Online research in philosophy

Conceptions of analytic truth -- Hume's fork -- Kant and the analytic/synthetic distinction -- Synthetic a priori propositions -- Bolzano and analyticity -- Analyticity in frege -- Russell's paradox and the theory of descriptions -- The Vienna circle -- Carnap and logical empiricism -- Carnap and Quine -- Demise of the aufbau -- Philosophy as logical syntax -- Logical and descriptive languages -- Physical languages -- Analyticity in syntax -- Carnap's move to semantics -- Explications -- Analyticity in a semantic setting -- Eliminating metaphysics : Carnap's final try -- W.V. Quine : explication is elimination -- Behaviorists ex officio -- Analyticity in the crosshairs -- Analyticity and its discontents -- Questioning analyticity -- Quine's two dogmas of empiricism -- Objections to the coherence of analytic -- Quine's coherence arguments : Carnap's reply -- Other responses to the coherence objection : Grice and Strawson on Quine -- A second dogma of empiricism -- Responses to the existence objections to analyticity -- Analyticity by convention -- Quine's developed attitude toward analyticity -- Analyticity and ontology -- Quine's naturalized ontology -- The indeterminacy of translation -- Some consequences of the indeterminacy arguments : ontological relativity and analyticity -- Responses to Quine's indeterminacy arguments -- Carnap's empiricism, semantics, and ontology -- Some Quinean and other responses to empiricism, semantics, and ontology -- Some recent connections between conceptual truths and ontology -- Quine's criterion of ontological commitment, causality, and exists -- Eli Hirsch and Ted Sider on mereological principles -- The Canberra Project : a resurrection of Carnap's aufbau -- Analyticity and epistemology -- Analytic truths and their role in epistemology : the classical position -- Objecting to the classical position -- Bonjour on moderate empiricism -- Quine's epistemology naturalized -- Quine and evidence : responses to circularity -- Kripke on a priority, analyticity, and necessity -- Analyticity repositioned -- The concept analytic -- One type of statement that might be reasonably called analytic -- Aside on two dimensionalism -- Analyticity and T-analyticity -- How analyticity avoids many common objections to analyticity -- Some brief comments on two other approaches to analyticity -- Mathematical claims as T-analytic -- A further potential application : pure and impure stipulata.

The author describes an interpreted modal language and produces some clear examples of logical and analytic truths that are not necessary. These examples: (a) are far simpler than the ones cited in the literature, (b) show that a popular conception of logical truth in modal languages is incorrect, and (c) show that there are contingent truths knowable ``a priori'' that do not depend on fixing the reference of a term.

Despite my concerted efforts to formulate the linguistic doctrine of (first-order) logical truth, explicitly not as a claim that stipulations governing logical particles suffice to generate the logical truths (LD(I)), but as a determination thesis (LD(III))--that stipulations that certain particles behave as the classical logical particles suffice to determine uniquely the class of logically valid sentences, whose emptiness is clear and relatively unproblematic--, Quine2 nevertheless managed to read me as having claimed “that the logical truths can be generated (sic!) by stipulations--hence conventions--without the regress”! This was all the more disheartening as Quine had also written, in earlier correspondence,3 concerning my answer to the regress argument of Quine’s “Truth by Convention”: “I think it a good answer.” That answer turned on the point that neither the conventionalist nor anyone else need justify the logical truths--as empty, they require no justification--but rather that logical rules are needed, and perfectly in order, to justify of any logical truth that indeed it requires no justification in virtue of membership in the privileged class. Evidently, Quine must have changed his mind about this, for he disparages my notion of a “stipulated universal trait”, such as ‘being red or not red’, by rhetorically asking: “But how, without prior logic, do we then infer, in particular, that the Taj Mahal has the trait?” (P. 206.) We agree that inferences cannot be made without logic, but why..

I consider the well-known criticism of Quine's characterization of first-order logical truth that it expands the class of logical truths beyond what is sanctioned by the model-theoretic account. Briefly, I argue that at best the criticism is shallow and can be answered with slight alterations in Quine's account. At worse the criticism is defective because, in part, it is based on a misrepresentation of Quine. This serves not only to clarify Quine's position, but also to crystallize what is and what is not at issue in choosing the model-theoretic account of first-order logical truth over one in terms of substitutions. I conclude by highlighting the need for justifying the belief that the definition of first-order logical truth in terms of models is superior to its definition in terms of substitutions.

