Philosophical Essays, Volume 2: The Philosophical Significance of Language - Softcover

Soames, Scott

Softcover

ISBN 10: 0691136831 ISBN 13: 9780691136837

Verlag: Princeton University Press, 2009

Inhaltsangabe

The two volumes of Philosophical Essays bring together the most important essays written by one of the world's foremost philosophers of language. Scott Soames has selected thirty-one essays spanning nearly three decades of thinking about linguistic meaning and the philosophical significance of language. A judicious collection of old and new, these volumes include sixteen essays published in the 1980s and 1990s, nine published since 2000, and six new essays.


The essays in Volume 1 investigate what linguistic meaning is; how the meaning of a sentence is related to the use we make of it; what we should expect from empirical theories of the meaning of the languages we speak; and how a sound theoretical grasp of the intricate relationship between meaning and use can improve the interpretation of legal texts.


The essays in Volume 2 illustrate the significance of linguistic concerns for a broad range of philosophical topics--including the relationship between language and thought; the objects of belief, assertion, and other propositional attitudes; the distinction between metaphysical and epistemic possibility; the nature of necessity, actuality, and possible worlds; the necessary a posteriori and the contingent a priori; truth, vagueness, and partial definition; and skepticism about meaning and mind.


The two volumes of Philosophical Essays are essential for anyone working on the philosophy of language.

Über diesen Titel

Über die Autorin bzw. den Autor

Scott Soames is director of the School of Philosophy at the University of Southern California. His books include "Reference and Description" (Princeton), "Philosophical Analysis in the Twentieth Century", Volumes 1 and 2 (Princeton), "Beyond Rigidity", and "Understanding Truth".

Von der hinteren Coverseite

"Soames's work is of an exceptionally high quality, the selections made here are truly excellent, and the organization is well thought out. Having these papers available in this form is a great boon to scholars."--Stephen Neale, CUNY Graduate Center

"Since many of these important papers are relatively inaccessible, it is particularly useful to have them collected together, and Soames has done an excellent job of selecting and arranging them. These two volumes are really terrific."--Alex Byrne, Massachusetts Institute of Technology

Auszug. © Genehmigter Nachdruck. Alle Rechte vorbehalten.

PHILOSOPHICAL ESSAYS

THE PHILOSOPHICAL SIGNIFICANCE OF LANGUAGEBy Scott Soames

PRINCETON UNIVERSITY PRESS

Copyright © 2009 Princeton University Press
All right reserved.
ISBN: 978-0-691-13683-7

Contents

The Origins of These Essays.............................................................................................ixIntroduction............................................................................................................1Part One Reference, Propositions, and Propositional Attitudes..........................................................31Essay One Direct Reference, Propositional Attitudes, and Semantic Content..............................................33Essay Two Why Propositions Can't Be Sets of Truth-Supporting Circumstances.............................................72Essay Three Belief and Mental Representation...........................................................................81Essay Four Attitudes and Anaphora......................................................................................111Part Two Modality......................................................................................................137Essay Five The Modal Argument: Wide Scope and Rigidified Descriptions..................................................139Essay Six The Philosophical Significance of the Kripkean Necessary A Posteriori........................................165Essay Seven Knowledge of Manifest Natural Kinds........................................................................189Essay Eight Understanding Assertion....................................................................................211Essay Nine Ambitious Two-Dimensionalism................................................................................243Essay Ten Actually.....................................................................................................277Part Three Truth and Vagueness.........................................................................................301Essay Eleven What Is a Theory of Truth?................................................................................303Essay Twelve Understanding Deflationism................................................................................323Essay Thirteen Higher-Order Vagueness for Partially Defined Predicates.................................................340Essay Fourteen The Possibility of Partial Definition...................................................................362Part Four Kripke, Wittgenstein, and Following a Rule...................................................................383Essay Fifteen Skepticism about Meaning: Indeterminacy, Normativity, and the Rule-Following Paradox.....................385Essay Sixteen Facts, Truth Conditions, and the Skeptical Solution to the Rule-Following Paradox........................416Index...................................................................................................................457

Chapter One

Direct Reference, Propositional Attitudes, and Semantic Content

What do we want from a semantic theory? A plausible answer is that we want it to tell us what sentences say. More precisely, we want it to tell us what sentences say relative to various contexts of utterance. This leads to the view that the meaning of a sentence is a function from contexts of utterance to what is said by the sentence in those contexts. Call this the propositional attitude conception of semantics.

