A Priori Knowledge: Toward a Phenomenological Explanation

219 Pages·2006·1.131 MB·English

by  Tommaso Piazza

Tommaso Piazza
A Priori Knowledge
Toward a Phenomenological Explanation
P H E N O M E N O L O G Y & M I N D

Herausgegeben von / Edited by

Arkadiusz Chrudzimski • Wolfgang Huemer

Band 10 / Volume 10
Tommaso Piazza

A Priori Knowledge

Toward a Phenomenological Explanation

ontos

verlag

Frankfurt I Paris I Ebikon I Lancaster I New Brunswick
Bibliographic information published by the Deutsche Nationalbibliothek
The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliographie;
detailed bibliographic data is available in the Internet at http://dnb.ddb.de

North and South America by
Transaction Books
Rutgers University
Piscataway, NJ 08854-8042
[email protected]

United Kingdom, Ire, Iceland, Turkey, Malta, Portugal by
Gazelle Books Services Limited
White Cross Mills
Hightown
LANCASTER, LA1 4XS
[email protected]

Livraison pour la France et la Belgique:
Librairie Philosophique J.Vrin
6, place de la Sorbonne ; F-75005 PARIS
Tel. [phone] ; Fax [phone]
www.vrin.fr

2007 ontos verlag
P.O. Box 15 41, D-63133 Heusenstamm nr. Frankfurt
www.ontosverlag.com

ISBN 10: 3-937202-92-7
ISBN 13: 978-3-937202-92-1

2007

No part of this book may be reproduced, stored in retrieval systems or transmitted
in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise
without written permission from the Publisher, with the exception of any material supplied specifically for the
purpose of being entered and executed on a computer system, for exclusive use of the purchaser of the work

Printed on acid-free paper
ISO-Norm 970-6
This hardcover binding meets the International Library standard

Printed in Germany
by buch bücher dd ag
Table of Contents

Introduction v

1. A Priori, Analyticity, and Implicit Definition

Empiricism, Analyticity, and the A Priori 1
Reductive and Non-Reductive Conceptions of Analyticity 3
Implicit Definition, Logical Truth, and the Recalcitrant A Priori 5
Problems with Implicit Definition 8
BonJour’s Objection 10
Fodor and Lepore’s Objection 13
Horwich’s Objection 23
Hale and Wright’s defence of the traditional connection 31
Logic and Convention 46
Coda 52

2. Realism about Logic

Introduction 57
Logical Principles, Justification and Epistemic Relativity 60
Objective Truth 64
Resnik’s Attack 65
Wittgenstein on the necessity of ‘1 inch = 2.54 cm’
and logical inference 75
Dummett’s Objection 79
Rule Following considerations and the adoption of a convention 84
Summarising Remarks 87
Wright’s Attack 89
Conclusion 109
iv TABLE OF CONTENTS
3. Objective Knowledge

Introduction 111
What the Tortoise Said to Boghossian 115
What Boghossian would say to the Tortoise 117
Rule-circular Arguments 119
The Side-Argument 122
Rejecting the Side-Argument 123
First Horn: Simple Internalism and Rational Insight 123
Second Horn: Epistemic Responsibility and the Lack
of Epistemic Irresponsibility 126
Realism, the A priori and Rational Insight 131
Boghossian’s Argument against Relativism 132
Epistemological Realism about Justification 134
Conclusion 135

4. Phenomenology and Rational Insight

Naturalism and Justification 138
Phenomenology, Justification, and Eidetic Seeing 145
Is Holism a Possibility for the Empiricist? 150
Intuition of Essences and the Analytic/Synthetic Distinction 156
Husserl’s Conception of the Analytic/Synthetic Distinction 157
Eidetic Variation 164
Passive synthesis and Concept Constitution 168
Knowledge of Reality and Conceptual Truth 174
Absolute vs Relative Objectivity 177
Are Conceptual Truths True? 179

