We set out a doctrine about truth for the statements of mathematics—a doctrine which we think is a worthy competitor to realist views in the philosophy of mathematics—and argue that this doctrine, which we shall call ‘mathematical relativism’, withstands objections better than do other non-realist accounts.

The idea underlying the Begriffsschrift account of identities was that the content of a sentence is a function of the things it is about. If so, then if an identity a=b is about the content of its contained terms and is true, then a=a and a=b have the same content. But they do not have the same content; so, Frege concluded, identities are not about the contents of their contained terms. The way Frege regarded the matter is that in an (...) identity the terms flanking the symbol for identity do not have their ordinary contents, but instead have themselves as their contents. In ?Uber Sinn und Bedeutung? Frege became convinced that if an identity a=bis about the signs aand b, then it expresses no proper knowledge. So, since it is evident that many such identities do express proper knowledge, Frege concluded that identities are not about their contained signs. So he became convinced that his Begriffsschrift account was incorrect. What was the error in the argument that led Frege to that account? It was thinking that the content of a sentence is a function of the contents of its constituent signs, that is, the things it is about. (shrink)

This book says Prior claims: (1) that a sentence never names; (2) what a sentence says cannot be otherwise signified; and (3) that a sentence says what it says whatever the type of its occurrence; (4) and that quantifications binding sentential variables are neither eliminable, substitutional, nor referential. The book develops and defends (1)-(3). It also defends (4) against the sorts of strictures on quantification of such philosophers as Quine and Davidson.

A popular view is that the great discovery of Pythagoras was that there are irrational numbers, e.g., the positive square root of two. Against this it is argued that mathematics and geometry, together with their applications, do not show that there are irrational numbers or compel assent to that proposition.

Conversational implicatures are easy to grasp for the most part. But it is another matter to give a rational reconstruction of how they are grasped. We argue that Grice's attempt to do this fails. We distinguish two sorts of cases: (1) those in which we grasp the implicature by asking ourselves what would the speaker have to believe given that what he said is such as is required by the talk exchange; (2) those in which we grasp the implicature by (...) asking ourselves why it is that what the speaker said is so obviously not such as is required by the talk exchange. We argue that Grice's account does not fit those cases falling under (2). (shrink)

The significance of the semantical paradoxes for natural languages is examined. If Tarski’s reflections on the issue are correct, English is inconsistent. Paul Ziff responds to Tarskian reflections by arguing to the conclusion that no natural language is or can be inconsistent. The authors reject Ziff’s argument, but they defend something similar to its conclusion: no language, natural or otherwise, is or can be inconsistent in the way that Tarski holds languages capable of formulating the Epimenides are inconsistent.

The purpose of this paper is to uncover and correct several confusions about expressions, tokens and the relations between them that crop up in even highly sophisticated writing about language and logic.

It is argued that if there are truth-value gaps then the disquotational theory of truth is false. Secondly, it is argued that the same conclusion can be reached even without the assumption that there are truth-value gaps.

In this paper the authors recapitulate, justify, and defend against criticism the extension of the redundancy theory of truth to cover a wide range of uses of ‘true’ and ‘false’. In this they are guided by the work of A. N. Prior. They argue Prior was right about the scope and limits of the redundancy theory and that the line he drew between those uses of ‘true’ which are and are not susceptible to treatment via redundancy serves to distinguish two (...) important and mutually irreducible types of truth: redundancy truth and predicative truth. Only the latter serves for semantic theorizing. (shrink)

A doctrine that occurs intermittently in Quine’s work is that there is no extra-theoretic truth. This paper explores this doctrine, and argues that on its best interpretation it is inconsistent with three views Quine also accepts: bivalence, mathematical Platonism, and the disquotational account of truth.

In this book a non-realist philosophy of mathematics is presented. Two ideas are essential to its conception. These ideas are (i) that pure mathematics--taken in isolation from the use of mathematical signs in empirical judgement--is an activity for which a formalist account is roughly correct, and (ii) that mathematical signs nonetheless have a sense, but only in and through belonging to a system of signs with empirical application. This conception is argued by the two authors and is critically discussed by (...) three philosophers of mathematics. (shrink)

Here we give a semantical account of propositional quantification that is intended to formally represent Russell’s view that one cannot express a proposition about "all" propositions. According to the account the authors give, Russell’s view bears an interesting relation to the view that there are no sets which are members of themselves.

