Hardcover
-
PICK UP IN STORECheck Availability at Nearby Stores
Available within 2 business hours
Related collections and offers
Overview
Originally published in 1992.
The Princeton Legacy Library uses the latest print-on-demand technology to again make available previously out-of-print books from the distinguished backlist of Princeton University Press. These editions preserve the original texts of these important books while presenting them in durable paperback and hardcover editions. The goal of the Princeton Legacy Library is to vastly increase access to the rich scholarly heritage found in the thousands of books published by Princeton University Press since its founding in 1905.
Product Details
| ISBN-13: | 9780691632384 |
|---|---|
| Publisher: | Princeton University Press |
| Publication date: | 04/19/2016 |
| Series: | Princeton Legacy Library , #194 |
| Pages: | 302 |
| Product dimensions: | 6.20(w) x 9.30(h) x 0.90(d) |
Read an Excerpt
A Prosentential Theory of Truth
By Dorothy Grover
PRINCETON UNIVERSITY PRESS
Copyright © 1992 Princeton University PressAll rights reserved.
ISBN: 978-0-691-07399-6
CHAPTER 1
Introductory Essay
At the beginning of the century, it was common to write about the nature of truth. Correspondence theorists advocated an analysis of truth in terms of a correspondence relation between linguistic entities, or beliefs, and extralinguistic entities. Pragmatists have been described as reducing truth to "what works" or to "what scientists are destined to agree upon"; and coherence theorists offered analyses in terms of consistency and comprehensiveness. Frequently different schools of philosophical thought are associated with these theories, e.g., realists are often identified as correspondence theorists, while instrumentalists and anti-realists are assumed to endorse a pragmatic account of truth. In §1,1 present a brief survey of some versions of these theories of the nature of truth, thereby providing the background against which the prosentential theory of truth was presented.
In §2 I review the papers on the prosentential theory included in this volume. A principal claim of the prosentential theory of truth is that truth talk—all the truth talk we need—can be explained without appeal to any kind of analysis of the nature of truth. I claim the truth predicate helps facilitate certain kinds of discourse and that this role can be explained in terms of the concept of a "prosentence." Joseph Camp, Nuel Belnap, and I developed this idea in part by drawing analogies with bound variables in formal languages and pronouns in English and in part by exploring the philosophical implications of this prosentential characterization. For example, with respect to the implications, I show that the prosentential theory offers a simple account of "paradoxical" expressions like the Liar sentence ('This is false') according to which they are not contradictory. The prosentential theory can also explain, without appeal to a substantive truth property, cases where philosophers have mistakenly claimed an explanatory role for truth.
In §3 I consider the implications of the prosentential theory for the so-called correspondence, pragmatic, and coherence theories of truth. In denying that we need an analysis of the nature of truth, the prosentential theory denies a basic assumption of these three theories as they have been traditionally construed. I show that the prosentential theory offers a new slant on some issues over which these theories contend. I show that among the diverse theories that have been classified as theories of truth, there are theories that have something to say about the nature of truth (Russell's correspondence theories), theories that have something to say about meaning (again, correspondence theories), theories that have something to say about the truth predicate (Tarski and the prosentential theory), theories that have something to say about what-is-true (some pragmatic and coherence theories), and theories that have something to say about epistemological issues concerning what-is-true (again, some pragmatic and coherence theories). These last two cases need further explanation because we are going to read 'true' in 'what-is-true' prosententially.
It emerges that a prosentential theorist has the option of endorsing significant aspects of the traditional theories. For example, even though the nature-of-truth issue is eliminated on the prosentential scheme, she has the option of endorsing the idea that language-world connections may have an explanatory role. Also forthcoming is an explanation of how a philosopher (like Quine) can endorse a so-called deflationary account of the truth predicate while at the same time endorsing views that resemble those we associate with coherence theorists.
