(Bloomsbury Studies in Philosophy) Delbert Reed-Origins of Analytic Philosophy - Kant and Frege-Bloomsbury Academic (2008)

Origins of Analytic Philosophy
Continuum Studies in Philosophy

Series Editor: James Fieser, University of Tennessee at Martin, USA
Continuum Studies in Philosophy is a major monograph series from
Continuum. The series features first-class scholarly research monographs
across the whole field of philosophy. Each work makes a major
contribution to the field of philosophical research.
Aesthetic in Kant, James Kirwan
Analytic Philosophy: The History of an Illusion, Aaron Preston
Aquinas and the Ship of Theseus, Christopher Brown
The Challenge of Relativism, Patrick Phillips
Demands of Taste in Kant's Aesthetics, Brent Kalar
Descartes and the Metaphysics of Human Nature, Justin Skirry
Descartes' Theory of Ideas, David Clemenson
Dialectic of Romanticism, Peter Murphy and David Roberts
Hegel's Philosophy of Language, Jim Vernon
Hegel's Philosophy of Right, David James
The History of Intentionality, Ryan Hickerson
Kierkegaard's Analysis of Radical Evil, David A. Roberts
Leibniz Reinterpreted, Lloyd Strickland
Metaphysics and the End of Philosophy, H. O. Mounce
Nietzsche and the Greeks, Dale Wilkerson
Origins of Analytic Philosophy, Delbert Reed
Philosophy of Miracles, David Corner
Platonism, Music and the Listener's Share, Christopher Norris
Popper's Theory of Science, Carlos Garcia
Role of God in Spinoza's Metaphysics, Sherry Deveaux
Rousseau and the Ethics of Virtue, James Delaney
Rousseau's Theory of Freedom, Matthew Simpson
Spinoza and the Stoics, Firmin DeBrabander
Spinoza's Radical Cartesian Mind, Tammy Nyden-Bullock
St. Augustine and the Theory of Just War, John Mark Mattox
St. Augustine of Hippo, R.W. Dyson
Thomas Aquinas & John Duns Scotus, Alex Hall
Tolerance and the Ethical Life, Andrew Fiala

Kant and Frege
Delbert Reed
continuum
Continuum Logo
Continuum International Publishing Group

The Tower Building
11 York Road
London SE1 7NX
80 Maiden Lane
Suite 704
New York
NY 10038
www.continuumbooks.com
Delbert Reed 2007
All rights reserved. No part of this publication may be reproduced or transmitted in any form
or by any means, electronic or mechanical, including photocopying, recording, or any
information storage or retrieval system, without prior permission in writing from the
publishers.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library.
ISBN-10: HB: 0-8264-9337-8
ISBN-13: HB: 978-0-8264-9337-8
Library of Congress Cataloging-in-Publication Data
Reed, Delbert, 1940Origins of Analytic Philosophy: Kant and Frege / Delbert Reed.
p. cm.
ISBN 978-0-8264-9337-8
1. Analysis (Philosophy) 2. Kant, Immanuel, 1725-1804. 3. Frege, Gottlob,
1848-1925. I. Title.
B808.5.R423 2008
146'.4 dc22
2007020441
Typeset by BookEns Ltd, Royston, Herts.

Printed and bound in Great Britain by Biddies Ltd, King's Lynn, Norfolk
This book is dedicated with love and affection to Betsy
This page intentionally left blank
Contents
Acknowledgements
ix
Abbreviations
xi
Introduction
Part 1 Logic Old and New

1 Kant's Logic
2 Frege's Logic
23
33
Part 2 Kant on Concepts, Intuitions and Arithmetic

3 Kant on Concepts, Intuitions and Arithmetic
67
Part 3 Analyticity and Fruitful Definitions

4 Analyticity, Generality and Content
89
Part 4 From Substance to Object

Introduction to Part 4
5 Leibniz and Kant on Substance and Relations
6 The Context Principle and Numbers as Logical Objects
7 Two Views on Existence
8 Leibniz, Kant and Frege on the A Priori
9 Frege on Logic and Objectivity
123
125
139
157
172
183
Bibliograpahy
195
Index
199
Acknowaledgements
Writing is a lonely endeavour. This is especially true in the absence of a

forum where one can share ideas with colleagues with similar philosophical
interests. This problem has been acute for me, given my decision not to
pursue an academic career after gaining my doctorate in philosophy from
the University of Minnesota in 2000. Without departmental colleagues
and without regular attendance at philosophical conferences, I have not
had the opportunity to present my work to the critical scrutiny of others
interested in the origins of early analytic philosophy. I hope this work has
not suffered too much as a result.
Nevertheless, I have received some feedback and encouragement on
certain issues discussed in this book and I would like to take this
opportunity to thank those who were kind enough to provide it.
In 1988, while a graduate student at the University of Minnesota, I sat
in on a class on the philosophy of Gottlob Frege by Tony Anderson. It
gave me my first real glimpse of the Frege beyond the essays 'On Sense and
Reference' and 'Thoughts': works I had studied because of my interest in
the philosophy of language. In particular I received my first exposure to
the primary elements of Frege's logicist programme and how his
distinction between sense and reference was connected to this programme.
In 1993, just as I was beginning to write my dissertation, Erich Reck
came to the University of Minnesota on a post-doctoral fellowship. We had
a series of regular meetings to discuss Frege, as well as some of the current
literature on Frege, during his nine-month stay. These discussions were
extremely helpful in forming my early understanding of Frege.
In 1996 I read a paper at the central division meeting of the American
Philosophical Association entitled 'Reason and Objectivity in Frege's
Foundations of Arithmetic'; Hans Suga, the commentator, and Thomas
Ricketts, the chair of the session, both gave me helpful feedback and
encouragement, as did Robert Hanna. Professor Ricketts was also kind
enough to send me some copies of some of his more recent articles and
Professor Hanna later sent me some written comments on sections of my
dissertation dealing with Kant with which I had provided him.
Acknowledgements
The two men whose writings have most shaped my view of Frege have
been Tyler Burge and Michael Dummett. This is particularly true of the
writings of Tyler Burge. I have had the fortune to correspond briefly with
Professor Burge through e-mail while writing my dissertation and to
discuss Frege with him on two different occasions when he visited the
University of Minnesota to read a paper.
Closer to home I would like to thank Norm Dahl and Stephen Donaho
for recent discussions on the topic of my book along with reassurances that
the project I was engaged in was worth pursuing. I would also like to
thank the members of my dissertation committee, Betty Belfiore, Joe
Owens, Mischa Penn and John Wallace, with special thanks to my adviser
Sandra Peterson. I thank Geoffrey Hellman and the members of a seminar
on the philosophy of mathematics whom Geoffrey taught in the early
1990s for feedback on a paper on Frege's context principle that I presented
to the seminar. I would also like to thank members of the Greek Reading
Group, Jim Dankert, Heidi Lee, Mike Tiffany, Pete Wahlstrom and
especially Bob Skovbroten for philosophical feedback and moral support. I
owe a special thanks to Bill Magdalene whose unsparing philosophical
feedback and keen editorial eye has helped make this a clearer more
coherent work than it otherwise would have been. I thank Jim Feiser, a
series editor for Continuum, for originally suggesting that I write this book,
as well as the other editors at Continuum for their assistance and patience.
Finally, I thank my wife Betsy who endured yet another round of
dissertation widowhood with the same patience and good nature as she
endured the first.
Note on translations: I have used the translation of Kemp-Smith with one
central exception: I have followed the translation of the Critique by Guyer
and Wood and have translated the word 'Erkenntnis' and its cognates asaaaaa
'cognition' instead of'knowledge'. I have used two different translations of
Kant's logic. The translation by Hartmann and Schwartz is labelled
'Logic'] that by Young is labelled 'Jdsche Logic'.
The author is grateful to the following publishers for permission to quote
from copyrighted material:
Blackwell Publishing and Northwestern University Press for Gottlob
Frege, Foundations of Arithmetic, 1968.
Oxford University Press for Gottlob Frege, Conceptual Notation and Related
Articles, 1972.
Abbreviations
Bdf
BAL
CCL
CCK
CN
CP
EP
FA
FPL
FPM
GB
L
NE
PE
PMC
PPL
PW
Frege, Be griffsschraaaaaaaaaaaa
Gottlob Frege, Basic Laws of Arithmetic: Exposition of the System.
Nicholas Jolley (eel.), The Cambridge Companion to Leibniz.
Paul Guyer, (eel.),AA
Gottlob Frege, Conceptual Notation and Related Articles.
Gottlob Frege, Collected Papers on Mathematics, Logic and
Philosophy.
APaul Edwards (ed.), The EAncyclop
Gottlob Frege, Foundations of Arithmetic.
Michael Dummett, Frege: Philosophy of Language.
Michael Dummett, Frege: Philosophy of Mathematics.
Gottlob Frege, Translations from the Philosophical Writings of
Gottlob Frege.
Gottfried L
Gottfried Leibniz, New Essays on Human Understanding.
Gottfried Leibniz, Philosophical Essays.
Gottlob Frege, Philosophical Mathematical Corre
Gottfried Leibniz, Philosophical Papers and Letters.
Gottlob Frege, Posthumous Writings.
Introduction
Gottlob Frege (1848-1925) is considered to be one of the founding fathers

of analytic philosophy. The logical notation he developed in his
Be griffsschrift, published in 1879, has secured his place as the founder ofAAAAAA
modern logic, providing philosophers, logicians and linguists with a
powerful tool with which to analyse the logical structure of natural
language. Indeed, the 'rediscovery' of Frege's work in the 1950s
corresponded to the rise of philosophy of language as a distinct discipline
in which the philosophical study of ordinary language and its relationship
to logic is pursued for its own sake. At Oxford in the 1950s Austin's work in
'ordinary language philosophy', as well as the work of philosophers such as
Peter Geach, Paul Grice and P. F. Strawson, brought a concern with the
relationship between logic and ordinary language to the forefront of
philosophical interest. The work of Geach1 and Strawson,2 in particular,
has shown the strong influence of Frege. Beginning in the 1940s the
American logician Alonzo Church stressed the importance of Frege's
distinction between sense and reference and started developing a
conception of intensional logic based upon his study of Frege.3 In the
1960s American philosophers such as W. V. O. Quine4 and Donald
Davidson,5 both influenced by Frege, Russell and the logical positivists,
engaged in extensive explorations of the relationship between science,
philosophy, logic and ordinary language. In 1973 Michael Dummett's
monumental book Frege: Philosophy of Language was published. In it
Dummett argues that it was Frege who first made the linguistic turn that
forms the basis of modern analytic philosophy. More recently 'direct
reference' theorists such as Saul Kripke6 and 'description theorists' such as
Dummett 7 have debated whether a proper name must have both a sense
and a reference in order to perform its role properly. Frege's exploration of
demonstratives in the essay 'Thoughts' is the starting point of recent work
on demonstratives.8 Frege's logical calculus, his definition of concepts as
functions whose values are truth-values, his distinction between sense and
reference, his context principle (a word has meaning only within the
context of a sentence) and compositionality principle (the meaning of a

sentence is a function of the meanings of the words that compose it) have
all played centre stage in the development of analytic philosophy.
This is a work in the history of early analytic philosophy, where analytic
philosophy is exemplified by the work of the above-listed philosophers. For
despite the disparate views espoused by them, there is a central core of
goals, interests and problems revolving around the relationship between
modern predicate logic, natural language, the world and our knowledge of
the world that merits the label 'analytic philosophy'. Given the influence
that Frege has had on analytic philosophy, it is my hope that a better
understanding of Frege's relationship to his philosophical predecessors can
lead not only to a better understanding of his philosophical achievements
but also to a better understanding of the relationship of analytic
philosophy to its past. This, finally, can contribute to a better understanding of the nature of analytic philosophy itself.
Recent work in the history of analytic philosophy has shown that it
developed largely in reaction to Immanuel Kant's (17241804) critical
philosophy as first expounded in the Critique of Pure Reason (henceforth
Critique), first published in 1781 with a second edition produced in 1787.
Kant's goal in the Critique is to answer the question: 'How are a priori
synthetic judgements possible?' (B19). Empiricists such as David Hume
(171176) were right in claiming that all cognition begins with experience
but wrong in claiming that cognition arises only out of experience.
Rationalists such as Rene Descartes (1596-1650) were correct in their
claim that the exercise of reason is a necessary condition for the acquisition
of knowledge, but they failed to recognize that intuition is also necessary.
Kant's notion of synthetic a priori judgement was designed to meet just
these deficiencies in his predecessors' accounts. He agreed with the
rationalists that a priori reason is a necessary condition for cognition and
with the empiricists that all cognition begins with experience. If, however,
all cognition rests upon a priori synthetic judgements, then cognition does
not arise from experience alone nor is it based solely upon principles of
pure reason. Understanding and sensibility are both necessary for
cognition to be possible.
As Robert Hanna explains, the analytic tradition emerged from Kant's
philosophy in the sense that its members were able to define and legitimate
their views only by means of an intensive, extended engagement with, and
partial or complete rejection of the first Critique' (Hanna 2001, 5). Given
Frege's role as one of the founding fathers of analytic philosophy it should
come as no surprise that in his early works especially, he explicitly
Introduction
confronts a central tenet of Kant's critical philosophy: the claim that all
knowledge is grounded in synthetic a priori judgements.
With the development of his new logical language, Frege believed that
he had the means to expand the scope and power of conceptual thought
beyond the limits imposed by Kant's critical philosophy. In particular he
believed that the understanding could acquire knowledge of objects such
as numbers, which transcend the conditions of space and time. In this
respect, Frege was returning to a conception of the power and scope of
pure reason similar to the views of Gottfried Leibniz (16461716), and
Christian Wolff (16791754). Leibniz was a German philosopher,
mathematician, logician, scientist and historian. He is credited, along
with Isaac Newton, with developing the calculus. A gifted logician,
Leibniz prefigured many of the developments in algebraic logic made by
the British mathematician and logician George Boole (181564) over a
century and a half after Leibniz's death. Leibniz also made significant
advances in modal logic, laying the groundwork for modern possible world
semantics.
Wolff was an influential German rationalist philosopher and proponent
of Enlightenment thought. He was a mathematician by training and
developed an interest in philosophy because he believed it needed more
precise and rigorous foundations. Influenced by Leibniz, with whom he
carried on a 12-year correspondence, Wolff developed a rigorous
philosophical system modelled after the method of mathematics. Beginning
with a few simple self-evident axioms, he sought to build a philosophical
system by deriving ever more complex propositions from these axioms
using strict rules of inference based upoAn principles of pure reason.
Both of these precursors to Kant believed that knowledge could be
advanced solely through the employment of conceptual reasoning.
However, Leibniz and Wolff, like Kant, adhered to a traditional
conception of logic. And this logic had limited applicability to the
propositions of arithmetic. While Leibniz worked hard to develop a logical
language that could be used to justify the truth of mathematical
propositions, he met with limited success, primarily due to his adherence
to the traditional subject and predicate form of judgement. While he
recognized that arithmetical propositions are intrinsically relational he was
unable successfully to reduce such propositions to the categorical
propositions of traditional logic, a necessary step, in his eyes, to the
ultimate justification of the truth of any proposition.
Kant and Frege: An Initial Comparison

Kant
Kant introduces the distinction between the a priori and the a posteriori in
the opening pages of the Critique:
In what follows, therefore, we shall understand by a priori cognition, not
cognition independent of this or that experience, but cognition
absolutely independent of all experience. Opposed to it is empirical
cognition, which is cognition possible only a posteriori, that is, through
experience. (B2-3)
Kant first uses the term 'cognition' (Erkenntnis) to express the distinction
between the a priori and the a posteriori. However, he also uses the terms
'proposition' (Satz) and 'judgement' (Urteil) to express the distinction. Thus, a
cognition, a proposition and a judgement can be either a priori or a posteriori.
In what follows the distinction will mainly be discussed in terms of judgement.
In a passage that will be discussed in more detail below, Kant claims
that: 'A judgement is the representation of the unity of the consciousness of
various representations, or the representation of their relation so far as they
constitute a concept' (Jdsche Logic 17, 597). A judgement has both matter
and form. The matter of a judgement is the concepts designated by the
subject and predicate terms that compose the sentence that expresses the
judgement. The form of a judgement is signified by how the copula 'is'
combines the predicate concept to the subject concept.
Kant distinguishes a priori judgements from a posteriori judgements in
terms of how they are justified. If the truth of a judgement is necessary and
strictly universal, then it is a priori. An a priorijudgement is one whose truth
is justified absolutely independent of all experience. If the truth of a
judgement holds with only assumed or comparative universality, then it is
a posteriori. The truth of an a posteriori judgement can only be justified
throughA experience.
In contrast, analytic judgements are distinguished from synthetic
judgements in terms of the relationship between the concepts that make up
the content of the judgement. As Kant explains:
In all judgements in which the relation of a subject to the predicate is
thought (I take into consideration affirmative judgements only, the
subsequent application to negative judgements being easily made), this
Introduction
relation is possible in two different ways. Either the predicate B belongs

to the subject A as something which is (covertly) contained in the
concept A; or B lies outside the concept A, although it does indeed stand
in connection with it. In the one case I entitle the judgement analytic,
in the other synthetic. (A6/B10)
Thus the judgement that all bodies are extended is analytic because
extension is part of what it is to be a body while the judgement that this
body is heavy is synthetic. Since not all bodies are heavy (e.g. a helium
balloon) a proper analysis of the concept of body will not yield the concept
of weight.
Since Aristotle first developed the subject, says Kant, 'logic has not been
able to advance a single step, and is thus to all appearance a closed and
completed body of doctrine' (Bviii). According to Kant:
The sphere of logic is quite precisely delimited; its sole concern is to give
an exhaustive exposition and a strict proof of the formal rules of all
thought, whether it be a priori or empirical, whatever be its origin or its
object, and whatever hindrances, accidental or natural, it may
encounter in our minds. (Bix)
Logic is able to give an exhaustive exposition of its subject matter only
because it abstracts 'from all objects of cognition and their differences,
leaving the understanding nothing to deal with save itself and its form'
(Bix). For Kant, logic is completely general because it is formal. Logic
possesses no content of its own, providing the understanding only with the
forms of thought by which the content provided through intuition can be
ordered and understood.
The version of traditional logic that Kant accepted as complete
consisted in the following basic elements. To begin with, the basic form of
judgement in traditional logic is categorical (S is P), where 'S' signifies a
subject concept, 'P' a predicate concept and 'is' signifies the copula that
connects the two. Truth is defined in terms of concept-containment, where
a judgement is true just in case the predicate concept is contained in the
subject concept. Thus the judgement that every human is an animal is true
because the concept of animality is contained in the concept of humanity.
Kant, along with the tradition he embraced, posited three rules of
inference corresponding to three forms of judgement: the categorical (S is
P), the hypothetical (if Q, tnen ^)> and the disjunctive (Qor R). However,AAAA
for Kant and the logical tradition that he accepted the categorical form of
judgement was considered to be the most basic. Categorical judgements

are the basic building-blocks of Aristotle's syllogistic theory. A syllogism
consists of two categorical sentences as premises and one categorical
sentence as a conclusion: Every human is an animal. Every animal is a
substance. Therefore every human is a substance. A syllogism is a proof
through the exchange of terms designating concepts. It is because the
concept of humanity falls under the concept of animality that it falls under
the concept of substance. In summary then, according to Kant, the basic
logical form of judgement is categorical. Truth is defined in terms of
concept-containment. The basic method of proof is the syllogism, which is
proof through the exchange of terms designating concepts.
In an analytic judgement, the role that a concept plays in determining its
truth depends upon its place in a species/genus tree. Thus the concept of
humanity is a node in the tree of substance. A substance is either corporeal
or incorporeal. A corporeal substance is either animate or inanimate and an
animate substance is either sensible or insensible. An animal is a sensible
substance and animals are either rational or irrational. A human is a
rational animal. Each step down the tree is a result of applying a differentia
to a genus in order to form a species. Thus the species of man is determined
by applying the differentia rationality to the genus of animality.
All analytic judgements are a priori according to Kant. However, not
all a priori judgements are analytic. Kant believed that the mathematical
and natural sciences as well as metaphysics were all grounded in a priori
synthetic judgements. It is a central goal of the Critique to explain and
justify the possibility of an a priori synthetic science of metaphysics.
A paradigm example of an a priori synthetic science for Kant is
mathematics. Since mathematical propositions follow with strict necessity,
'which cannot be derived from experience' (B14), they are a priori, and since
mathematical propositions do increase our knowledge of the world, they
must be synthetic. As Kant explains in the opening pages of the Critique:
We might, indeed, at first suppose that the proposition 7 + 5 = 12 is a
merely analytic proposition, and follows by the principle of contradiction from the concept of a sum of 7 and 5. But if we look more closely
we find that the concept of the sum of 7 and 5 contains nothing save the
union of two numbers into one, and in this no thought is being taken as
to what that single number may be which combines both. The concept
of 12 is by no means already thought in merely thinking this union of 7
and 5; and I may analyse my concept of such a possible sum as long as I
please, still I shall never find 12 in it. We have to go outside these
Introduction
concepts, and call in the aid of the intuition which corresponds to one of
them, our five fingers, for instance, or, as Segner does in his Arithmetic,
five points, adding to the concept of 7 unit by unit, the five given in
intuition. For starting with the number 7, and for the concept of 5
calling in the aid of the fingers of my hand as intuition, I now add one
by one to the number 7 the units which I previously took together to
form the number 5, and with the aid of that figure [the hand] see the
number 12 come into being. That 5 should be added to 7, I have indeed
already thought in the concept of a sum = 7 + 5, but not that this sum
is equivalent to the number 12. Arithmetical propositions are therefore
always synthetic. This is still more evident if we take larger numbers.
For it is obvious that, however we might turn and twist our concepts,
we could never, by the mere analysis of them, and without the aid of
intuition, discover [the number that] is the sum. (B156)
For Kant, the truth of an analytic judgement can be justified solely from
the definition of its constituent concepts and through its reduction to the
principle of contradiction. In denying that the proposition 7 + 5 = 12 is
analytic Kant is breaking with the views of his rationalist predecessors.
Before looking at this passage in more detail it will help to examine briefly
the views of one of Kant's contemporaries concerning the relationship
between logic and mathematics.
According to Moses Mendelssohn (172986), in his prize-winning essay
on the question of the relationship between philosophy and mathematics
(Kant took secondA place in the contest):

The certainty of mathematics is based upon the general axiom that
nothing can be and not be at the same time. In this science each
proposition such as, for example, 'A is B', is proven in one of two ways.
Either one unpacks the concepts of A and shows 'A is B', or one unpacks
the concept of B and infers from this that not-B must also be not-A. Both
types of proof are thus based upon the principle of contradiction, and
since the object of mathematics in general is magnitude and that of
geometry in particular extension, one can say that in mathematics in
general our concepts of magnitude are unpacked and analysed, while in
geometry in particular our concepts of extension are unpacked and
analysed. (Philosophical Writings, 257)
Mendelssohn believed that all mathematical propositions could be reduced
to categorical propositions. The truth of these propositions, in turn, could be
justified by unpacking the meaning of their subject and predicate concepts

through analysis and then reducing the proposition to explicit identities
through definition and the interchange of equivalent terms in a proof.
Now any logic that hopes to give any kind of definition of the numbers
must be able to generate an infinity of objects. The logic that Kant
believed was complete and in no need of substantial improvement was a
monadic logic, where a predicate can take only one object (S is P] in
contrast to a polyadic logic, where a predicate can take more than one
object (F(ab)}. Monadic logic cannot explain the infinity of numbers
because it cannot account for the relational and multi-general structure of
propositions such as every number has a successor. Given the inability of
traditional logic to analyse the relational and multi-general structure of
arithmetical propositions, Kant believed that one must turn to intuition to
account for the sequential nature of the numbers. For Kant numbers
cannot be generated by discursive concepts alone but instead must be
constructed in intuition through a process of synthesis.
To better understand Kant's point here it will help to say something
about his use of a conclusion, drawn by Leibniz, that all relations are ideal,
to show that space and time are not real entities existing independently of
human understanding and sensibility but ideal forms of intuition that
condition all appearances.
According to Leibniz:
You will not, I believe, admit an accident which is in two subjects at
once. Thus I hold, as regards relations, that paternity in David is one
thing, and filiation in Solomon is another, but the relation common to
both is a merely mental thing, of which the modifications of singulars
are the foundation. (Mates 1986, 210)
An accident, according to Leibniz, depends upon a subject for its existence.
That is an accident is a modification of a subject that 'can come into being
and perish while the subject remains' (PPL 606). Thus Socrates'
complexion can change from pale to red to brown and then back to pale
in the course of a year. When the paleness in Socrates changes to red, it
ceases to exist and the same holds true of the red when it changes to brown.
In the case of a relation between a father and a son, say, the relation is 'a
merely mental thing' that depends for its existence upon the modifications
of the two substances that stand in the relation.
According to Leibniz, it is impossible for an accident to exist in more
than one subject at the same time:
Introduction
I shall give yet another example to show how the mind uses, upon
occasion of accidents which are in subjects, to fancy to itself something
answerable to those accidents out of the subjects. The ratio or proportion
between two lines L and M may be conceived in three ways: as a ratio of
the greater L to the lesser M, as a ratio of the lesser M to the greater L,
and, lastly, as something abstracted from both, that is, the ratio between
L and M without considering which is the antecedent or which the
consequent, which the subject and which the object... In the first way of
considering them, L the greater, in the second, M the lesser, is the subject
of that accident which philosophers call 'relation'. But which of them will
be the subject in the third way of considering them? It cannot be said
that both of them, L and M together, are the subject of such an accident;
for, if so, we should have an accident in two subjects, with one leg in one
and the other in the other, which is contrary to the notion of accidents.
Therefore we must say that this relation, in this third way of considering
it, is indeed out of the subjects, but being neither a substance nor an
accident, it must be a mere ideal thing, the consideration of which is
nevertheless useful. (Mates 1986, 210-11)
An accident is in a subject when it directly inheres in it. An accident is out
of a subject when it does not directly inhere in it. The accident of being a
philosopher is something that inheres in Socrates. Now the relational
proposition that Socrates is a greater philosopher than Antisthenes can be
reduced to a categorical proposition in which the name 'Socrates'
designates a subject and the predicate 'is a greater philosopher than
Antisthenes' designates an accident. The proposition can also be reduced
to the categorical proposition that Antisthenes is a lesser philosopher than
Socrates, where the name 'Antisthenes' designates a subject and the
predicate 'is a lesser philosopher than Socrates' designates an accident. In
the proposition that Socrates is a greater philosopher than Antisthenes,
'Antisthenes' does not designate a subject but a component of the accident,
X is a greater philosopher than Antisthenes', that is attributed to the subject
Socrates. In the proposition that Antisthenes is a lesser philosopher than
Socrates the cases are reversed. In the case of a relation that abstracts from
both subjects, it does not have a leg in either subject. And since an
accident, of which a relation is a species, exists only in so far as it inheres in
a subject, such a relation is only ideal. For an accident to exist it must
inhere in some subject. If a relation inheres in neither of the subjects that it
relates to, then it does not exist. Instead, as Leibniz says in the above
quoted passage, 'it must be a mere ideal thing'.
10
In the 'Transcendental Aesthetic', Kant uses the ideality of relations

and their non-reducibility to non-relational propositions to establish the
ideality of space and time as the a priori forms of intuition:
In confirmation of this theory of the ideality of both outer and inner sense,
and therefore of all objects of the senses, as mere appearances, it is
especially relevant to observe that everything in our cognition which
belongs to intuition ... contains nothing but mere relations ... Now a thing
in itself cannot be known through mere relations; and we may therefore
conclude that since outer sense gives us nothing but mere relations, this
sense can contain in its representation only the relation of an object to the
subject, and not the inner properties of the object in itself. (B667)
All appearances, being objects of intuition conditioned by the forms of space
and time, contain nothing but relations; while things in themselves,
unconditioned by the forms of space and time, contain nothing but intrinsic
properties related solely to the object in itself with no relation to any other
object and its internal properties. Thus we are only able to understand that
the teapot is behind the cup or that this building is taller than that one by
presupposing space: the outer sense of intuition. The forms of inner and
outer sense are the necessary conditions by which we can understand and
differentiate the material provided to the mind through sensible intuition.
Even though the material of sensible intuition is provided to the mind by
outer sense, inner sense too is presupposed. For inner sense 'contains [only]
relations of succession, coexistence, and of that which is coexistent with
succession, the enduring' (B67). An intuition is a representation that 'can be
antecedent to any and every act of thinking anything' (B67). A formal
intuition is an intuition that contains nothing but relations. Since an
intuition is antecedent to any act of thinking and since a formal intuition
contains nothing but relations, our ability to understand relations
presupposes the formal intuition of space and time. As Paul Guyer explains,
Kant argues from the premise that relations are not real to 'the conclusion
that space and time are not properties of things in themselves, which in turn
implies that they must be mere forms of our own representation of things
(Guyer 1987, 352). For Kant, the phenomenal world consists entirely of
relations. Since the understanding, on its own, cannot grasp relations,
intuition is required. Since things in themselves fall outside the bounds of
space and time we cannot apprehend any relational properties that they
might have, and without some knowledge of an object's relational properties
it is impossible to acquire any knowledge of the object itself. Guyer
Introduction
11
comments that 'After Frege and Russell succeeded in clarifying the logic of
relations, a metaphysical prejudice against relations could certainly derive
no comfort from their earlier logical obscurity, and so this metaphysical
argument could hardly be persuasive today' (Guyer 1987, 352). According
to Guyer, then, a central component in Kant's argument for transcendental
idealism rests on Kant's lack of an adequate logic of relations. The third part
of this book will seek to clarify why Kant believed that the ideality of
relations entails the ideality of space and time and how that view shaped his
philosophy of mathematics. For the moment, the point to keep in mind is
that for Kant the fact that arithmetical equations are relations that cannot
be reduced to the categorical propositions of traditional logic is an
important reason why arithmetic is a synthetic and not an analytic science.
Thus, the reason why we cannot justify the truth of 7 + 5 = 12
analytically is because the relational structure of such an equation cannot
be adequately captured by the logic available to Kant. For example, the
first step in determining the truth of a mathematical proposition
analytically would be to reduce the mathematical proposition to a
categorical proposition. Thus, if one were to attempt to reduce 7 + 5 = 12
to a categorical judgement the most natural move would be to treat 7 + 5
as the subject concept, the equal sign as the copula and 12 as the predicate
concept. But it is unclear how such a reduction could be accomplished
with the resources available through traditional logic.
One problem is this: In traditional logic, the concept of humanity falls
under the concept of animality because animal is a genus of the species
human. However, the concept of 12 is surely not the genus of the concept of 7
+ 5. For if the concept of 7 + 5 is a species of the concept of 12, then so is 6
+ 6 or 20 8. But then the concept of 12 would have an infinite number of
species. As Kant explains in a letter to his former student Johann Schultz:
I can form a concept of one and the same quantity by means of many
different additions and subtractions; (notice that both of these processes
are syntheses, however.) Objectively, the concepts I form are identical (as
in every equation). But subjectively, depending on the type of
combination that I think, in order to arrive at that concept, they are
very different. So that at any rate my judgement goes beyond the concept
I get from the synthesis, in that the judgement substitutes another
concept (simpler and more appropriate to the construction) in place of
the first concept, though it determines the same object. Thus I can arrive
at a single determination of a quantity by means of 3 + 5, or 12 4, or 2
X 4, or 23, namely 8. But my thought '3 + 5' did not include the
12

thought '2 X 4'. Just as little did it include the concept '8', which is equal
in value to any of these. (Philosophical Correspondence [Kant 1967, 129])
It is possible to form a concept of the number 12 by means of a variety of

different equations. In such a case the different concepts that can be used to
express the number 12 (e.g. 6 + 6, 20 8, etc.) are objectively identical.
However, because the different equations expressing the number 12 are
different ways that the number can be thought, Kant concludes, perhaps a
bit too hastily, that these different ways of expressing the number are
subjective. Because we can arrive at the concept of a given number from a
potentially infinite number of ways, these different ways cannot be how the
specific number is defined. Thus these different equations cannot be a part
of the partial concepts that make up the concept of a specific number. What
uniquely characterizes the number cannot consist merely in the wide
variety of ways that the number can be arrived at. My concept of the
number 12 is the product of an act of synthesis. I can form the concept of a
given number through an act of synthesis in an infinite variety of ways.
However, my concept of the number 12 cannot include the many different
equations by which it can be expressed. For if the concept of the number 12
consists in all of the different determinations it can have, there would be no
way to arrive at a complete analysis of the concept.
According to Kant, an important task of logic is to make clear concepts
distinct (see Jdsche Logic VIII, 568). But this raises the question: in what
way does logic make concepts distinct? Followers of Christian Wolff
believed that the distinctness of a concept can be determined solely
through an act of analysis. But analysis can only account for our ability to
make a concept distinct. It cannot account for our ability to make a
distinct concept. Analysis can make a given concept distinct only by
analysing the marks that are already contained in a concept. However, in
the case where a mark is attributed to a concept that does not already
contain it, analysis cannot determine whether the mark belongs to the
concept. This requires an act of synthesis. In analysis we make a concept
distinct. In synthesis we make a distinct concept. To make a clear concept
distinct is to identify all of the constituent concepts that make up the
concept through analysis. To make a distinct concept one must use more
than the marks that are revealed in a concept through analysis. One must
also employ synthesis to show how a judgement can be true when the
predicate concept is not contained in the subject concept.
In constructing a number in intuition one is making a distinct concept.
Thus the concept of the number 12 can be attained by starting with the
Introduction
13
concept of the number 7, say, and using some kind of sensible object as the
mark of number count from 7 to 12. To count or calculate, according to
Kant, we must step outside of our concepts of the numbers and employ
intuition, either through counting our fingers or calculating with symbols,
in order to arrive at the concept of one number from the concept of
another. Because knowledge of arithmetical equations cannot be achieved
through the analysis of the concepts that make up the equations,
arithmetic cannot be an analytic a priori science. Because the truths of
arithmetic follow with necessity and strict universality, arithmetic is an a
priori science. However, because the truth of arithmetical propositions
cannot be justified solely through an analysis of the concepts that make it
up, the understanding must turn to synthesis, in particular the
construction of concepts in intuition, to justify the truths of arithmetic.
Frege
One of the central aims of Frege's two great early works, the BegriffsschriftAA
(1879) and the Foundations of Arithmetic (Foundations] (1884) was to refute
Kant's claim that arithmetic is a synthetic a priori science and establish the
probability that arithmetic was an a priori analytic science. He then set out
in his two-volume Grundgesetze der Arithmetik (1893, 1903) (Basic Laws of
Arithmetic (Basic Laws]) to offer a formal proof that all of the theorems of
arithmetic can be derived solely from logical laws and all of the concepts of
arithmetic can be given purely logical definitions.
In the Begriffsschrift Frege had something at his disposal that Kant did not,
namely a theory of relations that was not tied to the traditional conception of
judgement as a relationship between a subject (concept), a predicate
(concept) and a copula that binds them together. In short, what Frege had
and Kant lacked was a polyadic logic with a quantifier/variable notation that
could account for the relational and multi-general structure of arithmetical
propositions. This gave Frege's concept script the power to account for the
logical structure of general arithmetical propositions, such as every number
has some number as a successor. As Michael Friedman explains:
A central difference between monadic and polyadic logic is that the
latter can generate an infinity of objects while the former cannot . . .
Hence, monadic logic cannot serve as the basis of a serious
mathematical theory, for any theory aiming to describe an infinity of
objects. (Friedman 1992, 59)
14
Frege explicates his new logical language in the B egriffsschr ift and showsA
how it can be used to give a purely logical definition of following in a series.
He developed his concept script in order to give a more detailed analysis of
the concepts of arithmetic and to 'provide a deeper foundation for its
theorems' (CN 107). He continues this project in the Foundations, where he
gives an informal exposition of his project of grounding the truths of
arithmetic in the laws of logic and defining arithmetical concepts in terms
of logical concepts. Here he rejects Kant's claim that arithmetic is a
synthetic a priori science and argues that arithmetic rests solely upon
analytic a priori principles. Where Kant's account of the sequence of
numbers presupposes the pure intuition of time, Frege accounts for the
sequence of numbers using mathematical induction: a purely logical
operation. Where Kant accounts for our ability to grasp the individual
numbers in terms of the schematism of the categories of quantity, Frege
accounts for our ability to grasp numbers by invoking the context
principle: a word gets its meaning only in the context of a proposition.
Where Kant thinks that logic is a purely formal science with no content of
its own, Frege thinks that, while logic provides the forms of judgement that
the judgements of all other sciences must adhere to, the derivation of truths
from the basic laws of logic can lead to fruitful discoveries that can only be
explained by attributing both form and content to the laws of logic.
The context principle is invoked to explain how a number term can
mean something even though numbers are not ideas and are not accessible
through intuition or sense perception. For if words acquire their meaning
only within the context of a sentence, then the way to determine what
number words mean is to define the sense (establish the truth conditions)
of a proposition in which a numerical expression occurs. That is, if we can
establish the truth conditions of a numerical identity, since, by the context
principle, the meaning of the parts of a proposition is determined by the
meaning of the whole proposition, then we can determine the meaning of
the numerical expressions by determining whether the proposition they are
contained within is true.
Frege invokes the context principle in the Foundations in an attempt to
give a contextual definition of number such that the number of Fs is the
same as the number of Gs if and only if there are just as many Fs as Gs
( (Nx:Fx = Nx:Gx) = (Fx Gx)), where '', signifies that the concepts F
and G are equinumerous, i.e. that there are just as many Fs as there are Gs.
Such a definition consists solely of second-level concepts. That is, the
definition contains only concepts (functions) whose arguments are other
concepts. No proper names are deployed in such a definition and hence
Introduction
15
there is no need to invoke the power of intuition to explain how the

understanding grasps individual numbers. However, this attempt fails over
the inability of the principle to distinguish numbers from things that are
clearly not numbers, such as Julius Caesar. While Frege acknowledges that
no one could possibly mistake Julius Caesar for a number, there is nothing in
this definition that enables us to determine this. Frege's attempt to solve the
so-called 'Julius Caesar problem' leads him to abandon the attempt to give a
contextual definition of numbers. Instead he gives an explicit definition of
numbers in terms of extensions of concepts. The number ofFs is the same as
the number of Gs if and only if the extension of the concept F is the same as
the extension of the concept G. Frege incorporated this definition in Basic
Law (V) in the Basic Laws, where he attempted formally to establish his
logicist thesis. Just as the second volume was going to press in 1902, Frege
received the famous letter from Bertrand Russell informing him of a paradox
Russell had discovered that threatened his, as well as Frege's, attempt to
establish arithmetic upon purely logical foundations. Frege added a
postscript to the second volume of the Basic Laws acknowledging the threat
that Russell's paradox posed to his logicist thesis. As he explains in the
opening sentence of the postscript: 'Hardly anything more unwelcome can
befall a scientific writer than that one of the foundations of his edifice be
shaken after the work is finished' (BLA 127). After outlining the content of
Russell's letter and acknowledging that Basic Law (V) lacked the selfevidence possessed by the other laws in his work, he lays out the paradox.
Mr Russell has discovered a contradiction, which may now be set out.
No one will want to assert of the class of men that it is a man. Here we
have a class that does not belong to itself. That is, I say that something
belongs to a class if it falls under the concept whose extension that class
is. Now let us fix our attention upon the concept class that does not belong
to itself. The extension of this concept (if we may speak of its extension)
is accordingly the class of classes that do not belong to themselves. For
short we shall call it the class C. Now let us ask whether this class C
belongs to itself. First let us suppose that it does. If something belongs to
a class, then it falls under the concept whose extension the class is;
accordingly if our class C belongs to itself then it is a class that does not
belong to itself. Thus our first supposition leads to self-contradiction.
Second, let us suppose that our class C does not belong to itself; then it
falls under the concept whose extension it itself is, and thus does belong
to itself: here again a contradiction. (BLA 127-8)
16
Frege tried for a while to avoid the dire consequences of the paradox, but
was ultimately unsuccessful. In a short, posthumously published paper
entitled 'Numbers and Arithmetic' (1924/5), written near the end of his
life, Frege acknowledges that Kant was correct to invoke intuition to
account for the conceptual content of an arithmetical judgement. 'The
more I have thought the matter over', he says, 'the more convinced I have
become that arithmetic and geometry have developed on the same basis
a geometrical one in fact so that mathematics in its entirety is really
geometry. Only on this view does mathematics present itself as completely
homogeneous in nature. Counting, which arose psychologically out of the
demands of business life, has led the learned astray' (PW277). In another
paper, written around the same time, entitled 'A New Attempt at a
Foundation for Arithmetic', he says:
I have had to abandon the view that arithmetic does not need to appeal
to intuition either in its proofs, understanding by intuition the
geometrical source of knowledge, that is, the source from which flow
the axioms of geometry. (PW 278)
And on the next page he explains:
Since probably on its own the logical source of knowledge cannot yield
numbers either, we will appeal to the geometrical source of knowledge.
This is significant because it means that arithmetic and geometry, and
hence the whole of mathematics flows from one and the same source of
knowledge - that is the geometrical one. (PW 279)
Frege explains that '[i]f one wished to restrict oneself to the real numbers,
one could take these to be ratios of intervals on a line, in which the
intervals are to be regarded as oriented, and so with a distinction between
a starting point and an end point' (PW 279). If one wishes to explain
complex numbers, however, then such an approach won't work. He then
begins a brief and incomplete sketch of how to explain complex numbers in
terms of ratios of intervals on a plane.
So ends Frege's last known written remarks on a project that dominated
his scholarly career: the attempt to rid the science of arithmetic of any
dependence upon intuition and establish it upon purely logical foundations.
Virtually all of Frege's major accomplishments (the invention of the first
modern logical calculus, the functional analysis of concepts, the context and
compositionality principles, the sense/reference distinction) were spin-offs
Introduction
17
from this project. The project started with the rejection of Kant's claim that
arithmetic is an a priori synthetic science, and hence grounded in intuition,
and ended, after the blow of Russell's paradox, with Frege's acknowledgement that intuition is a necessary component to any attempt to establish
the foundations of arithmetic. However, his death in 1925 prevented him
from developing this new conception of arithmetic any further.
Outline of the Book

In Part 1 of this book Kant's conception of logic will be compared and
contrasted with Frege's concept script as explicated in the Be griffsschrift.
On the traditional view of logic that Kant follows the most basic elements
are concepts. Judgements are constructed out of concepts and arguments
are constructed out of judgements. This order is reflected in the
composition of traditional logic texts from Aristotle's Organon: the
Categories (concepts), De Interpretatione (judgements) and the Prior and
Posterior Analytics (arguments) to the Port-Royal Logic, a very influential
logic handbook by Antoine Arnauld and Pierre Nicole, first published in
1662. The first part of the Port-Royal Logic concerns ideas, the second
judgements and the third reasoning. With the development of his new
polyadic predicate logic and his quantifier/variable notation grounded in
the distinction between function and argument instead of subject and
predicate, Frege is able to develop a logic with the capacity to analyse the
relational and multi-general structure of arithmetical propositions to an
extent beyond the reach of the power of traditional logic. Particular
attention will be paid to Frege's development of the assertion stroke to
distinguish the content of a judgement from its assertibility conditions. The
difference between the forms of judgement in Kant's logic as exemplified in
the 'Table of Judgements' and the corresponding forms of judgement in
Frege's new logical language will also be of concern.
Part 2 will consist in a more detailed look at Kant's critical turn in
philosophy, concentrating on the role that the distinction between concepts
and intuitions plays in this turn. Next, Kant's criticism of Leibniz and Wolff
as dogmatic metaphysicians will be examined. I shall argue that Leibniz's
inability to capture logically the relational structure of arithmetical
expressions was one of the factors that led to Kant's claim that arithmetic
is a synthetic science. In rejecting Leibniz's attempt to give an analytic
definition of number, Kant argues that the justification of the truth of
arithmetical propositions requires the construction of concepts in intuition.
18
Next I shall examine the relations between the judgements of quantity and
the categories of quantity before considering Kant's definition of number as
a schema of the categories of quantity. Here special emphasis will be placed
on the role that time plays in such a schema.
The focus of Part 3 will be on Frege's criticism in the Foundations of
Kant's claim that arithmetic is a synthetic a priori science and with Frege's
claim that with his new logic, he can expand the scope of analytic
judgements to include the truths of arithmetic. Before looking at Frege's
criticism, however, Kant's distinction between analytic and synthetic
judgements will be examined along with Leibniz's definition of truth in
terms of concept-containment, which Kant rejects as too narrow. Because
the predicate concept of a synthetic judgement is not contained in the
subject concept, its truth cannot be justified in terms of conceptcontainment. From there I shall explore Frege's criticism of Kant's notion
of intuition and analytic judgements. Particular emphasis will be paid to
Frege's arguments for the possibility of fruitful analytic definitions and to
his logical definition of following in a series, which serves as an example of
such a possibility. This allows Frege to capture the iterative nature of
numbers without invoking the intuition of time, and thus to reject Kant's
claim that the apprehension of numbers requires the intuition of time.
In his book Frege: Philosophy of Language, Michael Dummett claims that
Frege replaced 'the Cartesian thesis that epistemology is the foundation of
philosophy' (FPL 667) with the thesis that philosophy of language is the
foundation of philosophy. According to Dummett:
Descartes made the question 'What do we know, and what justifies our
claim to this knowledge?' the starting-point of all philosophy: and,
despite the conflicting views of the various schools, it was accepted as
the starting-point for more than two centuries . . . Frege's basic
achievement lay in the fact that he totally ignored the Cartesian
tradition, and was able, posthumously, to impose his different
perspective on other philosophers of the analytic tradition. (FPL 666-7)
For Frege, according to Dummett, the first task in any philosophical
enquiry is the analysis of meanings. For, 'until we have first achieved a
satisfactory analysis of the meanings of the relevant expressions, we cannot
so much as raise questions of justification and of truth' (FPL 667). It was
Frege, according to Dummett, who first recognized this and showed that
the theory of meaning 'is the foundation of all philosophy, and not
epistemology as Descartes misled us to believe' (FPL 669). In so doing
Introduction
19
Frege 'effected a revolution in philosophy as great as the similar revolution

previously effected by Descartes . . . We can therefore, date a whole epoch
in philosophy as beginning with the works of Frege, just as we can do with
Descartes' (FPL 669).
The aim of Part 4 is to show that the scope of the revolution that Frege has
wrought upon philosophy stretches further back than Descartes. In
particular, in replacing the traditional subject/predicate-based Aristotelian
syllogistic logic with his function/argument-based polyadic predicate logic
with its quantifier/variable notation, Frege also replaced the traditional
substance/attribute metaphysical perspective with a concept/object metaphysical perspective. In doing so, Frege replaced the paradigm conception of
an object as a substance consisting of substratum of essential properties
around which accidental properties and external relations are predicated.
On such a view substances are independently existing objects and all other
entities depend for their existence upon their inherence in substances. Frege's
turning away from this conception was considerably helped by Kant, who
replaced the traditional conception of substance as an independently existing
object with a conception of substance as a schematized category that when
applied to appearances provided by sensible intuition produces objects of
experience. Part 4 will chart, albeit briefly, the history of this transition,
beginning with an account of Aristotle's conception of substance in the
Categories. Next to be examined will be the transition of this conception of
substance in the hands of philosophers such as Leibniz and Locke, in reaction
to the new modern mechanical worldview arising out of the development of
modern science. Next Kant's new conception of substance as a permanent
underlying substratum of change will be examined together with his criticism
of Leibniz's conception of substance as a thing in itself that transcends the
limits of space and time: a transcendental object that can be thought but not
known.
In Part 4 Frege's new conception of an object and the resulting new
conception of number as a logical object will then be examined. Particular
interest will be paid to his use of the context principle in the Foundations to
give a contextual definition of number. Frege's new conception of an
object and his inability to distinguish numbers from other kinds of objects
as revealed by the so-called Julius Caesar objection will then be explored,
along with a Kantian diagnosis of what led Frege astray. In the course of
this discussion it will be shown how Kant's rejection of substance as an
independently existing object together with his new conception of an
object, and his claim that existence is not a real predicate set the stage for
Frege's new conception of an object. However, this strong Kantian
20
influence on Frege should not lead to an interpretation of Frege as some

kind of Kantian idealist, as Michael Ayers has charged. Once Frege's
similarities to Leibniz over the nature of the a priori are examined in
contrast to Kant's conception of the a priori along with Frege's views on the
relationship between logic and objectivity, it will be seen that Ayers'
interpretation of Frege is misdirected.
Finally some concluding remarks will be made about the nature of
Frege's revolution of philosophy and the influence that it had on the rise of
analytic philosophy.
Notes
13
14
15
16
17
18
19
20
21
22
23
24
See, for example, the Geach 1961 chapter on Frege in Three

