This chapter argues that while quotation marks are polysemous, the thread that runs through all uses of quotation marks that involve reference to expressions is pure quotation, in which an expression formed by enclosing another expression in quotation marks refers to that enclosed expression. We defend a version of the so-called disquotational theory of pure quotation and show how this device is used in direct discourse and attitude attributions, in exposition in scholarly contexts, and in so-called mixed quotation in indirect (...) discourse and attitude attributions. We argue that uses of quotation marks that extend beyond pure quotation have two features in common. First, the expressions appearing in quotation marks are intended to be understood, and that they are intended to be understood is essential to the function that such quotations play in communication, though this does not always involve the expressions contributing their extensional properties to fixing truth conditions for the sentences in which they appear. Second, they appeal to a relation to the expression appearing in quotation marks that plays a role in determining the truth conditions of the sentences in which they appear. (shrink)

In his classic 1936 essay "On the Concept of Logical Consequence", Alfred Tarski used the notion of satisfaction to give a semantic characterization of the logical properties. Tarski is generally credited with introducing the model-theoretic characterization of the logical properties familiar to us today. However, in his book, The Concept of Logical Consequence, Etchemendy argues that Tarski's account is inadequate for quite a number of reasons, and is actually incompatible with the standard model-theoretic account. Many of his criticisms are meant (...) to apply to the model-theoretic account as well. In this paper, I discuss the following four critical charges that Etchemendy makes against Tarski and his account of the logical properties: (1) (a) Tarski's account of logical consequence diverges from the standard model-theoretic account at points where the latter account gets it right. (b) Tarski's account cannot be brought into line with the model-theoretic account, because the two are fundamentally incompatible. (2) There are simple counterexamples (enumerated by Etchemendy) which show that Tarski's account is wrong. (3) Tarski committed a modal fallacy when arguing that his account captures our pre-theoretical concept of logical consequence, and so obscured an essential weakness of the account. (4) Tarski's account depends on there being a distinction between the "logical terms" and the "non-logical terms" of a language, but (according to Etchemendy) there are very simple (even first-order) languages for which no such distinction can be made. Etchemendy's critique raises historical and philosophical questions about important foundational work. However, Etchemendy is mistaken about each of these central criticisms. In the course of justifying that claim, I give a sustained explication and defense of Tarski's account. Moreover, since I will argue that Tarski's account and the model theoretic account really do come to the same thing, my subsequent defense of Tarski's account against Etchemendy's other attacks doubles as a defense against criticisms that would apply equally to the familiar model-theoretic account of the logical properties. (shrink)

This paper concerns a key point of decision in Donald Davidson's early work in philosophy of language — a fateful decision that set him and the discourse in the area on the path of truth-theoretic semantics. The decision of moment is the one Davidson makes when, in the face of a certain barrier, he gives up on the idea of constructing an explicit meaning theory that would parallel Tarski's recursive way with truth theory. For Davidson there was little choice: he (...) tells us he does not see how to deal with the difficulty except in the radical way he proposes. But there is a way to give such a meaning theory — a meaning theory proper which, using classical logic only, meets a meaning-theoretic analogue of Convention T, satisfies Davidson’s three key desiderata for a theory of meaning, reflects linguistic competence, and avoids quantifying over meanings. The meaning theory sketched here uses Tarskian strategies, as Davidson proposed, but differs from Davidson's approach in not going by way of a truth theory for the target language. (shrink)

In this paper, we outline an approach to giving extensional truth-theoretic semantics for what have traditionally been seen as opaque sentential contexts. We outline an approach to providing a compositional truth-theoretic semantics for opaque contexts which does not require quantifying over intensional entities of any kind, and meets standard objections to such accounts. The account we present aims to meet the following desiderata on a semantic theory T for opaque contexts: (D1) T can be formulated in a first-order extensional language; (...) (D2) T does not require quantification over intensional entities­i.e., meanings, propositions, properties, relations, or the like­in its treatment of opaque contexts; (D3) T captures the entailment relations that hold in virtue of form between sentences in the language for which it is a theory; (D4) T has a finite number of axioms. If the approach outlined here is correct, it resolves a longstanding complex of problems in metaphysics, the philosophy of mind and the philosophy of language. (shrink)