1. On "the Nature of Truth"
Correspondence theorists seek an analysis of truth that will capture the idea that "true sentences agree with reality" (or the idea that "it is the world that makes sentences true"). A correspondence theory of truth typically presents truth as a relational property: Truth is a correspondence relation that connects bearers of truth and falsity (e.g., sentence types or tokens, propositions, or belief states) with extralinguistic entities (e.g., facts) with which the bearers correspond. Though the intuitions behind this theory appealed to many, the problem of providing an adequate account of the correspondence relation, and of identifying the two kinds of entities between which the correspondence relation is alleged to obtain, has proved recalcitrant.
Russell (1906–07), for example, struggled with several versions of the correspondence theory. First there was the idea that beliefs or sentences are true if there are facts in the world that make them true. The simplest version of this theory is the existence theory—so-called by Prior (1967). The existence theory treats sentences a bit like names, except that the objects with which they are related are facts, not objects. For example, 'The cat is on the mat' is true, it is said, if the world contains the fact that the cat is on the mat. A major problem for this theory is posed by false beliefs (or sentences), as both Plato (in the Theaetetus) and Russell realized. For example, Russell, in "On the Nature of Truth," says, "When (beliefs] do correspond, the beliefs are true, and are beliefs in facts; when they do not, the beliefs are erroneous, and are beliefs in nothing." The problematic question is, How is false belief possible? Indeed, falsity poses a variety of challenges for correspondence theorists.
Modifications of the existence theory were introduced that yielded slightly more satisfactory versions of a correspondence theory. At various stages Russell first tried adding to his ontology objective falsehoods and then negative facts. The idea was to have something with which false sentences would correspond. But the problems do not go away without much more explanation. For suppose a situation where there is a dog on the mat and the cat is up the tree: With what does 'The cat is on the mat' correspond? The dog's being on the mat? The cat's being up the tree? Or does it correspond with the objective falsehood, the cat's being on the mat, or with the negative fact, the cat's not being on the mat? What are facts, objective falsehoods, or negative facts? I have not (yet) seen an account of any of these that leads to a satisfactory analysis of "the nature of truth."
Both Russell and Plato seemed to think the answer might lie in the grammar of the sentences used in expressing beliefs. For then a sentence can be both about something—the subject of the sentence (the cat)—yet false (not on the mat). Russell himself recognized that saying only this much does not yield an account of the correspondence relation. What is the correspondence relation?
A "picture theory" of truth (or meaning) may be one way to make sense of the correspondence relation. For one is supposed to just see that a linguistic entity pictures the object with which it corresponds. Unfortunately people do not just see what a picture is a picture of; furthermore, what would a "false picture" (e.g., the false sentence 'the cat is on the mat' in the case where the dog is on the mat and the cat is up a tree) be a picture of? We need an account of 'picture of.'
Some recent proposals have tried to wrestle a correspondence theory out of Tarski's truth definition. I know of three such attempts. However, because two of these accept Tarski's (1936) definition (or something like it) as providing all there is to know about truth, they promote only a "deflationary" account of truth. For this reason, I will review those two versions in §3.
Other philosophers have thought there is more to the concept of truth than is captured by a Tarski-style truth definition. For example, Dummett (1978) complains that a Tarskian definition fails to give an account of the point of the predicate. He says, "What has to be added to a truth-definition for the sentences of a language, if the notion of truth is to be explained, is a description of the linguistic activity of making assertions." (p. 20) Davidson (1990) claims that, in addition to "the formal properties of the concept," we need "to indicate how a theory of truth can be applied to particular speakers or groups of speakers." (p. 314) Although I agree that we need an account of assertion and linguistic activity generally, I doubt that such an account has much to do with "the truth concept" itself. One thing I think we need beyond what Tarski has given us is an understanding of the utility of the truth predicate. This should help us understand how so-called theories of truth have been many and various. This supplementation is different from that which others have described.
Though Tarski claimed to capture the intuitions of the correspondence theory, most who call themselves correspondence theorists also think we need more on the truth concept than a Tarskian definition provides. For example, Sellars (1962) developed a version of the picture theory as supplementation to Tarski.