Another semantic picture that has enjoyed considerable popularity is the truth-conditional conception. According to it, the job of a semantic theory is to tell us what the truth conditions of sentences are. On this view, the meaning of a sentence can be thought of as a function from contexts of utterance to truth conditions of the sentence as used in those contexts.

Suppose now that we put the propositional attitude and the truth-conditional conceptions together. If we do this, it is virtually irresistible to conclude that what is said by a sentence in a context consists in its truth conditions relative to the context. But what are truth conditions?

One natural idea, embraced by the ruling semantic paradigm, is that the truth conditions of a sentence, relative to a context, are the metaphysically possible worlds in which the sentence, as used in the context, is true. Such truth conditions can be specified by a recursive characterization of truth relative to a context and a world. This characterization implicitly associates with each sentence a function representing its meaning. The value of the function at any context as argument is the set of metaphysically possible worlds in which the sentence, as used in the context, is true. It is this that is identified with what is said by the sentence in the context, when the propositional attitude conception of semantics is combined with this version of the truth-conditional conception.

This identification is, of course, highly problematic. The first difficulty one notices is that if S and S are necessarily equivalent relative to a context, then they are characterized as saying the same thing, relative to the context. However, it is highly counterintuitive to hold that all necessary truths say the same thing, that the conjunction of a sentence with any necessary consequence of it says the same thing as the sentence itself, and so on.

A plausible pragmatic principle extends this difficulty to the propositional attitudes of speakers.

(1) A sincere, reflective, competent speaker who assertively utters S in a context C says (or asserts), perhaps among other things, what S says in C.

This principle reflects an incipient relational analysis of the attitude of saying, or asserting—an analysis that sees it as a relation between speakers and things which serve as the semantic contents of sentences. Once this analysis is accepted, it is a short step to view propositional attitude reports in accord with (2) and (3).

(2) An individual i satisfies x says (asserts) that S relative to a context C iff i stands in a certain relation R to the semantic content of S in C.

(3) An individual i satisfies x v's that S (where v = 'believes', 'knows', 'proves', 'expects', etc.) relative to a context C iff i stands in a certain relation R to the semantic content of S in C.

But now our difficulties are surely unmanageable. Let us characteriz distribution over conjunction and closure under necessary consequence a follows:

Distribution over Conjunction

If an individual i satisfies x v's that P&Q relative to C, then i satisfies x v's that P and x v's that Q relative to C. (For example, anyone who asserts that P&Q asserts that P and asserts that Q.)

Closure under Necessary Consequence

If an individual i satisfies x v's that P relative to C, and if every possible world in which P is true relative to C is a possible world in which Q is true relative to C, then i satisfies x v's that Q relative to C. (For example, anyone who asserts that P asserts everything that necessarily follows from P.)

The second main difficulty with our combined truth-conditional and propositional attitude conception of semantics is that it equates distribution of a propositional attitude verb over conjunction with closure of the attitude under necessary consequence. For if Q is a necessary consequence of P, then the set of metaphysically possible worlds in which P&Q is true is the same as the set of worlds in which P is true. Given the identification of truth conditions with semantic content, this means that their semantic contents are the same. But then, a relational semantics of propositional attitude reports together with distribution over conjunction will yield closure under necessary consequence.

The problem is that for many propositional attitude verbs distribution over conjunction is a fact whereas closure under necessary consequence is not. My four year old son Greg has said many things, and whenever he says that P&Q he says that P and he says that Q. However, there are lots of necessary consequences of things he has said that he has left unasserted, for example that 2 to the ninth = 512, that first order logic is complete but undecidable, and that stones are made up of molecules.

A third difficulty with our semantic conception takes this problem one step further. The same considerations that lead to the view that beliefs and assertions are closed under necessary consequence lead to the view that no one has ever believed or asserted anything that couldn't have been true (in any metaphysically possible world). Since every Q is a necessary consequence of an impossible P, anyone who believes or asserts what P expresses believes or asserts everything. And surely, no one ever has, or could have, done that.

