore, what is
necessarily God is or can be creating
follows.
[2.6.31] Note that affirmatives de necessario can be converted into

affirmatives de possibili; for example, this follows:
(280) B necessarily is A; therefore, A can be B
The reason for this is that a sentence de possibili follows from a sen-
tence de necessaria <by the Remark in [2.6.22]>. For example, this
follows:
(281) B necessarily is A; therefore, B can be A
thence by conversion <of the consequent (Theorem 11-5) and Theorem
1-4>
(281 *) <B necessarily is A>; therefore, A can be B

[Theorem II-71 (a) Any sentence contingent ad utrumlibet
with the mode affirmed <can be> converted into <a sen-
tence> with the opposite quality with the mode affirmed, but
(b) none <can be> thus converted if the converted sentence
or its converse has a denied mode.
[2.6.33] Theorem 11-7(a) is clear by the nominal definition, for some-

thing is called 'contingent <ad utrumlibet>' if it can be the case and can
not-be the case. Thus the <sentences>
(282) Every B is contingently A; therefore, every B is contingently
notA
(283) Some B is contingently A; therefore, some B is contingently
notA
follow, as do their converses.
[2.6.34] Theorem II-7(b) is equally clear: the sentence

(284) ... is contingently- - -; therefore, ... IS not con-
tingently- - -
does not follow; rather, the opposite of (284) <follows>. Indeed, these
are contraries:
(285) Every B is contingently A
(285*) No B is contingently A
[2.6.35] It is clear, therefore, that

[Remark] From any contingent sentence having an affirmed
mode there follows a sentence de possibili, one <which is>

negative as well as one <which is> affirmative. lx
this is why it is correctly said that the contingent exludes the necessary
and the impossible; hence
[Remark] From any sentence de necessario having an
affirmed mode there follows a contingent sentence having a
denied mode.
[2.6.36] The eighth theorem is:

[Theorem II-8] (a) No contingent sentence can be converted
in its terms to another contingent sentence, but (b) any
having an affirmed mode can be converted into one de
possibili.
[2.6.37] Theorem II-8(a) is clear: the <sentence>

(286) God is contingently the one creating; therefore, the one
creating is contingently God
does not follow, for the first is true and the second false, because any-
thing creating is necessarily God. And if the affirmative is not con-
verted then the negative is not converted either, since they are
equipollent, or follow from one another mutually. Similarly, with the
denied mode,
(287) Nothing creating is contingently God; therefore, no God is
contingently creating
<does not follow>, because the first is true and the second false.
[2.6.38] Theorem II-8(b) is clear because from any contingent <sen-

tences> having an affirmed mode there follows an affirmative de pos-
sibili <by the first Remark in [2.6.35]», which is converted into another
<sentence> de possibili <by Theorem 11-5>.
CHAPTER 11-7. THEOREMS ABOUT COMPOSITE MODALS
[2.7.1] With these matters settled, then, we should deal with composite
modals. It should be pointed out first of all that (no matter what some
people say) they can be universal, particular, indefinite, and singular.

For example, this is universal:
(288) Every possible is that B is A (omne possihile est B esse A)
and this is indefinite:
(289) It is possible that B is A (possihile est B esse A)
and thus I can say 'some possible' or 'this possible' <as in (288), mak-
ing particular and singular sentences>.
[2.7.2] Here 'possible' is taken not for 'possible-to-be' but for a possi-
ble sentence, which is called possible because howsoever it signifies
can be the case. 19 And so the sentences
(290) Every possible is that B is A
(290*) Every possible sentence is that B is A
say the same thing. If 'possible' were taken otherwise, then there would
be an equivocation and I would not call such sentences 'de possihili.'
And the same should be said for the other modes, <namely 'necessary'
and 'contingent'>.
[2.7.3] It should be noted that in (288) the predicate 'that B is A' sup-
posits materially for a sentence such as "B is A" It does not supposit
for itself, because "that B is A" is not a well-formed sentence (oratio).
[2.7.4] If the mode is predicated and the dictum put as the subject, the
sentence can still be universal, particular, <indefinite, or singular>. In
that case, this is universal:
(291) Anything which is that B is A is possible (omne quod est B
esse A est possihile)
this is particular:
(292) Something which is that B is A is possible
this is indefinite:
(293) That B is A is possible
this is singular:
(294) That which is that B is A is possible
Alternatively, we may use <the following equivalent forms>:

(291 *) Any sentence "B is A" is possible
(292*) Some sentence "B is A" is possible
(293*) A sentence "B is A" is possible
(294*) This sentence "B is A" is possible
[2.7.5] But if we ask what form 20 the sentence

(295) Every B to be A is possible (omne B esse A est possibi/e)
has, then I say that the whole <expression> 'Every B to be A' is the
subject, suppositing materially for a sentence such as "every B is A,"
and so (295) is indefinite. But if we were to say that the subject is 'B
to be A,' suppositing for such a sentence as "B is A" and the quantifier
'every' is taken significatively, then (295) would be a universal sen-
tence having as its sense
(295*) Every sentence "B is A" is possible
Similarly, if I say
(296) No B to be A is possible
if the whole <expression> 'no B to be A' is the subject then (296) is an
indefinite affirmative sentence, but if the subject is only 'B to be A'
and the world 'no' is taken significatively then (296) is a universal
negative.
[2.7.6] the ninth theorem is:

[Theorem II-9J In all composite modals in which the dictum
is put as the subject, the universal follows from the particu-
lar, with all else unchanged. For example, this follows:
"Some sentence 'B is A' is possible; therefore, every sen-
tence 'B is A' is possible" and the same goes for truth and
falsity, contingency and necessity.21
[2.7.7] The reason for this is that for all such sentences "B is A," one
signifies whatever and however the other signifies. Therefore, if it is
the case as one signifies it is the case as the other signifies, and if not
then not; thus if one is true the other is true and if <one is> false <the
other is> false. Similarly for possibility, necessity, and the other modes.
[2.7.8) Just as <in Theorem 11-9> a universal <follows from> a particu-

lar, so too a universal <follows from> a singular, for the sentence
(297) This sentence "B is A" is possible
cannot be true unless
(298) Every sentence "B is A" is possible
is true.
[2.7.9) This is also clear for negatives, since if some sentence "B is A"
is not true then
(299) No sentence "B is A" is true
follows. The causes of the truth of this sentence
(300) Some sentence "B is A" is not true
exist, or there is no sentence "B is A" and so the subject does not sup-
posit for anything (and hence the negative sentence is true), or there is
some sentence "B is A" but it is not true. Now it follows from the first
of these causes not only that some such <sentence> is not true, but
also that none such is true, and we have <the desired conclusion about
negatives>. Similarly, it follows from the second cause of truth that no
such <sentence> is true, since any cause of the truth of one is a cause
of the truth of any of them, if there are several; thus we have the
<desired conclusion>.
[2.7.10) The tenth theorem is:

[Theorem 1I-10) Any given composite modal in which the
dictum is the subject is converted simply in its terms,
except for the universal affirmative which is only converted
accidentally,zz
[2.7.11) First of all, a particular affirmative is converted into a particu-

lar affirmative, as for example
(301) Some sentence "B is A" is possible; therefore, some <sen-
tences> is possible, i.e. some possible sentence is the sent-
tence "B is A"
This is proved as in the assertoric case, and can be proved by an

expository syllogism.
[2.7.12) Having proved this, it is clear that a universal is converted

into a particular, since whatever follows from the consequent follows
from the antecedent. But the given universal is not simply converted,
since it does not follow that if every sentence "B is A" is possible then
any possible sentence is the sentence "B is A."
[2.7.13) Next, the <universal> negative is converted simply, by the same

<argument> as in the case of assertoric <sentences>. There is the same
supposition of terms in each, and no ampliation is present.
[2.7.14) Next, the particular negative is converted into a particular

negative. The first <negative> is not false unless the subject supposits
for nothing, and so neither is the second; if the subject sup posits for
something then it is equipollent to a universal, which is converted <as
in [2.7.13». Thus the first cannot be true without the second.
[2.7.15) Next, the case of singulars is apparent from <the argument> in

the case of assertoric sentences.
[2.7.16) The eleventh theorem is:

[Theorem //-11) Any composite modal in which the mode is
put as the subject is converted simply, except for the parti-
cular negative, which is not convertedP
[2.7.17) Theorem II-ll is obvious in all sentences just as was the case
for assertoric sentences <in Theorem 1-14 and Theorem I-IS>, except
that the universal affirmative for such <composite modals> is converted
simply, because it is obviously converted <accidentally> into a particu-
lar and from the particular there follows the universal <by Theorem
11-9>.
[2.7.18) The twelfth theorem is:

[Theorem //-12) Any affirmative composite modal de vero
or de possibili or de necessario is converted with respect to
the dictum as the dictum is itself converted.
[2.7.19] I call a 'conversion with respect to the dictum' that which

comes about by transposing each of the terms in the dictum, and
not by transposing the whole subject and the whole predicate. For
example,
(302) It is possible that some man is running; therefore, it is pos-
sible that something running is a man
(303) It is necessary that any God is just; therefore, it is necessary
that something just is God
(304) That no man is an ass is true; therefore, that no ass is a
man is true
These conversions hold by Theorem 1-5, namely that if the antecedent
is true then the consequent is true, and if the <antecedent> is possible
then <the consequent> is possible, and if <the antecedent> is of necess-
ity then <the consequent> is of necessity, provided that they are formed
together.
[2.7.20] Whence it should be remarked that these conversions are not

simply acceptable consequences unless we add the hypothesis of the
nonemptiness (constantia)of the terms.24 For example,it is possible that
a sentence such as "B is A" is true and that <the sentence> "A is B"
is not true, because <the latter> is not formed. Nor does what some
people say hold, namely that the consequence cannot be formed unless
each <of its parts> is also formed, because, without the aforesaid
assumption, if the consequence is formed the terms are nonempty. <In
reply>, I say first that the consequence is formed without the formation
of those sentences for which the terms <of the consequence> supposit,
for example by saying
(305) It is possible that B is A; therefore, it is possible that A is B
for "that B is A" is not any sentence, although it supposits for some
sentence. I say secondly that it does not suffice for a consequence to
be simply acceptable that it is impossible to be as the antecedent signi-
fies to be the case unless it is the case as is signified by the conse-
quent, when the consequence is formed; rather, it is simply required
that it is impossible to be the case as is signified by the antecedent
unless it is the case as signified by the consequent. 25 Whence the
sentence
(306) Some sc:ntence is affirmative; therefore, some sentence is

particular
does not follow, which is clear because the contradictory of the
antecedent does not follow from the contradictory of the consequent
<as required by Theorem 1-3>.26
[2.7.21] The thirteenth theorem is:

[Theorem 11-13] In a composite affirmative modal sentence
de fa/so or de impossibili, (a) any particular dictum is con-
verted into a universal, but the universal is not converted
into the particular; (b) a universal negative dictum and a
particular affirmative are converted simply, but the univer-
sal affirmative is not convertedP
[2.7.22] All these claims hold because if the consequent is false or

impossible the antecedent must be false or impossible. The converse is
not necessary, since a truth may correctly follow from a falsehood and
a possible <sentence> from an impossible. (A<;ld to this conclusion the
hypothesis that the terms are nonempty, i.e. that the sentences are
formed, as in Theorem II-12.)

[Theorem 11-14] Every dictum in a composite affirmative
contingent sentence is converted according to the opposite
quality in the dictum of the contradictory, not of the
contrary.28
[2.7.24] The sense ofthe sentence

(307) It is contingent (ad utrumli~et) that B is A
is that "B is A" can be true and can be false. Its contradictory must
thus also be of this sort, since contradictories are neither true together
nor false together. But its contrary <need not be of this sort>; although
"Anything intelligent is God" can be true and can be false, neverthe-
less "Nothing intelligent is God," <which is its contrary>, cannot be
true. Thus
(308) It is contingent that anything intelligent is God; therefore, it

is contingent that nothing intelligent is God
does not follow.
[2.7.25) The fifteenth theorem is:

[Theorem II-IS) (a) From any composite affirmative sen-
tence de vera there follows its dictum, and conversely;29 (b)
from any <composite affirmative sentence> de necessario
there follows its dictum and there follow a sentence de vero
and de possibili, and not conversely; (c) from any <compo-
site affirmative sentence> de vero there follows a sentence
de possibili and not conversely; (d) from any sentence there
follows a sentence de possibili of which it will be the dic-
tum, and not conversely; and (e) from any such contingent
<sentence> there follows a sentence de possibili, and not
conversely.
[2.7.26) I have put all the above claims together because they are obvi-
ous, and perhaps should be taken as principles rather than as theo-
rems. Any necessary sentence is true and possible, but not conversely.
The sentence
(309) B is A; therefore, it is possible that B is A
correctly follows, but not the converse; the sentence
(310) B is A; therefore, it is true that B is A
follows, as does its converse, at least assuming the nonemptiness of the
terms. And stating these examples seems to me to suffice for Theorem
11-15.
[2.7.27) The sixteenth theorem is:

[Theorem II-I6] (a) If a composite affirmative sentence de
necessario is true, whatever follows from its dictum is
necessary; and whatever <follows from the dictum of a true
composite affirmative sentence> de possibili is possible; and
whatever <follows from the dictum of a true composite
affirmative sentence> de vero is true. 30 But (b) this is not the
case for divided <affirmative sentences which are true>,

except for <such sentences> de vero.
[2.7.28) Theorem II-16(a) is evident by the rule "If the antecedent is

true then the consequent is true; if possible, possible; if necessary,
necessary."
[2.7.29) Theorem II-16(b) is clear, because although

(311) Anything creating is necessarily God
is true, nevertheless
(312) The one creating is God
must not follow of necessity, since perhaps it is false, <as in the case
where God is not now creating>. Similarly, although
(313) A white thing can be black
is true, nevertheless
(314) A white thing is black
is impossible.
[2.7.30) Note that the general rule "Nothing impossible follows from
an assertoric <sentence> put forward as possible" should be interpreted
as in Theorem 11-16.
[2.7.31) However, there is a doubt whether any divided sentence de

possibili puts forward an assertoric <sentence>. That is, it is said that it
is dubious whether and how any true divided sentence de possibili
corresponds to a possible assertoric sentence.
[2.7.32) I say firstly <in reply to [2.7.31) >that

[Remark) It is not necessary that if a divided universal de
possibili is true that an assertoric universal is possible.
Although every star in the Zodiac is possibly shining above our hemis-
phere, nevertheless
(315) Every star in the Zodiac is shining above our hemisphere
is physically impossible. Therefore, if such a universal is put forward

assertorically, it should be put forward assertorically by the singular
<claims> it divides. 3'
[2.7.33] I say secondly <in reply to [2.7.31]» that

[Remark] It is not necessary that if a particular or indefinite
divided <sentence> de possibili is true then the assertoric
sentence <corresponding to it by> keeping the same terms is
possible.
Although a white thing can be black, nevertheless "a white thing is
black" is not possible. Thus it is said that if such <a sentence>, namely
one where the subject is a connotative term, is put forward assertori-
cally, it should be put forward with a demonstrative pronoun pointing
out what the subject supposits for. For example, if a white thing can be
black, then it follows that "This is black" is possible, pointing out that
for which 'white' supposits. Still, this is perhaps not universally true,
unless sometimes a change in the predicate is made, namely when the
predicate involves an opposition to an assertoric present-time sent-
ence. 32 For example, this man can be white at a future time, and "This
man is white at a future time" does not follow. Similarly, although the
matter of water can take the form of air which does not <now> exist,
the <sentence>
(236) The matter of water can take the form of the air which
does not exist
is not possible. Thus one must remove the expressions 'at a future
time' and 'which does not exist' from those predicates.

[Theorem Il-l7] From no composite affirmative de possibili
does there follow a divided sentence de possibili in the
affirmative mode, nor conversely, except that a particular
divided affirmative follows from a composite affirmative
where the dictum is affirmative. 33
[2.7.35] First of all, the exception is obvious: if "B is A" is possible

then "Some B can be A" clearly follows. Nevertheless, although
"Everything running is a horse" is possible, it does not follow univer-

sally that everything running can be a horse, since perhaps an ass
which cannot be a horse is running. Thus only the particular follows.
[2.7.36] It is also clear that nothing follows from the converse, for
although anyone sleeping can be awake, "One sleeping is awake" is not
possible. Similarly and negatively, although every star of the Zodiac
shining above our hemisphere can be not shining on our hemisphere,
nevertheless
(317) A star of the Zodiac shining on our hemisphere does not
shine on our hemisphere
is not possible.
[2.7.37] Similarly and conversely, although "Nothing creating is God"

is possible, nevertheless "The one creating is possibly not God" is not
true, since anything creating is necessarily God.
[2.7.38] The eighteenth theorem is:

[Theorem II-18] From no composite affirmative de neces-
sario does there follow a divided sentence de necessario
with the mode afirmed, nor conversely, except a composite
universal with the dictum denied follows from a divided
universal negative. 34
[2.7.39] The exception is proved first. If "B is A" is possible then by

Theorem 11-17 it follows that B can be A, and so by Theorem 1-3 this
follows:
(318) No B can be A; therefore, "B is A" is not possible
Therefore it follows by equipollence that
(318*) Every B is necessarily not A; therefore, "B is A" is not
possible
from which it follows that the contradictory of ("B is A") is necessary,
since anything impossible has a necessary contradictory. Therefore it
follows that "No B is A" is necessary, and this is the desired conclu-
sion.
[2.7.40] The first part of Theorem II-IS is clear: although "Every

horse is an animal" is according to Aristotle necessary, nevertheless no
horse is necessarily an animal, for any horse can fail to exist and con-
sequently fail to be an animal. Similarly and conversely, although any-
thing creating is necessarily God, nevertheless "The one creating is
God" is not necessary. And so too, even though it is necessary that
nobody asleep is awake, nevertheless nobody asleep is necessarily
awake.
[2.7.41] <Theorem II-IS> is also clear for the particular negative, for
(319) Some runner is necessarily not some horse; therefore, "A
runner is not a horse" is necessary
does not follow, for positing that only a man is running the first is true
and the second is false.
[2.7.42] The nineteenth theorem is:

[Theorem //-19] (a) From no sentence which is assertoric or
de possibili or de necessario does there follow a contingent
sentence with both modes being affirmed; similarly, (b)
from no contingent sentence does there follow any sentence
which is assertoric or de necessario, but (c) <there does fol-
low> a similar <sentence> de possibili.
[2.7.43] Theorem II-19(a) is clear: from no affirmative sentence which

is assertoric or de possibili or de necessario does there follow a nega-
tive in the same terms; nor an affirmative from a negative; and any
contingent sentence contains within itself an affirmative and a negative
either formally or consecutively.35
[2.7.44] Theorem II-19(b) is clear: 36 the contingent excludes the neces-

sary and the impossible. Since a contingent can be and can not-be, it
neither follows that it is nor follows that it is not.
[2.7.45] Theorem II-19(c) is also clear: a contingent is neither neces-

sary nor impossible; therefore, it possibly is and possibly is not, such
that these all follow:
(320) It is contingent; therefore it possibly is
(321 ) <It is contingent>; therefore, it possibly is not

(322) It is contingently not; therefore, it possibly is
(323) <It is contingently not>; therefore, it possibly is not
[2.4.46] These claims stated in Theorem 11-19 hold for divided as well
as for composite <modals>.
[2.7.47] Therefore, let all take caution lest they take divided modals
for composites, or conversely, and lest they take divided <modals> with
the mode denied for divided <modes> with the dictum affirmed, since
they are completely different, as is clear. And so ends Book II, which
deals with the consequences of one kind of modal <sentence> to
another.
BOOK III
SYLLOGISMS WITH ASSERTORIC SENTENCES
CHAPTER III-I. THE KINDS OF CONSEQUENCES
[3.1.1] Book III will deal with syllogistic consequences, about which
much needs to be said.
[3.1.2] First of all, there are many kinds of consequences. Some hold
only by their matter, so that they are not formal: examples are en thy-
memes, complete inductions, examples, and perhaps many others,l as
for instance <the rule that> you may infer whatever you please from the
impossible. None of these are worthy of the name of syllogism.
[3.1.3] Now some formal consequences are of one categorical from a

single categorical, and then each <categorical> must have the same terms,
leaving the syncategoremata out of account. Examples are equipol-
lences, subalternations, conversions, and consequences which come
about by changing the terms according to finite and infinite, <discussed
in the preceding chapters>.
[3.1.4] As all agree, these are neither complete inductions nor examples
nor yet syllogisms, since a syllogism must have several premisses. But
they are not strictly speaking enthymemes, for an enthymeme is an
imperfect or truncated syllogism. An enthymeme should move from a
single premise to a conclusion, and that conclusion is apt to follow syllo-
gistically from that premise and another added to it. Nevertheless,
broadly speaking, <equipollences and the like> can be called enthy-
memes, if we want to call every consequence from a single categorical to
another categorical an enthymeme. Enough has been said about these
consequences.
[3.1.51 Other consequences are formal by virtue of a single conjunct

following from a conjunctive sentence or a disjunctive sentence follow-
ing from a single disjunct: from any conjunctive sentence any of the
conjuncts follows, and from any sentence at all there follows any dis-
junctive <sentence> with that sentence as one of the disjuncts.
258
III. SYLLOGISMS WITH ASSERTORIC SENTENCES 259
[3.1.6] But these are not syllogistic consequences, since in a syllogism

something should follow from the premisses, and the conslusion does
not have both terms as found in either of the <sentences> in the antece-
dent, nor conversely.
[3.1.7] Other consequences are formal by virtue of a condition, that is,

by the addition of some sentence to a conditional sentence. 2 And this
can be by (i) adding to a conditional its antecedent and so inferring the
consequent; (ii) adding to a conditional the contradictory of its conse-
quent and inferring the contradictory of its antecedent; (iii) adding to a
conditional a conditional where the antecedent of one is the consequent
of the other, and inferring the other conditional in virtue of the principle
"whatever follows from the consequent follows from the antecedent."
An example of (i) is that if A exists then B exists, and A exists; therefore
B exists. An example of (ii) is that if A exists then B exists, and no B
exists; therefore no A exists. An example of (iii) is that if A exists then B
exists, and if B exists then C exists; therefore if A exists then C exists.
[3.1.8] Still, these consequences are not strictly called syllogisms, at

least, not in the way I want to treat them: I want to treat simple syllo-
gisms, which are simple 3 consequences that do not contain a consequ-
ence or consequences in their antecedents.
[3.1.9] Some consequences are formal by the formal impossiblity of the

antecedent or the formal necessity of the consequent. Since anything
follows from the impossible in its form then there will be a formal con-
sequence from it to any other <sentence>.
[3.1.10] (The 'impossible in its form' would be a conjunctive sentence

made up of two contraries or contradictories or even a <sentence> in
which some finite term is affirmed along with its infinite <version> or
conversely; and the 'formally necessary' would be a disjunctive sentence
made up of contradictories or subcontraries or even <a sentence> in
which an infinite term is denied of itself in its finite <version> or con-
versely/
[3.1.11] But I still do not call these syllogistic consequences, for a syl-
logism should proceed to some conclusion or conclusions determi-
nately, not to any <sentence at all>, and from a determinate antecedent,
not indifferently from any <antecedent>.
[3.1.12] Again, consequences are formal through the analysis of their

syncategoremata, either from the analysans to the analysandum or vice-
versa.
[3.1.13] I do not mean to treat of these either, for analysis of this sort is
nothing but the explication of the signification of the syncategoremata.
[3.1.14] Next of all, consequences are formal by division, which Aristo-

tle calls 'infinite syllogisms.'5 If one member is removed the rest may be
inferred by giving a sufficient division. For example:
(324) Every A is B or every A is C, and A is not B; therefore,
every Ais C
(325) Every A is B or C, and no A is B; therefore, every A is C
(326) Every A is B or C, and some A is not B; therefore, some A is
C
[3.1.15] But whether these should be called syllogisms or not, I do not
intend to deal principally with them in what follows, because in them the
conclusion has both of the terms which appear in one of the premisses.
Nevertheless I wanted to list these consequences in this chapter lest it
seem that I have forgotten to deal with them.
[3.1.16] Still others may be given, but these seem to me to be the most
principal ones other than the syllogistic <consequences> which we do
intend to discuss in Book III.
[3.1.17] I shall mean by 'syllogism' in the following only a formal conse-

quence to a single categorical conclusion by a middle <term> from each
extreme <which> differs from the given conclusion.
CHAPTER III-2. THE SYLLOGISM
[3.2.1] In Chapter II1-2 I spell out how any such syllogism requires in its
premisses the conjunction of each of the extremes of the conclusion with
the middle, according to which conjunction the conjunction among
themselves of the extremes may be inferred, either affirmatively or
negatively. And so it is clear that any syllogism, as we here understand
'syllogism,' is made up of only three terms,6 namely the two extremes
which are the terms of the conclusion and the middle term with which
those extremes are conjoined in the premisses.
[3.2.2] It is also clear that there are two premisses in a syllogism, where
in one the middle is conjoined with one extreme and in the other with
the other; it is obvious that the middle <term> is found in each premiss
and not in the conclusion.
[3.2.3] The first premiss is called the 'major sentence' and the second
the 'minor <sentence>: The extreme <term> found in the major sentence
is called the 'major extreme' and <the extreme term> found in the minor
is called the 'minor extreme:
[3.2.4] It further follows from this that there are only four figures of this
kind of syllogism. We name a 'syllogistic figure' the ordering of the mid-
dle to the extremes in the premisses by being put as the predicate and by
being put as the subject (subiectionem et praedicationem), and this can
only occur in four combinations.
[3.2.5] The first figure is where the middle is put as the subject of the
major sentence and is predicated in the minor.
(3.2.6] The second figure is where the middle is predicated in each <pre-
miss>.
[3.2.7] The third figure is where the middle is put as the subject in each
<premiss>.
[3.2.8] The fourth figure is the converse of the first figure, namely where
the middle is predicated in the major and put as the subject of the
minor.
[3.2.9] It should be noted that the fourth figure only differs from the
first figure in the transposition of its premisses - which transposition
makes no difference to the conclusion or prevents the entailment, but
only matters in that the entailed conclusion, if it is direct in the first
figure, would be indirect in the fourth figure, and conversely.?
[3.2.10] (I call a conclusion 'direct' in which the major extreme is predi-

cated of the minor, and <I call a conclusion> 'indirect' in which the minor
extreme is predicated of the major.)
[3.2.11] Now that these points are clear, it is obvious that if we deter-
mine matters for the first figure it will be superfluous to settle them for
the fourth figure as well, which is why Aristotle does not discuss it.
CHAPTER 111-3. FINITE AND INFINITE TERMS
[3.3.1] Chapter III-3 explains that some of the terms which make up a
syllogism are finite and some are infinite. One syllogizes from them in
the same way, provided that a term is not changed according to finite
and infinite in the two sentences in which it is found. x For example, just
as
(327) Every B is A, and every C is B; therefore, every C is A
follows, so too do
(328) Every B is non-A, and every non-C is B; therefore, every
non-C is non-A
(329) Every non-B is A, and every C is non-B; therefore, every C
isA
[3.3.2] But when some term is changed according to finite and infinite
in the two sentences in which it is found, then there will be other syllo-
gistic moods, which we shall determine in [3.7.31]-[3.7.45] <in Theorem
III-18 and Theorem III-19> - for example, if we were to say
(330) No non-B is A, and no C is B; therefore, no C is A
[3.3.3] Similarly, syllogisms are sometimes made up of nominative terms

and sometimes of oblique terms.
[3.3.4] We shall first of all discuss those made up of nominative terms,
and of these we shall first discuss those with assertoric <premisses,
throughout Book III, where syllogisms with oblique terms are discussed
in Chapter III-5 and after>, and then those with modal <premisses, in
Book IV>.
[3.3.5] Furthermore, in some syllogisms the middle is a common term

and in others the middle is a discrete term, and the latter case is custom-
arily called an 'expository syllogism,' which we shall discuss first of all

<in Theorem III-5 and [3.4.23]-[3.4.27]>.
[3.3.6] It should also be remarked that sometimes in the sentences of a

syllogism the subject is ampliated by the predicate, and sometimes it is
not so ampliated, which makes a great deal of difference to the syllogism
<and will be examined in [3.5.54]-[3.5.72]>.
CHAPTER III-4. THEOREMS
[3.4.1] In Chapter III-4 we assume that affirmative syllogisms hold in

virtue of this principle:
[Rule 1] Whichever are the same as one and the same are the
same as each other (quaecumque uni et eidem sunt eadem
inter se sunt eadem).
From the fact that the extremes are indicated in the premisses to be
called the same as the one middle, they are deduced in the conclusion to
be called the same as each other.
[3.4.2] Negative syllogisms hold by another principle:

[Rule 2] For any two things, when one is the same as someth-
ing else to which the other is not the same, then they are not
the same as each other.
For this reason an affirmative conclusion must be concluded from two
affirmatives, and a negative from one affirmative and the other negative,
since an affirmative sentence indicates an identity and a negative sen-
tence non-identity. We must analyze these rules so that their sense is not
misunderstoodY
[3.4.3] About Rule 1 of [3.4.1] I say firstly that if the expressions 'which-
ever (quaecumaque)' and 'same (eadem),' which are plural in number,
are taken collectively then Rule 1 is not true such that one thing is the
same as another, for matter and form are the same as one and the same
composite, and the matter is not the same as the form.
[3.4.4] I say secondly <about Rule 1> that nothing divided 10 is the same
as one and the same <thing>, namely such that each is the same as it
(unless there is a counterexample in the case of God, which we shall

deal with in [3.4.8] and [3.4.14]). For there are no things which are not
many and different from each other, and if this is so, then no one is
numerically the same as any of the others.
[3.4.5] But then how does Rule 1 of [3.4.1] hold? I answer that many
names are correctly called the same dividedly as one and the same dis-
crete term. I do not say 'are the same' but 'are called the same; i.e. they
are truly predicated affirmatively of the same discrete term even with
the addition of the word 'same; and so are truly put as subject to it. For
example, we correctly say
(331) Socrates is the same as animal and Socrates is the same as
man
or even the converse
(332) Animal is the same as Socrates and man is the same as Soc-
rates
from which it can be inferred that animal is the same as man, or that
man is an animal. Therefore Rule 1 should be literally stated in this way:
[Rule 1 (aJ] Whichever terms are dividedly truly called the
same as one discrete term are truly called the same as each
other.
And, as I said, I do not say they are the same but that they are called the
same.
[3.4.6] However, if there is no discrete term present, then the rule must
be put like this:
[Rule 1 (bJ] Whichever terms are dividedly truly called the
same as a common term by reason of the same thing for
which the common term supposits, those terms are called the
same as each other.
For example, although 'Socrates' and 'Plato' are dividedly said of the
same common term 'man,' since Plato is the same as a man and Socrates
is the same as a man, nevertheless it cannot be deduced that Socrates is
the same as Plato, for Socrates and Plato are not called 'man' by reason
of the same man - rather Socrates is the same as one man and Plato is
the same as another. But if Socrates were the same as a man and Plato
were the same as the same man, then it could be deduced that Socrates
is Plato.
[3.4.7) About Rule 2 for negatives in [3.4.2) I say in the same way that
[Rule 2) (a) For whichever two terms of which one is truly
called the same as some discrete term, of which the other is
truly called not the same, the one can be inferred negatively
of the other; (b) if the middle is a common term, it is neces-
sary to add that one of the extremes is called the same and
the other called not the same for the same thing.
[3.4.8) But it should be carefully noted that these rules do not hold in
the case of God, <the terms for Whom> supposit for a simple thing one
and triune at the same time. Whence although the Father is the same as
the simple God and the Son is the same as the simple God, the Father is
nevertheless not the Son; and although the same Father is God and not
the Son, it is false nevertheless that the Son is not the same as God.
[3.4.9) Nevertheless, in other things in which it is impossible that the

same simple thing is both single and triune, those rules hold, and Aristo-
tle believed all those rules to hold because a counterexample cannot be
found to them by human reason but through faith alone.
[3.4.10) Now that these matters have been clarified, some theorems fol-
low.
[3.4.11) The first theorem is:

[Theorem III-I) No syllogisms are formal in the customary
idiom.
[3.4.12) I call 'the customary idiom' the manner of speaking without the
addition of the "what is ..." <locutiOn>, or anything like it. For example,
when I say
(333) Socrates is a man
(334) Every man is an animal
and not when I say

(335) He who is Socrates is a man
(336) Anything which is a man is an animal
or sentences like these.
[3.4.13] The reason for Theorem III-1 is that terms can be found in
which such a form does not hold, since they do not hold according to
Rule 1. For example, these do not hold:
(337) God is the Father and the same God is the Son; therefore,
the Son is the Father
(338) Any God is the divine Father and any divine Son is God;
therefore, any divine Son is the divine Father
(339) No Son is the divine Father and any God is the Son; there-
fore, no God is the divine Father
The same objection can be made for the other <syllogistic> moods.
[3.4.14] I leave to the theologians whether there is another style of

speaking of a syllogism with a form that holds in the case of God, and
what the form <of such a syllogism> is. Note and remember that it is not
my business, <the business of a Master of> Arts, to settle <this matter>
beyond the foregoing remarks: since names are conventionally signifi-
cant, in the rest <of this treatise> I shall term those syllogisms 'formal' to
which no counterexample to the form can be given, except in the case of
God. I do not mean that they are simply formal, but <they are formal>
under the assumption that they are not made up of terms which pertain
to the distinctness of the Divine Persons.

[Theorem III-2] No syllogism made up of two negatives
holds.
Where, as noted, I here speak of a syllogism in which none of the terms
is changed according to finite and infinite.
[3.4.16] The reason for Theorem III-2 is that in a syllogism so disposed,

<namely with two negative premisses>, none of the rules though which
we said that syllogisms hold are observed, whence terms which are
called different as well as terms which are called the same can truly be
denied of the same. Brunellus is not a stone, Brunellus is not a man,
Brunellus is not risible; nevertheless, it cannot be affirmatively deduced
that a stone is a man nor negatively <deduced> that a man is not risible.
[3.4.17J The third theorem is:

[Theorem III-3J (a) In any acceptable syllogism, from either
premiss with the contradictory of the conclusion there
follows the contradictory of the other premiss; and (b) any
syllogism is acceptable in which the contradictory of one pre-
miss follows from the contradictory of the other premiss
along with the contradictory of the conclusion. I I
[3.4.18J Theorem III-3 holds by Theorem 1-3; to see how this is evident,
note that neither of the premisses alone in a syllogism is called the
antecedent, since with the truth of one the conclusion may still be false
through the falsehood of the other <premiss>. Rather, the whole antece-
dent is the conjunctive sentence made up of the two premisses. There-
fore, from the contradictory of the conclusion there follows the
contradictory of the conjunctive <sentence>, which is the disjunction of
the contradictories of the premisses. Now the premiss taken with the
contradictory of the conclusion cancels one of the disjuncts, and there-
fore the other disjunct must be inferred, for a disjunction is not true
unless one of its disjuncts is true. Thus Theorem III -3(a) is clear.
[3.4.19J Theorem III-3(b) is similarly clear: if the opposite of one pre-

miss follows from the contradictory of the conclusion and the other
premiss, then the premisses are not compatible (non possunt stare) with
the contradictory of the conclusion. <The premisses make up> the
antecedent, and thus the opposite of the antecedent is inferred from the
opposite of the consequent.
[3.4.20J The fourth theorem is:

[Theorem III-4J For any given syllogism, whatever conclu-
sion follows from its premisses follows from one of them
with <a sentence> antecedent to the other; and whatever con-
slusion does not follow from those premisses does not follow
from one with <a sentence> consequent to the otherP
[3.4.21] From Theorem III-4 we introduce the rule that if there is no

syllogism from two universals then there will be no syllogism from one
universal and one particular, the rest remaining the same, and whatever
conclusion follows from a particular and a universal follows from two
universals.
[3.4.22] Theorem 1lI-4 holds by Theorem 1-4: any conjunctive sentence

is consequent to a conjunctive sentence made up of one part of the
former and the antecedent of the other part, and is antecedent to the
conjunctive sentence made up of one part of the <original> and the
consequent of the other part.