Perhaps the most ambitious, most provocative, way of wrestling a correspondence theory out of Tarski has been suggested by Hartry Field (1972). Field is one among those who have claimed Tarski has failed to capture much that is important about truth. A problem with Tarski's definition, according to Field, is that it defines truth for only one language at a time. He says that if we are to have an account of truth that applies to all languages, we must add, at the level of the base clauses, physicalistically acceptable reductions of 'refers' and 'applies'. This, Field claims, would yield a definition that incorporates the kinds of causal language-world relations that correspondence theorists think we need.
Field's most recent argument for an explanatory role is based on a complex case. He speculates that correspondence truth is required to explain a certain kind of systematic reliability that an individual person may have concerning a certain topic or group of topics. An example he appeals to is that of an expert's reliability on social movements, including the Kronstadt rebellion. Field says: "We will need an explanation of the striking correlation between her belief-states and the events involving the Kronstadt sailors. It seems likely the explanation would have to involve something like causal networks of information...." Let us grant Field's point, that some kind of causal network is needed to explain T-reliability. What does this have to do with truth? Field continues:
A familiar proposal for giving a correspondence theory of truth (perhaps most explicit in Field 1972) is to explain the truth conditions of sentences in terms of reference or reference-like relations for the basic parts of the sentence, and then to give explanations of the reference-like relations for the parts; and it is often assumed that something like causal networks of information of the sort just alluded to will play a role in the theories of reference for the parts; ... a central feature of the proposal is to build into the theoretical explication of correspondence truth just the sort of thing needed to explain the striking T-reliability and converse T-reliability that agents have (about some matters), under the interpretations that we find useful in explaining their success. (Field 1986, p. 104)
Field's point is that the language-world connections he proposes incorporating, in his version of the correspondence theory, are needed to explain T-reliability.
I have argued elsewhere that Field has failed to establish his point because he has not really given truth itself an explanatory role. The primary problem is that in Field's scheme the causal languageworld connections are the same for true and false sentences, because the language-world connections of causal-historical theories of reference (the connections he suggests utilizing in unpacking 'refers' and 'applies') are the same for true and false sentences; indeed, they are also the same for questions and commands. On Field's scheme, a difference between true and false sentences arises in a clause that says something about only extralinguistic reality. This "difference" is captured by T-sentences. Because no causal relations are uniquely associated with truth, truth as a substantive language-world relation does not have its own explanatory role.
What Field's suggestion would perhaps show—if Field's reductions of 'refers' and 'applies' were successful—is that the language-world connections of reference have an explanatory role. But even then, the most such connections might contribute is something toward an explanation of the conditions under which people become articulate, or perhaps an explanation of the content of a person's beliefs. They can hardly explain "the striking T-reliability and converse T-reliability that agents have (about some matters)"; other kinds of causal connections are needed for that. Of course one can call any kind of explanatory relation a correspondence truth relation; however, Field does suggest ties with traditional correspondence theories, and traditional theories have different language-world connections for true and false sentences. That is what one would expect if truth, as a language-world relation, is to have an explanatory role. So Field's claim that his correspondence theory gives truth an explanatory role is questionable.
I am inclined to think that those sympathetic to Field's concerns should separate an explanatory role for reference from the issue of truth. After all, in Field's case, the only reason for tacking truth onto some kind of reference theory seems to be to give truth an explanatory role. Given that reference connections provide the substance of the explanation Field envisages, there is really no reason for claiming that truth itself provides the explanation, unless the motivation is to save the correspondence theory.
By the end of his paper (1986), Field himself entertains the possibility of separating the issues, stating, "Indeed, explanations of those phenomena (e.g., the 'success phenomena' ... or the fact that an agent is highly T-reliable and converse T-reliable under certain specific interpretations) is clearly important, whether or not one thinks that it will lead to a correspondence theory of truth." (p. 105) Michael Devitt (1984) also appeals to Field's suggestions as to the structure of a correspondence theory of truth. In his attempt to establish an explanatory role for truth, Devitt presents a case where he thinks we need to ascribe wide content to speakers. Though I find this a helpful discussion in general, Devitt's concern is really with establishing an explanatory role for reference connections—as it was for Field.