A sorites argument is a symptom of the vagueness of the predicate with which it is constructed. A vague predicate admits of at least one dimension of variation (and typically more than one) in its intended range along which we are at a loss when to say the predicate ceases to apply, though we start out confident that it does. It is this feature of them that the sorites arguments exploit. Exactly how is part of the subject of this paper. (...) The majority of philosophers writing on vagueness take it to be a kind of semantic phenomenon. If we are right, they are correct in this assumption, which is surely the default position, but they have not so far provided a satisfactory account of the implications of this or a satisfactory diagnosis of the sorites arguments. Other philosophers have urged more exotic responses, which range from the view that the fault lies not in our language, but in the world, which they propose to be populated with vague objects which our semantics precisely reflects, to the view that the world and language are both perfectly in order, but that the fault lies with our knowledge of the properties of the words we use (epistemicism). In contrast to the exotica to which some philosophers have found themselves driven in an attempt to respond to the sorites puzzles, we undertake a defense of the commonsense view that vague terms are semantically vague. Our strategy is to take fresh look at the phenomenon of vagueness. Rather than attempting to adjudicate between different extant theories, we begin with certain pre-theoretic intuitions about vague terms, and a default position on classical logic. The aim is to see whether (i) a natural story can be told which will explain the vagueness phenomenon and the puzzling nature of soritical arguments, and, in the course of this, to see whether (ii) there arises any compelling pressure to give up the natural stance. We conclude that there is a simple and natural story to be told, and we tell it, and that there is no good reason to abandon our intuitively compelling starting point. The importance of the strategy lies in its dialectical structure. Not all positions on vagueness are on a par. Some are so incredible that even their defenders think of them as positions of last resort, positions to which we must be driven by the power of philosophical argument. We aim to show that there is no pressure to adopt these incredible positions, obviating the need to respond to them directly. If we are right, semantic vagueness is neither surprising, nor threatening. It provides no reason to suppose that the logic of natural languages is not classical or to give up any independently plausible principle of bivalence. Properly understood, it provides us with a satisfying diagnosis of the sorites argumentation. It would be rash to claim to have any completely novel view about a topic so well worked as vagueness. But we believe that the subject, though ancient, still retains its power to inform and challenge us. In particular, we will argue that taking seriously the central phenomenon of predicate vagueness—the “boundarylessness” of vague predicates—on the commonsense assumption that vagueness is semantic, leads ineluctably to the view that no sentences containing vague expressions (henceforth ‘vague sentences’) are truth-evaluable. This runs counter to much of the literature on vagueness, which commonly assumes that, though some applications of vague predicates to objects fail to be truth-evaluable, in clear positive and negative cases vague sentences are unproblematically true or false. It is clarity on this, and related points, that removes the puzzles associated with vagueness, and helps us to a satisfying diagnosis of why the sorites arguments both seem compelling and yet so obviously a bit of trickery. We give a proof that semantically vague predicates neither apply nor fail-to-apply to anything, and that consequently it is a mistake to diagnose sorites arguments, as is commonly done, by attempting to locate in them a false premise. Sorites arguments are not sound, but not unsound either. We offer an explanation of their appeal, and defend our position against a variety of worries that might arise about it. The plan of the paper is as follows. We first introduce an important distinction in terms of which we characterize what has gone wrong with vague predicates. We characterize what we believe to be our natural starting point in thinking about the phenomenon of vagueness, from which only a powerful argument should move us, and then trace out the consequences of accepting this starting point. We consider the charge that among the consequences of semantic vagueness are that we must give up classical logic and the principle of bivalence, which has figured prominently in arguments for epistemicism. We argue there are no such consequences of our view: neither the view that the logic of natural languages is classical, nor any plausible principle of bivalence, need be given up. Next, we offer a diagnosis of what has gone wrong in sorites arguments on the basis of our account. We then present an argument to show that our account must be accepted on pain of embracing (in one way or another) the epistemic view of “vagueness”, i.e., of denying that there are any semantically vague terms at all. Next, we discuss some worries that may arise about the intelligibility of our linguistic practices if our account is correct. We argue none of these worries should force us from our intuitive starting point. Finally, we cast a quick glance at other forms of semantic incompleteness. (shrink)