[Theorem III-5] In every figure there is a valid expository syl-
logism l3 (a) from two affirmatives to an affirmative conclu-
sion; (b) from an affirmative and a negative to a negative
conclusion (no matter which premiss is affirmative).l4
[3.4.24] Theorem 1lI-5(a) and III-5(b) hold by Rule 1 and Rule 2.

<Expository> syllogisms are most obvious in the third figure: for ex-
ample,
(340) This C is A and this same C is B; therefore, B is A
Since C and A are the same, if B is the same as C and not <the same> as
A, then B will be the same and not the same as the same <thing>, which is
impossible. Similarly and negatively,
(341) This C is not A and this same C is B; therefore, B is not A
must hold, for otherwise the same impossibility as before would follow.
[3.4.25] It follows from the syllogisms permitted in the third figure that
they are valid in the second figure, since the third figure comes about
immediately from converting the premisses <of the second figure>. Sim-
ilarly, the <expository syllogisms> are valid in the first figure, since the
third figure comes about <from the first figure> by converting the minor
premiss.
[3.4.26] Nevertheless, it should be noted that in negative syllogisms

where the major extreme is not distributed the direct conclusion cannot
be inferred in the common idiom <for negatives, discussed above in

[1.8.70]>, in which the negation precedes the predicate, for the major is
distributed in the conclusion although it is not distributed in the <major>
premissY Similarly, if the minor <extreme> is not distributed then the
indirect conclusion cannot be inferred in this idiom, for the same rea-
son. But when the major is distributed the direct conclusion can be
inferred, and when the minor is distributed the indirect conclusion can
be inferred. If neither <the major extreme nor the minor extreme> are
distributed, then one must draw an inference without the distribution of
the predicate - although not in the common idiom for negatives. For
example,
(342) Animal is not Socrates <and> man is Socrates; therefore man
animal not is
[3.4.27] From the foregoing one can infer the following:

[Coro/lary1In order to deduce a conclusion by an expository
syllogism negatively and directly in the common idiom <for
negatives> the major <premiss> must be negative (in the first
and the third figures), and <the major premiss> must be nega-
tive and universal in the second figure.
[3.4.28] In what I say hereafter I speak of syllogisms formed with

common terms. And so some theorems may be stated.

[Theorem III-6] There is no valid syllogism in which the
middle is not distributed in either premiss, unless the middle
term were to be taken in the minor with an identificatory
relative-term (cum relativo identitatis),'f'
[3.4.30] The rules through which syllogisms hold require that if the
middle term is common, then the extremes are connected with it by
reason of the same thing for which that common term supposits, as
remarked before. When the middle is distributed in neither <premiss>, it
is possible that its connection with the major extreme is true for <the
case of> one thing and its connection with the minor extreme is true <for
the case of> another thing, and so no connection of the extremes to one
another can be inferred unless the middle term were forced by an

identificatory relative-term to hold for the same thing in the minor sent-
ence as what is in the verification of the major. But then the syllogism
holds, and clearly holds according to the rules given above, and is equiv-
alent to an expository syllogism, as for example in
(348) B is A and C is that same thing as B; therefore, C is A
Therefore, such syllogisms hold in all the moods in which the expository
syllogisms hold. And this is the single way in which we can argue from
two particular <premisses>. In the rest <of the chapter> we speak of the
syllogism assuming that there is no such identificatory relative-term
present.

[Theorem III-7J In every figure, if the middle is distributed in
one of the premisses then there is always a valid syllogism
with a conclusion connecting the two extremes, unless ampli-
ation prevents it or both <premisses> are negative.
[3.4.32] The reason for this is that one of the rules given above through
which syllogisms hold always applies in such a case. If the middle is
distributed in one of the premisses, then it must be that if the sentence is
true then it is so for all that for which the term supposits. Thus the other
premiss cannot be true unless it is true for something the same as that
for which the <first premiss> is true, and so the extremes are connected
through the middle by reason of something the same for which the
middle supposits. If the middle is distributed in each premiss then the
extremes are connected by reason of everthing which the middle sup-
posits for.
[3.4.33] How and why ampliation is an exception will be dealt with <in
[3.4.54]-[3.4.72], in Theorem II1-9 through Theorem IlI-12>.
[3.4.34] But one must pay close attention to what should be deduced;
thus, in order to draw our attention to this, another theorem follows.

[Theorem III-8] If the minor extreme is distributed in a
premiss 17 then a universal direct conclusion can be inferred,

and if not not; if the major extreme is distributed in a pre-
miss 1x then a universal indirect conclusion can be inferred,
and if not not; and if the predicate of a negative conclusion is
distributed in a premiss then the conclusion should be
formed in the common idiom <for negatives>, and if it is not
so distributed then the conclusion should be formed by
putting a negation after the predicate.
[3.4.36) All the claims <of Theorem III-8> are shown by Theorem 1-10,
for a term can in no way be distributed in the conclusion unless it is
distributed in a premiss. But it can be distributed <in the conclusion> if it
is distributed <in a premiss>, because for which things the extreme is
connected to the middle it can be inferred for those things to be
connected to the other extreme, and for no others.
[3.4.37) It should be noted that these three results, namely Theorem III-
6, Theorem III-7, and Theorem III-8, along with Theorem III-2 make
clear the number of all useful (utile) moods for syllogizing in any of the
three figures, indirectly as well as directly.
[3.4.38) In any figure there are sixteen conjugations,19 by combining

universal and particular by the affirmative and negative in the two
premisses: either both are universals, or both are particulars, or the
major is universal and the minor is particular, or <the major is particular
and the minor is universal>. And each of these four moods is divided
into four again, for either both <premisses> are affirmative, or both are
negative, or the major is affirmative and the minor is negative, or <the
major is negative and the minor is affirmative>.
[3.4.39) Now for all of those conjugations, in each figure there are four
useless (inutile) moods, namely those which are made up of a pair of
negatives. 20
[3.4.40) Also, in each figure there is a fifth useless mood, namely where
there are two particular affirmatives, for the middle is not distributed in
either premiss. 21
[3.4.41) In the first figure, if the major is particular (whether affirmative

or negative) and the minor a universal affirmative, the syllogism is not

valid, because the middle is not distributed. Thus there are seven useless
moods <of the first figure>. 22
[3.4.42] And there is an eighth useless mood <of the first figure>, where
the major is a particular negative and the minor a particular affirmative,
for the middle is not distributed. 23
[3.4.43] But the other eight moods <of the first figure> are usefu1. 24
[3.4.44] First of all, if the minor is negative and the major affirmative,
there are four useful moods,25 since the middle is distributed in the
minor sentence. But in these four moods if the minor is particular,26
then neither of the extremes is distributed in the premisses. In this case
no conclusion may be inferred, either directly or indirectly, in the com-
mon idiom for negatives, but a negation must be placed after 27 the pre-
dicate in the conclusion. 2H If on the the other hand the minor is
universal,29 then since the minor extreme is distibuted and the major
<extreme> is not, a conclusion may be inferred in the common idiom for
negatives, although indirectly and not directly. And these are thus the
two moods Fapesmoand Frisesomorum. 3o
[3.4.45] If each <premiss> is affirmative and the major is universal, then

there is a valid syllogism no matter what the minor is like, since the mid-
dle is distributed in the major. Thus there are two moods 3l in which a
conclusion may be inferred, and this again either directly or indirectly.
The direct moods of deduction are Barbara and Darii; the indirect
moods of deduction are Baralipton and Dabitis. 32
[3.4.46] If the major is a universal negative and the minor affirmative

(whether universal or particular), then there are two moods for valid
direct deduction, namely Celarent and Ferio. 33 If in these two moods the
minor is universal there can also be an indirect conclusion, and this is
Celantes;34 if the minor is particular there can only be an indirect con-
clusion which is not in the common idiom for negatives, for the minor
extreme is not distributed. 35
[3.4.4 7] And so we have eight useful moods, but only six of them are
valid in the common idiom for negatives, namely Barbara and Celarent
and Darii, which conclude indirectly3" as well as directly, and Ferio,

concluding directly alone, and Fapesmo and Frisesomorum, concluding
indirectly alone.
[3.4.48] It seems to me that Aristotle did not believe a syllogism to be

made up of premisses and conclusion, but rather only of premisses from
which a conclusion can be inferredY This is why he called it a power of
the syllogism that from one syllogism several things may be deduced
[Prior Analyties II.i 53a2-53b2]. Hence in the first figure, besides the
four moods which conclude directly in the common idiom for negatives,
Aristotle only posited those other two moods which conclude indirectly
in the common idiom, namely Fapesmo and Frisesomorum, and dismis-
sed those which do not conclude in the common idiom. Nor did he
enumerate Baraiipton, Celantes, and Dabitis as opposed to Barbara,
Ceiarent, and Darii, since according to his <definition of the syllogism>
the former do not differ from the latter.
[3.4.49] Now in the second figure nothing follows from purely negative
<premisses> nor from purely affirmative <premisses>, since the middle is
not distributed <in either case>; thus there are eight useless moods. 3H
[3.4.50] The other eight moods <of the second figure> are useful: Cesare
and Camestres, which conclude indirectly as well as directly,3Y and
Festino and Baroeo, which only conclude directly in the agreed-upon
manner of forming a conclusion. 40 Furthermore, if the premisses of
Festino and Baroeo are transposed, there are two other moods which
only conclude indirectly. These may be called Tifesno and Robaeo,41
and are proved by reduction to Festino and Baroeo merely by transpos-
ing the premisses. The two other moods do not conclude in the common
idiom for negatives, and only conclude in the uncommon manner of
speaking, namely when each premiss is particular one is affirmative and
the other negative. 42
[3.4.51] In the third figure, nothing follows from purely negative <pre-
misses> nor from purely particular <premisses>, since the middle is not
distributed <in either case>, and so there are seven useless moods. 43
[3.4.52] The other nine moods are useful for deduction, <and they are>
even in the common idiom for negatives: Darapti, Disamis, and Datisti
validly conclude indirectly as well as directly;44 Felapton, Bocardo, and

Ferison only validly conclude directly.45 Conversions of these last three,
namely Lapfeton, Carbodo, and Rifeson, only validly conclude indirectly
and are reduced to the direct <moods> by transposing the premisses:~6
[3.4.53] Now that these matters are settled, note well that I intend to
speak only of moods which support their conclusions in the common
idiom for negatives.

[Theorem III-9] An ampliative predicate does not prevent
the syllogistic moods given above if the subject of any of the
premisses or the conslusion is taken with the addition of the
"what is ... " <locution>, as for example: "Everything which is
B is A and everything which is C is B; therefore, everything
which is C is A" and in like manner for the other figures and
moods.
[3.4.55] The reason for <Theorem III-9> is because <using the "what
is ..." locution> prevents the ampliation.
[3.4.56] It is clear that with such a locution the first four moods which
Aristotle gives are perfect, holding evidently and explicitly by the dic-
tum de omni et nullo. The other moods which conclude in the common
idiom for negatives given by Aristotle are reduced to these <first> four
moods in the way he says they are reduced, since Aristotle holds the
conversions to be formal with such a locution. Although
(344) Aristotle is dead; therefore, a dead thing is Aristotle
does not follow, nevertheless
(345) He who is Aristotle is dead; therefore, what is dead was or is
Aristotle
does correctly follow.
[3.4.57] Now all other moods are reduced to the four perfect moods of
the first figure by conversion, except for Baraca and Bacarda. 47 But then
Baroca and all the other moods of the second figure can be reduced and
proved per impossibile, that is, from the major and the contradictory of
the conclusion the contradictory of the minor can be inferred by the
first figure, and so, by Theorem III-3, they are acceptable syllogisms.
Bocardo and all the other moods of the third figure are also proved in
this way, since from the minor and the contradictory of the conclusion
the contradictory of the major can be inferred <by the first figure>, and
so they are all acceptable <syllogisms>.
[3.4.58] The tenth theorem is:

[Theorem III-JO] An ampliative major extreme prevents a
universal direct conclusion, but it does not prevent a particu-
lar or a universal indirect <conclusion>.4x
[3.4.59] Theorem III-lO is given where no term is ampliative other than

the major extreme. And then the reason for Theorem III-I0 is that the
minor extreme, which is not ampliated in the premisses, is ampliated in
the conclusion. Thus if the major extreme is ampliative there will be a
change from the less ample to the more ample (processus a minus amplo
ad amplius). Such a change is invalid if the <case of> more is deduced
with distribution, but it is valid if deduced without distribution. For
example,
(346) Every man ... ; therefore, every animal - - -
is not valid, but
(347) Man ...; therefore, animal - - -
is correctly valid.
[3.4.60] The reason why the indirect universal conclusion is not pre-
vented is because the minor extreme is not ampliated in such a conclu-
sion, and so the only change is from the non-ampliated to the
non-ampliated.
[3.4.61] In order that these matters are clear even to beginning students,
I show that universal conclusions cannot be inferred in Barbara or in
Celarent <as required in Theorem III-I0> by giving counterexamples.
Against Barbara I posit the case that everything which is or will be my
horse is going to go to Rome, and that right now every horse is mine,
but tomorrow many horses will be born which will neither be mine nor
go to Rome. Then take the following syllogism:
(348) Every horse of mine is going to go to Rome and every horse
is a horse of mine; therefore, every horse is going to go to
Rome
the premisses are true and the conclusion false in the posited case, and
so the syllogism is invalid. Similarly, against Celarent I posit that no
horse of mine will go to Rome, and that every horse is now mine, and
that tomorrow there will be some <horse> which goes to Rome which will
not be mine. Then take the following syllogism:
(349) No horse of mine is going to go to Rome and every horse is a
horse of mine; therefore, no horse is going to go to Rome
[3.4.62] It should also be noted that when the predicate is not amplia-
tive if nevertheless the copula in the major sentence is ampliative, as for
example past-time or future-time, the same result should be given as in
the case of Theorem III-10, since the same arguments apply to each
case.
[3.4.63] The eleventh theorem is:

[Theorem III-ll] An ampliative middle does not prevent the
third figure.
[3.4.64] <The reason for Theorem III-II is> that the middle is put as the
subject and the extremes are predicated, and the predicate is not ampli-
ated by the subject. Thus the extremes are no more or less ample in the
premisses than in the conclusion. And you should understand Theorem
III-II so that only the middle term is ampliative.

[Theorem III-I2] An ampliative middle in the first or second
figure does not prevent the universal negative moods but
does prevent all the others. 49
[3.4.66] The reason why it does not prevent the universal negatives is
that the minor extreme is ampliated in the premisses and not ampliated
in the conclusion. Now it happens that there is negatively an acceptable

consequence from the more amply distributed to the less ample; thus
(351) Every animal is living; therefore, a horse is living
does not follow.
[3.4.67] A counterexample can be given in Barbara. 50 Although it is dif-

ficult to find a case, nevertheless let us posit that God, by His absolute
power, at some time before the end of the world brings it about that one
day all the people who are at that time old die and the young remain
<alive>, and that afterwards nothing living is generated and the young
ones live until they become old, and then they die, and then the world is
ended. Then on the day on which we posited the old people to die it is
argued:
(352) Everyone going to die is young, and everyone old is going to
die; therefore, everyone old is young
The premisses are true and the conclusion is false in the posited case;
the minor will clearly appear as true when analyzed.
[3.4.68] It is similarly apparent why particular syllogisms in the first or

second figure are not valid. This is because in particular syllogisms (I
speak here of those which conclude directly) the minor sentence is par-
ticular, since in these two figures the major must be universal, and so the
minor extreme is not distributed. There is no valid change from the un-
distributed number to the lesser.
[3.4.69] It is sufficiently clear that an ampliative middle does not pre-

vent indirect syllogisms in these two figures where the minor is a univer-
sal negative, for the same reason it does not prevent universal direct
negatives.
[3.4.70] It is also clear that an affirmative universal syllogism is accept-

able on the hypothesis or assumption of the nonemptiness of the subject
in the conclusion, i.e. that in the conclusion the subject supposits for
something.
[3.4.71] From the last three theorems given the careful reader can see
how syllogisms are prevented or are not prevented if the major extreme
and the middle are both ampliative terms. If they have the same mode
of ampliation, as for example each to the past or each to the future, then
no syllogism is prevented, for we do not shift from the more to the less
ample but from equally ample to equally ample. If the major extreme
ampliates in a way differnt from the middle, then it prevents what
Theorem Ill-I0 says is prevented by the major extreme. If the middle
ampliates in a way different from the major extreme, then it prevents
what Theorem III-II says is prevented. And if each ampliates in a way
different from the rest, then each are prevented so that there is no
formal consequence (although perhaps a counterexample will not be
immediately obvious for some).
[3.4.72] It should also be noted that the same can be said if there is an
ampliative copula in the minor sentence of the first or second figure, as
for example past-time or future-time, <that is, we can say the same as if>
the middle term were ampliative, for the same arguments apply in each
case.
[3.4.73] Therefore, these theorems settle questions aabout syllogisms

with past-time or future-time sentences. It only remains now to settle
matters about syllogisms with oblique terms.
CHAPTER III-5. SENTENCES WITH OBLIQUE TERMS
[3.5.1] We now turn our attention to syllogisms with oblique terms. We

shall first of all deal with oblique terms which are combined with nomin-
ative terms, which they govern as a determination of the nominative
term, just as an adjective is a determiination of a substantive <term>.
When we say "A white horse is running" the word 'white' dtermines the
word 'horse' to supposit for the white ones; so too when I say "The
horse of-Socrates is running" the word 'of-Socrates' contracts the word
'horse' to supposit for only those which are Socrates's
[3.5.2] For this reason I am of the opinion that an adjective, unless it be

of the neuter gender of substantives, cannot by itself take the place of
the verb or be the entire subject of a categorical sentence, and likewise
neither can an oblique term. 51 Nevertheless, I do not say this authori-
tatively, since I need not assume it in what follows.
[3.5.3] However, it is clear from [3.5.1]-[3.5.2] that sometimes the

whole made up of a nominative and an oblique <term> is distributed by a
single distribution, and neither <is distributed> individually. For example,
if I say
(353) Every ass of-a-man is running
neither 'ass' nor 'of-a-man' is distributed, but the whole aggregate 'ass
of-a-man.' Hence we cannot syllogize by taking something <which falls
under> 'ass' or <under> 'of-a-man.' Therefore
(354) Every ass of-a-bishop is running and Brunellus is an ass;
therefore, Brunellus is running
does not follow, and neither does
(355) Every ass of-a-bishop is running and Socrates is a bishop;
therefore, an ass of-Socrates is running
because if Socrates does not have an ass the conclusion is false, although
the premisses may be true. But if we said in the minor sentence <of
(355» that Brunellus is the ass of a bishop,52 then we could correctly
deduce that Brunellus is running.
[3.5.4] It is nevertheless true that sometimes an oblique term is distrib-

uted without the nominative term, as for example when a distributive
sign is added to the oblique term in the same case as the oblique term
(and so different from the case of the nominative term), as in
(356) Of-any-man an ass is running
in (356) 'man' is distributed and 'ass' is not distributed, and the whole
'ass of-a-man' is not distributed either. Similarly for
(357) The ass of-any-man is running
In (357) 'ass supposits determinately, since no cause of it <suppositing>
confusedly precedes it, yet it is true that 'ass' does not supposit taken by
itself but is contracted by the succeeding distributed oblique term.
Hence (357) is not true if no asses are running, or if no running ass is
the ass of any man.
[3.5.5] Next, it sometimes happens that the nominative and not the obli-
que term is distributed, nor the whole aggregate of nominative and obli-
que. Rather the oblique term retains determinate supposition, namely if

it precedes a distributive sign as in
(358) Of-a-man every ass is running
Nor does
(359) Of-a-man every ass is running; therefore, every ass of-a-man
is running
follow. Nevertheless, in (359) the term 'ass' is not distributed simply but
rather is contracted by term 'of-a-man' suppositing determinately. This
is why
(360) Of-a-man every ass is running, and Brunellus is an ass; there-
fore, Brunellus is running
<does not follow>. Similarly, since the whole aggregate is not distributed,
(361) Of-a-man every ass is running, and the ass of-the-bishop is
the ass of-a-man; therefore, the ass of-the-bishop is running
does not follow, since perhaps the bishop's ass remains in the stable and
Socrates is the man whose every ass is running.
[3.5.6] And so if we want to syllogize here by taking what falls under the
distribution of the given sentence, we must take the minor sentence with
an identity-relation added to the term 'of-a-man' so that the middle is
forced to hold for the same in <both> premisses, e.g.
(362) Of-a-man every ass is running, and the ass of-the-bishop is
an ass of-that-man; therefore, the ass of-the-bishop is run-
ning
correctly follows. In this way the rule by which affirmative syllogisms
were said to hold is preserved.
CHAPTER IIJ-6. SYLLOGISTIC EXTREMES AND MIDDLE
[3.6.1] Sometimes in a syllogism with oblique terms the syllogistic ex-

tremes or the middle need not be an extreme of some premiss.
[3.6.2] I call 'syllogistic extremes' what are inferred in the conclusion to

be connected in that they are each connected to the middle in either
premiss, and I name 'the extremes of a sentence' its subject and predic-
ate.
[3.6.3] I say therefore that the syllogistic middle is sometimes neither

the subject nor the predicate in the major sentence, but is a part of the
subject or <a part> of the predicate, and similarly neither of the extremes
are the subject or predicate in either the premisses or the conclusion.
[3.6.4] For example, this syllogism is acceptable:

(363) A man is seeing every horse s3 and Brunellus is a horse;
therefore, a man is seeing Brunellus
but in (363) 'horse' is the middle, which is neither the subject nor the
predicate in the major sentence. The major extreme is 'man' and 'see-
ing,' which are connected with the middle term 'horse' in the major sen-
tece. And it is clear that one of these two terms is found in the sentence
on the side of the subject and the other on the side of the predicate, and
hence they are not the subject nor are they the predicate. Furthermore,
the term 'Brunellus' is the minor extreme, which is connected with the
given middle in the minor sentence.
[3.6.5] And so it is clear that neither the major nor the minor extreme is
the subject or the predicate of the conclusion, but the major extreme is
connected to the minor in the conclusion in the same order as it is con-
nected to the middle in the major sentence.
CHAPTER III-7. THEOREMS
[3.7.1] Finally, we suppose in this chapter that the predicate is restricted

by the preceding verb to supposit for things (pro suppositis) of the time
of the verb.
[3.7.2] For example, if I say

(364) Socrates saw a white man yesterday
(365) I struck a running man
then (364) is true only if the man whom Socrates saw was white when
he saw him, <and (365) is true only if the man> whom I struck was runn-
ing when I struck him. Hence if the predicate or a term found in the pre-
dicate is distributed then <a sentence taking something falling under the
predicate as its predicate> is not validly <inferred>, except at the same
time for which the distribution is made. Thus these do not follow:
(366) I did not see a running man yesterday, and Socrates is a
running man; therefore I did not see Socrates yesterday
(367) I never struck an old man, and Socrates is an old man; there-
fore, I never struck Socrates
One must rather argue
(366*) I did not see a man running yesterday, and whenever I saw a
man running yesterday Socrates was a man running; there-
fore, I did not see Socrates yesterday
(367*) I never struck an old man, and whenever I struck a man Soc-
rates was an old man; therefore, I never struck Socrates
[3.7.3] Again, as was touched on in [1.6.12], there are some verbs which
go over into their accusative objects,54 which they govern such that the
act indicated by the verbs do not simply go over into the things for
which those objects supposit, but they go over into them by certain me-
diating concepts indicated by those terms suppositing for them.
[3.7.4] Verbs <of the sort described in [3.7.3]> are: 'know; 'understand;
'comprehend; 'opine; 'appear; 'judge; 'see; 'hear; 'imagine,' etc., and
consequently also 'desire; 'will,' 'love; 'want; 'hate; etc., and finally also
'promise; 'sell; 'buy; 'owe; 'give; 'condemn; etc. In like manner, the
participles of or the names derived from these verbs also have this ca-
pacity <described in [3.7.3]> with respect to the accusative objects which
they govern.
[3.7.5] Therefore, such verbs or participles, or names derived from

them, restrict their accusative terms which they govern to sup posit for
those things for which they do not supposit absolutely but with the
appellation of a ratio or a concept according to which the given terms
signify what they signify. Hence one term cannot be changed into an-
other <term> of a different ratio by syllogizing in taking <something>
falling under the distribution or in taking the predication of that term of
another term or other terms.
[3.7.6) This mode of supposition <described in [3.7.5» is assimilated to

simple or material supposition, in which it is not permitted to take
something falling under a distributed term. For example, if it is neces-
sary that every man is an animal and a man is running, it does not
thence follow that it is necessary that everything running is an animal.
These similarly do not follow:
(368) Any God always was and is a God with three Persons, and
Averroes knew God; therefore, Averroes knew a God with
three Persons
(369) Socrates does not know about prime matter [or, <Socrates>
does not have knowledge of prime matter) and all prime mat-
ter is Nature; therefore, Socrates does not know about
Nature
And if what many people say - namely, that the light of the Sun has a
shape and a size - then I may see the Sun's light, although I do not see
the shape or the size of the light of the Sun, as when there appears a
rainbow or when I see the daylight before the Sun rises.
[3.7.7) However, if the aforementioned terms or <accusative> objects

which are governed by this sort of verb or participle precede those
verbs or participles, then they are not restricted to appellate those con-
cepts or rationes.
[3.7.8) The thirteenth theorem is:
[Theorem III-13) For any given sentence with a distributed

term, whether nominative or oblique, an acceptable syllogism
can be constructed by taking another term under the given
term as the minor sentence, unless the supposition is
material. 55
[3.7.9J It seems to me that Theorem III-13 holds by the nature of

distribution, which is clear if you know that 'distribution' means. These
syllogisms are just as perfect as those constructed in the first figure with
simple nominative terms. Hence they can be reduced to the first figure
by their likeness <to it>, since they hold directly and immediately by the
nature of distribution and the taking of something falling under a distri-
buted term.
[3.7.10] The exception given in <Theorem III-13> about material sup-

position does not interfere with the nature of distribution, since a dis-
tributive sign need not act in the same fashion when it is taken
materially in a sentence as when it is taken significatively. Similarly, the
exception about the appellation of a concept or a ratio does not inter-
fere with the nature of distribution, since then a term is not distributed
simply but rather with some addition or determination implied (subin-
tellecta) or indicated by the conditions of those verbs. For example, if I
assert
(370) Socrates does not know prime matter
the term 'prime matter' is not distributed simply. Hence it does not fol-
low that Socrates does not know any prime matter. Rather, explain the
sense of (370) as follows:
(370*) Socrates does not know prime matter according to the ratio
by which it is called prime matter
and (370*) is the sense of (370). Hence you take under the given distrib-
ution that Nature is prime matter, and conclude that Socrates does not
know Nature according to the ratio by which it is called prime matter.
Similarly (368) does not follow, but rather the conclusion should be that
A verroes does not know God according to the ratio by which God is
called triune.
[3.7.11] It should be remarked that the exception about the appellation

of a certain time does not remove the meaning of the condition about
distribution <in Theorem III-13>, since if we take something falling under
a distributed term at the time for which it is distributed, then, in the way
mentioned previously, the syllogism will be valid.
[3.7.12] Therefore ([3.7.11]) does not prevent Theorem III-13, but

means rather that we must carefully take the minor under the major and
carefully infer the conclusion. Thus I say that
(371) B is every A, and C is A; therefore, B is C
clearly follows, as would "<B is every A, and C is A>; therefore, C is B"
and similarly these all follow:
(372) A man sees every horse and Brunellus is a horse; therefore, a
man sees Brunellus
(373) Of-any-man an ass is running and Socrates is a man; there-

fore, the ass of-Socrates is running
(374) The earth is surrounded by every celestial sphere, and the
moon's orbit is in a celestial sphere; therefore, the earth is
surrounded by the Moon's orbit
(375) Every horse of-the-king is running, and Brunellus is a horse
of the same king; therefore, Brunellus is running

[Theorem IlI-I4] For any affirmative sentence in nominative
terms, anything affirmatively connected to the subject, nomi-
natively or obliquely, the same <term> taken in the same way
can be deduced to be connected to the predicate, if the dif-
ference of the time (alietas temporis) or the appellation of a
ratio does not prevent it.
[3.7.14] The reason for Theorem III-14 is that if one thing is the same
as another, then anything which has some attribution to the one also has
the attribution to the other. For example, all of these follow:
(376) Every man is running and you see a man; therefore, you see
a runner
(377) Every horse is black and you have a horse in the stable;
therefore, you have <something> black in the stable
(378) Every ass is running and an ass is of-a-man; therefore, <som-
ething> of a man is running
Aristotle syllogizes in this way in the Prior Analytics I [cap.xxxvi 48b22]:
(379) All wisdom is knowledge and wisdom is good; therefore,
<some> knowledge is good
[3.7.15] Nevertheless, this does not follow, because of the difference of

the time <involved>:
(380) Every horse is running and you had a horse in the stable;
therefore, you had <something> running in the stable
since perhaps the minor is true not for some horse for which 'horse' is
distributed in the major, but for another <horse>. If, however, we force a
term to stand for the same in the minor and the major by an identity-
relation, then
(381) Every horse is running and you have some one of those
horses; therefore, you have something running
is an acceptable syllogism.
[3.7.16] Similarly, the appellation of a ratio may prevent the syllogism.

For
(382) Coriscus is approaching and you know Coriscus; therefore,
you know the one approaching
does not follow, but only
(382*) <Coriscus is approaching and you know Coriscus>; therefore,
the one coming is one you know
[3.7.17] The fifteenth theorem is:

[Theorem 1II-I5] In any given sentence in which an oblique
term is put first, if we use it as the subject and the rest as the
predicate then for any figure those syllogisms are valid which
are valid for those in nominative terms.
[3.7.18] The reason for this is that many people hold the truth of the
matter to be that the oblique term is the subject and the rest is the pre-
dicate. And even if it is not the case as these people proclaim, neverthe-
less any such sentence is equivalent to another sentence in which the
nominative term is taken in the place of the oblique term, and is the sub-
ject, and the rest is the predicate, in the resolution others have given.
For example, these are equipollent:
(383) Of-a-man the ass is running
(383*) There is a man whose ass is running
and so we may form syllogisms from either equally well. Hence in the
first figure or its likeness, there is a syllogism like this:
(384) Of-any-man the ass is running, and Socrates is a man; there-
fore, the ass of-Socrates is running
and in the second figure <there is a syllogism> like this:

(385) Of-any-colonel the horse is running and of-no-clerk the
horse is running; therefore, no clerk is a colonel
and in the third figure <there is a syllogism> like this:
(386) Of-any-man the ass is running and of-any-man the horse is
walking; therefore, there is someone whose ass is running
and whose horse is walking
[3.7.19] The sixteenth theorem is:

[Theorem III-16] Whenever some hypothetical or quasi-hy-
pothetical sentence is formed with a <term> of some category
distributed in the first categorical and its relative <term> in the
second categorical,56 then for any term taken under the dis-
tributive <term>, with regard to the same <thing> attributed to
it there can be deduced what the relative <term> attributes, by
forming such sentences:
Whatever you bought, that thing you ate
However much Socrates is, so much <too> is Plato
However Socrates is related to Plato, so is John <related>
to Robert
Wherever Socrates lectures, Plato is there listening
Whenever the king is angry, his servants are then fearful
[3.7.20] I call such a sentence 'hypothetical or quasi-hypothetical' be-

cause some people say that such are complex (and I believe this is true),
while others say that they are categorical, although they are very close
to hypothetical sentences. But since I am dealing with consequences I
don't care what they are called; nevertheless, I shall conventionally
name them 'hypothetical relatives' (hypotheticas relativas). It is com-
monly allowed that temporal and locational <sentences> are hypotheti-
cals, for the same reason.
[3.7.21] And so, just as I sometimes express in such sentences what is

asked or distributed in the category of Substance or Quantity or Quality
with its relative <term>, so too I express these two <temporals and loca-
tionals> in the categories of Time and Place. For example, just as I say
(387) Howsoever Socrates is, such is Plato

(388) However much Socrates is, so much is Plato
thus 1 say or 1 might say
(389) Wherever Socrates is, Plato is there
(390) Whenever <Socrates> exists, <Plato> exists
Similarly, sometimes 1 express only what is asked in the category of Sub-
stance or Quantity of Quality and take the relative-term as implicit
(subintelligo relativum), and 1 often do likewise for <the categories of>
Time and Place. For example, just as 1 say
(391) Socrates is what Plato is
(392) Socrates is as much as Plato is
(393) <Socrates is> as Plato is
1 may also say
(394) Socrates is when Plato is
(395) Socrates is where Plato is
Thus there is an equal reason to call them all hypotheticals or all cate-
goricals.
[3.7.221 It should be noted that just as we have in the Latin language (in
nostro idiomate) question with appropriate relative-terms in the catego-
ries of Substance, Quantity, Quality, Time, and Place,57 so perhaps the
Greeks had other categories (which we may impose), and perhaps used
the common question 'how-is-it-relatedT (quomodo aut qualiter se
habet) in them. And 1 don't much care about this, but rather only that if
there are such questions and relative terms corresponding to them ap-
propriate to a category, then Theorem 111-16 will be true for them.
[3.7.231 It seems to me that the syllogisms <discussed in Theorem III-

16> hold by the meaning of 'distribution' and 'relation: For <they hold>
by distribution in that the distributive term appropriate to a category
distributes the category into all the suppositing terms in it, and hence
any of them may be taken to fall under it. The identificatory relative-
term corresponding to it indicates that what is atributed to it can be at-
tributed to that which falls under the distributive term. Hence it seems
to me that such syllogisms ought to be called as perfect as any, by the
nominal definition.
[3.7.24] It should be noted that these syllogisms are valid no matter

what the categoricals from which the hypothetical-relative sentence is
constructed are like. For example, these all follow:
(396) However Socrates is, Plato is such
(397) However much you do not see I do not see, and you do not
see a two-foot-length; therefore, I do not see a two-foot-
length
(398) Whenever Socrates existed Plato did not exist then, and
yesterday Socrates existed; therefore, Plato did not exist
yesterday
(399) Wherever Socrates is Plato is not there, and Socrates is in the
church; therefore, Plato is not in the church
Nor must one give the same situation in both of the categoricals in the
first hypothetical in syllogisms of this sort; indeed, it is permissable to
argue as follows:
(400) Whomsoever you see, he is white; and you see John; there-
fore, John is white
(401) However-long is the ruler, you can only measure <that long>;
and the ruler is a two-foot-<length>; therefore, you can <only>
measure two-foot-<lengths>
This is how Aristotle syllogizes in the Prior Analytics I [cap.xxxvi,
48b22]:
(402) Of-whatsoever there is knowledge, of this there is a genus;
and knowledge is good; therefore, the genus is good
[3.7.25] It should also be noted that in syllogizing this way special care
should be taken with ampliations, namely lest we deduce from a more
ample but not distributed term a less ample <term>, and lest from a less
ample <term> we deduce a more ample distributed <term>. For these do
not follow:
(403) Whatever is dead, that thing is not alive; and a man is dead;
therefore, a man is not alive
(404) Whatever is alive, that thing is not dead; and every man is
alive; therefore, no man is dead
And particular care should be taken over ampliations of time, form, and
ratio with respect to the predicate; indeed, these do not follow:
(405) Whatever you bought so you ate, and you bought raw meat;
therefore, you ate raw meat 58
(406) Whatever you know, that I know; and you know matter;
therefore, I know matter
since perhaps it is possible that I know matter and I do not know matter.
[3.7.26J The seventeenth theorem is:

[Theorem 1//-17] In the aforesaid manner of arguing, namely
by terms which distribute with respect to categories and their
<corresponding identificatory> relative-terms, it is not permis-
sible to take under a term distributive of one category some
term of another category, except for the case of the subject
falling under a term distributive of Substance where terms of
other categories may be taken if ampliation does not prevent
it.
[3.7.28J <For example>, these do not follow:

(407) However-much is running, so much is disputing; and Soc-
rates is running; therefore, Socrates is disputing
(408) <However-much is running, so much is disputing>; and a
white thing is running; therefore, a white thing is disputing
Nevertheless, this follows acceptably:
(409) Whatever you see I see, and <there is> a white thing <which>
you see; therefore, <there is> a white thing <which> I see
although it does not follow that, therefore, I see a white thing. Nor does
this follow:
(410) Whatever you saw, that you see; and <there is> a white thing
<which> you saw; therefore, <there is> a white thing <which>
you see
for ampliation prevents this consequence.
[3.7.29J The reason why terms of other categories can be taken under a
term distributive of Substance, and not conversely, is that the relative-
term for <the category of> Substance simply indicates identity, but the
relative-terms of the other categories do not simply indicate identity but
only similarity, equality, same-place, and so on.59 And an affirmative

sentence through which <a term of another category> is taken under the
distributive term indicates the simple identity of that for which the sub-
ject supposits with that for which the predicate supposits, no matter
what category the terms are.
[3.7.30] Many more things could be said about this particular manner of
syllogizing, some of which I leave to the thoughtful investigations of the
careful reader.60
( [Theorem lII-18] In every figure, from two affirmatives or

. from two negatives there follows a negative conclusion, al-
tering the middle according to finite and infinite.61
[3.7.32] In Theorem III-18 it should be observed that the middle is dis-

tributed in at least one of the premisses and that <the syllogism> con-
cludes universally or particularly, directly or indirectly, in the common
idiom for negatives or not, as these questions were settled in Theorem
III-6, Theorem III-7, and Theorem III-8.
[3.7.33] The reason for Theorem III-18 is that for whatever a finite
term supposits the infinite term does not supposit, and for whatever the
finite term does not supposit the infinite term does supposit, and con-
versely.62 Hence that by which the finite term and the infinite term are
connected affirmatively should not be called the same, for from the af-
firmatives a negative follows. Also, a negative follows from negatives,
since either of the terms by which the finite term and the infinite term
are connected negatively supposit for something or one of them does
not supposit for anything; if one of them does not supposit for anything,
then it is clear that the conclusion must be negative. If each supposits for
something, though, then it is impossible that the finite term and the infi-
nite term are denied of the same; hence they must supposit for different
things, and so the negative conclusion will still be true.
[3.7.34] Now, for greater clarity, Theorem III-18 is shown <to apply> in
each of the three figures. First of all, I say that this is an acceptable syl-
logism <in the first figure>:63
(411) Any B is A and any C is non-B; therefore, any C A not is

and we should not deduce "no C is A," since 'A' is not distributed in the
premisses. The syllogism (411) is proved: from the major and the
opposite of the conclusion there follows the opposite of the minor; the
contradictory of the conclusion is "Some C is any A," from which it
follows that any A is C; then we argue as follows:
(412) Any A is C and any B is A; therefore, any B is C
therefore, by conversion, "Some C is B," and this is inconsistent with
<"any C is non-B">, the minor of (411). I also state that in (411) an
indirect conclusion such as "Some A is not C" can be deduced, since it
is immediately obvious that from the opposite of the major and the
conclusion there follows the opposite of the minor, and this is because
'C' is distributed in the minor sentence.
[3.7.35] Similarly, this is an acceptable syllogism in the second figure: 64

(413) Any B is A and any C is non-A; therefore, no Cis B
Proof: from the minor "any C is non-A" there follows "no C is A," and
then the syllogism is acceptable without changing the middle according
to finite and infinite.
[3.7.36] Similarly, this follows in the third figure: 65

(414) Any C is A and any non-C is B; therefore, some B is not A
Proof: when the minor is converted this is in the first figure, and this is
proved per impossibile, since the contradictory of the conclusion is
"Any B is any A," from which it follows that any A is any B, from which
along with "Any C is A" it follows that any C is any B, and this is incom-
patible with the minor <of (414».
[3.7.37] Next we show that from negatives there follows a negative.
[3.7.38] First of all, I say that this is an acceptable syllogism <in the first
figure>:66
(415) No B is A and no C is non-B; therefore, no C is A
from the major and the opposite of the conslusion there follows the op-
posite of the minor, like so:
(416) No B is A and some C is A; therefore, some C is not B

Now the conclusion of (416) is inconsistent with the minor of (415),
assuming that C exists, since "C exists and it is not B; therefore it is
non-B" certainly follows, but "C is non-B" obviously contradicts the
minor of (415). The given assumption, namely that C exists, is taken
from positing the contradictory of (415) (if there is no ampliation).
Indeed, if there is no ampliation, then "Some C is A; therefore, some C
exists" follows.
[3.7.39) From [3.7.38) it is immediately apparent that in the second fi-

gure this syllogism is acceptable: 67
(417) NoB is A and no C is non-A; therefore, no C is B
since when the major is converted it is in the first figure.
[3.7.40) Similarly, this syllogism in the third figure is acceptable: 6K

(418) No C is A and no non-C is B; therefore, no B is A
since when the minor is converted we have again <a syllogism in> the
first figure, and those universal negatives <of (418» are simply converti-
ble unless ampliation prevents it.
[3.7.41) It should be noted that we frequently argue in this way, though

we do not note the form <of our arguments>. We often argue thus:
(419) What is not animal is not man, and a stone is not animal;
therefore, stone is not man
and this way of arguing is not formally valid unless we understand the
major to be taken universally; and if it is taken universally, it is equiva-
lent to "No non-animal is man," and so is obviously a syllogism in the
first figure with the middle changed according to finite and infinite.
[3.7.42) The nineteenth theorem is:

[Theorem 1II-19) (a) From a major affirmative and a minor
negative in the first figure with the middle changed according
to finite and infinite there follows an affirmative conclusion,
assuming the nonemptiness of the negative terms; similarly,
(b) from one <premiss> affirmative and the other negative in
the third figure, no matter which is affirmative, <there follows

and affirmative conclusion like before>; but (c) nothing like
this follows in the second figure.
[3.7.43) The reason for Theorem IlI-19(a) is that from the minor, which
is negative, there follows an affirmative by changing the predicate ac-
cording to finite and infinite (assuming the nonemptiness of the terms),
and then we hasve a syllogism in the first figure with the middle un-
changed.6 <i
[3.7.44) And there is a similar reason for Theorem III-19(b), namely

that from a negative there follows, assuming the non emptiness of the
terms, <an affirmative with the predicate changed according to finite and
infinite>.70
[3.7.45) And Theorem III-19(c) is clear, that it does not hold for the
second figure, since every man is an animal and no ass is a non-animal,
but it does not follow that an ass is a man. Still, it holds in the second
figure if the major extreme is changed according to finite and infinite.
BOOK IV
SYLLOGISMS WITH MODAL SENTENCES
CHAPTER IV-I. SYLLOGISMS WITH COMPOSITE MODAL

SENTENCES
[4.1.1] Finally, in Book IV we are concerned with syllogisms made up

of modal sentences.
[4.1.2] First of all, remember that some modals are composite and
others divided. l And in all divided <modals> de necessario or de possibili
the subject is ampliated to supposit for those things which can be, unless
ampliation is prevented by the locution "what is ..." added to the sub-
ject. 2 Those in which that given addition is not made are called 'de pos-
sibil!' or 'de necessaria' simply, and those in which the given addition is
made are called 'de possibili-for - those-things-that-are' or 'de neces-
sario <-for-those-things-which-are>.'
[4.1.3] Aristotle was content to distinguish assertoric sentences, calling

some 'simply assertoric' and others 'assertoric ut nunc' [Prior Analytics
I.xv 34b7-8]. But he does not mean the same thing by 'simply' here as
he does in Topics II.xi [115b29J, where he says "what I say without addi-
tion I say simply." In that case, "Socrates is running" would be simply as-
sertoric, since without addition it is true to say that it is assertoric. But
Aristotle means by 'simply assertoric sentence' an assertoric sentence
which is necessary, and by 'assertoric sentence ut nunc he means an as-
sertoric sentence which is contingent.
[4.1.4] And so we should hold that when an assertoric sentence ut nunc

is simply and absolutely speaking literally assertoric, a syllogism with
only assertoric premisses does not hold in virtue of its form if a counter-
example can be found in an assertoric sentence ut nunc. Nevertheless,
although it is not formal simply speaking, it can be acceptable and for-
mal on the hypothesis or assumption that it is simply assertoric, i.e. that
it is necessary.3
295
[4.1.5] It should be noted that adding the hypothesis of [4.1.4] to a

syllogism makes it just as valid as if one put a composite de necessaria
<sentence> in the place of the assertoric <sentence>. For example, it is
virtually valid to syllogize and say that something is true or false, stating
that man is an animal with the assumption that such a sentence is simply
assertoric, as it is to say "That man is animal is necessary."
[4.1.6] Although Aristotle by way of example seems to posit that such

sentences are true:
(420) Every man of necessity is an animal
(421) Everything white is necessarily not black
it should be pointed out that verification is not required for these kinds
of examples, for these sentences are simply false. Anything which can
fail to exist can fail to be an animal, but a man (e.g. Socrates or Plato)
can fail to exist, and so a man can fail to be an animal, and this contra-
dicts (420). Similarly for (421), since a white thing can become black,
and hence a white thing can be black.
[4.1.7] I do not believe that the sentence "A horse is an animal" is ne-
cessary, speaking of a sentence simply necessary, because God could
annihilate all horses at the same time so that no horse would exist, and
thus "A horse is an animal" would be false. Therefore it is not necessary,
though nevertheless such <sentences> can be allowed to be necessary
with a conditional or temporal necessity, in this way: Every man is of
necessity an animal if <at least one man> exists and every man is of ne-
cessity an animal while one exists.
[4.1.8] It should be remarked that Aristotle thought such sentences to

be simply necessary, for he held that the eternity of the world and uni-
versal Nature to not allow that at some time no horse or no dog would
exist. It is true that this is not possible by natural actions, although it
may well come about through a supernatural miracle that at some time
no horse and no fire and no earth exist. Hence speaking only naturally
these sentences "A horse is an animal" and "Fire is hot" are necessary,
i.e. it is not possible for them to be falsified by nature, leaving miracles
to one side. And this is the way in the rest <of the treatise> we shall speak
of such <sentences> as necessary.
IV. SYLLOGISMS WITH MODAL SENTENCES 297
[4.1.9] With these matters settled, some theorems follow.

[Theorem IV-I] In every figure, (a) nothing follows from a
pair of composite <modal sentences> which are either de pos-
sibili or contingent; but (b) from a pair <of premisses> com-
posite de necessario or de vero, there follows a composite
conclusion in the same mode.
[4.1.11] Theorem IV-1 should be understood as attending to figures
and modes with respect to the dictum, not with respect to the whole
sentence, and that the whole sentence is affirmative with the mode af-
firmed, although perhaps the dictum is denied. 4
[4.1.12] Then the reason for Theorem IV-1 is that, as said in [3.4.18],
neither of the premisses in a syllogism is the entire antecedent; rather,
the conjunctive sentence made up from the two premisses is the whole
antecedent. Now every conjunctive sentence made up from truths is
true, and every conjunctive sentence made up from necessary <sen-
tences> is necessary, but some conjunctive sentences made up from pos-
sible <sentences> are impossible. For example, this is made up from
possible <sentences>:
(422) Every man runs and some man does not run
and nevertheless (422) is impossible. Therefore, if both premisses are
necessary the whole antecedent is necessary, and the conclusion is
necessary if formed. And so too for the case of truth. But if both pre-
misses are possible, then the whole antecedent need not be possible
because of this; hence it does not follow that the consequent is possible.
For example, this does not follow:
(423) 'Everything running is a horse' is possible (or contingent),
and 'Every man is running' <is possible (or contingent»;
therefore, 'Every man is a horse' is possible (or contingent)
since the premisses are true and the conclusion is false.

[Theorem IV-2J In every syllogism made up of composite
<modal sentences> de necesssario or de vera it is permitted to

put in the place of the sentence de necessario a sentence de
impossihili with the contradictory of the dictum and <to put>
in place of the sentence de vero a sentence de falso with the
contradictory of the dictum.
[4.1.14] The reason for this is that these quasi-equipollences follow

from the Law of Contradictories." If one of the contradictories is neces-
sary then the other must be impossible, and conversely; if one <of the
contradictories> is true then the other must be false, and conversely.
Therefore, this is an acceptable syllogism:"
(424) It is impossible that some B is not A, and it is impossible that

some C is not B; therefore, it is impossible that some B is not
A
Similarly, by <Theorem II-IS>, from a sentence de vera there follows its

dictum and from a sentence de falso there follows the contradictory of
its dictum, this is an acceptable syllogism:
(425) It is false that some B is not A, and it is false that some C is

not B; therefore, it is false that some C is not A
alternatively, we may deduce "<It is false that some B is not A, and it is

false that some C is not B; therefore>, every C is A."
[4.1.15] The third theorem is:
(426) That every B is A is known by Socrates, and that every C is

B is known by Socrates; therefore, that every C is A is
known by Socrates
because although the two <premisses> are known by Socrates, he never-

theless may perhaps fail to order them in a syllogism, or fail to see that
the conclusion follows from them. 7
CHAPTER IV-2. SYLLOGISMS WITH DIVIDED MODAL

SENTENCES
[4.2.1] I now stop discussing composite modals and turn to divided

[v. SYLLOGISMS WITH MODAL SENTENCES 299
<modals>. Whenever I speak of <divided modals> de passibili or de neces-

saria, I only mean those which have an affirmed mode, even if they have
a negation falling under the dictum. We saw in <the Corollary given in>
[2.6.7] that by equipollence if the matter is settled for these it is settled
for those with a denied mode.
[4.2.2] I also only speak here of conclusions which are direct and
formed in the common idiom for negatives, since, by settling the matter
for these, careful readers can easily extend the results to others.
[4.2.3] The fourth theorem is:
[Thearem IV-4] In the first figure, there is always a valid

syllogism from a pair of premisses de necessaria or <a pair of
premisses> de passibili or from <a pair of premisses> one of
which is de necessaria and the other <of which> is de passibili
to a conclusion with the same mode as there is in the major
sentence.
[4.2.4] These claims are clear from the dictum de amni et nul/a, <stated
in Rule 1 and Rule 2 in [3.4.5]-[3.4.7]>, and they are all perfect (or
quasi-perfect) syllogisms. 8 If the major is explicitly expressed by a
disjunction of the verb 'is' with the verb 'can,' then if the minor is de
passibili taken under the distribution of the major <Theorem IV -4> is
clear, and if the minor is de necessaria then the same holds true, since a
sentence de passibili follows from a sentence de necessaria.
[Thearem IV-5] In the second figure, (a) there is always a

valid syllogism from a pair of premisses de necessaria or
from <a pair of premisses> one of which is de necessaria and
the other de passibili to a conclusion <which is> de necessaria;
but (b) there is no valid syllogism from two sentences de
passibili. 9
[4.2.6] It is clear that there can be no valid syllogism with two de

passibili premisses, because in these terms 'creating,' 'God,' and 'first
cause' there is a false negative conclusion and in these terms 'running,'
'man,' and 'horse' there is a false affirmative conclusion (where the

middle terms are 'creating' and 'running'). Therefore no affirmative
conclusion and no negative conclusion can follow formally,IO <and so
Theorem IV-5(b) holds>.
[4.2.7] Proof that a conclusion de necessaria follows if one or both of

the premisses is de necessaria: the opposite of the minor always follows
from the major and the opposite of the conclusion, and so <Theorem IV-
5(a» is apparent from Theorem IVA ifthe syllogisms are formed.
[Thearem IV-6] In the third figure, (a) a conclusion de

passibili always follows from two premisses which are de
passibili; (b) a conclusion de necessaria always follows from
two premisses which are de necessaria; (c) from one premiss
de necessaria and the other de passibili there always follows
a conclusion with the same mode as the mode of the majoL l1
[4.2.9] The whole <theorem> is clear by expository syllogisms, since in

affirmed modes if both premisses are de passibili then a conclusion such
as "what can be B can be A" clearly follows, where 'C' is the middle.
Now <such a conclusion> follows like so:
(427) C can be A, and C can be B; therefore, <B can be A>

(428) What can be C can be A, and what can be C can be B; there-
fore, <what can be B can be A>
It further follows that what can be B can be A; therefore, B can be A.

And when the mode is denied the conclusion "what can be B can not-be
A" follows, from which it follows "B can not be A"
[4.2.10] If both premisses are de necessaria, then when the modes are
affirmed "what necessarily is B necessarily is A" follows; hence B is
necessarily A When the modes are denied "what necessarily is B
necessarily is not A" follows; hence B is necessarily not A.
[4.2.11] But if the major is de necessaria and the minor is de passibili

then in affirmed modes "what can be B necessarily is A" follows; hence

B necessarily is A. When the modes are denied "what can be B neces-
sarily is not A" follows; hence B necessarily is not A.
[4.2.12] But if the major is de possibili and the minor is de necessario

then in affirmed modes "what necessarily is B can be A" follows; hence
B can be A. When the modes are denied "what necessarily is B can not
be A" follows; hence B can not be A.
[4.2.13] Again, the whole <of Theorem IV-6> is proved per impossibile,
since from the minor and the opposite of the conclusion the opposite of
the major is deduced by the first figure.
[4.2.14] Since many people say that the subject in these sentences de
possibili or de necessario is restricted to supposit only for those things
which exist, I put forward some theorems about this.

[Theorem IV-7J When sentences de possibili or de necessario
are restricted by the "what is ..." <locution>, then in the first
figure (a) when the minor is de necessario while the major is
de possibili or de necessario there is a valid syllogism, but (b)
there is no valid <syllogism> if <the minor> is de possibili. L2
[4.2.16] I call it a 'restriction by the "what is ..." <locution>' when a sent-

ence is formed like this:
(429) What is B necessarily is A
(430) What is B can not be A
and it is said that the subjects in such sentences de possibili or de neces-
sario only supposit for those things which exist.
[4.2.17] Therefore, if the minor is de necessario, <Theorem IV-7> will be

clear, taking it under the distribution of the major. For example, if I say
(431) Anything which is B is of necessity A and every C is of ne-
cessity B
I take under 'B' only those things which are <now> B. For anything which
of necessity is B is as a matter of fact B and always will be B, and it

happens that 'B' is distributed for all those things which are B in the
major sentence. From this it is clear that although the major is restricted,
nevertheless if the minor is not restricted then the conclusion can be de-
duced without restriction. But if the minor is restricted then the
conclusion ought to be restricted if a universal is deduced, lest we go
from the more ample distribution to the less ample.
[4.2.18] If the minor is de possibili, however, then it cannot be taken

under the distribution of the major restricted by the "what is ... " <locu-
tion>. Hence there will be no valid syllogism. Indeed, these do not
follow:
(432) Everything which is shining is necessarily different from the

Moon, and every Moon can be shining; therefore, every
Moon can be different from the Moon
(433) Everything which runs can ride, and every horse can run;
therefore, every horse can ride
[Theorem IV-8] In the second figure, (a) when the premisses

are restricted by the "what is ... " <locution>, from a pair of
premisses which are de necessario or from one de necessario
and the other de possibili there follows a conclusion de pos-
sibili or an assertoric <conclusion> restricted by the "what is
..." <locution>, but (b) no conclusion de necessario follows."
[4.2.20] It is clear that a conclusion de necessario does not follow, since

if anything which is a planetary body shining above our hemisphere is
necessarily the Sun, and anything which is a planetary body shining be-
low our hemisphere is necessarily not the Sun, it does not thereby follow
that therefore anything which is a planetary body shining above our
hemisphere is necessarily not a planetary body shining below our hemis-
phere, for the premisses are naturally possible in a posited case and the
conclusion false. Nevertheless, it does follow <from the premisses in
[4.2.20]> that therefore nothing which is a planetary body shining above
our hemisphere is a planetary body shining below our hemisphere.
[4.2.21] As for the rest <of Theorem IV-8>, from this conclusion a con-
clusion de possibili follows, since
(434) Nothing which is B is A; therefore, anything which is B can

notbeA
is an acceptable consequence. Anything which is not A can be not A;

hence, just as the given assertoric conclusion follows, so too does this
<conclusion> de possibili.
[4.2.22] Moreover, it can be proven per impossibile that the assertoric

conclusion follows, since the opposite of the minor follows from the
major and the opposite of the conclusion in a mixture of the de neces-
sario and the assertoric, as will be clear in Theorem IV -12.
[Theorem IV-9] In the third figure, when the premisses are

restricted by the "what is ..." <locution>, if either <premiss> is
de necessario or de possibili there always follows a conclu-
sion in the same mode as the mode of the major. 14
[4.2.24] The entire <theorem> can be known <to be true> through exposi-
tory syllogisms and <by reduction> per impossibile.
[4.2.25] Nevertheless, it should be noted that if the minor is de possibili

the conclusion should not be restricted by the "what is ..." <locution>,
since the minor extreme is not restricted. But if the minor is de neces-
sario, then the conclusion can be deduced as restricted or non-res-
tricted, since (as has been said in [4.2.17]) anything which is of necessity
B is as a matter of fact B. Thus, although this does not follow:
(435) The one creating is necessarily God; therefore the one creat-
ing is God
nevertheless this follows:

(436) What <is creating is necessarily God>; therefore, <the one
creating> is God
[4.2.26] The tenth theorem is:
[Theorem IV-lO] In the first figure, (a) no syllogism with an

assertoric major and a minor de possibili is formally valid;
but (b) from a major de possibili and an assertoric minor
there is a valid syllogism whose conclusion is a particular de
possibili, but not universal. 15
[4.2.27] Theorem IV-IO(a) is clear: if everything running is a horse and

every man can run, then it does not follow that a man can be a horse.
Similarly, if nothing creating is God and any first cause can create, it
does not follow that the first cause can fail to be God.
[4.2.28] Furthermore, it is clear that a particular de possibili conclusion

follows from a major de possibili and an assertoric minor, if the major is
put explicitly. There will be a perfectly evident syllogism taken under
the distribution of the major, as for example
(442) Anything which is or can be B can be A, and C is B; there-

fore, that C can be A
The same applies if the major is negative de possibili. If the major is res-
tricted by the "what is ..." <locution>, the syllogism is no less evident.
[4.2.29] It is also clear that a universal conclusion does not follow: every
Moon can be the closest planetary body and every planetary body shin-
ing above our hemisphere is the Moon (let us assume); but it does not
follow that every planetary body shining above our hemisphere can
therefore be the closest planet. The conclusion is false, since if it is ana-
lyzed by a disjunction of 'is' with 'can' there is a clear counterexample
<in the case of> the Sun, which can be a planetary body shining above
our hemisphere and cannot be the closest planetary body. One can
similarly object to Celarent by putting in the place of the aforesaid
affirmative major the negative "Every Moon can not be the Sun." The
reason why the universal conclusion does not follow is that the minor
extreme is more ampliated in the conclusion than in the minor sentence,
and from the less ample, however much distributed, to the more amply
distributed is not a valid consequence.
[4.2.30) Nevertheless, it should be known that if the minor sentence and

the conclusion are formed with the "what is ..." <locution> so that each
of them prevents ampliation, then a universal conclusion can be de-
duced in both Barbara and Celarent, and they are perfect syllogisms.
But if the minor is not formed with the "what is ..." <locution>, then the
middle ampliative in the affirmed mode can prevent any conclusion res-
tricted by "what is ..." Nor does this follow:
(438) Anything going-to-die can be alive, and the Antichrist is
going-to-die; therefore, he who is the Antichrist can be alive.
But when the mode is denied the conclusion so restricted is not pre-
vented, for if the subject of the conclusion does not presently supposit
for anything, then that existing negative will be no less true, just as the
affirmative would not be true.
[4.2.31) The eleventh theorem is:

[Theorem IV-lI) In the first figure, from an assertoric major
and a minor de possibili there is always a valid syllogism to a
conclusion de possibili, under the hypothesis <that the major
is simply assertoric>, i.e. that the major is necessary.16
[4.2.32) Just as I said in [4.1.3), the same thing is signified by calling an

assertoric sentence necessary and calling it what Aristotle wanted to call
it [Prior Analytics Lxv 34b7-8), namely 'simply assertoric.'
[4.2.33) Proof of Theorem IV-ll: the opposite of the hypothesis or as-

sumption follows from the minor and the opposite of the conclusion; if
we have this sort of syllogism:
(439) 'Every B is A' is necessary and any C can be B; therefore any
Ccan beA
then from the minor and the opposite of the conclusion we get this syl-
logism:
(440) Some C is necessarily not A, and any C can be B; therefore,
some B is necessarily not A
This syllogism is acceptable, by Theorem IV-6(c);17 nevertheless the
conclusion is incompatible with the <necessity of> the major in (439).
That is clear for this reason: the sentence "Some B is necessarily not A"
can have three causes of its truth: (i) because 'A' or 'B' do not supposit
for anything, and so the major of (439), stating that every B is A, is false
and so not necessary; (ii) because something is B and this thing is neces-
sarily not A, and so again the major of (439) is false; (ii) because some-
thing can be B although it is not yet B, and that thing is necessarily not
A, assume that it becomes B, which is why it is possible, but is still not
yet A; then the major of (439) is false, and hence it is not necessary,
although it may perhaps be true.
[4.2.34] Similarly, I say that there is negatively an acceptable syllogism

here:
(441) 'No B is A' is necessary, and every C can be B; therefore,
every C can not-be A
For from the minor and the opposite of the conclusion another conclu-
sion is deduced which is incompatible with the necessity of the major,
just as was the case with the affirmed modes; the conclusion "Some B is
necessarily A" is incompatible with the necessity of "No B is A".

[Theorem IV-12] In the second figure, (a) there is no formally
valid syllogism with one <premiss> de possibili and the other
assertoric; nevertheless, (b) under the hypothesis that the as-
sertoric <premiss> is necessary, there is always a valid syllog-
ism with a conclusion de possibili if the major is assertoric
(and it is not valid if the major is de possibill).l x
[4.2.36] In order to prove that no <syllogisms of the second figure> are

thus formally valid, whether the major is assertoric and the minor de
possibili or conversely, we can give as an objection to all those moods
the terms 'shining,' 'Moon,' and 'closest planetary body,' where 'shining'
is always the middle, and 'Moon' the major extreme and 'closest plane-
tary body' the minor extreme.
[4.2.37] But that <the syllogism> is always valid under the hypothesis
with the major being assertoric is proved first of all for Cesare and
Festino, for by converting the major they are first figure.
[4.2.38J Next, all of the moods are proved together per impossibile. 19 In
Cessare and Festino from the minor and the opposite of the conclusion
there not only follows the conclusion "Some B can be A," but also the
conclusion "Something which of necessity is B can be A," which is clear
from an expository syllogism. For example, if the first syllogism is in Ce-
sare "No B is A (this is necessary)," then "and every C can be A; there-
fore, every C can not be B." Then we get a syllogism ad impossibile like
so: "Every C can be A, and some C is necessarily B"; choose the B
which is C, so that we say 'This C can be A and this same C is necessar-
ily B." It clearly follows from this that "Something which is necessarily B
can be A," and this conclusion is incompatible with the necessity of "No
B is A." Similarly, in Camestres and in Baroeo from the minor and the
opposite of the conclusion we draw the conclusion "Something which of
necessity is B can not be A," which is incompatible with the necessity of
"Every B is A" - form the syllogisms for yourself.
[4.2.39J However, that none of these syllogisms is valid if the minor is

assertoric though necessary is shown by counterexamples, which seem
difficult to find. Nevertheless, let us posit that our world is inside a
sphere such that every star sometimes has a fixed position relative to us
and sometimes <only has> a position relative to us. Then we give the
counterexample to Cesare and Festino: every star is possibly not above
our hemisphere, and every star elevated above our horizon is above our
hemisphere; nevertheless, it does not therefore follow that a star ele-
vated above our horizon can not be a star. The counterexample to Cam-
estres and Baroeo is similar: every star is possibly above our horizon,
and no star existing below our hemisphere is above our hemisphere; it
does not therefore follow that a star existing below our hemisphere can
not be a star.
[4.2.40J The thirteenth theorem is:
[Theorem IV-I3J In the third figure, (a) with the mode af-
firmed there is a valid syllogism with a mixture of assertoric
and de possibili <premisses> to a conclusion de possibili, if the
premiss de possibili is universal (otherwise it is not formally
valid);20 and (b) with the mode denied there is a valid syllog-
ism if the major is universal de possibili, and otherwise not. 21
[4.2.41] Theorem IV -13 is firstly proved for the useful moods: if the
major is universal de possibili, as in Darapti and Fe/apton and Datisi and
Ferison, it is immediately clear by conversion of the minor to the first
figure. Next, with the mode affirmed, if the minor is universal de pos-
sibili, as in Darapti and Disamis, then we get the first figure with a major
de possibili by converting the major and the conclusion, and transposing
the premisses. And so it is clear that all these moods are valid in this
way.
[4.2.42] Theorem IV-13 is proved for the useless moods: if the de pos-
sib iii premiss is particular, then with the mode affirmed there is a coun-
terexample in the terms 'Sun: 'Moon,' and 'planetary body shining
above our hemisphere,' where the major extreme is 'Sun' and the minor
extreme is 'Moon,' with 'planetary body shining above our hemisphere'
as the middle. In every combination of premisses a case can be posited
where the premisses are true and the conclusion is false. Similarly, with
the mode denied, if the de possibili premiss is particular, then there is a
counterexample in these terms: 'Sun: 'brightest planetary body,' and
'planetary body shining above our hemisphere,' where the latter is the
middle.
[4.2.43] Next we show that the denied modes are not valid if the minor
is de possibili, even if it is universal, since nothing creating is God (let us
assume), and everything creating can be the first cause. Then it does not
follow that therefore the first cause can not be God.
[4.2.441 The fourteenth theorem is:

[Theorem IV- J41 In the third figure, there is always a valid
<syllogism> from a mixture of assertoric and de possibili <pre-
misses> to a conclusion de possibili, under the assumption
that the assertoric <premiss> is necessary.""
[4.2.451 Theorem IV-14 is clear for the moods which in Theorem IV-13
were found to be valid without the hypothesis; adding the hypothesis
does not prevent <Theorem IV-13>. But Felapton and Datisi and Fer-
ison, if they have a minor de possibili, are reduced to the first figure by
converting the minor. And in the first figure, the conclusion de possibili
follows from the given hypothesis. Similarly, Disamis with a major de
passibili is reduced to the first figure by converting the major and the
conclusion, and transposing the premisses.
[4.2.46] We prove Bacarda per impassibili: if the major is de passibili,

then the conclusion "Some C is necessarily not 8" follows from the
major and the opposite of the conclusion (by Theorem IV-5), and this
conclusion is incompatible with the necessity of the minor of the first,
which asserts "Every C is B." But if the minor is de passibili, then the
conclusion "Every C is necessarily A" follows from the minor and the
opposite of the conclusion (by Theorem IV -4), and this conclusion is
incompatible with the necessity of the major of the first. which asserts
"Some B is not A."
[4.2.4 7] The fifteenth theorem is:

[Theorem IV-IS] In the first figure, from an assertoric major
and a minor de necessaria there does not follow a conclusion
de necessaria, nor does there even follow an assertoric con-
clusion, except in Celarent. 23
[4.2.48] That no conclusion de necessaria follows is clear, because every

God is creating (let us assume), and any first cause is of necessity God,
but it does not follow that the first cause is of necessity creating. Sim-
ilarly, no God is creating (let us assume), and any first cause is of ne-
cessity God, but it does not follow that the first cause necessarily is not
creating.
[4.2.49] We prove that an assertoric conclusion does not follow in

Barbara or in Darii: posit that God is not now creating; this does not
follow:
(442) Any God is just, and anything creating is necessarily God;
therefore, the one creating is just
since the premisses are true and the conclusion is false in the posited
case.
[4.2.50] It is clear that an assertoric conclusion does not follow in Feria:

let us posit that the Moon is not now shining, and then argue as follows:
no planet is closer than the Moon, and something shining of necessity is
the closest planetary body; but it does not follow that the Moon is not
something shining, since the conclusion is false and the premisses are
true. Even if the Moon were not to shine, nevertheless "Something
shining is of necessity the Moon" <is false>, since something which can be
shining is of necessity the Moon or the closest planetary body.
[4.2.51) We prove that an assertoric conclusion follows in Celarent let

us put forward the syllogism
(443) No B is A, and every C of necessity is B; therefore, no C is A
It happens that "Nothing which is B is A" follows from "No B is A" (by
Theorem 1-12); express the minor explicitly as "Everything which is or
can be C of necessity is B." From this it follows that everything which is
or can be C is B, since everything which of necessity is B is as a matter
of fact B. Then let us put forward the syllogism:
(444) Nothing which is B is A, and everything which is or can be C
is B; therefore, nothing which is or can be C is A
where the conclusion is inferred by the dictum de omni et nullo. Thus
this conclusion follows from the principal <syllogism>.
[4.2.52) Nevertheless, it should be known that if the minor which is de

necessario is restricted by the "what is ..." <locution>, then an assertoric
conclusion follows in all moods of the first figure, because from a minor
de necessario so restricted there follows an assertoric minor also res-
tricted. This follows:
(445) What is B necessarily is A; therefore, what is B is A
And so both premisses would be assertoric <in this case>.
[4.2.53) The sixteenth theorem is:

[Theorem IV-16) In the first figure, (a) from a major de ne-
cessario and an assertoric minor there is always a valid syl-
logism to a particular conclusion de necessario, but (b) not to
a universal <conclusion de necessario>.24
[4.2.54) Theorem IV-16 is proved and analyzed or modified in exactly

the same way as Theorem IV-12(b) was proved and analyzed or modi-
fied. Nevertheless, it should be said that assertoric universal <conclu-

sions> do acceptably follow. 25

[Theorem IV-l7] In the second figure, (a) there is always a
valid syllogism from a negative major de necessaria and an
assertoric minor to a particular conclusion de necessaria, but
not to a universal <conclusion de necessaria>;26 but (b) there
is no valid syllogism if the major is affirmative, de necessaria
or assertoric, to a conclusion de necessaria, although <there is
a valid syllogism> to an assertoric conclusion.27
[4.2.56] Theorem IV-17(a) says that Cesare and Festina are valid with a
conclusion de necessaria which is not universal. The reason for this is
that by converting the major of Cesare we immediately get Ceiarent, and
by converting the major of Festina we get Feria. Hence Theorem IV-17
is proved <in these cases> just as Theorem IV -16 was proved for Cela-
rent and Feria.
[4.2.57] We prove <Theorem IV-17(b», that no mood is valid if the ma-

jor is assertoric: we give a counterexample to Cesare and Festina as fol-
lows. Let us posit that nothing creating is God, and that any first cause is
of necessity God. It does not follow that the first cause is necessarily not
creating. There is a similar counterexample for Camestres and Baraca:
let us posit that every planetary body shining above our hemisphere is
the Sun, and every Moon necessarily is not the Sun. It does not follow
that therefore the Moon is necessarily not a planetary body shining
above our hemisphere.
[4.2.58] If the major is affirmative, then it is similarly shown that <the

syllogism> is not valid - even if <the major> is de necessaria, as in Cam-
estres and Baraca. Any first cause is of necessity God, and nothing
creating is God (let us posit); it does not follow that therefore the one
creating is necessarily not the first cause.
[4.2.59] Moreover, it is clear that the given moods are valid with an as-
sertoric conclusion, because from a sentence de necessaria there always
follows an assertoric <sentence>, except for the case in which there is
verification only for those things which can exist; but this case does not
prevent the truth of a negative assertoric conclusion.