Philosophers, 129-62. Frege's influence on Geach is evident in his
Reference and Generality (Geach 1980a) and in a collection of his logical
essays Logic Matters (Geach 1980b).
See Part 2 of Strawson 1959 for a discussion of the relationship
between subject and predicate that is clearly influenced by Frege.
See, for example, Church 1951.
See Word and Object (Quine 1960) and From a Logical Point of V
logico-philosophical essays, for examples of Frege's influence on Quine.
Davidson has been particularly influenced by Frege's context and
compositionality principles. See Inquiries into Truth and Interpretation, in
particular the essays 'Truth and Meaning', 'The Method of Truth in
Metaphysics' and 'Reality Without Reference'.
See Kripke 1972.
See the appendix to Chapter 5 in Dummett 1981.
See Evans 1985; Kaplan 1989; and Perry 1977.
See the article on Wolff by Giorgio Tonelli in, The Encyclopedia of
Philosophy, v. 8, pps. 340-4 and 'The reception of Leibniz in the
eighteenth century', by Catherine Wilson in Jolley 1995, 442-74.
Kant qualifies his claim that an a priori cognition is 'absolutely
independent of all experience' to make it consistent with his claim that
all cognition begins with experience. This qualification will be
discussed further below.
See 2b in Prolegomena to any Future Metaphysics (Kant 1997).
Quoted from Guyer 2006, 96.
Part 1
Logic Old and New
Kant's Logic
Kant on Pure General Logic

Kant lectured on logic throughout his long scholarly career. While he never
wrote a logical textbook, a number of his lectures were recorded and
published. In addition Kant instructed his student Gottlob Benjamin Jasche
to prepare a handbook on logic based on Kant's notes to Friedrich Meier's
(1718-77) textbook on logic Auszug aus der Vernunftlehre (Excerpts from the
Doctrine of Reason), published in 1752. Kant used Meier's textbook as a basis
for his logical lectures. In what follows I will concentrate my exposition of
Kant's views on logic on this textbook, turning occasionally to the evidence
of his other lectures on logic. Kant follows the traditional order of exposition
of logic texts of his times. After a lengthy introduction he divides the work
into two parts: 'I. General Doctrine of Elements' and 'II. General Doctrine
of Method'. He begins Part I with a discussion of the formation and nature
of concepts, then discusses judgements and, finally, inferences.
Kant opens the Logic with the claim that 'Everything in nature, both in
the lifeless and in the living world, takes place according to rules, although we
are not always acquainted with these rules' (Jasche Logic I, 527). After
listing several examples (water falls according to the rules of gravity,
animals move according to rules of motion) he turns to the human
understanding.
The exercise of our powers also takes place according to certain rules
that we follow, unconscious of them at first, until we gradually arrive at
cognition of them through experiments and lengthy use of our powers,
indeed, until we finally become so familiar with them that it costs us
much effort to think them in abstracto. (Jasche Logic I, 527)
He illustrates this with an analogy. While there is a 'universal grammar'
that underlies anyone's ability to speak, speakers can express grammatically correct sentences without any formal knowledge of this grammar.
24
'Since the understanding is the source of rules', concludes Kant, 'the

question is thus, according to what rules does it itself proceed?' (Jdsche Logic
I, 527). For Kant then, the understanding is a power whose actions are
bounded by rules. To discover these basic rules of the understanding, it is
necessary to abstract from all applications of the rules and determine their
form alone. This produces two broad categories of rules: those that are
necessary and those that are contingent. The rules of logic are necessary.
Some philosophers might think that since logic is concerned with the
necessary laws of thinking that its principles are grounded in psychology.
Kant disagrees. As he explains in his discussion of pure general logic in the
Critique:
1. As general logic, it abstracts from all content of the cognition of the
understanding and from all differences in its objects, and deals with
nothing but the mere form of thought.
2. As pure logic, it has nothing to do with empirical principles, and does
not, as has sometimes been supposed, borrow anything from
psychology, which therefore has no influence whatever on the canon
of the understanding. Pure logic is a body of demonstrated doctrine,
and everything in it must be certain entirely a priori. (A54/B78)
Pure general logic is a priori in two ways. Since it abstracts from all content
of a judgement and considers only its form, it is general. Since it is a body
of demonstrated doctrine that is certain entirely a priori, it is pure. Hence
pure general logic has no empirical principles. And since psychology is an
empirical science, the laws of logic do not depend upon the laws of
psychology. For if they did, then logic would only be able to explain 'how
thinking does take place and how it is under various subjective obstacles
and conditions; this would lead then to cognition of merely contingent
laws' (Jdsche Logic I, 529). Kant insists that:
In logic, however, the question is not about contingent but about necessary
rules; not how we do think, but how we ought to think. The rules of
logic must thus be derived not from the contingent but from the necessary
use of the understanding, which one finds in oneself apart from all
psychology. In logic we do not want to know how the understanding is
and does think and how one has previously proceeded in thinking, but
rather how one ought to proceed in thinking. Logic is to teach us the
correct use of the understanding, i.e., that in which it agrees with itself.
(Jdsche Logic I, 529)
Kant's Logic
25
Psychology is an empirical science that seeks to explain how humans think.

Logic is a normative science that seeks to explain how humans should think.
While the rules of logic are not discovered through the use of psychological
laws and principles they are discovered through introspection. 'Logic', Kant
concludes, 'is thus a self-cognition of the understanding and of reason, not as
to their faculties in regards to objects, however, but merely as to form . . . In
logic the question is only, How will the understanding cognize itself?^ (Jdsche Logic
I, 52930). We discover the laws of logic, according to Kant, through selfreflection on our own mental capacities. In this respect he has a conception
of logic at odds with most modern conceptions. As Beatrice Longuenesse
explains, recent criticisms of Kant's use of the term 'logical forms of
judgement' have been hindered by a failure to understand the difference
between Kant's notion and modern notions of logical form:
Kant's notion of logical form is not that of modern logic, in which the form
refers to the logical constants and the rules of composition and derivation
adopted in a given calculus . . . [F]or Kant 'logical form' refers to
something different, namely the universal rules of discursive thought . . .
What Kant claims to display in his table of the logical forms of judgement
are forms of mental activities, and the transcendental deduction of the
categories consists in showing that these mental activities are necessary for
any representation of an object. (Longuenesse 1998, 5)
Kant believed that pure logic is entirely a priori. However, while logic is
not grounded in the contingent laws of psychology, what gives logic its a
priori character, and thus its necessity, is its origin in the human
understanding. For Kant, the foundations of the laws of logic lie not in
some objective third realm but in the inner recesses of the human mind.
We can grasp the laws of logic not through contemplation of a timeless
realm of eternal truths but through self-reflection on our own power of
thinking. In this respect, Kant follows other early modern philosophers in
conceiving of the mind as possessing normative powers of thinking. As
Gary Hatfield explains, for most early modern philosophers:
The intellect or faculty of understanding . . . was regarded as a
'knowing' or a 'truth discerning' power. The deliverances of this power
were conceived to be successful epistemic achievements, that is,
essentially normative. The nature of the intellect was to perceive truth,
and authors such as Descartes maintained that, left undisturbed, it
could do nothing else. (Hatfield 1990, 3)
26
On Kant's conception of logic, then, the normative force of logic lies in its a
priori origins in the mind. The logical forms of judgement are forms of
mental activity such that the proper use of a specific logical form
guarantees its successful use. Thus the successful application of the rule of
modus ponens, say, depends upon the proper use of the hypothetical form of
judgement in the activity of drawing an inference. The necessity of logical
inference lies not in some normative constraint imposed upon the mind
from the outside but in the proper use of the basic a priori mental faculties
explicated by Kant in his Table of Judgements.
Concepts in Kant's Logic

According to Kant, every concept has both matter and form. The matter
of a concept is an object; its form is its generality. 'The origin of concepts as
to mere form rests on reflection and abstraction from the difference of things
that are designated by a certain representation' (Jdsche Logic 5 91). The
mere form of a concept, its logical origins, is a product of a process of
reflection and abstraction of the differences of things designated under the
same representation.
The logical form of a concept is not determined through a characteristic
by means of which a concept relates to an object but in terms of general
concepts under which more specific general concepts fall. The concern of
general logic is not the source of the representations that make up the
content of a concept but only with concepts that have already been
formed. It is the task of transcendental logic to explain how concepts are
formed from the appearances provided by intuition. The logical origin of a
concept is the form that is discovered when one abstracts from all sources
of the content of a concept and determines what is common to the several
objects that fall under a given concept. Kant calls this process of
discovering the logical form of a concept reflection. In reflection a
representation that is common to several objects is formed. The power of
judgement must be furnished with these forms of concepts in order for it to
operate effectively. Without empirical concepts we would be unable to
organize the stream of appearances provided by intuition into objective
representations.
In section 6 of the Logic, Kant explains how concepts are generated by
logical acts of the understanding. These logical acts or operations of the
understanding come in three forms: comparison, reflection and abstraction. These acts form the essential and general conditions for the
Kant's Logic
27
generation of any concept. They perform their function by bringing

various concepts under a higher concept through the unity of an act of
consciousness. Kant gives an example in which the determinate concept of
a tree is generated from the determinate concepts of different kinds of trees
through the act of comparing the differences in various objects, reflecting
on the similarities and abstracting from their differences.
A more universal concept can be created from more specific concepts
through logical acts of the understanding. Thus by comparing, reflecting
and abstracting between the characteristics of a tree, a bush and grass one
can form the concept of a plant, and so on. It is through the employment of
these three acts of the understanding that the characteristics or marks that
make up a concept can be analysed.
Judgements in Kant's Logic

Following the order of traditional logical texts, Kant next turns to an
examination of judgements: 'A Judgement is the representation of the
unity of the consciousness of various representations, or the representation
of their relation so far as they constitute a concept' (Jdsche Logic 17, 597).
Kant presents two characterizations of judgement that he seems to treat as
equivalent. A judgement is a representation of:
1. the unity of the consciousness of various representations
2. the relation between various representations in so far as they constitute
a concept.
Now to help make sense of Kant's logical notion of judgement, it will help
to say something about two of its key terms: 'consciousness' and
'representation'. The two terms are counterparts in Kant's explanation
of cognition:
All our cognition has a twofold relation, first a relation to the object, second
a relation to the subject. In the former respect it is related to
representation, in the latter to consciousness, the universal condition of all
cognition in general. (Consciousness is really a representation that
another representation is in me.) (Jdsche Logic V. 544)
Cognition in relation to an object is called a representation. Cognition in
relation to a subject is called consciousness. On the one hand,
28
consciousness is the universal condition that makes any representation

whatsoever possible. For Sara to have a representation of anything, be it
empirical or a priori, an object of sense perception or imagination, she must
be conscious. In this sense consciousness is the capacity to have a
representation. On the other hand, consciousness is the representation that
a representation is in me. Here consciousness is the product of a selfreflective act in which I am aware that that any representation that I have
is a representation in me.
In the Critique, Kant characterizes representations as 'inner determinations of our mind in this or that relation of time' (A197/B242). A passage
from the 'Transcendental Aesthetic' can perhaps clarify Kant's reason for
invoking time in his definition of 'representation'. 'Time', according to
Kant, 'is nothing but the form of inner sense, that is, of the intuition of
ourselves and of our inner state. It cannot be a determination of outer
appearances; it has to do neither with shape nor position, but with the
relation of representations in our inner state' (A33/B4950). Thought and
intuition provide the mind with a stream of representations that follow one
another in time. Intuitions, concepts and judgements are all representations (see A320/B376). In presenting an argument or drawing an
inference, the mind leaps from one judgement to another until a
conclusion is reached. In such a situation these judgements are not only
related to each other in terms of the logical role they play in the inference
but also in terms of a series of representations that follow one another in
time. Similarly, the perceptible world impinges upon our senses and
understanding as a multifaceted stream of sense perceptions following one
another in time.
According to (1), then, a judgement is the consciousness that certain
representations form a unity, where, according to (2), these various
representations are unified by being brought under a concept. In (1) a
judgement is characterized as a relationship between a judging subject and
representations that are unified by a conscious act of the understanding. If
we turn to the two senses of consciousness as discussed above it would seem
that Kant's second sense of consciousness is more applicable to his
characterization of judgement. Judgement is the awareness that various
representations are united under a higher representation. In (2) there is no
mention of such a relationship. Instead judgement is characterized as a
relationship between various representations in so far as they constitute a
concept. Thus the concept of a body is composed of the concepts of
extension, impenetrability, figure, etc. The judgement that all bodies have
extension is a representation of the relationship between the concept of a
Kant's Logic
29
body and one of the various concepts that make up the concept of a body.
(1), then, gives a mentalistic account of the unity of a judgement. To have
a representation of a judgement, the subject must be conscious that various
representations fall under a higher representation. (2) gives a logical
account of the unity of a judgement. A judgement forms a unity when
various representations fall under a single concept. Kant treats these two
characterizations as equivalent. Since a concept is a product of the
understanding, the unity of the various concepts that make up a
judgement corresponds to the unity of consciousness of the various
representations that make up a judgement.
On this account of the unity of a judgement there appears to be no
significant difference between a judgement and a complex concept. The
complex concept of a body as an entity that is extended, impenetrable,
etc., is something that can be unified in one act of consciousness. Similarly,
the concepts of extension, impenetrability, etc., can be unified under the
concept of a body. It is only when Kant turns to a discussion of the matter
and form of a judgement that the nature of a judgement is shown to be
different from the nature of a complex concept.
According to Kant, the matter of a judgement is expressed by the
subject and the predicate while its form is expressed by the copula (see
Blomberg Logic [Kant 1992], 274, 221). In a pre-critical essay entitled, 'The
Mistaken Subtilty of the Four Syllogistic Figures', Kant defines a
judgement as 'the comparison of a thing with some mark. The thing itself
is the subject, the mark is the predicate. The comparison is expressed by
the copula 'is' . . . which when used alone indicates that the predicate is a
mark of the subject' (quoted from Hanna 1990, 340-1). For Kant, a
judgement is a comparison between a thing (subject) and an attribute
(predicate). A judgement results when a predicate is connected to a subject
by the copula.
In the Logic Kant says:
Matter and form belong to every judgment as essential constituents of it.
The matter of the judgment consists in the given representations that are
combined in the unity of consciousness in the judgment, the form in theA
determination of the way that the various representations belong, as
such, to one consciousness. (Ja'sche Logic 18, 598)
If we combine these passages we get the following account. The matter of a
judgement consists of the representations signified by the subject and
predicate of the sentence that expresses the judgement. These representations
30
are combined into a unity by an act of judgement. The form of a

judgement is signified by the copula. The copula determines the way in
which the representation designated by the predicate belongs to the
representation designated by the subject. It is under the form of judgement
that the matter provided by appearances is combined into a unity of
consciousness. For Kant, then, it is an act of the understanding that
combines the matter designated by a subject concept and a predicate
concept into a judgement. It is the copula that signifies this combination
and indicates what kind of relation holds between the concepts.
Inferences in Kant's Logic

According to Kant, an inference is 'the derivation of one judgement from
another' (Jdsche Logic 41, 609). He claims that there are three different
rules of inference, categorical, hypothetical and disjunctive, corresponding
to the three forms of judgement under the heading of relation. That is,
there are three conditions under which one judgement follows from one or
more other judgements according to a rule. Since the disjunctive inference
can be defined in terms of the hypothetical inference and falsehood (see
Jdsche Logic 77, 623), the following account will concentrate on the
relation between categorical and hypothetical inferences.
The difference between categorical and hypothetical inferences is
grounded in the difference between categorical and hypothetical judgements. As Kant explains:
The matter of a categorical judgment consists of two concepts, the form
in the relation in which the one concerns the subject, the other the
predicate. E.g., All men are mortal. - In the hypothetical judgment the
matter consists of two judgments. E.g., If the soul is corporeal, then
there is no hope of the necessity of another life. The if expresses theAA
relation. (Vienna Logic [Kant 1992] 373)
Concepts are the matter of a categorical judgement and the form is the
manner in which the predicate concept is related to the subject concept, as
expressed by the copula. The matter of a hypothetical judgement consists of
two judgements. The form consists of a relationship between a ground
(antecedent) and a consequence (consequent) as expressed by 'if... then ...'
All categorical inferences rest upon the principle that ' What belongs to the
mark of a thing belongs also to the thing itself; and what contradicts the mark of a
Kant's Logic
31
thing contradicts also the thing itself (Jdsche Logic 63, 61718). All
hypothetical inferences rest upon the principle that the grounded follows
from the ground and the negation of the ground follows from the negation
of the grounded (see Jdsche Logic 76, 623).
All categorical inferences are syllogisms. As Kant explains in the
Critique:
In every syllogism I first think a rule (the major premiss) through the
understanding. Secondly, I subsume something known under the
condition of the rule by means of judgment (the minor premiss).
Finally, what is thereby known I determine through the predicate of the
rule, and so a priori through reason (the conclusion). The relation,
therefore, which the major premiss, as the rule, represents between
what is known and its condition is the ground of the different kinds of
syllogism. (A304/B360-1)
All syllogisms contain three essential elements: a rule, a condition and a
conclusion. The first element is a rule of the understanding or the major
premise of the inference. The second element is the minor premise or
condition. Finally, the conclusion is drawn by subsuming the subject
concept of the minor premise under the predicate concept of the major
premise. Take the syllogism:
All humans are mortal.
All philosophers are human.
Therefore, all philosophers are mortal.
The rule that all humans are mortal together with the condition that all
philosophers are human implies the conclusion that all philosophers are
mortal.
A hypothetical judgement consists of a relationship between two
judgements, a ground and a consequence. In a hypothetical inference the
conclusion is inferred either from the ground to the consequence, in the
case of modus ponens, or from the negation of the consequence to the
negation of the ground, in the case of modus tollens. As Kant explains: 'The
hypothetical inference of reason is one where the major propositio is a conditioned
proposition. Modus ponens infers from the ground to the consequence. Modus
tollens from the consequence to the ground' (Dohna-Wundlacken Logic [Kant
1992] 508). Take the following example of modus ponens.
32

If all philosophers are human, then all philosophers are mortal.
All philosophers are human.
Therefore, all philosophers are mortal.
In modus ponens the premise is a hypothetical judgement and the conclusion

follows from the truth of the ground of the hypothetical judgement. Kant
fails to point out that modus ponens and modus tollens were forms of inference
developed by Megarian and Stoic logicians independently of Aristotle's
syllogistic theory. Indeed, he even seems to claim that the hypothetical
form of inference is reducible to the categorical form. As he explains in the
Vienna Logic:
Categorical judgments constitute the basis of all the remaining ones.
Here the relation of subject and predicate is indicated. The hypothetical
judgment is composed of two problematic ones . . . In the hypothetical
judgment I consider the combination of two judgments as ground and
consequence. (Vienna Logic 373)
It is unclear exactly what Kant means when he says that the categorical
judgement constitutes the basis of a hypothetical judgement. For while it is
certainly true that a hypothetical judgement can be constructed from two
categorical judgements it is not necessary that they be so constructed. The
basis of a categorical inference revolves around a relationship between
concepts. The basis of a hypothetical inference revolves around a
relationship between judgements. While the two rules of inference are
compatible, Kant offers no explanation of how the latter can be reduced to
the former.
Frege's Logic
Preface to the Begriffsschrift

The goal of the Begriffsschrift was to develop a logical language precise and
rigorous enough to determine whether the theorems of arithmetic could be
derived from purely logical laws and whether its concepts and definitions
could be reduced to concepts and definitions of logic. As he explains in the
Preface: 'Arithmetic, as I said at the beginning, was the starting point of
the train of thought which led me to my "conceptual notation". I intend,
therefore, to apply it to this science first, trying to analyse its concepts
further and providing a deeper foundation for its theorems' (CN 107).
Frege turned to logic because only it could supply the degree of rigour and
precision necessary for him to analyse properly the concepts of arithmetic.
As he explains on the opening page of the Preface:
The firmest method of proof is obviously the purely logical one, which,
disregarding the particular characteristics of things, is based solely upon
the laws on which all knowledge rests. Accordingly we divide all truths
which require a proof into two kinds: the proof of the first kind can
proceed purely logically, while that of the second kind must be
supported by empirical facts. It is quite possible, however, for a
proposition to be of the first sort and still be one that could never come
to the consciousness of a human mind without activity of the senses [ftn:
Since without sense perception no mental development is possible for
beings known to us, the latter point holds for all judgements].
Therefore, not the psychological mode of origin, but the most perfect
method of proof underlies the classification. (CN 103)
If the goal is to establish the foundations of arithmetic, then the firmest
method of proof must be employed. Since logic disregards the particular
characteristics of things and 'is based solely upon the laws on which all
knowledge rests', it provides us with the firmest method of proof. Frege
34
then presents a brief argument to show that it is not psychological origin

but the best method of proof that underlies the classification of the two
different kinds of justification. All truths that require proof can be divided
into two kinds: those that admit of a purely logical justification and those
that are supported by empirical facts. Since sense perception is a necessary
condition for mental development, a logical truth could 'never come to the
consciousness of a human being without activity of the senses'. Since a
truth supported by empirical facts, also, obviously, could never enter the
consciousness other than through the senses, how a truth comes to our
consciousness (psychological mode of origin) cannot be what underlies the
difference between the two kinds of justification. The premise of the
argument Frege uses seems to be inspired by Kant's well-known claim that
while all cognition begins with experience not all cognition arises from
experience (see B1). According to Kant any kind of cognition be it a priori
or a posteriori ultimately has its origin in experience. Frege accepts Kant's
claim that sense perception is a necessary condition for knowing even
logical truths, and uses it to show that it cannot be the psychological mode
of origin that distinguishes an empirical truth from a logical truth but the
degree of precision that is required to prove the truth. Logic offers the
firmest method of proof.
So to determine whether the truths of arithmetic are purely logical or
based upon some other source of knowledge, the first step should be:
to test how far one could get in arithmetic by means of logical
deductions alone, supported only by the laws of thought, which
transcend all particulars. The procedure in this effort was this: I sought
first to reduce the concept of ordering-in-a-sequence to the notion of
logical ordering, in order to advance from here to the concept of
number. So that something intuitive could not squeeze in unnoticed
here, it was most important to keep the reasoning free of gaps. (CN 104)
Frege was compelled to develop his conceptual notation in order to reduce
the concept of ordering-in-a-sequence, an essential element in the analysis
of the concept of number, to the concept of a logical ordering. Only then
could a logical analysis of the concept of number be undertaken. In order
to keep logical reasoning free of gaps, the concept script was designed to
prevent anything intuitive from creeping unnoticed into a proof. That it is
something like Kantian intuition that Frege has in mind is confirmed later
where he argues that his logical definition of ordering in a sequence shows
that arithmetic is not synthetic a priori, as Kant has claimed.
Frege's Logic
35
Frege likens his new logical notation to Leibniz's attempt to develop a

'universal characteristic':
Leibniz also recognized perhaps overestimated the advantages of an
adequate method of notation. His idea of a universal characteristic, of a
calculus philosophicus or ratiocinator, was too ambitious for the effort to realize
it to go beyond the mere preparatory steps. The enthusiasm that overcomes
its [would be] creator when he considers what an immense increase in the
mental power of mankind that a method of notation directly appropriate
to objects themselves would bring about lets him underestimate the
difficulty which such aan undertaking confronts. (CN 105)
What impressed Frege about Leibniz's idea of a universal characteristic
was the increase in intellectual power that such a system of notation could
bring about by being directly applicable to objects themselves.
As Leibniz explains in a letter to Walter von Tschirnhaus:
No one should fear that the contemplation of characters will lead us
away from the things themselves; on the contrary, it leads us into the
interior of things. For we often have confused notions today because the
characters we use are badly arranged; but then, with the aid of
characters, we will easily have the most distinct notions, for we will have
at hand a mechanical thread of meditation, as it were, with whose aid
we can very easily resolve any idea whatever into those of which it is
composed. In fact, if the character expressing any concept is considered
attentively, the simpler concepts into which it is resolvable will at once
come to mind. Since the analysis of concepts thus corresponds exactly to
the analysis of a character, we need merely to see the characters in order
to have adequate notions brought to our mind freely and without effort.
We can hope for no greater aid than this in the perfection of the mind.
(PaaPL 193)
A proper notation, far from leading us away from the nature of the thing
under investigation, will reveal its true nature by giving us a precise
characterization of it based not upon the confused notions of things found
in ordinary discourse but in terms of the strict rules of the 'general
characteristic', as Leibniz calls his logical calculus. In such a language
there is a strict correspondence between the marks of a character and the
properties of the concept it denotes, so that to analyse the character is just
to analyse the concept.
36
Leibniz vacillated between two different conceptions of a universal

characteristic. As Donald Rutherford explains, in some places, especially
in his earlier writings, 'primary emphasis is laid on the universal
characteristic as a system of real characters, i.e. signs which possess a
determinate content, and exactly correspond in their structure to the
analysis of thoughts'. Attempts to establish such a language revealed that
the project was more difficult than Leibniz originally thought. Thus, in his
later writings, 'the universal characteristic is identified with a plan for a
general science of forms . . . a symbolic calculus (or collection of calculi)
whose principle object would be the formalization of patterns of inference
in different domains of knowledge' (CCL 227). As has been shown above,
Frege believed that in his concept script he had a logic of real characters
that possessed content and not just form.
Kant believed that since logic was completely general it could possess
no content and was thus purely formal. Frege, by contrast, believed that
complete generality does not imply that logic is completely formal. Leibniz
wavers over the question of whether logic has a content all its own. Early
in his career he believed that it did, but as his attempts to develop a
universal characteristic whose symbols possessed determinate conceptual
content faltered his goal shifted to developing a universal characteristic as
a general science of forms. While Leibniz made numerous important
innovations in logic, his adherence to Aristotle's syllogistic theory, with its
traditional structure of a predicate term connected to a subject term by a
copula, prevented him from making the kind of logical advances that
Frege accomplished.
Frege believed that his concept script finally fulfilled Leibniz's dream of
developing a universal characteristic that possessed content as well as form.
As he explains in two different essays aimed at contrasting his concept
script with George Boole's formula language:
I did not wish to present an abstract logic in formulas, but to express a
content through written symbols in a more precise and perspicuous way
than is possible with words. In fact, I wished to produce, not a mere
calculus ratiocinator, but a lingua characteristica in the Leibnizian sense. (CN
90-a1)
Right from the start I had in mind the expression of a content. What I am
striving after is a lingua characterica in the first instance for mathematics,
not a calculus restricted to pure logic. But the content is to be rendered
more exactly than is done by verbal language. For that leaves a great
deal to guesswork, even if only of the most elementary kind. There is
Frege's Logic
37
only an imperfect correspondence between the way words are

concatenated and the structure of the concepts. (PW 12-13)
Frege sought to develop a logical language with a content of its own.
Because ordinary language leaves so much to guesswork in the expression
of content, he was forced to abandon the rules and structure of ordinary
language in order to express the content of a concept with the precision
required for his project of establishing the foundations of arithmetic. So he
developed his conceptual notation: a language designed to account only
for the conceptual content of a judgement.
Frege compares his concept script to ordinary language by way of an
analogy. The human eye, 'because of the range of its applicability and
because of the ease with which it can adapt itself to the most varied
circumstances, has a great superiority over the microscope' (CN 105).
However, for the purpose it was designed the magnification of microscopic
objects the microscope is far superior to the human eye. What makes a
microscope useful for this purpose makes it useless for any other purpose. In
a similar way, Frege's concept script does not have the wide range of use and
expressive ability of ordinary language because it was designed for one sole
purpose: to ensure precise gap-free chains of inferences.
Frege then asks logicians not to allow the unfamiliarity of his conceptual
notation to prevent them from appreciating the advances in analytic
power that his new logic provides:
The mere invention of this 'conceptual notation', it seems to me, has
advanced logic. I hope that logicians, if they do not allow themselves to
be frightened off by the first impression of unfamiliarity, will not refuse
their assent to the innovations to which I have been driven by a necessity
inherent in the subject matter itself. These deviations from the
traditional find their justification in the fact that logic up to now has
always confined itself too closely to language and grammar. In
particular, I believe that the replacement of the concepts of subject and
predicate by argument and function will prove itself in the long-run. It is easy
to see how regarding a content as a function of an argument leads to the
formation of concepts. Furthermore, the demonstration of the connection between the meanings of the words: if, and, not, or, there exists,
some, all,2 and so forth, may deserve notice. (CN 106-7)
Frege's replacement of the traditional subject/predicate form of judgement
by a function/argument form of judgement was what enabled him to
38
expand the analytic power of logic beyond the scope of traditional logic.
Take the sentence 'Socrates is wise.' We can symbolize it, using the
subject/predicate form of judgement, as S is P, where '' stands for Socrates
and 'P' stands for wisdom. We can symbolize it, using the function/
argument form of judgement, as G(a), where 'G" stands for wisdom and V
stands for Socrates.
Now take the sentence 'Alexander is greater than Philip.' Using the
subject/predicate form of judgement we would again symbolize it as S is P,
where '$' stands for Alexander and 6P' stands for the concept of being
greater than Philip. However, using modern logical symbolism in place of
Frege's symbolism, we can symbolize the sentence as G(a,b) where theaaaaa
function 'G" stands for the relationship x is greater than y and the
arguments 'a' and 'b' stand respectively for Alexander and Philip. Frege's
function/argument form of judgement is able to account for the relational
structure of the sentence in a way that the traditional subject/predicate
form cannot.
Frege then introduces the conditional A B, if A then B, as his basic
form of inference. He defines it as affirmed unless the antecedent is
affirmed and the consequent is denied. (In later works Frege replaced the
terms 'affirmed' and 'denied' with 'true' and 'false'.) Frege's claim can be
illustrated using the method of truth-tables.
Table 2.1
Truth-table
A -> B
T
T
F
F
T
F
T
T
F
T
Using the truth-table method a conditional is defined as true in every case

except where the antecedent is true and the consequent is false. Frege then
introduces the symbol for negation '= ' and defines conjunction 'A and B'
and disjunction 'A or By in terms of the conditional and negation. Where '&'
stands for 'and' and ' V ' stands for 'or', '(A & B)' is defined as '- (A > -aaa
B)\ 'it is not the case that if A then not B9 and 6(A V B)' is defined as '(-A
B)\ 'if not A then B\ Frege's new function/argument conception of
judgements together with these forms of inference enables him to give a
logical analysis of arguments such as this: 'If Alexander is greater than
Frege's Logic
39
Philip and Aristotle is greater than Alexander, then Aristotle is greater than
Philip.' as '((G(ab) & G(ca)) -> G(cb))\ where V stands for Aristotle.
Frege then introduces the universal quantifier 6(x)\ 'all', 'every'. With
the help of the negation sign the existential quantifier 6(3x)\ 'some xy can
be defined as ' (x) F(x)\ 'it is not the case that every x does not fall
under the concept F.3 Take the sentence 'Romeo loves Juliet.' We can
symbolize this as 6L(rj)\ where 'L' stands for the concept of love, V for
Romeo and 'j' for Juliet. If we replace V with the variable V we obtain
the open sentence 'L(xj)' and if we replace 'j' with the variable 'j' we
obtain the open sentence 6(Lxy)\ If we attach the universal quantifier (y)
to this open sentence we obtain the open sentence '(y) (Lxy)\ If we attach
the universal quantifier (x) we get the closed sentence 6(x)(y) (Lxy)\
'Everyone loves everyone.' If we replace '(y)' with the existential
quantifier '(3y)' we get the sentence 6(x)(3y) F(xy)\ 'Everyone loves
someone.' If, instead we replace '(x)' with '(^x)' we get the sentence '(^x)
(y) (Lxy)\ 'Someone loves everyone.' Traditional logic was unable to
analyse the relational structure and multiple generality expressed by such
sentences. Frege's conceptual notation could. With these logical advances
Frege set out to explore how far logic could go in explaining the concepts,
definitions and truths of arithmetic. However, before this logical enterprise
was carried out more philosophical groundwork was required.
Definition of the Symbols

(i) Constants and variables
Frege begins his definition of the symbols of his new logic by distinguishing
between two different kinds of symbols. Letters represent 'either a number
left undetermined or a function left undetermined. This indeterminateness
makes it possible to use letters for the expression of the general validity of
propositions, as in
(a + b)c = ac + be
The other kind consists of such symbols as + , , ^/, 0, 1, 2;5 each of which
has its own specific meaning' (Bgf 111). Frege, then, makes a fundamental
distinction in his conceptual notation between variables and constants.
The distinction is made upon the basis of whether the symbol in question
represents a number or a function left undetermined, or whether it 'has its
40
own specific meaning'. While the letters in '(a + b)c = ac + be' are
specific in so far as they represent numbers in general, as opposed to people
in general or fish in general, they are indeterminate in so far as they do not
refer to any specific numbers. A paraphrase of the sentence would go
something like this: 'Adding any number a to any number b and
multiplying by any number c yields the same results as multiplying a and b,
b and c and adding the resulting two sums.' In contrast, symbols such as
' +', '', '2' refer to a determinate number or function. As he explains a few
lines later:
I ... divide all the symbols I employ into those which one can take to signify
various things and those which have a completely fixed sense. The first are the
letters, and these are to serve mainly for the expression of generality. But we
must insist that a letter, for all its indeterminateness, should retain
throughout the same context the meaning which we first gave it. (CN 111)
According to Frege, a symbol is determinate if it has its 'own specific
meaning' and if it possesses a 'completely fixed sense'. Presumably, a specific
numeral like '2' has its own specific meaning because there is one and only
one specific object, the number 2, that it refers to. Presumably, a term has a
completely fixed sense when whatever it refers to remains the same
throughout all of the sentences that it is contained within. Variables are
letters that can be taken 'to signify various things' and which represent either
a number or a function left undetermined. They are used 'for the expression
of generality'. If the variables 'a' and 'b' are said to range over the natural
numbers, say, then the letter 'a' can signify various things depending upon
the value it is assigned. Thus whatever meaning it does have is determined
not through its designating a determinate object but by ranging over a class
of objects. It is not one particular positive integer that gives the variable its
significance but the class of positive integers that it is said to range over.
What must always be kept in mind, Frege warns, is that while there is
no determinate positive number that the variable signifies, whatever
course of values 'a' is said to range over must remain the same in whatever
sentence of a proof it is contained within, 'a' cannot range over the positive
numbers when contained in one sentence of a proof and the negative
numbers when contained in another sentence of the same proof. For a
proof to be cogent, the symbols used in its construction must maintain the
same meaning in every instance that they are used.
Later in his career, Frege rejected the explanation of variables as
signifying various things depending upon the context of use. In the
Frege's Logic
41
posthumously published essay, 'Logical Defects in Mathematics', he cites

the definition of a variable from a recent textbook on higher analysis by a
mathematician named Emanuel Gzuber. ' "By a real variable, we
understand a number that is indeterminate at the outset, and which,
depending on the problem in which it occurs, can assume indefinitely
many real values" ' (PW 160).
This definition seems suspiciously close to the one Frege gives in the
Begriffsschift, where a variable signifies various things depending on the
kind of judgement it is contained within, while a constant has a completely
fixed sense. The problem with such a conception of a variable, complains
Frege, is that the notion of an indeterminate number is incoherent. Is there
a class of indeterminate prime numbers as well as a class of determinate
prime numbers? Is there a class of indeterminate rational and irrational
numbers that is distinct from the class of determinate rational and
irrational numbers? In another essay, 'Logic in Mathematics', he says:
'Where they do not stand for an unknown, letters in arithmetic have the
role of conferring generality of content on sentences, not of designating a
variable number; for there are no variable numbers' (PW 237). A symbol
can play a role in a judgement without referring to an object or concept.
As Benson Mates explains: 'Variables are best regarded simply as letters of
the alphabet and not as things that vary (or as names of things that vary).
They are used for a number of purposes, one of the most important of
which is to facilitate the expression of generalizations' (Mates 1972, 24).
(ii) Judgements
In sections 2, 3 and 4 of the Begriffsschrift Frege introduces his new
conception of judgement. He begins, in section 2, with the introduction of
the judgement-stroke:
A judgement will always be expressed with the aid of the symbol '|-'
which stands to the left of the symbol or combination of symbols giving
the content of the judgement. If we omit the small vertical stroke at the
left of end of the horizontal one, then the judgement is to be
transformed into a mere combination of ideas of which the writer does
not state whether or not he acknowledges its truth. (CN 111)
Frege says if we take '|-A' to designate the judgement: 'Opposite magnetic
poles attract each other', then 'A' would designate the circumstance that
42
opposite magnetic poles attract each other. This should not be understood
as the expression of a judgement, 'but should simply evoke in the reader
the idea of the reciprocal attraction of opposite magnetic poles, perhaps,
say, in order to derive some conclusions from it and with these test the
correctness of the thought' (CN 112). To attach a symbol to the
judgement-stroke ' -A', is to assert that the judgement expressed by the
symbol is true. To place the horizontal stroke by itself in front of the
judgement, 'A' is to entertain the possibility that the judgement might be
true. Frege suggests that the relationship between the horizontal stroke
and a proposition can be paraphrased, 'by means of the words
"circumstance that" or "the proposition that"' (CN 112). One can take
the thought expressed by a proposition as a mere combination of ideas
whose truth is not acknowledged or one can assert that the judgement is
true. In both cases it is the same content that is being considered.
Frege then makes a distinction between 'assertable and unassertable
contents'. A sentence expresses assertable content if it has a truth-value.
Only sentences expressing assertable content can form a judgement when
attached to the assertion symbol. So, on the one hand, there is assertable
content expressed by sentences. On the other hand, there is unassertable
content, expressed by words or phrases out of which sentences are
composed. The truth of the same assertable content can be merely
entertained or it can be asserted.
The two parts of Frege's judgement-stroke show similarities to the
distinction between assertoric and problematic judgements in Kant. To
understand Kant's distinction between assertoric and problematic
judgements it will help to start with an examination of his 'Table of
Judgements'. According to Kant,
all judgments are functions of unity among our representations; instead
of an immediate representation, a higher representation, which
comprises the immediate representation and various others, is used in
knowing the object, and thereby much possible knowledge is collected
into one. Now we can reduce all acts of the understanding to judgments,
and the understanding may therefore be represented as a faculty of
judgment. For, as stated above, the understanding is a faculty of thought.
Thought is cognition by means of concepts. But concepts as predicates
of possible judgments, relate to some representation of a not yet
determined object. (A69/B94)
A judgement is a function that unifies the immediate representation
Frege's Logic
43
(concept of a subject denoting an intuition or a concept) with various

mediate representations (concepts of the predicate) into a higher
representation. Thus the judgement expressed by the sentence 'This ball
is big, red and round' unifies various representations of the ball into a
judgement whose truth can be evaluated. The concepts of being a ball and
of being big, round and red by themselves are neither true nor false. It is
only when an object, the ball, is said to fall under these concepts that a
judgement is produced that can be either true or false. According to Kant,
all acts of the understanding can be reduced to acts of judgement, and thus
the understanding can be characterized as a faculty of judgement.
Concepts are predicates of possible judgements that can be related to an
indeterminate range of objects. For example, one can understand what
metal is by bringing it under the concept body. An object can only be
understood by being brought under a concept, and this act of bringing an
object under a concept is an act of judgement. Thus the understanding is
the faculty of judgement.
Kant uses this conclusion that the understanding is the faculty of
judgment as a clue to the discovery of the categories or the pure concepts
of the understanding. As he explains, the 'functions of the understanding
can . . . be discovered if we can give an exhaustive statement of the
functions of unity in judgments' (A69/B94). He begins with the claim that
'If we abstract from all content of a judgment, and consider only the mere
form of understanding, we find that the function of thought in judgment
can be brought under four heads, each of which contains three moments'
(A70/B95):
1.
2.
3.
4.
Quantity: universal, particular, singular

Quality: affirmative, negative, infinite
Relation: categorical, hypothetical, disjunctive
Modality: problematic, assertoric, apodeictic.
Robert Paul Wolff gives a helpful explanation of the Table of Judgements:

The Table of Judgments can best be understood if we develop it step by
step, indicating the meaning of each of the headings. Let us begin with
the subject term S and a predicate term P, and the connecting word or
copula 'is'. The basic framework for a judgment is then 'S is P'. In
specifying this matrix, we can assert the predicate P of all the objects to
which S refers, or only of a portion of them. This gives us the universal
from of judgment, 'All S are P', and the particular form, 'Some S are P'.
44