Conclusion 183

References 185
Introduction

Do we know anything a priori, that is to say independently of experi-
ence? Do we know a priori anything objective concerning reality? Do
we know logic? Do we know it a priori? This book is devoted to criti-
cally analysing these problems, to inquiring into existing answers to
these questions, and to suggesting a plausible way out of the difficulties
my analysis will hopefully bring to the fore.
Addressing the general question about whether we do know anything
a priori involves addressing two distinct sub-questions. The first one
concerns the very nature of the statements allegedly known a priori, that
is to say concerns the question whether these statements do express the
kind of things that can in principle be known, and the question concern-
ing what kind of knowledge one acquires when knowing them. The sec-
ond subquestion deals with the way, if these statements do express the
kind of things that can in principle be known, they can, and indeed are
(at least sometimes and locally), known a priori.
Contemporary empiricism has delivered a unified answer to both
questions with the notion of analyticity. In accordance with this tradi-
tional answer, all a priori statements are analytic statements. As syn-
thetic statements, analytic statements are truth-apt (indeed, true) state-
ments. Therefore, a priori statements are the kinds of things that can be
known. However, unlike synthetic statements, analytic statements do not
have cognitive content, they do not say anything about reality. This fea-
ture of analytic statements is due to the fact that their truth, on the em-
piricist reading of the notion of analyticity, is entirely due to the mean-
ings of their constituting expressions. So, a priori knowledge is not any
kind of knowledge of reality. However, the distinctive nature of analytic
truth yields a satisfactory account of our knowledge of it. For a natural
suggestion is that if the truth of a statement is entirely determined by the
meaning of its constituting expressions, then its truth can be known sim-
ply by understanding the statement. Accordingly, the truth of an analytic
statement can be known merely by understanding it, therefore a priori.
This empiricist solution has exerted a considerable influence among
empirically minded philosophers; in fact it avoids the intuitive draw-
backs of Mill’s solution according to which alleged pieces of a priori
knowledge, like mathematical and logical knowledge, is empirical and
inductive in nature. Mill’s solution is consistent with the empiricist prin-
vi INTRODUCTION
ciple according to which every piece of knowledge of reality stems from
and is justified on the basis of experience. However, this solution has the
unpalatable feature of imposing a conception of mathematic and logic as
just contingently true, for just beliefs in contingent statements can be jus-
tified, to a degree sufficient for knowledge, by inductive and empirical
means. The solution based on analyticity allows the empiricist to defend
her epistemological principle without loosing the necessity of mathe-
matical or logical statements. If mathematic and logic say nothing about
reality, in fact, to admit that our mathematical or logical knowledge is a
priori is not to admit that our knowledge of reality has sources other than
our experience of it.
The viability of this empiricist solution clearly depends on the very
notion of analyticity called into question. Notoriously, the empiricist no-
tion of analyticity is not Kant’s notion. However, admitting that Kant’s
notion constitutes an improvement on such notion is probably not that
far from the truth. Kant held that an analytic statement is one character-
ized by the fact that its predicate term expresses a concept which is con-
tained by the concept expressed by its subject term. Kant’s analyticity
has been found wanting for two reasons. It is too narrow, for it applies
just to statements of the subject-predicate form: statements like

(K1) Everything is spatio-temporal or is not spatio-temporal

is not analytic according to Kant’s definition, because its predicate con-
cept is not contained in its subject concept.
Secondly, it is too wide, for a statement like

(K2) Every daughter of a professional philosopher is a professional phi-
losopher