[Theorem IV-IB] In the third figure, (a) there is always a
valid syllogism from a universal major de necessario and an
assertoric minor to a conclusion de necessario, but (b) it is
not valid from an assertoric major to a direct conclusion de
necessario, nor (c) even from a major de necessario if it is
particular. 28
[4.2.61] Theorem IV-18(a) is clear for all the moods through expository
syllogisms, that is, <it is clear fop Darapti, Fe/apton, Datisi, and Ferison.
It can also be proved per impossibile, since the opposite of the major is
inferred from the minor and the opposite of the conclusion (by Theo-
rem IV-IO).
[4.2.62] <Theorem IV-18(b», that <the syllogism> with the major being
assertoric to the conclusion de necessario is not valid is clear, first of all,
for those with the modes affirmed: any God is creating (let us assume),
and any God necessarily is the first cause; but it does not follow that the
first cause is necessarily creating. Nevertheless, an indirect conclusion
de necessario can be deduced, since when the premisses are transposed
the major is de necessario and the conclusion is direct.
[4.2.63] It is similarly clear for those with the modes denied: no God is
creating (let us assume), and any God of necessity is the first cause; but
it does not follow that the first cause is necessarily not creating.
[4.2.64] Likewise, Theorem IV-18(c) is clear: if the major is particular,

even if it is de necessario, <the syllogism> with the conclusion de neces-
sario is not valid. We give a counterexample to Disamis first of all: posit
that the Moon is not now shining, and then this does not follow
(446) Something shining is necessarily the Moon, and everything
shining is different from the Moon; therefore, something
other than the Moon is necessarily the Moon
indeed, it is clear from (446) that no assertoric or de possibili conclu-
sion follows. Nevertheless, if the major is restricted by the "what is ..."

<locution>, the conclusion de necessario follows, which can be seen by an
expository syllogism.
[4.2.65] We give a counterexample to Bocardo: posit that nothing is

shining now except a star, and nevertheless that tomorrow there will be
a shining fire, which is necessarily not a celestial body. Then this does
not follow:
(447) Something shining is necessarily not a celestial body, and
anything shining is a star; therefore, a star necessarily is not a
celestial body
Nevetheless, if the major were restricted by the "what is ..."<locution>,
then a conclusion de necessario would acceptably follow.
[4.2.66] The nineteenth theorem is:

[Theorem IV-J9] In the first and second figure, from one as-
sertoric <premiss> and the other de necessario there follows a
conclusion de necessario under the hypothesis that the asser-
toric <premiss> is necessary.
[4.2.67] Theorem IV-19 is proved for the first figure: if the major is de
necessario and the minor is assertoric, then a conclusion de necessario
follows without the hypothesis (by Theorem IV-16). The hypothesis
does not prevent the given consequence, but rather supports it. It also
holds for universal modes, since it makes it that a universal conclusion
de necessario follows, which would only follow as a particular without
the hypothesis. For example, if every B is of necessity A, and every C is
B (this is necessary), then it follows that every C is of necessity A, since
from the major and the opposite of the conclusion there follows the op-
posite of the hypothesis, like so: since every B is of necessity A, and
some C can not be A, then it follows (by Theorem IV-5) that therefore
some C is necessarily not B, and this is incompatible with the necessity
of the minor of the first, asserting that every C is B. Similarly in Celar-
ent
(448) Every B necessarily is not A, and every C is B [this is neces-
sary]; therefore, every C is necessarily not A
Now (448) follows, since from the major and the opposite of the conclu-
sion there follows, as before, the <conclusion that> some C is necessarily
notB.
[4.2.68] If however the major is assertoric and the minor is de neces-

sario, there is still an acceptable syllogism like so:
(449) Every B is A [this is necessary], and every C is of necessity B;
therefore, every C of necessity is A
From the minor and the opposite of the conclusion there follows not
only the conclusion "Some B can be not A," but there also follows the
conclusion "Something which of necessity is B can be not A," as can be
seen through an expository syllogism; and this conclusion is incompati-
ble with the necessity of "Every B is A." It is similarly apparent"for Ce/-
arent there is an acceptable syllogism like so:
(450) No B is A [this is necessary], and every C of necessity is B;
therefore, every C is necessarily not A
since from the major and the opposite of the conclusion it follows that
"Something which of necessity is B can be A," and this conclusion is in-
compatible with the necessity of the major asserting "No B is A." And
the same can be proved for Darii and Ferio.
[4.2.69] Next, we prove Theorem IV-19 for the second figure. First, if
the major is de necessario and the minor is assertoric, then since Cesare
and Festino are reduced to the first figure by converting the major, <The-
orem IV-19 holds>. Moreover, the universal negative de necessario
converts simply.
[4.2.70] But all four moods <of the second figure> can be proved to-
gether per impossibi/e, for from the major and the opposite of the con-
clusion there always follows a conclusion which is incompatible with the
necessity of the minor.
[4.2.71] Similarly, if the major is assertoric and the minor de necessario,

then Cesare and Festino are reduced as before to the first figure by
converting the major. Camestres is reduced by converting the minor and
the conclusion and transposing the premisses. But Baroco and all of the
aforesaid moods are proved per impossibiie, since from the minor and
the opposite of the conclusion there follows (in the third figure) a con-
clusion which is incompatible with the necessity of the major.
[4.2.72] The twentieth theorem is:

[Theorem IV-20] In the third figure, (a) there is always a
valid syllogism from a major de necessario and an assertoric
minor to a conclusion de necessario under the hypothesis
that the assertoric <premiss> is necessary, but (b) it is not
valid if the major is assertoric.2~
[4.2.73] Theorem IV-20(a) is clear if the major is universal, since a con-

clusion de necessario follows without the hypothesis (as is clear from
Theorem IV -18), and the hypothesis does not prevent the consequence.
But if the major is particular, as in Disamis and Bocardo, the syllogism is
proved per impossibile, since from the major and the opposite of the
conclusion there follows a conclusion which is incompatible with the ne-
cessity of the minor.
[4.2.74] Theorem IV-20(b) is proved by counterexamples. In <cases

where> the modes are affirmed we give the following counterexample:
(451) Every sign of the Zodiac elevated above our horizon is ele-
vated above our hemisphere [this is necessary], and every
sign of the Zodiac elevated above our horizon is of necessity
a sign of the Zodiac; therefore, some sign of the Zodiac of
necessity is elevated above our horizon
which does not follow. Similar counterexamples apply to denied modes:
since no sign of the Zodiac existing below our hemisphere is above our
hemisphere [<this is necessary>], and every sign of the Zodiac existing
below our hemisphere is of necessity a sign of the Zodiac, it neverthe-
less does not follow that therefore some sign of the Zodiac is necessarily
not above our hemisphere.
CHAPTER IV-3. DIVIDED CONTINGENT MODALS
[4.3.1] If we want to next deal with syllogisms made up of sentences

contingent ad utrumlibet, Theorem 1I-7 should be recalled, namely that
every sentence contingent ad utrumlibet is converted in the opposite
quality of the dictum, but never in the opposite quality of the mode, so
that an affirmative is equivalent to a negative and a negative <is equiva-
lent to> an affirmative.
[4.3.2] It should also be recalled that some contingent sentences are

composite and some are divided, just as <sentences> de passibili and de
necessaria, and to this recollection I add that I do not intend to say any
more of composites, but everything I say in the rest is to be understood
for divided <contingents>.
[4.3.3] It should also be recalled that some divided contingents have the
mode affirmed and others have the mode denied, just as with other
modals.
[4.3.4] Now, however, since settling matter about these contingents is

reduced to settling matters about <sentences> de passibili and de neces-
saria, we should first of all say how contingents are related to <sen-
tences> de passibili and de necessaria. It should be known about this that
from any contingent sentence with the mode affirmed there follows a
sentence de passibili with the mode also affirmed, negative as well as
affirmative. From "'B is contingently A" or "B is contingently not A," it
follows that "B is possibly A" and "B is possibly not A" Whence these
are all equivalent:
(452) B contingently is A
(452*) B is contingently not A
(452**) B can be A and can be not A
since affirmative and negative contingents are equipollent.
[4.3.5] Nevertheless, it should be noted that a particular or indefinite

contingent is not equivalent to a conjunctive sentence made up of an af-
firmative and a negative de passibili, unless the second de passibili is
taken with an identificatory relative-term. Although the conjunctive
sentence "Some planetary body can be the Moon and some planetary
body can be not the Moon" is true, nevertheless "Some planetary body
is contingently the Moon" is false.
[4.3.6] Next, with the mode affirmed, from every sentence de necessaria,
whether affirmative or negative, there follows a contingent sentence with
the mode denied. From "B necessarily is A" or "B necessarily is not A"
there follows "B is not contingently A" and "B is not contingently not
A." Whence "No B is contingently A" is equivalent to "Every B neces-
sarily is A or necessarily is not A."
[4.3.7] Nevertheless, we should not believe that a contingent universal

with the mode denied is equivalent to a disjunctive sentence made up
from an affirmative and negative de necessario with the mode affirmed.
For "No planetary body is contingently the Moon" is true, and neverthe-
less "Every planetary body is necessarily the Moon or every planetary
body is necessarily not the Moon" is false.
[4.3.8] Further, it should be mentioned that in a contingent sentence

the subject is ampliated to supposit for things which exist and those
which can exist, and it is not necessary that the subject supposit for
those things which happen to exist. God creates contingently, and yet
nothing is contingently God and contingently creates, since nothing is
contingently God - rather, anything is necessarily God or is necessarily
not God.
[4.3.9] With these matters settled, we give some more theorems.
[4.3.10] The twenty-first theorem is:

[Theorem IV-21] In any syllogism to some given conclusion
which has some premiss de possibiJi with the mode affirmed,
(a) if a contingent <sentence> with the mode affirmed
(whether affirmative or negative) is put in the place of that
premiss, then the same conclusion follows; and (b) from
whatever premisses from which there does not follow a con-
clusion de possibiJi with the mode affirmed there does not
follow a contingent with the mode affirmed.'o
[4.3.11] The whole <theorem> is proved by Theorem 1-4, since a sen-

tence de possibili is consequent to a contingent sentence, as said. Hence
we may also conclude the following:
[Corollary] Any syllogism valid with two contingent nega-
tives is also valid for affirmatives de possibiJi.
[4.3.12] The twenty-second theorem is:

[Theorem IV-22] From whichever premisses there follows a
conclusion de necessario with the mode affirmed, there fol-
lows from the same <premisses> a contingent conclusion with
the mode denied. 3 I
[4.3.13] Theorem IV-22 is proven by Theorem 1-4, just like Theorem

IV-21, since sentences de necessarioare antecedent to such contingents.
[4.3.14] The twenty-third theorem is:

[Theorem I V- 23] In the first and the third figure, with a con-
tingent major (whether the mode is affirmed or denied),
there follows a similar conclusion which is contingent, if the
minor is de n ecessa rio, and which is de possibili, if it is con-
tingent.
[4.3.15] Theorem IV-23 is clear with respect to the first figure by the
dictum de omni et nul/o, as was clear in Theorem IV-4. 32
[4.3.16] <Theorem IV-23> can be clarified for the third figure by exposi-
tory syllogisms and <reduction> per impossibile, as was clear in Theorem
IV-6. 33
[4.3.17] The twenty-fourth theorem is:

[Theorem IV-24] In the first figure, <a syllogism> from a con-
tingent major and an assertoric minor to a particular contin-
gent conclusion is valid, but not to a universal <contingent
conclusion>.34
[4.3.18] Theorem IV-24 is shown just as Theorem IV-10(b) was shown.

That no universal conclusion follows is clear: every man is contingently
riding, and everything running is a man (let us assume); then the univer-
sal conclusion is false. If the major has a denied mode there is a counter-
example: no horse contingently rides, and everything running is a horse
(let us assume); the universal conclusion is false.
[4.3.19] The twenty-fifth theorem is:

[Theorem IV-25] In the third figure, (a) from a universal con-
tingent major and an assertoric minor there follows a sim-
ilarly contingent conclusion, but (b) if the major is particular
the contingent conclusion does not follow. 35
[4.3.20] Proof of Theorem IV-25(a): in every mood of the third figure

having a universal major, if the assertoric minor is converted then we get
the first figure, which is valid, as stated in Theorem IV-24.
[4.3.21] Theorem IV-25(b) is clear: although someone running is con-

tingently riding and and everything running is a horse, nevertheless no
horse is contingently riding. Similarly if the major is taken as negative,
since it is equivalent to an affirmative.
[4.3.22] If however we take the mode as denied then there is a counter-

example: some intelligent being does not contingently create, and every-
thing intelligent is God (let us assume this); it does not therefore follow
that God does not contingently create.
[4.3.23] The twenty-sixth theorem is:

[Theorem IV-26] In the first and the third figure, a syllogism
with a contingent conclusion with the mode affirmed is not
valid unless the major is contingent with the mode affirmed;
nor is it of any avail if the minor is contingent.
[4.3.24] Theorem IV-26 is clear, first of all, for the first figure: although
anything creating is God and can be God and is of necessity God, and
any first cause contingently creates, nevertheless it is not true that the
first cause is contingently God.
[4.3.25] Similarly, there is a counterexample for denied modes: nothing

running is a stone and every horse is contingently running, but it is not
true that a horse is contingently not a stone since it is contingent that it
is a stone.
[4.3.26] Similarly, in the third figure, although any planetary body

lacking light is the Moon and every planetary body lacking light is con-
tingently below our hemisphere, nevertheless it is not true that someth-

ing existing below our hemisphere is contingently the Moon. Similarly if
the major were de necessaria or de possibili, since then the premisses
would be true and the conclusion false. Similarly for denied modes: no
planetary body lacking light is the Sun, and if we take the minor as be-
fore, the conclusion will be false if it is as a contingent with the mode
affirmed.
[4.3.27J The twenty-seventh theorem is:

[Theorem IV-27J No contingent conclusion with the mode af-
firmed follows in the second figure.
[4.3.28J <Proof of Theorem IV-27>: if either premiss is de necessaria

there will be a counterexample in these terms: 'planet; 'Moon; and
'stone,' where 'planet' is the middle. And in the same terms the pre-
misses will be no less true if they were to be assertoric or de possibili,
and the conclusion with the mode affirmed would still be false. And if
both premisses are contingent, then there will still be a counterexample
in these terms: 'running; 'horse; and 'man,' where 'running' is the
middle.
CHAPTER IV-4. SYLLOGISMS WITH REDUPLICATIVE

SENTENCES
[4.4.1 J The twenty-eighth and final theorem:

[Theorem IV-28J (a) The first figure and the third figure are
valid for a reduplicative conclusion if the major is reduplica-
tive, and otherwise not; and (b) the second figure is not valid
for a reduplicative conclusion. 3!>
[4.4.2J It should be noted that there are four principal elements in a re-
duplicative sentence: (i) the principal subject; (ii) the principal predicate;
(iii) the reduplicated term; (iv) the reduplication. For example, in
(453) Man is sensible qua animal
'man' is the principal subject, 'sensible' is the principal predicate, 'ani-
mal' is the reduplicated term, and the word 'qua' is the reduplication.
[4.4.31 It should be known that reduplication in an affirmative sentence

indicates an immediacy or a precisiveness of the principal predicate to
the reduplicated term. Hence the principal predicate and the redupli-
cated term should not be the same, since it is fruitless to indicate the
precisiveness or immediacy of a thing to itself. Thus it is not proper to
say that man is animal qua animal, or that animal is sensible qua sensi-
ble. Nevertheless, sometimes the subject and the reduplicated term are
the same, namely if the principal predicate prescinds from the princi-
pal subject by the reduplication indicating the precisiveness or the im-
mediacy through the reduplication indicating immediacy. And so we say
that man qua man is risible, but that Socrates is not risible qua Socrates
but rather qua man.
[4.4.4] Note that although a reduplicative sentence is sometimes permit-

ted according to immediacy alone, as "man is animal qua man," and
sometimes is allowed according to precisiveness alone, as "triangle qua
triangle has three angles equal to two right angles," nevertheless strictly
a <sentence> is reduplicative in which there are both precisiveness and
immediacy.
[4.4.51 However, sometimes reduplicative words are taken improperly,

namely specificatively and not reduplicatively, and then they indicate
the ratio or the sense in which the sentence is to be understood, which,
without that specification, would not be true literally. For example, if we
say that man is a species in that the term 'man' sup posits materially, and
form in that what is in pure potency is prime matter, and Homer is an
example of a poet. Sentences of this sort are not properly reduplicative,
and we do not understand them to be such.
[4.4.6] But even with these matters noted, there is a doubt whether the
reduplicated term in reduplicative <sentences> is taken on the side of the
subject, so that it is a part of the total subject, or on the side of the pre-
dicate, so that it is a part of the total predicate. It is also not clear how
<reduplicative sentences> convert.
[4.4.7] In reply to [4.4.6]: according to Aristotle, the reduplicated term

should be taken as the major extreme, and in the first figure is predi-
cated as much of the minor extreme as of the middle; hence the redupli-
cated term is part of the total predicate. It is also clear from this <claim>
that if it were part of the subject in the major sentence, then it would be
part of the syllogistic middle, and then it would not appear in the con-
clusion. Hence the natural order ought to be "Man is risible qua man." If
we sometimes say "Man qua man is risible" then this should be under-
stood as if the order of the terms were otherwise <namely, as in the pre-
ceding sentence). In this case, 'man' is the subject and 'risible qua man' is
the predicate, notwithstanding that as propounded 'qua man' is put
before the copula.
[4.4.8] Hence the sentence "Man qua man is risible" is not converted as
'The risible is man qua man," for this is improper (as was said), or even
false and unintelligible, as Aristotle says in Prior Analytics I [cap.xxxvii
49a22]. It should rather be converted as "What is risible qua man is
man." Similarly, "Socrates is sensible qua animal" should be converted
as "Something which is sensible qua animal is Socrates."
[4.4.9] These issues settled, I <re)state Theorem <lV-28). First of all, it is

clear by the dictum de omni <et nullo) that the <syllogisms of) the first
figure are valid when the major is reduplicative, since we may take
something as subsumed under the distributed middle obviously and
directly in the minor sentence. The force is the same whether the minor
is taken with reduplication or without reduplication, since each is suffi-
ciently subsumed under the middle.
[4.4.10] That the third figure is valid with the major being reduplicative
can be shown by expository syllogisms and per impossibile.
[4.4.11] Yet it is clear that the first figure is not valid for a reduplicative
conclusion if the major is not reduplicative, for there should be no redu-
plication in the conclusion unless it is found in one of the premisses, and
if it is found in the minor premiss, where the middle is predicated, then
<the reduplication) is part of the middle, and so is not found in the con-
clusion; hence a conclusion with reduplication should never be deduced
in this case.
[4.4.12] The same is clear for the third figure: if the conclusion is redu-
plicative, then the reduplication should be a part of the predicate, as was
said, and the predicate is the major extreme if the conclusion is direct;
hence the major must be reduplicative. If however we want an indirect
conclusion, then the minor should be reduplicative. For example, this

follows:
(454) Every man is animal, and every man is risible qua man;
therefore, some animal is risible qua man
since we get the direct mood by transposing the premisses. It should
also be allowed that from a reduplicative minor we may conclude as
follows:
(455) Every man is animal, and every man is risible qua man;
therefore, something which is risible qua man is animal
but the conclusion of (455) is not reduplicative, although it has a redu-
plicated subject.
[4.4.131 It is clear that the second figure is not valid for a reduplicative
conclusion: (i) because it does not conclude affirmatively; (ii) because
whichever premiss is put as reduplicative, the reduplication is a part of
the middle, in that in each <premiss> the middle is predicated, and hence
there should never be a reduplication in the conclusion for this reason.
[4.4.141 Nevertheless, it should be noted that every figure and mood in-
ferring a negative assertoric <conclusion> without reduplication can from
the same premisses infer a negative <conclusion> with reduplication. This
is because a negative with reduplication follows from an assertoric nega-
tive without reduplication; for example, these follow:
(456) B is not A; therefore, B is not A qua C
(457) No stone is risible; therefore, no stone is risible qua man
NOTES
NOTES TO BURIDAN'S PHILOSOPHY OF LOGIC
I Unless one identifies Plato as Buridan's Ass: "He [Plato] said that if I am indifferent
and able to go to the left or right, for whatever reason I go to the right by the same rea-
son I go to the left, and conversely; therefore, either I go to each <direction>, which is
impossible, or I go to neither until another determining sufficient cause comes along ..."
(Quaestiones in Metaphysicen Aristotelis [hereafter QM], Paris 1518 [the publication
year is listed as 1588 on the title page, which is incorrect], reprinted by Minerva
G.M.B.H. Frankfurt-am-Main 1964, Book VI q.5, ff.35vb-36ra).
2 The classic biographical work is Edmond Faral [1949], the source of the factual
knowledge of Buridan's life described in the succeeding paragraphs.
3 The edition of Peter of Spain I have used to check against Buridan's text of TS - the
points of comparison are virtually nonexistent - is De Rijk [1972].
4 There is a modem edition of this work (although not a genuinely critical edition) by
T. K. Scott, "Johannes Buridanus: Sophismata," in Grammatica Speculativa Vol. I,
Stuttgart-Bad Cannstatt: Fromann-Holzboog 1977. An English translation is available
in Scott [1966]. The last chapter, Sophismata VIII, has been newly edited and translated
by G. E. Hughes [1982]. Buridan's references to the Sophismata in TS are not consist-
ent: they are future and past, frequently within a short compass - see e.g. TS 3.7.16
(past) and 3.7.23 (future).
5 Faral [1949] p. 496 lists several lost works of Buridan which probably dealt with
logic: De syl/ogismis, De relationibus, and works pertaining to metaphysics as well as
logic.
6 I have not translated Aristotle's original Greek, but rather the Latin version given by
Boethius in Minio-Paluello [1965]; Boethius conflates certain distinctions present in the
original text: see Norman Kretzmann [1974].
7 This suggestive analogy is made by Calvin Normore [1976] p. 13.
8 Note that this formulation does not necessarily involve denying the existence of
objects not in space and time, or immaterial objects - God is an example; nor does it
require the denial of possibilia or future objects. What it does entail is that any such
objects must be particulars.
9 This useful term was coined by Hughes [1982] p. 5; we shall use it to refer to ex-
pressions which are sufficiently similar in relevant respects, although distinct by other
criteria.
10 In the rest of Section 3.2 I shall, for simplicity's sake, only discuss inscriptions; the
same points can and should be made for utterances and individual acts of thought.
II See TC 1.3.8 and TS 1.8.72, in which Buridan states how his logical laws apply to the
Mental correlate of inscriptions, and the discussion of consequences in Section 7.2

below.
12 This point was first made with respect to Ockham, not Buridan, by Trentman [1970],
but it holds generally for fourteenth-century logicians. Trentman does not consider
[1]-[5], but they seem the necessary requirements for a language to be counted as 'ideal.'
327
13 This point is made by Joan Gibson [1976].

14 "Subordination" is a technical term used by Ockham; as used here it expresses (i)
which concept(s) of Mental are immediately signified by an utterance; (ii) which
concept(s) of Mental are immediately signified by the utterance which an inscription
immediately signifies.
15 That is, they are always in personal and never in material supposition; these terms
will be explained in Section 6.2. For this point, see for example Summulae de dialectica
VII.iii.4, where Buridan writes, "Nevertheless, it should be known that there is only
material supposition in the significative utterance, as it seems to me, for the reason that
no term in a Mental sentence supposits materially, but always <supposits> personally, for
we do not use Mental terms conventionally as we use Spoken or Written <terms>: no
Mental expression has several signifcations or acceptations; the passions of the soul are
the same for all, as are those things of which they are the likenesses, as is said in De
interpretatione cap.i [16a3-8, cited in Section 3.1]." The text translated here is edited by
EbbeseninPinborg(1976[p.155.
16 This answer needs to be made more precise, since I might acquire the concept of a
lion by using Simba as a paradigmatic instance while you use Leo as a paradigm. But
that is a difference which depends on you and I having different concepts, whereas the
question we are faced with is about whether a single person can have two such concepts
internally different. The claim that a single individual cannot form two different con-
cepts by choosing two different paradigmatic instances - which he perhaps does not
realize are of the same kind - is a substantive thesis, endorsed by Buridan, in the
philosophy of mind, and relies on a theory of concept-acquisition.
17 Buridan explicitly states these principles in the Summulae de dialectica l.iii.2: "A
Mental sentence consists in the complexion of concepts, hence presupposing simple
concepts in the mind and superadding a complexive concept by which the intellect
affirms or denies one of the concepts of the remainder ... that compJexive concept is
called the copula of the Mental sentence. And so it is clear how the subject and the
predicate of a Spoken sentence immediately signify the Mental subject and predicate,
while this copula 'is' signifies an affirmative complexive concept and the copula 'is not'
signifies a negative complexive concept." This text is edited by Jan Pinborg in Pinborg
(1976]p.87.
IH I have translated Buridan's examples so that these rules apply directly to the trans-
lated version; there are some awkward results, such as "Some B A not is" rendering the
uncommon idiom for negatives described in TC \.8.70, or the rules describing confused
supposition; I beg the reader's indulgence.
IY The argument is alluded to in OM VII q.21 fol.54vb, but the best presentation is
given in his Quaestiones subtilissimae super octo Physicorum libros Aristotelis (hereafter
OSP), Paris 1509, reprinted as Johannes Buridanus: Kommentar Zur Aristotelischen
Physik, Minerva G.M.B.H. Frankfurt-am-Main 1964, Book I q.4 fol.5ra.
20 Obviously, the relevant form of simplicity in question is something like logical sim-
plicity; the concept 'man' is not simple in regard to containing distinguishable physical
parts (legs and arms, for example). Roughly, we may regard all terms appearing on each
category-tree as prima facie candidates for simple concepts. Exactly what makes such a
simple concept "simple" is a very difficult question; we shall have something to say
about this matter below.
21 For example, OM VII q.5 fol.44va: "Some definitions are simply quidditative, which
NOTES 329
precisely indicate what <a thing> is, such that they do not indicate that of which it is or
that from which it is ... There are other definitions expressing the quid nominis, indeed,
often some name involves (imp/ieat) exceedingly many diverse concepts of diverse
things, and a definition expressing the quid nominis ought to designate those diverse
concepts explicitly. Such definitions are fitting for substantial as well as accidental
terms."
22 The distinctions absolute/non-absolute and non-appellative/appellative are often
identified, but this is a substantive semantic thesis and is open to question: in TS 1.4.8
Buridan says that 'white' is appellative, and in TS 2.4.5 he explicitly says that 'white'
corresponds to a simple concept. Therefore some appellative terms are absolute.
23 See TS 1.4.1,5.1.1,5.2.5; Sophismata I Theorem 6. Modern scholars often state the
Remainder Principle as a biconditional, but Buridan neve;. states it in this way - nor
should he, since he countenances 'non-external' appellation (discussed below).
24 Rule App-6 in TS 5.4.1 and the discussion in 5.4.4-5.4.7; such terms, like attribu-
tive adjectives, also appellate in combination with a term in the nominative case (TS
1.4.4-5).
25 Nonsubstantial abstract terms are more difficult; they need not have appellation, and
are categorized by reduction (OM IV q.6 foI.17va).
26Buridan's usage is not regular; sometimes he characterizes terms as connotative,
particularly in his commentaries on Aristotle, and sometimes he calls such indirect signi-
fication appellation, particularly in his independent logical works. This looseness is not
vague; his intention is usually clear in context.
27 There is a problem here: we might take 'rational animal' to be the real definition of
'man,' and hence to correspond to a simple concept. But in OM VII q.21 fol.54vb
Buridan says that definitions by genus and differentia are nominal, not real. There are
two reasons for rejecting this claim: (i) if definition by genus and differentia is not 'real:
then we have no example of real definitions, and indeed may lose our intuitive grasp on
this notion altoghether; (ii) 'man' cannot be synonymous with 'rational animal' by the
Additive Principle.
2H See Calvin Normore [1984J who discusses this text and the metaphysical problems in
detail.
29 The qualification 'apt to' is necessary, because (i)-(iii) may fail in actual practice: the
term may fail to refer; what is connoted may not exist; what is connoted may fail to
inhere, i.e. the disposition be destroyed (OM IV q.9 fol.19ra; TS 5.2.6). This last case is
theologically crucial, since it describes the Eucharist, in which the qualities of the bread
remain without inhering in a subject. In each of these cases the truth-value of the
sentence containing the term is affected.
30 The semantic version of this principle is that 'F-ness' is not synonymous with 'what it
is to be F,' which Buridan defends in OM IV q.6 fol.30vb.
31 By "intentional verbs" I mean verbs which are (i) cognitive or epistemic, such as
'know,' 'understand,' 'believe,' and the like; (ii) verbs of desire, such as 'want,' 'intend'
'hope,' and the like; (iii) promissory-verbs, such as 'owe' or 'promise' and the like. The
fullest list, though Buridan acknowledges its incompleteness, is found in TC 3.4.7; their
characteristics are discussed in Sophismata IV sophisms 7-15, TS 3.8.24-31 and
5.3.1-8, TC 1.6.12-16 and 3.7.3-10, OM IV q.8 fol.19ra and q.14 fol.23va, OSP II
q.12 fol.38va.
32 This is Buridan's grammatical way of drawing scope distinctions: a term appears a
parte post and so in the scopc of the verb, or a parte priori and so outside the scope of
the verb,
JJ Strictly speaking it is incorrect to call these 'opaque' and 'transparent,' for they are
not alternative ways of reading one and the same sentence but rather Buridan's way of
regimenting the difference between the logical form of two different sentences; I shall
use these terms as convenient abbreviations for distinguishing the two classes of
sentences, which should not occasion any confusion.
J4 Note that Buridan is careful only to state the Converse-Entailment Principle for scire:
it clearly fails for most intentional verbs, which allow for intentional inexistence.
J5 Which is exactly what Geach [19721 p.134 calls it.
Jo By the Square of Opposition, negative sentences with empty subject-terms are
automatically true. Hence 'The present King of France is bald" is false, and "Pegasus is
not a winged horse" is true.
J7 Buridan will speak loosely about the 'sentences' which make up a consequence, but
this is a permissible looseness, not a deviation from his doctrine. It is therefore in-
correct, strictly speaking, to represent Buridan's logic by means of the propositional
calculus; it is equally incorrect to metalinguistically quote sentential letters. But as
Buridan often speaks loosely, relying on his general rule that terms have the same
semantic role in equiform sentences (TC 3.7.41), I shall occasionally use such quasi-
technical devices in my exposition.
JK This method for interpreting universal and existential sentences so that existential
import is validated I take from Belnap 119731; it works because 'each' presupposes the
existential import of the subject-term - see TS 3.7.l9 and the discussion of anaphoric
reference in Section 6.4 .
.N Buridan admits a class of sentences which he first calls "quasi-hypothetical," but
immediately says that such sentences are in fact hypothetical, and they are only so-called
because they are very similar to certain categorical sentences (TC 3.7.20).
411 Buridan actually suggests such frames in several places: TC 1.8.62, 2.6.34, 3.4.59, for
example.
41 This may be the key intuition behind the mediaeval view that quantity is irrelevant to
ontological commitment, but quality is not. See Section 6.9 below.
4, Strictly speaking this definition is too loose, for the dictum of "It is possible that a
man is running" is "that a man is running," and so includes an occurrence of the copula
'is.' Buridan's intention is nonetheless clear: the dictum includes only non-primary
occurrences of logical particles. It is less clear how to explain 'non-primary' in a non-
trivial fashion; this is precisely Frege's distinction between 'force' and 'assertion.'
4J In Latin this may be either a subordinate clause introduced by quod or an accusative-
infinitive construction, neither of which is grammatically a sentence.
44 See A. N. Prior 119761 for an excellent and pioneering discussion of the distinction
between the possible and the possibly-true.
45 See OM VI q.! 0 fol.40rb: "It should be known that a conjunctive (sentence> requires
for its truth that any of its categoricals are true, and it suffices for its falsity that one is
false. Conversely for disjunctives: their falsity requires that each part is false and it
suffices for their truth that either part is true."
40 There may be as many conjunctions or disjunctions in a conjunctive or disjunctive
sentence as one likes; they may even be related by DeMorgan's Laws, Because of
associativity, we may treat them as pairwise sentence-functors without loss of generality.
NOTES 331
47 Most mediaeval logicians considered bivalence to be part of the very definition of