Each of these forms, now, can be either asserted or denied, so there are
two types of universal judgments and two types of particular judgments,
namely: 'All S are P', 'It is not the case that all S are P', 'Some S are P',
'It is not the case that some S are P.' The two clumsy negative
judgments can be reworded, respectively, as 'Some S are not P' (the
negative of'All S are P') and 'No S are P', (the negative of'Some S are
P'). Finally, any one of these four forms of judgment can be asserted
either problematically ('Possibly S is P'), assertorically ('S is P'), or
apodeictically ('Necessarily S is P'). (Wolff 1973, 64-5)
Every judgement, according to Wolffs account of Kant, falls under one of

the logical forms of judgement from each of the four groups. For example,
this A is B is a singular, affirmative, categorical, assertoric form of
judgement.
In the Logic, Kant says:
As to the modality by which moment is determined the relation of the
entire judgment to the faculty of cognition, judgments are either
problematic, assertoric, or apodeictic. Problematic judgments are accompanied with the consciousness of the mere possibility, assertoric judgments
with the consciousness of the actuality, apodeictic judgments, lastly, with
the consciousness of the necessity of judging. (Logic 30, 11415)
If I am unsure whether a given judgement is true or false, then my
understanding of the meaning of the judgement is accompanied by a
consciousness of the mere possibility of the judgement being true. As Kant
explains:
This moment of modality indicates the manner in which something is
asserted or negated in judgment: Whether one does not make out
anything about the truth or untruth of a judgment, The soul of man may be
immortal] or whether one determines something about its truth or untruth,
as in the assertoric judgment, The human soul is immortal] or lastly, whether
one expresses the truth of a judgment even with the dignity of necessity,
as in the apodeictic judgment, The soul of man must be immortal. This
determination of the merely possible, actual or necessary truth, thus
concerns only the judgment itself, not at all the matter that is judged. (Logic
30, 115)
Kant claims that the difference between a proposition and a judgement
Frege's Logic
45
rests upon the difference between a problematic and an assertoric

judgement. A proposition is a judgement whose truth-value has yet to
be determined. In other words a proposition is a problematic judgement.
Once a proposition is thought to be true, it becomes an assertoric
judgement. It has a truth-value. Once an assertoric judgement is thought
to be necessary it becomes an apodeictic judgement. To think of a
proposition as necessary is to bring it under one of the twelve forms of
judgement. As Kant explains in the Critique:
The apodeictic proposition thinks the assertoric as determined by these
laws of the understanding, and therefore as affirming a priori] and in this
manner it expresses logical necessity. Since everything is thus
incorporated in the understanding step by step inasmuch as we first
judge something problematically, then maintain its truth assertorically,
and finally affirm it as inseparably united with the understanding, that
is, as necessary and apodeictic we are justified in regarding these three
functions of modality as so many moments of thought. (A76/B101)
A moment of thought under the heading of modality is a function of
modality. A function, according to Kant, is 'the unity of the act of bringing
various representations under one common representation' (A68/B93). Since
the understanding is a faculty of judgement, the act of bringing various
representations under one common representation is an act of judgement.
Hence a function is an act of imposing a form of judgement on already
existing concepts. Where the function is apodeictic under the heading of
modality the form of judgement would be something like 6S is necessarily P\
Here it should be pointed out that unlike the headings of quality,
quantity and the categorical form of judgement under the heading of
relation, where the central concern is with the relationship between the
subject concept and the predicate concept in a judgement, modality
concerns the whole judgement. As Kant explains:
The modality of judgments is a quite peculiar function. Its distinguishing
characteristic is that it contributes nothing to the content of the
judgment . . . but concerns only the value of the copula in relation to
thought in general. Problematic judgments are those in which
affirmation or negation is taken as merely possible (optional). In
assertoric judgments affirmation or negation is viewed as real (true),
and in apodeictic judgments as necessary. (A74-5/B99-100)
46
The modality of a judgement has nothing to do with the relationship

between its subject and predicate concepts but with how that relationship
is thought. The same judgement can be thought under three different
modalities. Tt is possible that the human soul is immortal.' 'The human
soul is immortal.' Tt is necessary that the human soul is immortal.' The
copula acts as the bond that unifies the subject and predicate concepts into
a judgement. The form of judgement 6S is necessarily P' expresses a certain
relationship between a proposition and the person who is thinking the
proposition. If she thinks the proposition as necessary, then the copula has
the value of necessity. It is the person's thinking the proposition as
necessary that converts it into an a priori judgement.
For example if Sara states that 'The human soul must be immortal', she
is expressing a judgement that falls under the form of judgement 'S is
necessarily P\ The act of judgement is possible because Sara, like all other
human beings, possesses the a priori faculties of understanding and
sensibility. In this act of judgement she combines the various concepts
into a unity by bringing them under logical forms of judgement. The
conceptual content of the judgement consists of the concepts of being
human, having a soul and immortality. These concepts are unified under
the singular, affirmative, categorical forms of judgement to produce the
judgement 'The human soul is immortal.' It is conceivable that when Sara
first encountered this claim she thought that it might be true. Upon further
reflection she determined that it is true, and finally, after more reflection,
she determined that the human soul must be immortal. In all three cases it
is the same judgement that she is reflecting on. She just thinks the
judgement under a different modality each time. As Kant explains: 'The
apodeictic proposition thinks the assertoric as determined by these laws of
the understanding [the forms of judgement], and therefore as affirming a
priori] and in this manner it expresses logical necessity' (A76/B101).
Both Kant and Frege, then, draw a distinction in their logic between a
truth being entertained and a truth being asserted. Kant accounts for this
distinction in terms of two forms of judgement under the heading of
modality: S is possibly P and S is (actually) P. Frege accounts for the
distinction by decomposing his judgement-stroke into two components. The
content-stroke '' signifies that the truth of a judgement is being entertained
but not acknowledged. When the judgement-stroke is attached to the
content-stroke, the resulting symbol ' -', also called the judgement-stroke,
signifies that a judgement is being asserted. Kant's forms of judgement and
Frege's content and judgement-strokes concern the relationship between the
content of a possible judgement and a judging subject.
Frege's Logic
47
For Kant, the difference between a problematic judgement and an

assertoric judgement concerns not the relationship between a subject
concept and predicate concept but the value that the copula assigns to that
relationship. As Kant explains: 'The modality of judgments is a quite
peculiar function. Its distinguishing characteristic is that it contributes
nothing to the content of the judgment . . . but concerns only the value of
the copula in relation to thought in general' (A74/B99100). In this
respect a more apt way of expressing Kant's understanding of the
problematic form of judgement would be 6S possibly is P' instead of 6S is
possibly P\ If the relationship between the subject concept and the
predicate concept is thought by the judging subject to possibly be true,
then the judgement is problematic. If the judgement is actually thought to
be true, then the judgement is assertoric.
The copula serves a variety of functions for Kant. As he explains in the
Vienna Logic, notes of a series of lectures given by Kant in the early 1780s:
All affirmative propositions show their affirmation through the copula
est, which copula indicates the relation of two concepts. When the copula
est occurs simpliciter, it means the connection of two concepts when the
copula est is affected with the non, it means the opposition of the two
concepts and indicates that the one concept does not belong to the
other, or is not contained in the sphaera of the other. (Vienna Logic 930,
370)
The copula connects the predicate concept to the subject concept and it
indicates what kind of relationship holds between the concepts. If the
predicate concept is attached to the subject concept with the copula alone
(S is P) the copula affirms the connection of the two concepts. Negating the
copula (S is-not P) indicates that the predicate concept is denied of the
subject concept. Here, from Frege's perspective, Kant conflates affirmation
and assertion. To affirm that a judgement is true is not the same as
asserting it is true. For while the opposite of affirmation is denial, it is
possible to assert both an affirmative and a negative judgement. Assertion
plays a different role in a judgement than affirmation does. To assert a
judgement is to claim that it is true. To affirm a judgement is to claim that
the subject concept falls under the predicate concept. To negate a
judgement is to deny that the subject concept falls under the predicate
concept. Since Socrates falls under the concept of wisdom, the judgement
that Socrates is wise is true. Because Alcibiades does not fall under the
concept of wisdom the judgement that Alcibiades is not wise is true. For
48
Kant, then, the copula performs a variety of interrelated but nevertheless

distinct functions. It signifies the subsumption of a concept or object under
a higher concept. It is used to affirm or deny that a predicate belongs to a
subject, and it is used to assert that a judgement is true. Kant and the
logical tradition that he followed did not clearly distinguish these different
logical roles that were mixed together in the copula or predicate of a
sentence.
Frege is able to distinguish clearly these different logical roles for the
first time because he replaces the subject/predicate form of judgement with
his function/argument conception of judgement. In doing so, he denies the
copula any significant logical role in a judgement. Frege's conceptual
notation distinguishes between an object falling under a concept (F(<z))
from the subsumption of a concept under a higher concept ((x) F(x)).
He makes an affirmative judgement (F(a), 'Socrates is wise') basic and
detaches negation from the copula and creates the negation sign, a
logical operator that attaches to the whole judgement instead of the
copula (-F(b), 'It is not the case that Alcibiades is wise'). He then creates
the judgement-stroke to insulate the content of a judgement from the act
of asserting a judgement.
In a later posthumously published essay, Frege distinguishes between a
proper name that refers to an object, which is a complete whole, and a
predicative expression that refers to a concept, which is unsaturated.
We count the copula 'is' as belonging to this [predicative] part of the
sentence. But there is usually something combined with it which here
must be disregarded: assertoric force. We can of course express a
thought, without stating it to be true. The thought is strictly the same,
whether we merely express it or whether we also put it forward as true.
(PW 111]
Here Frege has incorporated the copula into the functional part of a
judgement. To account for the assertoric force of a judgement, a force
expressed by the copula in traditional logic, Frege employs the judgementstroke.
In the Be griffsschrift, Frege had not yet entirely separated the role of the
copula from his judgement-stroke. Thus he says that the horizontal stroke,
'', 'ties the symbols which follow it into a whole; and the assertion, which is
expressed by means of the vertical stroke at the left end of the horizontal one, relates to
this whole' (CN 112). Presumably what Frege means by this is that
application of the content-stroke transforms a concatenation of symbols
Frege3 s Logic
49
without a truth-value into a unified whole, a judgement, with a truthvalue. However, unlike the copula, which is placed between subject and the
predicate in a judgement of traditional logic, the content-stroke does not
appear between the function and the argument but in front of the
judgemental content. In later writings Frege offers a different explanation
of the content-stroke. Thus in 'Boole's Logical Calculus and the ConceptScript', an essay written shortly after the publication of the Begriffsschrift,
Frege says that, '[t]he content-stroke is horizontal, it is always prefixed to
the expression of a content of possible judgement' (PW 11). Here Frege does
not say that the role of content-stroke is to tie symbols into a whole but to
signify that what follows the content-stroke is the content of a possible
judgement. This signifies an important difference. For the latter explanation gives no role to the content-stroke in unifying the symbols into a whole
judgement. As he explains in a later article 'Negation' (published in 1919),
a common mistake about the content of a judgement is
the view that the judging subject sets up the connexion or order of the
parts in the act of judging and thereby brings the judgment into
existence. Here the act of grasping a thought and the acknowledgment
of its truth are not kept separate . . . But even the act of grasping a
thought is not a production of the thought, is not an act of setting its
parts in order; for the thought was already true, and so was already
there with its parts in order, before it was grasped. (GB 1267)
Here Frege stands in sharp contrast to Kant who believed that it was an
act of the understanding that combines concepts into a judgement. As
Kant explains in the Critique:
all combination be we conscious of it or not, be it a combination of the
manifold of intuition, empirical or non-empirical, or of various concepts
- is an act of the understanding. To this act the general title 'synthesis'
may be assigned, as indicating that we cannot represent to ourselves
anything as combined in the object which we have not ourselves
previously combined, and that of all representations combination is the
only one which cannot be given through objects. Being an act of the selfactivity of the subject, it cannot be executed save by the subject itself.
(B130)
All combination, whether of intuitions or concepts, is an act of the
understanding. A judgement is an act of the understanding that combines
50
concepts and/or intuitions into a unity that can be judged to be either true
or false. Unlike Kant and the logical tradition that considered an act of
judgement as an essential condition for the unity of a judgement, Frege
believed that the unity of a thought is not a product of an act of
judgement. Thoughts already exist as wholes that are grasped by an act
of judgement. As D. Greimann explains:
To judge is, therefore, not to unite ideas, but to acknowledge something
which is already united as true. In particular, the basic cognitive
operation is, for Frege, not 'to say something of something' (TI KATA
TINOS), but to judge something as true. (Griemann 2000, 220)
For Frege, judgement is not an act of saying something about something
but an act of judging that something is true. Later in his career, after he
had drawn his famous distinction between the sense and the reference of an
expression, Frege introduced thoughts as the senses of sentences, with truth
or falsehood being their reference. As he explains in 'Thoughts' (1919):
We are not owners of thoughts as we are owners of our ideas. We do not
have a thought as we have, say, a sense-impression, but we also do not
see a thought as we see, say, a star. So it is advisable to choose a special
expression; the word 'grasp' suggests itself for the purpose. To the
grasping of thoughts there must then correspond a special mental
capacity, the power of thinking. In thinking we do not produce
thoughts, we grasp them. For what I have called thoughts stand in the
closest connection with truth. What I acknowledge as true, I judge to be
true quite apart from my acknowledging its truth or even thinking
about it. That someone thinks it has nothing to do with the truth of a
thought. (CP 368)
Frege's new conception of judgement requires a new conception of how a
judgement is understood as well as a new conception of the relationship
between the content of a judgement and the judging subject. Judgement is
no longer seen as an act of affirming or denying a predicate of a subject but
of grasping a thought.
Whereas the activity of the understanding plays an essential role in
Kant's account of the unity of judgements, Frege seeks to avoid any kind of
psychological explanation, or for that matter an epistemological explanation, of the unity of a judgement. A judgement forms a unity because
concepts are unsaturated and objects are not. The unity of a judgement,
Frege's Logic
51
for Frege, can be likened to two pieces in a puzzle that together form a
larger whole. On such an account there is no need to invoke the
synthesizing power of the mind to explain the unity of a judgement. More
will be said on this topic later.
(Hi)
Conceptual content
Frege introduces his notion of conceptual content in section 3 of the

Begriffsschrift by contrasting it with the traditional logical conception of a
judgement. 'A distinction between subject and predicate does not occur in my
way of representing a judgement' (CN 112). Frege illustrates his new way
of representing a judgement with an example. In the two propositions: 'At
Plataea the Greeks defeated the Persians' and 'At Plataea the Persians
were defeated by the Greeks' there is little difference in sense between the
two sentences even though the active and passive constructions result in a
reversal of the roles of subject and predicate. Despite this shift in roles,
however, the two propositions express the same content. 'Now I call the
part of the content which is the same in both the conceptual content. Since only
this is meaningful for our 'conceptual notation', we need not distinguish
between propositions which have the same conceptual content' (CN 113).
Frege claims that the conceptual content is not sensitive to the distinction
between active and passive voice because active and passive voice, in this
example, do not affect the truth-value of the two sentences. The judgement
is true because of the meaning of the words, the logical structure of the
judgement, and the fact that in 479 BC the Greeks defeated the Persians at
the Battle of Plataea. Whether you express this fact in the active or the
passive voice does not affect the truth-value of the sentence. Hence the two
sentences have the same conceptual content.
Frege then goes on to isolate conceptual content 6 not only from its
assertability conditions and the grammatical (and traditional logic)
distinction between subject and predicate but from all contextual and
psychological factors that may contribute to the meaning of a sentence:
Now all aspects of language which result only from the interaction of
speaker and listener for example, when the speaker considers the
listener's expectations and tries to put them on the right track even
before speaking a sentence have nothing corresponding to them in my
formula language, because here the only thing considered in a judgment
is that which influences its possible consequences. Everything necessary
52

for a correct inference is fully expressed; but what is not necessary
usually is not indicated; nothing is left to guessing. (CJV 113)
Ordinary language cannot be adequately relied upon to isolate that

content of an expression that contributes to the truth-value of whatever
judgement it is contained within. Frege's concept script was designed to
assign significance only to this conceptual content of a judgement. The
development of the concept script was meant to take all guesswork out of
logical inference. Only that content that contributes to the truth-value of a
judgement and thus to its role in a proof is relevant. In order for
conceptual content to maintain the integrity of its role it must be isolated
from a host of possibly infecting factors. The psychological and contextual
factors that enable human beings to communicate with each other in
ordinary discourse can play no significant role in the nature of such
content. Neither can the distinction between subject and predicate, nor
can mood or feelings expressible in ordinary discourse.
(iv) Conditionality and negation

Having introduced his notion of judgement, Frege moves on to a
discussion of inference. In the Be griffsschrift, Frege conceived of
judgements in terms of affirmation and denial instead of truth and
falsity (as he later conceived of them). He defines the affirmation
conditions of the conditional, Tf P, then QJ, in terms of the different
possible affirmation conditions of the component sentences of the
conditional. A conditional is affirmed in every case except where the
antecedent is affirmed and the consequent is denied.
Frege observes that in cases where the consequent of a conditional is
affirmed, the conditional is true by default. The same holds true in cases
where the antecedent is denied. He then shows how conditional
judgements of more than two elements can be constructed. On the basis
of this definition of the conditional, Frege introduces the logical rule of
inference modus ponens (If P is affirmed, then Q^is affirmed, P is affirmed,
therefore Q^is affirmed) as the basic rule of inference in his logical system.
Frege observes that Tn logic people enumerate, following Aristotle, a
whole series of modes of inference. I use just this one . . . Accordingly, since
it is possible to manage with a single mode of inference, perspicuity
demands that we do so' (CJV 119-20).7
Frege then introduces the sign for negation as an independent logical
Frege's Logic
53
operator that is attached to the complete judgement instead of to the

copula or predicate of the judgement as in traditional logic. He then uses
his affirmation tables to define conjunction (A and E) and disjunction (A
or B] in terms of the conditional and negation. Frege observes that he
could just as easily have started with disjunction or conjunction as his basic
rule of inference and defined conditionality in terms of one of these other
rules. 'I chose the other way because deduction seemed to me to be
expressed more simply that way' (CN 123).
(v) Identity of content

In the Begriffsschrift Frege defines 'identity of content' to mean that 'the
symbol A and the symbol B have the same conceptual content, so that we
can always replace A by B and vice versa' (CN 126). Conceptual content
for Frege was that content of a judgement that contributes to its truthvalue. So if A and B have the same conceptual content, then they both
play the same role in any inference they are contained within, and thus
they can be interchanged without affecting the truth of the judgements
they are contained within. Frege opens section 8, which is entitled
'Identity of Content', with the claim that
Identity of content differs from conditionality and negation by relating
to names, not to contents. Although symbols are usually only
representatives of their contents - so that each combination [of symbols
usually] expresses only a relation between their contents - they at once
appear in propria persona as soon as they are combined by the symbol for
identity of content, for this signifies that the names have the same
content. Thus, with the introduction of a symbol for identity of content,
a bifurcation is necessarily introduced into the meaning of every
symbol, the same symbols standing at times for their contents, at times
for themselves. (CN 124)
In a judgement involving negation or conditionality, the symbols in the
corresponding sentence of the concept script denote objects. However, in the
case of identity the symbols indicate themselves. So in the sentence 'Venus is a
planet', the word 'Venus' refers to the planet Venus. However, in the sentence
'Venus is the Morning Star', 'Venus' does not refer to the planet Venus but to
the word 'Venus'. What the sentence is saying is that the word 'Venus' and
the word 'Morning Star' refer to the same object. In the Begriffsschrift, of
54
course, Frege had not yet made his famous distinction between the cognitive
significance (sense) and truth-value (reference) of a sentence. Conceptual
content accounts both for the cognitive significance of an expression as well as
the inferential role the expression plays in any sentences it is contained within.
So when he comes to the problem of accounting for the difference of cognitive
value between 6a = a' and 6a = b' (where 6a = b' is true), he sees his only
option as claiming that identity is a relationship between symbols. Since 6a =
rf and '<2 = b' have the same truth-value, and thus play the same inferential
role in whatever proofs they are contained in, they must have the same
cognitive significance. Therefore sentences expressing identity cannot be
described as a relation between objects. Frege concludes that the only
alternative is that identity is a relation between words.
Frege turns to a rather obscure geometrical example to explain his
point. The gist of the example is that a point on a circle can be determined
in two different ways either as the point of the intersection of line A and
line B or 'directly through intuition' (CN 125, replacing 'perception' with
'intuition' as a translation of 'Anschauung7). According to Frege:

A separate name corresponds to each of these two modes of
determination. Thus, the need of a symbol for identity of content rests
upon the following fact: the same content can be fully determined in
different ways; but, that the same content in a particular case, is actually
given by two {different} modes of determination is the content of a judgement.
Before this [judgement] can be made, we must supply two different
names, corresponding to the two [different] modes of determination, for
the thing thus determined. But the judgement requires for its expression
a symbol for identity of content to combine the two names. It follows
from this that different names for the same content are not always
merely an indifferent matter of form; but rather, if they are associated
with different modes of determination, they concern the very heart of
the matter. In this case, the judgement as to identity of content is, in
Kant's sense, synthetic. (CN 125-6)
To say that A = B is not just to say that the same content has two different
names but that there are two different modes of determination by which
the same content or object is presented to a subject. The content of a
judgement of identity is that the object signified by 'A' is the same object
signified by 6B\ The cognitive significance of 'A = B' lies not just in a
difference in words but in a difference in the manner in which the same
content can be determined. If there are two different ways of determining
Frege's Logic
55
the same content of a judgement, then there can be two different names
that correspond to the two different modes of determination. So the fact
that the same content can have two different names need not be just a
linguistic feature about the use of two different symbols, but shows
something much more important: that there are different ways in which
the same conceptual content can be determined. And since these different
modes of determination can differ in cognitive significance, this makes the
identity of the content synthetic in Kant's sense of the term.
(vi) Functions and Generality

In section 9 Frege characterizes a function by way of example. Suppose, he
says, 'that the circumstance that hydrogen is lighter than carbon dioxide'
is expressible in the conceptual notation (CN 126). In such a case, he
observes, the symbol for hydrogen can be replaced by the symbol for
oxygen or nitrogen:
By this means, the sense is altered in such a way that 'oxygen' or
'nitrogen' enters into the relations in which 'hydrogen' stood before. If
we think of an expression as variable in this way, it divides into (1) a
constant component which represents the totality of the relations and
(2) the symbol which is regarded as replaceable by others and which
denotes the object which stands in these relations. I call the first
component a function, the second its argument. This distinction has
nothing to do with the conceptual content, but only with our way of
viewing it. (CN 126)
A function is that 'constant component' of a judgement that 'represents the
totality of the relations' contained in it. An argument is a symbol that
designates an object. Thus, in the sentence 'Hydrogen is lighter than
carbon dioxide', 'hydrogen' and 'carbon dioxide' are arguments that refer
to objects; whereas 6x is lighter than j' is the function-symbol that
represents the relation of one thing being lighter than another. He then
claims that the distinction between function and argument lies not with
the actual internal structure of a judgement's conceptual content but with
our way of viewing the content. The fact that we can analyse a judgement
in terms of a particular relationship between argument and function need
not entail that such an analysis is revealing the objective structure of the
content of the judgement in question. His reason for holding this view lies
56
in his realization that any given conceptual content can be analysed in

different ways. Thus he claims that
The circumstance that carbon dioxide is heavier than hydrogen and the
circumstance that carbon dioxide is heavier than oxygen are the same
function with different arguments if we regard 'hydrogen' and 'oxygen' as
arguments. On the other hand, they are different functions of the same
argument if we consider 'carbon dioxide' as the argument. (CN 1267)
Similarly the sentence 'Carbon dioxide is heavier than oxygen' can be
analysed as speaking of one or two one place functions: x is heavier than
oxygen or carbon dioxide is heavier than j, or as speaking of the two place
function: x is heavier than y. Because of such variability in analysis Frege
considered the function/argument distinction to be something that is
imposed upon the conceptual content of a judgement in the act of analysis
and not something intrinsic to that content. The compositional structure of
the two judgements plays no objective role in determining their conceptual
content. For since the two judgements have the same semantic value (they
can be interchanged in a proof without affecting its soundness), their
compositional structure is irrelevant to their semantic value. Such a view is
consistent, it would seem, with a view that the content of a judgement is a
seamless whole whose parts are purely a product of logical analysis.
However, Frege does not think that this holds true for the content of all
judgements. To illustrate this he compares two propositions:
1. The number 20 can be represented as the sum of four squares.
2. Every positive integer can be represented as the sum of four squares.
Frege argues that 'it appears possible to consider "being representable as
the sum of four squares" as a function whose argument is "the number 20"
one time, and "every positive integer" the other time' (CN 127). However,
he observes, it is clear that
'the number 20' and 'every positive integer' are not concepts of the
same rank. What is asserted of the number 20 cannot be asserted in the
same sense of [the concept] 'every positive integer'; though, of course, in
some circumstances it may be asserted of every positive integer. The
expression 'every positive integer' by itself, unlike [the expression] 'the
number 20', yields no independent idea; it acquires a sense only in the
context of a sentence. (CN 128)
Frege's Logic
57
The concept symbolized by 'the number 20' and the concept symbolized
by 'every positive integer' are not of the same logical type.8 'The number
20' is a proper name with a determinate meaning, namely the number 20,
whereas 'every positive integer' does not name a determinate object but
ranges over a class of objects, namely the positive integers.
In section 10 Frege introduces the notion of a function with more than
one argument place: F(a,b). At the end of the section he observes, 'that the
concept of a function in analysis, which I have in general followed, is far
more restricted than the one developed here' (CN 129). Later in 'Function
and Concept', Frege, following the implications of this expanded conception
of a function, characterizes a concept as a function whose value is a truthvalue and characterizes an argument as an object. This completes his turn
away from the traditional conception of a judgement as consisting of a
subject concept combined with a predicate concept by a copula.
In section 11 Frege introduces the universal quantifier. If we let 'a'
represent 'the number 20' and 'F9 the predicate 'can be represented as the
sum of four squares', then (1) can be represented by 'F(a)'. (1) simply
claims that a determinate object possesses a specific property. To represent
(2) adequately, however, is not as simple a task. What (2) says is that if
something is a positive integer, then it is the sum of four squares, or to put it
in modern notation: 6(x) (Px Fx)\ where 6P' represents the predicate 'is a
positive integer'. While 'the number 20' has a significance independently of
its role in a sentence the same cannot be said of 'every positive integer',
which is represented by a quantifier or second-level concept which 'yields no
independent idea' outside the context of the sentence in which it is used.
There is not one specific positive integer that V is said to designate; instead
it is said to range over the class of positive integers.
(vii) Forms of judgement versus a logical calculus

In examining Frege's introduction of the symbols of his new logical language
I followed the general order of his explication with one exception. I skipped
section 4, which is the last section in Frege's discussion of judgement. In this
section Frege contrasts his new conception of logical judgement with that of
Kant and the older logical tradition. He opens the section with the claim
that 'The following remarks are intended to explain the significance, for our
purposes, of the distinctions which people make with regard to judgements'
(CN 114). While Frege does not mention Kant by name here, the list closely
corresponds to Kant's Table of Judgements. Indeed, in the discussion that
58
follows, Frege mentions nine of Kant's 12 forms of judgement in the same

order that Kant lists them in the Critique.
According to Kant, as shown above, there are 12 basic logical forms of
judgement, falling under four headings:
Table 2.2
1.
2.
3.
4.
Kant's 12 logical forms of judgement
Quantity: Universal, Particular, Singular

Quality: Affirmative, Negative, Infinite
Relation: Categorical, Hypothetical, Disjunctive
Modality: Problematic, Assertoric, Apodeictic
With the exception of the hypothetical and disjunctive forms of

judgements, which concern a relation between judgements, all of the
other forms of judgement in Kant's table concern a relation between a
subject concept and a predicate concept (see B140-1). Thus the
categorical form of judgement: 6S is P' supplies the root for all of the
forms of judgement except the hypothetical and the categorical forms. This
can be seen in the following table, which compares the manner in which
Kant's logic and Frege's (in terms of modern notation) would characterize
the relevant forms of judgement in Kant's Table of Judgements.
Table 2.3 Comparison of Kant's and Frege's logical forms of judgement
Quantity:
Universal
Particular
Singular
KANT:
FREGE:
All S is P
(x) F(x)
Some S is P
(3x) F(x)
This S is P
F(a)
Quality:
Affirmative
Negative
Infinite
KANT:
S is-not P
-F(a)
S is non-P
FREGE:
SisP
F(a)
Relation:
Categorical
Hypothetical
Disjunctive
KANT:
FREGE:
Sis P
F(a)
If A then
A -> B
A or B
A vB
Modality:
Problematic
Assertoric
Apodeictic
KANT:
S possibly is P
S actually is P
S necessarily is P
FREGE:
-F(a)
\-F(a)
Frege's Logic
59
There are several key differences between Kant's conception of a

judgement and Frege's conception. To begin with, Kant conceived
of judgements in terms of a subject and a predicate connected by a copula
in a mental act of judging. Frege conceived of judgements in terms of the
mathematical notions of function and argument. He did not believe that
the unity of a judgement is a function of a mental act of judging. Kant's
logic is grounded in the Table of Judgements, mental faculties that enable
us to form and understand judgements by unifying concepts or judgements
in one act of consciousness. It is an act of judgement that provides the form
of judgement by which the content of the judgement is unified and
understood. Frege's logic is a language or calculus, not a set of mental
faculties. It consists of several basic symbols: constants (2), letters or
variables (a), the horizontal stroke (^4) and judgement-stroke (|A), the
conditional (>), negation (A), identity of content (a = b), the function
and argument(s) (F(a), F(a,b)) and the universal quantifier ( ( x ) ) . From
these basic elements, Frege defines other logical concepts such as the
existential quantifier, conjunction and disjunction. This basic vocabulary,
together with rules of formation and inference, enables Frege to construct
gap-free proofs of impressive scope and complexity.
One of Frege's most important advances over Kant's conception of
judgement was his treatment of quantity. Kant treated the quantity of
a judgement in terms of three forms of judgement where the quantity of
the judgement is inseparably attached to the subject of the judgement.
Frege detached the quantity of a judgement from the subject of the
traditional subject/predicate form of judgement and turned it into an
independent logical operator. Just as the relational structure of a
judgement such as 'Alexander is greater than Philip' cannot be captured
by any of Kant's logical forms of judgement, neither can judgements
expressing multiple generality such as 'Some man is greater than all
men.' These advances enabled Frege to give a logical analysis of a
numerical sequence. Take the sentence 'Two is followed by three in the
number sequence.' This could be symbolized as '2/3'. We can introduce
the predicate 6F to stand for the function ( ) is a number and replace '2'
with the variable V and '3' with the variable '_/ to produce the logical
formula: (Fx & xfy) Fy, Tf x is a number and y follows x in the
number sequence, thenj is a number.' We can then attach quantifiers to
produce a judgement such as (x) (y) ((Fx & xfy) > Fy) 'For every x
and every j, if x is a number and y follows x in the number sequence, then
y is a number. Frege's logic can also quantify over properties as well as
objects. Thus one could define identity as a = b <-> (F) (Fa <-> Fb), 'a is
60
identical to b if and only if for every property F, a has F if and only if, b
has F\
Frege observes that the traditional distinction between universal and
particular judgements (he does not mention singular judgements) is not a
distinction between the judgements but between contents of judgments.
While the tradition speaks of a particular or universal judgment, Frege
insists that '[t]hey should say, "a judgement with universal content", "a
judgment with a particular content" ' (CN 114). A judgment is universal
or particular regardless of whether or not it is asserted. 'These properties',
he explains, 'belong to the content even when it is put forth, not as a
judgment, but as an [unasserted] proposition' (CN 114). With his
introduction of the judgement-stroke, and the replacement of the
subject/predicate distinction with the function/argument distinction, Frege
is able to isolate the conceptual content of a judgement from its
assertability conditions. This enables him to attach quantifiers to the
content of a judgement regardless of whether the content is asserted or not.
Similarly negation is attached to the content of the judgement, not the
asserted judgement. There is no need for an explicit distinction between an
affirmative and a negative form of judgement. A negative judgement
results when a negation-stroke is attached to the content of an affirmative
judgement (Frege does not mention infinite judgements).
Frege then says that the 'distinction between categorical, hypothetical,
and disjunctive judgments appears to me to have only grammatical
significance [ftn. The reason for this will be brought out by the whole of
this work]' (CN 114). Because the categorical form of judgement is
grounded in the subject/predicate distinction it can sometimes fail properly
to account for the logical form of a judgement. Thus the move from subject
and predicate to function and argument enables Frege to account for
relational expressions such as 'Alexander is greater than Philip', (G(ab)}.
As shown above, such expressions are impervious to analysis on the
traditional subject/predicate conception of judgement. Not only does the
categorical form of judgement fail properly to account for the logical form
of some judgements, in other cases it attributes a difference in logical form
where the difference is not logically significant. As Frege observed earlier
in a discussion of conceptual content, the difference between the active and
passive voice of a judgement has no effect on its conceptual content even
though the roles of the subject and predicate are reversed in the two types
of judgement.
According to Frege, the difference between an apodeictic judgement
and an assertoric judgement is that the former 'suggests the existence of
Frege's Logic
61
general judgments from which the proposition can be inferred, while the
assertoric lacks such an indication' (CN 114). Frege concludes that to call a
proposition necessary is merely to give 'a hint about my grounds for
judgment' (CN 114). But this is just a psychological attitude that I have
toward a judgement and thus does not affect its conceptual content. Thus
the apodeictic form of judgement, together with the concept of necessity, is
dismissed as logically insignificant, while the problematic and assertoric
forms are incorporated into the judgement-stroke to signal the difference
between the entertainment of a judgement and its assertion.
Another key difference between Kant and Frege is the order of
exposition of their respective Logics. Kant follows the order of traditional
logical texts, starting with an explication of concepts, moving to judgement
and then to inferences. In his exposition of concepts Kant distinguishes
between their form (generality) and their matter (content). The form of a
concept is a subjective faculty of the understanding. The matter is its
objective content supplied ultimately through intuition. The goal of logic
is not to determine the source of a concept but how a given concept can be
produced from already formed concepts. Concepts are formed from other
concepts through a threefold process of comparison, reflection and
abstraction through which a representation common to several objects
arises. Lower concepts (species) fall under higher concepts (genus).
Concepts are related to each other in terms of co-ordination and
subordination.
A judgement is the unity of consciousness of various representations or
the relationship between various representations in so far as they constitute
a concept. The matter or content of a judgement is the set of concepts and
intuitions designated by the subject and predicate expressions. The form of
a judgement is signified by the copula, which binds the predicate concept
to the subject concept. The copula also signifies the manner in which the
predicate concept belongs to the subject concept (e.g. it affirms or denies
that the predicate concept belongs to the predicate concept). For Kant,
both the conceptual content and the unity of a judgement are products of
mental acts of judging. A judgement is the recognition that a subject
concept and a predicate concept are related in such a way as to form a
unity.
Finally, Kant distinguishes three different kinds of inference. They
correspond to the three moments under the heading of relation in his
Table of Judgements. The categorical form of judgement involves a
relationship between a subject and the marks that inhere in it (S is P). The
categorical form of inference is the syllogism has three essential parts: a
62
major premise (All animals are mortal), a minor premise (all humans are
animals) and a conclusion (therefore all humans are mortal). The
hypothetical form of judgement concerns the ground of dependence of
one cognition upon another (If A then B}. The hypothetical form of
inference is modus ponens (If A then B, A, therefore B}. Finally the
disjunctive form of judgement concerns the 'combination of parts in a
whole (logical division)' (A or B) (Jdsche Logic 60, 616). The disjunctive
form of inference is defined in terms of the hypothetical form of inference
and falsehood: 'if A is false, then B is true' or 'if B is false, then A is true'.
Frege does not follow the traditional order of exposition of logical
textbooks. He does not, in other words, begin with concepts, move on to
judgements and then to inferences. Instead, he begins with the
introduction of a distinction between two kinds of symbols: constants
('2', ' + ') with a fixed determinate sense and letters ('a', 'b') with an
indeterminate sense. He then moves immediately to a discussion of the
judgement distinguishing the assertability conditions of a judgement from
its content. He then introduces his notion of the conceptual content of
a judgement, and shows how it differs from the subject/predicate form of
judgement. Finally, he completes his discussion of judgement by
contrasting his conception of judgement with the Kantian Table of
Judgements. He then moves to a discussion of inference, defining the
conditional, If P, then Q? truth functionally, and then introducing his basic
rule of inference: modus ponens (A > B, A therefore B ). From there he
introduces his negation-stroke and defines conjunction and disjunction in
terms of negation and the conditional. He then moves to a discussion of
identity of content, giving a provisional solution to the puzzle about
informative identity statements in terms of different modes of determination. He invokes Kant's distinction between analytic (uninformative) and
synthetic (informative) judgements in the course of the solution. It is only
near the end of the 'Definition of the Symbols', the second to last major
heading, that he introduces what he will later use to characterize concepts:
his notion of a function. Finally, Frege introduces the universal quantifier,
defines the existential quantifier and moves on in Part Two to prove some
judgements of pure thought with his new logical language.
Despite his introduction of the judgement-stroke and his replacement of
the subject, predicate and copula conception of judgement with his
function argument conception, there are still some residual traditional and
Kantian elements to Frege's B egriffsschr ift account of judgements. To begin

with, Frege claims that the horizontal stroke, '-', 'ties the symbols which follow
it into a whole; and the assertion, which is expressed by means of the vertical stroke at
Frege's Logic
63
the left end of the horizontal one, relates to this whole^ (CJV 112). Thus he has not
completely purged the role of the copula as the force that converts a
concatenation of concepts into a unified judgement from his conception of
the horizontal stroke. He would soon correct this by introducing the term
'content of a possible judgement' to characterize the result of attaching the
horizontal stroke to a proposition. Second, he conceived of the assertable
contents of a judgement in terms of affirmation and denial instead of truth
and falsity. Because affirmation and denial are acts of judgement, a key
element of Frege's logical system still maintained a mentalistic feature.
Finally, in his conception of identity of content, Frege characterizes the
difference in cognitive significance between an uninformative and an
informative identity statement in terms of Kant's distinction between
analytic and synthetic judgements. With his development of the distinction
between the sense and the reference of an expression Frege is later able to
give an account of the difference between the two kinds of statements
without invoking this famous Kantian distinction.
Notes
1
A2
Translation slightly altered.

There should be quotation marks surrounding 'and', 'not', etc. This is
an instance where Frege seems unaware of use-mention conventions
that he was later to follow scrupulously.
Frege does not explicitly define the existential quantifier in the
Begriffsschrift.
An open sentence is one where at least one variable in a sentence is not
bound by a quantifier. A closed sentence is one where every variable
in a sentence is bound by a quantifier.
Here again Frege fails to distinguish properly between the use and the
mention of a symbol.
This is an instance in the Begriffsschrift where ambiguity seems to arise
because Frege has yet to make the sense-reference distinction. Does
Frege mean here that the two sentences possess the same conceptual
content because they have the same sense or the same reference? It
would appear to be the former, given Frege's example. For if it were
the latter, he need only invoke two sentences that have the same truthvalue. If, however, for two sentences to have the same conceptual
content they need to have the same sense, then Frege must explain
what sameness in sense amounts to.
64
A7
88

Despite his claim that he only uses one rule of inference in the
Begriffsschrift, he uses other rules of inference without explicit
acknowledgement. As Terrell Ward Bynum observes in a footnote
to his translation of the Begriffsschrift, 'Frege frequently says (for
example in the Preface of the present work) that he uses only one
mode of inference modus ponens even though he actually also uses
substitution, confinement of generality to consequent, and several
others' (CN 119, fn. 6).
That Frege has not made the firm conceptobject distinction yet is
apparent here where he claims that the two singular terms in question
signify a concept.
Part 2
Kant on Concepts, Intuitions and

Arithmetic
Kant on Concepts., Intuitions and

Arithmetic
Concepts and Intuitions in Kant's Critique

Our nature is so constituted that our intuition can never be other than
sensible; that is, it contains only the mode in which we are affected
by objects. The faculty, on the other hand, which enables us to think
the object of sensible intuition is the understanding. To neither of
these powers may a preference be given over the other. Without
sensibility no object would be given to us, without understanding no
object would be thought. Thoughts without content are empty,
intuitions without concepts are blind . . . The understanding can
intuit nothing, the senses can think nothing. Only through their
union can cognition arise. (B75/A51)
Human beings possess two fundamental faculties of the mind out of which
all cognition arises: sensibility and understanding. Sensibility provides the
understanding with objects through intuition; the understanding enables
us to think these objects by bringing them under concepts.
Having established that sensibility is the faculty of intuition and
understanding is the faculty of concepts, Kant claims:
The cognition yielded by understanding, or at least by the human
understanding, must therefore be by means of concepts, and so is not
intuitive, but discursive. Whereas all intuitions, as sensible rest on
affections, concepts rest on functions. By 'function' I mean the unity of
the act of bringing various representations under one common
representation. Concepts are based on the spontaneity of thought,
sensible intuitions on the receptivity of impressions. (A68-9/B93)
Kant spells out the relationship between intuitions and concepts in terms
of a series of contrasts. Intuitions are sensible and rest on affections;
68
concepts rest on functions of the understanding. Intuitions are singular (see

A320/B377) and are based on the receptivity of impressions; concepts are
general and are based on the spontaneity of thought. An intuition is in
immediate relation to its object; a concept is indirectly related to its object
either through intuitions or through other concepts.
In claiming that the mind consists of two fundamentally different
faculties Kant is parting ways with his immediate predecessors and
contemporaries. As Lome Falkenstein explains:
At least since Descartes, who had argued that the soul is simple and
without parts and has but a single faculty, the spirit of the time had
been to unite, not separate, sense and intellect. In the generations after
Descartes theorists in both the rationalist and empiricist camps
advocated 'one faculty' theories of cognition albeit in diametrically
opposed fashions. Thus, Leibniz, Wolff, and Baumgarten treated all
representation as the more or less clear or obscure, distinct or confused,
apprehension of the specific differences (notae, Merkmale) distinguishing
species from genera and individuals within species. Sensory representations were taken to differ from intellectual ones only in the greater
degree of confusion (Verworrenheit) with which the collected differentiae
were apprehended. Hume, Gondillac, and Helvetius, on the other
hand, treated all representations as more or less vivacious 'traces' left
behind by past experience, or more or less complete replications of the
content of such experience. (Falkenstein 1991, 168)
As Kant himself explains:
He [Leibniz] compared all things with each other by means of concepts
alone, and naturally found no other differences save those only through
which the understanding distinguishes its pure concepts from one
another. The conditions of sensible intuition, which carry with them
their own differences, he did not regard as original, sensibility being for
him only a confused mode of representation, and not a separate source
of representations . . . In a word, Leibniz intellectualised appearances, just
as Locke . . . sensualised all concepts of the understanding^ i.e. interpreted
them as nothing more than empirical or abstracted concepts of
reflection. Instead of seeking in understanding and sensibility two
sources of representations which, while quite different, can supply
objectively valid judgments of things only in conjunction with each other,
each of these great men holds to one only of the two, viewing it as in
Kant on Concepts, Intuitions and Arithmetic
69
immediate relation to things in themselves. The other faculty is then

regarded as serving only to confuse or to order the representations
which this selected faculty yields. (A270-1/B326-7)
Leibniz intellectualized appearances in that he thought of sensibility as a
confused mode of conceptual representation and a sense perception as a
combination of confused and indistinct concepts. John Locke (16321704)
sensualized appearances in that he regarded concepts of the understanding
as abstracted from empirical concepts. Kant's solution was to introduce
two faculties: sensibility and understanding. Sensibility is the faculty of
receptivity. It supplies the faculty of the understanding with objects as
appearances. Understanding is the faculty of spontaneity (see A68/B93). It
produces representations from the appearances provided by intuition by
bringing them under concepts.
The faculty of sensibility provides the mind with the pure forms of
intuition: space and time. The faculty of understanding provides the mind
with the pure concepts of the understanding: the categories. Intuition and
concept are the two conditions that make the cognition of objects possible.
As Kant explains:
Now there are two conditions under which alone the cognition of an object
is possible, first intuition, through which it is given, though only as
appearance; secondly, concept, through which an object is thought
corresponding to this intuition. It is evident from the above that the first
condition, namely, that under which alone objects can be intuited, does
actually lie a priori in the mind as the formal ground of the objects. All
appearances necessarily agree with this formal condition of sensibility, since
only through it can they appear, that is, be empirically intuited and given.
The question now arises whether a priori concepts do not also serve as
antecedent conditions under which alone anything can be, if not intuited,
yet thought as object in general. In that case all empirical cognition of
objects would necessarily conform to such concepts, because only as thus
presupposing them is anything possible as object of experience. (A923/B125)
If therefore, we seek to discover how pure concepts of understanding are
possible, we must enquire what are the a priori conditions upon which
the possibility of experience rests, and which remain as its underlying
grounds when everything empirical is abstracted from appearances. A
concept which universally and adequately expresses such a formal and
objective condition of experience would be entitled a pure concept of
understanding. (A95-6)
70
Space and time, as the a priori forms of intuition, are the formal conditions
of sensibility. Appearances are objects given to the understanding through
intuition. To be an appearance an object must conform to the a priori
conditions of space and time. The categories are pure concepts of the
understanding. They are the a priori conditions through which appearances provided by intuition are brought under concepts in order to
produce concepts of the objects of experience.
Leibniz, Wolff and the Dogmatic Use of Reason

In drawing his distinction between analytic and synthetic judgements
Kant is rejecting a conception of the scope and power of conceptual
analysis championed by two of his predecessors: Leibniz and Wolff. Kant's
basic criticism of Leibniz in the 'Amphiboly' is that he only acknowledged
one faculty of cognition, the understanding, and failed to recognize that
the faculty of sensibility is also a necessary condition in the acquisition of
knowledge. Kant levels the same charge against Wolff.
Kant claims that his critique of pure reason is not opposed to the use of
'the dogmatic procedure of reason' to prove conclusions from secure principles
in a strictly a priori fashion. What it is opposed to is 'dogmatism, that is, to
the presumption that it is possible to make progress with pure cognition,
from concepts alone . . . without having first investigated in what way and
by what right reason has come into possession of these concepts.
Dogmatism is thus the dogmatic procedure of pure reason, without previous
criticism of its own powers' (Bxxxv). Kant has no disagreement with those
who wish to employ the method of deriving conclusions from secure
principles through the use of pure reason. What he objects to is the attempt
to advance knowledge solely through the analysis of concepts without first
subjecting the source of those concepts to a critique.
Kant insists that 'such criticism is the necessary preparation for a
thoroughly grounded metaphysics, which as science must necessarily be
developed dogmatically, according to the strictest demands of system . . .
For that is a demand to which it stands pledged, and which it may not
neglect, namely, that it carry out its work entirely a priori, to the complete
satisfaction of speculative reason' (Bxxxvi). Once the critique of pure
reason has been completed, the attention of philosophy can turn to the
development of a system of metaphysics. In such an endeavour, Kant
recommends that we
71
follow the strict method of the celebrated Wolff, the greatest of all the
dogmatic philosophers. He was the first to show by example . . . how the
secure progress of science is to be attained only through the orderly
establishment of principles, clear determination of concepts, insistence
upon strictness of proof, and avoidance of venturesome, non-consecutive
steps in our inferences. (Bxxxvi)
Wolffs mistake was not in the strict method of reasoning that he advocated or
his demand for clarity of concepts and gap-free proofs but in assuming that
pure cognition of an object is possible without first conducting a critique of the
powers through which the cognition of an object is possible in the first place.
Kant's goal in the Critique, by contrast, is not the analysis of already
existing concepts but an analysis of the faculties of understanding and
sensibility that make concepts possible:
By 'analytic of concepts' I do not understand their analysis, or the
procedure usual in philosophical investigations, that of dissecting the
content of such concepts as may present themselves, and so of rendering
them more distinct; but the hitherto rarely attempted dissection of the
faculty of the understanding itself, in order to investigate the possibility of
concepts a priori by looking for them in the understanding alone, as their
birthplace, and by analysing the pure use of this faculty . . . We shall
therefore follow up the pure concepts to their first seeds and dispositions
in the human understanding, in which they lie prepared, till at last, on
the occasion of experience they are developed, and by the same
understanding are exhibited in their purity, freed from the empirical
conditions attaching to them. (A65-6/B90-1)
In his analysis of concepts Kant is not concerned with dissecting concepts
already given and making them more distinct but with the 'dissection of the
faculty of the understanding itself. Kant's goal is to explain the possibility of a
priori concepts by discovering their source in the human understanding.
The mistake that Wolff makes is to think that the project of dissecting the
content of concepts through logical analysis can proceed without first
analysing the pure concepts of the understanding that lie prepared as
dispositions in the human understanding awaiting exposure to experience
for their proper employment.
Kant continues his criticism of Wolff in the 'Transcendental Aesthetic',
this time in conjunction with a criticism of Leibniz. Both men believed,
according to Kant, that the difference between sensible and intelligible
72
representations is merely a matter of logic. A concept whose marks can be

clearly determined through conceptual analysis is intelligible, whereas a
concept whose marks are so confused that they cannot be so analysed is
sensible. However, counters Kant, the difference between the intelligible
and the sensible is transcendental not logical. It is not a question of
whether the logical form of a judgement is clear or confused, but a question
of the origin and content of our representations. It is not just our sensibility
that can know things in themselves only in a confused fashion, but our
understanding as well. It is through our subjective constitution that we
impose the forms of sensibility and understanding on the matter of
experience. If this subjective constitution were to be removed, it would not
even be possible to represent an object for there would be no forms of
appearance by which it could be represented (see A44/B612).
A central feature of Kant's critical turn in philosophy is his claim that
we cannot account for our ability to have a priori cognition of objects if we
assume that our intuitions must conform to objects but only if we assume
that objects must conform to our intuitions. Kant compares this change in
perspective to the change in perspective wrought by the Gopernican
revolution in astronomy:
Failing of satisfactory progress in explaining the movements of the
heavenly bodies on the supposition that they all revolved round the
spectator, he [Copernicus] tried whether he might not have better
success if he made the spectator to revolve and the stars to remain at
rest. A similar experiment can be tried in metaphysics, as regards
intuition of objects. If intuition must conform to the constitution of the
objects, I do not see how we could know anything of the latter a priori]
but if the object (as object of the senses) must conform to the
constitution of our faculty of intuition, I have no difficulty conceiving
such a possibility. (Bxvi-xvii)
Kant's critical turn in philosophy rests upon his insight that only if we
treat objects as conditioned by sensibility and understanding can we
account for the possibility of cognition. The traditional conception of
cognition requires that our cognition must conform to objects. Kant
suggests that we reverse this priority and insist instead 'that objects must
conform to our cognition' (Bxvi). As a result, we cannot cognize things-inthemselves (e.g. objects that exist independently from the a priori sensible
conditions of space and time), but only objects as they are conditioned by
sensibility and understanding.
73
Leibniz and Kant on Relations and Number