Anderson and Belnap devise a model theory for entailment on which propositional identity equals proposional coentailment. This feature can be reasonably questioned. The authors devise two extensions of Anderson and Belnap’s model theory. Both systems preserve Anderson and Belnap’s results for entailment, but distinguish coentailment from identity.

The standard response is illustrated by E, J. Lemmon's claim that if all objects in a given universe had names and there were only finitely many of them, then we could always replace a universal proposition about that universe by a complex proposition. It is because these two requirements are not always met that we need universal quantification. This paper is partly in agreement with Lemmon and partly in disagreement. From the point of view of syntax and semantics we can (...) replace a universal proposition about any universe (finite or infinite, countable or uncountable) by a complex proposition (= sentence built up from atomic sentences and the connectives). But from the point of view of communication such a replacement is not possible if the universe is infinite. (shrink)

This paper deals with the question of whether there is objectivist truth about set-theoretic matters. The dogmatist and skeptic agree that there is such truth. They disagree about whether this truth is knowable. In contrast, the relativist says there is no objective truth to be known. Two versions of relativism are distinguished in the paper. One of these versions is defended.

Robert cummins has recently attacked this line of argument: if p is a semantically primitive predicate of a first order language l, then p requires its own clause in the definition of satisfaction integral to a definition of truth of l. thus if l has infinitely many such p, the satisfaction clause cannot be completed and truth for l will remain undefined. against this cummins argues that a single clause in a general base theory for l can specify satisfaction conditions (...) for even infinitely many semantically primitive predicates of l. against cummins we argue that a general base theory contains a primitive expression 'applies to' and that to make his case cummins would need to prove this expression stands for a semantical relation sufficient to a truth definition for l. not only is such a proof lacking but we show the claim to be altogether dubious. (shrink)

The following syllogism is considered: a string is not an expression unless it is tokenable; no one can utter, write, or in anyway token an infinite string; so no infinite string is an expression. The second premise is rejected. But the tokenability of an infinite sentence is not sufficient for it being an infinite expression. A further condition is that no finite sentence expresses that sentence’s truth-conditions. So it is an open question whether English contains infinite expressions.

Let C1 and C2 be distinct moral codes formulated in English. Let C1 contain a norm N and C2 its negation. The paper construes the moral relativist as saying that if both codes are consistent, then, in the strongest sense of correctness applicable to moral norms, they are also both correct in the sense that they contain only correct moral norms. If we believe that the physical statements of English are true (false) in English, we will reject an analogous statement (...) made of physical theories. We will hold that the strongest sense of correctness applicable to physical statements is not system-relative. The moral relativist denies that there is any corresponding sense of correctness applicable to moral norms. That is, there is no notion of moral correctness that is not system-dependent. It is argued that, while the position may not be true, there is not a strictly logical basis for refuting it. (shrink)

In this paper the authors argue that if Tarski’s definition of truth for the calculus of classes is correct, then set theories which assert the existence of proper classes (classes which are not the member of anything) are incorrect.

Peter Geach proposed a substitutional construal of quantification over thirty years ago. It is not standardly substitutional since it is not tied to those substitution instances currently available to us; rather, it is pegged to possible substitution instances. We argue that (i) quantification over the real numbers can be construed substitutionally following Geach's idea; (ii) a price to be paid, if it is that, is intuitionism; (iii) quantification, thus conceived, does not in itself relieve us of ontological commitment to real (...) numbers. (shrink)

A strong and weak version of the redundancy theory of truth are distinguished. An argument put forth by Michael Dummett concludes that the weak version is vitiated by truth-value gaps. The weak version is defended against this argument. The strong version, however, is vitiated by truth-value gaps.