'sentence,' so the question did not arise. However, reflections on the problem of future
contingents stimulated discussion of bivalence, leading some (e.g. Peter de Rivo) to give
it up. Buridan, however, in his discussion of future contingents in OM VI q.5, maintains
bivalence.
48 Buridan's arguments for this thesis will be discussed in Section 5.5 and Section 6.9.
49 Buridan devotes q.15 to exploring whether the Law of Non-Contradiction holds for
tensed sentences, a difficult case; they will be discussed in Section 6.8.
50 We should also note here that common terms appearing in sentences are all quanti-
fied, explicitly or implicitly: an unquantified term may be read as involving distributive
or particular quantification, depending on the requirements of the sentence, context, and
good sense: "Every S is P" is subordinated in Mental to either "Every S is every P" or
"Every S is some P" (typically the latter). Indefinite sentences are quadruply ambiguous.
51 For the next series of arguments see Buridan's Quaestiones in Libros Priorum AnaLyt-
icorum I q.5 and OM VI q.8.
52 The claim that supposition theory is in fact a unified theory, and that it should be
construed as I suggest in Section 6, is a matter of controversy. Some of the arguments
are taken up in the following sections. My interpretation has the virtue of making
supposition theory a reasonable philosophical enterprise, as well as fitting in well with
other braches of mediaeval logic, such as fixing coreferentiality by the Doctrine of
Distribution in the theory of the syllogism. The distinction of referential and attributive
uses of the particular sign given below is, it seems to me, well-confirmed by the remarks
made by Buridan in his discussion of determinate and non-distributive confused supposi-
tion. But the reader should note that the interpretation I offer is a matter of dispute.
53 This is surprising, since Buridan holds that the speculative sciences each study a
single Mental term, to which other terms are related as individual men in an army are
related to their leader (OM IV q.3 fo1.14ra and OSP I q.2 foI.3rb-va): such attributions
to a primary element are semantic but not exactly 'literal,' and we would expect a
theoretical account of such uses.
54 This is too loose: the term 'utterance' personally supposits for signs. But it may serve
as a handy initial characterization.
55 Do not worry about whether the examples of material supposition need quotation-
marks; that is one of the jobs performed by the theory of material supposition.
56 Sometimes Buridan uses the mediaeval French word 'Ii' or 'Ly' prefixed to a term: this
however was not a mediaeval quote-functor but rather a metalinguistic comment indi-
cating that the following term has material supposition.
57 Provided that no syntactic ambiguity arises, as would be the case in "p' & 'q": but a
scope convention will clarify such cases.
58 Unless such iteration can be accomplished through anaphoric reference: see the
discussion of relative supposition in the next section.
59 In general the Scholastics did distinguish descriptive knowledge from direct knowl-
edge, rather like Russell's "knowledge by description" and "knowledge by acquaintance."
In OM VII q.20 fol.54rb-va Buridan explores how names actually correspond to
descriptions (conjunctions of properties) in the absence of the thing named. See also
Alan R. Perreian [1972].
60 Buridan considers the division of supposition into absolute/relative as a division of
personal common supposition because (i) the terms 'this' may be either a demonstrative
or a relative pronoun, but when it fails to have discrete SUppOSItIon it is used as a

relative pronoun and so has relative supposition (TS 4.5.1); (ii) the distinguishing
features of relative-terms only show up in personal supposition, since any expression,
conventionally significant or not, may be in material supposition.
o I This is very loose of Buridan: strictly, the relative-term supposits for what its ante-
cedent supposits for - in "Man is a noun and it has three letters" the relative-term 'if
does not have material supposition, but personally supposits for the term 'man' in
material supposition. Perhaps this provides a way of iterating material supposition.
02 This reveals a deep connection between the behavior of quantifiers, the subject and
predicate of a sentence, and attributive combination: taken referentially, an attributive
combination like 'white man' acts to restrict the supposition of 'man' to those who are
white (Rule Res-2 in TS 6.3.1), which are the supposits of the sentence ",Some> men are
white." Perhaps this is why Buridan tries to minimize the difference between the
complexio indistans (attributive combination) and the complexio distans (copulation of
subject and predicate) in QM IV q.14 fol.23va and VI q.6 fol.37vb. This subject is
touched upon in Gabriel Nuchelmans [1973J pp 244-246.1
oj This may be why such supposition was called 'natural,' namely the supposition coin-
cides with the complete signification of the term, to its full or 'natural' extent.
64 John Buridan, Quaestiones in decem libros Ethicorum Aristotelis ad Nicomachllm
(hereafter Q NE), Oxford 1637.
6, In QNE VI q.6 p.501 he says "This sentence 'Man is an animal' or 'Every man is an
animal' is about all 'men> through natural supposition: of whatever and whenever it is
true to say 'man' then it is true to say 'animal,' and so 'Thunder is a sound in the clouds'
is true, by referring singulars to singulars." I take this to be equivalent to the reconstruc-
tion proposed above, and T K. Scott 119651 to be in error when he suggests that such
sentences are disguised conditional statements. They are not, but rather complex
categoricals, as Buridan says.
66 There is a further problem here: such a sentence may have truth-at-all-times, but that
might not seem enough to capture the necessity involved in knowledge and science: do
we not need to require that every possible instance of thunder is equally a sound in the
clouds? Buridan does not mention this. On his behalf we might suggest either (i) natural
supposition includes ampliation to possibles; (ii) there are no never-actual possibles -
but each suggestion obviously has further problems.
67 In Geach 11962J pp. 6-7 the problem is discussed at some length; I can only treat it
briefly here.
6R It is only fair to point out that existential quantification may be in no better shape:
the variables bound by a quantifier range over the entire domain, so that universal and
existential quantifiers equally 'refer' to the entire domain. The mediaeval and modern
cases only appear different; the modern case seems simpler because the work of fixing
the reference of a term is embodied in the interpretation-function, after which determin-
ing truth-value is simple. As we shall see, this is but to embrace one possible solution.
Mediaeval logic will take another path.
69 This is Geach's central claim, and he argues against the Doctrine of Distribution by
means of such puzzles. But Geach is more abusive than conclusive; his objections, as I
shall argue, are ill-founded.
70 This distinction was first noted by Keith Donnellan 119661, who only applied it to de-
NOTES 333
finite descriptions; he was able to do so in virtue of the fact that definite descriptions are
disguised existential quantification - which have an attributive and a referential use.
71 Strictly speaking, this is not true; the same term may appear several times with a sent-
ence. Coreferentiality in such cases is established (or not) anaphorically, by relative sup-
position.
72 Buridan uses these terms only rarely, as in TS 2.6.75-76 and 3.8.22-23, but does
frequently talk about 'descending' from a general sentence to a string of less general
sentences. Despite the fact that it was a standard part of supposition-theory, though,
Buridan does not give any systematic or theoretical account of the doctrine.
73 Buridan argues for this point in TC 2.4.1-7. The nonequivalence mentioned is
established for tensed sentences and for modal sentences. Buridan's argument proceeds
as follows: consider the sentence (S*) "What is B can not-be A or what can be B can
not-be A" - an equivalence Buridan will deny. He argues that when there are no actual
B, the first disjunct of (S*), 'what is B can not-be A,' will be true, since it is a negative
sentence with an empty subject-term. (Equally "unicorns do not have wings" is true for
the same reason.) Since one of the disjuncts of (S*) is true, (S*) is true. Buridan now
claims that (S), however, may be false while (S*) is true, clearly showing the nonequival-
ence of (S) and (S*). He suggests that (S) would be false for the case of a possible (but
non-actual) B which must be A, that is, cannot not-be A. Since (S) ampliates its subject
to stand for possibles, (S) would be false. A simple example will help: suppose that God
destroys the human race, and consider the sentence (S) "A man can not-be rational."
The proposed equivalent (S*) "What is a man can not-be rational or what can be a man
can not-be rational" will be true, since its first disjunct is a negative sentence with an
empty subject-term. But it is part of the nature of man to be rational; it is part of the hu-
man essence. Thus any possible man must be rational, that is, cannot not-be rational,
and so (S) is false while (S*) is true. Buridan opts for understanding (S) by the equival-
ence (S**) "What is or can be B can not-be A," whose truth-conditions are the same as
for (S).
74 Consequences, and hypothetical sentences generally, are assessed for truth-value
with each of their constituent parts relativized to the same temporal interval.
75 Quality is not completely irrelevant for contemporary philosophers; there is no onto-
logical commitment if a negation precedes the existentially-quantified sentence on the
modern view. But "Some unicorns do not have horns" would be represented as (3x) (UX
& - Hx), which does commit one to the existence of unicorns.
76 It makes no difference whether the reference to possible worlds is included in the
sentence (e.g. as an operator) or is part of the semantic apparatus, included in the inter-
pretation function assigning truth-values to sentences; the net result is the same - an
assertoric sentence true in an expanded ontology.
77 I borrow this name from Hughes [1982[ p. 22.
78 Things were not always so: the twelfth-century philosopher and logician Peter
Abailard clearly distinguishes arguments and conditionals, and even argues for a Princi-
ple of Conditionalization and Deconditionalization - a mediaeval "deduction theorem."
But Abailard's work seems to have been completely lost to the fourteenth century, for
unknown reasons. The unpublished studies of Christopher Martin on Abailard and
Boethius on conditionals are invaluable.
79 The problem is exacerbated if Spoken language is taken as primary, for the distinc-
tion is usually not drawn verbally; if drawn at all, it typically relies on non-verbal cues.
Buridan mentions such cues in TS 2.6.79 when distinguishing composite and divided
senses.
xo While the theory of consequences explores some principles about commitment, the
theory of obligationes explores in greater detail which inferential connections obtain
when the sense of a term or a sentence is altered, all else remaining the same - this is
the form known as institutio. See Spade and Stump [1982] for a discussion of obliga-
tiones.
XI Buridan will occasionally call a grammatically consequential sentence a "consequ-
ence," even if it fails to satisfy his complex conditions, but this is no more than a mere
abbreviation, a harmless way of talking e1iminable upon request. I shall also indulge in
this looseness when there is no danger of being misunderstood.
X2 Buridan is careful to argue that the supposition and appellation of terms in the prota-
sis or apodosis is the same as their supposition in the equiform sentences p* and q* (TS
3.7.41). But note that the definition is not stated by talking about the truth of p* and q*;
this is due to complications which arise from taking Liar-sentences into account: what a
sentence signifies to be is the case is a necessary, but not a sufficient, condition for the
truth of the sentence.
XJ In a few places Buridan says "consequentia valet," which I have translated as ·valid.'
X~ In the spirit of Carnap's "meaning-postultes" we could modify our substitution-rules

to preserve relations among terms, as was done by the later Scholastics, but Buridan's
refusal to do so indicates an admirable attempt to sever logic from metaphysics.
X5 Conversions are simple when the subject-term and predicate-term are reversed and
the result is equipollent. E-form and I-form sentences convert simply, e.g. "No S is P"
and "No P is S" are equipollent. Conversions are accidental when the subject-term and
the predicate-term are reversed and the quantity is changed; the result is not equipollent.
Thus an A-form is converted accidentally to an I-form, such as "Every S is P" to "Some
P is S," and an E-form is converted accidentally to an O-form, such as "No S is P" to
"Some P is not S."
H6 The technical names for these conversions are as follows. First, if the subject-term
and predicate-term are not reversed, then (i) if the predicate-term is changed according
to finite and infinite the result is called obversion; (ii) if the subject-term is changed
according to finite and infinite the result is called partial inversion; (iii) if both subject-
term and predicate-term are changed according to finite and infinite the result is called
full inversion. Second, if the subject-term and the predicate-term are reversed, (i) the
obversion of the resulting sentence is called obverted conversion; (ii) the partial inver-
sion of the resulting sentence is called partial contraposition; (iii) the full inversion of the
resulting sentence is called full contraposition. (Do not confuse the latter case with
'contraposition' as applied to consequences themselves.)
X7 The equipollence of necessity and possibility requires divided modals de necessario
to ampliate their subject-terms to stand for possibles, as Buridan points out in Theorem
11-2; this in turn supports his remark in TC 2.6.22 that a de possibili divided modal
follows from a de necessario divided modal, but not conversely.
xx Note that this principle requires strong accessibility among possible worlds: only in
S5, in which every possible world is accessible from every other possible world, is a
claim like this acceptable.
NOTES 335
89 See the discussion of various definitions of the fourth figure and the related con-
troversies in Nicholas Rescher [1966[ Chapter II.
9() Two terms are used collectively if a sentence in which they are the conjoint subject
does not entail the conjunctive sentence in which each conjunct has only one of the
terms as the subject. See further Buridan's discussion of collective terms in TS 3.2.3.
91 We have seen Buridan make the same point in the material mode, for example when
he says that an affirmative sentence "indicates" that its terms supposit for the same
thing(s), as determined by the correspondence truth-conditions listed in Section 6.9.
92 Classically, a syllogism is acceptable if and only if (i) the middle term is distributed
exactly once; (ii) an extreme term is distributed at most once; (iii) if the conclusion is
negative exactly one premiss is negative; (iv) if the conclusion is affirmative neither
premiss is negative.
9.1 The traditional name for each mood indicates the method of reduction in the follow-
ing way: (i) the initial letter of the name indicates which of the four basic first-figure
syllogisms it is reduced to; (ii) the first three vowels characterize the quantity and quality
of each premiss (and all other vowels are ignored); (iii) the letter's' following a vowel
indicates that the premiss characterized by the vowel is to be converted simply, that is,
the terms transposed; (iv) the letter 'p' following a vowel indicates that the premiss
characterized by the vowel is to be converted per accidens, that is, the terms transposed
in the subalternate; (v) the letter 'm' indicates that the premisses are to be transposed;
(vi) the letter 'c' indicates that the reduction is a reductio ad absurdum. For example,
take Camestres, of the second figure. The vowels indicate that the form is "All P is M,
and no S is M; therefore, no S is P." To reduce Camestres we apply (iii) to the minor
and to the conclusion, and by (v) we transpose the premisses, giving us "No M is S, and
all P is M; therefore, no P is S" - a syllogism in Celarent.
94 Buridan discusses the perfection of the assertoric syllogism in TC 3.4.56-57 (with
an aside on ampliation and the 'what is .. .' locution in 3.4.56), the oblique syllogism in
TC 3.7.9, syllogisms including identificatory relative-terms in TC 3.7.23, and the
divided modal syllogism in TC 4.2.4 and 4.2.28.
95 For the sake of convenience, examples of each kind of syllogism Buridan finds
acceptable are given in the notes to each theorem. Some difficulties are also noted.
96 'Useful' is a predicate of a figured conjugation, indicating that a conclusion can
be added to produce a syllogistic consequence; a figured conjugation is in fact what
Aristotle called a "syllogism" (TC 3.4.48). When a conclusion is so added the resulting
syllogism is called a mood.
97 Such syllogisms have four terms rather than three, but Buridan's extended definitions
of syllogistic extremes and middle permit this case.
98 Note that some of these variation syllogisms only have conclusions which are not in
the common idiom for negatives, namely l-AA, I-IE, 1-10, 1-00, 2-II, 2-00, 3-II,
3-JE, 3-10, 3-01.
99 The correspondence is not exact, since the various modal systems are usually distin-
guished by two-way iterations, not the one-way iterations we have derived from modal
syllogistic. Later mediaeval logicians, such as Strode, will argue over two-way iterations
explicitly, but Buridan does not.
100 It is so replaced according to Theorem IV-6, applied to (9) in the third figure with a
direct conclusion.
NOTES TO TREATISE ON SUPPOSITION
I See 11.2.5J: naturally significant utterances are, for example, cries of pain or inter-
jections.
2 'Ad placitum' technically means at (one's) pleasure, without reference to a linguistic
community, and so the translation 'by convention' is inaccurate; but in his Quaestiones
super Perihermenias q.3 Buridan writes: "But if you were to ask about utterances with
are nouns and verbs ad placitum, whether they are at my pleasure or yours (ad placitum
meum vel tuum), <I say' that some nouns and verbs signify the same in the same way to a
single large community: Latin utterances to all Latins and French utterances to all
French. It is not in my power or in yours to give or to change this common signification
of these" (MS Leipzig 1372 fol.50rb, from the Latin text edited by Jan Pinborg in
Pinborg 11976J p. 89). See further [1.2.4J.
] Here Buridan is criticizing Peter of Spain's arrangement of his chapters; Peter has his
treatise on supposition after his treatise on the syllogism, which Buridan finds irrational.
4 Etymologically 'supposition' derives from sub-ponere, "standing under," which was
originally a grammarian's term, used to specify pronomial reference in giving the

(grammatical) subject of a sentence. Buridan is concerned with the logical notion of sup-
position, parting company with the grammarians over empty terms; see also Sophismata
III, first remark.
S That is, does not stand for itself (see [1.2.61); this was ruled out in [0.0.1 J. Such a word
is said to have material supposition.

6 A pars orationis such as noun, verb, preposition, and the like.
7 In [1.2.3J-I1.2.4J Buridan asks whether the concept of the one speaking or the one
hearing determines the signification of an utterance; his answer is that in most cases
these coincide and deviant cases need not be explained by his theory: an utterance
"gives rise to a concept," i.e. has causal force, only through habit, which would not have
been established if the attempts at communication had not generally been successful in
the past. This is a causal theory, not an ideal-case analysis, which would hardly fit
Buridan's nominalism. The emphasis on communication should not be overemphasized,
though; Mental is a private language. In [1.2.4J Buridan lists several impediments to his
theory, namely when words are not being used for communication, such as singing in the
shower for the pleasure of hearing one's voice.
H By this last sentence I take Buridan to say that the sentence "The term 'bu' is running"
is grammatically well-formed, that is, the sentence" 'Bu' is running" is an oratio congrua,
although nonsense.
9 That is, personal supposition, where a term is taken to stand for (at least some of) the
things it signifies.
10 This does not mean a term in the category of Relation, but a term which is a gram-
matical relative, such as a pronoun: see TC Chapter IV for Buridan's analysis of such
relative-terms.
336
NOTES 337
II Strictly this "new criterion" is inadequate to account for personal supposition; a term
may personally supposit for past or future items currently nonexistent, as well as for
merely possible items - see the Introduction, Section 6.
12 Buridan uses 'verification' to refer to the causes of the truth or falsity of a given
sentence, discussed in greater detail (though without the name 'verification') in TC

1.2.2-1.2.10 and QM VI q.8. Here [1.3.31-[1.3.71 explains how only a Mental sentence
is true or false, strictly speaking, and [1.3.71-11.3.91 emphasizes the point by underlining
how the grammatical form of an utterance or inscription may not accurately reflect the
structure of the expression to which it is subordinated in Mental.
J] Reading qualibet with the manuscripts here instead of Reina's emendation Quaelibet
istarum repraesentatur vera vel falsa.

14 In TC 1.8.72 Buridan says that the logical form of a categorical sentence includes the
subject, the copula, and the predicate; see the Introduction, Section 5.2.
15 This is a version of the Remainder Principle; several points about appellation are
discussed in Section 4.2 of the Introduction.
16 Buridan is here speaking of personal supposition, and his point is that terms not in
the nominative case fail the test for personal supposition laid down in [1.2.91; oblique
terms are treated at length in Rule Sup-3 of [2.6.11 in the discussion [2.6.51-[2.6.481; see
also TC 1.8.92.
17 Buridan's point is that 'rational' is part of the essence of man, and so nothing 'exter-
nal' to man is involved in the expression 'rational animal'; there is no connotation of a
. distinct disposition. See the discussion of this point in Section 4.2 of the Introduction.
18 The phase 'taken per se' indicates that the expression is in personal supposition and
not material supposition, i.e. is taken significatively and not mateially, with the significa-
tion conventionally given to it: see [2.2.4[.
19 Buridan's discussion is as follows: in 12.2.5[ he explains why albus and niger cannot
be subject or predicate, and in [2.2.61-[2.2.151 he argues that quantifiers (signs of

quantity) cannot be subject or predicate but instead are "conditions" of the whole
sentence. In [2.2.51 Buridan argues that a non-substantive adjective such as albus must
be understood attributively, and so a sentence such as "Albus est animal" is not well-
formed, although "Albus Socrates est animal" and "Album est animal" are.
20 Buridan reiterates this claim in Rule Sup-5 in [2.6.501; see also TC 1.8.86 and TC
3.5.2.
21 Reading substantivo, which is clearly required by the context, rather than Reina's
adiectiva possunt apponi ad designandum quid subiaceat substantiae vel subiecto ...
22 The pattern of Buridan's aregument is as follows: in 12.2.61-[2.2.7] he argues that
negative signs of quantity imply negations, and if the copula is within their scope the
implied negation affects the copula; the negative sign should thus be taken as part of the
copula. If the copula is not within their scope they act as infinitizing negations and not as
signs of quantity at all. In [2.2.81-[2.2.15] Buridan examines affirmative signs of
quantity, [2.2.81-[2.2.91 dealing with affirmative distributive signs (universal quantifiers)
and [2.2.10)-[2.2.14) with affirmative particular signs (existential quantifiers). See
Section 4.1 of the Introduction.
23 In the sentence "Every man is an animal" the predicate '(an) animal' has non-distribu-
tive confused supposition; see Rule NDC-l in [3.8.1). Thus the quantifier acts on the
copula and so the entire sentence.
24 This is the second of the "many" answers which may be given (12.2.111), and is the
answer Buridan endorses; 12.2.12] is rejected.
25 Buridan's argument is that the particular sign is either a vacuous quantification
([2.2.131) or a condition of the whole sentence when taken with the subject ([2.2.141).
The only case not covered is an affirmative distributive sign taken with the predicate,
which apparently affects only the predicate, and so is part of the predicate ([2.2.91).
26 Buridan only raises this distinction to ignore it; he classifies terms as categorematic
by signification, not predication.
27 Syncategorematic terms are discussed in Section 4.1 of the Introduction.
2M The point of [2.3.7]-[2.3.9] is that pure syncategoremata lack an ultimate significa-
tion; the list here is surely incomplete - see TC 1.7.3 for a more complete list of
syncategoremata.
2Y This is Buridan's Additive Principle: the signification of a sentence is the sum of the
signification of its categorematic parts. See Section 4.2 of the Introduction.
30 The immediate signification of a (purely) syncategorematic term is a "complexive
concept," i.e. a concept which combines other concepts: they are semantic functors, as
argued in the Introduction. Buridan gives examples for the copula ([2.3.121) and other
particles ([2.3.13]).
31 Note that Buridan is careful here to distinguish the affirmative copula and the nega-
tive copula: they correspond to distinct complexive concepts in Mental.
32 A 'first operation' of the intellect is simply thinking; a 'second operation' is thinking
about thinking - a move to a higher-order concept.
D Buridan has a long discussion of the distinction between complex and incomplex
concepts in QSP I q.4 (foI.5ra-va), in which he argues that (i) simple or incomplex
concepts correspond to indefinable substantial terms; (ii) there must be such simple
concepts, on pain of infinite regress; (iii) such concept are only acquired through
experience, by a process of abstraction; (iv) terms with a nominal definition correspond
to complex concepts; (v) one can replace a complex concept by a simple concept only
through experience, if at all. Roughly, incomplex concepts correspond to the basic meta-
physical simples: concrete and abstract terms in the category of Substance and certain
abstract terms in the category of Quality. The term 'man' corresponds to a simple con-
cept, even though the concept of a man includes 'parts' such as feet, hands, the ability to
reason, and the like. I am not sure there is any non-trivial way to specify what makes a
concept 'simple' for Buridan; see Section 4.3. of the Introduction.
34 The inclusion of 'white' on this list seems like a slip, for 'white' supposits for (white)
substances and appellates whiteness as inhering in those substances: but see the discus-
sion in Section 4.3 of the Introduction.
35 This was not said in 12.1.1], but the point is clear: Buridan wants to exclude non-
declarative sentences from consideration.
36 Only Rule Sup-3 and Rule Sup-9 are discussed at any length; the rest are simply
explained.
37 Rule Sup-3 is discussed in [2.6.5]-[2.6.48]; the order of the discussion is as follows:
in 12.6.5]-12.6.7] the trditional view, that an oblique term can be the subject of a
sentence, is stated; four arguments are given in support. The first argument is stated at
12.6.8] and is refuted by Buridan in 12.6.36]-12.6.39]; the second argument is stated at
[2.6.9]-[2.6.11] and refuted in 12.6.40]-12.6.42]; the third argument is stated at
12.6.12]-12.6.17] and refuted in 12.6.43]-12.6.46]; the fourth argument is stated at
NOTES 339
[2.6.18[ and refuted in [2.6.47[. Buridan's own view, that an oblique term cannot be the
subject of a sentence, is put forward in [2.6.19[ and argued for in [2.6.20[-[2.6.34[ .
.1H The example is difficult to render in English because 'equum homo videni perspicu-
ously illustrates (i) the grammatical order, relevant for Buridan's practice of determining
the subject grammatically if possible; (ii) the accusative (oblique) case of 'every horse,'
obscured in translation. This difficulty persists throughout the third argument. For
example, in [2.6.17[ the suggestion is that the whole 'equum homo' is the subject, a
combination of accusative and nominative terms, which cannot be well-represented by
translation. I beg the reader's indulgence and more care than usual to grammatical cases
while reading this argument.
_W Question 6 of Buridan's Quaestiones super librum Perihermenias is "whether in the
sentence 'Of-any-given-man the ass is running' an oblique term, or a nominative term, or
the aggregate of the two is the subject" (Leipzig MS lat.1372 fol.51 rb; see Pinborg
[1976[ p. 90).
40 This is the subject of Theorem 1-14 in TC 1.8.69 and the subsequent discussion.
41 Again, in this example 'Brunellus' is accusative and hence the object of the verb; so
too for 'chimaera' in the next sentence; this holds for the next few paragraphs.
42 Buridan's point here is obscure; the reference is to the fifth treatise (as in [2.6.39[).
4.1 Buridan is correct. Syllogisms with oblique terms ('oblique syllogisms') are investi-
gated by Buridan in TC 3.6.1-3.7.30; see Section 8.4 of the Introduction.
44 Rule Sup-9 is discussed in [2.6.54[-[2.6.81[; the discussion is as follows: in [2.6.54[
the traditional view that only the first part of a conjunctive or disjunctive categorical is
the subject of the sentence; each of the next three paragraphs puts forward a separate
argument for the traditional view. In [2.6.58[ Buridan rejects the traditional view, and
each of the succeeding four paragraphs puts forward an argument supporting Buridan's
position. To support his view Buridan investigates conjunctive subjects in [2.6.64)-
[2.6.67) and disjunctive subjects in [2.6.69[-[2.6.79[, with a digression on composite and
divided senses in [2.6.72[-[2.6.74[.
45 The terms 'where; 'such; 'what; and the like are what Buridan calls categorial
relative-terms: they are discussed in Chapter IV-8; see the Introduction, Section 4.1.
46 The difference between the two cases is whether the negation is sentence-negation
(negating) or term-negation (infinitizing).
47 Buridan's point is clear - the predicate must match the subject in number -
although his way of putting the point is extremely misleading. Socrates and Plato are not
a being; they are (two) beings. The discussion in [2.6.88[ depends on a play between
infinitizing and negating interpretations of est non ens.
4H Supposition, the mediaeval theory of reference, is discussed at length in Section 6 of
the Introduction; see especially Section 6.1 for a survey of the different forms of suppo-
sition and a brief discussion of 'improper' supposition, which Buridan mentions here
only in passing.
49 Reading causis for casis here, as required by context.
50 The distinction of personal and material supposition is discussed in Section 6.2 of
the Introduction. Buridan here rejects simple supposition as a third form of supposition,
whether intepreted realistically, as by e.g. Walter Bureigh in his De puritate artis logicae
([3.2.4J-[3.2.5[), or non-realistically, as by e.g. William of Ockham in his Summa logicae
([3.2.6[).
51 This is Buridan's nominalism: the refusal to countenance any metaphysically real
shared entities. It is not defended here, but merely expounded.

52 This point is why material supposition is not merely a fragment of quotation theory:
quotation marks form a new term which is the name of the old term, but material
supposition (i) does not create a new term; (ii) may not name the term itself but
something "similar," as Buridan says in the next sentence.
5J Buridan's point here is that the whole phrase 'that man is a stone' is not a sentence,
but an infinitival phrase (hominem esse lapidem), and so cannot supposit for itself.
54 Buridan raises six doubts and resolves them: (i) in [3.2.151 he asks which sort of
supposition a term ought to have in a given sentence; in each of [3.2.171-[3.2.181 a
view is put forward, and in [3.2.191-[3.2.221 Buridan states and defends his own view;
(ii) the second doubt is stated and then resolved in [3.2.241-[3.2.251; (iii) the third doubt
is stated and then resolved in [3.2.261-[3.2.271; (iv) the fourth doubt is stated and then
resolved in [3.2.281-[3.2.291; (v) the fifth doubt is stated and then resolved in
[3.2.301-[3.2.311; (vi) the final doubt is raised in [3.2.321 and discussed in [3.2.331-
[3.2.381.
55 The phrase 'de virtute sermonis is very difficult to translate; literally, it means
something like 'by the power of the word,' and perhaps the closest idiom in English is
'by the meaning of the terms.' But this is not very close, and is misleading in a technical
treatise, for 'virtus' is not meaning, but 'power' or 'force' (see [3.2.211 for Buridan's
discussion of this point). However, Buridan's worry is clear: where a phrase might be in
either personal or mateial supposition, how do we decide which sort of supposition it
has?
56 Obligationes were a mediaeval logical practice the exact nature and purpose of which
is disputed, although the characteristic Buridan here alludes to - asking which in-
ferences remain acceptable if a term is given a new stipulated meaning within an
obligational dispute - is known to be the kind of obligation called 'institutio.'
Obligations have nothing to do with moral commitment; they are rather what one is
'obliged' to assent to on the basis of sentences previously assented to; see Spade and
Stump [19821.
57 See Section 4.4 of the Introduction for a discussion of the 'ratio' behind utterances or
imposition of a term.
58 Reading voco here for Reina's video, which makes no sense: quorum terminorum
genus universale video?
59 The examples are difficult because they are purely Aristotelian. The 'primary object'
of a faculty is that object which of itself perfects the faculty. Thus the faculty of vision
has as its object the visible, and what is paradigmatically visible is color; hence "color is
the primary object of vision." This is informative, since several faculties may have the
same object, though accidentally; vision and touch, for example, both perceive shape (or
any of the common sensibles). Buridan's discussion quibbles on the sense of 'primary'
meant here, ambiguous in Latin as the term 'fir!it' is in English: something may be
logically primary or prior; it may simply be principal, though not logically prior; it may
simply be temporally first (which is the point of [3.2.341). Such problematic sentences
seem to involve a mode of distribution not readily accounted for.
60 Buridan is alluding to Aristotle's modes of per se predication in Posterior Analytics
I.iv 73a35ff.: a sentence is true per se and primarily when the subject is included in the
definition of the predicate.
NOTES 341
61 This sense of 'absolute term' is only as opposed to 'relative term,' and has nothing to
do with absolute terms as opposed to appellative terms.

62 See Section 6.3 of the Introduction for a discussion of discrete and common
supposition.
63 Natural supposition is discussed at length in Section 6.5 of the Introduction.
64 Buridan is here suggesting that there may be a copula in Mental which is timeless
(though whether omnitemporal or atemporal he does not say); this suggestion was first
made in [2.4.3[, and is discussed in [3.4.8[-[3.4.12[.
65 The following example is discussed by Buridan in greater detail in QM IV q.9
fol.l9vb.
66 In QM IV q.9 fol.l9vb Buridan argues at length that F and to-be-F are the same in a
thing, though the concepts of each are imposed by different rationes, in that the term 'F'
is imposed according to a concept which connotes time, while 'to-be-F' is imposed by a
simple concept free from the connotation of time. Hence temporality marks the distinc-
tion between essence and existence.
67 The distinction of confused and determinate supposition is discussed at length in
6X Distributive and non-distributive supposition are discussed at length in Section 6.7 of
the Introduction, which covers the material Buridan presents in TS Chapter 3.6-8.
69 This is the Rule of Existential Import: from the universal to infer the particular.
7(1 Rule DC-l is given in [3.7.1[; Rule DC-2 is given in [3.7.34[; Rule DC-3 is given in
[3.7.42]; Rule DC-4 is given in [3.7.45]; Rule DC-5 is given in [3.7.50]. The rules are
systematized in Section 6.7 of the Introduction.
71 The discussion of Rule DC-l covers [3.7.3]-[3.7.33], and is devoted to exploring the
many forms of universal affirmative signs; in [3.7.6] Buridan introduces the notion of
categorial quantifiers, discussed in Section 4.1 of the Introduction.
72 Reading substantivum in Reina's sive subiectum praecedat adiectivum sive sequatur,
to agree with the earlier part of the sentence complexum ex nomine substantivo et
adiectivo.
73 See Buridan's earlier discussion of Rule Sup-3 in [2.6.5]-[2.6.48].
74 The example is obscured by the fact that 'homo totus and 'totus homo' have no ready
English equivalent, but Buridan's point is clear enough: the latter corresponds to a
distributive sign, the former to a particular sign.
75 Buridan does not use the terminology 'mobile' and 'immobile' supposition, but the
context makes it clear: distributive supposition is said to be immobilized if the descent of
the singulars is prevented or frozen.
76 The sentence 'Sortes est homine fortior is practically untranslateable so that
Buridan's grammatical strictures about scope apply. Again, the reader should take care
to note the grammatical case of the terms involved, since they are crucial to Rule DC-4
though not readily apparent in English.
77 Rule NDC-l is given in [3.8.1]; Rule NDC-2 is given in [3.8.13]; Rule NDC-3 is
given in [3.8.19]; Rule NDC-4 is given in [3.8.24]. These rules are systematized in
78 Note again that 'seeing any ass' is oblique, in the accusative case, and the grammatical
scope-rules are being questioned; bear this in mind in the succeeding paragraphs.
79 The nature of such intentional verbs are discussed at length in Section 4.3 of the
Introduction; see this for more particular discussion of the points Buridan makes here.
xo Reading non for modo here; the abbreviations are similar, and the former seems
required by context while the latter is nonsense: Reina reads possum in mente habere
genera/em conceptum triangu/i modo distributivo.
HI Buridan's point is that intentional verbs operate by mediating concepts (see TC
1.6.12), but most verbs do not - their action goes immediately into their objects. We
know and desire only through concepts, but fire heats water directly and stones lie
directly on the ground, not through mediating concepts.
K2 Chapter IV treats what we now call 'anaphoric reference·; this is discussed in Section
6.4 of the Introduction.
X3 The problem is that ·it' does not have two syllables itself; equally, 'mirror' is not
similar to 'it' such that we can say 'it' materially supposits for 'mirror: Buridan's solution
is to say that such similarity is not required for material supposition.
X4 Appellation is discussed in Sections 4.2 and 4.3 of the Introduction.
" See the parallel discussion in TC 1.6.18-20.

K6 Intentional verbs are treated at length in Section 4.3 of the Introduction; see the
discussion there for many of the points Buridan here raises. For example, the Converse-
Entailment Principle debated in 15.3.71-15.3.131 is analyzed.
" Ampliation is discussed at length in Section 6.8 of the Introduction; restriction is
unproblematic.
xx Omitting terminus in Reina's reading si We terminus 'equid qui est praesens etiam
foit in praeterito et foerit a/bus . .. : the sentence makes no sense otherwise.
NOTES TO TREATISE ON CONSEQUENCES
BOOKI
I See the discussion of truth-conditions in the Introduction, Section 5.5. Buridan's point
here is that the formula of [1.1.3] requires one to treat the present tense of the verb
seriously; as he states in [1.1.6], for different kinds of sentence (affirmative/negative;
past/present/future; assertoric/modal) there are different truth-conditions, not reducible
to a single formula.
2 This is a semantic form of the Law of Non-Contradiction; see the Introduction,
Section 5.4
3 This is the Law of Bivalence .
.j Equally (though Buridan does not say so), any affirmative sentence has, or can have,
some negative which is contradictory to it. The clause "or can have" is part of this
inscriptionalism: it is possible to form the contradictory of any given sentence.
S The clause "hy any power" includes divine power. Notice here how Buridan gives a
syntactic form of the Law of Non-Contradiction (sometimes called the Law of the
Excluded Middle) based on the semantic principles given above.
6 That is, Buridan will use the literally incorrect expression as a handy abbreviation; it
is not to be taken literally.
7 Sentences are 'equiform' (similis) if they are sufficiently similar in the relevant
respects; which respects are relevant, and the degree of similarity required, will depend
on context. Equiformity is how Buridan, as a good nominalist, will group inscriptions
and utterances into classes; see the Introduction, Section 3.2. In this chapter all that is
required for equiformity is that two inscriptions or utterances have the same logical
form (see examples (5) and (6) in [1.2.5]).
x A common term is "distributed" if it is used to refer to everything to which it can
refer; see TS Chapter III for Buridan's discussion, and Sections 6.1 and 6.7 of the
Introduction.
" Buridan defines the 'form of a sentence' in Chapter 1-7 below; roughly, two sentences
have the same form if one may be changed into the other by altering only non-logical
constants, that is, the categorematic terms.
10 See TS Chapter III for confused supposition and determinate supposition; roughly, a
term has determinate supposition when it is used to refer to at least one of the things to
which it may refer, and a term has confused supposition when it is used to refer to
several of the things to which it may refer. These are discussed in the Introduction,
Sections 6.5-7.
11 For the division of sentences into categorical and hypothetical, see the Introduction,
Section 5.2. Buridan discusses the definition of 'consequence' in [1.3.4/-/1.3.11]; see the
Introduction, Section 7.2.
12 In [1.3.4]-[1.3.6/ Buridan calls consequences 'true' (vera), a usage he immediately
343
drops in favor of 'acceptable' (bona) in [1.3.61 and thereafter; see the Introduction, Sec-
tions 5.3 and 7.2.
tJ Note how careful Buridan is to refer to the parts of the (putative) consequence as
"the first <sentence>" and "the second <sentence>": (i) the parts of a sentence are not
themselves sentences, strictly speaking, but only equiform to sentences; (ii) as noted
in [1.3.7) they are correctly called the 'antecedent' and 'consequent' only if the
consequence is acceptable. The problem raised in 11.3.81 is about whether the given
inscription or utterance actually exists.
14 Buridan rejects [1.3.91 for reasons connected with his distinction of the possible and
the possibly-true; see [1.8.26) and the Introduction, Section 5.2.
IS See the Introduction, Section 7.2, for an exact formulation of the definition of "con-
sequence" given here.

10 Consequences are thus formal if they satisfy the Uniform Suibstitution Principle: the
consequence is acceptable no matter what terms may be uniformly substituted for its
categorematic terms. See Buridan's discussion of the 'form' of a sentence in Chapter 1-7
and the Introduction, Section 7.2. (See e.g. [1.8.351 for an example.)
17 The point of (15) is that the genus-species relationship between 'animal' and 'man' in
(14) need not be preserved by substitution.