In the New Essays on Human Understanding, Leibniz argues that even simple
arithmetical equations are not immediate truths. He gives an example of
how he thinks the truth of an arithmetical equation can be proved using
only definitions and one logical axiom:
That two and two are four is not quite an immediate truth. Assume that
Tour' signifies 'three and one'. Then we can demonstrate it, and here is
how.
ADefinitiond
Definitions. 1. Two is one and one.
2. Three is two and one.
3. Four is three and one.
Axiom. If equals be substituted for equals, the equality remains.
Demonstration. 2 and 2 is 2 and 1 and 1 (def. 1 ) 2 + 2
2 and 1 and 1 is 3 and 1 (def. 2) 2 + 1 + 1
3 and 1 is 4 (def. 3) 3 + 1 = 4
Therefore (by the Axiom) 2 and 2 is 4. Which is what

was to be demonstrated. (NE 414)
In the Foundations, Frege comments on this proof from Leibniz:
This proof seems at first sight to be constructed entirely from
definitions and the axiom cited . . . If we look more closely, however,
we can discover a gap in the proof which is concealed owing to the
omission of the brackets. To be strictly accurate, that is, we should
have written:
2 + 2 = 2+ (1 + 1)
(2+1) + 1 = 3 + 1 = 4
What is missing here is the proposition
2 + (1 + 1) = (2 + 1) + 1,
which is a special case of
a + (b + c) = (a + b) + c (FA 7).
What Leibniz fails to recognize, according to Frege, is that his proof tacitly
presupposes the associative law of addition.1 Since the proof presupposes a
basic law of arithmetic, Leibniz is unsuccessful in reducing the truth of 2 +
74
2 = 4 to a law of logic using only definitions and logical axioms. The lesson
that Frege takes from Leibniz's failure, according to Harold Noonan, is that
'not only definitions but general laws are required to prove numerical
formulae, and their credentials need an investigation which Leibniz does not
give' (Noonan 2001, 89).
Despite this criticism, Frege nevertheless thinks that Leibniz is on the
right track in seeking to define every number in terms of its predecessor.
Indeed, since Frege has shown the difficulty of establishing the foundations
of arithmetic upon empirical principles or upon Kant's notion of pure
intuition, he believes that the procedure proposed by Leibniz is the only
method that has hope of success. As he explains:
I do not see how a number like 437986 could be given to us more aptly
than in the way Leibniz does it. Even without having any idea of it, we
get it by this means at our disposal none the less. Through such
definitions we reduce the whole infinite set of numbers to the number
one and increase by one, and every one of the infinitely many numerical
formulae can be proved from a few general propositions. (FA 8)
Leibniz, in seeking to define number in terms of increase by one, provides a
key element in any attempt to define number. Because Leibniz's logic
allowed only monadic predication, he needed to reduce the relational
propositions found in arithmetic to categorical propositions in order to
justify their truth.
When Leibniz claims that relational sentences are reducible he could
mean a couple of different things depending upon his attitude toward
the reality of relational predicates. To clarify this point some groundclearing is in order. A relational sentence is a sentence in which a
relational expression has a foot in two different subjects (e.g. aRb, Five is
greater than four). A relational predicate is a predicate with only one foot
in a subject, the other subject being contained in the predicate of the
sentence (e.g. a is P where P= xRb, Five is greater than four]. Finally, an
absolute predicate is a predicate with one foot in a subject, where the
predicate contains no reference to any other subject (e.g. S is P, Socrates
is white).
When Leibniz claims that all relational sentences are reducible, he
could mean:
1. All relational sentences are reducible to subject-predicate sentences,
where the predicate is absolute (non-relational), or
75
2. All relational sentences are reducible to subjectpredicate sentences,

where the predicate is either absolute or relative.2
Take the sentence 'Socrates and Plato are similar.' Leibniz could reduce it
to the sentences 'Plato is wise' and 'Socrates is wise.' Two subjects are
similar if they both fall under the same concept. Hence Socrates and Plato
are similar in that they both fall under the concept of wisdom. In such a
case Leibniz is able to reduce the relational sentence 'Socrates and Plato
are similar' to a pair of sentences with non-relational predicates. If it is true
that Socrates is wise and that Plato is wise, then it is true that they are
similar in regard to wisdom. Similarity is a symmetrical relation. If
Socrates is similar to Plato, then Plato is similar to Socrates.
However, what Leibniz cannot do is reduce asymmetrical relational
propositions to categorical propositions. At best Leibniz could reduce the
relational proposition that 7 is greater than 5 to the two categorical
propositions that 7 is a number and 5 is a number. For the latter two
propositions to be logically equivalent to the relational proposition being
analysed there must be something that represents the fact that 7 is a
number that is greater than the number 5. However, it seems that there is
nothing that can accomplish this other than another relational proposition. That 7 is greater than 5 is a relational proposition whose logical form
cannot be captured through a reduction to one or more categorical
propositions. Thus, Leibniz is unable to reduce such asymmetrical
relational propositions to one or more categorical sentences with purely
non-relational predicates that are logically equivalent to it. But without
this ability he does not have the logical resources to justify the truths of
arithmetical propositions in terms of his conception of truth as conceptcontainment.
Kant on the Construction of Concepts in Intuition

In some pre-critical lectures on mathematics delivered in the early 1760s,
Kant proves the truth of the proposition 8 + 4 = 12 in a way very similar
to the proof given by Leibniz in the New Essays on Human Understanding?
Commenting on Kant's proof, Longuenesse claims that 'Just like Leibniz's
proof, this proof proceeds by successive substitutions of terms equivalent by
definition ( 3 + 1 replaces 4, then 9 replaces 8 + 1 . . . ) . This series of
substitutions corresponds to what Kant describes in the Critique as one
synthetic operation (7 + 5 = 12)' (Longuenesse 1998, 279). By the time
76
Kant has come to write the Critique, he has deserted the Leibnizian-like
proof in terms of the substitution of equivalent definitions, insisting instead
that the truth of such arithmetical propositions must be grounded in
synthesis. While he gives no explanation for rejecting his earlier
Leibnizian-like demonstration, it is not difficult to guess what his reasons
for dissatisfaction were. Kant would no doubt observe that intuition is
necessary not only to acquire a representation of a unit but also to acquire
the representation of an enumeration of units. To form the representation
of a unit requires the ability to form the representation of an individual
object. However, on Kant's principles, it is only through intuition that one
can form a representation of an individual object. So to be able to use the
symbol '!' as the representation of a singular unit requires the use of
intuition. Since it is intuition not concepts that provide the understanding
with individual representations of objects, it is only through intuition that
one can have the representation of a unit. Intuition is also required to
account for our ability to start from the number 1, or from a single stroke,
and through a process of enumeration arrive at ever greater numbers.
Such an enumeration cannot be accomplished through the analysis of the
concept of number or of the concepts of individual numbers or sums but
only through an act of synthesis by which a concept is constructed in
intuition.
While Kant discusses the relationship between mathematics and
philosophy in a variety of passages and contexts throughout the Critique,
the closest he comes to explicating fully the basic elements in his
philosophy of mathematics in general and his philosophy of arithmetic in
particular occurs near the end of the Critique in the 'Transcendental
Doctrine of Method'. In the opening paragraph of the section entitled
'The Discipline of Pure Reason in its Dogmatic Employment', Kant cites
mathematics as a prime example of a discipline that expands itself solely
through the principles of pure reason with no intrinsic reliance upon
experience. This has led some philosophers to attempt to establish
philosophy on the same sure footing as mathematics by borrowing the
methods of mathematics and applying them to philosophy. The problem
with such an approach, in Kant's eyes, is that it leads to the illegitimate
extension of reason beyond the limits of possible experience.
Like Leibniz, Kant acknowledges the essential role that symbols play in
arithmetical calculation. However, unlike Leibniz, he does not believe that
arithmetical symbols can be understood by concepts alone. Kant believed
that any attempt to justify the truths of both mathematics and philosophy
through conceptual analysis and syllogistic inference alone fails to
77
understand the fundamental difference between philosophical cognition

and mathematical cognition. While philosophical cognition is achieved
through the analysis of concepts, mathematical cognition is achieved
through the construction of concepts. To construct a concept is to show
that the connection between the concept and the intuition holds with
necessity and strict universality. Such a construction requires a nonempirical or pure intuition. An empirical intuition is the singular
representation of a sensible object. A pure intuition is a singular
representation of the forms of space or time that condition all empirical
intuitions. A concept is a universal representation. Because construction
involves intuition it is an individual object that corresponds to the concept
in the construction. Because construction involves concepts this individual
object must serve as a universal representation of the concept in question
(see A713/B741).
Kant turns to an example of the construction of a triangle in geometry.
One can construct a triangle in geometry either with an empirical
intuition where the triangle is constructed on paper with pencil, compass
and ruler, or in pure intuition using the imagination alone and borrowing
nothing from experience. But even in the case of an empirical intuition, the
individual figure drawn on paper serves as a universal representation of the
concept of a triangle. As Kant explains:
The single figure which we draw is empirical, and yet it serves to express
the concept, without impairing its universality. For in this empirical
intuition we consider only the act whereby we construct the concept,
and abstract from the many determinations (for instance, the
magnitude of the sides and of the angles), which are quite indifferent,
as not altering the concept 'triangle'. (A713-14/B741-2)
In the construction of a triangle on paper, the points, lines and figures on
the paper are singular representations. Yet, by abstracting from their
colour, width, etc., it is possible to arrive at a universal representation of a
triangle as a pure spatial intuition. This representation is synthetic because
it represents an intuition. However, it is a priori because it is ultimately
grounded not in the empirical intuition of the figure on paper but in the
pure intuition of space. Elsewhere Kant observes that the representation of
a geometrical figure presupposes time as well as space:
I cannot represent to myself a line, however small, without drawing it in
thought, that is, generating from a point all its parts one after another.
78

Only in this way can the intuition be obtained similarly with all times,
however small. In these I think to myself only that successive advance
from one moment to another, whereby through the parts of time and
their addition a determinate time-magnitude is generated. (A1623/
B203)
For Kant, then, the construction of a geometrical figure in pure intuition

presupposes not only space but time as well. The use of figures in a
geometrical proof for Kant is not a mere heuristic device but a necessary
feature of the proof. Corresponding to the points and lines on paper are
points and lines in the pure intuition of space. And corresponding to the
construction of any geometrical figure is the extension of a line through the
addition of points, which presupposes not just the forms of space but the
forms of time as well.
In contrast to geometrical construction, where a geometrical object is
constructed, in algebraic construction an arithmetical magnitude is
constructed:
But mathematics does not only construct magnitudes (quanta) as in
geometry; it also constructs magnitude as such (quantitas), as in algebra.
In this it abstracts completely from the properties of the object that is to
be thought in terms of such a concept of magnitude. It then chooses a
certain notation for all constructions of magnitude as such (numbers),
that is, for addition, subtraction, extraction of roots, etc. Once it has
adopted a notation for the general concept of magnitude so far as their
different relations are concerned, it exhibits in intuition, in accordance
with certain universal rules, all the various operations through which
the magnitudes are produced and modified. (A717/B745)
In geometry we construct spatial objects (quanta) such as triangles,
squares and circles in intuition. In algebra we construct magnitudes or
quantities in intuition through the successive iteration of homogenous
units. In Lectures on Metaphysics, Kant gives the following characterization
of quantity and quantum: 'Quantity: determination of a being, how many
times it is posited'; 'Quantum: it is one thing, in which there is quantity'
(quoted from Longuenesse 1998, 264). A quantum is a unit, a single object.
A quantity is the determination of how many times a given unit is posited.
In the geometrical construction of an object we abstract from all features of
the representation of the triangle except for those spatial and temporal
properties needed to construct the object in pure intuition. In contrast to
79
geometrical construction, an object is not constructed in algebra or

arithmetic.
In algebra, symbols are chosen to represent the different ways that a
magnitude can be constructed. Thus there are signs for addition,
subtraction and so on. Once such a notation has been developed, then it
is possible to exhibit some magnitude in intuition by constructing it in
accordance with certain universal rules as expressed by the notation. Thus
once I have defined the operations of addition and equality, say, I can
introduce symbols such as ' + ' or ' = ' to signify the universal rules of
addition. I can then construct equations such as 2 + 3 = 5 in which the
relation between numbers and operations is expressed in symbols. Such
relations cannot be understood solely through the use of concepts but must
be exhibited in intuition.
Kant then draws a distinction between symbolic construction and
ostensive construction to reflect this difference: 'and thus in algebra by
means of symbolic construction, just as in geometry by means of an
ostensive construction (the geometrical construction of the objects
themselves), we succeed in arriving at results which discursive cognition
could never have reached by means of mere concepts' (A717/B745).
He returns to the distinction a few pages later:
Even the method of algebra with its equations from which the correct
answer, together with its proof, is deduced by reduction, is not indeed
geometrical in nature, but is still constructive in a way characteristic of
the science. The concepts attached to the symbols, especially concerning
the relations of magnitudes, are presented in intuition; and this method,
in addition to its heuristic advantages, secures all inferences against
error by setting each one before our eyes. (A734/B762, translation
slightly revised)
In ostensive construction it is objects themselves (quanta) that are
constructed, while in symbolic construction a quantity is constructed. The
procedure of algebra is not a geometric (ostensive) construction but a
(symbolic) construction in terms of characters in which the truth of an
algebraic equation is justified by reduction through proof to a more basic
truth. But this reduction is not purely logical in nature for the concepts
that are attached to the symbols concern relations of magnitudes. And
since relations cannot be expressed by concepts alone, but only through
sensibility, the use of symbols to represent relations of magnitudes
presupposes sensible intuition.
80
According to Kant, the mathematician is not 'concerned with analytic

propositions, which can be produced by mere analysis of concepts . . . but
with synthetic propositions that can be known a prior? (A718/B746). To
determine the truth of a synthetic proposition one must go beyond the
properties that are contained in the subject concept to properties that belong
to the subject concept even though they are not contained within it. Thus, in
attempting to justify the truth of a mathematical proposition, a philosopher
will get nowhere by merely reflecting upon the concepts that make up the
proposition. And while a philosopher qua philosopher can perform a
transcendental synthesis on concepts alone, such a synthesis relates only to a
thing in general, as defining the conditions under which the perception of it
can belong to possible experience. 'But in mathematical problems there is no
question of this, nor indeed of existence at all, but only of the properties of
the objects in themselves, solely in so far as these properties are connected
with the concept of the objects' (A719/B747).
Arithmetic is not concerned with the existence of numbers as objects.
For numbers are not objects. Instead they are properties assigned to
sensible objects in order that those objects can be used as representations of
numbers.
For Kant, then, numbers have no independent existence. Instead they
are the result of a cognitive process by which the concept of a number is
constructed through the application of the categories of quantity to a
manifold of homogeneous units provided to the understanding through
intuition. As Charles Parsons explains: 'What plays the role of
mathematical existence in Kant's usage is constructibility' (Parsons
1992, 137). For Kant it makes no sense to speak of the independent
existence of numbers as abstract objects. The objects of arithmetic for Kant
are sensible representations of numbers, be they arithmetical symbols or
strokes on a page.
Kant on Judgement, Concepts and Quantity

A number, according to Kant, is a schema of the categories of quantity. So
to understand what Kant means by a number we must understand what
he means by a schema of the categories of quantity. His most extended
discussion of a schema in the Critique occurs appropriately enough in the
'Schematism of the Pure Concepts of the Understanding'. However, before
moving to an examination of the 'Schematism', it will help to say
something about the relationship between the three judgements of
81
quantity: universal, particular and singular and the three categories

of quantity: unity, plurality and totality.
The quantity of a judgement is defined in terms of the inclusion and
exclusion of concepts. As Kant explains in his Logic:
In the universal judgment, the sphere of one concept is wholly enclosed
within the sphere of another; in the particular, a part of the former is
enclosed under the sphere of the other; and in the singular judgment,
finally, a concept that has no sphere at all is enclosed, merely as part
then, under sphere of another. (Jdsche Logic 21, 598)
The quantity of a judgement is determined by the relationship between the
constituent subject and predicate concepts. To say that all humans are
mortal is to say that the concept of humanity is wholly enclosed within the
sphere of the concept of mortality. To say that some humans are wise is to
say that only some members of the concept of humanity are enclosed in the
sphere of the concept of wisdom. Finally, to say that Socrates is wise is to
say that the individual concept of Socrates, a concept with no sphere at all,
is enclosed as a part under the sphere of the concept of wisdom.
Every judgement of quantity is thus either universal (All S is P),
particular (Some S is P] or singular (This S is P). Corresponding to these
three moments under the heading of Quantity in the Table of Judgements
(see p. 58) are the three categories of Quantity: Unity, Plurality and
Totality in the table of categories.
If the order in which Kant lays out the three moments of quantity in the
Table of Judgements corresponds to the order of the three categories of
quantity in the table of categories, then the singular form of judgement will
correspond to the category of totality while the universal form of judgement
will correspond to the category of unity. However, the relationship between
the judgements of quantity and the categories of quantity makes more sense
if singular judgements correspond to the category of unity and universal
judgements correspond to the category of totality. Direct support that Kant
did in fact see the relation in that way can be found in the Prolegomena to any
Future Metaphysics (see Longuenesse 1998, 116).
In 20 of the Prolegomena, Kant argues that synthetic judgements are
objectively valid only in so far as the pure concepts of the understanding
are applicable to the concepts abstracted from intuition. He turns to an
example from geometry to illustrate his claim:
Even the judgments of pure mathematics in their simplest axioms are not
82

exempt from this condition. The principle that a straight line is the
shortest distance between two points presupposes that the line is
subsumed under the concept of quantity, which certainly is no mere
intuition but has its seat in the understanding alone and serves to
determine the intuition (of the line) with regard to the judgments which
may be made about it in respect to the quantity, that is, to plurality (as
judica plurativa). For under them it is understood that in a given intuition
there is contained a plurality of homogeneous parts. (Prolegomena 20)
In the course of justifying the objective validity of the categories, Kant argues
that the pure concept of quantity is presupposed even in the simplest axioms
of pure mathematics. Thus to understand the axiom that a straight line is the
shortest distance between two points presupposes the concepts of a line and a
point. But, according to Kant, to understand these concepts presupposes the
concept of a plurality of homogeneous parts (e.g. the points that make up a
line). But a plurality of parts presupposes the categories of quantity. In
particular it presupposes the category of plurality. Kant then appends a
footnote to the second to last sentence of this passage:
This name [judica plurativa] seems preferable to the term particularia, which
is used for these judgments in logic. For the latter already contains the
thought that they are not universal. But when I start from unity (in
singular judgments) and proceed to totality, I must not [even indirectly
and negatively] include any reference to totality. I think plurality merely
without totality, and not the exclusion of totality. This is necessary, if the
logical moments are to underlie the pure concepts of the understanding. In
logical usage one may leave things as they were. (Prolegomena 20)
So corresponding to the singular judgement that this card is a spade is the
category of unity. Treating a card as a unit, it is possible to represent
enumeration as a series of judgements: this card is a spade, this card is a
spade, etc. This can be represented as the particular judgement that some
cards are spades, which is true when there is a plurality of cards that are
spades. We finally arrive at the universal judgement that every card is
a spade. This judgement is true if the totality being counted is a suit of
spades but false if the totality is a deck of cards. The enumeration of units up
to a given totality presupposes the application of the three categories of
quantity. An intuition is brought under the concept of unity to form the
representation of a unit. This unit can then be used as the base step of an
enumeration of units by being brought under the concept of plurality until
83
the understanding finally arrives at the representation of a number by being

brought under the category of totality. As Kant explains in the Critique,
our counting, as is easily seen in the case of larger numbers, is a synthesis
according to concepts, because it is executed according to a common
ground of unity, as, for instance, the decade. In terms of this concept, the
unity of the synthesis of the manifold is rendered necessary. (A78/B104)
Counting by tens, say, or hundreds is to synthesize the representation of
symbols on a page according to a specific concept. This concept by which
units are counted forms a common ground of unity under which sensible
intuitions (a plurality of units) are brought under a general concept: the
total quantity being counted (e.g. suits or decks of cards).
Kant on the Schematism of the Categories of Quantity

According to Kant:
philosophical cognition considers the particular only in the universal,
mathematical cognition the universal in the particular, or even in the
single instance, though still always a priori and by means of reason.
Accordingly, just as this single object is determined by certain universal
conditions of construction, so the object of the concept, to which the
single object corresponds merely as its schema, must likewise be thought
as universally determined. (A714/B742)
Mathematical cognition enables us to treat an empirical intuition of a
written symbol as a universal mark of a mathematical concept. Thus I can
treat the number '12' imprinted on the page in front of me as a universal
mark of the concept of the number 12. Since no human can have a
representation of an object except through intuition, and since every
intuition is subject to the transcendental conditions of space and time, the
understanding can only form representations of objects that are subject to
the transcendental conditions of space and time. Thus it is possible to
abstract from all of the empirical attributes of a representation and to treat
it as an object only in so far as it has a certain spatial structure, as in the
case of a triangle, say, or as a certain place in a temporal sequence, say, as
in the case of a number. Kant calls such an object the schema of a concept.
After setting out to prove in the transcendental deduction that experience
84
is possible only if the categories are applicable to empirical objects, Kant sets
out in the 'Analytic of Principles' to explore in more detail the relationship
between the categories and the objects to which they are applicable. Kant
begins with the more manageable task of explaining how an empirical
object can be subsumed under an empirical concept:
In all subsumptions of an object under a concept the representation of
the object must be homogeneous with the concept; in other words, the
concept must contain something which is represented in the object that
is to be subsumed under it. This, in fact, is what is meant by the
expression, 'an object is contained under a concept'. Thus the empirical
concept of a plate is homogeneous with the pure geometrical concept of a
circle. The roundness which is thought in the latter can be intuited in the
former. (A137/B176)
The goal of the schematism is to explain not how an object can be
subsumed under an empirical concept but how an object can be subsumed
under a pure concept of the understanding. Because an empirical concept,
by its nature, acquires its content only through empirical intuition, it is no
great difficulty to chart the relationship between the empirical concept
and the forms of intuition. The roundness that I am able to think in the
concept of a plate presupposes the pure geometrical concept of a circle.
But I am able to form a representation of a circle because I have
experience of various round objects in the sensible world. However, since
the categories, in contrast to empirical concepts, are 'quite heterogeneous
from empirical intuitions', they can never be found in an intuition. 'How,
then', Kant asks, 'is subsumption of intuitions under pure concepts, the
application of a category to appearances, possible?' (A138/B177). His
answer is that 'there must be some third thing, which is homogeneous on
the one hand with the category, and on the other hand with the
appearance, and which thus makes the application of the former to the
latter possible' (A138/B177). He calls this mediating representation,
which is in one respect intellectual and in another respect sensible, a
transcendental schema.
Kant stresses that time, as the formal condition of inner sense, plays a
crucial role in the application of the categories to the objects of
appearances. For a transcendental determination of time forms a unity
only in so far as it is homogeneous with a category, while a category is
supplied with a representation only in so far as it is homogeneous with an
appearance. And appearances, as empirical representations of a manifold
85
of inner sense, are subject to the conditions of time. 'Thus an application of

the category to appearances becomes possible by means of the
transcendental determination of time, which, as the schema of the
concepts of the understanding, mediates the subsumption of the
appearances under the category' (A139/B178).
Kant then claims that while a schema is a product of the imagination, it
is not an image. The difference between an image and a schema is that
while an image is a representation of an object in the imagination, a
schema is a method by which a representation of an object is brought into
conformity with a concept. Kant turns to the concept of number as an
example. It is certainly possible to have the image of the number 5, say, as
five points ( ) or five strokes (| |) set alongside one another. In the
first instance we have an image of the number 5 in terms of the concept of a
point. In the second instance we have an image of the number 5 in terms of
the concept of a stroke. It is of no concern which image is chosen. What is
important is that the image is in conformity with a concept under which
the number 5 can be brought. A schema provides the understanding with
a rule by which a given number can be brought under a given concept
through the use of the imagination.
According to Kant:
The pure image of all magnitudes (quantorum) for outer sense is space; that
of all objects of the senses in general is time. But the pure schema of
magnitude (quantitatis), as a concept of the understanding is number, a
representation which comprises the successive addition of homogeneous
units. Number is therefore simply the unity of the synthesis of the manifold
of a homogeneous intuition in general, a unity due to my generating time
itself in the apprehension of the intuition. (A142-3/B182)
The schema of quantity is a rule that applies the pure general concept of
number to a manifold of singular homogeneous intuitions in order to
produce a determinate number. The mediating factor that makes this
connection between a general concept and a singular intuition possible is
the pure intuition of time. It is through the application of a schema of
quantity to a manifold of homogeneous intuitions that we are able to
produce individual numbers from the concept of number. Temporal
synthesis is what connects the pure concepts of quantity with the manifold
of sensible intuitions that provides the understanding with the matter
necessary for the generation of numbers.
However, one should not conclude from this that time furnishes the
86
objects of arithmetic in the way that space furnishes the objects of

geometry. As Kant explains in a letter to his student Schiiltz:
Time, as you correctly notice, has no influence on the properties of
numbers (considered as pure determinations of quantity), as it may
have on the character of those changes (of quantity) that are possible
only relative to a specific state of inner sense and its form (time). The
science of numbers, notwithstanding the succession that every
construction of quantity requires, is a pure intellectual synthesis, which
we represent to ourselves in thought. But in so far as specific quantities
(quanta) are to be determined in accordance with this science, they
must be given to us in such a way that we can grasp their intuition
successively; and thus this grasping is subjected to the time condition.
So that when all is said and done, we cannot subject any object other
than an object of a possible sensible intuition to quantitative, numerical
assessment, and it thus remains a principle without exception that
mathematics can be applied only to sensibila. (Kant 1967, 1301)
According to Kant, when we consider number as a pure determination of
quantity, time has no influence upon its properties. Presumably his point is
that time has no influence on whatever properties a given number may
possess, say being prime in the case of the number 5. However, time does
have an influence in the calculation of numbers. Thus to arrive at a number
through counting or addition requires that my representations of the
numbers be given to me successively. But successive iteration of homogeneous
units that underlies our ability to count and calculate depends upon our
intuition of time as the formal condition of possible experience.
Notes
1
2
3
See Pap 1958, 12.

The above discussion is based upon, Mugnai 1992, 93-4.
The lectures were recorded by J. G. Herder, who attended Kant's
lectures from 1762 to 1764. Leibniz's New Essays on Human Understanding were posthumously published in 1765. Thus Kant's proof
predates the publication of the similar proof offered by Leibniz. It is
certainly possible that the proof was well known before the
publication of the new essay and that Kant did not discover the
proof independently. See Longuenesse 1998, 278-80, esp. fn. 76.
Part 3
Analyticity and Fruitful Definitions
Analyticity, Generality and Content
Introduction
Frege's goal in his 1884 book Foundations of Arithmetic was to establish the
probability that arithmetic is an a priori analytic science. He opens the first
chapter of the book by pointing to the recent advances in rigour and
precision that have been made in the mathematical sciences:
After deserting for a time the old Euclidean standards of rigour,
mathematics is now returning to them, and even making efforts to go
beyond them . . . The discovery of higher analysis only served to confirm
this tendency; for considerable, almost insuperable, difficulties stood in
the way of any rigorous treatment of these subjects, while at the same
time small reward seemed likely for the efforts expended in overcoming
them. Later developments, however, have shown more and more clearly
that in mathematics a mere moral conviction, supported by a mass of
successful applications is not good enough. Proof is now demanded of
many things that formerly passed as self-evident . . . The concepts of
function, of continuity, of limit and of infinity have been shown to stand
in need of sharper definition . . . In all directions these same ideals can be
seen at work rigour of proof, precise delimitation of extent of validity,
and as a means to this, sharp definition of concepts. (FA 1)
Given such advances it should come as no surprise, Frege observes, that
'we are bound eventually to come to the concept of Number' (FA 2). It is
here, especially, in an enquiry into the very foundations of arithmetic, that
the proper standards of rigour and precision must be met. Indeed it was
the need for such standards that led Frege to develop his concept-script.
'The aim of proof is, in fact', Frege continues, 'not merely to place the
truth of a proposition beyond all doubt, but also to afford us insight into
the dependence of truths upon one another' (FA 2). It is a truth's
relationship to the primitive laws that it depends upon and the status of
90
those laws, more than indubitability, which most concerned Frege. Here
he echoes Leibniz's distinction between the order of discovery and the
order of justification of a truth. As Leibniz explains, 'we are not concerned
. . . with the sequence of our discoveries, which differs from one man to
another, but with the connection and natural order of truths, which is
always the same' (NE 412). Frege believed that in his concept script he
had developed a method powerful enough to show that the truths of
arithmetic could be derived from purely logical axioms and thus be shown
to be a priori and analytic. What he is offering in these opening pages, then,
is a reason why an enquiry into the nature of numbers is important, and
why such an enquiry requires an even higher degree of deductive rigour
and precision than has hitherto been offered.
However, it was not just a concern with mathematical rigour that led
Frege to this enquiry. He was also motivated by philosophical concerns:
Philosophical motives too have prompted me to enquiries of this kind.
The answers to the questions raised about the nature of arithmetical
truths are they a priori or a posteriori? synthetic or analytic? must lie
in this same direction. For even though the concepts concerned may
themselves belong to philosophy, yet, as I believe, no decision on these
questions can be reached without assistance from mathematics. (FA 3)
Frege observes earlier in the Foundations that the disciplines of mathematics
and philosophy have hitherto not cooperated as much as they could. This
lack of cooperation, he claims, 'is due in my opinion to the predominance
in philosophy of psychological methods of argument, which have
penetrated even into the field of logic. With this tendency mathematics
is completely out of sympathy' (FA v). As an example he points to a
proponent of psychologism named Strieker who 'calls our ideas of numbers
motor phenomena and makes them dependent on muscular sensations'
(FA v). But sensations are not the concern of arithmetic, says Frege.
No more are mental pictures, formed from the amalgamated traces of
earlier sense-impressions. All these phases of consciousness are characteristically fluctuating and indefinite, in strong contrast to the definiteness
and fixity of the concepts and objects of mathematics. (FA v-vi)
To combat this tendency to think of the foundations of arithmetic as
psychological, Frege recommends that we never confuse a psychological
description of how an idea of a proof originated with the grounds that
91
justify it. 'A proposition', he tells us, 'may be thought, and again it may be
true; let us never confuse these two things' (FA vi).
One constant feature of Frege's philosophical outlook from the
beginning to the end of his career was his firm and unwavering stance
against the improper incursion of psychological laws or principles into the
sciences of logic and arithmetic. Frege considered attempts to explain our
knowledge of mathematics or logic in terms of psychological principles as a
threat to the objectivity of those sciences. Psychologism, as he understood
it, can be characterized as the view that concepts like truth, validity, and
knowledge can and must be given mentalistic or subjectivistic explanations. Any legitimate account of knowledge or truth must be grounded in
the truths of psychology. Thus science in general and logic in particular
can be properly grounded only through self-observation of and reflection
on our own ideas. The problem with such a view, according to Frege, is
that ideas are private and essentially unshareable, and thus cannot
account for the objectivity of truth and so of science.
It is Frege's claim, then, that we must turn to mathematics for
assistance in order to answer the philosophical question of whether the
truths of arithmetic are a priori or a posteriori, analytic or synthetic. As a first
step toward acquiring the standards of rigour required for mathematics
and to understand the nature of arithmetical truth it is important, Frege
warns, to distinguish the way that we discover the content of a judgement
from the way that we justify its truth:
It not uncommonly happens that we first discover the content of a
proposition, and only later give the rigorous proof of it, on other and
more difficult lines; and often this same proof also reveals more precisely
the conditions restricting the validity of the original proposition. In
general, therefore, the question of how we arrive at the content of a
judgement should be kept distinct from the other question, Whence do
we derive the justification of its assertion? (FA 3)
Frege claims that:
these distinctions between a priori and a posteriori, synthetic and
analytic, concern, as I see it, not the content of the judgment but the
justification for making the judgment. Where there is no such
justification, the possibility of drawing the distinctions vanishes. [Here
Frege appends a footnote: 'By this I do not, of course, mean to assign a
new sense to these terms, but only to state accurately what earlier
writers, Kant in particular, have meant by them.'] (FA 3)
92
To understand the extent to which Frege's conception of this pair of

distinctions corresponds to Kant's conception and the extent to which it
differs will be the concern of the next three chapters. However, to
appreciate fully the significance of Frege's departure from philosophical
presuppositions of his day it is not enough just to look at his departure from
Kant. It is necessary to consider his relationship to Leibniz as well. Leibniz
and Frege share a common conception of the nature of the a priori and its
relation to logic that puts them both at odds with Kant. Frege and Leibniz
shared a trust in the power of pure reason to provide the understanding
with cognitive content through pure conceptual thought without any
intrinsic reliance upon experience. In this respect they both stood in
contrast to Kant whose critical philosophy sought to limit the legitimate
use of reason to objects of possible experience. Kant believed that logic is
completely general because it is purely formal. Indeed, his distinction
between synthetic and analytic judgements and his claim that all
knowledge is grounded in synthetic judgements arises, at least in part,
from his dissatisfaction with Leibniz's unstinting trust in the power of
conceptual thought by itself to furnish the understanding with reliable
representations of objects. While Leibniz does not explicitly distinguish
between analytic and synthetic judgements, the roots of Kant's conception
of the distinction lie in Leibniz's conception of truth as conceptcontainment. When Frege rejects Kant's claim that all knowledge is
grounded in synthetic principles he is returning to a conception of
philosophy with strong similarities to certain views of Leibniz that Kant
rejects. Despite these similarities between Frege and Leibniz there are also
some important differences. While Leibniz was a gifted and influential
logician, like Kant he was never able to free himself from the traditional
conception of a proposition or judgement as a predicate connected to a
subject by a copula. Corresponding to this logical distinction between a
subject and a predicate is the metaphysical distinction between a substance
and an attribute. In replacing the subject/predicate/copula conception of a
judgement with a function/argument conception of judgement, Frege not
only departed from the traditional logical conception of a judgement but
he also departed from the traditional metaphysical conception of a
substance, together with its attributes, as the paradigm example of an
object. (Part 3 of this work will deal with this issue.) On the one hand,
then, Frege and Leibniz share strong similarities in philosophical outlook,
as briefly shown above. On the other hand, in replacing a subject/
predicate/copula conception of a judgement with a function/argument
conception, Frege parts company with both Kant and Leibniz. This
93
departure frees Frege from the traditional conception of an object as an

individual substance and thus enables him to argue for a conception of
numbers as non-actual objects.
In breaking with Kant's claim that arithmetic is an a priori synthetic
science, Frege returns to an outlook with strong similarities to Leibniz's
outlook as Frege himself acknowledges. Leibniz never explicitly claims that
the laws of number are analytic, the distinction between analytic and
synthetic having first been made by Kant. Nevertheless, according to
Frege, statements made by Leibniz imply it. To begin with, for Leibniz the
analytic and the a priori coincide. This attitude is illustrated in Leibniz's
belief that algebra is grounded in logic and his belief that all necessary
truths can be proved by being reduced to explicit identities.
While Leibniz does not explicitly claim that arithmetic is analytic he
does claim that the truths of arithmetic, as well as all other truths for that
matter, can in principle be justified solely in terms of concept-containment
and the reduction to explicit identities. To better appreciate the import of
Leibniz's belief that arithmetic is analytic, his definition of truth in terms of
concept-containment needs to be examined. Next Kant's distinction
between analytic and synthetic judgements will be examined. Particular
attention will be paid to his contention that the definition of truth in terms
of concept-containment cannot account for the cognitive significance of
arithmetical judgements. Since in the truths of arithmetic the predicate
concept is not contained in the subject concept, they are impervious to
analysis by traditional logical means. Instead, one must turn to synthetic
judgements to see how a predicate concept can belong to a subject concept
when it is not contained in the subject concept. Finally, we will turn to
Frege's criticism of Kant and his claim that his new logical language makes
fruitful analytic concepts and definitions possible.
Leibniz on Truth
Leibniz famously defines truth in terms of concept-containment:
[I]t is common to every true affirmative proposition, universal and
particular, necessary or contingent, that the predicate is in the subject,
that is, that the notion of the predicate is involved somehow in the
notion of the subject. And this is the source [principium] of infallibility in
every sort of truth for that being who knows everything a priori. (P E 95)
94
In this passage Leibniz says merely that the predicate concept 'is involved
somehow' in the subject concept. Elsewhere he is more explicit in claiming
that the relationship between the subject concept and a predicate concept
in a true affirmative proposition is one of concept-containment:
In every proposition, the predicate is said to be in the subject, that is, the
notion of the predicate is contained in the notion of the subject. For, in a
universal affirmative proposition, when I say 'every man is an animal' I
mean 'the concept of animal is contained in the concept of man' (for the
concept of man is to be a rational animal). And when I say 'every pious
person is happy' I mean that whoever understands the nature of piety
will also understand that it contains within itself true happiness. And so
in a universal affirmative proposition, it is obvious that the predicate is
contained in the subject considered by itself. (PE 11)
Leibniz believed that the basic form of a proposition is categorical. As he
explains in a short logical essay of 1679 entitled Elements of a Calculus: 'By
'proposition' I understand here categorical propositions, unless I make
special mention to the contrary. However, the categorical proposition is
the basis of the rest, and modal, hypothetical, disjunctive and all other
propositions presuppose it' (LP 17). In another essay, written in the same
year, he claims that:
Every categorical proposition has a subject, a predicate, a copula, a
quality and a quantity. Subject and predicate are called 'terms'. For
example, in 'The pious man is happy', 'the pious man' and 'happy' are
terms, of which 'the pious man' is the subject, 'happy' the predicate,
and 'is' is the copula. The 'quality' of a proposition is affirmation or
negation . . . The 'quantity' of a proposition is its universality or
particularity. (LP 25)
Truths for Leibniz are either primary or derivative: 'The primary truths
are those which assert the same thing of itself or deny the opposite of its
opposite.' Leibniz lists several examples of such truths, 6A is A\ 'A is not
not-^4', and 'Every thing is similar or equal to itself He claims that all
such primary truths are identities (see PE 30-1). Derivative truths are of
two kinds: those that can be resolved through demonstrative reasoning
into identities and those that cannot. Two passages from the essay
'Primary Truths' expand on this claim. Derivative truths
95
are reduced to primary truths with the help of definitions, that is,
through the resolution of notions . . . The connection and inclusion of
the predicate in the subject is explicit in identities, but in all other
propositions it is implicit and must be shown through the analysis of
notions; a priori demonstration rests on this. (PE 31)
When an analysis of the relationship between the subject and the predicate
reveals that the notion of the predicate is contained in the notion of the
subject and that the truth of the proposition can be reduced to an explicit
identity, then the truth is necessary. When the analysis proceeds to infinity
and never gains completion, then the truth is contingent. Contingent
propositions can be known completely only by God who alone
comprehends the infinite (see PE 989).
For Leibniz, the most basic truths of all are explicit identities. All
derivative truths are implicit identities whose truth can be justified
through an analysis of the concepts that make up the truth into clear and
distinct ideas and the reduction of the truth, through the substitution of
logically equivalent terms, to an explicit identity. Take, for example, the
proposition that Socrates is rational. Leibniz could give the following
argument to justify that the proposition is true:
Socrates is a man.
Man is a rational animal.
A rational animal is a rational animal.
Therefore, Socrates is a rational animal.
Therefore, Socrates is rational.
Here the truth of the proposition that Socrates is a rational animal is
justified by reducing it, through the use of analysis of notions and the
interchange of logically equivalent terms, to the explicit identity that a
rational animal is a rational animal.
According to Leibniz: 'The great foundation of mathematics is the
principle of contradiction or identity', that is, that a proposition cannot be true
and false at the same time, and that therefore A is A and cannot be not A.
This single principle is sufficient to demonstrate every part of arithmetic
and geometry, that is, all mathematical principles' (PE 321). Leibniz
believed that the foundations of mathematics are grounded in the principle
of contradiction or identity and thus just as the truth of the proposition
that man is an animal can be justified through analysis of its constituent
concepts and reduction to an explicit identity, so too is it possible to justify
96
the truths of arithmetic. In short, Leibniz believed that the truths of all of
mathematics are reducible to the truths of logic.
Kant on Analytic Judgements

Kant's distinction between analytic and synthetic judgements is a
centrepiece of his quest to answer the question of how a priori synthetic
judgements are possible. The nature of synthetic judgements will be
examined in more detail in the next section. The concern here is with
analytic judgements.
Kant contrasts analytic judgements with synthetic judgements in terms
of three criteria: concept-containment, identity and contradiction. He uses
the first two criteria in his initial discussion of the distinction in the
'Introduction' to the Critique.
In all judgments in which the relation of a subject to the predicate is
thought (I take into consideration affirmative judgments only, the
subsequent application to negative judgments being easily made), this
relation is possible in two different ways. Either the predicate B belongs
to the subject A as something which is (covertly) contained in the
concept A; or B lies outside the concept A, although it does indeed stand
in connection with it. In the one case I entitle the judgment analytic, in
the other synthetic. Analytic judgments (affirmative) are therefore those
in which the connection of the predicate to the concept of the subject is
thought through identity; those in which this connection is thought
without identity should be entitled synthetic. (A6-7/B10)
The judgement that all bodies are extended is analytic.
For I do not require to go beyond the concept which I connect with
'body' in order to find extension as bound up with it. To meet with this
predicate, I have merely to analyse the concept, that is, to become
conscious to myself of the manifold which I always think in that
concept. The judgment is therefore analytic. (A7/B11)
To be a body is to be extended. So if I understand what a body is, then an
analysis of the concept of body into the concepts that compose it will yield
the concept of extension without any need to turn to experience. Here
Kant seems to treat conceptual containment and identity as interchangeable.
97
If the concept of extension is contained in the concept of body, then the

connection between the two concepts is thought through identity.
Analytic judgements, according to Kant, by 'adding nothing through
the predicate to the concept of the subject, but merely breaking it up into
those constituent concepts that have all along been thought in it, although
confusedly, can also be entitled explicative' (A7/B11). Whereas synthetic
judgements are labelled ampliative because they 'add to the concept of the
subject a predicate which has not in any way been thought in it, and which
no analysis could possibly extract from it' (A7/B11). The judgement that
all bodies are extended is analytic because the concept of extension is
contained in the concept of body. However, in the judgement that some
bodies are heavy, the concept of weight is not contained in the concept of
body. So to determine whether such a judgement is true I must go beyond
what is contained in the concept of body and turn to experience to
determine whether some bodies are heavy. In synthetic judgements the
predicate concept is not contained in the subject concept. So the discovery
that a predicate concept can belong to a subject concept even though it is
not contained in the subject concept advances our knowledge of the world
in a way that an analytic judgement cannot. While the discovery that a
predicate concept is contained in a subject concept can elucidate or clarify
a concept, it cannot advance our knowledge of the world.
Kant claims that all judgements of experience are synthetic, for one
must appeal to the testimony of the senses to determine whether the
predicate concept is contained in the subject concept. In analytic
judgements, in contrast, one need never appeal to the testimony of
experience:
That a body is extended is a proposition that holds a priori and is not
empirical. For, before appealing to experience, I have already in the
concept of body all the conditions required for my judgment. I have
only to extract from it, in accordance with the principle of contradiction, the required predicate, and in so doing can at the same time
become conscious of the necessity of the judgment - and that is what
experience could never have taught me. (A7/B11-12)
The concept of body is composed of the characters of extension,
impenetrability and figure. Something without extension cannot be a
body. The same holds true for impenetrability and figure. Here Kant
introduces the third criterion of analyticity: the principle of contradiction.
Kant explicates the principle of contradiction later in the Critique: 'The
98
proposition that no predicate contradictory of a thing can belong to it, is

entitled the principle of contradiction, and is a universal, though merely
negative, criterion of all truth' (A151/B190). As a negative criterion of
truth the principle of contradiction holds for judgements in general
irrespective of their content. However, the principle also has a positive
employment.
For, if the judgment is analytic, whether negative or affirmative, its truth
can always be adequately cognized in accordance with the principle of
contradiction .. . The principle of contradiction must therefore be
recognised as being the universal and completely sufficient principle of
all analytic cognition ... The fact that no cognition can be contrary to it
without self-nullification, makes this principle a conditio sine qua non, but
not a determining ground, of the truth of our [non-analytic] cognition.
(A151-2/B190-1)
If a judgement is analytic, then the denial that the predicate concept
belongs to the subject concept leads to contradiction. This leads Kant to
label the principle of contradiction 'the universal and completely sufficient
principle of all analytic cognition'. Presumably Kant means something like this.
If the judgement that all bodies have extension is analytic, then to say that
a body has no extension is to claim that a body can both have and not
have extension and this is a contradiction in thought. Therefore, extension
belongs to the concept of body analytically.
Kant's discussion of the distinction between analytic and synthetic
judgements is filled with phenomenological vocabulary. He talks about
one concept being 'covertly' contained in another. In successfully using the
principle of contradiction to determine whether a judgment is analytic he
claims that we 'become conscious of the necessity of the judgment' (A7/
B12). And a few lines later he says: 'From the start I can apprehend the
concept of body analytically through the characters of extension,
impenetrability, figure, etc., all of which are thought in the concept'
(A8/B12).
Kant's remarks seem to imply that what makes a judgement analytic or
synthetic, at least in part, is the state of mind of the person making the
judgement. If the person can arrive at the predicate concept through a
simple analysis of the subject concept, then the judgment is analytic. If one
cannot think the predicate concept in the subject concept, then the
predicate concept is not contained in the subject concept and the
judgement is synthetic. This in turn seems to imply that the difference
99
between an analytic and synthetic judgement for Kant is variable. Thus a

key difference between synthetic and analytic judgements is that synthetic
judgements increase our knowledge while analytic judgements do not. But
it is surely possible for Jones, with little understanding of physics or
philosophy, to be unaware that the concept of extension is a characteristic
mark of the concept of body. Upon being informed of this by a more
informed friend his knowledge of the concept of body would be increased.
If what makes a judgement analytic or synthetic depends upon whether
one's knowledge of the concept has increased, then it would seem that he
has understood the relationship between the subject concept and the
predicate concept synthetically upon learning that all bodies have
extension.
There are a couple of things that can be said in Kant's defence. First, in
a passage immediately preceding his introduction of the analytic/synthetic
distinction, Kant says,
a great, perhaps the greatest, part of the business of our reason consists
in analysis of the concepts which we already have of objects. This
analysis supplies us with a considerable body of cognitions, which while
nothing but explanation or elucidation of what has already been
thought in our concepts, though in a confused manner, is yet prized as
being, at least as regards its form, new insight. But so far as the matter
or content is concerned, there has been no extension of our previously
possessed concepts, but only an analysis of them. (A56/B9)
When we analyse the concept of an object, we discover the partial concepts
that make up the concept we are analysing. While such an analysis offers
'nothing but explanation or elucidation of what has already been thought
in our concepts, though in a confused manner', it can provide new insight
into the nature of the concept being analysed. However, such an analysis
does not in any way change the content of the concept but only reveals an
aspect of the concept's content that was unclear before the process of
analysis began.
Kant could then claim that when Jones learns that the concept of
extension is an essential attribute of the concept of body, the concept of
body has been elucidated. But this does not amount to a synthetic
extension of Jones's knowledge of the concept of body but merely an
analytic elucidation. However, this merely pushes the question back a step.
What distinguishes an analytic elucidation of a concept from a synthetic
expansion of our knowledge of a concept? How can I tell the difference
100
between a judgement where the predicate concept is contained in the

subject concept from one where it is not?
Answers to these questions can be found in the Logic, where Kant
explicates the distinction between analytic and synthetic judgements in
terms of identity, without any appeal to phenomenological descriptions:
Analytic propositions one calls those propositions whose certainty rests
on identity of concepts (of the predicate with the notion of the subject).
Propositions whose truth is not grounded on identity of concepts must
be called synthetic . . . To every X which appertains the concept of
body (a + b) appertains also extension (b) is an example of an
analytic proposition . . . To every X to which appertains the concept of
body (a + b) appertains also attraction (c) is an example of a
synthetic judgment. (Logic 36, 117)
For Kant, identity is a relationship between concepts not objects. A
proposition expresses an identity if the concept of the predicate is
contained within the concept of the subject. To claim that all bodies are
extended is to express an identity, for the concept of extension is an
essential characteristic of the concept of a body. Extension is part of what it
is to be a body. This provides us with a way of expressing the distinction
between analytic and synthetic judgements that does not depend on any
phenomenological descriptions. If a concept is defined in terms of a set of
essential characteristics, then any time one of these characteristics is
predicated to the subject concept they belong to, the two concepts are
identical and the judgement is analytic. Since the concept of body is just a
complex concept composed of the characteristics of extension, impenetrability, figure, etc., and since the predicate concept is identical to the
subject concept, the judgement that all bodies are extended is analytic.
In the 'Introduction' to the Logic, Kant claims that necessary
characteristics are called essential while contingent characteristics are
called non-essential. Kant characterizes the essence of a thing as '[t]he
complex concept of all essential components of a thing or the sufficiency of
its characteristics as to coordination and subordination' (Logic VIII. G 5,
67). He then explains:
For to the logical essence belongs nothing but the cognition of all
predicates in respect of which an object is determined by its concept... If,
for example, we want to determine the logical essence of a body, we do
not have to search out the data for this in nature; we only need to direct
101
our reflection to those characteristics which as essential elements . . .