The semantic assumptions that lead to these difficulties can be summarized as follows:

A1a. The semantic content of a sentence (relative to a context) is the collection of circumstances supporting its truth (as used in the context).

A1b. The collection of circumstances supporting the truth of a sentence (as it is used in a context) = the set of metaphysically possible worlds in which it is true (relative to the context).

A2. Propositional attitude sentences report relations to the semantic contents of their complements—i.e., an individual i satisfies x v's that S (relative to a context C) iff i bears R to the semantic content of S (relative to C).

A3. Many propositional attitude verbs, including 'say', 'assert', 'believe', 'know', and 'prove' distribute over conjunction.

Since these assumptions lead to unacceptable results, one or more of them must be rejected.

The crucial assumptions are A1 and A2, which, in turn, are direct descendants of the two conceptions of semantics mentioned earlier. A1 (a and b) represent the truth-conditional conception, with metaphysically possible worlds taken as truth conditions. A2 represents the propositional attitude conception, with the relational analysis of 'say', and 'assert' extended to propositional attitude reports generally. The need to give up one or the other of these assumptions makes it necessary to rethink the fundamental issues underlying these semantic conceptions.

I will focus on the truth-conditional conception. Much of the support it has enjoyed comes from the familiarity of the possible worlds machinery plus the fact that the semantic content of a sentence (relative to a context) should determine the possible worlds in which it is true. However, there is a big difference between admitting that semantic content determines such truth conditions and claiming that it should be identified with them. What we need is some conception of semantics in which the content of a sentence determines, but is not determined by, the metaphysically possible worlds in which it is true.

There are two main ways in which such a conception might be developed. One way is to retain the basic assumption, A1a, of the truth-conditional conception, while rejecting the characterization of truth conditions, or truth-supporting circumstances, as metaphysically possible worlds. The idea is to try and find some more finely grained circumstances that will distinguish among sentences true in the same worlds. The second way in which an appropriate semantic account might be developed is to give up A1a, thereby abandoning the fundamental tenet of the truth-conditional conception. In its place, one might substitute a conception of semantic contents as complex objects that encode much of the structure of the sentences that express them, and that determine sets of truth-supporting circumstances, without being identified with them.

In what follows, I will argue for the second approach. The heart of my argument involves the interaction of propositional attitudes with the phenomenon of direct reference. Let us say that a singular term is directly referential if its semantic content relative to a context (and assignment of values to variables) is its referent relative to the context (and assignment). Variables are the paradigm examples of such terms. In recent years, a number of arguments have been given for treating names and indexicals as directly referential as well. Later, I will show how this view can be defended against certain objections based on the behavior of such terms in propositional attitude ascriptions. To begin with, however, I wish to note the destructive consequences it has when added as a fourth assumption to A1–A3.

A4. Names, indexicals, and variables are directly referential.

This expanded set of assumptions has a number of clearly unacceptable consequences. Suppose, for example, that Mary assertively utters (4a) while pointing at me. On the assumptions we are considering, she cannot correctly be reported to believe, or to have said, that I am David Kaplan.

(4) a. He is David Kaplan. (Said pointing at Scott.)

b. Mary says (believes) that he (Scott) is David Kaplan.

The reason for this is that the semantic content of the complement sentence, relative to the context, is taken to be the set of metaphysically possible worlds in which two distinct objects are absolutely identical with one another—that is, the empty set. But then the third difficulty noted above—the impossibility of saying or believing the impossible—comes into play, ruling out the possibility that Mary said or believed what she seemed to say and believe. The same problem arises in a variety of cases, including those in (5).

(5) a. John says (believes) that Ruth Marcus is Ruth Barcan's sister.

b. Martin says (believes) that this table is made up of atomic particles with properties P, Q, and R. (Where it is later discovered that nothing made of such particles could be a table.)

The significance of these difficulties is not that they mar an otherwise unproblematic account of the attitudes. As we have seen, the conjunction of A1–A3 is problematic in its own right. Nevertheless, the difficulties arising from the addition of A4 are special.