IX Note that simple material consequences do not satisfy the Unform Substitution
Principle, although they are necessary, whereas we, following Bolzano, often define a
necessary truth as one which satisfies the Uniform Substitution Principle. See the Intro-
duction, Section 7.2.
14 See the Introduction, Section 7.2, for a discussion of consequences ut nunc.
20 It is not clear whether 'one such man; who is deceitful ut nunc, is Cardinal White;
Buridan's discussion is extremely compressed. Perhaps this was a stock figure in
examples, or even a person known to Buridan's students. But his point is clear.
21 A better characterization might have been 'non-necessary consequences: in view of
what Buridan says here; witness his impatience in [1.8.71, where he calls some conse-
quences 'ut nunc pro tunc.'
22 That is: an E-form sentence such as "No A is B" is true when 'A' does not supposit
for anything which is B; an O-form sentence such as "Some A is not B" is true when 'A'
supposits for at least one thing which is not B.
23 Affirmative sentences with empty subject-terms are all false, as Buridan says in
11.5.31-[1.5.4]; negative sentences with empty subject-terms, conversely, are true,
according to the Square of Opposition.
24 Buridan mentions the Liar in 11.5.5]-[1.5.71 only to dismiss it; it shows that his truth-
conditions are necessary but not sufficient, a fact he explicitly ignores in this treatise
([ 1.5.7)), The Liar and other insolubilia are discussed at length in Sophismata VIII.
2S By 'consecutively or implicitly' Buridan means that the sentence does not have as a
consequence or implicitly that it be false, as can happen with referential chains (such as
when Socrates says "What Plato is saying is false" and Plato says "What Socrates is
saying is true"). See also [2.7.43).
26 Buridan explains the locution 'the predicate appellates its form' in TS 5.2.5.
27 Such verbs are discussed in the Introduction, Section 4.3.

2H This sentence suggests that Buridan admits never-actual possibles into his ontology.
See the discussion of ontological commitment in the Introduction, Section 6.8.
29 See the parallel discussion in TS 5.2.10-12.
NOTES 345
30 See the Introduction, Section 4.1, for a discussion of categorematic and syncategore-
matic terms.
31 Strictly speaking Buridan offers two principles governing pairs of contradictory sen-
tences, and two principles governing sentences in general: "any sentence is true or false"
is the Law of Bivalence, and "it is impossible for the same sentence to be true and false
at the same time" is the Law of Noncontradiction. See Section 5.4 of the Introduction.
32 Theorem 1-3 describes the Contraposition Rule for consequences.
33 Theorem 1-4 describes the Rule of Transitivity for consequences.
34 Theorem 1-5(c) may appear surprising at first, but it is perfectly acceptable when we
recall that Buridan is here talking about the modal quality of an assertoric sentence, and
by the definition of 'consequence' the non-necessary cannot follow from the necessary,
as argued in [1.8.24] (although the non-necessary will follow as a consequence ut nunc).
35 The distinction drawn here, reflected in (80)-(81), is the distinction between the
possible and the possibly-true: see [2.7.2] and the Introduction, Section 5.2.
36 That is, what fails is not possibility but compossibility.
37 Such consequences will be formal, that is, satisfy the Uniform Substitution Principle
of [1.4.2], provided that the definition of 'contradictory' is formal: as we should say,
"syntactic." The issue Buridan discusses in [1.8.34]-[1.8.39] is which semantic properties
are relevant to syntactic form.
3H This is what Buridan will call an "infinite syllogism": see [3.1.14]. The rule it validates
is disjunction-elimination.
39 The antecedent of (87) might be thought to be impossible, since 'father' is a relational
term and nothing is a father unless something is its son or daughter. Buridan's point in
(88) is that the semantic property of being a relational term is irrelevant to legitimate
substitution.
40 'Something exists' is necessary, and 'nothing exists' (the antecedent of (93» is impos-
sible, because God necessarily exists.
41 Theorem 1-8 is one of Buridan's central theorems: it establishes the basic principle
for all equipollences and conversions, as Buridan himself remarks in [1.8.47]. Indeed,
Buridan remarks that Theorem 1-9 is just a pedagogical restatement of Theorem 1-8, as
are the next six theorems ([1.8.52]). A version of Theorem 1-8, revised and simplified
in accord with [1.8.43], is given in Section 7.3 ofthe Introduction.
42 The principle used to establish these claims is the symmetry of identity-statements
(the asymmetry of distinction-statements); Buridan is here speaking in the material
mode corresponding to affirmative and negative sentences.
43 See the discussion of distributive confused supposition in Section 6.7 of the Intro-
duction.
401 See the discussion of non-distributive ("merely confused" or "only confused") suppo-
sition in Section 6.7 of the Introduction.
45 This is Buridan's first use of the "uncommon idiom" for negatives, which is important
in his later theory of syllogistic. The uncommon idiom for negatives is discussed in
Section 8.4 of the Introduction; although Buridan is nowhere very explicit about it, such
locutions should be read as branching quantification: "Some B A not is" should be read
as "Some B is not some A (or other)," i.e. "This B is not that A,"
46 The discussion of Theorem 1-14 is indeed "long and difficult," covering
[1.8.72]-[1.8.95], including several digressions. It may help to have a general outline of
Buridan's discussion. In [1.8.72], Buridan distinguishes four kinds of assertoric sentence:
(1) sentences in which the subject-term, copula, and predicate-term are explicit and
distinct; (2) sentences in which the subject-term is combined with another term in which
the copula and predicate are implicit; (3) sentences in which the subject and copula are
implicit in a term and combined with the predicate-term; (4) sentences in which the
subject, copula, and predicate are all implicit in a single term. (Buridan never considers
the case in which the subject and predicate are implicit in a term and combined with the
copula; it is not clear that such cases are possible.)
Buridan discusses (1) in [1.8.73]-[1.8.80], distinguishing (1) further in [1.8.73] into
cases in which (Ia) the copula is present-time, (Ib) past-time, or (Ic) future-time. Now
(Ia) is distinguished into cases in which neither the subject-term nor the predicate-term
is ampliated ([1.8.74]), in which one or the other is ampliated ([1.8.75]-[1.8.77]) and
hence must be coverted by the 'what is .. .' locution, or in which both are ampliated
([1.8.78]). Both (Ib) and (Ic) are treated summarily in [1.8.79], being reducible to spe-
cial cases of (1 a).
Buridan discusses (2) in [1.8.80]-[1.8.86] by reduction to the preceding case through
paraphrase, although there are two difficulties: (2a) grammatical agreement may be lost
in such conversions ([1.8.81]-[ 1.8.83]); (2b) how to resolve existentials ([ 1.8.84]-
[1.8.851). (There is also a digression in [1.8.86] on collective terms, which is not very
clear.)
Both (3) and (4) are treated quickly, in [1.8.87] and [1.8.88] respectively, by
reduction to the preceding cases.
After this discussion Buridan takes up three related problems: first, conversion may
violate the rules of grammar ([1.8.89]); second, sentences containing oblique terms may
require special treatment ([ 1.8.90]-[ 1.8.93]), and Buridan gives an alternative view and
then opposes it; finally, he reminds us that certain syncategorematic terms may require
special treatment ([1.8.94]). And that is the proof of Theorem 1-14 ([1.8.95]).
47 In Mental all assertoric categoricals have these three logical constituents, including
existence-statements, as Buridan argues in [1.8.84]-[1.8.85] below. Much of Buridan's
subsequent discussion will concern parsing the logical form of sentences whose gram-
matical form is not perspicuous.
48 See the brief discussion of the same examples given in TS 2.6.88.
49 The point in [1.8.86] is obscure, partly from the obscurity of the examples
(165)-(166); the translation reflects this unclarity. Perhaps Buridan means to be dis-
cussing the formations of an army (e.g. on a parade-ground), but then he should reject
the conversions because they do not make this explicit, not for the reason he gives.
Buridan's intentions here elude me.
so The rest of [1.8.90] expounds one of these alternative views, the good points of which
Buridan appreciates in [1.8.91]. But in [1.8.92] Buridan rejects it in favor of his own
view, there stated. In any event, Buridan tells us that nothing much rides on the decision
here ([1.8.94]).
SI Reading iste homo with the manuscript A rather than homo with Hubien; the sense
requires it.
S2 For conversion by contraposition see Section 7.3 of the Introduction; the contraposi-
tion of a sentence reverses the subject-term and the predicate-term and changes each ac-
cording to finite and infinite.
NOTES 347
BOOK II
I Buridan is too loose here: in a composite modal sentence such as "It is possible that
Socrates is running" the dictum is 'that Socrates is running,' in which the copula 'is'
appears; negations and quantifiers may also appear in the dictum. Buridan needs to find
a way to characterize those terms which belong to the dictum from those which do not
without a petitio principii.
2 Here and in the argument of [2.3.4] below Buridan is relying on the interdefinability
of 'necessarily' and 'possibly,' so that e.g. 'necessarily' is equivalent to 'not possibly not.'
This is not established as a formal result until Theorem II-I in [2.6.4] and the concomit-
ant analysis of necessity given by Theorem 11-2 in [2.6.8].
3 The example is obscure, although Buridan returns to it (e.g. in ]2.7.33]). Perhaps it
involves principles about the four elements and the analysis of change.
4 The argument for the nonequivalence of (231 *) and (231 **) is discussed in the Intro-
duction, Section 6.8; it turns on the difference in truth-conditions apparent in negative
sentences, and is most clearly stated by Buridan in ]2.4.5].
S This theorem establishes the equipollence of the modes 'necessarily' and 'possibly'
presupposed in [2.3.3]-[2.3.4] above.
6 This corollary is crucial for the sufficiency of Buridan's theorems dealing with divided
modals in Chapter II and Chapter IV.
7 Buridan should also take questions about contingents to be solved by the equipoll-
ence of 'contingently' and 'possibly and possibly not' (see Theorem 11-7 in [2.6.32], but
he does not; perhaps the conjunctive mode introduces special problems.
8 Since 'necessarily' and 'not possibly not' are equipollent, the sentences "A is necessar-
ily B" and "A is not possibly not B" are also equipollent, and the latter is analyzed as
"What is or can be A is not possibly not B," i.e. "What is or can be A is necessarily B."
9 Buridan is referring to the proposed equivalence of (231) and (231 **) discussed in
Chapter 11-4, which he rejects.
10 The analysis of (263) is either (263*) "What is or can be the one creating can be
God" or (263**) "What is creating can be God or what can be creating can be God."
Both (263*) and (263**) will be true if there is something which can create and can be
God, which is true since God always was and is and will be ([2.6.13]).
II This last case is covered as example (270) in the text of [2.6.17].
12 The first part of (268) is analyzed as a sentence with a disjunctive subject, "Anything
which is or can be creating is necessarily God," and so (268) is true in the posited case
because whatever can be creating is necessarily God, even though nothing is now creat-
ing.
I3 The following consequences are acceptable: "Some S is P; therefore, some S is possi-
bly P" and "All S is P; therefore, some S is possibly P."
14 The following consequences are acceptable: "Every A is possibly B; therefore, some B
is possibly A" (accidental conversion) and "Some A is possibly B; therefore, some B is
possibly A" (simple conversion).
15 The following consequence is acceptable: "Every A is necessarily not B; therefore,
every B is necessarily not A" (simple conversion).
16 Since (278) is a consequence, its contraposition is a consequence (by Theorem 1-3);
the contradictory of "Some S can be P" is "All S is necessarily not P" (see [2.6.19]); thus
the contrapositive of (278) is "All B is necessarily not A; therefore, all A is necessarily

not B."
'7 The following consequences are acceptable: "Every A is necessarily B; therefore,
what is B is necessarily A" and "Some A is necessarily B; therefore, something which is
B is necessarily A"
,g The following consequences are acceptable: "A is contingently B; therefore, A is
possibly B" and "A is contingently B; therefore A is possibly not B." They hold by the
definition of 'contingent' in [2.6.33].
'" Buridan discussed 'possible sentences,' contrasting them with 'possibly-true' sen-
tences, in [1.8.26].
20 The problem in (295)-(296) is that the term 'omne' may be construed as accusative
with 'B' or as nominative with 'possihile'; Buridan sensibly says that either construal is
possible.
2' Consequences of the form "Some [dictum] is [mode]; therefore, every [dictum] is
[mode]" are acceptable: the quantifications are across classes of inscriptions or utter-
ances. See the Introduction, Section 7.5, for a discussion of this theorem.
22 The following consequences are acceptable: (i) "No [dictum] is [mode]; therefore, no
[mode] is [dictum],,; (ii) "Some [dictum] is [mode]; therefore, some [mode] is [dictum],"
example (301) in the text; (iii) "Some [dictum] is not [mode]; therefore, some [mode] is
not [dictum]," where (i)-(iii) are simple conversions, and (IV) "Every [dictum] is [mode];
therefore, some [mode] is [dictum]," converting accidentally.
2.1 The following consequences are acceptable: (i) "Every [mode] is [dictum]; therefore,
every [dictum] is [mode]"; (ii) "Some [mode] is [dictum]; therefore, some [dictum] is
[mode]"; (iii) "No [mode] is [dictum]; therefore, no [dictum] is [mode]."
24 Note that one of the terms of a composite modal is the dictum; Buridan requires that
the dictum have supposition, i.e. that there be some sentence for which it supposits: it is
possible that no sentence equiform to 'Some A is B' is ever formed, but a sentence
equiform to 'Some A to be B is possible' is formed.
25 See the definition of 'consequence' in [1.3.11] and the discussion in the Introduction,
Section 7.2, for the existential requirement.
21> Recall the distinction between the possible and the possibly-true in [1.8.26]: 'Some
sentence is affirmative' is possible because it describes a possible situation, such as the
world in which only universal sentences are ever formed-in which case 'some sentence
is particular' is false, and so (306) is not a consequence. See Section 5.2 of the Introduc-
tion.
27 The following consequences are acceptable (giving the indefinite de falso composite
modal as the paradigm): (i) "It is false that some A is B; therefore, it is false that every A
is B"; (ii) "It is false that some A is not B; therefore, it is false that no A is B"; (iii) "It is
false that no A is B; therefore, it is false that no B is A"; (IV) "It is false that some A is
B; therefore, it is false that some B is A"
28 The following consequences are acceptable (taking the indefinite contingent com-
posite modal as the paradigm): (i) "It is contingent that every A is B; therefore, it is
contingent that some A is not B," and conversely; (ii) "It is contingent that no A is B;
therefore, it is contingent that some A is B," and conversely.
29 This is Buridan's version of the Tarski Biconditional: "It is true that [dictum]; there-
fore, [dictum]" and "[Dictum]; therefore, it is true that [dictum]." Note that when com-
bined with a mode the dictum is not a sentence, but when freestanding it is - hence
NOTES 349
Buridan's version of Tarski is rather different from "It is true that p '" p." Note also that
Buridan here ignores complications which the Liar-paradox raises: see 11.5.51-11.5.7]
and the Introduction, Section 7.5.
30 Theorem II-16(a) corresponds closely to standard distribution-principles of modal
logic: (i) from "It is necessary that p" and the consequence "p; therefore, if' there follows
"It is necessary that q," and (ii) from "It is possible that p' and the consequence "p;
therefore, if' there follows "It is possible that q."
31 Understand this last sentence in light of what Buridan says in [2.7.33]: if the universal
divided modal de possibili "Every A is possibly B" is true, then, although "Every A is B"
may not be possible, assertoric sentences of the form "This A is B" will each be possible
(though perhaps not compossible), which are the "singular claims" the universal "di-
vides."
32 That is, the preceding formula may fail when the predicate is ampliative, and so the
predicate itself must be modified. See the related discussion in TS 5.3.14.
33 The following mixed consequences are acceptable (taking the indefinite as the para-
digm): (i) "It is possible that every A is B; therefore, some A is possibly B"; (ii) "It is
possible that some A is B; therefore, some A is possibly B"; (iii) "It is possible that no A
is B; therefore, some A is possibly not B"; (iv) "It is possible that some A is not B;
therefore, some A is possibly not B."
34 The argument given in [2.7.39] shows that the consequence "Every B is necessarily
not A; therefore, it is not possible that B is A," example (318*) in the text, is acceptable.
3S The clause 'either formally or consecutively' means that a contingent simply defined
as a conjunction of affirmative and negative de possibili sentences (see [2.7.33]), which is
how as contingent is 'formally' an affirmative and a negative, or else such a conjunction
follows as a consequence, which is how a contingent is 'consecutively' an affirmative
and a negative ('consecutively' here is the adverbial form of consecutio, which has to do
with consequence, and not 'following in time,' the customary English usage). See also
[1.5.8].
36 See the remark in [2.6.35] for this point.
BOOK III
1 A consequence is an enthymeme if it is acceptable when some sentence is added; a

complete induction (inductio) is a consequence with argues from particular claims about
each member of a complete list to the general claim, "Socrates is mortal, and Plato is
mortal, et sic de singulis; therefore, all men are mortal"; an example passes from a single
instance to the general claim, e.g. "Socrates is mortal; therefore, all men are mortal."
2 This is one of the rare times Buridan speaks of the conditional rather than of conse-
quences; the rules are modus ponens, modus tollens, and transitivity.
3 This use of 'simple' has nothing to do with 'simple consequence' as discussed in
11.4.6]; Buridan means first-degree consequences, whose antecedent and consequent are
not themselves consequences. Note his move from conditionals to consequences in
[3.1.7]-[3.1.8).
4 These consequences will be formal only if their semantic relations as contradictories,
contraries, subcontraries, finite and infinite, and so forth are included in the Uniform
Substitution Principle (e.g. they are specified syntactically). The same holds for [3.1.12].
In each case the consequence is necessary, but it is not clear that it must also be formal:
see [1.4.2]-(1.4.6].
5 Despite the name these syllogisms do not necessarily have anything to do with infinite
terms; we customarily call them "disjunctive syllogisms."
6 This requirement will be given up, and the definitions of the syllogistic extremes and
middle generalized, when Buridan turns to the oblique syllogism in Chapter 111-6.
7 Buridan is correct: only when alternative definitions of the syllogistic figures are given
does the 'controversy' over the fourth figure have any point.
8 Buridan is thinking of a case in which a term appears in finite form in one sentence
and in infinite form in the other sentence, as for example "Some man is running" and "A
stone is a non-man."
9 The rules are Buridan's semantic formulation of the dictum de omni et nullo, and
their analysis is given in [3.4.5]-[3.4.71. See the Introduction, Section 8.2, for some com-
ments on Buridan's analysis.
10 By things which are 'divided' Buridan means distinct items which do not compose
some one thing; the linguistic analogue is counterpart to terms taken collectively.
11 Theorem 111-3 is the principal theorem in the Reductio Method for proving a syllog-
ism acceptable; see the Introduction, Section 8.3.

12 Buridan's text reads: "omnis syllogismi dati quaecumque conclusio sequitur ad eius
praemissas ilia sequitur ad unam ilIarum cum antecedente alterius; et quaecumque con-
clusio non sequitur ad ilIas praemissas ilia non sequitur ad a/iquam earum cum conse-
quente alterius." I have translated cum antecedente alterius not as "with the antecedent
of the other" but as "with a sentence which is antecedent to the other," and similarly for
cum consequente alterius. I suspect Buridan is careless here (one of the rare instances);
this is surely the sense demanded by [3.4.221. Theorem III-4 is the principal theorem in
the Method of Reduction for proving a syllogism acceptable and the sense I have given
it allows it to be so used. See the Introduction, Section 8.3.
13 An 'expository syllogism' is a syllogism in which the middle is a discrete term: see
13.3.41 above.
14 The following expository syllogisms are acceptable by Theorem III-5, taking 'this M'
as the example of a discrete term: (i) in the first figure, "This M is P, and S is this M;
therefore, S is P"; "This M is not P, and S is this M; therefore, S is not P"; and "This M is
P, and S is not this M; therefore, S is not P"; (ii) in the second figure, "P is this M, and S
is this M; therefore, S is P"; "P is not this M, and S is this M; therefore, S is not P"; and
"P is this M, and S is not this M; therefore, S is not P"; (iii) in the third figure, "This M is
P, and this M is S; therefore, Sis P"; "This M is not P, and this M is S; therefore, S is not
P"; and "This M is P, and this M is not S; therefore, S is not P." The conclusion of each
expository syllogism may be converted simply.
15 This is Buridan's first use of the Doctrine of Distribution to show the unacceptablility
of a syllogism; details about distribution are given in (3.4.291-[3.4.361. See also the In-
troduction, Sections 6.7 and 8.2, for details.
16 See TS Chapter IV for details about identificatory relative-terms.
17 This is the case when the minor is A-form or E-form in the first or second figures,
and when the minor is E-form or a-form in the third figure.

18 This is the case when the major is E-form or a-form in the first and third figures,
and when the major is A-form or E-form in the second figure.
19 A 'conjugation' is a pair of premisses specified as to quantity and quality. The possi-
NOTES 351
ble combinations are: (1) AA; (2) At; (3) AE; (4) AO; (5) EA; (6) EI; (7) EE; (8) EO;
(9) IA; (10) II; (11) IE; (12) 10; (13) OA; (14) 01; (15) OE; (16) 00. See the Introduc-
tion, Section 8.4. We shall refer to these conjugations by number hereafter.
20 This holds by Theorem III-2, because the middle is not distributed; we reject the
combinations (7)-(8) and (15)-(16).
21 That is, we reject the combination (10).
22 The five useless combinations given above, and (9) and (13), rejected for the first
figure here.
2, That is, we reject (14) for the first figure.
24 The figured conjugations useful for the first figure are (1 )-(6) and (11 )-(12).
25 Namely (3)-(4) and (11)-(12).
26 That is, in (4) and (12).
27 Reading here postponere for Hubien's praeponere: if the negation were placed before
the predicate, the sentence would have an infinitizing negation. The negation has to be
placed after the predicate for the sentence to be in the uncommon idiom for negatives.
2R The following are acceptable syllogisms, not in the common idiom for negatives (for
which see [1.8.70]), in the first figure: (4) "All M is P, and some S is not M; therefore,
some S P not is [i.e. some S is not some P]"; (12) "Some M is P, and some S is not M;
therefore, some S P not is li.e. some S is not some PI."
29 That is, in the four moods mentioned above, namely in (3) and (11).
,0 The following are acceptable syllogisms: (3) (Fapesmo) "All M is P, and no Sis M;
therefore, some P is not S"; (11) (Frisesomorum) "Some M is P, and no S is M; there-
fore, some P is not S."
31 That is, (1) and (2) in the first figure.
,2 Thus the following are acceptable syllogisms: for (1) the direct conclusion (Barbara)
is "All M is P, and all S is M; therefore, all S is P," and the indirect conclusion (Baralip-
ton) is "All M is P, and all S is M; therefore, some P is S"; and for (2) the direct conclu-
sion (Daril) is "All M is P, and some S is M; therefore, some S is P," and the indirect
conclusion (Dabitis) is "All M is P, and some S is M; therefore, some P is S."
" The following are acceptable syllogisms in direct conclusions: (5) (Celarent) "No Mis
P, and all S is M; therefore, no S is P"; (6) (Ferio) "No M is P, and some S is M; there-
fore, some S is not P."
,4 Thus (5) with an indirect conclusion is acceptable, namely (Celantes) "No M is P, and
all S is M; therefore, no P is S."
,5 Thus (6) with an indirect conclusion is acceptable, though not in the common idiom
for negatives, namely "No M is P, and some S is M; therefore, some P S not is [i.e. some
P is not some SI."
,6 When they conclude indirectly they are given the names Baralipton, Celantes, and
Dabitis respectively; see the end of [3.4.481.
,7 That is, Aristotle identified a 'syllogism' with a figured conjugation. Most scholars
agree with Buridan.
,8 That is, in addition to the five moods useless in all three figures, in the second figure
(1), (2), and (9) are also useless, because the middle is not distributed; hence the useful
moods in the second figure are (3 )-(6) and (11 )-( 14).
,9 The following are acceptable syllogisms: (5) (Cesare) "No P is M, and all S is M;
therefore, no S is P," with a direct conclusion, or "No P is M, and all S is M; therefore,
no P is S," with an indirect conclusion; (3) (Camestres) "All P is M, and no S is M; there-
fore, no S is P," with a direct conclusion, or "All P is M, and no S is M; therefore, no Pis

S," with an indirect conclusion.
40 That is, in the common idiom for negatives. Thus the following syllogisms are
acceptable: (6) (Festino) "No P is M, and some S is P; therefore, some S is not P"; (4)
(Baroco) "All P is M, and some S is not M; therefore, some S is not P."
41 The following syllogisms are acceptable: (11) (Tifesno) "Some P is M, and no S is M;
therefore, some P is not S"; (13) (Robaco) "Some P is not M, and all S is M; therefore,
some P is not S." Note that Buridan's coined names do not agree with the requirement
that the initial letter of the name of the syllogism be the initial letter of the first-figure
syllogism it is reduced to. See Section 8.3 of the Introduction.
42 The following syllogisms are acceptable: (12) "Some P is M, and some S is not M;
therefore, some S P not is [i.e. some S is not some Pj"; (14) "Some P is not M, and some
S is not M; therefore, some S P not is [i.e. some S is not some Pl." The conclusions may
also be simply converted.
43 In addition to the five moods useless in all figures, (12) and (14) are rejected for the
third figure since the middle is not distributed; thus the nine useful moods are (1 )-(6),
(9), (11), and (13).
44 The following syllogisms are acceptable: (1) (Daraptz) "All M is P, and all M is S;
therefore, some S is P," with a direct conclusion, and "All M is P, and all M is S; there-
fore, some P is S," with an indirect conclusion; (2) (Datisl) "All M is P, and some M is S;
therefore, some S is P," with a direct conclusion, and "All M is P, and some M is S;
therefore, some P is S," with an indirect conclusion; (9) (Disamis) "Some M is P, and all
M is S; therefore, some S is P," with a direct conclusion, and "Some M is P, and all M is
S; therefore, some P is S," with an indirect conclusion.
4, The following syllogisms are acceptable: (5) (Fe/apton) "No M is P, and all M is S;
therefore, some S is not P"; (13) (Bocardo) "Some M is not P, and all M is S; therefore,
some S is not P"; (6) (Ferison) "No M is P, and some M is S; therefore, some S is not P."
40 The following syllogisms are acceptable: (3) (Lapfeton) "All M is P, and no M is S;
therefore, some P is not S"; (4) (Carbodo) "All M is P, and some M is not S; therefore,
some P is not S"; (11) (Rifeson) "Some M is P, and no M is S; therefore, some P is not
S." Note again that Buridan's coined names do not agree with the rules for standard
syllogistic nomenclature.
47 The proof-procedure is contained in the traditional name for a syllogism; see the
Introduction, Section 8.3, for examples.
4H If the major extreme is ampliative (and no other term is), then in the first figure Bar-
bara and Celarent are not acceptable - Buridan gives counterexamples to these in
[3.4.61[ - and in the second figure the direct form of Cesare and the direct form of
Camestres are not acceptable; all the other nonampliative acceptable moods are accept-
able.
4" If the middle term is ampliative (aand no other term is), then by Theorem III-II the
nonampliative acceptable moods of the third figure are all acceptable, and by Theorem
III-12 in the first figure only Celarent and Celantes are acceptable, and in the second
figure only Cesare and Carnestres are acceptable.
,0 This counterexample is quite similar to one given in sophism 4 of Sophismata V.
,I See TS 2.2.5 for the same view.
52 Reading episcopi for hominis here: otherwise the proposed syllogism is not accept-
NOTES 353
able. Buridan proposes to replace the minor of (355) to give us "Every ass-of-a-bishop
is running, and Brunellus is an ass-of-a-bishop; therefore, Brunellus is running."
53 That is, 'omnem equum.' Here as throughout most of Chapter III-7 Buridan
discusses terms which are oblique in the accusative case; there is no reasonable way to
indicate that such terms are accusative in English. Part of the point of [3.6.4[ is to
indicate the difference between 'equum' in the major and 'equus in the minor. The
reader should be careful about these points of grammar in what follows.
54 See Section 4.3 of the Introduction for a discussion of intentional verbs.
55 In [3.7.1O[ Buridan clearly indicates that the appellation of a ratio will prevent it, and
in [3.7.11[ indicates that a temporal difference will prevent it: in both cases he refers to
the 'exception,' suggesting that the final clause of Theorem III -13 ought to be "... unless
the supposition is material, <or the ratio of a term is appellated, or the difference of a
time prevents it>." (There is no manuscript evidence for such an emendation, which is
why I have not put it in the text.)
56 Buridan refers to the categorial quantifiers and their relative-terms: see, for example,
TS 3.7.4, TS 4.2.3, TS 4.8.1, and the Introduction, Section 4.1.
57 That is, the Latin interrogative particles quid (what is it?), quantum (how-much is
it?), quale (how is it?), quando (when is it?), and ubi (where is it?). There are no simple
interrogative particles for the other categories.
58 Buridan treats (405) as sophism 4 of Sophismata VI.
59 Buridan's point is that there is an appropriate identity-relation for items of each cate-
gory: two quantities which are the same are called 'equal; not identical, and so for each
category. Sameness in the case of substances is simply identity, though, which is why
'whatever' is an unrestricted quantifier (as well as a categorial quantifier).
60 This marks the end of Buridan's discussion of the oblique syllogism; the rest of
Chapter III-7 is given over to the variation syllogism, in which the middle term appears
as finite in one premiss and infinite in the other premiss.
61 Buridan suggests that in every figure the conjugations (1)-(2), (7)-(10), and
(15)-(16) are useful, each figured conjugation generating a pair of variation syllogisms:
one in which the middle is finite in the major and infinite in the minor, and the other
conversely. Note that Buridan says in [3.7.32J, contrary to [3.4.53J, that some variation
syllogisms only have conclusions which are not in the common idiom for negatives. I
shall give examples for each figure, though only one for each pair of variation syllo-
gisms, namely where the middle is finite in the major and infinite in the minor, ignoring
conversions of the conclusion.
62 This explains why conjugations ruled out as useless before produce useful variation
syllogisms: the middle term is distributed in the premiss in which it is infinite.
63 The following are acceptable variation syllogisms, with affirmative premisses, in the
first figure: (1) "All M is P, and all S is non-M; therefore, all S P not is [i.e. all S is not
some PJ," which is the example given in (411) in the text; (2) "All M is P, and some S is
non-M; therefore, some S P not is [i.e. some S is not some PI"; (9) "Some M is P, and all
Sis non-M; therefore, some P is not S"; (10) "Some M is P, and some S is non-M; there-
fore, some P is not S."
64 The following variation syllogisms, with affirmative premisses, are acceptable in the
second figure: (1) "All P is M, and all S is non-M; therefore, no S is P," which is the
example in (413) in the text; (2) "All P is M, and some S is non-M; therefore, some S is
not P"; (9) "Some P is M, and all S is non-M; therefore, some S is not P"; (10) "Some P is
M, and some S is non-M; therefore, some S P not is [i.e. some S is not some PJ."
65 The following variation syllogisms, with affirmative premisses, are acceptable in the
third figure: (1) "All M is P and all non-M is S; therefore, some S is not P," which is the
example in (414) in the text; (2) "All M is P, and some non-M is S; therefore, some S is
not P"; (9) "Some M is P, and all non-M is S; therefore, some S is not P"; (10) Some
Mis P, and some non-M is S; therefore, some S P not is [i.e. some S is not some PJ."
66 The following variation syllogisms, with negative premisses, are acceptable in the first
figure: (7) "No M is P, and no S is non-M; therefore, no S is P," which is the example in
(415) in the text; (8) "No M is P, and some S is not non-M; therefore, some S is not P";
(15) "Some M is not P, and no S is non-M; therefore, some S is not P"; (16) "Some M is
not P, and some S is not non-M; therefore, some non-S non-P not is [i.e. some non-S is
not some non-PJ."
67 The following variation syllogisms, with negative premisses, are acceptable in the sec-
ond figure: (7) "No P is M, and no S is non-M; therefore, no S is P," which is the exam-
ple (417) in the text; (8) "No P is M, and some S is non-M; therefore, some S is not P";
(15) "Some P is not M, and no S is non-M; therefore, some S is not P"; (16) "Some P is
not M, and some S is not non-M; therefore, some S P not is [i.e. some S is not some PJ."
68 The following variation syllogisms, with negative premisses, are acceptable in the
third figure: (7) "No M is P, and no non-M is S; therefore, no S is P," which is the
example (418) in the text; (8) "No M is P, and some non-M is not S; therefore, some S is
not P"; (15) "Some M is not P, and no non-M is S; therefore, some S is not P"; (16)
"Some M is not P, and some non-M is not S; therefore, some non-S non-P is [i.e. some
non-S is not some non-PJ."
69 The following are acceptable variation syllogisms in the first figure: (3) "All M is P,
and no S is non-M; therefore, all S is P"; (4) All M is P, and some S is not non-M; there-
fore, some S is P"; (11) "Some M is P, and no S is non-M; therefore, some non-S P
not is [i.e. some non-S is not some PI"; (12) "Some M is P, and some S is not non-M;
therefore, some non-S P not is [i.e. some non-S is not some PJ."
70 The following variation syllogisms are acceptable in the third figure: (3) "All M is p,
and no non-M is S; therefore, all S is P"; (4) "All M is P, and some non-M is not S;
therefore, some S is P"; (5) "No M is P, and all non-M is S; therefore, all S is P"; (6) "No
Mis P, and some non-M is S; therefore, some S is P"; (11) "Some M is P, and no non-M
is S; therefore, some non-S P not is [i.e. some non-S is not some PI"; (12) "Some M is P,
and some non-M is not S; therefore, some non-S P not is [i.e. some non-S is not some
PI"; (13) "Some M is not P, and all non-M is S; therefore, some S is P"; (14) "Some M is
not P, and some non-M is S; therefore, some S non-P not is [i.e. some S is not some non-
PJ."
BOOK IV
I See Chapter 11-2 for this distinction.