originally constitute its basic concept. For the logical essence itself is
indeed nothing else but the first fundamental concept of all necessary
characteristics of a thing. (Logic VIII. G 5, 67)
To determine whether a predicate concept belongs to a subject concept
analytically one need only determine whether the predicate concept
belongs to the logical essence of subject concept. For the logical essence of a
body is the set of partial concepts that are the necessary characteristics of
the concept of body. By defining analyticity in terms of conceptcontainment, Kant has a way to determine whether a judgement is
analytic without any essential dependence upon phenomenological
language.
A hierarchical structure based on species, genus and differentia
underlies Kant's understanding of the logical relationship between
concepts. This, in turn, informs his understanding of analytic judgements.
An analysis of the essence of a concept consists of a journey through the
hierarchical tree structure in which the concept is a node in the structure.
To determine whether a judgement is analytic requires locating the place
of the subject concept in a given hierarchical structure and then
determining whether the predicate concept is appropriately related to
the subject concept in the hierarchical structure. If I want to discover
whether the judgement that all humans are animals is analytic I can work
my way up the appropriate tree from the location of the concept of
humanity until I discover the concept of animality. I am then justified in
claiming that the judgement is analytic.
For Kant, the only way to justify the truth of an analytic judgement is
to show that the predicate concept is contained in the subject concept and
that the sentence can be reduced to an explicit identity. In this respect
Kant's conception of how the truth of analytic judgements is justified
shows the strong influence of Leibniz's definition of truth in terms of
concept-containment. Where Kant disagrees with Leibniz is over the scope
of Leibniz's definition of truth in terms of concept-containment. Whereas
Leibniz thinks that such a definition of truth holds for all categorical
judgements, Kant believes that it holds only for analytic categorical
judgements. However, such a definition of truth cannot be used as
adequate justification of the truth of a synthetic judgement. For in a true
synthetic judgement, while the concept of the predicate belongs to the
concept of the subject it is not contained in the concept of the subject.
102
Kant on Synthetic Judgements

The central difference between analytic and synthetic judgements lies in
the question of whether or not the predicate concept is contained in the
subject concept. As Kant explains:
In the analytic judgment we keep to the given concept, and seek to
extract something from it. If it is to be affirmative, I ascribe to it only
what is already thought in it. If it is to be negative I exclude from it only
its opposite. But in synthetic judgments I have to advance beyond the
given concept, viewing as in relation with the concept something
altogether different from what was thought in it. This relation is
consequently never a relation either of identity or of contradiction; and
from the judgment, taken in and by itself, the truth or falsity of the
relation can never be discovered. (A154-5/B193-4)
The truth of an analytic judgement can be determined solely through an
analysis of the concepts that compose it. Since the concept of extension is
contained in the concept of body, an analysis of the concept of body is all
that is required to determine whether the judgement that all bodies have
extension is true. Since the concept of weight is not contained in the
concept of body, the judgement that all bodies are heavy is synthetic. To
determine the truth of a synthetic judgement requires that one advance
beyond the concepts contained in the subject concept and determine how
it is that a predicate concept can be true of a subject concept that does not
contain it. As Kant explains: in order to determine the truth of a synthetic
judgement, 'we must advance beyond a given concept in order to compare
it synthetically with another, a third something is necessary, as that
wherein alone the synthesis of two concepts can be achieved' (A155/B194).
To advance beyond a given concept and compare it synthetically with
another is not a task that can be undertaken by general logic. For logical
analysis can determine the truth of a judgement only in cases where the
predicate concept is contained in the subject concept. What general logic
cannot explain is how the matter provided by sensible intuition furnishes a
concept with content in the first place. As Kant observes: 'The explanation
of the possibility of synthetic judgments is a problem with which general
logic has nothing to do. It need not even so much as know the problem by
name. But in transcendental logic it is the most important of all questions'
(A154/B193).
Whereas general logic abstracts from all content of a judgement and
103
considers only its logical form, transcendental logic abstracts only from the
empirical content of a judgement. Taking a clue from the 'Transcendental
Aesthetic', where the pure intuitions of space and time are distinguished
from empirical intuitions, Kant distinguishes between 'the pure and
empirical thought of objects'. Transcendental logic is a science of how
concepts relate a priori to objects. The proper use of transcendental logic is
to 'determine the origin, the scope, and the objective validity' (A57/B81)
of the a priori modes of cognition by which we are able to cognize objects of
experience.
Because general logic abstracts from all content of cognition, we must
look to some other source to explain how concepts are provided with
content. Because transcendental logic only abstracts from the empirical
content of a concept, it has lying before it the transcendental conditions of
space and time. They provide the understanding with a manifold of a priori
intuition. Without the material provided to the understanding by this
manifold, concepts would be without content. Space and time, the a priori
forms of sensibility, are the conditions that make the receptivity of the
representations furnished by intuition possible. The task of transcendental
logic is to explain how concepts of objects can be produced from the
material provided to the understanding through a manifold of sensible
intuitions. But to turn a manifold of sensible intuitions into objects of
experience requires that the intuitions be gone through, taken up and
connected by an act that Kant calls synthesis:
By synthesis, in its most general sense, I understand the act of putting
different representations together, and of grasping what is manifold in
them in one cognition. Such a synthesis is pure, if the manifold is not
empirical but is given a priori, as is the manifold of space and time.
Before we can analyse our representations, the representations must
themselves be given, and therefore as regards content no concepts can
first arise by way of analysis. Synthesis of a manifold (be it given
empirically or a priori) is what first gives rise to a cognition. This
cognition may, indeed, at first be crude and confused, and therefore in
need of analysis. Still the synthesis is that which gathers the elements for
cognition and unites them [to form] a certain content. (A77/B103)
Synthesis is a mental act by which several representations are combined
into a single cognition. In empirical synthesis a manifold of sensible
appearances is combined into a single cognition. Thus in a judgement such
as, 'this rose is red', a manifold of impressions, the red of the flower, the
104
shape and configuration of its petals, the green of the stem, a pleasantly
sweet smell, and so on, are all combined under the concept of a rose in an
act of synthesis. In pure synthesis the a priori forms of space and time are
combined into a single cognition. Presumably what Kant means is that
through a process of abstraction one can arrive at the representation of the
spatio-temporal framework that provides the sensible conditions under
which all objects of experience must fall if they are to be recognized by the
understanding. Kant believes that this framework provides transcendental
logic with the content it requires to deduce the objective validity of the
categories from the conditions of possible experience.
There is a dispute in the literature as to whether Kant's account of
synthesis should be given a psychological or an epistemological reading.
On the psychological reading, Kant's explication of synthesis is an
explanation of how representations of objects are produced from a
manifold of appearances supplied by sensible intuition. Influenced in a
large part by Frege, many analytic philosophers have been deeply
suspicious of attempts to give psychological explanations of epistemological
and logical problems. They have thus shown little sympathy with synthesis
understood in a psychological fashion.
The problem with a psychological reading of synthesis is that it seems to
place Kant between the horns of a dilemma. Either he is giving an
empirical account, through self-reflection on his own cognitive powers, of
how human beings are able to produce objects of experience from sensible
intuitions, or he is describing activities and faculties that do produce
objects of experience but that are not empirically accessible. If Kant takes
the first horn of the dilemma, then the best that he can hope to establish in
the transcendental deduction is the empirical truth that human beings
have a subjective need to employ the pure concepts of the understanding.
What it does not establish is that these concepts are objectively valid. If
Kant takes the second horn, then he can give a satisfactory account of
synthesis only by positing a cognitive power that transcends the limits of
human experience. But this violates a fundamental tenet of Kant's critical
philosophy that we cannot acquire knowledge of things-in-themselves.
These issues have led to attempts to give an epistemological reading of
Kant's notion of synthesis. On the epistemological reading, Kant's account
of synthesis is an explanation of the epistemic conditions that must be met
for a concept to have representational content. As J. Michael Young
explains: 'Kant's theory of synthesis is best interpreted, not as an attempt
to explain how experience gets produced, but as an account of what it is for
concepts, and in particular pure concepts, to have representational
105
content' (Young 1994, 354). Young wishes to concentrate on 'important

logical and epistemological doctrines' that Kant connects with his notion
of synthesis and not on the causal account of how representational content
is produced. Instead of looking to the transcendental deduction, especially
the A edition, where what he calls the 'mental manufacture' (Young 1994,
332) conception of synthesis is most prevalent, Young concentrates on
Kant's initial introduction of the notion of synthesis in the passages cited
above. For it is there, he thinks, that the logical and epistemological
doctrines connected with synthesis are more prevalent.
To illustrate the epistemological role that synthesis plays in Kant's
philosophy, Young turns to the contrast between the analysis and the
synthesis of a concept. One way to define a concept is just to produce a list
of predicates that are true of it. Thus a triangle is a figure that is rectilinear
and three-sided. Now this is a good enough definition as far as it goes, for if
a figure is rectilinear and three-sided, it is a triangle. However, a triangle
cannot be adequately characterized just by a list of predicates. As Young
explains:
For something to be a triangle is not merely for it to satisfy certain
predicates. It is for there to be three line segments, suitably joined.
More carefully, it is for there to be three noncolinear points, joined by
line segments, and for the composite entity constituted by those lines to
be identical with the thing in question. (Young 1992, 114)
The difference between the two kinds of definitions can be expressed in
modern logical notation. The first would be expressed as (x) [Tx <-> (Fx &
Rx & TSx &'...)]. The second, in contrast, would be expressed as (x) {Tx
<-> (3w) (3y) (3z) [Pw & Py & Pz& -C (w,y,z) & (wy U J U zw =
x)y}2 (see Young 1992, 114). The first definition merely consists of a list of
predicates bound by a universal quantifier. The second definition, in
contrast, is not a mere list of predicates but the logical construction of an
object through the use of existential quantifiers nested within the scope of a
universal quantifier (For all x there exists somej . . . ) . A triangle is an object
consisting of three non-colinear points connected by three straight lines.
The first definition merely lists a set of properties possessed by triangles.
The second definition explicates how it is that the points and lines that
make up a triangle are related to each other to produce an object.
Young acknowledges that obviously Kant would not have expressed the
contrast in this way. The point that Kant seeks to make, according to
Young, is that
106
We cannot capture the content of a mathematical concept merely by

listing predicates that the instances of that concept must satisfy. Instead
we must posit objects and represent them as standing in certain relations.
Representing such objects involves intuition. In Kant's characteristic
claim, it involves representing a manifold, or multiplicity in intuition.
(Young 1992, 115)
Kant's distinction between the analysis and synthesis of a concept rests
upon his recognition of the limited scope of logical analysis. An analytic
definition of a triangle is merely a list of predicates that are true of
triangles. A synthetic definition defines a triangle by positing objects and
representing them as standing in a certain relationship. According to
Kant, the latter definition cannot be expressed analytically. For general
logic is only concerned with what partial concepts make up a concept. It is
not concerned with the origin of the partial concepts or how they are
related to each other in order to form a concept of an object. To explain
how a concept acquires its content and how the partial concepts that it
consists of are related to each other to form a concept of an object requires
an account of how the manifold provided by sensible intuition is taken up
and unified by imagination and understanding. Since it is synthesis that
performs this task, any account of how a concept requires its content must
include an account of how synthesis transforms the manifold provided by
sensible intuition into objects of experience.
One problem with Young's epistemological reading of synthesis is that
it does not adequately address the overwhelming textual evidence in
favour of the psychological reading. Thus in the passage of the
metaphysical deduction that Young believes favours the epistemological
reading of synthesis, Kant says that
Synthesis in general . . . is the mere result of the power of imagination,
a blind but indispensable function of the soul, without which we
should have no cognition whatsoever, but of which we are scarcely
ever conscious. To bring this synthesis to concepts is a function which
belongs to the understanding, and it is through this function of the
understanding that we first obtain cognition properly so called. (A78/
of the a
Here synthesis is characterized as a power of the imagination and as a
blind function of the soul 'of which we are scarcely ever conscious'. It is
clear that synthesis for Kant is a mental activity: an activity that he
107
explains in great detail in the A deduction. For Young's epistemological

reading of Kant's notion of synthesis to be more than just a modern
reconstruction of Kant's notion, we need a further explanation than is
given by Young of how an epistemological reading of synthesis squares
with the strong textual evidence in favour of the psychological reading of
synthesis.
As R. Lanier Anderson explains in a clear and helpful overview of the
problem:
both the psychological and the epistemological readings get something
right about the Critique. The psychological reading is right that Kant
offers us a theory of the mental actions involved in cognition, and the
epistemological reading is right that he is trying to provide a normative
account of cognitions. But this is just the puzzle. Precisely the
epistemological reader's emphasis on the normative status of Kant's
account seems to rule out the psychological reader's insistence on the
theory of mental actions. (Anderson 2001, 279)
Anderson believes that the way to resolve this tension between the
psychological and epistemological readings of synthesis lies in the
recognition that for Kant, along with most other early modern
philosophers, the mind was conceived as an instrument with normative
force internal to it. As he explains:
On the early modern conception, normativity is internal to any adequate
description of the mind . . . because the mind is thought of as an
instrument with a correct (intended) use 'built in'. By contrast, on the
currently standard way of thinking, activities and products of mind are
typically evaluated on the basis of standards applied from an outside
viewpoint. So conceived, norms are mind-independent achievements of
culture, binding on particular minds in virtue of their participation in
that culture, or in virtue of the norms' objectivity, or what have you; they
are not binding in virtue of mindedness as such. (Anderson 2001, 289)
Anderson claims that Kant accepted this early modern conception of
the mind as a site of intrinsically normative cognitive capacities, but his
work is paradigm-shattering, in the sense that he sees this basic
assumption as standing in need of special explanation: an adequate
philosophical theory of cognition must explain in detail how it is possible
108
for the mind to operate as a normative knowing power. (Anderson

2001, 289)
Like other early modern philosophers, Kant believed that normative
constraints are imposed upon a judgement through the correct operation
of the mind. Given such a conception of the mind, it is natural for Kant to
conceive of synthesis in both normative and psychological terms. If the
mind is a normative knowing power whose correct use can ensure the truth
of a judgement, then it should not be surprising that the act of synthesis by
which the mind converts the raw data of experience into determinate
objects of experience should possess normative force. A significant feature
of Frege's thought, spurred on by his strong stance against the legitimacy
of psychological explanations of epistemological and logical problems, was
to move the normative force of judgement outside of the mind and into a
third realm of being distinct from the mental and physical realms.
Frege on Arithmetic and Intuition

Given the distinctions between the analytic and the synthetic and the a
priori and a posteriori, there seem to be three different ways, according to
Frege, that the truths of arithmetic may be grounded. Either they are a
posteriori, as empiricists such as Mill have argued, or they are synthetic a
priori, as Kant has argued, or they are a priori and analytic, as Leibniz has
argued.
Frege begins section 12 of the Foundations by observing that if the truths
of arithmetic are synthetic a priori, then, 'there is no alternative but to
invoke a pure intuition as the ultimate ground of our knowledge of such
judgments, hard though it is to say of this whether it is spatial or temporal,
or whatever else it may be' (FA 17-18). Frege takes as a contemporary
example of such a view Hankel's locution 'pure conception of a
magnitude'. But what could such a term mean?
If we consider all the different things that are called magnitudes:
Numbers, lengths, areas, volumes, angles, curvatures, masses, velocities,
forces, illuminations, electric currents, we can quite well understand
how they can all be brought under the single concept of magnitude; but
the term 'intuition of magnitude', and still worse 'pure intuition of
magnitude' cannot be admitted as appropriate. (FA 1819)
109
What could an intuition of 100,000 be like? asks Frege, or for that matter
number or magnitude in general? In short, Frege cannot see how attaching
the description 'intuition' or 'pure intuition' to the concept of magnitude
can be of any assistance in the attempt to define the concept of a magnitude.
This leads him to an examination of Kant's conception of intuition.
Since Kant's claim that arithmetic is synthetic a priori rests on his claim
that we apprehend numbers through pure intuition, Frege examines
Kant's notion of intuition, beginning with a quote from the Logic: ' "An
intuition is an individual idea . . . a concept is a general idea . . . or an idea of
reflexion [i.e. discursive]"' (FA 19). Frege observes that this definition of
intuition is different from the one given in the Critique: ' "It is therefore
through the medium of sensibility that objects are given to us and it alone
provides us with intuitions'" (FA 19 [A19/B33]). Frege concludes that the
sense of intuition in the Logic is wider than the sense in the Critique, in that
it might be possible to call 100,000 an object of intuition in the former
sense but not in the latter, since 100,000 is not a general concept. However,
he claims that even 'an intuition in this sense cannot serve as the ground of
our knowledge of the laws of arithmetic' (FA 19).
In section 13 he explains why. He begins by pointing in agreement to
Leibniz's warning not to 'overestimate the extent to which arithmetic is
akin to geometry' (FA 19). He observes that one geometrical point by itself
cannot be distinguished from another, and the same holds true for lines
and planes. It is only when several points, lines and planes are included in
a single intuition that we can distinguish them. Frege concludes that:
In geometry, therefore, it is quite intelligible that general propositions
should be derived from intuition; the points or lines or planes which we
intuit are not really particular at all, which is what enables them to
stand as representatives of the whole of their kind. But with the numbers
it is different; each number has its own peculiarities. To what extent a
given particular number can represent all the others, and at what point
its own special character comes into play, cannot be laid down
generally in advance. (FA 19-20)
In section 14 Frege returns to a comparison of'the various kinds of truths in
respect of the domain that they govern' (FA 20). Empirical propositions
are concerned with the domain of the 'physically or psychologically
actual', while the truths of geometry govern all that is spatially intuitable.
The spatially intuitable governs not only what is physically actual but also
anything imaginable, no matter how fantastic or fanciful. It is only
110
through conceptual thought, however, that we can actually transcend

spatial intuition and study such things as non-Euclidean geometries. While
the postulate that space is curved may conflict with intuition (i.e. the
transcendental conditions of space and time), there is nothing selfcontradictory about it:
The fact that this is possible shows that the axioms of geometry are
independent of one another and of the primitive laws of logic, and
consequently are synthetic. Can the same be said of the fundamental
propositions of the science of number? Here, we have only to try
denying any one of them, and complete confusion ensues. Even to think
at all seems no longer possible. The basis of arithmetic lies deeper, it
seems, than that of any of the empirical sciences, and even than that of
geometry. The truths of arithmetic govern all that is numerable. This is
the widest domain of all; for to it belongs not only the actual, not only
the intuitable, but everything thinkable. Should not the laws of
number, then, be connected very intimately with the laws of thought?
(FA 21)
While the denial of an axiom of geometry may well conflict with intuition
it is not self-contradictory. However, the same cannot be said for the basic
laws of number. To say that 1 + 1 does not equal 2 is to utter a
contradiction.
The basic laws of arithmetic must lie deeper than the laws of the
empirical sciences, or even the science of geometry. For the truths of
arithmetic govern all that is numerable. Since the domain of countable
objects ranges over all that is actual (i.e. perceptible) as well as all that is
intuitable (i.e. the principles of geometry), imaginable or even conceivable, the subject matter of arithmetic 'is the widest domain of all', its laws
cover everything that is thinkable. Thus Frege concludes that an intimate
connection exists between arithmetic and the laws of thought. For both
sciences are a priori and analytic. And since arithmetic is the science of all
that is numerable, it is the science of all that is thinkable.
Frege on Analytic Judgements and Fruitfulness

Having settled that not all numbers can be known through intuition, Frege
next examines the claim that the laws of arithmetic are analytic:
111
Statements in Leibniz can only be taken to mean that the laws of

number are analytic, as was to be expected, since for him the a priori
coincides with the analytic. Thus he declares that the benefits of algebra
are due to its borrowings from a far superior science, that of the true
logic. In another passage he compares necessary and contingent truths
to commensurable and incommensurable magnitudes, and maintains
that in the case of necessary truths a proof or reduction to identities is
possible. (FA 21)
Frege admits that the view that the laws of arithmetic can be generated
from purely analytic principles 'has its difficulties. Can the great tree of the
science of number as we know it, towering, spreading, and still continually
growing, have its roots in bare identities? And how do the empty forms of
logic come to disgorge so rich a content?' (FA 22). He thinks that such a
criticism is justified if the claim that arithmetic is analytic amounts to an
espousal of formalism where mathematical symbols are thought of as
empty symbols and mathematics amounts to a kind of elaborate game.
However, Frege observes,
it is possible for a mathematician to perform quite lengthy calculations
without understanding by his symbols anything intuitable, or with
which we could be sensibly acquainted. And that does not mean that
the symbols have no sense; we still distinguish between the symbols
themselves and their content, even though it may be that the content
can only be grasped by their aid. We realize perfectly that other
symbols might have been assigned to stand for the same things. All we
need to know is how to handle logically the content as made sensible in
the symbols and, if we wish to apply our calculus to physics, how to
effect the transition to the phenomena. (FA 22)
Kant's mistake, according to Frege, was that he defined the analytic too
narrowly and thus 'underestimated the value of analytic judgements,
though it seems that he did have an inkling of the wider sense in which I
have used the term' (FA 99-100). On the basis of Kant's own definition,
according to Frege, 'the division of judgements into analytic and synthetic
is not exhaustive' (FA 100). Kant expresses the distinction in terms of
universal affirmative judgements, and here it is indeed correct to ask
whether or not a predicate concept is contained in a subject concept:
But how can we do this, if the subject is an individual object? Or if the
112
judgement is an existential one? In these cases there can simply be no

question of a subject concept in Kant's sense. He seems to think of
concepts as defined by giving a simple list of characteristics in no special
order. (FA 100)
Frege finds this a fruitless way of forming concepts and claims that the
reader will find no such definitions in his work. Any fruitful definition in
mathematics does not merely list the elements of a concept but deduces
other truths from the definition. Frege offers a geometrical illustration to
show how his new conception of concepts and definitions can advance our
knowledge without employing synthetic judgements:
If we represent the concepts (or their extensions) by figures or areas in a
plane, then the concept defined by a simple list of characteristics
corresponds to the area common to all the areas representing the
defining characteristics; it is enclosed by segments of their boundary
lines. With a definition like this, therefore, what we do in terms of our
illustration is to use the lines already given in a new way for the
purpose of demarcating an area. Nothing essentially new, however,
emerges in the process. (FA 100)
On the traditional (Kantian) interpretation, then, a definition of a concept
merely demarcates an area common to a set of objects that the definition is
true of. It lists the defining characteristics. Take, for example, the concept
of a square. In one case, draw a line from the midpoint of one side of the
square to the midpoint of the opposite side; in the other case, draw a line
from one corner of the square to the opposite corner. In the first case, two
rectangles will result; in the second, two triangles. Frege thinks that in both
examples no new knowledge is gained: existing boundaries are simply
redrawn. Thus one square can be defined as two triangles or two
rectangles, depending on where one draws a boundary. As Frege says in
'Boole's Logical Calculus and the Concept-script' (an essay soon to be
discussed in more detail):
If we look at what we have in the diagrams, we notice that in both cases
the boundary of the concept, whether it is one formed by logical
multiplication or addition, is made up of parts of the boundaries of the
concepts already given . . . In this sort of concept formation, one must,
then, assume as given a system of concepts, or speaking metaphorically
a network of lines. These really already contain the new concepts: all
113
one has to do is to use the lines that are already there to demarcate
complete surface areas in a new way. (PW 334)
The more fruitful definitions, on the other hand, draw boundaries that
have not previously been given.
What we shall be able to infer from it [the definition of a concept],
cannot be inspected in advance . . . The conclusions we draw from it
extend our knowledge, and ought therefore, on Kant's view, to be
regarded as synthetic; and yet they can be proved by purely logical
means and are thus analytic. (FA 101)
Frege eventually turns to metaphor to try to describe the nature of these
new definitions. The conclusions, contained in such definitions, he says,
should be likened to the way that 'plants are contained in their seeds, not
as beams are contained in a house' (FA 101).
Frege turns to a similar analogy in his posthumously published essay,
written much later in his career, 'Logic in Mathematics':
Science demands that we prove whatever is susceptible of proof and
that we do not rest until we come up against something unprovable. It
must endeavour to make the circle of unprovable primitive truths as small
as possible, for the whole of mathematics is contained in these primitive
truths as in a kernel. Our only concern is to generate the whole of
mathematics from this kernel. The essence of mathematics has to be
defined by this kernel of truths, and until we have learnt what these
primitive truths are, we cannot be clear about the nature of
mathematics. If we assume that we have succeeded in discovering
these primitive truths, and that mathematics has been developed from
them, then it will appear as a system of truths that are connected with
one another by logical inference. (PW 204-5)
The primitive truths of mathematics are kernels from which the entire
science of mathematics can be derived. The essence of mathematics is
contained in these kernels. In deriving theorems from these kernels new
information is gained about the science of mathematics, and thus new
knowledge is acquired. Like Kant, Frege agrees that knowledge requires
content as well as form. However, he objects to Kant's claim that content
can only be acquired through sensibility.
Frege discusses the importance of the fruitful definitions, and how his
114
logical language leads to them in more detail, in the posthumously

published essay 'Boole's Logical Calculus and the Concept-script'. Written
shortly after the publication of the Begriffsschrift, the essay aims to show
critics of his concept-script that its power to provide a clear logical analysis
of certain arithmetical expressions far exceeds what can be accomplished
by traditional logic. In particular, as the title suggests, he sets out to show
the superiority of his logical calculus over George Boole's algebraic logic.
Boole was an English mathematician and logician who applied the
methods of algebra to the logic of classes to develop a logical calculus more
powerful than traditional logic. Boole's views were championed and
expanded in Germany by Ernst Schroeder (1841 1902). Indeed,
Schroeder's critical review of Frege's Begriffsschrift, was a strong factor in
motivating Frege to write the article. Frege opens the essay with praise for
the genius of Leibniz:
In his writings, Leibniz threw out such a profusion of seeds of ideas that
in this respect he is virtually in a class of his own. A number of these
seeds were developed and brought to fruition within his own lifetime
and with his collaboration, yet more were forgotten, then later
rediscovered and developed further. This justifies the expectation that
a great deal in his work that is now to all appearance dead and buried
will one day enjoy a resurrection. As part of this, I count an idea which
Leibniz clung to throughout his life with the utmost tenacity, the idea of
a lingua characterica, an idea which in his mind had the closest possible
links with that of a calculus ratiocinatur. (PW 9)
In particular what Leibniz recognized, according to Frege, was the
'advantage of a script which compounded a concept out of its constituents
rather than a word out of its sounds' (PW 9). In developing his conceptscript, Frege adopts Leibniz's idea that an adequate logical notation
requires that the meaning of the symbols of the notation must be defined
purely conceptually. Thus it is best to avoid using words from ordinary
language as symbols of the logical calculus lest some aspect of the word's
ordinary meaning, not relevant to the conceptual role of the word, should
creep in unnoticed.
While Boole too can be seen as following in the footsteps of Leibniz,
Frege observes that a primary difference between his logic and Boole's is
that while Boole's logic is purely formal, his own aim is to develop a
language that expresses a content. Here Boole is in essential agreement
with Kant that logic is purely formal. Another place where Frege's
115
concept-script diverges from Boole's (and Kant's) logic is that while Boole
follows traditional logic and begins with concepts and builds judgements
from them, Frege starts 'from judgments and their contents, and not from
concepts' (PW 16). As Frege explains in an 1882 letter, 'I do not believe
that concept formation can precede judgement because this would
presuppose the independent existence of concepts, but I think of a concept
as having arisen by decomposition from a judgeable content' (PMC 101).
This enables him to take a given judgement and analyse its content in such
a way that a variety of different concepts can be produced from the same
judgemental content.
Speaking of the use of Venn diagrams, in which the relationships
between classes are illustrated by relationships of intersection, inclusion
and exclusion between circles, to illustrate the relationship between
concepts, Frege observes:
If we compare what we have here with the definitions contained in our
examples, of the continuity of a function and of a limit, and again that
of following a series which I gave in 26 of my Begriffsschrift, we see that
there's no question there of using the boundary lines of concepts we
already have to form the boundaries of new ones. Rather, totally new
boundary lines are drawn by such definitions and these are the
scientifically fruitful ones. Here too, we use old concepts to construct
new ones, but in so doing we combine the old ones together in a variety
of ways by means of the signs for generality, negation, and the
conditional. (PW 34)
Frege's procedure for providing a fruitful definition of a concept, then, is to
begin with the content of a complete judgement and then analyse the
judgemental content into a function/argument structure. From the
concepts produced by such an analysis, new and more complex
judgemental content can be constructed through the application of
generality, negation and the conditional. Take the proposition that 5 is
greater than 4. It can be analysed into the concept 5 is greater than x or
the relationy is greater than x. To these concepts and relations quantifiers
can be attached to create the propositions 5 is greater than some x or every
y is greater than some x. With the use of negation and the conditional,
arguments can be constructed that can produce significant and even
surprising results that could never be achieved by the traditional logical
methods practised by Kant and Boole. In particular Frege is able to give a
logical analysis of following in a series. It is this breakthrough that allows
116
him to account for the iterative nature of the natural numbers without
needing to invoke the pure intuition of time. What enables Frege to
accomplish this is the ability of his concept-script to account for the
relational structure and multiple general nature of the general truths of
arithmetic.3
Frege on Logic and Following in a Sequence

Frege rejects Kant's claim that 'without sensation no object would be
given to us'. He observes that numbers are objects and even those that
claim that smaller numbers are intuitable cannot explain how larger
numbers are intuitable. 'I have no wish', Frege says of Kant, 'to incur the
reproach of picking quarrels with a genius to whom we must all look up
with grateful awe' (FA 101). In drawing the distinction between analytic
and synthetic judgements Kant did a great service to philosophy. While
Kant was right, in Frege's eyes, in claiming that geometry is synthetic a
priori, he was wrong in his claim that arithmetic was as well. What Frege's
new logic has provided is a means, through a properly constructed
conceptual notation, to acquire access to the content of arithmetic without
the use of intuition. With such a notation all logical gaps in an inference
which might otherwise be explained in terms of intuition will be
eliminated. Frege's concept-script is designed to ensure that no jumps in
one's logical reasoning occur. Given the power of his new conceptual
notation, Frege thinks that he can prove a proposition that one would
think is synthetic, since it advances our knowledge, without any reliance
upon intuition. He then introduces the ancestral of a relation, a purely
logical notion first formulated in the Be griffsschrift, that he thinks will
enable him to generate the numbers from purely logical axioms. 'From this
proof, he observes, 'it can be seen that propositions which extend our
knowledge can have analytic judgements for their content' (FA 104).
As Frege explains:
Time is only a psychological necessity for numbering, it has nothing to
do with the concept of number. We do represent objects which are nonspatial and non-temporal by spatial or temporal points, and this may
perhaps be of advantage in carrying out the procedure of numbering;
but it presupposes, fundamentally, that the concept of number is
applicable to the non-spatial and the non-temporal. (FA 53)
111
Frege's logical developments provided him with a purely logical procedure

by which an infinite number of objects could be generated through the use
of relational propositions, variables and nested quantifiers. As Charles
Parsons summarizes: 'Thus the temporal notion of the successive addition
of units, or the even more concrete one of combining groups of objects, is
replaced in Frege's construction by the timeless relation of one class being
the union of two others' (Parsons 1969, 583).
Frege introduces the concept of following in a series in Part III of the
Be griffsschrift. In particular he proves that: 'If x has a property F which is
hereditary in the f-sequence, and if y follows x in thef-sequence, theny has the property
F (CN 111}. In some general remarks made before his proof he observes
that
we see in this example how pure thought (regardless of any content
given through the senses or even given a priori through an intuition) is
able, all by itself, to produce from the content which arises from its own
nature judgements which at first glance seem to be possible only on the
grounds of some intuition. We can compare this to condensation by
which we succeed in changing air, which appears to be nothing to the
childlike mind, into a visible drop-forming fluid. The propositions
about sequences developed in what follows far surpass in generality all
similar propositions which can be derived from any intuition of
sequences. (CN 167)
Without the resources necessary to analyse the multiple generality or
relational structure of a judgement, traditional logic only has the power
to determine whether one concept is contained in another by locating the
concepts in their proper place in the species/genus tree structure they are
contained within. Because Frege has the use of quantifiers and relational
expressions at his disposal, he is able to analyse the content of
arithmetical concepts such as that of following in a series without
invoking sense perception or intuition. Instead, using only principles of
pure thought, Frege is able to define the arithmetical concept of following
in a series.
In the Foundations', Frege invokes the definition of following in a series to
prove that for any natural number there is another number that follows it
in the series of natural numbers:
Now in order to prove that after every Number (ri) in the series of
natural numbers a Number directly follows, we must produce a concept
118
to which this latter Number belongs. For this we shall choose the
concept 'members of the series of natural numbers ending with n\ which
requires first to be defined. (FA 92)
After informing the reader that he originally introduced the concept of
following in a series in the Begriffsschrift, Frege sets out to use it to prove
that every natural number has a successor:
'if every object to which x stands in the relation (p falls under the
concept F, and if from the proposition that d falls under the concept Fit
follows universally, whatever d may be, that every object to which d
stands in the relation (p falls under the concept F, theny falls under the
concept F whatever concept F may be' is to mean the same as 'j follows
in the ^-series after x9 and again the same as 'x comes in the ^-series
before/. (FA 92)
Frege then remarks that 'It will not be time wasted to make a few
comments on this. First, since the relation (p has been left indefinite, the
series is not necessarily to be conceived in the form of a spatial and
temporal arrangement, although these cases are not excluded' (FA 92).
As he did in the Begriffsschrift, Frege observes that his definition has no
need to invoke a spatial or temporal arrangement in order to succeed. He
grants that it is certainly possible to state his definition of following in a
series in terms of focusing our attention on an initial object and then
shifting our attention from one object to the next until we reach the
object we wish to define. But, he insists, such a formulation is
unnecessary:
Whether y follows in the ^-series after x has in general absolutely
nothing to do with our attention and the circumstances in which we
transfer it; on the contrary, it is a question of fact, just as much as it is a
fact that a green leaf reflects light rays of certain wave lengths whether
or not these fall into my eye and give rise to sensation . . . (FA 93).
Frege concludes that:
My definition lifts the matter onto a new plane; it is no longer a
question of what is subjectively possible but of what is objectively
definite. For in literal fact, that one proposition follows from certain
others is something objective, something independent of the laws that
119
govern the movements of our attention, and something to which it is

immaterial whether we actually draw the conclusion or not. What I
have provided is a criterion which decides in every case the question
Does it follow after?, wherever it can be put; and however much in
particular cases we may be prevented by extraneous difficulties from
actually reaching a decision, that is irrelevant to the fact itself.
(FA 93)
Frege's definition provides a criterion for deciding whether or not a given
object follows another object in a series. The objective truth that is
revealed by such a definition is not something that is limited to sensation or
even the transcendental conditions of space and time. The ability to
provide such a definition of following in a series is a central component in
his attempt to ground the truths of arithmetic upon firm logical
foundations. As Frege explains: 'Only by means of this definition of
following in a series is it possible to reduce the argument from n to (n + 1),
which on the face of it is peculiar to mathematics, to the general laws of
logic' (FA 93). In other words, it is now possible to construct the natural
numbers without any reliance upon the inner intuition of time as a
necessary condition to formulating such a sequence.
So far then it has been shown that once we have defined the concept of
number and determined what kind of objects numbers are, it is possible to
account for the iterative nature of numbers without any recourse to the
pure intuition of time. Frege's logic has given him the ability to give a
logical definition of following in a series. All that is needed now to reduce
the truths of logic to the truths of arithmetic and to define the concepts of
arithmetic in terms of purely logical concepts is to give a purely logical
definition of the concept of number. If it can be shown that numbers are
logical objects, and if following in a series can be given a purely logical
definition, then the basic resources to establish arithmetic on a purely
logical foundation have been provided. What is left is the tedious work of
establishing this claim in a gap-free manner using the resources provided
by Frege's new logic.
Notes
For all x, x is a triangle if and only if x is a figure and x is rectilinear
and x is three-sided and . . .
For all x, x is a triangle if and only if there exist three points w^y and <:
120
that are non-colinear and the line segment joined at points wy, the line
segment joined at points yz and the line segment joined at points zw
together equal x.
The preceding discussion was influenced by Ruflino 1991 (esp. p. 189)
and Tappenden 1995.
Part 4
From Substance to Object
Introduction to Part 4
Questions such as whether or not there are any abstract objects, what
abstract objects there are, what abstract objects are and how we
know that they exist, what is the criterion for their existence, where
the dividing line comes between concrete and abstract objects - all
these are modern questions. At first sight, such a contention appears
ludicrous: one might well think such questions to be as old as
philosophy. But the fact is that the notion of an 'object' itself, as it is
now commonly used in philosophical contexts, is a modern notion,
one first introduced by Frege. (FPL 471)
As has been shown above, a key factor that enabled Frege to make the
advances in logic that he did was his decision to ignore the traditional
subject/predicate form of judgement and replace it with a function/
argument form of judgement. By replacing the traditional subject/
predicate form of judgement with a function/argument form of
judgement, Frege also freed himself from the traditional distinction
between substance and attribute, with its accompanying conception of
the ontological dependence of attributes upon substances. On Aristotle's
conception of substance from the Categories, individual substances are
beings in the primary sense of the term. All other entities, be they
qualities, relations or any other entities falling under one of the categories
other than substance, depend for their existence upon their inherence in
an individual substance. In place of the distinction between substance
and attribute, Frege introduces the distinction between concept and
object.
As Reinhardt Grossmann explains:
In holding that all concepts are related to objects through the fallingunder nexus, Frege rejects all substance-philosophies. Objects have no
natures, they are 'externally' related to the concepts under which they
fall. Hence Frege's objects are of the same kind as the so-called bare
124
particulars some contemporary philosophers defend. To say that an

entity is 'bare' is merely to say that it has no nature and not, of course,
that it has no properties. (Grossmann 989)
By ignoring the concept of substance and the metaphysics of inherence
associated with it, Frege is able to define a variety of different relational
expressions without the felt need to reduce such relations to a categorical
form of judgement. In doing so he escapes the shackles of the metaphysics
of inherence and treats a relational expression as designating a real relation
between two objects and not merely an accident of an accident that derives
its existence through adherence in an accident of a substance. According to
Frege, a function of one argument whose value is a truth-value is a
concept. 'A function of two arguments whose value is always a truth value,
we call a relation' (PW 244). Just as a concept requires an object for its
completion, a relation requires two objects for its completion. This signifies
a key difference between concepts and relations in Frege's eyes: 'As
functions of one argument are fundamentally different from objects, so
functions of two arguments are fundamentally different from functions of one
argument' (PW 240). Just as a concept must have sharp boundaries such
that it is necessary to know for any object whether it falls under the
concept, so too with relations.
This shift also enabled Frege to expand the scope of what counts as an
object, so that numbers, while not substances in the traditional sense of the
term, are nevertheless objects. According to Frege, some objects are actual
and some are not. So just because a number is not an actual physical object
like the individual man or horse that Aristotle cites as the paradigm case of
an individual substance in the Categories this does not mean that it is not an
object. The goal of Part 4 is to assess the nature and significance of Frege's
new conception of an object by comparing it to the traditional conception
of an object as a substance.
Leibniz and Kant on Substance and