As a way of dealing with the semantical paradoxes Quine has suggested: that semantical expressions such as ‘true’ and ‘true of’ be used with numerical subscripts; that when a truth locution T is applied to a sentence S, the subscript on T is greater than any within S; otherwise, the result of applying T to S is ill formed. A problem is that this introduces infinitely many semantical primitives. The paper suggests a way around the problem. The paper raises a (...) further problem, leaving it open whether this further problem has a satisfactory solution. (shrink)

Whereas arithmetical quantification is substitutional in the sense that a some-quantification is true only if some instance of it is true, it does not follow (and, in fact, is not true) that an account of the truth-conditions of the sentences of the language of arithmetic can be given by a substitutional semantics. A substitutional semantics fails in a most fundamental fashion: it fails to articulate the truth-conditions of the quantifications with which it is concerned. This is what is defended in (...) the paper. In particular, it is defended against remarks to the contrary in a well known paper on the subject. (shrink)

Let A, B, C stand for sentences expressing propositions; let A be a component of C; let C A/B be just like C except for replacing some occurrence of A in C by an occurrence of B; let = be a binary connective for propositional identity read as ‘the proposition that __ is the very same proposition as …’. Then authors defend adding ‘from C = C A/B infer A = B’ to Prior’s rules for propositional identity, appearing in OBJECTS (...) OF THOUGHT. (shrink)

It may be that all that matters for the modalities, possibility and necessity, is the object named by the proper name, not which proper name names it. An influential defender of this view is Saul Kripke. Kripke’s defense is criticized in the paper.

Frege held that the result of applying a predicate to names lacks reference if any of the names lack reference. We defend the principle against a number of plausible objections. We put forth an account of consequence for a first-order language with identity in which the principle holds.

A case against Prior’s theory of propositions goes thus: (1) everyday propositional generalizations are not substitutional; (2) Priorean quantifications are not objectual; (3) quantifications are substitutional if not objectual; (4) thus, Priorean quantifications are substitutional; (5) thus that Priorean quantifications are not ontologically committed to propositions provides no basis for a similar claim about our everyday propositional generalizations. Prior agrees with (1) and (2). He rejects (3), but fails to support that rejection with an account of quantification on which there (...) could be quantifications that are neither substitutional nor objectual. The paper draws from the work of Lorenzen an alternative conception of quantification in terms of which that needed account can be given. (shrink)

In a series of papers R. L. Martin propounds a theory for dealing with the semantical paradoxes. This paper is a criticism of that theory.

It is argued that Wittgenstein’s remarks 6.2-6.22 Tractatus fare well when one focuses on non-quantificational arithmetic, but they are problematic when one moves to quantificational arithmetic.

A common assumption among philosophers is that every language has at most denumerably many expressions. This assumption plays a prominent role in many philosophical arguments. Recently formal systems with indenumerably many elements have been developed. These systems are similar to the more familiar denumerable first-order languages. This similarity makes it appear that the assumption is false. We argue that the assumption is true.

An account of the logic of bivalent languages with truth-value gaps is given. This account is keyed to the use of tables introduced by S. C. Kleene. The account has two guiding ideas. First, that the bivalence property insures that the language satisfies classical logic. Second, that the general concepts of a valid sentence and an inconsistent sentence are, respectively, as sentences which are not false in any model and sentences which are not true in any model. What recommends this (...) approach is (1) its relative simplicity, and (2) the fact that it leaves the fundamental features of classical logic intact. (shrink)

The aim of this paper is to provide a nondenotational semantics for first-order languages which will match one for one each distribution of truth-values available in terms of a denotational semantics.

In Tractatus, Wittgenstein held that there are null sentences – prominently including logical truths and the truths of mathematics. He says that such sentences are without sense (sinnlos), that they say nothing; he also denies that they are nonsensical (unsinning). Surely it is what a sentence says which is true or false. So if a sentence says nothing, how can it be true or false? The paper discusses the issue.

For Rudolf Carnap the question ‘Do numbers exist?’ does not have just one sense. Asked from within mathematics, it has a trivial answer that could not possibly divide philosophers of mathematics. Asked from outside of mathematics, it lacks meaning. This paper discusses Carnap ’s distinction and defends much of what he has to say.

Fundamental to Quine’s philosophy of logic is the thesis that substitutional quantification does not express existence. This paper considers the content of this claim and the reasons for thinking it is true.