2 See Chapter 11-4 for the analysis of de passibili divided modals, Theorem 11-2 in
[2.6.8J-[2.6.9J for the analysis of de necessaria composite modals, and [2.6.2J for the
"what is ..." locution.
NOTES 355
J In 14.1.4]-14.1.8] Buridan is concerned with the fact that Aristotle often gives demon-
strative syllogisms with merely assertoric premisses, thought to compromise the neces-
sity required for a science; he resolves the issue straightforwardly, admitting that such
syllogisms are not in fact formal consequences of themselves but are considered to be so
under the hypothesis of their necessity.
4 For any acceptable assertoric syllogism we may prefix premisses and conclusion with
the mode 'necessary' or 'true' and obtain an acceptable composite modal syllogism. For
example, we alter Baroeo as follows: "It is necessary that all P is M, and it is necessary
that some S is not M; therefore, it is necessary that some S is not M." Problems with
compossibility do not allow us to prefix the mode 'possible,' as Buridan points out in
(423), although prefixed to the whole antecedent (the conjunctive sentence) we do get
an acceptable syllogism.
s See 12.5.1]-12.5.2] for this reference. Buridan calls them 'quasi-equipollences' because
they are not relations among sentences; rather, they are relations between a sentence
and a sentence equiform to part of the original (its dictum).
6 The corresponding form of this syllogism is "It is necessary that all M is P, and it is
necessary that all S is M; therefore, it is necessary that all S is P": a modalized form of
Barbara.
7 These are errors the ideal reasoner would avoid, not difficulties with substitution in
opaque contexts. Buridan is here thinking of obligationes again: see the third group of
sophisms in Sophismata VIII.
8 The following divided modal syllogisms are acceptable in the first figure, where those
starred require relatively strong modal principles (see Section 8.6 of the Introduction):
for (1) with a direct conclusion there is (a) "All M is necessarily P, and all S is necessar-
ily M; therefore, all S is necessarily P," (b) "All M is possibly P, and all S is possibly M;
therefore, all S is possibly P," (c)* "All M is possibly P, and all S is necessarily M; there-
fore, all S is possibly P," (d)* "All M is necessarily P, and all S is possibly M; therefore,
no S is necessarily not P," and for (1) with an indirect conclusion convert the preceding
conclusions accidentally; for (2) with a direct conclusion there is (a) "All M is necessar-
ily P, and some S is necessarily M; therefore, some S is necessarily P," (b) "All M is
possibly P, and some S is possibly M; therefore, some S is possbily P," (c) "All M is pos-
sibly P, and some S is necessarily M; therefore, some S is possibly P," (d)* "All M is ne-
cessarily P, and some S is possibly M; therefore, no S is necessarily not P," and for (2)
with an indirect conclusion convert the preceding conclusions simply; for (3) there is (a)
"All M is necessarily P, and no S is necessarily M; therefore, some P is not necessarily
S," (b) "All M is possibly P, and no S is possibly M; therefore, some P is not possibly S,"
(c) "All M is possibly P, and no S is necessarily M; therefore, some P is possibly not S,"
(d) "All M is necessarily P, and no S is possibly M; therefore, some P is necessarily not
S"; for (5) with a direct conclusion there is (a) "No M is necessarily P, and all S is neces-
sarily M; therefore, no S is necessarily P," (b)* "No M is possibly P, and all S is possibly
M; therefore, all S is possibly not P," (c) "No M is possibly P, and all S is necessarily M;
therefore, no S is possibly P," (d) "No M is necessarily P, and all S is possibly M; there-
fore, no S is necessarily P," and for (5) with an indirect conclusion convert the preceding
conclusions simply; for (6) with a direct conclusion (the indirect conclusion is not in the
common idiom for negatives) there is (a) "No M is necessarily P, and some S is neces-
sarily M; therefore, some S is not necessarily P," (b) "No M is possibly P, and some S is
possibly M; therefore, some S is not possibly P," (c) "No M is possibly P, and some S is
necessarily M; therefore, some S is not possibly P," (d) "No M is necessarily P, and some
S is possibly M; therefore, some S is not necessarily P"; for (11) there is (a) "Some M is
necessarily P, and no S is necessarily M; therefore, some P is not necessarily S," (b)
"Some M is possibly P, and no S is possibly M; therefore, some P is not possibly S," (c)
"Some M is possibly P, and no S is necessarily M; therefore, some P is possibly not S,"
(d) "Some M is necessarily P, and no S is possibly M; therefore, some P is necessarily
notS."
9 The following are acceptable divided modal syllogisms in the second figure: for (3)
with a direct conclusion there is (a) "All P is necessarily M, and no S is necessarily M;
therefore, no S is necessarily P," (b) "All P is necessarily M, and no S is possibly M;
therefore, all S is necessarily not P," (c) "All P is possibly M, and no S is necessarily M;
therefore, I?? - no conclusion seems to follow; see Section 8.6 of the Introduction]," and
for (3) with an indirect conclusion convert the preceding conclusions simply; for (4)
there is (a) "All P is necessarily M, and some S is necessarily not M; therefore, some S is
necessarily not P," (b)* "All P is necessarily M, and some S is possibly not M; therefore,
some S is necessarily not P," (c) "All P is possibly M, and some S is necessarily not M;
therefore, some S is necessarily not P"; for (5) with a direct conclusion there is (a) "No P
is necessarily M, and all S is necessarily M; therefore, no S is necessarily P," (b) "No Pis
necessarily M, and all S is possibly M; therefore, no S is necessarily P," (c) "No P is pos-
sibly M, and all S is necessarily M; therefore, all S is necessarily not P," and for (5) with
an indirect conclusion convert the preceding conclusions simply; for (6) there is (a) "No
P is necessarily M, and some S is necessarily M; therefore, some S is necessarily not P,"
(b) "No P is necessarily M, and some S is possibly M; therefore, I?? - no conclusion
seems to follow; see Section 8.6 of the Introduction]," (c) "No P is possibly M, and some
S is necessarily M; therefore, some S is necessarily not P"; for (11) there is (a) "Some P is
necessarily M, and no S is necessarily P; therefore, some P is necessarily not S," (b)
"Some P is necessarily M, and no S is possibly M; therefore, some P is necessarily not
S," (c) "Some P is possibly M, and no S is necessarily M; therefore, some P is not neces-
sarily S"; for (13) there is (a) "Some P is necessarily not M, and all S is necessarily M;
therefore, some P is necessarily not S," (b) "Some P is necessarily not M, and all S is
possibly M; therefore, some P is not necessarily S," (c) "Some P is possibly not M, and
all S is necessarily M; therefore, some P is not necessarily S."
10 Examples of such unacceptable syllogisms in these terms are: "Every man is possibly
running, and every horse is possibly running; therefore, every horse is possibly a man,"
and "Any God is possibly not creating, and no first cause is possibly not creating; there-
fore, no first cause is possibly God."
11 The following are acceptable divided modal syllogisms in the third figure: for (1) with
a direct conclusion there is (a) "All M is necessarily P, and all M is necessarily S; there-
fore, some S is necessarily P," (b) "All M is possibly P, and all M is possibly S; therefore,
some S is possibly P," (c) "All M is necessarily P, and all M is possibly S; therefore,
some S is necessarily P," (d) "All M is possibly P, and all M is necessarily S; therefore,
some S is possibly P," and for (1) with an indirect conclusion convert the preceding con-
clusions simply; for (2) with a direct conclusion there is (a) "All M is necessarily P, and
some M is necessarily S; therefore, some S is necessarily P," (b) "All M is possibly P,
and some M is possibly S; therefore, some S is possibly P," (c)* "All M is necessarily P,
and some M is possibly S; therefore, some S is necessarily P," (d) "All M is possibly P,
NOTES 357
and some M is necessarily S; therefore, some S is possibly P," and for (2) with an indi-
rect conclusion convert the preceding conclusions simply; for (3) there is (a) "All M is
necessarily P, and no M is necessarily S; therefore, some P is not necessarily S," (b) "All
M is possibly P, and no M is possibly S; therefore, some P is not possibly S," (c) "All M
is necessarily P, and no M is possibly S; therefore, some P is necessarily not S," (d) "All
M is possibly P, and no M is necessarily S; therefore. some P is possibly not S"; for (4)
there is (a) "All M is necessarily P. and some M is necessarily not S; therefore. some P is
necessarily not S," (b) "All M is possibly P, and some M is possibly not S; therefore.
some P is possibly not S." (c) "All M is necessarily P. and some M is possibly not S;
therefore. some P is not necessarily S." (d) "All M is possibly P, and some M is neces-
sarily not S; therefore, some P is not possibly S"; for (5) there is (a) "No M is necessarily
P, and all M is necessarily S; therefore. some S is not necessarily P," (b) "No M is possi-
bly P. and all M is possibly S; therefore. some S is not possibly P." (c) "No M is neces-
sarily P, and all M is possibly S; therefore. some S is not necessarily P." (d) "No M is
possibly p. and all M is necessarily S; therefore. some S is not possibly P"; for (6) there
is (a) "No M is necessarily P, and some M is necessarily S; therefore. some S is not ne-
cessarily P," (b) "No M is possibly P. and some M is possibly S; therefore. some S is not
possibly P," (c) "No M is necessarily p. and some M is possibly S; therefore. some S is
not necessarily P," (d) "No M is possibly p. and some M is necessarily S; therefore. some
S is not possibly P"; for (9) with a direct conclusion there is (a) "Some M is necessarily
P, and all M is necessarily S; therefore, some S is necessarily P," (b) "Some M is possibly
P, and all M is possibly S; therefore. some S is possibly P," (c)* "Some M is necessarily
P, and all M is possibly S; therefore. some S is necessarily P," (d) "Some M is possibly P.
and all M is necessarily S; therefore. some M is possibly P," and for (9) with an indirect
conclusion convert all the preceding conclusions simply; for (11) there is (a) "Some M is
necessarily P. and no M is necessarily S; therefore, some P is not necessarily S." (b)
"Some M is possibly P, and no M is possibly S; therefore, some P is not possibly S." (c)
"Some M is necessarily P, and no M is possibly S; therefore. some P is necessarily not
S," (d) "Some M is possibly P, and no M is necessarily S; therefore, some P is possibly
not S"; for (13) there is (a) "Some M is necessarily not P, and all M is necessarily S;
therefore, some S is necessarily not P," (b) "Some M is possibly not P. and all M is pos-
sibly S; therefore, some S is possibly not P," (c) "Some M is necessarily not P, and all M
is possibly S; therefore, some S is not necessarily P," (d) "Some M is possibly not P. and
all M is necessarily S; therefore. some S is possibly not P."
12 The following are acceptable divided modal syllogisms formed with the "what is ... "
locution in the first figure: for (I) with a direct conclusion there is (a) "All which is M is
necessarily P, and all S is necessarily M; therefore, all which is S is necessarily P." (b)
"All which is M is possibly P, and all S is necessarily M; therefore, all which is S is possi-
bly P." and for (1) with an indirect conclusion there is (a) "All which is M is necessarily
P, and all S is necessarily M; therefore. some P is necessarily S." (b) "All which is M is
possibly P. and all S is necessarily M; therefore, some P is possibly S"; for (2) with a di-
rect conclusion there is (a) "All which is M is necessarily P, and some S is necessarily M;
therefore. something which is S is necessarily P." (b) "All which is M is possibly P, and
some S is necessarily M; therefore, something which is S is possibly P," and for (2) with
an indirect conclusion there is (a) "All which is M is necessarily P. and some S is neces-
sarily M; therefore, something which is P is necessarily S." (b) "All which is M is possi-
bly P, and some S is necessarily M; therefore, something which is P is possibly S"; for (3)
there is (a) "All which is M is necessarily P, and no S is necessarily M; therefore, nothing

which is P is necessarily S," (b) "All which is M is possibly P, and no S is necessarily M;
therefore, something which is P is possibly not S"; for (5) with a direct conclusion there
is (a) "Nothing which is M is necessarily P, and all S is necessarily M; therefore, nothing
which is S is necessarily P," (b) "Nothing which is M is possibly P, and all S is necessarily
M; therefore, nothing which is S is possibly P," and for (5) with an indirect conclusion
there is (a) "Nothing which is M is necessarily p, and all S is necessarily M; therefore,
nothing which is P is necessarily S," (b) "Nothing which is M is possihly P, and all S is
necessarily M; therefore, nothing which is P is possibly S"; for (6) there is (a) "Nothing
which is M is necessarily P, and some S is necessarily M; therefore, something which is S
is not necessarily P," (b) "Nothing which is M is possibly P, and some S is nccessarily M;
therefore, something which is S is not possibly P"; for (II) there is (a) "Something which
is M is necessarily P, and no S is necessarily M; therefore, something which is P is not
necessarily S," (b) "Something which is M is possibly P, and no S is necessarily M; there-
fore, something which is P is possibly not S."
lJ Hence there are three acceptable divided modal syllogisms restricted by the "what is
... " locution in the second figure, as follows (here I only give the de possibili conclu-
sions): for (3) with a direct conclusion there is (a) "All which is P is necessarily M, and
nothing which is S is necessarily M; therefore, all which is S is possibly not P," (b) "All
which is P is necessarily M, and nothing which is S is possibly M; therefore, nothing
which is S is possibly P," (cl "All which is P is possibly M, and nothing which is S is
necessarily M; therefore, all which is S is possibly not P," and for (3) with an indirect
conclusion convert the preceding conclusions simply; for (4) there is (a) "All which is P
is necessarily M, and something which is S is necessarily not M; therefore, something
which is S is necessarily P," (h) "All which is P is necessarily M, and something which is
S is possibly not M; therefore, something which is S is possibly not P," (c) "All which is
P is possibly M, and something which is S is necesarily not M; therefore, nothing which
is S is possibly P"; for (5) with a direct conclusion there is (a) "Nothing which is P is
necessarily M, and all S is necessarily M; therefore. nothing which is S is necessarily P,"
(b) "Nothing which is P is possibly M. and all which is S is necessarily M; therefore,
nothing which is S is possibly P," (c) "Nothing which is P is necessarily M. and all which
is S is possibly M; therefore, all which is S is possihly not P," and for (5) with an indirect
conclusion convert the preceding conclusions simply; for (6) there is (a) "Nothing which
is P is necessarily M, and something which is S is necessarily M; therefore, something
which is S is possibly not P," (b) "Nothing which is P is necessarily M, and something
which is S is possibly M; therefore. something which is S is possibly not P." (c) "Nothing
which is P is possibly M, and something which is S is necessarily M; therefore, nothing
which is S is possibly P"; for (II) there is (a) "Something which is P is necessarily M,
and nothing which is S is necessarily M; therefore, something which is P is possibly not
S," (b) "Something which is P is necessarily M, and nothing which is S is possibly M;
therefore, something which is P is not possibly S." (c) "Something which is P is possibly
M, and nothing which is S is necessarily M; therefore, something which is P is possibly
not S"; for (13) there is (a) "Something which is P is necessarily not M, and all which is S
is necessarily M; therefore, something which is P is necessarily not S," (b) "Something
which is P is necessarily not S. and all which is S is possibly M; therefore, something
which is P is not possibly S," (c) "Something which is P is possibly not M, and all which
is S is necessarily M; therefore, something which is P is possibly not S."
NOTES 359
14 Hence there are four acceptable divided modal syllogisms restricted by the "what is
..." locution in the fourth figure, which are as follows (I give here only the unrestricted
conclusions where possible): for (1) with a direct conclusion there is (a) "All which is M
is necessarily P, and all which is M is necessarily S; therefore, some S must be P," (b)
"All which is M is necessarily P, and all which is M is possibly S; therefore, some S is
not necessarily not P," (c) "All which is M is possibly P, and all which is M is necessarily
S; therefore, some S is possibly P," (d) "All which is M is possibly P, and all which is M
is possibly S; therefore, some S is possibly P," and for (1) with an indirect conclusion
convert the preceding conclusions simply; for (2) with a direct conclusion there is (a)
"All which is M is necessarily P, and something which is M is necessarily S; therefore,
some S is necessarily P," (b) "All which is M is necessarily P, and something which is M
is possibly S; therefore, some S is not necessarily not P," (c) "All which is M is possibly
P, and something which is M is necessarily S; therefore, some S is possibly P," (d) "All
which is M is possibly P, and something which is M is possibly S; therefore, some S is
possibly P," and for (2) with an indirect conclusion convert the preceding conclusions
simply; for (3) there is (a) "All which is M is necessarily P, and nothing which is M is
necessarily S; therefore, some P is not necessarily S," (b) "All which is M is necessarily
P, and nothing which is M is possibly S; therefore, some P is not possibly S," (c) "All
which is M is possibly P, and nothing which is M is necessarily S; therefore, some P is
not necessarily S," (d) "All which is M is possibly P, and nothing which is M is possibly
S; therefore, some P is not possibly S"; for (4) there is (a) "All which is M is necessarily
P, and something which is M is necessarily not S; therefore, some P is necessarily not S,"
(b) "All which is M is necessarily not P, and something which is M is possibly not S;
therefore, some P is not necessarily S," (c) "All which is M is possibly P, and something
which is M is necessarily not S; therefore, some P is not possibly S," (d) "All which is M
is possibly P, and something which is M is possibly not S; therefore, some P is possibly
not S"; for (5) there is (a) "Nothing which is M is necessarily P, and all which is M is ne-
cessarily S; therefore, some S is not necessarily P," (b) "Nothing which is M is necessar-
ily P, and all which is M is possibly S; therefore, some S is not necessarily P," (c)
"Nothing which is M is possibly P, and all which is M is necessarily S; therefore, some S
is not possibly P," (d) "Nothing which is M is possibly P, and all which is M is possibly
S; therefore, some S is not possibly P"; for (6) there is (a) "Nothing which is M is neces-
sarily P, and something which is M is necessarily S; therefore, some S is not necessarily
P," (b) "Nothing which is M is necessarily P, and something which is M is possibly S;
therefore, some S is not necessarily P," (c) "Nothing which is S is possibly P, and some-
thing which is M is necessarily S; therefore, some S is not possibly P," (d) "Nothing
which is M is possibly P, and something which is M is possibly S; therefore, some S is
not possibly P"; for (9) with a direct conclusion there is (a) "Something which is M is ne-
cessarily P, and all which is M is necessarily S; therefore, some S is necessarily P," (b)
"Something which is M is necessarily P, and all which is M is possibly S; therefore, some
S is not necessarily not P," (c) "Something which is M is possibly P, and all which is M is
necessarily S; therefore, some S can be P," (d) "Something which is M is possibly P, and
all which is M is possibly S; therefore, some S is possibly P ," and for (9) with an indirect
conclusion convert the preceding conclusions simply; for (11) there is (a) "Something
which is M is necessarily P, and nothing which is M is necessarily S; therefore, some P is
not necessarily S," (b) "Something which is M is necessarily P, and nothing which is M is
possibly S; therefore, some P is necessarily not S," (c) "Something which is M is possibly
P, and nothing which is M is necessarily S; therefore, some P is possibly not S;' (d)
"Something which is M is possibly P, and nothing which is M is possibly S; therefore,
some P is not possibly S"; for (13) there is (a) "Something which is M is necessarily not
P, and all which is M is necessarily S; therefore, some P is necessarily not S," (b)
"Something which is M is necessarily not P, and all which is M is possibly S; therefore,
some P is necessarily not S," (c) "Something which is M is possibly not P, and all which
is M is necessarily S; therefore, some P is possibly not S," (d) "Something which is M is
possibly not P, and all which is M is possibly S; therefore, some P is possibly not S,"
" The following 'mixed' syllogisms (syllogisms with assertoric and divided modal
premisses) are acceptable in the first figure: for (I) there is "All M is possibly P, and all
S is M; therefore, some P is possibly S"; for (2) with a direct conclusion there is "All M
is possibly P, and some S is M; therefore, some S is possibly P," and for (2) with an
indirect conclusion there is "All M is possibly P, and some S is M; therefore, some P can
be S"; for (3) there is "All M is possibly P, and no S is M; therefore, some P is possibly
not S"; for (6) there is "No M is possibly P, and some S is M; therefore, some S is not
possibly P"; for (II) there is "Some M is possibly P, and no S is M; therefore, some P is
possibly not S," Only the modalized versions of Barbara, Celarent, and Celantes are
prevented; Buridan gives counterexamples to the first two in 14.2.291 and shows how to
combine them with the "what is ... " locution so they are acceptable in 14.2.301.
In The following mixed syllogisms, assuming the major to be necessary (though in asser-
toric form), are acceptable in the first figure: for (I) with a direct conclusion there is "All
M is P (necessarily), and all S is possibly M; therefore, all S is possibly P," which is the
example (439) in the text, and for (I) with an indirect conclusion there is "All M is P
(necessarily), and all S is possibly M; therefore, some P is possibly S"; for (2) with a
direct conclusion there is "All M is P (necessarily), and some S is possibly M; therefore,
some S is possibly P," and for (2) with an indirect conclusion there is "All M is P
(necessarily), and some S can be M; therefore, some P is possibly S"; for (3) there is "All
Mis P (necessarily), and no S is possibly M; therefore, some P is not possibly S"; for (5)
with a direct conclusion there is "No M is P (necessarily), and all S is possibly M;
therefore, all S is possibly not P," which is the example (441) in the text, and for (5) with
an indirect conclusion there is "No M is P (necessarily), and all S is possibly M; there-
fore, all P is possibly not S"; for (6) there is "No M is P (necessarily), and some S is
possibly M; therefore, some S is possibly not P"; for (11) there is "Some M is P
(necessarily), and no S is possibly M; therefore, some P is not possibly S."
17 It is a mod ali zed version of Bocardo, corresponding to (13) in the third figure.
IX The following mixed syllogisms, where the major is necessary (though stated asser-
torically), are acceptable in the second figure: for (3) with a direct conclusion there is
"All P is M (necessarily), and no S is possibly M; therefore, no S is possibly P," and for
(3) with an indirect conclusion there is "All P is M (necessarily), and no S is possibly M;
therefore, no P is possibly S"; for (4) there is "All P is M (necessarily), and some S is
possibly not M; therefore, some S is possibly not P"; for (5) with a direct conclusion
there is "No P is M (necessarily), and all S is possibly M; therefore, no S is possibly P,"
and for (5) with an indirect conclusion there is "No P is M (necessarily), and all S is
possibly M; therefore, no P is possibly S"; for (6) there is "No P is M (necessarily), and
some S is possibly M; therefore, some S is not possibly P"; for (II) there is "Some P is M
(necessarily), and no S is possibly M; therefore, some P is not possibly S"; for (13) there
NOTES 361
is "Some P is not M (necessarily), and all S is possibly M; therefore, some P is not

possibly S."
10 Here, as throughout the rest of the chapter, Buridan (i) uses the names of the
traditional assertoric syllogistic to discuss the modalized versions; (ii) ignores his non-
traditional, but perfectly acceptable, syllogisms Tifesno, Robaco, Lapfeton, Carbodo,
and Rifeson; we shall note any problems which arise on this account.
20 The following mixed syllogisms in the third figure are acceptable: for (I) with a direct
conclusion, there are (a) "All M is possibly P, and all M is S; therefore, some S is
possibly P," (b) "All M is P, and all M is possibly S; therefore, some S is possibly P," and
for (I) with an indirect conclusion convert the precedign conclusions simply; for (2)
with a direct conclusion there is "All M is possibly P, and some M is S; therefore, some
S is possibly P," and for (2) with an indirect conclusion convert the preceding conclusion
simply; for (3) there are (a) "All M is possibly P, and no M is S; therefore, some P is
possibly not S," (b) "All M is P, and no M is possibly S; therefore, some P is not possibly
S"; for (4) there is "All M is possibly P, and some M is not S; therefore, some P is
possibly not S"; for (5) there are (a) "No M is possibly P, and all M is S; therefore, some
S is possibly not P," (b) "No M is P, and all M is possibly S; therefore, some S is possibly
not P"; for (6) there is "No M is possibly P, and some M is S; therefore, some S is
possibly not P"; for (9) with a direct conclusion there is "Some S is P, and all M is possi-
bly S; therefore, some S is possibly P," and for (9) with an indirect conclusion convert
the preceding conclusion simply; for (11) there is "Some M is P, and no M is possibly S;
therefore, some P is not possibly S"; for (13) there is "Some M is not P, and all M is
possibly S; therefore, some S is possibly not P."
" The following mixed syllogisms are acceptable in the third figure: for (I) with a direct
conclusion there is "All M is not possibly P, and all M is S; therefore, some S is not
possibly P,"and for (I) with an indirect conclusion convert the preceding conclusion
simply; for (2) with a direct conclusion there is "All M is not possibly P, and some M is
S; therefore, some S is not possibly P," and for (2) with an indirect conclusion convert
the preceding conclusion simply; for (3) there is "All M is not possibly P, and no M is S;
therefore, I?? - no conclusion of the form Buridan prescribes seems to follow; see
Section 8.6 of the Introduction],,; for (4) there is "All M is not possibly P, and some M is
not S; therefore I?? - no conclusion of the sort Buridan prescribes seems to follow; see
Section 8.6 of the IntroductionI"'; for (5) there is "No M is not possibly P, and all M is S;
therefore, some S is not possibly not P"; for (6) there is "No M is not possibly P. and
some M is S; therefore, some S is not possibly not P."
22 The following mixed syllogisms are acceptable in the third figure (not listing those
acceptable even when the assertoric premiss is not necessary given above in Theorem
IV-l3): for (2) with a direct conclusion there is "All M is P (necessarily), and some M is
possibly S; therefore, some S is possibly P," and for (2) with an indirect conclusion
convert the preceding conclusion simply; for (4) there is "All M is P (necessarily), and
some M is possibly not S; therefore, some P is possibly not S"; for (6) there is "No M is
P (necessarily), and some M is possibly S; therefore. some S is possibly not P"; for (9)
with a direct conclusion there is "Some M is possibly P, and all M is S (necessarily);
therefore, some S is possibly P," and for (9) with an indirect conclusion convert the
preceding conclusion simply; for (II) there is "Some M is possibly P, and no M is S
(necessarily); therefore, some P is possibly not S"; for (13) there is "Some M is possibly
not P, and all M is S (necessarily); therefore, some S is not possibly P."

2.1 The acceptable mixed syllogism, a variant of Celarent, is given as example (443) in
the text.
24 The following mixed syllogisms are acceptable in the first figure: for (1) with a direct
conclusion there is "All M is necessarily P, and all S is M; therefore, some S is neces-

sarily P," and for (1) with an indirect conclusion convert the preceding minor simply; for
(2) with a direct conclusion there is "All M is necessarily P, and some S is M; therefore,
some S is necessarily P," and for (2) with an indirect conclusion convert the preceding
conclusion simply; for (3) there is "All M is necessarily P, and no S is M; therefore,
some P is not necessarily S"; for (5) with a direct conclusion there is "No M is neces-
sarily P, and all S is M; therefore, some S is not necessarily P," and for (5) with an
indirect conclusion convert the preceding conclusion simply; for (6) there is "No M is
necessarily P, and some S is M; therefore, I?? - no assertoric conclusion seems to
follow; see Section 8.6 of the Introduction!"; for (11) there is "Some M is necessarily P,
and no S is M; therefore, some P is not necessarily S."
2, The following mixed syllogisms are acceptable in the first figure: for (1) with a direct
conclusion there is "All M is necessarily P, and all S is M; therefore, all S is P," and for
(1) with an indirect conclusion convert the preceding conclusion accidentally; for (2)
with a direct conclusion there is "All M is necessarily P, and some S is M; therefore,
some S is P," and for (2) with an indirect conclusion convert the preceding conclusion
simply; for (3) there is "All M is necessarily P, and no S is M; therefore, some P is not
S"; for (5) with a direct conclusion there is "No M is necessarily P, and all S is M; there-
fore, [~? - no assertoric conclusion seems to follow; see Section 8.6 of the Introduction[,"
and for (5) with an indirect conclusion see Section 8.6 of the Introduction; for (6) there
is "No M is necessarily P, and some S is M; therefore [ - I can find no acceptable
conclusion here; see Seclion 8.6 of the Introduction!"; for (11) there is "Some M is
necessarily P, and no S is M; therefore, some P is not S."
21> The following mixed syllogisms are acceptable in the second figure: for (5) with a
direct conclusion there is "No P is necessarily M, and all S is M; therefore, some S is not
necessarily P," and for (5) with an indirect conclusion convert the preceding conclusion
simply; for (6) there is "No P is necessarily M, and some S is M; therefore, some S is not
necessarily P"; for (13) there is "Some P is not necessarily M, and all S is M; therefore,
some P is necessarily not S."
27 The following mixed syllogisms are acceptable in the second figure: for (3) with a
direct conclusion there is "All P is necessarily M, and no S is M; therefore, no S is P,"
and for (3) with an indirect conclusion convert the preceding conclusion simply; for (4)
there is "All P is necessarily M, and some S is not M; therefore, some S is not P"; for
(11) there is "Some P is necessarily M, and no S is M; therefore, some P is not S,"
2K The following mixed syllogisms are acceptable in the third figure: for (1) with a direct
conclusion there is "All M is necessarily P, and all M is S; therefore, some S is neces-
sarily P," and for (1) with an indirect conclusion convert the preceding conclusion
simply; for (2) with a direct conclusion there is "All M is necessarily P, and some M is S;
therefore, some S is necessarily P," and for (2) with an indirect conclusion convert the
preceding conclusion simply; for (3) there is "All M is necessarily P, and no M is S;
therefore, some P is not necessarily S"; for (4) there is "All M is necessarily P, and some
M is not S; therefore, some P is not necessarily S"; for (5) there is "No M is necessarily P
NOTES 363
and all M is S; therefore, some S is not necessarily P"; for (6) there is "'No M is neces-
sarily P, and some M is S; therefore, some S is not necessarily P."
2" The following mixed syllogisms are acceptable in the third figure: for (I) with a direct
conclusion there is "All M is necessarily P, and all M is S (necessarily); therefore, some
S is necessarily P," and for (I) with an indirect conclusion convert the preceding con-
clusion simply; for (2) with a direct conclusion there is "All M is necessarily P, and some
M is S (necessarily); therefore, some S is necessarily P," and for (2) with an indirect con-
clusion convert the preceding conclusion simply; for (3) there is "All M is necessarily P,
and no M is S (necessarily); therefore, some P is necessarily not S"; for (4) there is "All
M is necessarily P, and some M is not S (necessarily); therefore, some P is necessarily
not S"; for (5) there is "No M is necessarily P, and all M is S (necessarily); therefore,
some S is not necessarily P"; for (6) there is "No M is necessarily P, and some M is S
(necessarily); therefore, some S is not necessarily P"; for (9) with a direct conclusion
there is "'Some M is necessarily P, and all M is S (necessarily); therefore, some S is
necessarily P," and for (9) with an indirect conclusion convert the preceding conclusion
simply; for (II) there is "Some M is necessarily P, and no M is S (necessarily); therefore,
some P is necessarily not S"; for (13) there is "Some M is necessarily not P, and all M is
S (necessarily); therefore, some S is necessarily not P."
311 Theorem IV-21 applies to syllogisms shown acceptable by Theorem IV-4 through
Theorem IV-14 (except for Theorem IV-6(b».

" Theorem IV-22 applies to syllogisms shown acceptable by Theorem IV-4 through
Theorem IV-9 (except for Theorem IV-6(a) and (c» and Theorem IV-16 through
Theorem IV-20.
32 The following mixed contingent syllogisms are acceptable in the first figure (only
giving the case where thc mode is affirmed; for the denied mode simply replace
'contingently' by 'not contingently' wherever it occurs): for (I) with a direct conclusion
there is (a) "All M is contingently P, and all S is necessarily M; therefore, all S is
contingently P," (b) "'All M is contingently P, and all S is possibly M; therefore, all S is
possibly P," and for (I) with an indirect conclusion convert the preceding conclusions
accidentally; for (2) with a direct conclusion there is (a) "'All M is contingently P, and
some S is necessarily M; therefore, some S is contingently P," (b) "All M is contingently
P, and some S is possibly M; therefore, some S is possibly P," and for (2) with an
indirect conclusion convert the preceding conclusions simply; for (3) there is (a) "'All M
is contingently P, and no S is necessarily M; therefore, some P is contingently not S," (b)
"All M is contingently P, and no S is possibly M; therefore, some P is not possibly S";
for (5) with a direct conclusion there is (a) "No M is contingently P, and all S is
necessarily M; therefore, no S is contingently P," (b) "'No M is contingently P, and all S
is possibly M; therefore, no S is possibly P," and for (5) with an indirect conclusion
convert the preceding conclusions simply; for (6) there is (a) "'No M is contingently P,
and some S is necessarily M; therefore, some S is not contingently P," (b) "No M is
contingently P, and some S is possibly M; therefore, some S is not possibly P"; for (II)
there is (a) "Some M is contingently P, and no S is necessarily M; therefore, some P is
contingently not S," (b) "'Some M is contingently P, and no S is possibly M; therefore,
some P is not possibly S."
33 The following mixed contingent syllogisms are acceptable in the third figure (only
giving cases with the mode affirmed): for (I) with a direct conclusion there is (a) "All M
is contingently'p, and all M is necessarily S; therefore, some S is contingently P," (b) "All
M is contingently P, and all M is possibly S; therefore, some S is possibly P," and for (1)
with an indirect conclusion convert the preceding conclusions simply; for (2) with a
direct conclusion there is (a) "All M is contingently P, and some M is necessarily S;
therefore, some S is contingently P," (b) "All M is contingently P, and some M is
convert the preceding conclusions simply; for (3) there is (a) "All M is contingently P,
and no M is necessarily S; therefore, some P is contingently not S," (b) "All M is con-
tingently P, and no M is possibly S; therefore, some P is not possibly S"; for (4) there is
(a) "All M is contingently P, and some M is necessarily not S; therefore, some P is not
contingently S," (b) "All M is contingently P, and some M is possibly not S; therefore,
some P is possibly not S"; for (5) there is (a) "No M is contingently P, and all M is neces-
sarily S; therefore, some S is not contingently P," (b) "No M is contingently P, and all M
is possibly S; therefore, some S is not possibly P"; for (6) there is (a) "No M is contin-
gently P, and some M is necessarily S; therefore, some S is not contingently P," (b) "No
M is contingently P, and some M is possibly S; therefore, some S is not possibly P"; for
(9) with a direct conclusion there is (a) "Some M is contingently P, and all M is neces-
sarily S; therefore, some S is contingently P," (b) "Some M is contingently P, and all M is
convert the preceding conclusions simply; for (11) there is (a) "Some M is contingently
P, and no M is necessarily S; therefore, some P is contingently not S," (b) "Some M is
contingently P, and no M is possibly S; therefore, some P is not possibly S"; for (13)
there is (a) "Some M is contingently not P, and all M is necessarily S; therefore, some S is
contingently not P," (b) "Some M is contingently not P, and all M is possibly S; there-
fore, some S is possibly not P."
3. The following mixed contingent syllogisms are acceptable in the first figure: for (1)
there is "All M is contingently P, and all S is M; therefore, some P is contingently S"; for
(2) with a direct conclusion there is "All M is contingently P, and some S is M; there-
fore, some S is contingently P," and for (2) with an indirect conclusion convert the
preceding conclusion simply; for (3) there is "All M is contingently P, and no S is M;
therefore, some P is contingently not S"; for (6) there is "No M is contingently P, and
some S is M; therefore, some S is not contingently P"; for (11) there is "Some M is
contingently P, and no S is M; there4)fe, some P is contingently not S."
35 The following mixed contingent syllogisms are acceptable in the third figure: for (1)
with a direct conclusion there is "All M is contingently P, and all M is S; therefore, some
S is contingently P," and for (1) with an indirect conclusion convert the preceding
conclusion simply; for (2) with a direct conclusion there is "All M is contingently P, and
some M is S; therefore, some S is contingently P," and for (2) with an indirect con-
clusion convert the preceding conclusion simply; for (3) there is "All M is contingently
P, and no M is S; therefore, some P is contingently not S"; for (4) there is "All M is
contingently P, and some M is not S; therefore, some P is contingently not S"; for (5)
there is "No M is contingently P, and all M is S; therefore, some S is not contingently P";
for (6) there is "No M is contingently P, and some M is S; therefore, some S is not
contingently P."
36 The following reduplicative syllogisms are acceptable in the first figure: for (1) there
is "All M is P qua R, and all S is M; therefore, all S is P qua R"; for (2) there is "All M is
P qua R, and some S is M; therefore, some S is P qua R"; for (5) there is "No M is P
NOTES 365
qua R, and all S is M; therefore, no S is P qua R"; for (6) there is "No M is P qua R, and
some S is M; therefore, some S is not P qua R," The following reduplicative syllogisms
are acceptable in the third figure: for (1) there is "All M is P qua R, and all M is S; there-
fore, some S is P qua R"; for (2) there is "All M is P qua R, and some M is S; therefore,
some S is P qua R"; for (5) there is "No M is P qua R, and all M is S; therefore, some S
is not P qua R"; for (6) there is "'No M is P qua R, and some M is S; therefore, some S is
not P qua R"; for (9) there is "Some M is P qua R, and all M is S; therefore, some S is P
quaR."
BIBLIOGRAPHY
I. BURIDAN: PRIMARY SOURCES
Treatise on Supposition (TS): Edited by Maria Elena Reina, 'Giovanni Buridano:

Tractatus de suppositionibus', Rivista critica di storia della filosofia XIV (1959), pp.
175-208 and pp. 323-352.
Treatise on Consequences (TC): Edited by Hubert Hubien, lohannis Buridani tractatus
de consequentiis: Edition critique in the series Philosophes medievaux Vol. XVI,
Universite de Louvain, 1976.
Questions on the Metaphysics (QM): Quaestiones in Metaphysicen Aristotelis. Paris
1518; photoreprint Minerva G.M.B.H., Frankfurt-am-Main 1964.
Questions on the Physics (QSP): Quaestiones subtillisimae super octo Physico rum libros
Aristotelis, Paris 1509, photoreprint Minerva G.M.B.H., Frankfurt-am-Main 1964.
Questions on the Nichomachean Ethics (QNE): Quaestiones in decem libros Ethicorum
ad Nicomachum, Oxford 1637.
Questions on the De caelo et mundo (QCM): Quaestiones super De caelo et mundo,
edited by Ernest A. Moody, Mediaeval Academy of America, Cambridge, 1942.
Sophismata: Edited by T. K. Scott, Johannes Buridanus: Sophismata in the series
Grammatica speculativa Vol. I, Fromann-Holzboog, Stuttgart 1977. A translation
is given in Scott (1966). The eighth chapter has been re-edited and translated in
Hughes (1982).
2. SECONDARY SOURCES
Belnap, Nuel D: 1973, 'Restricted Quantification and Conditional Assertion', in Hughes