Relations
Aristotle on Substance
In the Categories Aristotle draws a distinction between present-in and saidof predication. Present-in predication is a relationship between a substance
and one of the other categories (e.g. 'Socrates is white'). Said-of
predication is a relationship between a more specific entity denoted by
the subject of a sentence and a more general term denoted by the
predicate, where both terms denote entities of the same category (e.g.
'Socrates is a man', or 'White is a colour').
Some things are said-of a subject but not in a subject. The example that
Aristotle gives is in the category of substance. Man is said-of the individual
man. Similarly, animal is said-of man (Ib34). Said-of predication is
classificatory and intracategorical. It is classificatory because saying of
Socrates that he is a man, tells us what kind of thing he is in a way that
saying that he is white or musical does not. It is intracategorical because
'Socrates', 'man' and 'animal' all denote entities that fall under the
category of substance just as 'knowledge-of-grammar' and 'knowledge'
designate entities that fall under the category of quality.
Some things, according to Aristotle, are in a subject but not said-of a
subject. By 'in a subject', he says, 'I mean what is in something, not as a
part, and cannot exist separately from what it is in' (la24-5). White is not
a part of Socrates in the way that his nose or eyes are parts of him. To say
that x cannot exist separately from y is presumably to say that a colour
white cannot exist without there being white things that it inheres in.
However, while Socrates can exist without the colour white inhering in
him, he nevertheless does have some colour inhering in him. Similarly he
has to be at some place or in some position at some time, and so on, it
would appear, for the rest of the categories. The difference between the
two cases is that while the white in Socrates depends upon the existence of
126
an individual in which to inhere, Socrates does not depend for his existence
upon the particular shade of white that inheres in him. Instead he must
have some colour or be in some position, at some place, and so on. All of
the categories other than substance exist only in so far as they inhere in
individual sensible substances. And all specific and general substances exist
only in so far as primary substances exist. If there are no substances that
are white, then white does not exist.
Some things can both be in a subject and be said-of a subject.
Knowledge is in the soul but is said-of knowledge-of-grammar. Both
Socrates and man are substances, knowledge and knowledge-of-grammar
are qualities. Present-in predication is a relationship between substance
and the other categories. Said-of predication is a relationship between
individuals, species, and genera in the same category. Socrates is
an individual and man is a species and both fall under the category of
substance. Knowledge-of-grammar is a species of the genus knowledge;
both fall under the category of quality. Knowledge-of-grammar is presentin Socrates; man is said-of him.
Finally, some things are neither said-of nor present-in a subject. As
examples Aristotle gives the individual man or horse. Because they are
individuals they are not said-of a subject, for only universals can be said-of
subjects. Nothing that is 'individual and numerically one' (Ib6) is said-of
a subject. However, individuals in categories other than substance can be
in a subject. The only entities that are neither present-in nor said-of a
subject are primary substances:
A substance that which is called a substance most strictly, primarily,
and most of all is that which is neither said of a subject nor in a
subject, e.g. the individual man or the individual horse. The species in
which the things primarily called substances are, are called secondary
substances, as also are the genera of these species. For example, the
individual man belongs in a species, man, and animal is a genus of the
species; so these both man and animal are called secondary
substances. (2all-18)
It is because individual substances are always subjects and never
predicates that they are called substances 'most strictly, primarily, and
most of all'. As Aristotle says in Chapter 5, 'it is because the primary
substances are subjects for all the other things and all the other things are
predicated of them or are in them, that they are called substances most of
all' (2bl6-18). 2
Leibniz and Kant on Substance and Relations
127
While the forms of Plato and the atoms of Leucippus and Democritus
offered competing metaphysical conceptions, Aristotle's conception of
substance remained the most dominant metaphysical perspective in the
Western world until the rise of modern science and philosophy. Even then
substance played a central role in the philosophical thought of most early
modern philosophers. Indeed, a central philosophical concern of the
seventeenth and eighteenth centuries was how to situate Aristotle's
conception of substance with modern scientific methods and discoveries.
All recognized the need to radically revise or even reject Aristotle's
conception of substance in light of the new mechanical conception of the
universe arising out of the scientific discoveries and breakthroughs of
Galileo, Copernicus, Newton and other so-called natural philosophers.
Leibniz on Substance
In a passage from the Discourse on Metaphysics, Leibniz presents two
definitions of an individual substance, the first of which shows the clear
influence of Aristotle's definition from the Categories:
It is of course true that when a number of predicates are attributed to a
single subject while this subject is not attributed to any other, it is called
an individual substance. But this is not enough, and such a definition is
merely nominal. We must consider, then, what it means to be truly
attributed to a certain subject . . . So the subject term must always
include the predicate term in such a way that anyone who understands
perfectly the concept of the subject will also know that the predicate
pertains to it. This being premised, we can say that it is the nature of an
individual substance or of a complete being to have a concept so
complete that it is sufficient to make us understand and deduce from it
all the predicates of the subject to which the concept is attributed. An
accident, on the other hand, is a being whose concept does not include
everything that can be attributed to the subject to which the concept is
attributed. (PPL 307)
While Leibniz does not reject the criterion of substance put forth by
Aristotle in the Categories, he does think that it does not provide a
sufficiently complete criterion. What needs to be added is a further
criterion, namely his own conception of substance as a complete concept.
This conception is grounded in Leibniz's definition of truth as concept-
128
containment. To determine whether Alexander the Great was a king or

whether he conquered the Persian empire one need only determine
whether his complete concept contains these concepts. Of course no
human being would ever be able to determine the complete concept of any
individual substance. God, being infinite, is the only being who could
know the complete concept of an individual substance. Nevertheless, what
distinguishes Alexander from other famous conquerors such as Julius
Gaesar or Attila the Hun, or for that matter any other human being, is his
complete concept. No two individual substances possess the same complete
concept. Such a conception of substance does not seem to rely on
Aristotle's distinction between essential and accidental prediction. What
makes Alexander the Great the very being he is (or was) is not some
essential core of defining attributes but the sum total of every attribute, no
matter how small, that makes up one individual concept.
But already in the Discourse there lurks another conception of substance:
I believe that anyone who will meditate about the nature of substance
. . . will find that the nature of body does not consist merely in extension,
that is, in size, shape, and motion, but that we must necessarily
recognize in body something related to souls, something we commonly
call substantial form. (PE 44)
In his later thought, this conception of substance as substantial form is
developed more fully. As Donald Rutherford explains in an article on
Leibniz's later metaphysics:
From the 1680s onward, Leibniz remains committed to a set of basic
assumptions about substance. To be a substance is, minimally, to be an
individual principle of action, which persists through change and which
serves as a ground for the existence and properties of all other things.
The latter requirement, which sees substance as an ultimate explanatory principle, implies for Leibniz that whatever is true of a substance
must be true in virtue of its own nature, and not the nature of
something else. (CCL 126)
Such a conception of substance, according to Rutherford, commits Leibniz
to the position that a substance must be a 'source of action' that
is sufficient to produce all and only those modifications that are
predicable of that substance. In Leibniz's terminology, the nature of
129
any substance must be spontaneous, or causally self-sufficient, such that

it is dependent for the production of its states on no other created being.
(CCL 126)
He calls these simple substances monads.
According to Leibniz, a monad 'is nothing but a simple substance that
enters into composites simple, that is, without parts' (PE 213). Monads
have no parts, extension or shape and they are indivisible. They 'are',
according to Leibniz, 'the true atoms of nature and, in brief, the elements
of things' (PE 213). Given their simple nature, Leibniz continues,
There is also no way of explaining how a monad can be altered or changed
internally by some other creature, since one cannot transpose anything in
it, nor can one conceive of any internal motion that can be excited,
directed, augmented, or diminished within it, as can be done in
composites, where there can be change among the parts. The monads
have no windows through which something can enter or leave. (PE 2134)
Despite their simple natures, Leibniz insists that monads have intrinsic
qualities by which one monad is distinguishable from another. Because the
nature of a monad consists solely of intrinsic properties, 'the monad's
natural changes come from an internal principle, since no external cause can
influence it internally' (PE 214).
Leibniz is faced with the challenge of determining what the relationship
is between these two different conceptions of substance. On one conception
a substance is the complete concept of an individual. Thus the complete
concept of Alexander the Great consists in all of the accidents and relations
to other substances that make up a complete description of every aspect of
his life. On the other conception, one that leads to Leibniz's conception of
individual substances as monads, an individual substance is a simple,
immaterial, self-sufficient source of action. The challenge, then, is to
explain the relationship between simple individual substances such as
monads and the complex individual substances such as Alexander the
Great.
One of Leibniz's solutions to this problem was to treat individual
substances as phenomena that can be reduced to simple substances. As he
explains in a letter:
I don't really eliminate body, but reduce it to what it is. For I show that
corporeal mass, which is thought to have something over and above
130
simple substances, is not a substance, but a phenomenon resulting from

simple substances, which alone have unity and absolute reality. I
relegate derivative forces to the phenomena, but I think that it is
obvious that primitive forces can be nothing but the internal strivings of
simple substances . . . (PE 181)
On this conception individual substances such as Alexander the Great or
his horse Bucephalus are not substances in the true sense of the term but
phenomenal substances that are reducible to the simple substances that
compose them. To connect the conception of substance as the complete
concept of an individual such as Alexander the Great with the conception
of substance as simple immaterial sources of action, Leibniz needs to show
how substance as a complete concept is either identical to or reducible to
substance as a phenomenal body. He then needs to show that a
phenomenal substance is reducible to simple substances. But any chance
that Leibniz might have had in successfully carrying out such a project
requires that he reduce relational propositions to non-relational propositions, and it is just this that Kant believes cannot be done.
Kant on Substance
The goal of Kant's metaphysical deduction was to show that there are 12
basic forms of judgement that underlie our ability to judge and thus
understand anything. The ability to use these forms of judgement, in turn,
presupposes the use of the corresponding 12 pure concepts of the
understanding or categories. Thus corresponding to the categorical form
of judgement (S is P] is the category of inherence and subsistence, or
substance and accident. The categorical form of judgement falls under the
heading of relation along with the hypothetical and the disjunctive forms
of judgement. Thus the categorical form of judgement consists of a subject,
a predicate and a copula that binds them. This form of judgement imposes
no conditions upon the kind of concept that can serve as a subject concept.
Thus in the two sentences 'Every philosopher is human' and 'Every human
is an animal', the word 'human' designates a predicate concept in the first
sentence and a subject concept in the second. However, according to the
category of substance and inherence, a substance is that 'which can exist as
subject and never as a mere predicate' (B149). On such a criterion of
substance the word 'human' would not designate a substance because it
can designate a predicate concept as well as a subject concept. Thus the
131
pure category of substance narrows eligible candidates for substance from

anything that can be designated by a subject term in a categorical
judgement to anything that can only be designated by a subject term and
can never be used as a predicate term. Kant then narrows this conception
of substance even further when he schematizes the category of substance.
According to Kant: 'The schema of substance is permanence of the real in
time, that is, the representation of the real as a substrate of empirical
determination of time in general, and so as abiding while all else changes'
(A143/B183).
Kant, then, seeks to refine further the pure category of substance as that
'which can exist as subject and never as a mere predicate' (B149) to that of
permanent substratum of change. As he explains in the 'First Analogy', a
chapter dedicated to proving that substance is the permanent substratum
of change:
Our apprehension of the manifold of appearance is always successive, and
is therefore always changing. Through it alone we can never determine
whether this manifold, as object of experience, is coexistent or in
sequence. For such determination we require an underlying ground
which exists at all times, that is, something abiding and permanent, of which
all change and coexistence are only so many ways (modes of time) in
which the permanent exists. And simultaneity and succession being the
only relations in time, it follows that only in the permanent are relations
of time possible. In other words, the permanent is the substratum of the
empirical representation of time itself; in it alone is any determination of
time possible. (A182-3/B225-7)
In order to distinguish whether our apprehension of a manifold of
appearances is either successive (following one another in time) or
coexistent (related to one another in space) requires that such appearances
possess an underlying substratum that remains permanent in time. As
Kant explains in his concluding remarks to the 'First Analogy':
'Permanence is thus a necessary condition under which alone appearances
are determinable as things or objects in a possible experience (A189/B232).
Kant's Critique of Leibniz's Conception of Substance

Kant's most direct and sustained criticism of Leibniz in the Critique occurs
in the 'Amphiboly of Concepts of Reflection'. A reflection, according to
132
Kant, is not concerned with the derivation of concepts from objects but
with the subjective conditions that make the derivation of concepts
possible in the first place. Kant calls transcendental reflection, 'the
comparison of representations with the cognitive faculty by which it
belongs, and by means of which I distinguish whether it is as belonging to
the pure understanding or to sensible intuition that they are to be
compared with each other' (A261/B317). A transcendental reflection,
then, is concerned with determining the sources of our concepts by
determining to which cognitive faculties they owe their origin. An
amphiboly is simply an ambiguity. So Kant's basic criticism of Leibniz is
that in his reflection on concepts he conflates concepts whose source lies in
the understanding with concepts whose source lies in sensibility.
Kant claims that there are four basic relations by which concepts in the
mind can be compared with each other: (1) identity and difference; (2)
agreement and opposition; (3) inner and outer; and (4) form and matter.
In contrast to Leibniz, he insists that to determine in what way concepts
are the same or different, in agreement or disagreement, and so on,
requires not just a comparison of the concepts but a determination of
whether they owe their origin to the understanding or sensibility. Kant
thus contrasts logical reflection that is concerned solely with the
comparison of concepts with transcendental reflection that is concerned
solely with determining to which cognitive faculty the concepts being
compared owe their origin.
According to Kant,
Leibniz's monadology has no basis whatsoever save his mode of
representing the distinction of inner and outer merely in relation to the
understanding. Substances in general must have some internal nature
which is therefore free from all outer relations, and consequently also
from composition. The simple is therefore the basis of that which is
inner in things-in-themselves. But that which is inner in the state of a
substance cannot consist in place, shape, contact, or motion (these
determinations being all outer relation, and we can therefore assign to
substances no inner state save that through which we ourselves inwardly
determine our sense, namely, the state of the representations. (A274/B330)
The basis of Leibniz's monadology, according to Kant, rests upon his
contention that the distinction between inner and outer is characterized
solely in terms of concepts of the understanding. But to characterize a
substance solely in terms of concepts is to characterize it solely in terms of
133
intrinsic determinations. But since external relations cannot be reduced to

internal determinations, it is impossible to give a purely conceptual
account of a phenomenal substance's relation to its own parts, to other
substances, or of the relation between a phenomenal substance and the
understanding. For Leibniz, the intrinsic nature of a simple substance
consists solely of internal determinations. By stripping simple substances of
all extrinsic relations and identifying them solely in terms of their intrinsic
determinations, Kant claims that Leibniz is led to the view that
phenomenal substances are not real. For they are not conditioned by the
transcendental forms of space and time but simple substances that are
produced solely through inner determination of the understanding. He
conflates an intrinsic logical property of being a non-relational property of
a categorical judgement with an inner determination of the mind.
The solution to this problem of how to reduce phenomenal substances
to simple substances is to accept that phenomenal substance is substance in
the primary sense even though it consists in nothing but relations and is
thus not real:
. . . an abiding appearance in space (impenetrable extension) can
contain only relations and nothing at all that is absolutely inward, and
yet be the primary substratum of all outer perception. Through mere
concepts I cannot, indeed, think what is outer without thinking
something that is inner; and this for the sufficient reason that concepts
of relation presuppose things which are absolutely [i.e. independently]
given, and without these are impossible. (A284/B340)
A phenomenal substance is an abiding appearance in space. As such it
consists solely in external relations. Because it is impossible adequately to
account for the nature of external relations solely through the use of
concepts, it is impossible to account for the nature of phenomenal
substances by reducing them to simple substances consisting solely of
intrinsic determinations.
Whatever intrinsic determinations a phenomenal substance in space
might have, then, consists merely in the sum of relations that makes up its
matter. As Kant explains, the only means we have to become acquainted
with the matter of a phenomenal substance is 'through forces which are
active in this and that space, either bringing other objects to it (attraction),
or preventing them from penetrating into it (repulsion and impenetrability)' (A265/B321).
However, that does not mean that a phenomenal substance consists
134
solely in our acquaintance with external relations. Underlying these

relations is the schematized category of substance:
All that we know in matter is merely relations (what we call the inner
determinations of it are inward only in a comparative sense), but
among these relations some are self-subsistent and permanent and
through these we are given a determinate object . . . It is certainly
startling to hear that a thing is to be taken as consisting wholly of
relations. Such a thing is, however, mere appearance, and cannot be
thought through pure categories; what it itself consists in is the mere
relation of something in general to the senses. (A285/B341)
While a phenomenal substance is an appearance consisting solely of external
relations, underlying these external relations is a self-subsistent and
permanent substratum by which the appearances provided to the understanding through sensible intuition are transformed into determinate
objects. Without application of the schematized category of substance to
the material provided to the understanding through sensible intuition, it
would be impossible to have a representation of an object. As Kant explains:
If by merely intelligible objects we mean those things which are thought
through pure categories, without any schema of sensibility, such objects
are impossible. For the condition of the objective employment of all our
concepts of understanding is merely the mode of our sensible intuition,
by which objects are given us; if we abstract from these objects, the
concepts have no relation to any object. (A286/B342)
By treating both inner determinations and extrinsic relations as products of
the understanding, Leibniz is led to claim that true substance consists
solely of internal determinations: qualities whose inherence in the
substance are in no way affected by any other substances. However, this
leads to the problem of explaining the relationship between simple
substances (with their intrinsic determinations) and phenomenal substances (with their extrinsic relations). Simple substances consist solely of
intrinsic determinations. Phenomenal substances consist solely of external
relations. Since external relations are not reducible to internal determinations, phenomenal substances cannot be reduced to simple substances.
Here Kant seems to draw a metaphysical and epistemological conclusion
based largely upon a limitation of formal logic. It is not possible to analyse
the logical structure of certain kinds of relational expressions solely
135
through analytic judgements, hence the only way to account for our ability
to know such objects is to invoke synthesis. One must choose, then,
whether simple substance, a product of pure concepts, or phenomenal
substance, a product of concepts and intuitions, is to be taken as primary.
Kant comes down on the side of phenomenal substance, but the price he
pays is to settle for a conception of substance consisting solely of relations
with no internal nature of its own other than permanence through time.
Leibniz's sin, in Kant's eyes, was that he believed that things in themselves
could be known purely through conceptual thought. As Kant explains:
Leibniz erected an intellectual system of the world, or rather believed that
he could obtain cognition of the inner nature of things by comparing all
objects merely with the understanding and with the separated, formal
concepts of its thought . . . He compared all things with each other by
means of concepts alone, and naturally found no other differences save
those only through which the understanding distinguishes its pure
concepts from one another. The conditions of sensible intuition, which
carry with them their own differences, he did not regard as original,
sensibility being for him only a confused mode of representation, and
not a separate source of representations. Appearance was, on his view,
the representation of the thing in itself. (A270/B326)
Leibniz, according to Kant, believed that the understanding can cognize
things in themselves through the logical analysis of concepts. What he
failed to recognize was that the cognition of an object requires more than
the logical analysis of concepts. It also requires an act of synthesis by which
intuitions provided to the understanding through sensibility are taken up
and combined into the concepts of empirical objects that make up our
experience of the world. Whereas Leibniz believed that objects could be
acquired solely through logical analysis, Kant believes that a concept can
be supplied with an object only through synthesis.
The epistemological constraints necessary to guarantee the reliability of
our claims to knowledge entail that objects of cognition cannot exist
independently of the synthesizing power of the mind. As Kant explains:
we cannot represent to ourselves anything as combined in the object
which we have not ourselves previously combined, and that of all
representations combination is the only one which cannot be given
through objects. Being an act of the self-activity of the subject, it cannot
be executed save by the subject itself. (B130)
136
No object can be represented to the mind prior to it being conditioned by

the forms of sensibility and understanding. It is the self-activity of the mind
that is responsible for a representation of an object. Thus it makes no sense
to think of an object as something existing independently of the conditions
of sensibility and understanding. For Kant, all objective content
presupposes the self-activity of a subject who must synthesize the content.
To appreciate the difference between Kant's conception of substance and
the traditional conception of substance arising out of the thought of
Aristotle, it will help to take a brief look at how John Locke, the great British
empiricist, developed a conception of substance in light of the new
mechanical world-picture arising out of the development of modern science.
In An Essay concerning Human Understanding, John Locke's conception of
substance shows the influence of Aristotle's Categories conception of
substance:
So that if any one will examine himself concerning his Notion of pure
substance in general, he will find he has no other Idea of it at all, but only a
Supposition of he knows not what support of such Qualities as are
commonly called Accidents. If any one should be asked, what is the
subject wherein Colour or Weight inheres, he would have nothing to
say, but the solid extended parts . . . The Idea then we have, to which we
give the general name Substance, being nothing but the supposed, but
unknown support of those Qualities, we find existing, which we imagine
cannot subsist, sine re substante, without something to support them, we
call that Support Substantial which, according to the true import of the
Word, is in plain English, standing under, or upholding. (I I. XXI11.2)
Locke observes that we form ideas of particular kinds of substances by
combining ideas which through experience and observation are usually
seen, 'to exist together, and are therefor supposed to flow from the
particular internal Constitution, or unknown Essence of that Substance'
(II.XXIII.3). Thus, according to Locke, we come to have an idea of a
man or a horse by having an idea of a collection of simple ideas existing
together around an underlying substratum.
Locke claims that there are three ideas that make up our complex
notion of a substance. One is the idea of primary qualities discoverable by
the senses and in a substance even when we don't perceive them. As
examples, he lists ideas of bulk, figure, number situation and motion of
parts of bodies. Such ideas are in substances whether or not we recognize
them. Next is the idea of secondary qualities, which 'are nothing but the
137
Powers, those substances have to produce several Ideas in us by our Senses;

which ideas are not in the things themselves' (II.XXIII.9). For example,
colour is not something that is a primary quality of a substance, as bulk or
figure are, but an idea added to the substance by the mind. Finally, there
are active and passive powers to give or receive alterations of primary
powers. The only notion we have of these powers comes from the sensible
simple ideas that they produce. Locke concludes that 'our specifick Ideas of
Substances are nothing else but a Collection of a certain number of simple Ideas,
considered as united in one thing'' (I I. XXI11.14).
Locke distinguishes between substance as an immaterial spirit and
substance as a body. 'Our Idea of a Body, as I think, is an extended solid
Substance, capable of communicating Motion by impulse: and our Idea of
our Soul, as an immaterial Spirit, is of a Substance that thinks, and has a
power of exciting Motion in Body, By Will, or Thought' (II.XXIII.22).
Locke argues that both conceptions of substance are obscure and that
ultimately our idea of one is just as obscure as our idea of the other. Despite
his scepticism over the question of whether we can ever arrive at a clear
and complete understanding of the nature of material and immaterial
substance, Locke believed, like Aristotle, in the existence of substance as an
underlying substratum that is distinct from the secondary qualities and
powers through which we are able to perceive sensible substances.
Like Locke, Kant is sceptical that we can ever acquire knowledge of the
pure underlying substratum around which all of the sensible attributes by
which we cognize the substance are ordered. However, Locke conceives of
substance as something with independent existence that underlies the
attributes of phenomenal substances, even if he doubts that human beings
will ever acquire complete knowledge of the nature of such a substance.
Leibniz believed in the existence of simple substances or monads as the
reality underlying the existence of phenomenal substances. Leibniz agreed
with Locke that the substratum underlying phenomenal substances cannot
be known through experience, but he believed it can be known through
the exercise of pure reason. Kant, in contrast, takes the fact that pure
substance as an individual object characterized solely in terms of intrinsic
attributes is unknowable to the conclusion that substances are not real but
ideal.
In a discussion of Locke's conception of substance, Michael Ayers
claims that
the principles of pre-Kantian substance theories, including the notions
of dependent and independent existence, can only be understood in
138
terms of a conception of ontology as absolute. For the conceptualist, all

beings are concept-relative: for the realist, some beings are conceptrelative but others, substances, are not.
Acceptance of the famous logical criterion of substantiality, when
interpreted from a realist point of view, is simply equivalent to the
doctrine that an individual substance is a natural or real individual, in
some sense 'given' and not 'constructed'. (Ayers 1991, 113)
For Kant, there is ultimately no real distinction, at least not one that can
be recognized by human beings, between external reality and the
conceptual structure that we impose on the appearances provided to the
understanding through sensible intuition. This leads him to postulate a
conception of an object not as an independently existing entity but as a
product of the joint activity of sensibility and understanding.
Notes
I shall ignore the problem of whether or not a soul is a substance and
what its relationship is to the individual man or horse that Aristotle
cites as the paradigm examples of primary substance.
I have limited my discussion of Aristotle's conception of substance to
the views expressed in the Categories. Aristotle's conception of
substances in such works as the Physics, De Anima, and especially the
Metaphysics adds numerous complications to the picture of substance. I
could find no way of accurately discussing this other conception of
substance without expanding the scope of the discussion far beyond
what is needed in this work.
The Context Principle and Numbers as

Logical Objects
The Context Principle

Frege first mentions the context principle in the introduction to the
Foundations, where it is listed as the second of three fundamental principles
that underlie his attempt to define numbers. These principles advise the
reader
always to separate sharply the psychological from the logical, the
subjective from the objective;
never to ask for the meaning of a word in isolation but only in the
context of a proposition;
never to lose sight of the distinction between concept and object. (FA x)
Frege warns that if the first principle is not observed there will be nothing
to prevent the conflation of a subjective idea with an objective concept or
object. If the second principle 'is not observed, one is almost forced to take
as the meanings of words mental pictures or acts of the individual mind,
and so to offend against the first principle as well' (FA x). The third
principle entails that a concept cannot be made into an object without
altering it. The context principle, then, is first invoked as a methodological
principle, meant to avoid an atomistic and psychologistic conception of
meaning that treats subjective ideas as the meanings of words. But Frege's
employment of the principle consists of more than a means to avoid
treating the meanings of words as subjective ideas. In section 62 of the
Foundations, he employs the principle in an attempt to give a contextual
definition of number.
According to Dummett, Frege's employment of the context principle in
section 62 of the Foundations is the first clear expression of the 'linguistic
turn' in philosophy. Frege's problem, Dummett observes, was to explain
how numbers can be accessible if they are not intuitable or perceptible:
140
His solution was to invoke the context principle: only in the context of a
sentence does a word have meaning. On the strength of this, Frege
converts the problem into an enquiry how the senses of sentences
containing terms for numbers are to be fixed. There is the linguistic turn.
The context principle is stated as an explicitly linguistic one, a principle
concerning the meaning of words and their occurrence in sentences; and
so an epistemological problem with ontological overtones, is by its
means converted into one about meanings of sentences. (FPM 111)
An important consequence of the linguistic turn, on Dummett's reading of
Frege, is that logical categories determine ontological commitment.
Now Frege's use of the ontological term 'object' is strictly correlative to his
use of the linguistic term 'proper name': whatever a proper name stands
for is an object, and to speak of something as an object is to say that there
is, or at least could be, a proper name which stands for it. The question
then naturally arises in which realm, the linguistic or the ontological, the
primary principle of classification is to be applied. (FPL 556)
Dummett claims that for Frege the primary principle of classification is
linguistic: 'For a name introduced by a contextual definition, there simply
is no answer to the question what its reference is on its own; all we have is a
method of explaining the truth-condition of any sentence in which it
occurs' (FP 496).
Frege, then, attempts to use the context principle in the Foundations to
establish the existence of numbers as logical objects. But this attempt to use
logic alone to establish the existence of abstract objects leaves what
Dummett calls 'a residual uneasiness': 'In what sense are we entitled to
suppose that abstract objects are constituents of an external reality, when
the possession of reference by their names has been interpreted as a matter
wholly internal to the language?' (FPL 499).
But if the assignment of meaning to numerical expressions is a matter
that is wholly internal to the logical language, then it is unclear how the
numbers that are meanings of the numerical expressions can exist as selfsubsistent objects wholly independent of the logical language by which
they are grasped. Thus if one pushes the implications of using a contextual
definition to define numbers to its logical conclusion, then it would appear
that numbers cannot exist independently from whatever language we are
using to define them. 'Thus', Dummett concludes, 'in a certain sense . . . for
Frege, the world does not come to us articulated in any way; it is we who,
The Context Principle and Numbers as Logical Objects
141
by the use of our language (or by grasping the thoughts expressed in that
language), impose a structure on it' (FPL 5034).
We will return to section 62 of the Foundations, and Dummett's
assessment of its ontological implications later. Before doing so, however,
we need to see how Frege was led to invoke the context principle to
attempt a contextual definition of number. Prior to his own positive
attempt to define numbers Frege offers a critical assessment of the attempts
of his predecessors and contemporaries to provide a definition of number.
What follows will be a selective account of some of Frege's criticism.
Number is not a Property of External Things

Frege begins his critical examination with a discussion of the view that
numbers are the properties of external things.
In language, numbers most commonly appear in adjectival form and
attributive construction in the same sort of way as the words hard or
heavy or red, which have for their meanings properties of external
things. It is natural to ask whether we must think of the individual
numbers too as such properties, and whether, accordingly, the concept
of Number can be classed along with that of say colour. (FA 27)
However, Frege observes, while I am unable to change the colour of a
thing just by thinking of it differently, I can change the number of
something by thinking of it differently. For 'I am able think of the Iliad as
either one poem or as 24 Books, or as some large Number of verses' just by
bringing the object in front of me under a different concept.
If I give someone a stone with the words: Find the weight of this, I have
given him precisely the object he is to investigate. But if I place a pile of
cards in his hands with the words: Find the Number of these, this does
not tell him whether I wish to know the number of cards, or the
complete pack of cards, or even say of points in the game of skat. To
have given him the pile in his hands is not yet to have given him
completely the object he is to investigate; I must add some further
words - cards, or packs, or points. (FA 28-9)
That a number is not the property of an external thing is shown by the
difference in logical behaviour between the attribution of number and of a
142
property such as colour. Since we cannot change the colour of an object by

thinking of it in a different way but we can change the number that we
assign to an object by thinking of it in a different way, number is not a
property. Now this, Frege suggests, might lead someone to think that
number is something subjective.
Numbers are not Subjective

Frege thus examines the claim of a certain Lipschitz, who claims that
' "Anyone who proposes to make a survey of a number of things, will begin
with some one particular thing and proceed by continually adding a new one
to those previously selected"' (FA 33). Frege adamantly rejects any such
suggestion:
No description of this kind of the mental processes which precede the
forming of a judgement of number, even if more to the point than this one,
can ever take the place of a genuine definition of the concept. It can never
be adduced in proof of any proposition of arithmetic; it acquaints us with
none of the properties of numbers. For number is no whit more an object
of psychology or a product of mental processes than, let us say, the North
Sea is. The objectivity of the North Sea is not affected by the fact that it is
a matter of our arbitrary choice which part of all the water on the earth's
surface we mark off and elect to call the 'North Sea'. This is no reason for
deciding to investigate the North Sea by psychological methods. In the
same way too number is something objective. (FA 34)
Frege insists that a genuine definition of the concept of number can
contain no description of mental processes that precede the formation of a
judgement about number. He denies that mental processes can be deduced
from a proposition of arithmetic. Finally, he claims that a psychological
description of how we become acquainted with numbers cannot acquaint
us with any of the properties of numbers.
He compares the objectivity of numbers with the objectivity of the
North Sea:
If we say 'the North Sea is 10,000 square miles in extent' then neither by
'North Sea' nor by '10,000' do we refer to any state or process in our
minds: on the contrary, we assert something quite objective, which is
independent of our ideas and everything of the sort. (FA 34)
143
On the next page Frege explains what he means by objective:

I distinguish what I call objective from what is handleable or spatial or
actual. The axis of the earth is objective, so is the centre of mass of the
solar system, but I should not call them actual in the way the earth itself
is so. We often speak of the equator as an imaginary line] but it would be
wrong to call it a fictitious line] it is not a creature of thought, the product
of a psychological process, but is only recognized or apprehended by
thought. If to be recognized were to be created, then we should be able
to say nothing positive about the equator for any period earlier than the
date of its alleged creation. (FA 35)
Frege distinguishes what he calls objective from anything handleable,
spatial or actual. It is not required that what is objective be handleable,
spatial or have causal power. It is not the product of a psychological
process, and in being recognized it is not created. Here Frege is clearly
parting ways with Kant's claim that the concept of a number is the result
of a mental process of constructing the concept in intuition:
The concept of magnitude in general can never be explained except by
saying that it is that determination of a thing whereby we are enabled
to think how many times a unit is posited in it. But this how-many-times
is based on successive repetition, and therefore on time and the synthesis
of the homogeneous in time. (A242/B 300)
For Frege, a number is something objective and mind-independent even
though it is not actual. For Kant, in contrast, what exists is what is actual
and what is actual is that 'which is bound up with the material conditions
of experience, that is, with sensation' (A218/B265-6). As he explains a few
pages later:
The postulate bearing on the cognition of things as actual does not,
indeed, demand immediate perception (and, therefore, sensation of which
we are conscious) of the object whose existence is to be known. What we
do, however, require is the connection of the object with some actual
perception, in accordance with the analogies of experience, which
define all real connection in an experience in general. (A225/B272)
For Kant, the paradigm conception of an object is a concrete perceptible
object existing in space and time and having causal efficacy. On such a
144
conception numbers would not be objects since they do not exist in space
and time and they do not have causal efficacy. For Frege, there is no reason
to limit our conception of an object to the realm of the actual, and thus no
need to deny that numbers are objects just because they are not actual.
Frege on One and Unity

In Part II of the Foundations, Frege establishes that numbers are not
properties of external things, they are not subjective and they are not sets.
In Part III he turns to a discussion of the views of some of his predecessors
and contemporaries on the nature and relationship between the concept of
number and the concepts of unity (Einheit) and one (Eins). He begins with
the view that the number 1 stands for the property of an object. Thus we
would treat the word 'one' in the expression 'one city' in the same way that
the word 'wise' is treated in the expression 'wise man'. He immediately
raises a number of objections to this view. To begin with he observes that
since the property of being one is true of every single thing, it is unclear
what the use of such a property would be.
He then raises an objection based on the difference in grammatical
behaviour of a word such as 'one' and a word such as 'wise'. If I say that
Solon is wise and that Thales is wise, then I can say that Solon and Thales
are wise. However, if I say that Solon is one and that Thales is one, I
cannot say that Solon and Thales are one. But if the word 'one' designates
a property in the same way that the word 'wise' does, it is hard to explain
this difference in grammatical behaviour.
The content of a concept diminishes as its extension increases; if its
extension becomes all-embracing, its content must vanish altogether. It
is not easy to imagine how language could have come to invent a word
for a property which could not be of the slightest use for adding to the
description of any object whatsoever. (FA 40)
Such difficulties in defining the property designated by the word 'one'
have led some to define 'one' as whatever is grasped in one act of the
understanding. Frege points to Leibniz and Baumann as examples of those
who seek such a definition of'one'. The problem with such attempts is that
it is also possible to grasp many objects in one grasp of the understanding.
On such a view our conception of numbers depends not upon objective
facts but upon 'our way of regarding them' (FA 41).
145
Frege next turns to a suggestion by Baumann that being undivided and

being isolated are suitable criteria for being one. But, he observes, even
animals are able to distinguish things based on these criteria. We surely
don't want to claim that a dog understands the word 'one'. 'This is hardly
credible', he claims, 'and yet it certainly distinguishes individual objects:
another dog, its master, a stone it is playing with, these certainly appear to
the dog every bit as isolated, as self-contained, as undivided as they do to
us' (FA 42). Frege concludes that being isolated and undivided 'cannot be
what is essential to our concept' (FA 42). Yet such characteristics do seem
to be at least part of what it is to be one.
Some writers have thus sought to add further properties to define the
concept of one. Thus to be one is to be undivided and indivisible.
'Obviously', Frege observes of such writers,
by tightening up its internal cohesion without limit, they hope to arrive
at a criterion for their unit which is independent of any arbitrary way of
regarding things. This attempt collapses because we are then left with
practically nothing fit to be called a unit and to be numbered. (FA 43)
Frege concludes that 'we must finally abandon the view that in designating
a thing a unit we are adding to a description of it' (FA 44). We are thus left
with a series of questions. How does calling something a unit add to calling
it a thing if any and every thing can be called a unit? 'Why do we ascribe
identity to objects that are to be numbered? And is it only ascribed to
them, or are they really identical?' (FA 44). Frege observes that while 'no
two objects are ever completely identical', we are always able to find some
attribute that even the most disparate things have in common. This has led
some writers to identify units as identical without qualification. Units are
simply things that are identical to one another. One way to define number
then would be to say what kind of thing a number is. The goal would be to
determine certain essential characteristics that are common to all numbers
such as indivisibility or unity, say, through a process of abstraction: 'For
suppose that we do, as Thomae demands, "abstract from the peculiarities
of the individual members of a set of items", or "disregard, in considering
separate things, those characteristics which serve to distinguish them" '
(FA 45).
Frege illustrates the problem of such an approach with an example: I can
arrive at the concept of a cat through a process of abstraction. By
disregarding the various accidental properties that distinguish one cat from
another it is possible through the mental process of abstraction to form the
146
concept of a cat. With such a concept I am able to distinguish cats from other
animals like dogs and gerbils, but such a concept does not allow me to
distinguish one cat from another. For having abstracted from all of the
features that are common to all cats, I am unable to use the resulting concept
to distinguish one cat from another. But that is just what we need the concept
of number to do: to be able to distinguish one number from another. So if
you want to define number as an object, as Frege seeks to do, the process of
abstraction will be of no use. For at best it enables us to distinguish an object
that falls under one concept, a cat, say, from an object that falls under
another, a dog, but it does not enable us to distinguish one cat from another.
Such an approach might well tell us what kind of thing a number is, but it
will not enable us to distinguish one number from another.
So to account for the nature of numbers in terms of units it is not
enough merely to claim that one can define a number simply by
abstracting out all differences between objects and treating numbers as
identical units. It is also necessary to explain the diversity of numbers.
Frege summarizes the problems faced by his predecessors and
contemporaries in their attempts to give an adequate definition of the
concept of number in terms of the concepts of one and unity:
If we try to produce the number by putting together different objects,
the result is an agglomeration in which the objects contained remain
still in possession of precisely those properties which serve to distinguish
them from one another; and that is not the number. But if we try to do
it in the other way, by putting together identicals, the results run
perpetually together we never reach a plurality.
If we use 1 to stand for each of the objects to be numbered, we make
the mistake of assigning the same symbol to different things. But if we
provide the 1 with differentiating strokes, it becomes unusable for
arithmetic. (FA 50)
Frege's definition of number is designed to meet these deficiencies in his
predecessors' and contemporaries' attempts at definition. He prefaces his
solution to the problem with a question. When a statement of number is
made, what are we asserting it of? Frege thinks that light can be shed on
this question by examining the context in which judgements involving
numbers are used:
While looking at one and the same external phenomenon, I can say
with equal truth both 'It is a copse' and 'It is five trees', or both 'Here
147
are four companies' and 'Here are 500 men'. Now what changes here
from one judgement to the other is neither any individual object, nor
the whole, the agglomeration of them, but rather my terminology. But
that is itself only a sign that one concept has been substituted for
another. This suggests as the answer to the first of the questions left open
in our last paragraph [see previous quote], that the content of a
statement of number is an assertion about a concept. (FA 59)
In Frege: Philosophy of Mathematics, Dummett suggests translating the last
clause as 'that the content of an ascription of number consists in
predicating something of a concept' (FPM 88). Dummett's translation
illuminates the point that for Frege to attribute a number to something is
to predicate an expression, containing a number word as a component, to
the concept under which the object or objects to be counted fall. If a
number of soldiers are standing in order in a field and a general points to
the formation and asks a subordinate 'How many?', the subordinate will
give a different answer depending upon whether he thinks the general is
asking him how many men or how many companies. In the sentence 'Four
companies consist of 500 men', '500' is an element in the predicate '500
men'. In particular it is a singular term that designates the number 500.
This predicate is, in turn, attributed to the predicate 'four companies'
which designates a concept under which the mustered soldiers fall.
For Frege, whether the same external phenomenon is treated as four or
500 depends upon what concept the external phenomenon falls under.
Commenting on Frege's claims that we can say with equal truth of the
same external phenomenon that 'Here are four companies' and 'Here are
500 men', Michael Ayers claims that Frege 'did so without reflecting on
the different ontological status of companies and men. Accordingly, he
regarded natural unity as a psychological matter' (Ayers 118). Ayers cites
an earlier passage from the Foundations to support this claim:
The more the internal contrasts within a thing fade into insignificance
by comparison with the contrasts between it and its environment, and
the more the internal connexions among its element overshadow its
connexions with its environment, the more natural it becomes for us to
regard it as a distinct object. For a thing to be 'united' means that it has
a property which causes us, when we think of it, to sever it from its
environment and consider it on its own. (FA 42)
Ayers concludes from these two passages that, 'For Frege, all unity is
148
concept-relative: nothing in the physical world is individual, or a whole

but thinking makes it so' (Ayers 1991, 118). Ayers' interpretation of these
passages can be divided into two elements.
1. For Frege all unity is concept-relative, and
2. For Frege what makes something an individual in the physical world
depends upon what property we choose to bring it under. And from
this he concludes that for Frege what makes something individual in
the physical world is a psychological matter.
Now it is certainly true that for Frege some phenomenon in the physical
world can be treated as a unity when it has some property that causes us,
when thinking of the phenomenon in terms of that property, to sever it
from its environment and treat it as an individual. And this is indeed a
psychological matter. However, it is only because 500 men fall under the
concept of four companies that it is possible to say with equal truth of the
same external phenomenon that it is Tour companies' or '500 men'. And
this is an objective matter for Frege, not a psychological one.
However, there is a broader point that Ayers is making. In broadening
his conception of what an object is to include anything that can be thought
and thus counted, Frege is abandoning the metaphysical outlook that
draws a firm ontological distinction between a man and a number. Since
they are both objects that can be counted, Frege assigns no ontological
priority to men over companies of men. For Aristotle, in contrast, a man is
a substance and a number is a quantity that depends for its existence upon
the existence of substances. While there can be 500 men without there
being four companies, there cannot be four companies without there being
500 men. Thus for Frege, on Ayers' reading, all unity is concept-relative
because Frege's conception of an object leaves no room for a distinction
between an object whose existence and identity conditions are absolute
and an object whose existence and identity conditions are relative to
whatever concept we choose to bring it under. More will be said on this
point in the final chapter.
The Context Principle and Frege's Definition of Number

Having established that the content of an ascription of number consists in
predicating something of a concept, Frege sets out 'to complete the Leibnizian
definition of the individual numbers by giving the definitions of 0 and 1':
149
the number 0 belongs to a concept, if the proposition that a does not fall
under that concept is true universally, whatever a may be.
Similarly we could say: the number 1 belongs to a concept F, if the
proposition that a does not fall under F is not true universally, whatever
a may be, and if from the propositions 6a falls under F and 6b falls under
F it follows that a and b are the same.
It remains still to give the general definition of the step from the given
number to the next. Let us try the following formulation: the number (n +
1) belongs to a concept F, if there is a number falling under F and such that
the number n belongs to the concept Tailing under F, but not a\ (FA 67)
Despite their initial plausibility, however, Frege rejects these definitions on
the somewhat obscure grounds that
we can never to take a crude example decide by means of our
definitions whether any concept has the number Julius Caesar
belonging to it, or whether that same familiar conqueror of Gaul is a
number or is not. Moreover we cannot by the aid of our suggested
definitions prove that, if the number a belongs to the concept F and the
number b belongs to the same concept, then necessarily a = b. Thus we
should be unable to justify the expression 'the number which belongs to
the concept F9, and therefore should find it impossible in general to
prove a numerical identity, since we should be quite unable to achieve a
determinate number. It is only an illusion that we have defined 0 and 1;
in reality we have only fixed the sense of the phrases
'the number 0 belongs to'
'the number 1 belongs to'
but we have no authority to pick out the 0 and 1 here as self-subsistent
objects that can be recognized as the same again. (FA 68)
The first problem with the proposed definition is that by itself it will not let us
distinguish an object, like Julius Caesar, who is clearly not a number, from an
object, like 1, which is a number. To know that T refers to a number and
Julius Caesar' does not requires that we at least know what kind of thing ' 1'
refers to. However, there is nothing in the above definitions that would
enable us to determine this. Secondly, from such a definition we are not able
to infer that if two numbers belong to the same concept, then the numbers
are identical. A criterion of identity is required to determine that if a numbers
the concept F and b numbers the concept F, then a is identical with b. As a
result of these two deficiencies the proposed definitions cannot produce a
150
determinate number and thus cannot justify that it is a determinate number

which falls under a given concept. At most the definitions have fixed the sense
of a phrase like 'the number 1 belongs to', as in the sentence 'The number 1
belongs to the concept number of the Earth's moons.' If'the number 1' does
in fact refer to a determinate object, then the above sentence is true.
However, the meaning of the singular expression 'the number 1' cannot be
determined solely in terms of the role that it plays in the above definition.
Frege begins his last attempt to answer the 'Julius Caesar problem' in
section 57. He opens the section by insisting that,
It is time to get a clearer view of what we mean by our expression 'the
content of a statement of number is an assertion about a concept'. In
the proposition 'the number 0 belongs to the concept F9, 0 is only an
element in the predicate (taking the concept F to be the real subject).
For this reason I have avoided calling a number such as 0 or 1 a
property of a concept. Precisely because it forms only an element in
what is asserted, the individual number shows itself for what it is, a selfsubsistent object. I have already drawn attention above to the fact that
we speak of'the number 1', where the definite article serves to class it as
an object. In arithmetic this self-subsistence comes out at every turn, as
for example in the identity 1 + 1 = 2 . (FA 68-9)
If one takes 'the concept F9 as the subject of the proposition 'the number 0
belongs to the concept F\ then 'the number O3 is not a predicate but an
element within the predicate. The role of 'the number 0' is to be an
element in the expression 'the number 0 belongs to'. Within that predicate
it serves as a singular term that refers to an object. The logical behaviour of
such a singular term in an arithmetical expression is evidence that
numbers are self-subsistent objects. As further evidence that numbers are
self-subsistent objects, Frege points to other linguistic behaviour associated
with number words: we can attach the definite article to them, they
behave as singular terms in arithmetical expressions like '1 + 1 = 2 ' .
Frege admits that numerical expressions often appear in the attributive
position in the language of everyday life, but claims such expressions can
always be transformed into an expression in which the numerical
expression appears in the subject position. Thus 'Jupiter has four moons'
can be transformed into the identity statement: 'The number of Jupiter's
moons is the number 4' where the 'is' is that of identity and not
predication. Identities, he further observes, are the most typical types of
expression in arithmetic.
151
Frege next turns to an objection that on his characterization of numbers

it is not possible to form an idea of, say, the number 4. 'It may be', he
admits, 'that every word calls up some sort of idea in us, even a word like
"only"; but this idea need not correspond to the content of the word; it
may be different in different men' (FA 70). It is in the context of an
objection against construing numbers as objects because we cannot form
an idea of them that Frege invokes the context principle for a second time:
Time and time again we are led by our thought beyond the scope of our
imagination, without thereby forfeiting the support we need for our
inferences. Even if, as seems to be the case, it is impossible for men such
as we are to think without ideas, it is still possible for their connexion
with what we are thinking of to be entirely superficial, arbitrary and
conventional.
That we can form no idea of its content is therefore no reason for
denying all meaning to a word, or from excluding it from our
vocabulary. We are indeed only imposed on by the opposite view
because we will, when asking for the meaning of a word, consider it in
isolation, which leads us to accept an idea as the meaning. Accordingly
any word for which we can find no corresponding mental picture
appears to have no content. But we ought always to keep before our
eyes a complete proposition. Only in a proposition have the words
really a meaning. It may be that mental pictures float before us all the
while, but these need not correspond to the logical elements in the
judgement. It is enough if the proposition taken as a whole has a sense;
it is this that confers on its parts also their content. (FA 71)
Even if the meaning of a concept used in an inference transcends the
imagination and no idea can be found to correspond to it, this need not
affect the meaning of the concept and thus the role that it plays in a
judgement. Even if it is impossible for humans to think without ideas, these
ideas need have no intrinsic connection with the nature of the objects that
are thought through them. Thus that content of a word that contributes to
determining the truth of the sentence that it is contained within (i.e. its
conceptual content) need have no intrinsic connection with any ideas by
which that content is grasped. The context principle is then invoked, again
as a methodological principle, as a means of avoiding confusing any
mental pictures that may float before one's eyes when making a judgement
with 'the logical elements in the judgement' that contribute to its truth.
Frege then goes on to insist that his claim that numbers are self-subsistent
152
objects is consistent with his claim that number words acquire their
meaning only within the context of a proposition:
The self-subsistence which I am claiming for number is not to be taken to
mean that a number word signifies something when removed from the
context of a proposition, but only to preclude the use of such words as
predicates or attributes, which appreciably alters their meaning. (FA 72)
Here Frege appears to hedge upon his claim that numbers are selfsubsistent objects. Unlike in the Begriffsschrift where numbers are
contrasted with variables and quantifiers in that their meaning is fixed
independently of their roles in propositions, Frege now believes that
numbers can be considered to be self-subsistent objects even though their
meaning is now acquired only in the context of a sentence. Does this mean
that the ontological status of numbers is somehow parasitic upon the
logical language through which they are grasped? Frege's claim that
numbers are self-subsistent is simply meant to contrast them with
predicative or attributive expressions, and certainly seems to encourage
such a reading. For the difference between a singular term and a predicate
expression is a logical contrast not an ontological one. Whereas the claim
that numbers are self-subsistent carries with it the connotation of
independent existence, the contrast between singular terms and predicates
or attributes need not imply that numbers have independent existence but
only that the numerical expressions that designate numbers serve a
different logical role in propositions than do predicate expressions.
Frege's Attempted Contextual Definition of Number