is false, and a fortiori not analytic, yet the concept [professional phi-
losopher] expressed by the predicate term is actually contained by the
concept [daughter of a professional philosopher] expressed by the sub-
ject term.
However, Kant notoriously also held that analytic statements are not
ampliative with respect to our knowledge of reality, while synthetic
statements are; for this reason he also claimed that the a priori epistemo-
logical status of analytic statement is not problematic: given the con-
tainment theory, to know an analytic statement it is sufficient to possess
INTRODUCTION vii
both its predicate- and the subject-concept, and to be acquainted with the
principle of non-contradiction. Both features are preserved within the
empiricist notion of analyticity, because according to this notion analytic
statements do not say anything about reality.
Given the wider nature of the empiricist conception with respect to
its applicability just to statements of the subject-predicate form, how-
ever, the epistemology of analytic statement is not anymore Kant’s.
Rather, we might say, it is Frege’s.
Frege held that a statement is analytic when it is either a substitu-
tional instance of a logical principle – much in the way “it rains → it
rains” is a substitutional instance of the principle “p → p” – or can be
reduced to a substitutional instance of a logical principle with the aid of
definitions – much in the way “every bachelor is an unmarried man” can
be reduced to “every unmarried man is unmarried”.
If, pace Quine, the question whether two expressions of a natural
language are synonymous is not intractably unintelligible, and if a com-
petent speaker of a natural language is indeed in a position to answer
such a question whenever it arises concerning two distinct expressions,
the question about how analytic statements can and are known reduces
to the question about how predicate and propositional logic is known.
Though Frege considered the latter epistemological task as completely
unproblematic – convinced, as he was, that logical principles are simply
self-evident – it is clearly vital for a sound empiricist theory of the a pri-
ori to provide such an account. Unless we are told how logic is known,
we cannot stay content with the contention that a priori knowledge of
analytic statement is unproblematic because, at bottom, it reduces to
knowledge of logical truths.
A simple suggestion is that logical principles are known either be-
cause they are implicit definitions of the logical constants they contain,
or because they are deducible from such principles. This is the contem-
porary proposal I shall consider at the beginning of the first chapter.
The basic idea is that the meaning of certain expressions – in the case
at issue the meaning of the logical constants – is determined by con-
straining those expressions to have whatever meaning is required for the
truth or the correctness of certain basic contexts that contain them. It
follows that no one understanding such context (sentences or rules of
inference) can fail to appreciate that they are true (valid), if they perform
the role of implicit definitions. So long as they do, understanding what
they mean coincides with (because it requires) appreciating that they are
viii INTRODUCTION
true (valid). More than this, the theory seems to make good sense of the
traditional suggestion according to which logical principles are self-
evident. Since accepting such principles is constitutive of the capability
of understanding what they say, it follows that no one fully understand-
ing what they say can fail to appreciate that what they say is true.
An empiricist should welcome this account for a third important rea-
son. Take the following statement:

(H) If something is entirely coloured of red it is not, at the same time
and under the same respect, entirely coloured of green.

(H) is not the instantiation of a logical principle, nor it is reducible to
the instantiation of a logical principle if synonymous are substituted by
synonymous. The meaning of “red”, in fact, is not the same as the mean-
ing of the (conjunctive) predicate “not green and not blue and not gray
…”. Given the potential infinity of colour discriminating expressions,
the meaning of any colour term “c” could not be grasped in the first
place if it were equivalent to the infinite conjunction of the negative
predicates constructed out of each colour expression other than “c”. Ac-
cordingly, statements like (H1) are to be counted as synthetic under the
standard (Fregean) empiricist notion of analyticity. The problem is that
synthetic a priori knowledge is not consistent with the empiricist episte-
mological principle. In contrast with analytic statements, synthetic
statements are about reality. So, the admission of a priori knowledge of
synthetic statements entails that experience is not the only source of
knowledge and justification.
The notion of implicit definition seemingly makes it available to the
empiricist a plausible way out. As it makes available a notion of logical
truth according to which logical principles are just definitions of a de-
vised sort of the logical constants they contain, it seemingly makes
available the view that statements like (H1) are just definitions of a de-
vised sort of the color predicates they contain, therefore acceptable from
an empiricist epistemological point of view.
The first chapter of this book will be devoted to taking into consid-
eration several criticisms, advanced by L. BonJour, J. Fodor and H.
Lepore, and by P. Horwich, against P. Boghossian’s idea that meaning
coincides with conceptual role, and against the idea that the meaning of
certain expressions is implicitly defined by the resolve to accept as true
given contexts featuring these very expressions. In accordance with a