Perhaps there are other contemporary philosophers who have worked out the details of a correspondence theory, but I am not aware of them. Indeed, in discussions of scientific and moral realism and in epistemology, there are many references to "truth" where a correspondence theory is assumed, but the writers do not describe the details of the correspondence theory (the relation and the relata) they have in mind. There are also references to "realist truth" by those who presumably think correspondence truth is constitutive of realism. But again, I am not sure what these philosophers have in mind as a correspondence theory.
Why did philosophers think it important to develop theories of "the nature of truth"? We have seen that Field argues we need such a theory because he thinks truth has an explanatory role. In our terms, we might think of Russell as another who at least sometimes assumed that truth has an explanatory role. For example, Russell sometimes raised questions about truth in the context of discussions of knowledge. In his view (and Plato's also), an account of knowledge called for an account of the difference between true and false belief. I presume Russell thought an account of the nature of truth would lead to an account of this difference. Russell's investigation of language and meaning also led him to take up the issue of "the nature of truth."
(Continues...)
Excerpted from A Prosentential Theory of Truth by Dorothy Grover. Copyright © 1992 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.
Table of Contents
Preface
1 Introductory Essay 3
1. On "the Nature of Truth" 4
2 The Prosentential Theory - Review and Reflections 15
3 Implications for Some Other Theories of Truth 27
4 Final Reflections 44
2 Propositional Quantifiers 46
A Preview 46
1 Introduction 47
2 A Language in Which Sentences Are the Substituends of Propositional Variables 48
3 A Language in Which Propositional Terms Are the Substituends of Propositional Term Variables 63
4 Summary 68
3 A Prosentential Theory of Truth 70
1 Ramsey 71
2 Prosentential Theory: Exposition 80
3 Prosentential Theory: Objections 97
4 Consequences and Applications 105
4 Inheritors and Paradox 121
1 Grounded Pronouns 124
2 Inheritors 125
3 Ungrounded Inheritors 126
4 Grounded Prosentences 127
5 Sentential Inheritors 127
6 Ungrounded Inheritors 128
7 Paradoxes 129
8 Generalized Versions of the Liar 129
9 Grounded/Ungrounded 131
10 A Cause of "Paradox" 132
11 A Strengthened Version of the Liar 133
12 Semantic and Logical Paradoxes 136
5 Prosentences and Propositional Quantification: A Response to Zimmerman 137
1 Prosentences 138
2 Propositional Quantification 140
3 Irredundant Uses of 'True' 143
6 Truth 146
1 Two Sources of Scepticism 152
2 Extensions for 'True' and 'False' 155
3 Would a Property-Ascribing 'True' Be Useful? 156
4 Review 164
5 Propositional Quantification in English 167
7 Truth: Do We Need It? 173
1 Problems of Expressibility 174
2 Is Truth a Property? 175
3 The Prosentential Theory 178
4 A Bit of Characterizing 181
5 Meaning and Truth 185
6 Logic and Truth 189
7 On 'Neither True nor False' 193
8 Properties Reconsidered 204
8 Berry's Paradox 207
1 Inheritors 208
2 Berry's Paradox 211
3 Variations on Berry 213
9 On Two Deflationary Truth Theories 215
A Preview 216
1 The Disquotational Theory 216
2 The Prosentential Theory 218
3 Generalization - A Comparison 222
4 Metalinguistic Uses of the Truth Predicate 225
5 Disquotation with Prosentences 230
6 Summary 233
10 Propositional Quantification and Quotation Contexts 234
1 Introduction 234
2 Grammar 235
3 Hierarchies 236
4 Semantics 238
5 Discussion 242
11 Quantifying in and out of Quotes 244
1 Logic as an Organon 244
2 0[subscript 2] and M[subscript 2] Completely Separate 249
3 0[subscript 3] a Proper Part of M[subscript 3] 260
4 0[subscript]4 and M[subscript 4] Overlap 275
Bibliography 277
Index 285