Given that numbers, according to Frege, are objective entities that cannot
be perceived through sense perception, intuition or ideas, the obvious
question to ask is:
How, then, are numbers to be given to us, if we cannot have any ideas
or intuitions of them? Since it is only in the context of a proposition that
words have any meaning, our problem becomes this: To define the sense
of a proposition in which a number word occurs. That, obviously, leaves
us a very wide choice. But we have already settled that number words
are to be understood as standing for self-subsistent objects. And that is
enough to give us a class of propositions which must have a sense,
153
namely those which express our recognition of a number as the same

again. If we are to use the symbol a to signify an object, we must have a
criterion for deciding in all cases whether b is the same as a, even if it is
not always in our power to apply this criterion. (FA 73)
Thus, in section 62 of the Foundations, Frege famously invokes the context
principle in an attempt to give a contextual definition of the numbers. He
claims that his demonstration that numbers are self-subsistent objects can
be used as a starting-point to define a class of propositions which express
our recognition of a number as the same again. The task that Frege sets for
himself, then, is,
to define the sense of the proposition
'The number which belongs to the concept F is the same as that which
belongs to the concept G";
That is to say, we must reproduce the content of this proposition in
other terms, avoiding the use of the expression
'The Number which belongs to the concept F\
In doing this, we shall be giving a general criterion for the identity of
numbers. When we have thus acquired a means of arriving at a
determinate number and of recognizing it again as the same, we can
assign it a number word as its proper name. (FA 73)
For numerical terms to be provided with a criterion of identity a noncircular condition for the truth of an identity statement must be discovered.
According to Dummett, Frege 'tacitly assumes that the fundamental type of
terms standing for numbers consists of those of the form "the number of 7*s",
or, in his jargon, "the number belonging to the concept T7'" (FPM 112).
Dummett labels this term 'the number of F^ 'the cardinality operator'.
Frege's goal then, according to Dummett, is to arrive at an explanation of
the cardinality operator through the specification of the truth-conditions of
propositions of the form 'the number of Fs = the number of Gs\ In order for
such an explanation to work, it must avoid circularity. That is in specifying
the truth-conditions no appeal can be made to 'singular terms denoting
numbers presented as objects' (FPM 114).
Frege hopes to establish this condition through the use of what has come
to be called (thanks to Frege) 'Hume's Law':
Hume long ago mentioned just such a means: 'When two numbers are so
combined as that the one has always an unit answering to every unit of
154
the other, we pronounce them equal.' This opinion that numerical

equality or identity may be defined in terms of oneone correlation, seems
in recent years to have gained widespread acceptance among mathematicians. But it raises at once certain logical doubts and difficulties, which
ought not to be passed over without examination. (FA 734-)
Adopting Hume's suggestion, then, Frege offers the following contextual
definition of number: the number of Fs is the same as the number of Gs if and
only if there are just as many Fs as Gs. ((Nx:Fx = Nx:Gx) = (Fx Gx)).
Frege raises three objections against such a definition of number, rejects
the first two but sustains the third, and as a result drops his attempt to give
a contextual definition of number. The first two objections need not
concern us. However, the third objection is of concern for it leads to the
rejection of Frege's attempted contextual definition of number.
Frege illustrates why such a definition will not work with an example
not from arithmetic but from geometry:
In the proposition 'the direction of a is identical with the direction of V
the direction of a plays the part of an object, and our definition affords
us a means of recognizing this object as the same again, in case it should
happen to crop up in some other guise, say as the direction of b. But this
means does not provide for all cases. It will not, for instance, decide for
us whether England is the same as the direction of the Earth's axis if I
may be forgiven an example which looks nonsensical. Naturally no one
is going to confuse England with the direction of the Earth's axis; but
that is no thanks to our definition of direction. (FA 77-8)
Frege, then, raises anew the Julius Gaesar problem. His attempt to define
the sense of a numerical identity in which a number word occurs by
fixing the sense of the proposition: 'The number which belongs to the
concept F is the same as that which belongs to the concept G" (FA 73),
does provide a criterion for determining an object as the same again, even
if the object should appear in another guise. For if line a is parallel with
line b, then it is possible to determine whether the direction of line a is
identical with the direction of line b by determining whether the two are
parallel. Frege suggests that we use this definition of direction as a model of
our definition of number. Thus the new definition solves one part of the
Julius Gaesar problem for it enables us to determine whether two numbers,
given in different ways (i.e. falling under different concepts), are identical.
For what the above definition enables one to do is to determine that if the
155
number of Waverley novels that Scott wrote is the same as the number of
counties in Utah, then knowing that the number of counties in Utah is 29
and that Scott wrote just as many Waverley novels enables one to infer
that Scott wrote 29 Waverley novels. Whatever number falls under the
first concept falls under the second.
However, the above definition does not solve the other half of the Julius
Caesar problem. For it does not tell us what kind of a thing a direction is
and thus gives us no means of distinguishing a direction from a country,
say, or any other thing that is not a direction. 'What we lack', says Frege,
'is the concept of direction; for if we had that, then we could lay it down
that, if q is not a direction, our proposition is to be denied' (FA 78). The
same problem holds true for the concept of number. The definition, then,
has given us a criterion for determining a number as the same again but it
has not established what kind of thing a number is. As long as an identity is
asserted between two numbers as falling under the same concept, then the
definition works. But when an identity is asserted between a number and,
say, Julius Caesar (e.g. the number of Fs = Julius Caesar), then the
definition gives us no means of determining the truth of this identity
because it offers no means of determining what kind of thing a number is
and thus no means of determining whether or not Julius Caesar is a
number. We can tell by the role that number-words play in true
arithmetical sentences that they refer to self-subsistent objects, but we
cannot tell from this role what kind of self-subsistent objects they are. On
the force of this objection Frege relinquishes his attempt to give a
contextual definition of number and instead offers an explicit definition of
number as the extension of a concept. The attempt to give a contextual
definition of number has failed because it cannot tell us what kind of thing
a number is. To account for this deficiency Frege offers an explicit
definition of numbers in terms of extensions of concept:
Seeing that we cannot by these methods obtain any concepts of
direction with sharp limits to its application, nor therefore, for the same
reasons, any satisfactory concept of Number either, let us try another
way. If line a is parallel to line b, then the extension of the concept 'line
parallel to line <2? is identical with the extension of the concept 'line
parallel to line V\ and conversely, if the extensions of the two concepts
just named are identical, then a is parallel to b. (FA 79)
He observes that if one replaces lines with concepts and parallelism with
'the possibility of correlating one to one the objects which fall under the
156
one concept with those which fall under the other', then one could define
the concept of number as follows: 'the Number which belongs to the
concept /MS the extension of the concept "equal to the concept F*''(FA 79
80). Frege thinks that this definition succeeds where the previous one failed
because it tells us what kind of thing a number is, namely the extension of a
concept. Of course it was just this move that eventually led to the fatal
consequences of Russell's paradox.
Note
1
In a footnote Frege justifies this claim by observing that the expression

contains a definite article.
Two Views on Existence
A Diagnosis of the Problem

Frege raises the Julius Caesar problem against his attempt to answer the
question of how numbers can be given to us if we cannot have any ideas or
intuitions of them. If a word acquires its meaning only in the context of a
proposition (context principle) and if the proposition of the form that there
are just as many F$ as Gs (Hume's principle) is true, we can use these two
principles to fix the sense of an identity statement of the form the number
of F$ is the same as the number of Gs. Frege suggests that such a definition
will yield a class of propositions that express 'our recognition of a number
as the same again' (FA 73). The hope is that since it has already been
established (at least to Frege's satisfaction) that numbers are self-subsistent
objects, such a definition will provide a criterion of identity for deciding in
all cases whether an object a is the same as an object b, even if it is not
always possible to apply such a criterion.
The problem is that Hume's principle together with the context
principle cannot fix the sense of mixed identity statements such as Julius
Caesar is the same as the number 4. While Frege's contextual definition of
number gives us a criterion for determining a number as the same again,
what it does not give us is a criterion for determining what kind of thing a
number is. As long as an identity is asserted between two numbers as
falling under the same concept, then the definition works. But when an
identity is asserted between a number and, say, Julius Caesar (e.g. the
number of Fs = Julius Caesar), then the definition gives us no means of
determining the truth of this identity because it offers no means
of determining what kind of thing a number is and thus no means of
determining whether or not Julius Caesar is a number. As Richard Heck
explains:
it is rarely mentioned, because it is not thought important, that Frege
takes for granted that we do recognize that Caesar is not a number . . .
158
The example that Frege chooses is one about which he takes us to have
strong intuitions: Whatever numbers may be, Caesar is not among
them. Thus, there must be more to our apprehension of numbers than a
mere recognition that they are objects that satisfy HP [Hume's
principle]. Something explains why 'no one is going to confuse [Caesar]
with the [number zero]'. Frege's conclusion is that any complete
account of our apprehension of numbers as objects must include an
account of how we recognize that Caesar is not a number. But HP alone
yields no such explanation. (Heck 2005, 1712)
Now Kant's diagnosis of what went wrong for Frege would undoubtedly
focus on the fact that Frege's definition seeks to derive an object solely from
a concept, without any mediating assistance from intuition. But the only
way to derive an object from a concept is through the schematism of the
categories. Since the category of substance is the category that produces
objects of experience out of the sensible data of intuition, Frege's failure to
invoke such a schematism explains his inability to produce a coherent
notion of number as a logical object. As Kant explains:
If I leave out permanence (which is existence in all time), nothing remains
in the concept of substance save only the logical representation of a subject
a representation which I endeavour to realise by representing to myself
something which can exist only as subject and never as predicate. But not
only am I ignorant of any conditions under which this logical preeminence may belong to anything; I can neither put such a concept to any
use, nor draw the least inference from it. For no object is thereby
determined for its employment, and consequently we do not know
whether it signifies anything whatsoever. (A242-3/B301)
For Kant, if you will recall, 'The schema of substance is permanence of the
real in time, that is, the representation of the real as a substrate of
empirical determination of time in general, and so as abiding while all else
changes' (A143/B183). And 'Permanence is . . . a necessary condition
under which alone appearances are determinable as things or objects in a
possible experience (A189/B232). For all of the attributes and relations
that make up an object, the red of a ball, the texture of its surface, its size,
and so on, there must be an underlying permanent substratum around
which those attributes and relations are ordered. When the colour of the
ball fades from red to pink, or it loses its roundness by being punctured, it
is the same object that is undergoing these changes. Without a permanent
159
underlying substratum of change there is nothing around which the

understanding can order various attributes that make up a concept of an
object. Such an object of experience presupposes an act of synthesis in
which the schematized category of substance is applied to the manifold of
appearances provided by sensible intuition. However, it is impossible to
produce such an object of experience through the application of the pure
category of substance, which is merely the concept of a subject that is never
a predicate (or in Frege's case an object is anything that is never a
function). But that condition is not enough to explain how the various
marks that make up the concept of an object can be unified in such a way
as to yield a determinate representation of an object. By making numbers
into logical objects that transcend the conditions of space and time, objects
accessible solely through conceptual analysis, Frege is left with a
conception of an object with no substratum around which the various
attributes that make up an object can be ordered. Frege's attempt to define
numbers in terms of a contextual definition is an attempt to give a purely
conceptual definition of number. But any object defined by such a
definition would merely be a thing in itself: an object that transcends the
conditions of space and time and is thus unknowable. Thus on Kant's
principles, if numbers are logical objects then they are unknowable. It
should come as no surprise, then, that Frege's contextual definition of
number leaves him with no means of distinguishing one kind of thing such
as a number from another kind of thing such as a human being.
In this respect Frege has the converse of the problem of the proponents of
definition by abstraction that he criticized earlier in the Foundations. Their
problem, according to Frege, is that while concept formation by way of
abstraction enables them to distinguish one kind of thing, say a cat, from
another kind of thing, say a dog, it does not allow them to distinguish two
objects from each other when they fall under the same concept. For them
definition by abstraction can only reach down to the level of concepts. It
cannot be used to define individual objects such as numbers. Thus such a
definition does not allow them to distinguish one cat from another, or more
to Frege's concern, one number from another. While Frege's attempted
definition of number enables him to distinguish one number from another if
it is only numbers that are in question, it does not allow him to distinguish a
number from something that is clearly not a number, such as a human being.
That is, it does not allow him to fix the sense of mixed identity statements.
Now the criticism of Frege that I am attributing to Kant relies upon a
feature of traditional logic and metaphysics, the ordering of attributes
around a central substantial core, that Frege has rejected in replacing the
160
traditional subject/predicate conception of a judgement with the function/

argument conception. This frees Frege's conception of an object from the
shackles of the internal/external controversy grounded in a conception of
substance as possessing an inner essential core upon which all external
relations are dependent.
Thus, according to Leibniz:
in the universe, place and position, quantity (number, proportion) are
only relations, resulting from other things which per se constitute or
terminate change . . . Considering the matter more accurately, I saw
that they are only mere results, which themselves do not constitute any
intrinsic denomination, and thus are only relations which need a
fundament from the predicament of quality or an intrinsic accidental
denomination (Mates 1986, 223).
A denomination is a concept that is expressed by a description. Take the
sentence 'Socrates is shorter than Theatetus'. The description 'shorter than
Theatetus' expresses a denomination, or a concept, under which the
individual Socrates falls if the sentence is true. Because the description in
this sentence makes reference to an individual other than Socrates, it
expresses an extrinsic denomination. In the sentence 'Socrates is wise', the
denomination of wisdom, expressed by the predicate 'wise' is an intrinsic
denomination because it makes reference to no individual other than
Socrates. An intrinsic denomination is a quality or accident that inheres in
an individual substance. For Leibniz,
there are no purely extrinsic denominations which have no basis at all in the
denominated thing itself. For the concept of the denominated subject
necessarily involves the concept of the predicate. Likewise, whenever
the denomination of a thing is changed, some variation has to occur in
the thing itself. (PPL 268)
Thus all relational propositions must be reduced to categorical propositions. But as has been show above, Leibniz is unable to reduce
asymmetrical relational propositions to categorical propositions, or even
a set of categorical propositions, that are logically equivalent to the
proposition being reduced. And, as was shown in the Introduction to this
book, Kant uses the irreducibility of relations as an argument in favour of
transcendental idealism:
161
In confirmation of this theory of the ideality of both outer and inner

sense, and therefore of all objects of the senses, as mere appearances, it is
especially relevant to observe that everything in our cognition which
belongs to intuition . . . contains nothing but mere relations . . . Now a
thing in itself cannot be known through mere relations; and we may
therefore conclude that since outer sense gives us nothing but mere
relations, this sense can contain in its representation only the relations of
an object to the subject, and not the inner properties of the object in
itself (B66-7)
Since a phenomenal substance consists of matter, it occupies space and
interacts with other phenomenal substances. Hence the matter of
phenomenal substance can be apprehended only through outer sense.
However, since an object of the pure understanding has only an inner
simple nature it has no relationship whatsoever to any outer objects.
Because objects of the pure understanding have only internal properties
and do not interact with other objects in any way, they are not real:
The absolutely inward [nature] of matter, as it would have to be
conceived by pure understanding, is nothing but a phantom; for matter
is not among the objects of pure understanding, and the transcendental
object which may be the ground of this appearance that we call matter
is a mere something of which we should not understand what it is, even
if someone were in a position to tell us. For we can understand only that
which brings with it, in intuition, something corresponding to our
words. (A277/B333)
Kant treats Leibniz's monads as examples of things-in-themselves or
noumena, transcendental objects that fall outside the conditions of space and
time. But since, at least from Kant's perspective, matter or conceptual
content can only be supplied to human understanding through sensible
intuition it is impossible to determine what, if any, matter a transcendental
object might possess. And since the matter that is supplied to the
understanding through sensible intuition is the only means that the
understanding has for acquiring conceptual content, things-in-themselves,
at least for human beings, can have no conceptual content.
This points to an important difference between Kant's conception of
substance as an underlying substratum and the traditional conception of
substance as an independently existing object. Both Leibniz and Locke, in
their own ways, agree with Aristotle that substances are independently
162
existing objects. Locke invokes a distinction between primary qualities that

act as the substratum of a substance and secondary qualities, the sensible
attributes through which the substance is experienced. Because we can
only experience a substance through its secondary properties, Locke is
sceptical of our ability ever to achieve a clear understanding of the primary
qualities: those intrinsic qualities that constitute the essence of a
phenomenal substance. Substance for Locke is an underlying substratum
consisting of primary qualities in which secondary qualities inhere. Since
we cannot perceive the underlying substratum, but only the secondary
qualities that inhere in it, Locke believed that we can have no real
knowledge of the primary qualities that make up a substance. Leibniz, in
contrast, believed that we could achieve a clear and distinct understanding
of the intrinsic properties of a substance through the exercise of pure
reason. Leibniz embraces Aristotle's idea of substantial forms that serve as
the essence, motivating force, and the cause of the being of phenomenal
substances. Simple substances cannot be perceived but they can be
understood through conceptual thought.
While Kant follows the tradition in thinking of a substance as a paradigm
example of an object, he rejects the conception of substances as independently
existing objects. So Kant wants to keep the notion of substance as a core
substratum around which all of the attributes that compose an object are
ordered, while rejecting the conception of substance as an independently
existing object. For Kant, substance is a schematized category through which
determinate objects are produced by the understanding through the synthesis
of appearances provided by sensible intuition.
The difference between Kant's new conception of substance and the
traditional conception is illuminated by Kant's conception of an object
and by his conception of existence. Let us examine each in turn.
Kant on Objects
We have stated above that appearances are themselves nothing but
sensible representations, which, as such and in themselves, must not be
taken as objects capable of existing outside our power of representation.
What, then, is to be understood when we speak of an object
corresponding to, and consequently also distinct from, the cognition?
It is easily seen that this object must be thought only as something in
general = x, since outside of our cognition we have nothing which we
could set over against this cognition as corresponding to it. (A 104)
163
Kant turns to his account of appearances to explain what an object of

representations is. Earlier it was shown that he characterized an
appearance as an object whose apprehension is not yet determined by
concepts but is immediately present to the understanding through
sensation. An appearance, then, is the indeterminate object of a sensible
intuition. As such it cannot exist outside our capacity to represent objects.
This raises the question of what the nature of this appearance is that
corresponds to but is distinct from the sensible representation. Although in
other contexts Kant uses the locution 'transcendental object = #' to speak
of a thing in itself (see for example A109), this object is not a thing in itself
because if it existed outside of our cognition there would be nothing for the
appearance to correspond to. Kant labels this object 'something in general
= x\ an object lacking any descriptive content serving as a general spatiotemporal placeholder for anything that corresponds to a sensible intuition.
Kant, however, sheds no further light on the issue in this passage. Instead
he turns to an explication of a more robust conception of an object:
Now we find that our thought of the relation of all cognition to its object
carries with it an element of necessity; the object is viewed as that which
prevents our modes of cognition from being haphazard or arbitrary,
and which determines them a priori in some definite fashion. For in so far
as they are to relate to an object, they must necessarily agree with one
another, that is, must possess that unity which constitutes the concept of
an object . . . It is only when we have thus produced synthetic unity in
the manifold of intuition that we are in a position to say that we cognize
the object. (A 104-5)
The question that the transcendental deduction seeks to answer is whether
empirical cognition of an object necessarily conforms to the categories.
Kant seeks to establish that the cognition of an object of experience
requires not only that an appearance be given through sensible intuition
but that the appearance be thought under the concept of an object. Thus
my perception of the computer in front of me is a product of manifold
sensible appearances being brought under a concept of an object, in this
case the concept of a computer. While the concept of a computer is an
empirical concept of an object, such a concept ultimately presupposes the
existence of pure concepts of the understanding that lie a priori in the mind.
As Kant explains, the categories 'are concepts of an object in general, by
means of which the intuition of an object is regarded as determined in
respect of one of the logical functions of judgment' (B128). In synthetic
164
judgements concepts of an object are applied to appearances provided

through intuition to produce objects of experience. It is objects of
experience that we perceive and interact with in everyday existence and in
scientific investigations, not things in themselves. The goal of the
transcendental deduction, then, is to show that the existence of objects
of experience presupposes the existence of the categories or pure concepts
of the understanding that lie a priori in the mind as the necessary conditions
for the possibility of experience.
However, Kant is quick to recognize that such a conception of an object
threatens the very existence of objects as the matter for intuition,
that we can therefore have no cognition of any object as a thing in itself,
but only in so far as it is an object of sensible intuition, that is an
appearance all this is proved in the analytical part of the Critique.
Thus it does indeed follow that all possible speculative cognition of
reason is limited to mere objects of experience. But our further contention
must also be duly borne in mind, namely, that though we cannot cognize
these objects as things in themselves, we must yet be in a position at
least to think them as things in themselves; otherwise we should be
landed in the absurd conclusion that there can be appearance without
anything that appears. (Bxxvixxvii)
While we cannot cognize things in themselves we must be able to claim
that such objects, removed though they may be from the sensible
conditions of space and time, are at least thinkable. For if reason could
not interact with such objects in any way, then Kant would be committed
to the claim 'that there could be appearances without anything that
appears', and his philosophy would be reduced to a radical form of
idealism where all that we can claim to exist are appearances.
Kant, then, has at least three conceptions of an object. The first
conception of an object is an appearance. An appearance is the
indeterminate object of a sensible intuition. It is the pre-conceptual
spatio-temporal content provided to the understanding by intuition before
it is gathered up and thought through the understanding. The second
conception of an object involves a concept under which a manifold of
intuitions is brought to a unity. Such an object is a product of both
sensibility and understanding. Sensibility provides the understanding with
appearances that conform to the pure conditions of space and time. These
appearances are synthesized by the imagination and the understanding to
produce concepts of objects. However, Kant also recognizes that if his
165
critical turn is to avoid lapsing into absurdity, he must give some account
of how the faculty of sensibility is supplied with content that is distinct
from the sensible conditions of space and time. He thus must admit that
there are also objects in the latter sense that he calls things-in-themselves or
transcendental objects. The third kind of object, because it is completely
independent of the conditions that make cognition of an object possible,
can only be thought by reason; it cannot be known.
Kant on Existence
The difference in ontological status that Kant's conception of substance
has in relation to the traditional conception of substance is also illustrated
by his conception of existence. As he explains in the Critique:
'Being' is obviously not a real predicate; that is, it is not a concept of
something which could be added to the concept of a thing. It is merely
the positing of a thing, or of certain determinations, as existing in
themselves . . . The small word 'is' adds no new predicate, but only
serves to posit the predicate in its relation to the subject. (A598/B626)
According to Kant, existence is not a real predicate because it is not a
property that can add anything to my concept of a thing. If I say 'That ball
is red' I am saying something determinate about the object in front of me by
bringing it under a concept. But if I say 'That ball is' I add nothing to my
concept of the ball. I merely state that the bundle of properties whose
combination makes up my concept of the ball is not merely possible but
actual. Existence is a predicate that when applied to the bundle of properties
that make up a possible object of experience signifies that the object is actual.
The relationship between existence and actuality in Kant's philosophy
can be illuminated by an examination of the categories of modality:
The categories of modality have the peculiarity that, in determining an
object, they do not in the least enlarge the concept to which they are
attached as predicates. They only express the relation of the concepts to
the faculty of cognition. Even when the concept of a thing is quite
complete, I can still enquire whether this object is merely possible or is
also actual, or if actual, whether it is not also necessary. No additional
determinations are thereby thought in the object itself; the question is
only how the object, together with all its determinations, is related to
166
understanding and its empirical employment, to empirical judgment,

and to reason in its application to experience. (A219/B266)
The modality of a judgement has nothing to do with the relationship
between its subject and predicate concepts but with the relationship
between the object, with all of its determinations, and the empirical
employment of the understanding. The form of judgement S is actually P
expresses a certain relationship between a proposition and the person who
is thinking the proposition. If he thinks the proposition as actual, then the
copula has the value of actuality.
Charles Parsons observes that Kant's conception of actual existence
shows similarities to the modern notion of an existential quantifier:
By actuality, Kant means actual existence, and it therefore has the logical
properties which, for existence per se, he appeals to in his famous criticism
of proofs of the existence of God. It is not a property of objects; we could
reconstruct it as what is expressed by the existential quantifier in empirical
judgments ... In our modern logic, where we readily give the existential
quantifier a more general sense, it is natural to render Kant's actuality as
a restricted quantifier, so that 'Fs are actual' in Kant's sense would be
rendered [(3x) (x is actual & F(x))]. (Parsons 1982, 494-5)2
Frege on Existence
Parsons sees Kant's claim that existence is not a real predicate as a
harbinger of the modern notion of an existential quantifier. Since it was
Frege who first developed modern quantification theory, it should come as
no surprise that he has a conception of existence very similar to Kant's. As
he explains in 'On Concept and Object':
I have called existence a property of a concept. How I mean this is to be
taken is best made clear by an example. In the sentence 'There is at
least one square root of 4', we are saying something, not about (say) the
definite number 2, nor about - 2, but about a concept, square root of 4;
viz. That it is not empty. (FR 189)
Frege had already developed this view of existence when he wrote the
Foundations. Thus in a discussion of whether numbers are derived by
abstraction from things, Frege criticizes those who make
167
the mistake of supposing that a concept can only be acquired by direct

abstraction from a number of objects. We can, on the contrary, arrive at
a concept equally well by starting from defining characteristics; and in
such a case it is possible for nothing to fall under it. If this did happen,
we should never be able to deny existence, and so the assertion of
existence would lose all content. (FA 62)
For both Leibniz and Kant, the paradigm process for acquiring concepts
is a process of abstraction. By abstracting away from the differences
between a certain class of objects and concentrating on certain similarities
one can arrive at a concept under which all of the objects fall. Frege does
not claim that this is necessarily an illegitimate way of acquiring a
concept, merely that it is not the only way. If one starts not from an object
but from the content of a judgement, then concepts can be acquired
through an analysis of this content. One advantage that Frege's method of
acquiring concepts has over the method of abstraction is that in the
former we can define a concept under which no object falls while in the
latter we cannot: something that will come in handy for Frege when he
seeks to define the number zero. And once he has defined the number
zero, he can use his definition of following in a series to define the rest of
the numbers. For Frege, existence is not a property of an object but a
property of a concept.
Kant and Frege, then, agree that existence is not a first-level concept.
Where they disagree is over the ontological status of the concept of
existence. For Kant, the concept of existence is one of the categories of
modality. Existence is conferred on an object, together with all of its
determinations, by an act of judgement in which a possible object of
experience is deemed to be actual. Like the other categories, it makes no
sense to think of the concept of existence as something that a subject can
have independently of the application of the categories to the appearances
provided to the understanding through sensible intuition.
Like all of the other basic components of his philosophical outlook,
existence is a logical notion for Frege. If an object falls under a concept,
then it exists. To appreciate this point, let us take a brief look at Frege's
later discussion of the nature of and relationship between concepts and
objects. In 'On Concept and Object', Frege observes that:
The word 'concept' is used in various ways; its sense is sometimes
psychological, sometimes logical, and sometimes perhaps a confused
mixture of both. Since this licence exists, it is natural to restrict it by
168
requiring that when once a usage is adopted it shall be maintained.

What I decided was to keep strictly to a purely logical use. (FR 181)
In 'On Function and Concepts', Frege characterizes a concept as a
function whose value is a truth-value: 'We thus see how closely that which
is called a concept in logic is connected with what we call a function.
Indeed, we may say at once: a concept is a function whose value is always a
truth-value' (FR 139). On the next page he elucidates his notion of an
object as anything that is not a function:
When we have just admitted objects without restriction as arguments
and values of functions, the question arises what it is that we are here
calling an object. I regard a regular definition as impossible, since we
have here something too simple to admit of logical analysis. Here I can
only say briefly: an object is not a function, so that an expression for it
does not contain any empty place. (FR 140)
Frege defines both concept and object, then, in terms of the notion of a
function. A concept is a function whose value is a truth-value. An object is
anything that is not a function, anything that does not contain an empty
place.
The best way to understand the role that existence plays in Frege's
philosophy is to look at the role it plays in his logic. From 'Julius Caesar is
a dictator', we can form the concept 6x is a dictator' by replacing 'Julius
Caesar' with the variable V. We can then quantify over this variable to
produce the proposition that there exists a dictator. A proposition not
about an individual object but about any object that falls under the
concept x is a dictator.
We can thus see Frege's influence on Quine's dictum that to be is to be
the value of a bound variable:
We can very easily involve ourselves in ontological commitments, by
saying, e.g., that there is something (bound variable) which red houses
and sunsets have in common; or that there is something which is the prime
number between 1000 and 1010. But this is essentially the only way we
can involve ourselves in ontological commitments: by our use of bound
variables . . . To be is, purely and simply, to be the value of a variable.
(Quine 1952, 199-200)
169
Two Views on Objects and Existence

For Frege, then, existence is a second-level concept. An object exists if it
falls under a concept. Existence is imposed, we might say, from above
through a concept. For Aristotle in contrast, to be in the primary sense is to
be a substance, where the existence of all other entities depends upon their
being predicated of a substance.
As Aristotle observes in the Metaphysics, there are several senses in which
something can be said to be but there is one sense that is primary and that
is substance. For:
All other things are said to be because they are, some of them, quantities
of that which is in the primary sense, others qualities of it, others
affections of it, and others some other determination of it. And so one
might raise the question whether 'to walk' and 'to be healthy' and 'to
sit' signify in each case something that is, and similarly in any other case
of this sort; for none of them is either self-subsistent or capable of being
separated from substance, but rather, if anything, it is that which walks
or is seated or is healthy that is an existent thing. Now these are seen to
be more real because there is something definite which underlies them;
and this is the substance or individual, which is implied in such a
predicate; for 'good' or 'sitting' are not used without this. Clearly then it
is in virtue of this category that each of the others is. Therefore that
which is primarily and is simply (not is something) must be substance.
(Metaphysics 1028a 18-31, in Aristotle 1984)
Substance in its primary sense is self-subsistent because while all other
things are said to be in virtue of their inherence in substance, the existence
of substance does not depend upon the existence of any of the other
categories. To call something healthy is ultimately to call a substance
healthy. If there were not substances, then there would not be anything to
be healthy or unhealthy, and hence being healthy cannot exist without the
existence of substance.
For Kant, existence is not a real predicate. To say that an object exists
does not add anything to its descriptive content: it merely attributes actuality
to a possible object of experience. Frege follows Kant in this regard, treating
existence as a second-level predicate that is predicated of concepts. Now
Kant's claim that existence is not a real predicate is tied up with his rejection
of Aristotle's conception of substance as a self-subsistent object existing
independently of the conditions of sensibility and understanding. Frege
170
follows Kant and treats existence as a second-level concept. Now Frege

certainly believed that some objects, such as numbers, exist independently of
the conditions of sensibility. Did he also think that they exist independently of
the conditions of the understanding? Or, in rejecting the traditional
substance-based conception of an object, is Frege also rejecting the
conception of an object as an independently existing entity?
Michael Ayers, as shown above, thinks so: 'For Frege, all unity is conceptrelative: nothing in the physical world is individual, or a whole but thinking
makes it so' (Ayers 1991, 118). Ayers makes this criticism in the context of
Frege's claim 'that the content of an ascription of number consists in
predicating something of a concept' (FA 67). Ayers cites Frege's claim that
the same object can be called either a company or 500 men, to show that
Frege does not give the existence of men any priority over the existence of a
company. For a substance-based metaphysician, since substances are basic
and a man is a substance whereas a company is not, the ontological status of
a man is more basic than that of a company of men. For the existence of a
company of men depends upon the existence of men, but the existence of men
does not depend upon the existence of companies. Because Frege imposes
existence from above through concepts there is no room in his philosophy for
a conception of an object as something that exists independently of concepts.
Dummett, in a passage cited earlier, seems to express a similar view of Frege:
'Thus, in a certain sense ... for Frege, the world does not come to us
articulated in any way; it is we who, by the use of our language (or by
grasping the thoughts expressed in that language), impose a structure on it'
(FPL 503-4). As Ayers explains:
Kant was perhaps the first philosopher to hold that all our distinctions
among objects are imposed or ideal, none are given or real. He held that
even space and time are mind-dependent forms of our sensibility, and
that we necessarily interpret the data of sense in terms of a set of
categories which constitute the form of our capacity for unitary
experience. In demonstrating what these ineluctable categories are, he
took himself to be elucidating a conceptual structure which is in us a
priori rather than a structure due to external reality. (Ayers 1991, 4-5)
Ayers later invokes a distinction between realism and conceptualism to
illustrate the difference between Kant's conception of substance as an
object that is constructed through the imposition of the categories on the
appearances provided to the understanding through sensible intuition and
a conception of substance as an independently existing object. In rejecting
171
the notion of substance as an independently existing object and claiming

that the number of objects is relative to whatever concept the objects fall
under, Frege too departs from the traditional realist conception of an
object as an independently existing object and should be classified along
with Kant as a conceptualist who believes that what constitutes an object
is always concept-relative.
It is my contention that Ayers' interpretation of Frege is inaccurate
both on textual grounds and upon historical grounds. On the first point,
there are a number of passages in Frege's work that are at odds with such
an interpretation. An examination of these passages will point the way to
an alternative interpretation of the passages cited by Ayers and Dummett
in defence of their views. On the second point, Frege and Leibniz share a
common conception of the nature and score of a priori knowledge that
stands in sharp contrast to Kant's conception. It will be argued that Frege
is heir to the great rationalist tradition inherited largely from Leibniz.
Aspects of Kant's thought can certainly be called rationalist, but the spirit
of his critical turn was to limit the scope of reason beyond the bounds that
Leibniz thought it could be stretched. Frege's outlook is much more in line
with Leibniz's views on the scope of reason, and shows no sympathy for
Kant's attempt to limit the scope of reason to objects of possible
experience. If a primary aim of idealism is to impose limits on the scope
of reason, then Frege is not an idealist even if he does claim that what is
objective is not independent of reason. Let us begin by looking at Leibniz,
Kant and Frege on the a priori.
Notes
1
Heck replaces 'England' with 'Caesar' and 'direction of the earth's

axis' with 'number zero' in the passage from FA 78 that he cites in the
quotation.
I have slightly revised Parsons' symbolism here.
Leibniz, Kant and Frege on the A Priori
Leibniz on the A Priori

In a letter to Queen Sophie Charlotte of Prussia, Leibniz distinguishes
between sensible, imaginable and intelligible objects. Sensible objects are
known through sense perception, imaginable objects through the imagination
and intelligible objects through the understanding. Sensation and imagination are unreliable partners in the search for truth. For our senses can deceive
us and we can imagine a myriad of things that are not true. It is only through
the exercise of the understanding that we discover 'what it is to affirm, deny,
doubt, will, and act. But above all, we find there the force of the conclusions of
reasoning, which are part of what is called the natural light' (PE 189). Because
logic follows the natural light of reason, it is not subject to the errors that
sensation and imagination are prone to. Through the use of logical reasoning
the understanding can demonstrate which conclusions follow from which
premises and which do not. The subject-matter of logical reasoning consists of
intelligible objects that can be discovered through the exercise of the
understanding according to the dictates of the natural light of reason. The
same holds true for the subject-matter of mathematical reasoning:
It is also by this natural light that the axioms of mathematics are
recognized, for example, that if we take away the same quantity from
two equal things, the things remaining are equal; similarly, that if
everything is equal on both sides of a balance, neither side will incline
a thing we can easily predict without ever having experienced it. And it
is on such foundations that we establish arithmetic, geometry,
mechanics, and other demonstrative sciences, where the senses are
indeed necessary for having certain ideas of sensible things, and
experience is necessary for establishing certain facts, and even useful for
verifying our reasonings as by a kind of proof. But the force of the
demonstrations depends upon intelligible notions and truths, which
alone are capable of allowing us to judge what is necessary. (PE 189)
173
For Leibniz, the intelligible objects that are apprehended through the light
of nature are innate ideas, and since the objects of mathematics are
intelligible objects, they are innate ideas. As he explains in New Essays on
Human Understanding:
the whole of arithmetic and of geometry should be regarded as innate,
and contained within us in an implicit way, so that we can find them
within ourselves by attending carefully and methodically to what is
already in our minds, without employing any truth learned through
experience or through being handed on by other people. (NE 77)
Leibniz contrasts pure ideas from images of the senses and necessary truths
or truths of reason with truths of fact. While he acknowledges that abstract
thought requires the use of sensible symbols, he still insists that since
experience cannot account for the necessity of the truths of reason it cannot
account for our ability to grasp innate ideas. Instead, the ground of
necessity lies within us. 'On any view of the matter', he insists, 'it is always
manifest in every state of the soul that necessary truths are innate, and that
they are proved by what lies within, and cannot be established by
experience as truths of fact are' (NE 79).
In contrast to John Locke's belief that all of our ideas come ultimately
from the senses, Leibniz claims that 'all the thoughts and actions of our
soul come from its own depths and could not be given to it by the senses'
(NE 74). Leibniz agrees that the outer senses 'can be said to be, in a
certain sense, partial causes of our thoughts', but he nevertheless insists
that 'there are ideas and principles which do not reach us through the
senses, and which we find in ourselves without having formed them,
though the senses bring them to our awareness' (NE 74). He insists that the
certainty of innate principles does not rest on universal consent but instead
on reflection upon what is certain within us:
I conclude that a principle's being rather generally accepted among
men is a sign, not a demonstration, that it is innate; and that the way
for these principles to be rigorously and conclusively proved is by its
being shown that their certainty comes only from what is within us'
(NE 74).
The objects of mathematics, then, according to Leibniz, are intelligible
objects apprehended by the understanding through the exercise of the
natural light of reason. Since all intelligible objects are innate ideas, and
174
since the objects of mathematics are intelligible objects, the objects of

mathematics are innate ideas.
Leibniz distinguishes between primary truths and derivative truths.
Examples of primary truths are such propositions as 6A is A' or 6A is not notA\ According to Leibniz, 'all remaining truths are reduced to primary
truths with the help of definitions, that is, through the resolution of notions;
in this consists a priori proof] proof independent of experience' (PE 3).
Leibniz claims that primary truths are truths that we are immediately
aware of. He distinguishes between two types of primary truth: truths of fact
and truths of reason. Truths of fact are called first experiences. They are a
posteriori and our cognition of them is immediate in that nothing stands
between the understanding and the object of perception. Truths of reason are
called first illuminations. They are a priori and our cognition of them is
immediate because nothing stands between the subject and the predicate of
the proposition (see NE 434). In grasping a truth of reason it is immediately
obvious that the predicate concept is contained in the subject concept.
Whether a derivative truth is a priori or a posteriori depends upon how it
is justified. If the justification of a truth depends upon the experience of a
particular fact, then the truth is a posteriori. If the truth of a proposition is
proved by reducing it, with the help of definitions, to an explicit identity,
without any intrinsic reliance upon experience, then the truth is a priori. As
Leibniz explains:
The primary truths which we know by 'intuition' are of two sorts, as are
the derivative ones. They are either truths of reason or truths of fact.
Truths of reason are necessary, and those of fact are contingent. The
primary truths of reason are the ones to which I give the general name
'identities' because they seem to repeat the same thing without telling us
anything. (NE 361)
In response to Locke's claim that we are more familiar with 'particular
truths and notions' than with 'general and abstract ideas', Leibniz
acknowledges that we become aware of particular truths before we become
aware of general ones. 'But', he insists, 'that doesn't alter the fact that in
the order of nature the simplest ones come first, and that the reasons for
particular truths rest wholly on the more general ones of which they are
mere instances' (NE 83).
For Leibniz, then, an a priori truth is a truth of reason that is necessary
and grounded in general principles. It is justified independently from
175
experience, through reduction and definition to explicit identities. Such

truths are innate and are grasped not through experience but through the
exercise of the natural light of reason.
Kant on the A Priori

Kant introduces the distinction between a priori and a posteriori cognition in
the opening pages of the Critique:
In what follows, therefore, we shall understand by a priori cognition, not
cognition independent of this or that experience, but cognition absolutely
independent of all experience. Opposed to it is empirical cognition, which
is cognition possible only a posteriori, that is, through experience. A priori
modes of cognition are entitled pure when there is no admixture of
anything empirical. Thus, for instance, the proposition 'every alteration
has a cause', while an a priori proposition, is not a pure proposition, because
alteration is a concept which can be derived only from experience. (B23)
It is important to understand just what Kant is and is not claiming when
he says that a priori cognition is 'absolutely independent of all experience'.
For while Kant does claim that not all cognition is derived from experience
(i.e. some modes of cognition are a priori), he also claims that all cognition
arises from experience. A priori cognition is possible because the human
mind possesses certain a priori modes of cognition: the a priori forms of space
and time and the pure concepts of the understanding. Unlike Leibniz, who
believed that things in themselves existing independently of all human
experience could be grasped through a priori reasoning, Kant believed that
the a priori forms of the understanding and sensibility are only applicable to
appearances and never to things in themselves. These forms can be
supplied with objects only by experience. As Kant explains: 'Pure a priori
concepts . . . cannot indeed contain anything empirical; yet, none-the-less,
they can serve solely as a priori conditions of possible experience. Upon this
ground only can their objective reality rest' (A95/B129). While pure a
priori concepts contain nothing empirical, their only legitimate use is to act
as conditions of possible experience. The objective reality of a pure a priori
concept rests upon its capacity to be applied to experience.
What distinguishes a priori cognition from a posteriori cognition is that
the former possesses necessity and strict universality, whereas the latter
does not. As Kant explains:
176
What we require is a criterion by which to distinguish with certainty

between pure and empirical cognition. Experience teaches us that a
thing is so and so, but not that it cannot be otherwise. First, then if we
have a proposition which in being thought is thought as necessary, it is an
a priori judgment. Secondly, experience never confers on its judgments
true or strict, but only assumed and comparative universality, through
induction. We can properly only say, therefore, that, so far as we have
hitherto observed, there is no exception to this or that rule. If, then, a
judgment is thought with strict universality, that is, in such a manner
that no exception is allowed as possible, it is not derived from
experience, but is valid absolutely a priori. Empirical universality is only
an arbitrary extension of a validity holding in most cases to one which
holds in all, for instance in the proposition 'all bodies are heavy'. When,
on the other hand, strict universality is essential to a judgment, this
indicates a special source of cognition, namely, a faculty of a priori
cognition. Necessity and strict universality are thus sure criteria of a
priori cognition, and are inseparable from one another. (B34)
The modality of a judgement, for Kant, consists of a relationship between
a judgement and the faculty of cognition. What makes a judgement
problematic, assertoric or necessary for Kant is not the content of the
judgement but the intentional attitude that the person making the
judgement has toward the content of the judgement. Since one of the
criteria of the a priori is necessity, and since what confers necessity on a
judgement is the intentional stance that a person making the judgement
has toward its content, there is an inextricable intentional element to
Kant's conception of the a priori.
Much of the best work on Kant in the second half of the twentieth
century has sought to separate the logical and the epistemological aspects in
his thought from psychological aspects. Analytic philosophers distinguish
these issues in the way they do thanks in large part to Frege's achievement
and influence. It was Frege, after all, who taught philosophers to make a
sharp distinction between the logical and the psychological, and the
subjective and the objective. But as Longuenesse warns, any attempt to
account for the aims of the transcendental deduction or to understand the
transcendental analytic as a whole while excluding any account of the
'psychological' or 'mental' dimensions in Kant's philosophy cannot
adequately capture his intentions in the Critique. 'Indeed', she says,
both in the Transcendental Deduction and in the Analytic of Principles,
177
which the Deduction is meant to ground, Kant's argument for the

applicability of categories to objects rests on the relation he tries to
establish between discursive syntheses or combinations (combinations of
concepts in judgments) on the one hand, and syntheses or combinations
of our sensible perceptions on the other. Such an argument is undeniably
'mentalist' or 'psychological' even though Kant's procedure is quite the
opposite of an introspective procedure, and even though psychological
hypotheses are always guided by a logical analysis of the conditions of
the truth or falsity of our judgments. (Longuenesse 1998, 6)
An accurate understanding of the relationship between Kant and Frege
requires that we keep in mind the different roles that psychological states,
activities and processes play in their respective accounts of the form and
content of judgements. Frege's reaction against Kant was in a large part
connected with differences over the question of whether an adequate
account of the conceptual content and unity of a judgement requires the
invocation of mentalistic states or processes. Hence the true nature of their
disagreement cannot be appreciated without a proper understanding of the
role that psychological states, activities and processes play in Kant's work.
Frege on the A Priori