Alfred Tarski (1944) wrote that "the condition of the 'essential richness' of the metalanguage proves to be, not only necessary, but also sufficient for the construction of a satisfactory definition of truth." But it has remained unclear what Tarski meant by an 'essentially richer' metalanguage. Moreover, DeVidi and Solomon (1999) have argued in this Journal that there is nothing that Tarski could have meant by that phrase which would make his pronouncement true. We develop an answer to the historical question (...) of what Tarski meant by 'essentially richer' and pinpoint the general result that stands behind his essential richness claim. In defense of Tarski, we then show that each of the several arguments of DeVidi and Solomon are either moot or mistaken. One of the fruits of our investigation is the reclamation of what Tarski took to be his central result on truth. This is a reclamation since: (i) if one does not understand 'essential richness', one does not know what that result is, and (ii) we must unearth a heretofore unrecognized change that occurs in Tarski's view - an alteration of his main thesis in light of a failing he discovered in it. (shrink)

I offer an interpretation of a familiar, but poorly understood portion of Tarskis work on truth – bringing to light a number of unnoticed aspects of Tarskis work. A serious misreading of this part of Tarski to be found in Scott Soames Understanding Truth is treated in detail. Soamesreading vies with the textual evidence, and would make Tarskis position inconsistent in an unsubtle way. I show that Soames does not finally have a coherent interpretation of Tarski. This is unfortunate, since (...) Soames ultimately arrogates to himself a key position that he has denied to Tarski and which is rightfully Tarskis own. (shrink)

Stephen Schiffer has argued that natural languages do not have compositional semantics. But it has been widely held that compositional semantics is required in order to explain how it is possible that we have the linguistic capacities that we do. In particular, our use of natural languages is productive in the sense that there are indefinitely many sentences that we have never heard or considered before, but which we are nonetheless capable of understanding. How is this possible? Compositionality evidently supplies (...) a clear answer to that question, because it guarantees that there is some way of determining the meaning of each sentence of the language from a fixed and finite base of semantic value assignments. This poses a serious challenge to Schiffer's negative thesis. -/- Schiffer proposes to answer this challenge in a way that will also provide a solution to the language-relation problem. This is the problem of specifying what relation must obtain between a population P and a language L in order for L to be a language of P. Schiffer's strategy is to reduce the problem for public languages to that of specifying the language-relation for languages of thought-specifying what it is to think in a language. -/- I will show in a precise way that Schiffer has neither met the productivity challenge nor solved the language-relation problem. Using Schiffer's characterization of what it is to think in a language, I show that if an agent thinks in some language L, then there is an infinity of languages that the agent also thinks in with the very same sentence tokens, but with arbitrarily different meanings. Thus, Schiffer has clearly not given a sufficient condition for an agent to think in a language, and Schiffer cannot do with less than a sufficient condition. Moreover, I will argue that Schiffer cannot avail himself of various attempts in the literature to address similar problems. (shrink)

In his 1993 Nicod Lectures (The Elm & the Expert), Jerry Fodor proposed a solution to a certain version of the problem of 'inscrutability of reference', which problem poses a challenge to a certain naturalistic, computational approach to cognition which Fodor has favored. The problem is that a purely informational account of an agent's mental contents cannot discriminate meanings finely enough. Fodor proposes a strategy of solution which appeals to the inferential dispositions of agents to discriminate contents more finely. After (...) a brief exposition of the problem and Fodor's bid for solution, I employ three counterexamples to argue that Fodor's proposal cannot succeed. (shrink)

The problem of negative existentials is one of the classic problems in philosophy of language. Latter-day developments in semantics resolved this problem without our help, but due to accidents of history no one noticed.

Famously, Saul Kripke proposes that there are contingent a priori truths, and has offered a number of examples to illustrate his claim. The most well-known example involves the standard meter bar in Paris. Purportedly, a certain agent knows a priori that the bar is one meter long. However, in response to a long-standing objection to such examples - the "existential complaint" - generally only modified examples having a conditional form are now considered candidates for the contingent a priori. Gareth Evans (...) argues that these conditionals must be understood free-logically, and on this basis argues against Keith Donnellan's analysis of the contingent a priori. I show Evans' argument mistaken. I also take issue with the existential complaint, and suggest a way of understanding Kripke's original examples that is not subject to it. My approach focuses the debate in its right place, and allows us to take Kripke's original examples seriously. (shrink)