Leblanc (ed.), Truth, Syntax, and Modality, North-Holland, Amsterdam, pp. 48-75.
DeRijk, L. M. (editor): 1972, Peter of Spain, Tractatus: Afterwards called Summulae
logicales, Van Gorcum, Assen.
Donnellan, Keith: 1966, 'Reference and Definite Descriptions', Philosophical Review
LXXV, 281-304.
Faral, Edmond: 1949, 'Jean Buridan: Maitre es arts de l'Universite de Paris' in the
Histoire litteraire de France, Vol. XXVIII (Part 2), Imprimerie Nationale, Paris,
pp.462-605.
Geach, Peter T.: 1962, Reference and Generality (second edition), Cornell University
Press, Ithaca.
Geach, Peter T.: 1972, 'A Mediaeval Discussion of Intentionality', in Logic Matters,
University of California Press.
Gibson, Joan: 1976, 'The Role of Mental Language in the Philosophy of William of
Ockham', unpublished Ph.D. Dissertation, University of Toronto.
367
368 BIBLIOGRAPHY
Hughes, G. E.: 1982, John Buridan on Self-Reference (edition and translation of

Sophismata VIII), Cambridge University Press.
Kretzmann, Norman: 1974, 'Aristotle on Spoken Sound Significant by Convention', in
John Corcoran (ed.), Ancient Logic and Its Modern Interpretations, in the Synthese
Historical Library Vol. IX, D. Reidel, Dordrecht, Holland, 1974, pp. 3-21.
Minio-Paluello, L.: 1965, Editor of De interpretatione vel Perihermenias translatio
Boethii in the series Aristoteles latin us, Union Academique Internationale, corpus
philosophorum medii aevii sect. II, vol. II, part 1, Desclee de Brouwer, Bruges.
Normore, Calvin: 1976, The Logic of Time and Modality in the Later Middle Ages:
The Contribution of William of Ockham', unpublished PhD. dissertation, University
of Toronto.
Normore, Calvin: .1984, 'Buridan's Ontology', in Bogen and McGuire (eds.), How
Things Are: Studies in Predication and the History and Philosophy of Science, D.
Reidel, Dordrecht, Holland, 1984, pp. 187-200.
Nuchelmans, Gabriel: 1973, Theories of the Proposition: Ancient and Medieval Con-
ceptions of the Bearers of Truth and Falsity, North-Holland, Amsterdam, 1973.
Perreiah, Alan R.: 1972, 'Buridan and the Definite Description', Journal of the History
of Philosophy X (1972), 153-160.
Pin borg, Jan: 1976, Editor of The Logic of John Buridan: Acts of the Third t-"uropean
Symposium on Medieval Logic and Semantics, Opuscula Graecolatina Vol. IX,
Museum Tusculanum, Copenhagen, 1976.
Prior, Arthur N.: 1976, The Possibly-True and the Possible' in Geach and
Kennedy (eds.), Papers in Logic and Ethics, University of Massachusetts Press,
Amherst, 1976, pp. 202-214.
Rescher, Nicholas: 1966, Galen and the Syllogism, University of Pittsburgh Press, 1966.
Reichenbach, Hans: 1947, Elements of Symbolic Logic, Macmillan, New York, 1947.
Scott, 1'. K.: 1965, 'John Buridan on the Objects of Demonstrative Science', in
Speculum XI (1965),654-673.
Scott, T. K.: 1966, John Buridan: Sophisms on Meaning and Truth (translation of the
Sophismata), AppIeton-Century-Crofts, New York, 1966.
Spade, Paul and Stump, Eleonore: 1982, 'Obligations', in Kretzmann et al. (eds.), The
Cambridge History of Later Mediaeval Philosophy, Cambridge 1982, pp. 315-340.
Trentman, John: 1970, 'Ockham on Mental', Mind LXXIX (1970), 586-590.
INDEXES
Note: The references in the Index are to section and paragraph number, not to page
number. The Index should be used in conjunction with the various cross-
references in the notes. Entries conform to the following rules for style: unla-
belled references refer to the introduction; references to the Treatise on
Supposition and the Treatise on Consequences are prefixed with the abbreviations
TS and TC; notes are indicated by the lower-case 'n'. Names appearing in
examples are not indexed.
INDEX OF NAMES
Abailard, Peter n.78 Faral, Edmond nn.2, 5

Aristotle §2, §3.1, and n.6; TS 1.2.3, Frege, Gottlob §5.3
2.3.7, 2.6.9, 2.6.33, 2.6.49, 3.2.5.
3.2.20, 3.2.34, 3.4.6, 3.4.9, 3.8.26, Geach, Peter §6.6 and nn.35, 67, 69
4.9.4, 5.2.3 (by implication), 5.3.9, Gibson, Joan n.13
5.3.12, 5.3.17 (by implication), and Gregory of Rimini §5.5
n.60; TC 1.4.12, 1.6.5, 1.8.104,
2.7.40, 3.4.9, 3.4.48, 3.7.24, 4.1.3, Hubien, Hubert §2; TC I n.51, III
4.1.6, 4.1.8, 4.2.32, 4.4.7-8, and III nn.27,52
n.37, IV n.3. Hughes, G. E. §6.9 and nn.4, 9, 77
Averroes TS 5.3.10
Kretzmann, Norman n.6
Belnap, Nuel n.38
Boethius nn.6, 78 Martin, Christopher n.78
Bolzano, B. TC I n.18 Minio-Paluello, Lorenzo n.6
Burleigh, Walter TS n.50
Normore, Calvin nn.7,28
Camap, Rudolf n.84 Nuchelmans, Gabriel n.62
Carroll, Lewis §7.1
Ockham: see William of Ockham
Davidson, Donald §6.1
De Rijk, L. M. n.3 Perreiah, Alan n.59
Donnellan, Keith n.70 Peter of Rivus (Petrus de Rivo) n.4 7
Peter of Spain (Petrus Hispanus) §2
Ebbesen, Sten n.15 and n.3; TS n.3
369
370 INDEXES
Pinborg, Jan nn.15, 17; TS nn.2, 39 Scott, T. K. nn ..4,65

Plato n.1 Spade, Paul n.80; TS n.56
Porphyry TS 4.9.4 Spinoza, Benedict de § I
Prior, Arthur n.44 Strode, Ralph n.99
Stump, Eleonore n.80; TS n.56
Quine, W. V. O. §3.2, §6.1, §6.8
Trentman, John n.12
Reichenbach, Hans §6.8
Reina, Maria Elena §2; TS nn.13, 21, VilIon, Fran"ois § I
58,72,80,88
Russell, Bertrand §3.3 William of Ockham § I, §6.2, and n.12;
TSn.50
Wodeham. Adam §5.5
INDEX OF SUBJECTS
Absolute Terms (see also Appellation, Connotation) §4.2 in toto

Additive Principle §3.3, §5.5; TS 2.3.10
Adjacence (see also Appellation) §4.2
Alienation (alienated supposition) TS 6.4 in toto, 6.5 in toto
Ampliation §6.8 in toto, §6.5, §8.4 (syllogistic); TS 4.3.2,5.2.13,6.1 in toto, 6.2 in toto,
6.4.1, 6.5.1; TC 1.6 in toto, 1.8.58-66, 1.8.73-80, 1.8.91, 1.8.97, 2.4 in toto,
2.6.8-13, 2.7.13, 3.3.6, 3.4.33, 3.4.58-72 (syllogistic), 3.5.54-72, 3.7.25-28,
4.1.2,4.3.8
Anaphora (see also Reference, Coreferentiality, Relative Term) §6.1; §6.4; TS 3.7.16-
33
Antecedent
in a consequence (see also Consequence, Consequent) TC 1.3.6-7,1.3.10-12,
1.8.4
of a relative-term (see also Relative Term) TS 4.3.-4.8 in toto
ofasyllogism TC3.4.18,4.1.12
Appellation (see also Absolute Term, Adjacence, Connotation) §4.2 in toto, §4.3 in
toto; TS 1.4 in toto, 3.4.3, 3.8.24-31, 5-6 in toto; TC 1.6.12-16, 3.7.3-13,
3.7.16
Ascent (see also Descent, Supposition) §6.7; TS 2.6.75-76
Assertion (see also Sentence) §5.1, §6.9
- conditional assertion §5.3
Bivalence §5,4; TC 1.1.7, 1.8.1

Bradley's Regress §4.2
Categorematic Term §4.1, §7.2; TS 2.3.1-5, 2.4.2, 2.4.5, 2.6.4, 3.2.8, 3.4.10; TC
1.7.2-3
Categorial-Relative Term (see also Anaphora, Antecedent, Quantity, Quantifier) TS
2.6.60,4.8 in toto
Categories §4.2, §5.5; TS 3.2.31
Collective Terms TS 2.6.64-69,2.6.88,3.8.10; TC 1.8.86,3.4.3
Comparatives §6.7; Rule DC-4 in TS 3.7.45-49, 3.8.16-17
Complexe significabile §5.5
Concepts §3.1, §3.3, §4.2
acquisition §3.2 (§3.3)
complex §3.3, §4.2; TS 2.4 in toto, Rule Sup-12 in 2.6.1,2.6.86-87
complexive §3.3, §4.1, §4,2; TS 2.3.12-13, 2.4.3, 3.4.8-12
formation §3.2; TS 1.2.3-4
mediating (see also Ratio) §3.3; TS 2.3 in toto, 3.2.8-9, 3.8.28-30, 5.3.7; TC
1.6.12-14,3.7.3-10
371
372 INDEXES
Conclusion
- direct/indirect (defined) TC 3.2.10, 3.4.35-37
Conditionals (see also Consequences, Inference-Rules) § 7.1; TC 3.1. 7, 4.1. 7
Conjugation §8.4; TC 3.4.38-52
Conjunctive Subject TS 2.6.64-67, 3.8.10
Connotation (see also Appellation) §4.2; TS 5.1.1, 5.2.11
Consequences §7 ill toto
assertoric §7.3 in toto; TC I
composite modal §7.5 ill toto; TC II
divided modal §7.4 ill toto; TC II
formal (see also Uniform Substitution Principle) §7.1; TC 1.3 ill toto, 1.4.2-4,
1.8.32-39, 1.8.58, 1.8.60-61, 1.8.65, 1.8.75, 1.8.100,3.1.9-14
material §7.1; TC 1.4.3-4
simple §7.1; TC 1.4.6
syllogistic: see Syllogism
utllllnc §7.1;TC 1.4.7-12, 1.8.6-7, 1.8.38
Consequent (see also Antecedent, Consequence) TC 1.3.6-7, 1.3.10-12, 1.8.4
Contingent TC 2.2.3, 2.6.15, 2.6.32-38, 2.7.23-25, 2.7.42-45,4.3 ill toto
Contradictories (see also Law of Non-Contradiction) §3.2, §5.4; TS 2.6.8, 2.6.69;
TC 1.1.7-10, 1.8.8, 1.8.32-33,2.6.6,2.7.39,3.7.17-19
Contraries (see also Law of Contraries) §5.4; TS 2.6.46, 2.6.68; TC 2.7.24
Converse-Entailment Principle §4.3, §6.5; TS 5.3.7-13
Conversion §5.4, §7.3; TS 2.6.21, 2.6.50, 2.6.62, 6.2.6; TC 1.8.47-52, 1.8.55, 1.8.70,
1.8.72-95,2.6.25-38,2.7.16-17,4.4.8
by contraposition TC 1.8.100-106
- with respect to the dictum §8.5; TC 2.7.18-24
- violating grammatical rules TS 2.6.21; TC 1.8.81-83, 1.8.89
Copula §4.1, §5.2, §6.8; TS 2.3.9-12, 2.4.3, 2.6.5, 2.6.15, 2.6.24, 2.6.28, 3.4.8-12,
4.9.8, 5.2.13; TC 1.6.6, 1.7.2-3, 1.8.48-49, 1.8.72-75, 1.8.79-83, 1.8.87, 2.2.3,
2.2.5,3.4.62,3.4.71,4.4.7
Coreferentiality (see also Anaphora, Doctrine of Distribution, Reference) §6.6, §8.2;
TC 1.8.89,3.4.1-10
Deduction Theorem §7.1

Definition §4.2
- nominal TS 1.4.7,2.4.1, Rule Sup-12 in 2.6.1, 2.6.86; TC 1.3.2,1.8.4,2.6.33
DeMorgan's Laws §5.3, §8.3; TC 1.8.76
Descent (see also Ascent, Supposition) §6.7; TS 2.6.75-76, 3.8.16, 3.8.22, 3.8.25,
3.8.29-30,4.6.1
de virtute sermollis TS 3.2.15-22, 3.4.7
Dictum §5.2, §7.5; TC 2.2.2-5, 2.7.4, 2.7.6-17, 2.7.25-28
- conversion with respect to §7.5, §8.5; TC 2.7.18-24
- syllogisms with respect to §8.5;TC4.1.10-12
Dictum de omlli et Ilullo (see also Distribution, Syllogism: Semantic Principles) §8.2;
TC 3.4.1-10, 3.4.56, 4.2.4, 4.2.51, 4.3.15
Disjunctive Subjects TS 2.6.59, 2.6.69-79, 6.2.3; TC 1.8.64,2.4.6,2.6.16
INDEX OF SUBJECTS 373
Disjunctive Predicates TS 2.6.62, 3.6.1, 3.8.25,6.2.4; TC 1.8.79, 1.8.82-85

Distribution (see also Coreferentiality, Dictum de omni et nullo, Supposition, Syllogism:
Semantic Principles, Quantity: Distributive Sign) TC 1.2 in toto, 1.7.2,
1.8.53-54, 1.8.56, 1.8.62, 1.8.70, 1.8.74,3.4.59,3.5.3-6,3.7.15,3.7.19-30
distributive supposition: see Supposition: Distributive
Doctrine of Distribution §8.2; TS 1.3.1, 2.4.4, 2.6.8, 2.6.36, 2.6.39, 2.6.42,
2.6.46, 2.6.79; TC 3.4.26, 3.4.29-37, 3.7.5-6, 3.7.9-11, 3.7.23, 4.2.17-18,
4.2.28,4.2.51,4.4.9
Divided Terms TS 2.6.68, 3.4.4-6
Divisibilism (Temporal) §6.8
Entailment Principle §4.3

Enthymeme TC 3.1.2-4
Equiformity §3.2, §5.2; TC 1.2.2, 1.2.5,3.7.4
Equipollence (see also Conversion) §5.4, §7.3-5; TS 2.2.12-13, 2.6.47; TC 1.6.8,
1.8.47-51, 1.8.104,2.3.3,2.5 in toto, 2.6.4-6, 2.6.9, 2.7.14, 2.7.39, 3.1.3-4,
3.7.18,4.2.1
quasi-equipollence TC 4.1.14
Expression
perfect TS 2.5.1, Rule Sup-6 in 2.6.1
imperfect TS 2.5.1, 2.5.4, Rule Sup-8, Rule Sup-9, Rule Sup-10, Rule Sup-II
in 2.6.1 .
Finite and Infinite Terms (see also Negation: Infinitizing, Syllogism: Variation) TC 3.3
in toto, 3.4.15
Form
- of a term TS 5.2.2-5; TC 1.6.7
- of a sentence TC 1.2.6, 1.4.2, 1.7.2-3
Foundation (of a relation) Rule RT-7 in TS 4.9.1, 4.9.3, 4.9.5
Impersonal Verbs see - Verbs: Impersonal

Induction TS 3.8.25; TC 3.1.2, 3.1.4
Inference-Rule (see also Conditional, Consequence) § 7.1
Infinite Term: see Finite and Infinite Terms
Infinitive Phrases Rule Sup-4 in TS 2.6.1, 2.6.49, 3.2.11-13
Intentional Verbs: see Verbs: Intentional
Law of Contraries §5.4; TS 2.6.8, 2.6.39; TC 3.1.1 °

Law of Non-Contradiction §5.4; TC 1.1.7-8, 1.8.1,2.7.24,3.1.10,4.1.4
Liar-Paradox §3.2; TC 1.5.5-7
Locus a divisione TC 1.8.34
Logical Form (see also Form: Of a Sentence, Sentence) §4.1; TC 1.7.3-4
Matter
of a term TS 5.2.3
- of a sentence TC 1.4.3, 1.7.2
374 INDEXES
Mental Language §3.1, §3.3 in toto, §6.2; TS 1.3.5-10

Method of Reduction §8.3; Theorem IIl-4 in TC 3.4.20-22
Modals TS 2.6.36, 2.6.39; TC 2.1-2 in toto
composite §5.2, §7.5 (consequences), §8.5 (syllogistic); TC 2.2.2-4, 2.7 in toto,
4.1 in toto
divided §5.2, §6.8, §7.4 (consequences), §8.6 (syllogistic); TC 2.2.5; 2.3-4 in
toto, 2.6 in toto, 4.1.2, 4.2 in toto
Mode §5.2, §7.5; TC 2.2.2-5, 2.3.1-5, 2.5.3, 2.6.7, 2.6.32, 2.6.35-38, 2.7.2-4, 2.7.7,
4.1.15
Negation §5.2; TS 2.3.9, Rule Sup-l0 in 2.6.1, 2.6.30-32, 2.6.82-84, Rule DC-5 in
3.7.50,3.8.18,4.9.8; TC 1.7.2-3,2.2.3,2.3 in toto, 2.5.2-3
negating §4.1; §6.7; TS 2.2.7, 3.7.7, Rule DC-2 in 3.7.34-37, 3.8.14
infinitizing §4.1, §6.7; TS 3.7.35, Rule DC-3 in 3.7.42-44,6.5.1-2
uncommon idiom §6.6, §8.4; TC 1.8.70,3.4.26,3.4.44,3.4.46,3.4.58
Obligations TS 3.2.19
Oblique Terms (see also Appellation, Syllogism: Oblique) §4.2; TS 1.4.4-5, Rule
Sup-3 in 2.6.1, 2.6.5-48, 2.6.81, 3.7.10-13, 3.8.3-8,4.1.5,4.2.6, Rule App-6 in
5.4.1-6; TC 1.7.3, 1.8.90-93,3.3.3-4,3.5 in toto, 3.7 in toto (syllogistic)
Ontological Commitment §6.8
Opaque contexts §4.3
Possibly-True (see also Modals: Composite) §3.2, §5.2, §7.5; TC 1.3.9-10, 1.8.26,
2.7.2
Predication, Modes of §4.2
Primary Object (of a faculty) TS 3.2.32-35
Proposition (see also Assertion, Complexe Significable, Sentence, Truth) §3.2
Quality §4.1, §5.2

Quantifiers §4.1, §6.1, §6.4, §6.7; TC 2.2.3-10
branching §6.6
- categorial (see also Categorial-Relative Quantifiers, Relative Term) §4.1, §6.4;
TS 2.6.60, 3.7.4-33; TC 3.7.19-30
Quantity (see also Categorial-Relative Term, Distribution, Quantifier) §4.1, §5.2, §5.5;
TS 2.6.55, 2.6.65-66; TC 1.7.2-3,2.2.3,2.2.6-15,2.7.5,3.7.19-30
distributive sign (see also Supposition: Distributive) TS 2.2.6-15, 2.3.9, 2.6.4,
2.6.72-74, Rule DC-I in 3.7.1-33, Rule DC-2 in 3.7.34-41, 4.4.8; TC
3.5.3-6
double distributives Rule NDC-2 in TS 3.8.1 3-15
particular sign TS 2.2.13-15, 4.4.8, Rule RT-3 in 4.5.1
Ratio (see also Appellation, Concepts: Mediating, Verbs: Intentional) §4.3; TS 3.2.27,
3.8.26-28,5.1.2,5.3 in toto; TC 3.7.5-7, 3.7.10, 3.7.13, 3.7.16, 3.7.25,4.4.5
Reduction, Method of see Method of Reduction
Reductio-Method (proof per impossibi/e) §8.3; TC Theorem III-3 in 3.4.17-19,
3.4.57,4.2.24,4.2.38,4.2.46,4.2.61,4.2.70, 4.2.73,4.3.16,4.4.10
INDEX OF SUBJECTS 375
Reduplication TC 4.4 in toto

Reference §6.1 in toto (supposition as reference)
Referential-Attributive Use §6.6
Relative-Term TS 4 in toto
Antecedent-Term (see also Antecedent) TS 4.3-4 in toto
Categorial Relative-Terms TS 2.6.60, 3.2.31, 3.7.4, 4.2.1-3, 4.4.8, 4.8 in toto,
TC 3.7.19-30
Differentiating Relative-Terms §6.4; TS 3.7.45, 3.7.50, 3.8.14,4.2.1,4.4.8,4.9
in toto
Identificatory Relative-Terms §4.1, §6.4 in toto; TS 4.2.1, 4.2.4; TC 3.4.29-30,
3.5.6,3.7.15,3.7.23,3.7.26,3.7.29,4.3.5
Remainder Principle §4.2; TS 1.4.1,5.1.1,5.2.5
Restriction TS 6.1.1, 6.3 in toto; TC 2.6.1-2, 2.6.26
Scope §4.1
Sentences (see also Assertion, Proposition, Truth) §5.1; TC 1.3.2, 1.3.5-10, 1.8.94
Categorical §5.2 in toto
- Hypothetical §5.3intoto;TC3.7.19-30
- obtaining TC 1.8.9
Square of Opposition §5.4, §6.7, §6.9, §7.3
Status (of a term) (see also Ampliation) TS 6.1.1, 6.1.4
Subalternation TC 1.8.55,2.6.25,3.1.3
Subcontraries §5.4
Supposition §6 in toto; TC 1.5 in toto
as reference §6.1 in toto
Absolute §6.4; TS 4 (by implication) in toto
Accidental §6.5; TS 3.4 in toto
Common §6.3 ff.; see Distributive and Non-Distributive supposition, Deter-
minate supposition
Confused §6.6; see Distributive and Non-Distributive Supposition
Determinate §6.6; TS 3.5 in toto; TC 1.2.7-9, 1.8.56, 1.8.92,3.5.5
Discrete §6.3; TS 3.3 in toto, 4.4.5-6; TC 3.4.5-7
Distributive (see also Distribution, Quantity: Distributive Sign) §6.7; TS 3.6-7
in toto, 4.6.1, 4.9.7-9; TC 1.2.2-6, 1.8.53-54, 1.8.56-57
Material §6.2 in toto; TS 0.0.1, 1.2.1, 1.2.6, 3.2 passim, 3.8.30, 4.4.3-4; TC
2.7.5,3.7.6,3.7.8,3.7.10,4.4.5
Natural §6.5 in toto; TS 3.4 in toto
Non-Distributive confused (Merely Confused) §6.7 in toto; TS 3.6 in toto, 3.8
in toto; TC 1.8.53-54, 1.8.56-57
Personal §6.2 in toto; TS 1.2.1-2, 1.2.7-9,3.2 in toto
Simple TS 3.2.4-6
Syllogism §8.1; TC 3.1.17, 3.2 in toto, 3.4.11-14, 3.4.48
Assertoric §8.4; TC 3.4.37-57
Composite Modal §8.5; TC 4.1 in toto
Divided Modal §8.6; TC 4.2 in toto
Expository §8.2; TS 3.8.26, 5.3.10; TC 1.8.79, 2.6.24, 2.7.11,3.3.5,3.4.23-27,
4.2.9,4.2.24,4.2.61,4.3.61,4.4.10
376 INDEXES
Oblique §8.4; TS 2.6.9-17, 2.6.40-43, 3.7.12; TC 3.3.3-4, 3.5 in toto, 3.6.1-

3.7.30
Proof Procedure §8.3; TC 3.4.57
Semantic Principles (see a/so Dictum de omni et nul/a, Distribution) §8.2; TC
3.4.1-10,3.4.24,3.4.29-30,3.4.56,3.5.6,3.7.5-6, 3.7.9-11, 3.7.23, 4.2.4
Temporal §8.4; TC 3.3.6, 3.5.54-72
Variation §8.4; TC 3.7.31-45
Syllogistic Extremes and Middle §8.1; TC 3.2.2-4, 3.4.1-10, 3.4.29-30, 3.6 in toto,
4.4.7
Syllogistic Figure §8.1; TC 3.2.4-11
Syncategorematic Terms (see a/so Negation, Quantity) §4.1; TS 2.3.1, 2.3.6. 2.4.2-4,
2.6.4, Rule Sup-2 in 2.6.1, 2.6.23,2.6.33,3.4.10,3.7.21,3.7.24; TC 1.7.2-3, 1.8.94,
3.1.3, 3.1.12-13
Terminus (of a relation) Rule RT-7 in TS 4.9.1,4.9.3,4.9.5-6

Transcendental Terms §4.2
Transparent Context §4.3
Truth (see a/so Sentence) TS 1.3.3-8,2.6.21-22; TC 1.1 in toto
- causes TC 1.2 in toto, 1.8.40-49
Truth-Conditions §5.3, §5.4, §5.5, §6.9; TC 1.1 in toto, 1.5 in toto, 1.6.17,2.4.4
Truth-Entailment Principle §6.9
Uniform Substitution Principle (see a/so Consequence: Formal) §7.2. §8.2; TC 1.4.2-3
Universals TS 3.2.4-5, 3.2.9, 3.2.28-29
Use-Mention §6.2
Verbs
impersonal TS 1.3.8,2.1.3;TC 1.8.72, 1.8.87-88
intentional (see a/so Appellation, Concepts: Mediating, Ratio) §4.3 in toto; TS
3.8.24-31, 5.3.1-8, Rule Amp-5 in 6.2.1: TC 1.6.12-16,3.7.3-10,3.7.25,
4.1.15-16
Verification TS 1.3 in toto, 2.2.14, Rule RT-I in 4.3.1,6.4.3; TC 3.4.30
'What is' locution §8.6; TC 1.8.60, 1.8.64, 1.8.75-79. 1.8.82, 2.6.1-3, 2.6.30, 3.4.12.
3.4.54-56,4.1.2.4.2.15-25.4.2.30.4.2.52
INDEX OF RULES AND THEOREMS
Rules of Supposition:
Rule Sup-I: TS 2.6.1, 2.6.3
Rule Sup-2: TS 2.6.1, 2.6.4
Rule Sup-3: TS 2.6.1, 2.6.5-48
Rule Sup-4: TS 2.6.1, 2.6.49
Rule Sup-5: TS 2.6.1, 2.6.50
Rule Sup-6: TS 2.6.1, 2.6.51
Rule Sup-7: TS 2.6.1, 2.6.52
Rule Sup-8: TS 2.6.1, 2.6.53
Rule Sup-9: TS 2.6.1, 2.6.54-81
Rule Sup-l0: TS 2.6.1, 2.6.82-84
Rule Sup-II: TS 2.6.1,2.6.85
Rule Sup-12: TS 2.6.1, 2.6.86
Rule Sup-13: TS 2.6.1, 2.6.87
Rule Sup-14: TS 2.6.1, 2.6.88
Rules for Distributive Confused Supposition:

RuleDC-l:TS 3.7.1, 3.7.3-33
Rule DC-2: TS 3.7.34-41
Rule DC-3: TS 3.7.42-44
Rule DC-4: TS 3.7.45-49
Rule DC-5: TS 3.7.50-51
Rules for Non-Distributive Confused Supposition:

Rule NDC-l: TS 3.8.1-12
Rule NDC-2: TS 3.8.13-18
Rule NDC-3: TS 3.8.19-23
Rule NDC-4: TS 3.8.24-31, 5.3.2
Rules for Relative-Terms:

Rule RT-l: TS 4.3.1-2, 4.4.1
Rule RT-2: TS 4.4.1-2, 4.4.9
Rule RT-3: TS 4.5.1-5
Rule RT-4: TS 4.6.1-4
Rule RT-5: TS 4.7.1-2
Rule RT-6: TS 4.8.1-3
Rule RT-7: TS 4.9.1
Rules of Appellation:
Rule App-l: TS 5.2.1-6
377
378 INDEXES
Rule App-2: TS 5.2.1, 5.2.7, 5.2.10-13

Rule App-3: TS 5.2.1, 5.2.8-9, 5.2.13
Rule App-4: TS 5.3.1
Rule App-5: TS 5.3.1
Rule App-6: TS 5.4.1,5.4.4-6
Rule App- 7: TS 5.4.1,5.4.7
Rule App-8: TS 5.4.1, 5.4.7
Rule App-9: TS 5.4.1, 5.4.7
Rules of Ampliation:
Rule Amp-I: TS 6.2.1
Rule Amp-2: TS 6.2.1
Rules of Restriction:
Rule Res-I: TS 6.3.1
Rule Res-2: TS 6.3.1
Theorems for Assertoric Consequences:

Theorem I-I: TC 1.8.3-5, 1.8.27
Theorem 1-1 *: TC 1.8.6-7
Theorem 1-2: TC 1.8.8-11
Theorem 1-3: TC 1.8.12-15, 1.8.39.2.6.19,2.6.21, 2.6.29, 2.7.20, 3.4.18
Theorem 1-4: TC 1.8.16-20, 1.8.34,2.6.25,2.6.31,3.4.22,4.3.11, 4.3.13
Theorem 1-5: TC 1.8.21-24,2.7.19
Theorem 1-6: TC 1.8.29-30
Theorem 1-7: TC 1.8.32-37
Theorem 1-7*: TC 1.8.38-39
Theorem 1-8: TC 1.8.40-47, 1.8.54, 1.8.59
Theorem 1-9: TC 1.8.48-52, 1.8.70, 1.8.74
Theorem 1-10: TC 1.8.53-55,3.4.36
Theorem 1-11: TC 1.8.56-57
Theorem 1-12: TC 1.8.58-63,4.2.51
Theorem 1-13: TC 1.8.66-68
Theorem 1-14: TC 1.8.69-95, 2.7.17
Theorem 1-15: TC 1.8.96-99,2.7.17
Theorem 1-16: TC 1.8.100-106
Theorem 1-17: TC 1.8.107-110
Theorems for Modal Consequences:

Theorem II-I: TC 2.6.4-7
Theorem 11-2: TC 2.6.8-13
Theorem 11-3: TC 2.6.14-17, 2.6.19-20
Theorem 11-4: TC 2.6.18-21, 2.6.24, 2.6.29
Theorem II-5: TC 2.6.23-26, 2.6.31, 2.6.38
INDEX OF RULES AND THEOREMS 379
Theorem 11-6: TC 2.6.27-29

Theorem 11-7: TC 2.6.32-34, 4.3.1
Theorem 11-8: TC 2.6.36-38
Theorem 11-9: TC 2.7.6-9, 2.7.17
Theorem II -10: TC 2.7.10-15
Theorem 11-11: TC 2.7.16-17
Theorem 11-12: TC 2.7.18-20, 2.7.22
Theorem 11-13: TC 2.7.21-22
Theorem 11-14: TC 2.7.23-24
Theorem 11-15: TC 2.7.25-26, 4.1.14
Theorem 11-16: TC 2.7.27-30
Theorem 11-17: TC 2.7.34-37, 2.7.39
Theorem 11-18: TC 2.7.38-41
Theorem 11-19: TC 2.7.42-46
Theorems for Assertoric Syllogistic:

Theorem III-I: TC 3.4.11-14
Theorem III-2: TC 3.4.15-16, 3.4.37
Theorem III-3: TC 3.4.17-19, 3.4.57
Theorem III-4: TC 3.4.20-22
Theorem III-5: TC 3.3.5, 3.4.23-27
Theorem III-6: TC 3.4.29-30, 3.4.32, 3.7.32
Theorem III-7: TC 3.4.31-32, 3.7.32
Theorem III-8: TC 3.4.35-36, 3.7.32
Theorem III-I 0: TC 3.4.58-62, 3.4.70
Theorem III-II: TC 3.4.63-64, 3.4.70
Theorem III-I 3: TC 3.7.8-12
Theorem 111-16: TC 3.7.19-25
Theorem III-18:TC 3.7.31-41
Theorems for Modal Syllogistic:

Theorem IV-I: TC 4.1.10-12
Theorem IV-2: TC 4.1.13-14
Theorem IV-4: TC 4.2.3-4, 4.2.7, 4.3.15
Theorem IV-6: TC 4.2.8-13, 4.2.33, 4.3.16
Theorem IV-10: TC 4.2.26-30, 4.3.18
380 INDEXES
Theorem IV-II: TC 4.2.31-34

Theorem IV-12: TC 4.2.35-39, 4.2.54
Theorem IV-21: TC 4.3.10-11, 4.3.13
Theorem IV-28: TC 4.4.1, 4.4.9-14

Burridan - Treatises

Transféré par

Informations du document

Titre original

Copyright

Formats disponibles

Partager ce document

Partager ou intégrer le document

Options de partage

Avez-vous trouvé ce document utile ?

Ce contenu est-il inapproprié ?

Droits d'auteur :

Formats disponibles

Burridan - Treatises

Transféré par

Droits d'auteur :

Formats disponibles

JEAN BURIDAN'S LOGIC

THE TREATISE ON SUPPOSITION

THE TREATISE ON CONSEQUENCES

TEXTS AND STUDIES IN THE HISTORY OF

LOGIC AND PHILOSOPHY

N. KRETZMANN, Cornell University

J. BERG, Munich Institute of Technology

F. DEL PUN T A, Linacre College, Oxford

J. E. MURDOCH, Harvard University

THE TREATISE ON SUPPOSITION

Translated, With a Philosophical Introduction

D. REIDEL PUBLISHING COMPANY

DORDRECHT / BOSTON / LANCASTER /TOKYO

(Synthese historical library ; v. 27)

Published by D. Reidel Publishing Company,

Sold and distributed in the U.S.A. and Canada

In all othcr countries, sold and distributed

All Rights Reserved

Softcover reprint of the hardcover 1st edition 1985

- Buridan, MS B.N.lat.l6128 fo1.232 VO

INTRODUCTION. BURIDAN'S PHILOSOPHY OF LOGIC

Section 1. John Buridan: Life and Times 3

TRANSLATION. THE TREATISE ON SUPPOSITION

Chapter 4. The Supposition of Relative Terms 148

TRANSLATION. THE TREATISE ON CONSEQUENCES

Book I. Consequences in General and Among Assertoric

2.2. Composite and Divided Modal Sentences 228

Notes. Buridan's Philosophy of Logic 327

Buridan was a brilliant logician in an age of brilliant logicians, sensitive

BURIDAN'S PHILOSOPHY OF LOGIC

Buridan is best-known to philosophers for the example of "Buridan's

The "Treatise on Supposition" [TS] is the fourth treatise of a much

We divide this work into nine treatises, of which the first

"Our author" is Peter of Spain,3 and "his book" is the Summulae

Wishing to say certain general things about the whole of

The Summulae de dialectica is written as a commentary on Peter of

3.1. LEVELS OF LANGUAGE

Buridan and other logicians of the fourteenth century were inspired by a

3.2. NOMINALIST SEMANTICS AND EQUIFORMITY

Logic, for Buridan, will include Aristotelian psychology as a component

Buridan's pragmatic interpretation also allows him to deal with

3,3. MENTAL AS AN IDEAL LANGUAGE

Mental Language is of the utmost importance for Buridan's logic.

[4] Mental is nonredundant.

Ad [1]. The universality of Mental is a matter of its structure, not its

Mental an inscription or utterance corresponds to. We may call

no synonyms. Suppose that 'vixen' and 'female fox' are in fact

Is there such a discriminatory medium for Mental? Buridan thinks

4.1. SYNCATEGOREMATIC TERMS

To have a theory of logical form, we have to distinguish logical from

In addition to such typical syncategoremata, there are also exceptives

4.2. ABSOLUTE AND APPELLATIVE TERMS

Purely categorematic terms have ultimate signification and do not imply

concept 'female fox' to which the inscription or utterance 'vixen' is sub-

If a term signifies something which it does not stand for,

A term appellates that which it connotes as in some way ad-

'Connotation' is a semantic relation, a kind of indirect or oblique. form

how one thing is related to another, and so forth, are taken

of definition was in that definition provides a generally reliable guide to

4,3. INTENTIONAL VERBS

One of the key uses of the doctrine of appellation is to analyze the

stars, though Socrates does not know which, The Converse-Entailment

5.1. SENTENCES AS ASSERTIONS