Many of Frege's characteristic views are deeply connected to his insistence
that psychological states, processes and acts play no essential role in the
science of logic or arithmetic. His firm stance against psychologism thus
puts him in direct conflict with Kant. While logic, for Kant, is not
grounded in principles of empirical psychology, it is grounded in the
existence and exercise of mental faculties.
According to Frege, once the question of a truth of a judgement has
been 'removed from the sphere of psychology' (FA 3), the question of how
or when the content of the truth was discovered is kept separate from the
question of what justifies its truth:
The problem becomes, in fact, that of finding the proof of the
proposition, and of following it up right back to the primitive truths. If,
in carrying out this process, we come only on general logical laws and
on definitions, then the truth is an analytic one, bearing in mind that we
must take account also of all propositions upon which the admissibility
of any of the definitions depends. If, however, it is impossible to give a
178
proof without making use of truths which are not of a general logical
nature, but belong to the sphere of some special science, then the
proposition is a synthetic one. For a truth to be a posteriori, it must be
impossible to construct a proof of it without an appeal to facts, i.e., to
truths which cannot be proved and are not general, since they contain
assertions about particular objects. But if, on the contrary, its proof can
be derived exclusively from general laws, which themselves neither need
nor admit of proof, then the truth is a priori. (FA 4)
To call a proposition analytic, say, or a posteriori, according to Frege, has
nothing to do with physiological, psychological or physical conditions that
make the judgement possible, but rather with 'the ultimate ground upon
which rests the justification for holding it to be true' (FA 3). He stresses
that viewing the distinctions in this way removes the question about the
justification of a judgement from the realm of psychology. To justify the
truth of a proposition it is inappropriate to talk about when or how the
truth was discovered. What is necessary is that we are able to follow the
proof of the proposition back to the primitive truths upon which it
depends. If the truth can be derived from general laws that neither need
nor admit of proof, then it is a priori.
For Kant, a priori cognition is cognition that is independent from
experience. The marks of the a priori are necessity and strict universality.
What makes a judgement necessary for Kant is not the content of the
judgement but the intentional attitude that the person making the
judgement has toward the content of the judgement. While Kant calls a
wide variety of things a priori, the basis of the a priori lies in faculties and
capacities of the human mind.
This outlook is grounded in Kant's belief that the only way to account
for the possibility of a priori cognition is to start with the premise that it is
the representation that an object must conform to:
There are only two possible ways in which synthetic representations and
their objects can establish connection, obtain necessary relation to one
another, and, as it were, meet one another. Either the object alone must
make the representation possible, or the representation alone must make
the object possible. In the former case, this relation is only empirical, and
the representation is never possible a priori. (A92/B124-5)
You cannot answer the question of how a priori cognition is possible if you
start from the premise that the object alone can make cognition possible. For
179
this relationship between object and representation is empirical and thus

does not have the requisite necessity to ground an a priori cognition. The
question can be answered only when we accept that objects must conform to
our representations. Judgements, propositions, concepts and intuitions can
all be a priori for Kant. But ultimately everything that is a priori is so because
it is a product of sensibility or understanding, the two fundamental faculties
of the mind. The epistemological role of the a priori in Kant is conditioned on
his belief that objects must conform to the subjective a priori conditions of
sensibility and understanding before they can be experienced.
Despite Frege's claim that in drawing the distinction between a priori
and a posteriorijudgements he is not assigning a new sense to the terms but
merely making more accurate what Kant meant by them, it is clear that he
is parting ways with Kant on a number of key issues. For Kant, one of the
two marks of the a priori is necessity. Frege parts ways with both Kant and
Leibniz and avoids characterizing the a priori in terms of necessity. Here he
agrees with Kant that to call a judgement necessary is simply to state what
your intentional attitude is to the judgement. As he explains in the
Begriffsschrift, to call a proposition necessary is merely to 'give a hint about
my grounds for judgment' (CN 114). For Frege, it is generality, not
necessity, that is the mark of the a priori. In this respect he breaks with both
Leibniz and Kant. For while generality is a mark of the a priori for Leibniz
and strict universality for Kant, it is not the only mark. For both, necessity
is an essential mark of the a priori. As Tyler Burge explains:
Neither Kant nor Leibniz gives any hint of defining a priori in terms of
generality. Both appeal, however, to generality in their elucidations of a
priori. Frege's use of generality (Allgemeinheit) in his definition is surely
inherited from them. Like them he believed that a priori is deeply
connected with some form of generality of application, or universal
validity. But he interpreted and used his notion of generality differently.
He departs from both Leibniz and Kant in defining a priority in terms of
generality. He departs from both in saying little about the relation
between a priori and necessity. Indeed his conception of generality
differs from both in that he does not connect it to modal notions, seen as
independent notions, at all. (Burge 2005, 368-9)
For Frege and Leibniz, a truth is a priori if its justification depends only
upon general logical laws that neither need nor admit of truth. While a
priori truths for Leibniz are innate and dependent upon the existence of
minds, he claims that they ultimately depend for the existence upon the
180
mind of God and not human beings. Frege declines to follow Leibniz's
characterization of a priori truths in terms of innate ideas. And while he
does not claim that a priori truths exist in the mind of God, he does think
their existence in no way depends upon the minds of human beings.
To appreciate the extent to which Frege's philosophical outlook is similar
to Leibniz's, and the extent to which it is different, let us take one last look at
Leibniz's claim that relations are ideal. According to Leibniz, an accident is
peculiar to the substance that it belongs to. While both Plato and Socrates
fall under the concept of wisdom, the wisdom that inheres in Socrates is
different from the wisdom that inheres in Plato. Hence, while two subjects
can fall under the same concept, the same accident cannot be in two subjects
at once. In the sentences 'David is a father' and 'Solomon is a son', the
attributes of being a father and being a son, designated by the predicates of
the sentences, are real attributes which inhere in David and Solomon
respectively. However, in the sentence that David is the father of Solomon,
the relational expression '... is the father o f ' does not designate a real
relation between David and Solomon. Instead, it is a mental construct
grounded in the attributes of being a father and being a son, which, in turn,
are modifications of David and Solomon. As Leibniz explains:
It is no wonder that the number of numbers, or that of all possibilities, all
reflexions or all relations, are not distinctly understood, for they are ideae
imaginariae, nor does anything correspond to them in reality; so if there is
a relation between A and B and this relation is called C, and a new
relation between A and C is considered, called D, and so on, ad infinitum,
it is clear that we would not say that all these relations are true and real
ideas. For those things only are intelligible which can be produced, that
is, which have been produced or are being produced. (Mates 1986, 222)
In the case of the name 'Socrates' there is something that corresponds to it
in reality: namely, Socrates. The word 'pale' corresponds to the pallor in
Socrates when Socrates is pale and when Socrates is no longer pale, the
particular pallor no longer exists. However, in the case of the sentence
'Theatetus is taller than Socrates', while there is something corresponding
to the words 'Theatetus' and 'Socrates' there is nothing in reality that
corresponds to the relational expression '... is taller than -'. As Leibniz
explains: T believe that qualities are just modifications of substances, and
that the understanding adds relations' (Mates 1986, 223).
However, when Leibniz says that relations are not real but ideal, he
does not mean that they are only subjective ideas that are predicated to an
181
accident of a substance solely through the operations of the human

understanding. 'Relations and orders are some kind of entia rationis\ he
explains, 'although they have their fundament in things; for one can say
that their reality, like that of eternal truths and of possibilities, comes from
the supreme reason' (Mates 1986, 223). Elsewhere he says: 'Relations have
a reality dependent upon the mind, like truths; but not the mind of men,
since there is a supreme intelligence which determines them all for all time'
(Mates 1986, 223). Relations do not exist just in human minds but in the
mind of God as well. As Mates explains, for Leibniz,
reality consists exclusively of individual substances or monads. One of
these monads, God, exists eternally and unchanging, outside the
temporal series. The remainder, including ourselves, exist through all
time, continually undergoing changes that range all the way from the
nearly imperceptible to the relatively drastic. Ideas, concepts, propositions, and so forth, are 'in God's mind', but this does not mean that his
mind is a kind of receptacle in which such entities reside or have some
sort of shadowy existence. It means only that he 'has' the ideas, which in
turn means only that he has the capacity or disposition to think in
certain ways. Thus, by including one very special individual in his
nominalistic ontology, Leibniz achieves some of the advantages
ordinarily thought to flow from Platonism. (Mates 1986, 49-50)
Thus there are two strains in Leibniz's thought. On the one hand, he is a
nominalist who believes that individual substances are the only things that
are fully real and that accidents acquire their existence only through
inherence in an individual substance. Relations, as accidents of accidents,
do not directly inhere in an individual substance but are merely ideal
things added to an accident by the understanding with nothing
corresponding to them in reality. On the other hand, whatever reality
relations do have does not depend merely upon the ability of human minds
to add them to accidents of individual substances. For all relations, like all
truths, exist in the mind of God.
For Aristotle, the paradigm example of an object is a substance, the
individual man or horse. Leibniz follows Aristotle here and thus considers
accidents and relations to be ontologically dependent upon their
connection to a substance. Since an accident depends for its existence
upon its inherence in a substance, it has a lesser degree of reality in relation
to the substance in which it inheres. And since a relation is not attributed
directly to a substance but to an accident of a substance it is two steps
182
removed from reality. Frege rejects, or more accurately ignores, the

substance-based conception of the relationship between a subject and an
attribute when he replaces the subject/predicate conception of judgement
with a conception grounded in the relationship between function (concept)
and argument (object). Thus he feels no compunction to try to explain the
relationship between the subject and predicate of a sentence in terms of an
attribute inhering in a substance. Thus, for Frege, a concept or relation
does not depend upon a substance for its existence.
Frege on Logic and Objectivity
The Objectivity of Numbers

According to Ayers, when Frege claims that we can say with equal truth of
the same external phenomenon that it is either four companies or 500 men,
depending upon what concept we choose to bring it under, he affords no
special ontological status to men over companies. Both are objects that can
be brought under concepts and thus counted. Ayers concludes from this
that what makes something a specific individual is always relative to
whatever concept we choose to bring the external phenomenon under.
Dummett makes a similar claim in declaring that for Frege, 'the world
does not come to us articulated in any way; it is we who, by the use of our
language (or by grasping the thoughts expressed in that language), impose
a structure on it' (FPL 504).
Now Frege certainly acknowledges that in any attempt to capture the
objective structure of truth the use of language is indispensable. But this
does not mean that the objective structure of truth is somehow dependent
upon language. Burge has shown that a proper understanding of Frege
must recognize that he distinguished between the conventional linguistic
meaning of a word and a word's sense. The conventional linguistic
meaning of a word is the meaning that it has in a dictionary, or as
understood by a competent speaker. In contrast, the sense of an expression
is its mode of presentation. As Burge explains:
Frege, in using the term 'sense', was primarily concerned with mode of
presentation, with the objective content of thoughts, rather than with
the meaning of linguistic expressions. The objective thought content
expressed with sentences containing proper names like 'Aristotle' was
regarded by Frege as (normally) publicly accessible, but not purely by
virtue of mastering the language. (Burge 2005, 2212)
For Frege, the sense of a definite description corresponds to the mode of
184
presentation of whatever object falls under the description, and not to the
meaning of the proper name. Frege identified the sense of a name with the
sense of a definite description not because he thought that a definite
description was synonymous with the conventional meaning of the name
but because a definite description gives linguistic expression to the mode of
presentation by which the object designated by the name can be identified.
For Frege, the mode of presentation of the name's bearer is contained in
the sense, and a sense is a mind-independent object. Senses do not depend
upon the existence of a language or of language-users. While human beings
can grasp senses only through the employment of language, senses exist
independently from the meaning of the words that express them. The sense
of a name, while accessible through language, is not dependent on
language. A sense can be grasped through the use of language, but it is not
a product of language. As Frege explains:
we distinguish the sentence as the expression of a thought from the thought
itself. We know we can have various expressions for the same thought. The
connection of a thought with one particular sentence is not a necessary one;
but that a thought of which we are conscious is connected in our mind with
some sentence or other is for us men necessary. But that does not lie in the
nature of the thought but in our own nature. There is no contradiction in
supposing there to exist beings that can grasp the same thought as we do
without the need to clad it in a form that can be perceived by the senses.
But still for us men there is this necessity. (PW 269)
While human beings can grasp a thought only through the use of
language, thoughts exist independently from whatever language is being
used to grasp them. The fact that human beings can grasp thoughts only
through the use of language does not mean that the objectivity of thoughts
is dependent upon the language used to grasp them. The construction of
ever more sophisticated logical languages brings the human mind closer
and closer to the objective structure of truth. But there is a fundamental
difference between the objective structure of truth and whatever logical
language is being used by human beings to grasp such a structure.
This last passage comes from a posthumously published essay written
late in Frege's career. However, the roots of such a view can be found as far
back as the Foundations:
What is known as the history of concepts is really a history either of our
knowledge of concepts or of the meanings of words. Often it is only after
185
immense intellectual effort, which may have continued over centuries,

that humanity at last succeeds in achieving knowledge of a concept in
its pure form, in stripping off the irrelevant accretions which veil it from
the eyes of the mind. (FA vii)
Frege believed that it is crucial to distinguish an account of when a given
concept was first discovered from a justification of the truth of a judgement
containing the concept. It may take centuries for a concept or logical
object to reveal itself in its purity; but once such a revelation occurs, the
history of our discovery of the concept and how our understanding of it
developed over time is, at best, of only secondary concern. It is logic that
tells us how a given truth is justified, not the history of our understanding
of the concepts contained in the judgement. To uncover the true nature of
a concept, the layers of misunderstanding that have accumulated around
our understanding of it over the years must be stripped away. Frege's
concept-script certainly advanced our knowledge of the concept of number
further than the traditional conception of logic that it replaced. But that
does not mean that it has captured the concept of number in its purity, as
Russell's paradox showed years after the publication of the Foundations.
Frege expresses similar sentiments in the posthumously published essay
'Logic in Mathematics', written in 1914:
The progress of the history of the sciences runs counter to the demands
of logic. We must always distinguish between history and system. In
history we have development; a system is static. Systems can be
constructed. But what is once standing must remain, or else the whole
system must be dismantled in order that a new one may be constructed.
Science only comes to fruition in a system. We shall never be able to do
without systems. Only through a system can we achieve complete
clarity and order. No science is in such command of its subject-matter as
mathematics and can work it up into such a perspicuous form; but
perhaps also no science can be so enveloped in obscurity as
mathematics, if it fails to construct a system. (PW 241-2)
The difference between the history of the sciences and the demands of logic
is that the accepted truths of a science can change over time, while logic is
a static system. Systems can be constructed as a science progresses in order
to chart the logical relationships between the currently accepted truths of
that science. However, a significant enough discovery can so change what
is accepted as true in a given science that the old system must be
186
dismantled and replaced with a new one. The construction of each new
system hopefully brings us closer and closer to an understanding of the
science's objects and concepts in their purest form. But each system is just
that: a construction that hopes to capture the concepts and objects of a
science in their purity but invariably falls short due to human limitations.
Frege certainly did not share Aristotle's interest in classifying entities
and explicating their relations to each other in terms of ontological
categories such as substance, quality, quantity, and so on. But that should
come as no great surprise given that he had a much narrower
philosophical goal, which was to define the concept of a number and
explain how numbers are applicable to things. Given this goal, and given
the nature of the new logical language that he had developed to achieve it,
he required an extremely wide conception of what constitutes an object.
Aristotle treats a man as a substance and a number as a quantity, whereas
Frege treats both men and numbers as objects. But this does not mean that
Frege failed to recognize a difference in ontological status between a man
and a number. Indeed, the Julius Caesar problem that ultimately leads to
Frege's rejection of his attempted contextual definition of numbers rests on
his acknowledgement that Julius Caesar is a different kind of entity than
the number 4. The problem with his attempted contextual definition of
number is that it cannot account for this difference. Frege's ontology
distinguishes actual objects from abstract objects instead of distinguishing
(as does Aristotle's ontology) between primary substances, secondary
substances and categories other than substance.
In rejecting Aristotle's substance-based conception of what constitutes an
object, Frege expanded the scope of what constitutes an object to include
numbers and other non-actual entities. But this does not imply that he is
committed to some form of radical conceptual relativism where what
constitutes an object is simply relative to whatever conceptual scheme
someone chooses to impose upon the world. Numbers for Frege are selfsubsistent, mind-independent objects existing in an abstract realm distinct
from whatever linguistic means human beings have at their disposal at any
point in time to access them. Frege did believe it possible to yield different
kinds of objects from the same external phenomenon by bringing the
phenomenon under different concepts. Nevertheless he believed that the
concepts we impose upon the world, as well as the numbers that we use to
number things in the world, are objective and mind-independent.
Conventional scientific meaning (the meaning of a scientific term as
understood by current practitioners of whatever science it is a term of) is a
cousin to conventional linguistic meaning. In both cases the meaning of the
187
expression only approximates the sense that it seeks to express. The distinction
between the conventional scientific meaning of a word and the sense it seeks to
express runs parallel to the distinction between the history of our grasp of a
concept and the concept itself. While a thought can be analysed in a variety of
different ways by whatever language a scientist has at her disposal, the
objectivity or existence of a thought being analysed does not depend upon the
existence of whatever scientific language is being used to analyse it.
Frege's Place in the History of Philosophy

Leibniz thought that by developing an adequate logical calculus, one
could have unprecedented control over the meaning of the concepts
involved in scientific reasoning. For every determinate concept of a given
science there will correspond a precisely defined symbol that characterizes
it. Whenever the truth of a science requires justification one can analyse
the proposition expressing the truth into the basic symbols that compose it.
Since the symbols have been assigned clear and distinct concepts as their
meanings, one can trace the concepts that the symbols designate back,
through a chain of inference, to the basic axioms that its truth ultimately
depends upon. In developing his concept-script Frege aimed for the same
goal, the creation of a concept-script where any significance that a word
might have over and above its logical significance was excluded. As he
explains in the Begriffsschrift:
Now all aspects of language which result only from the interaction of
speaker and listener - for example, when the speaker considers the
listener's expectations and tries to put them on the right track even
before speaking a sentence - have nothing corresponding to them in my
formula language, because here the only thing considered in a
judgement is that which influences its possible consequences. Everything necessary for a correct inference is fully expressed; but what is not
necessary usually is not indicated; nothing is left to guesswork. (CN 113)
However, underlying this shared dream of establishing a universal logic
applicable to the objects and concepts of any science is a different
conception of logic. The logic that formed the starting-point of Leibniz's
attempt to develop a general characteristic was Aristotle's syllogistic
theory. Leibniz believed that by first establishing the simple forms of
reasoning, as exemplified in Aristotle's syllogistic theory, it is possible to
188
advance to the analysis and justification of more complex and difficult

forms of reasoning. In particular, he claims that the truth of arithmetical
propositions, involving operations such as addition, multiplication and
division, can be demonstrated in virtue of their form. Just because an
argument cannot be proved valid through syllogistic logic does not mean it
is invalid. However, Leibniz was never able to free himself from the
influence of Aristotle's syllogistic theory to consider the possibility of
expanding the scope of logic beyond the constraints of the subject/
predicate form of judgement that lies at the basis of traditional logic.
Frege's development of variables, quantifiers and polyadic logic
enabled him logically to capture the relational and multi-general structure
of arithmetical propositions to an extent not possible by the syllogistic logic
embraced by Leibniz and Kant. As Michael Potter explains, in contrasting
Frege's conception of logic with Kant's:
Now the distinctive feature of reasoning that is ampliative rather than
merely explicative is that it depends on objects as its subject matter. If
polyadic logic is ampliative, it must therefore depend on objects in a
way that syllogistic does not. The feature which makes this dependence
manifest is the variable. Variables are what enable logic to handle the
two features that give it its distinctive complexity (and therefore its
power), polyadic predicates and nested quantification. Thus, for
example, the distinct concepts of murder and suicide can both be
extracted from the sentence 'Gassius killed Gassius' by using variables to
differentiate between 6x killed / and 6x killed x\ (Potter 2000, 64)
The quantifier/variable notation that Frege developed is a crucial
component in his attempted definition of number. For such a definition
of number to work requires objects as the values of the variables, and in the
case of numbers Frege needs an infinity of objects.
Frege needed numbers to be objects because in his contextual definition
of numbers using Hume's Principle (there are just as many Fs as Gs) to fix
the sense of an identity statement (number of Fs is the same as the number
of Gs), he needs the cardinality operator 'the number of Fs' to range over
everything. Similarly his later explicit definition in terms of extensions of
concepts requires an infinite number of logical objects for his variables to
range over. As Dummett explains:
It is only because Frege reckoned numbers among objects, that is, as
belonging to the domain of the individual variables, that he was
189
enabled to spin the infinite sequence of natural numbers out of nothing,

as it were. There must be at least 0 objects, and hence the number 0
exists. Since the number 0 exists, there is at least one object, and so the
number 1 exists: and so on indefinitely. It is in order to prove the
infinity of the natural-number sequence that Frege is compelled to
construe numbers as objects . . . (FPM 1323)
Here it should be observed that Frege's conception of logic is importantly
different from modern conceptions where the scope of quantifiers can be
relativized to a specified domain of discourse such as the natural numbers
or presidents of the United States. In contrast, as Leila Haaparanta
observes:
Frege holds a firm belief in the doctrine that his conceptual notation is a
universal language, which speaks about one single world . . . He does
not divide his universe into various sorts or categories. For him, any
object can be the value of an individual variable, and, indeed, any
function must be defined for all objects . . . Accordingly, for Frege there
is one single domain of discourse for all quantifiers, that is, the whole
universe, and it is not possible to confine oneself to considering any
subdomains of the given domain of objects. (Haaparanta 1985, 36)
The principles of arithmetic apply not only to objects that exist in space
and time but to mental processes, events and concepts. The only limit to
the application of the basic laws of arithmetic to any kind of object is that
its concepts be sharp and its objects determinate. For every object there
must be a yes or no answer as to whether it falls under a given concept. As
long as this degree of precision is adhered to, the laws of arithmetic are
applicable to everything thinkable. The class of things that are countable is
as wide as conceptual thought. Because of this, a source of knowledge that
is more restricted in scope, like the geometrical and temporal sources of
knowledge or sense-perception, cannot guarantee the general validity of
arithmetical statements. Only the laws of logic can do that.
So from Frege's perspective, the difference between actual objects, such
as human beings, and logical objects, such as numbers, was less important
than the fact that each was a self-subsistent object that could serve as the
value of a variable and as something that can be quantified and thus
counted. This is what makes the Julius Caesar objection so problematic for
Frege. If Frege did not need the laws of number to be applicable to
everything, he could have solved the Julius Caesar problem by limiting the
190
domain of discourse over which his cardinality operator ranged to

numbers or classes. He would then not be required to distinguish Julius
Caesar from the number 4, since Julius Caesar does not fall under the
stipulated domain of discourse. But then numbers would no longer be
about everything. As Michael Potter explains: 'a formal language in which
Julius Caesar cannot be spoken of is one in which he cannot be counted,
and in such a language the applicability of arithmetic remains
unexplained' (Potter 2000, 108).
Syllogistic logic depends upon a hierarchy of concepts logically related to
one another in terms of their place in species/genus tree structures, with
different categories at the headings of different tree structures. This logic,
rooted as it is in the metaphysics of substance, has at its disposal a
classificatory scheme by which to organize entities into kinds. At the basis of
the structure is the notion of substance as subject of both essential and
accidental attributes. Socrates is a man. To be a man is to be an animal. To
be an animal is to be a substance. Socrates is pale. To be pale is to be a
colour. To be a colour is to be a quality. To say of Socrates that he is a man is
to attribute an essential attribute to him. To say that he is pale is to attribute
an accidental attribute to him. If Socrates changes from pale to red during a
hot spring day's discussion in the Agora, he undergoes a change of quality,
but his essence remains the same. He is the same man who has undergone a
change in quality. The logical relationships between concepts can be charted
in terms of their essential attributes. A human being is an animal and an
animal is a substance. Four is a number and a number is a quantity.
In developing his new logic Frege eschews necessity and possibility;
following Kant he considered both to be mere psychological attitudes that
one can have toward the content of a judgement. Neither did he make any
use of the distinction between essential and accidental predication. These
are all tools that traditional logic uses to classify objects under different
kinds and thus to distinguish one kind of object from another. Socrates and
Rover are similar in so far as they are animals but different in so far as
Socrates is a human and Rover is a dog. Julius Caesar and the number 4
are different in so far as Julius Caesar is a substance and the number 4 is a
quantity. Thus numbers do not even fall under the same category as
substances and since only substances are objects, numbers are not objects.
Frege, then, finds himself in the difficult situation of requiring logical
objects to serve as the values of his variable in his attempt to establish the
logical foundations of arithmetic with his new quantifier/variable notation
but lacking the logical resources to sort objects into kinds in a way that can
answer the Julius Caesar objection.
191
Once the definition failed due to the Julius Caesar objection, Frege
showed no inclination to abandon his conception of numbers as logical
objects. In the closing paragraph of'Appendix IF of Volume II of the Basic
Laws, written in response to the discovery of Russell's paradox, Frege says:
The prime problem of arithmetic is the question, In what way are we to
conceive logical objects, in particular, numbers? By what means are we
justified in recognizing numbers as objects? Even if this problem is not
solved to the degree I thought it was when I wrote this volume, still I do
not doubt that the way to the solution has been found. (BLA 143)
While Frege never achieved his quest to catch the ever-elusive logical
object, the tools he developed, both logical and philosophical, to aid him in
his quest have proven to be highly influential. Frege did indeed shift the
focus of philosophical attention away from the Cartesian concern with the
relationship between the mind and the world to a concern with the
relationship between word or symbol and object, as Dummett has asserted.
His new logical language is employed in an attempt to establish arithmetic
upon firm epistemological foundations by grounding it in purely logical
foundations. His context principle is used to give a semantic answer to the
epistemological problem of how numbers can be grasped if they cannot be
perceived, imagined or intuited. This shift in focus did indeed initiate a
linguistic turn in philosophy, but to characterize this shift merely as the
overthrow of the Cartesian epistemological tradition is to paint an
incomplete picture of Frege's achievement.
Frege replaces the notion of an object as a substance with an inner core
of essential attributes around which accidents and external relations inhere
with a more general conception of an object as anything that is not a
function. Self-subsistence is all that is left of his notion of an object. Frege's
conception of objects as self-subsistent does not involve any talk of essential
attributes. Instead he uses purely syntactical criteria in his characterization of objects as self-subsistent. If a word acquires its meaning only in the
context of a proposition, and a numerical expression behaves like a
singular term (e.g. it can take the definite article, can be used in identity
statements, etc.), then if the proposition it is contained within is true, the
numerical expression designates a number.
Frege's logical developments opened up a new realm of discovery hitherto
out of reach by the methods of traditional logic. His new quantifier/variable
notation broadened the conception of what counts as an object beyond the
phenomenal substances or possible objects of experience of his predecessors.
192
He introduced the technical tools with his new logic as well as the
philosophical underpinnings with his context principle, sense/reference and
concept/object distinctions, by which this new realm could be explored.
Given its origins in reflection on the nature and relationship between logic
and arithmetic, the territory to be charted possessed a highly abstract and
general nature. This realm was a world apart, so to speak, from the world of
individual substances that formed the starting-point of Aristotle's philosophical speculations. Here Kant was the midwife. By disconnecting the
conception of an object from the notion of a substance as an independently
existing object (a thing in itself) and claiming that existence was not a real
predicate, Kant furnished Frege with the material to expand the scope of
what can legitimately be counted as an object beyond the limits imposed by
Aristotle's substance-based conception of philosophy as well as the different
limits imposed by Kant's critical philosophy.
However, Frege rejects the limits that Kant's critical philosophy places
upon reason and returns to a more robust Leibnizian conception of the scope
and power of pure reason as exemplified in the concept-script, a universal
characteristic meant to capture both the form and the content of judgements.
It is this spirit, I suggest, that lies behind Frege's attempt to give a contextual
definition of number. Frege saw logic as the means to free arithmetic from the
last vestiges of intuition and finally establish its truths on firm logical
foundations. His attempted contextual definition of number in the Foundations
marks the peak of his attempt to develop universal characteristic with the
power that Leibniz had dreamed of. If successful it would have shown how it
is possible to fix the sense of an expression designating a logical object using
only conceptual resources. It would have established the rationalist quest to
show that logic possesses content as well as form. However, Frege's failure to
provide a consistent definition of numbers as logical objects marked the end of
this dream but certainly not the end of his influence.
In a note attached to his will, bequeathing his unpublished papers and
letters to his son Alfred, Frege wrote:
Dear Alfred,
Do not despise the pieces I have written. Even if all is not gold, there is
gold in them, I believe there are things here which will one day be
prized more highly than they are now. Take care that nothing gets lost.
Your loving father.
It is a large part of myself which I bequeath to you herewith.
(PW ix)
193
While the start of a full appreciation of Frege's logical discoveries and

philosophical achievements was delayed until the middle of the twentieth
century, interest in his work, both technical and philosophical, is now in
full flower, and rightly so. In this respect, modern philosophers might
honour Frege with the same praise that he himself once bestowed upon
Leibniz:
In his writings, Leibniz threw out such a profusion of seeds of ideas that
in this respect he is virtually in a class of his own. A number of these
seeds were developed and brought to fruition within his own lifetime
and with his collaboration, yet more were forgotten, then later
rediscovered and developed further. This justifies the expectation that
a great deal in his work that is now to all appearance dead and buried
will one day enjoy a resurrection. (PW 9)
Bibliography
Anderson, R. Lanier (2001), 'Synthesis, cognitive normativity, and the

meaning of Kant's question, "How are synthetic cognitions a priori
possible?" ', European Journal of Philosophy 9.3: 275-305.
Aristotle, The Complete Works of Aristotle (1984), ed. Jonathan Barnes, 2
vols, Princeton, NJ: Princeton University Press.
Ayers, Michael (1991), Locke: Volume II Ontology, London: Routledge.
Burge, Tyler (2005), Truth, Thought, Reason: Essays on Frege, Oxford:
Clarendon Press.
Church, Alonzo (1951), 'A Formulation of the Logic of Sense and
Denotation', in Structure, Method and Meaning: Essays in Honor of Henry
Sheffer, ed. P. Henle, New York: Liberal Arts Press.
Davidson, Donald (1982), Inquiries into Truth and Interpretation, Oxford:
Oxford University Press.
Demopoulous, William (ed.) (1995), Frege7s Philosophy of Mathematics,
Cambridge, MA: Harvard University Press.
Dummett, Michael (1981), Frege: Philosophy of Language, 2nd edn,
Cambridge, MA: Harvard University Press (FPL in text).
(1991), Frege: Philosophy of Mathematics, Cambridge, MA:
Harvard University Press, Harvard (FPM in text).
Edwards, Paul (ed.) (1967), The Encyclopedia of Philosophy, 8 vols, London
and New York: Macmillan.
Evans, Gareth (1985), 'Understanding Demonstratives', in Collected Papers,
Oxford: Oxford University Press.
Falkenstein, Lome (1991), 'Kant's account of intuition', Canadian Journal
of Philosophy, 21.2: 165-93.
Frege, Gottlob (1960), Translations from the Philosophical Writings of Gottlob
Frege, ed. and trans. Peter Geach and Max Black, Oxford: Basil
Blackwell (GB in text).
(1964), Basic Laws of Arithmetic: Exposition of the System, trans, and
ed. Montgomery Furth, Berkeley, CA: University of California Press
(BLA in text).
(1970), Be griffsschrift, a Formula Language, Modeled upon that of
196
Arithmetic, for Pure Thought, in Frege and Godel: Two Fundamental Texts in
Mathematical Logic, ed. Jean van Heijenoort, Cambridge, MA: Harvard
University Press (Bgfin text).
(1972), Conceptual Notation and Related Articles, trans, and ed.
Terrell Ward Bynum, Oxford: Clarendon Press (CN in text).
(1979), Posthumous Writings, ed. Hans Hermes, et al., Chicago,
IL: University of Chicago Press (PW in text).
(1980a), Foundations of Arithmetic, trans. J. L. Austin, 2nd rev. edn,
Evanston, IL: Northwestern University Press (FA in text).
(1980b), Philosophical and Mathematical Correspondence, ed.
Gottfried Gabriel et al., Oxford: Basil Blackwell (PMC in text).
(1984), Collected Papers on Mathematics, Logic and Philosophy, ed.
Brian McGuinness, trans. Max Black et al., Oxford: Basil Blackwell (CP
in text).
(1997), Frege Reader, ed. Michael Beaney, Oxford: Basil
Blackwell Publishers (FR in text).
Friedman, Michael (1992), Kant and the Exact Sciences, Cambridge, MA:
Harvard University Press.
Geach, Peter and Anscombe, G.E.M. (1961), Three Philosophers, Oxford:
Blackwell & Mott.
Geach, Peter (1980a), Reference and Generality (3rd edn), Ithaca, NY:
Cornell University Press.
(1980b), Logic Matters, Berkeley, CA: University of California Press.
Greimann, D. (2000), 'The judgment-stroke as a truth-operator: a new
interpretation of the logical form of sentences in Frege's scientific
language', Erkenntnis, 5.52: 213-38.
Grossmann, Reinhardt (1969)^ Reflections on Frege's Philosophy, Evanston,
IL: Northwestern University Press.
Guyer, Paul (1987), Kant and the Claims of Knowledge, Cambridge:
Cambridge University Press.
(ed.) (1992), The Cambridge Companion to Kant, Cambridge:
Cambridge University Press (CCK in text).
(ed.) (2006), The Cambridge Companion to Kant and Modern
Philosophy, Cambridge: Cambridge University Press.
Haaparanta, Leila (1985), Frege's Doctrine of Being, Acta Philosophica Fennica,
39, Helsinki: Philosophical Society of Finland.
Hanna, Robert (1990), 'Kant's theory of empirical judgment and modern
semantics', History of Philosophy Quarterly, 7.3: 335-51.
(2001), Kant and the Foundations of Analytic Philosophy, Oxford:
Clarendon Press.
Bibliography
197
Hatfield, Gary (1990), The Natural and the Normative: Theories of Spatial
Perception from Kant to Helmholtz, Cambridge, MA: MIT Press.
Heck, Richard G. Jr (2005), 'Julius Caesar and basic law V, Dialectica,
59.2: 161-78.
Jolley, Nicholas (ed.) (1995), Cambridge Companion to Leibniz, Cambridge:
Cambridge University Press (CCL in text).
Kant, Immanuel (1929), Critique of Pure Reason, trans. Norman Kemp
Smith, New York: St Martin's Press.
(1967), Philosophical Correspondence 1759-99, ed. and trans. Arnulf
Zweig, Chicago, IL: University of Chicago Press.
(1974), Logic, trans. Robert S. Hartman and Wolfgang
Schwarz, New York: Dover (Logic in text).
(1992), Lectures on Logic, trans, and ed. J. Michael Young,
Cambridge: Cambridge University Press (Jdsche Logic in text).
(1997), Prolegomena to any Future Metaphysics, trans, and intro.
James W. Ellington, Indianapolis, IN: Hackett.
(1998), Critique of Pure Reason, trans, and ed. Paul Guyer and
Allen Wood, Cambridge: Cambridge University Press.
Kaplan, David (1989), 'Demonstratives', in Themes from Kaplan, ed. Joseph
Almog, John Perry and Howard Wettstein, Oxford: Oxford University Press.
Kripke, Saul (1972), Naming and Necessity, Cambridge, MA: Harvard
University Press.
Leibniz, Gottfried (1956), Philosophical Papers and Letters, trans, and ed. Leroy
Loemker, Dordrect-Holland: D. Reidel Publishing Company (PPL in text).
(1966), Logical Papers, trans, and ed. G.H.R. Parkinson, Oxford:
Clarendon Press (LP in text).
(1986), New Essays on Human Understanding, ed. and trans. Peter
Remnant and Jonathan Bennett, Cambridge: Cambridge University
Press (NE in text).
(1989), Philosophical Essays, ed. and trans. Roger Ariew and
Daniel Garber, Indianapolis, IN: Hackett (PE in text).
Locke, John (1975), An Essay Concerning Human Understanding, ed. and intro.
Peter H. Nidditch, Oxford: Oxford University Press.
Longuenesse, Beatrice (1998), Kant and the Capacity to Judge, trans. Charles
T. Wolfe, Princeton, NJ: Princeton University Press.
Mates, Benson (1972), Elementary Logic, 2nd edn, New York: Oxford
University Press.
(1986), The Philosophy of Leibniz: Metaphysics and Language, New
York: Oxford University Press.
Mendelssohn, Moses (1997), 'On Evidence in Metaphysical Sciences', in
198
Philosophical Writings, ed. Danield Dahlstrom, Cambridge: Cambridge

University Press.
Mugnai, Massimo (1992), Leibniz's Theory of Relations, Stuttgart: Franz Steiner.
Noonan, Harold W. (2001), Frege: A Critical Introduction, Cambridge: Polity
Press.
Pap, Arthur (1958)^ Semantics and Necessary Truth: An Inquiry into the
Foundations of Analytic Philosophy, New Haven, CT: Yale University Press.
Parsons, Charles (1969), 'Kant's Philosophy of Arithmetic', in Philosophy,
Science and Method: Essays in honor of Ernest Nagel, ed. Sidney
Morgenbesser et al., New York: St Martin's Press.
(1982), 'Objects and logic', Monist, 65: 491-516.
(1992), 'Arithmetic and the Categories', in Kant's Philosophy of
Mathematics, ed. Carl J. Posy, Netherlands: Kluwer Academic.
Perry, John (1977), 'Frege on demonstratives', Philosophical Review, 86: 47497.
Potter, Michael (2000), Reason's Nearest Kin: Philosophies of Arithmetic from
Kant to Carnap, Oxford: Oxford University Press.
Quine, Willard van Orman (1952), fOn What there Is', in Semantics and the
Philosophy of Language, ed. Leonard Linsky, Urbana, IL: University of
Illinois Press.
(1960), Word and Object, Cambridge, MA: MIT Press.
Ruflino, Marco Antonio (1991), 'Context principle, fruitfulness of logic
and the cognitive value of arithmetic in Frege', History and Philosophy of
Logic, 12: 185-94.
Shabel, Lisa (2006), in 'Kant's Philosophy of Mathematics', in The
Cambridge Companion to Kant and Modern Philosophy, ed. Paul Guyer,
Cambridge: Cambridge University Press.
Strawson, Peter (1959), Individuals: An Essay in Descriptive Metaphysics,
London: Methuen.
Tappenden, Jamie (1995), 'Extending knowledge and "fruitful concepts":
Fregean themes in the foundations of mathematics', Nous, 29.4: 427-67.
Wolff, Robert Paul (1973), Kant's Theory of Mental Activity: A Commentary on
the Transcendental Analytic of the Critique of Pure Reason, Gloucester, MA:
Peter Smith.
Wright, Crispin (1983), Frege7s Conception of Numbers as Objects, Aberdeen:
Aberdeen University Press.
Young, J. Michael (1992), 'Functions of thought and the synthesis of
intuitions', in The Cambridge Companion to Kant and Modern Philosophy, ed.
Paul Guyer, Cambridge: Cambridge University Press.
(1994), 'Synthesis and content of pure concepts in Kant's first
Critique', Journal of the History of Philosophy, 32.3: 331-57.
Index
a posteriori^, 174-5, 178-9

a priori 2-^, 11, 17-18, 89, 108, 116,
174-9
abstraction 167
accidents 9
active/passive voice 51
Alexander the Great 128-30
algebra 93
An Essay concerning Human Understanding
(Locke) 136
analytical judgements 6-7, 92, 101,
111-12
appearances 19, 84-5, 85, 163-4
arguments 17, 55
Aristotle
judgements/arguments 17
objects 186
predication 125
substance 127, 169, 181, 186
syllogistic theory 5-6, 187-8
arithmetic
a priori 11, 14, 17-18, 89, 108, 116
analytical 93, 110-11
concepts 117
and geometry 16
laws 110
and Leibniz 73, 109, 111
and logic 33-5
and mathematics 91
numbers as objects 80
principles 189
structure of propositions 17
symbols 76
Arnaud, Antoine 17
assertable/unassertable contents 42
Auszug aus der Vernunftlehre (Meier) 23
Ayers, Michael 20, 137-8, 147-8, 1701, 183

Basic Laws of Arithmetic (Frege) 13, 15,
191
Begriffsschrift (Frege) 1, 13-14, 33, 179,
187
being see existence
Boole, George 114-15
'Boole's Logical Calculus and the
Concept-script' (Frege) 114
Burge, Tyler 179
cardinality operator 153, 188, 190
categories 18
Categories (Aristotle) 125
cognition
a priori/a posteriori 175
experience 34
judgements 176
objects 72
philosophical 83
representation 27
significance 54
understanding 70
concept-script 114-17
concepts
analysis/synthesis 99, 106
arithmetical 117
content 51-3, 55-6, 63n.6
definition 112-13, 159
empirical 84
existence 167
form/matter 61
four relations 132
judgements 115, 151, 167
200
Index
Kant/Wolff 71-2
logical form 12, 26-7, 61, 115
mathematical cognition 76-7
relative 148
substance 124
and truth 18
as truth-value 168
understanding 26, 679
conceptualism/realism 170
consciousness 27-30
content-stroke 49
context principle 139-40, 148-52, 154
contextual definition 159
contradiction 978
copula 47-8, 61, 63
Critique of Pure Reason (Kant) 2, 75-6,
107, 175
denomination 160
Descartes, Rene 2, 18, 191
determinate/indeterminate numbers 41
determinations 133, 134
differentia 6
dogmatism 701
Dummett, Michael 18-19, 140, 147,
153, 170, 183, 188-9
'Elements of Calculus' (Leibnitz) 94
empirical concepts 84
entia rationis 181
essential/non-essential elements 1001
existence 165-70, 167
experience 34
external things 141-2
Foundations of Arithmetic (Frege) 1314,
89-90, 108, 153, 184-5
Frege: Philosophy of Language (Dummett)
1, 18
Frege: Philosophy of Mathematics
(Dummett) 147
Frege's logical judgement forms 589,
61-2
Friedman, Michael 13
function 55-7
generality 179
geometry 16, 73, 109-10, 111, 116, 154

Grossmann, Reinhardt 1234
Guyer, Paul 10-11
Haaparanta, Leila 189
Hanna, Robert 2
Hatfield, Gary 25
Heck, Richard 157-8
Hume, David 2
'Hume's Law' 153-4, 158
idealism 171
ideas 173
identity 53-4, 100
image 85
inference 30, 31, 32, 52, 61-2, 64n.7
intuition
empirical/pure 77Q, 84
formal 10
magnitude 108
sensibility 67-9
space/time 70, 78, 110
unit representation 76
judgements
act of judging 50
affirmation/denial 52
analytical 6-7, 92, 101, 111-12
analytical/synthetic 4-5, 18, 62,
96-102
apodeictic 44-6, 60-1
and Aristotle 17
assertoric / problematic 42-7, 60-1
categorical/hypothetical/disjunctive
6, 11, 30, 60, 130
cognition 176
concepts 115, 151, 167
consciousness 2830
content 49-50, 56, 91
copula 47-8, 57, 61, 63, 92
existence 167
Frege's new conception 412, 503
function/argument 55, 92-3, 115,
123-4
hypothetical 30-1
Kant/Frege 27, 57-9, 130
logical forms 58-9, 61-2
Index
201
modality 45-6, 166

numbers 1467
object intuition 163
out of concepts/arguments 17, 26
problematic 44-5, 47
quantity 18-19, 81
subject/predicate/copula 57, 92
synthetic/singular 81-2, 92-3, 163-4
universal/particular 60
'Julius Caesar problem' 15, 19, 154-5,
157-8, 186, 189-91
static system 185

traditional 11
transcendental 102-4
and truth 34
Logic (Kant) 100
'Logic in Mathematics' (Frege) 185
'Logical Calculus and the ConceptScript' (Boole) 49
'Logical Defects in Mathematics'
(Frege) 41
Longuenesse, Beatrice 75, 81, 176-7
knowledge-of-grammar 126
Mates, Benson 41, 181

mathematics
and arithmetic 91
cognition 76-8
concepts 76-7
magnitude construction 78
objects 173
truths 91, 113
Meier, Friedrich 23
Mendelssohn, Moses 7-8
mind 107-8
modality 45-6, 165-7
modes of determination 545
modus ponens 312, 52, 62
modus tollens 31-2
monads 8, 13, 129-33, 137, 161, 181
language 37
Lanier Anderson, R. 107
Lectures on Metaphysics (Kant) 78
Leibniz, Gottfried
and appearances 69
arithmetic/geometry 73, 109, 111
denomination 160
and Frege 73-5, 92-3, 192-3
innate ideas 173
introduction 3
and Kant 8-9, 17, 70, 76, 131-5
and Locke 137
logical notation 114
one 144
primary/derivative truths 174
scope of reason 171
substance 127-9, 133, 162, 181-2
truth 93-6, 101, 179-81
universal characteristic 35-7
Locke, John 69, 136-7, 162, 173
logic
and arithmetic 335
concept-script 11617
concepts 12, 61, 115
description 5-6
formal 36
Frege's new 37-40
general 102-3
Kant's conception 256, 177
monadic 8, 13
notation 1, 114
polyadic 8, 13, 17
rules 24
natural light 172-3

'Negation' (Frege) 49
New Essays on Human Understanding
(Leibniz) 73, 75, 86n.3, 173
Nicole, Pierre 17
nominalism 181
Noonan, Harold 74
notation 1, 17, 114
noumena see things-in-themselves
numbers
concept of 89-90
context principle 148-52, 154, 188
definition 154-6, 157
external things 1413
and Frege 116-19, 124, 146-7
identity 152-3
independent existence 140-1
judgements 146-7
202
Index
as logical objects/unknowable 19, 159

and objects 140
one 144-6
objectivity 143
objects
and Aristotle 186
cognition 72
concepts 84, 103
inner/outer 161
and Kant 72, 135-6, 143-4, 162-4
and mathematics 173
numbers 140
and substance 19, 124, 158-9, 171,
191
transcendental 161
one 144-5
Parsons, Charles 80, 117, 166
philosophical cognition 767
Plato 180
plurality 82
polyadic logic 8, 13, 17
Port-Royal Logic (Arnaud/Nicole) 17
Potter, Michael 188, 190
predicates 74, 94, 102, 125-6
Prolegomena to any Future Metaphysics
(Longuenesse) 81
propositions 80, 100, 160
psychology 25, 91
quantity/quantum 59, 78-9, 85
realism/conceptualism 170
reason 70
receptivity 69
reflection 26
representation
cognition 27
consciousness 28
various treatments 68-9
rules 23-4
Russell, Bertrand 15-17, 156
Rutherford, Donald 128
schema 80, 84, 85
Schroeder, Ernst 114
science 185-7
sense 10, 184
sensibility
faculty of receptivity 69
intuition 67, 69
and Kant 165
and Leibnitz 69
space/time 70
sentences, relational 745
Socrates 125-6, 180
soul 126, 138n.l
space/time 68-70, 77-8, 103-4, 110,
159
substance
and Aristotle 127, 169, 181, 186
attributes 123
and change 19
concepts 124
corporeal/incorporeal 6
independent 161-2
Leibniz 127-9
and Locke 136-7
and objects 19, 124, 158-9, 171, 191
phenomenal 161-2
primary/secondary 126, 1612
simple/phenomenal 1335
substantial form 12831
syllogistic theory 5-6, 31, 187-8, 190
symbols 39-40
and arithmetical calculation 76
symbolic/ostensive construction 79
synthesis 13, 103-8, 135, 178
systems 185
Table of Judgements 43-4, 81
things-in-themselves 72, 104, 135, 161,
165
thought 184
time
inner sense 28
and schema 85-6
'Transcendental Aesthetic' (Kant)
9-10, 71, 103
transcendental idealism 11, 1601
transcendental logic 1024
transcendental objects 161, 165
truth
Index
a posteriori I a priori 178
concept/containment 18
entertained/asserted 46
and logic 34
objective structure 183-4
primary/derivative 945, 174
truth-tables 38
truth-values 45, 48-9, 51-2, 124, 168
understanding 26, 679

unity 1478, 1634
universal characteristic 35-7
Wolff, Christian 3, 12
Wolff, Robert Paul 43-4, 71
unassertable/assertable contents 42
Young, Michael J. 104-7
Venn diagrams 115
203

(Bloomsbury Studies in Philosophy) Delbert Reed-Origins of Analytic Philosophy - Kant and Frege-Bloomsbury Academic (2008)

Uploaded by

Copyright:

Available Formats

(Bloomsbury Studies in Philosophy) Delbert Reed-Origins of Analytic Philosophy - Kant and Frege-Bloomsbury Academic (2008)

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

(Bloomsbury Studies in Philosophy) Delbert Reed-Origins of Analytic Philosophy - Kant and Frege-Bloomsbury Academic (2008)

Uploaded by

Copyright:

Available Formats

Origins of Analytic Philosophy

Continuum Studies in Philosophy

Origins of Analytic Philosophy

Continuum International Publishing Group

Typeset by BookEns Ltd, Royston, Herts.

This book is dedicated with love and affection to Betsy

This page intentionally left blank

Part 1 Logic Old and New

Part 2 Kant on Concepts, Intuitions and Arithmetic

Part 3 Analyticity and Fruitful Definitions

Part 4 From Substance to Object

This page intentionally left blank

Writing is a lonely endeavour. This is especially true in the absence of a

This page intentionally left blank

Gottlob Frege (1848-1925) is considered to be one of the founding fathers

Origins of Analytic Philosophy

context of a sentence) and compositionality principle (the meaning of a

Origins of Analytic Philosophy

Kant and Frege: An Initial Comparison

relation is possible in two different ways. Either the predicate B belongs

Origins of Analytic Philosophy

judgement was considered to be the most basic. Categorical judgements

(Kant took secondA place in the contest):

Origins of Analytic Philosophy

justified by unpacking the meaning of their subject and predicate concepts

Origins of Analytic Philosophy

In the 'Transcendental Aesthetic', Kant uses the ideality of relations

Origins of Analytic Philosophy

It is possible to form a concept of the number 12 by means of a variety of

Origins of Analytic Philosophy

there is no need to invoke the power of intuition to explain how the

Origins of Analytic Philosophy

Outline of the Book

Origins of Analytic Philosophy

Frege 'effected a revolution in philosophy as great as the similar revolution

Origins of Analytic Philosophy

influence on Frege should not lead to an interpretation of Frege as some

See, for example, the Geach 1961 chapter on Frege in Three

Logic Old and New

This page intentionally left blank

Kant on Pure General Logic

Origins of Analytic Philosophy

'Since the understanding is the source of rules', concludes Kant, 'the

Psychology is an empirical science that seeks to explain how humans think.

Origins of Analytic Philosophy

Concepts in Kant's Logic

generation of any concept. They perform their function by bringing

Judgements in Kant's Logic

Origins of Analytic Philosophy

consciousness is the universal condition that makes any representation

Origins of Analytic Philosophy

are combined into a unity by an act of judgement. The form of a

Inferences in Kant's Logic

Origins of Analytic Philosophy

In modus ponens the premise is a hypothetical judgement and the conclusion

Preface to the Begriffsschrift

Origins of Analytic Philosophy

then presents a brief argument to show that it is not psychological origin

Frege likens his new logical notation to Leibniz's attempt to develop a