The problem with model-theoretic modal semantics is that it provides only the formal beginnings of an account of the semantics of modal languages. In the case of non-modal language, we bridge the gap between semantics and mere model theory, by claiming that a sentence is true just in case it is true in an intended model. Truth in a model is given by the model theory, and an intended model is a model which has as domain the actual objects of (...) discourse, and which relates these objects in an appropriate manner. However, the same strategy applied to the modal case seems to require an intended modal model whose domain includes mere possibilia. Building on recent work by Christopher Menzel (Synthese 85 (1990)), I give an account of model-theoretic semantics for modal languages which does not require mere possibilia or intensional entities of any kind. Menzel has offered a representational account of model-theoretic modal semantics that accords with actualist scruples, since it does not require possibilia. However, Menzel's view is in the company of other actualists who seek to eliminate possible worlds, but whose accounts tolerate other sorts of abstract, intensional entities, such as possible states of affairs. Menzel's account crucially depends on the existence of properties and relations in intension. I offer a purely extensional, representational account and prove that it does all the work that Menzel's account does. The result of this endeavor is an account of model theoretic semantics for modal languages requiring nothing but pure sets and the actual objects of discourse. Since ontologically beyond what is prima facie presupposed by the model theory itself. Thus, the result is truly an ontology-free model-theoretic semantics for modal languages. That is to say, getting genuine modal semantics out of the model theory is ontologically cost-free. Since my extensional account is demonstrably no less adequate, and yet is at the same time more ontologically frugal, it is certainly to be preferred. (shrink)

The aim of the paper is twofold: i) to give a logically explicit formulation of a slight generalization of Quine's master argument about de re modality—an argument which imposes important constraints on modal semantics, ii) to briefly present my favored account of modal locutions (especially locutions of the de re metaphysical flavor) and show how it successfully copes with Quine's argument. Though Quine made this argument so many years ago, it is still widely misunderstood, and so careful attention to detail (...) seems warranted. From what I have seen, philosophers' attitudes towards Quine's master argument fall into two kinds: i) there are those that think that the argument has no force, because it is based on some mistake (usually, something about definite descriptions), and ii) there are those that think that the argument poses some insuperable barrier to any kind of de re modality. Neither of these attitudes is justified. So, I hope to make plain along the way that a) the original version of Quine's argument is sound, b) there is a version of this same basic argument which imposes very definite constraints on any proposed account of de re "metaphysical" modality in particular, an c) there is an account that satisfies these constraints. Part 1 of the paper is concerned with laying out and discussing three versions of Quine's argument, in the service of establishing points (a) and (b). -/- In Part 2 of the paper, I briefly sketch what I take to be a very promising, and also very Quinean account of de re modality -- one that respects the constraint on modal semantics that Quine's argument reveals and comports well with the few positive remarks Quine makes, for example, in Word and Object regarding our use of modal locutions. This will put us in a position to see that the proposed account does not fly in the face of Quine's master argument. (shrink)

According to Nancy Cartwright, a causal law holds just when a certain probabilistic condition obtains in all test situations which in turn satisfy a set of background conditions. These background conditions are shown to be inconsistent and, on separate account, logically incoherent. I offer a corrective reformulation which also incorporates a strategy for problems like Hesslow's thrombosis case. I also show that Cartwright's recent argument for modifying the condition to appeal to singular causes fails.Proposed modifications of the theory's probabilistic condition (...) to handle effects with extreme probabilities (0 or 1) are found unsatisfactory. I propose a unified solution which also handles extreme causes. Undefined conditional probabilities give rise to three good, but non-equivalent, ways of formulating the theory. Various formulations appear in the literature. I give arguments to eliminate all but one candidate. Finally, I argue for a crucial new condition clause, and show how to extend the results beyond a simple probabilistic framework. (shrink)

An influential view about the relationship between publicity and linguistic meaning is brought into question. It has been thought that since public languages are essentially public, linguistic meaning is subject to a kind of epistemic cap so that there can be nothing more to linguistic meaning than can be determinately known on the basis of publicly available evidence. Given the thinness of such evidence, a well-known thesis follows to the effect that linguistic meaning is substantially indeterminate. In this paper, we (...) consider the sort of reasons offered for the Epistemic Thesis and uncover an unexamined presupposition about the epistemic requirements of communication and the establishment of meaning conventions. We show this presupposition is undermined by independently motivated considerations about communication and convention, giving us good reason to reject the Epistemic Thesis and its corollary about indeterminacy. (shrink)