I will argue that these difficulties are intractable for theories that identify semantic contents of sentences with sets of truth-supporting circumstances. Although many of the problems encountered in standard, truth-theoretic accounts of the attitudes can be avoided by substituting fine-grained circumstances for metaphysically possible worlds, those posed by names and indexicals cannot. Not only do these problems resist such treatment, they remain even when assumptions A2, A3, and A4 are weakened substantially. In effect, directly referential singular terms can be used to show that semantic contents of sentences (relative to contexts) cannot be sets of truth-supporting circumstances, no matter how fine-grained.

The reason for this is that such terms require the introduction of structure into semantic contents. After establishing this, I will consider two different ways in which such structure might be constructed—one based on a modified version of the truth-theoretic approach, the other based on the introduction of structured, Russellian propositions. Although considerations involving directly referential singular terms are insufficient to decide between these alternatives, I shall argue that additional factors favor the Russellian approach. Thus, the end result is an argument for an expanded conception of semantics that includes Russellian propositions as semantic contents of sentences, over and above standard, truth-theoretic intensions and extensions.

2.

Let us begin with the strategy of substituting fine-grained truth-supporting circumstances for metaphysically possible worlds. These circumstances can be thought of as arising from the relaxation of certain constraints that hold for such worlds. Taking a cue from Carnap's notion of a state description, we can describe these constraints in terms of their role in constructing a semantics for a language L.

Let D be the set of individuals L is used to talk about, and B be the set of properties expressed by simple predicates of L plus their complements. Let us say that a C-description is a set each of whose members consists of an n-place property plus an n-tuple of objects drawn from D (for variable n). A C-description X is complete iff it contains a complete assignment of objects to properties—i.e., iff for every n-place property P in B, and every o₁, ..., o_n in D, either [P,o₁, ..., o_n] is a member of X or [[- P],o₁, ..., o_n] is a member of X, where [- P] is the complement of P. A C-description X is consistent iff no two of its members are negations of one another—i.e., iff for every n-place property P in B, [P, o₁, ..., o_n] is a member of X only if [[- P],o₁, ..., o_n] is not a member of X. A C-description is metaphysically possible only if it is metaphysically possible for the objects mentioned in the description to (jointly) instantiate the properties they are paired with in the description.

For present purposes, truth-supporting circumstances might either be identified with C-descriptions, or be taken to correspond to them. The classifications "complete," "consistent," and "metaphysically possible" can then be applied to circumstances.

Metaphysically possible worlds are truth-supporting circumstances that are metaphysically possible, complete, and consistent. Suppose the first of these constraints is relaxed, while we retain the second and third. This allows truth-supporting circumstances corresponding to every consistent and complete C-description. Thus, we allow metaphysically impossible circumstances in which Ruth Marcus is Ruth Barcan's sister, 2 to the ninth is not 512, and I am identical with David Kaplan ('=' being treated as a simple, nonlogical predicate in the object language). In effect, we substitute what might be called "logically possible" worlds or circumstances for "metaphysically possible" worlds or circumstances.

However, the structure of the semantic theory remains the same as before. It continues to be a recursive characterization of truth relative to a context and circumstance, with the recursive clauses retaining their standard specifications. The semantic content of a sentence relative to a context is identified with the set of circumstances in which it is true. But since these circumstances are more finely grained than metaphysically possible worlds, we no longer have the results that metaphysically equivalent sentences have the same semantic content, that distribution of a propositional attitude verb over conjunction requires closure of the attitude under metaphysically necessary consequence, or that no one can believe or assert the metaphysically impossible. In this way, substitution of A1b for A1b might be seen as alleviating the original difficulties with A1–A4.

(Continues...)

Excerpted from PHILOSOPHICAL ESSAYSby Scott Soames Copyright © 2009 by Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Details zur Ausgabe

• Verlag: Princeton University Press
• Erscheinungsdatum: 2009
• ISBN 10: 0691136831
• ISBN 13: 9780691136837
• Einband: Tapa blanda
• Sprache: Englisch
• Anzahl der Seiten: 474
• Kontakt zum Hersteller: Nicht verfügbar