According to Timothy Williamson 's epistemic view, vague predicates have precise extensions, we just don't know where their boundaries lie. It is a central challenge to his view to explain why we would be so ignorant, if precise borderlines were really there. He offers a novel argument to show that our insuperable ignorance ``is just what independently justified epistemic principles would lead one to expect''. This paper carefully formulates and critically examines Williamson 's argument. It is shown that the argument (...) does not explain our ignorance, and is not really apt for doing so. Williamson 's unjustified commitment to a controversial and crucial assumption is noted. It is also argued in three different ways that his argument is, in any case, self-defeating – the same principles that drive the argument can be applied to undermine one of its premises. Along the way, Williamson 's unstated commitment to a number of other controversial doctrines comes to light. (shrink)

Alfred Tarski's work on truth has become a touchstone for a great deal of philosophical work on truth. A good grasp of it is critical for understanding the contemporary literature on truth and semantics. In this paper, I present a fresh interpretation of Tarski's view, one which aims to draw it out more fully in areas of philosophical interest.

Alfred Tarski's (1936) semantic account of the logical properties (logical consequence, logical truth and logical consistency) makes essential appeal to a distinction between logical and non-logical terms. John Etchemendy (1990) has recently argued that Tarski's account is inadequate for quite a number of different reasons. Among them is a brief argument which purports to show that Tarski's reliance on the distinction between logical and non-logical terms is in principle mistaken. According to Etchemendy, there are very simple (even first order) languages (...) for which no such distinction can be made. This is a surprising result, and an important one, if true. Since Tarski's account does indeed depend on such a distinction, Etchemendy's argument, if correct, would rule out definitively the received view on logical truth (as well as logical consequence and logical consistency). But his argument is not correct, and it is the job of this paper to show that. (shrink)

Alfred Tarski’s work on truth has played such a central role in the discourse on truth that most coming to it for the first time have probably already heard a great deal about what is said there. Unfortunately, since the work is largely technical and Tarski was only tan- gentially philosophical, a certain incautious assimilation dominates many philosophical discussions of Tarski’s ideas, and so, examining Tarski on the concept of truth is in many ways an act of unlearning. -/- In (...) this paper I focus on those key ideas in Tarski’s work that have made a lasting impact on the philosophical discourse. These are the notions of T-sentence, Convention T, Tarskian truth definition, and Tarski’s general limiting theses on the expressibility and definability of truth. Though these ideas are in name familiar, we will seek in this essay to uncover and remove certain widespread misunderstandings of each. Tarski’s name also features prominently in discussions of the liar paradox, so we will take time out to explain Tarski’s connection to this ancient puzzle. (shrink)

I discuss one version of a puzzle about the identity of a statue with the lump of clay of which it is made. The case is one in which the statue and lump agree in all their non-modal features. While this is a favorable case for the claim that they are identical, we nonetheless have discrepant intuitions about their potentialities, which appear irreconcilable. Critical analyses are given of recent treatments by Allan Gibbard, Kit Fine, and Stephen Yablo. An ontologically conservative (...) solution to the puzzle is offered which has the following features: the statue and lump of clay are said to be identical, our discrepant modal intuitions about the statue and lump are validated, identity is necessary, and is a relation between objects, not concepts, Leibniz' Law is not constrained, only a standard modal logic and semantics is needed, and our modal logical intuitions are preserved. In part, this last consists in proving that standard modal logics have certain structural features. In addition, the proposed account allows that names have the same function in and out of modal contexts, namely just to refer. As well, it allows genuine de re modality, i.e. that things have genuine modal properties. How the account meets general concerns about de re modality and essentialism is discussed. The proposed method of solution is also shown to apply to a number of other modal puzzles. (shrink)

Tarski's argumentative use of the liar paradox is well-known, but officially it is the Grelling paradox that has final pride of place in Tarski's argument, not the Liar at all. Tarski explicitly gives argumentation that adverts to the liar argument, but it is an alternative argument—one he only hints at and which adverts to the Grelling—which he says has the advantage of removing any empirical element. In this paper, we will examine how the Grelling might be used in place of (...) the Liar in Tarski's argument for his exact indefinability thesis, and assess in what way the difference might be significant. If successful, Tarski's use of the Grelling puts pressure on how Convention T can be justified. (shrink)