It would be ever so nice if there were a viable analytic/synthetic distinction. Though nobody knows for sure, there would seem to be several major philosophical projects that having one would advance. For example: analytic sentences2 are supposed to have their truth values solely in virtue of the meanings (together with the syntactic arrangement) of their constituents; i.e., their truth values are supposed to supervene on their linguistic properties alone.3 So they are true in every possible world where they mean what they mean here.4 So they are necessarily true. So if there were a viable analytic/synthetic distinction (‘a/s distinction’ often hereafter), we would understand the necessity of at least some necessary truths. If, in particular, it were to turn out that the logical and/or the mathematical truths are analytic, we would understand why they are necessary. It would be ever so nice to understand why the logical and/or mathematical truths are necessary (cf. Gibson 1998; Quine 1998). Any account of necessity would be welcome, but one according to which necessary truths are analytic has special virtues. Necessity isn’t, of course, an epistemic property. Still, suppose that the necessity of a sentence arises from the meanings of its parts. It’s natural to assume that one of the things one knows in virtue of knowing one’s language is what the expressions of the language mean (cf., e.g., Boghossian 1994). A treatment of modality in terms of analyticity therefore connects the concept of necessity with the concept of knowledge; and knowledge is, of course, an epistemic property. So maybe if there is an a/s distinction, we could explain why the necessary truths, or at least some of the necessary truths, are knowable a priori by anybody who knows a language that can express them (cf. Quine 1991). It bears emphasis that not every theory of..

I want to analyse the Quine-Carnap discussion on analyticity with regard to logical, mathematical and set-theoretical statements. In recent years, the renewed interest in Carnap’s work has shed a new light on the analytic-synthetic debate. If one fully appreciates Carnap’s conventionalism, one sees that there was not a metaphysical debate on whether there is an analytic-synthetic distinction, but rather a controversy on the expedience of drawing such a distinction. However, on this view, there can be no longer a single analytic-synthetic distinction, because several kinds of statements could be regarded as analytic (L-determinate). L-equivalence between extra-logical linguistic predicates has already been heavily debated. The recent consensus states that Quine’s rejection of this analytic-synthetic is pragmatically grounded in his linguistic behaviorism. However, Carnap’s logical frameworks also contain other kinds of statements, and it is worthwhile to compare both Quine and Carnap’s grounds for considering these statements as analytic or not analytic. First, I will discuss logical statements. I will argue that Quine draws a very sharp distinction between first order logic and set theory, which should be regarded as a (pragmatic) analytic-synthetic distinction (as Quine admits in an interview, see Theoria, 40, 1994, p. 199). In fact, Quine’s major worry is whether identity statements are analytic. Second, I will discuss mathematical statements. In Carnap’s Foundations of Logic and Mathematics, it is clear that mathematical statements are analytic. For Quine, all mathematical statements are reducible to set-theoretical statements. Third, I discuss the analyticity of set-theoretical statements. For Quine, the membership predicate should be regarded as an interpreted extra-logical predicate. Quine’s work in set theory and his later philosophy of set theory naturally lead to the view that set-theoretical statements cannot be analytic. A major complication for the Quine-Carnap comparison is that Carnap has no elaborate reflections on set theory, while the influence of set theory on Quine’s views can hardly be underestimated. I conclude with some lessons for the contemporary debate on analyticity.

It is sometimes said that there are two, competing versions of W. V. O. Quine’s unrelenting empiricism, perhaps divided according to temporal periods of his career. According to one, logic is exempt from, or lies outside the scope of, the attack on the analytic-synthetic distinction. This logic-friendly Quine holds that logical truths and, presumably, logical inferences are analytic in the traditional sense. Logical truths are knowable a priori, and, importantly, they are incorrigible, and so immune from revision. The other, radical reading of Quine does not exempt logic from the attack on analyticity and a priority. Logical truths and inferences are themselves part of the web of belief, and the same global methodology applies to logic as to any other part of the web, such as theoretical chemistry or ordinary beliefs about ordinary objects. Everything, including logic, is up for grabs in our struggle for holistic confirmation. The purpose of this paper is to examine the law of non-contradiction, and the concomitant principle of ex falso quodlibet, from the perspective of the principles advocated by the radical Quine. I show that he has no compelling reason to accept either of these. To put it bluntly, neither the law of non-contradiction nor the rule of ex falso quodlibet is empirically confirmed, and these principles fare poorly on the various criteria for theory acceptance on the methodology of the radical Quine. So the radical Quine is led rather quickly and rather directly into something in the neighborhood of Graham Priest’s dialetheism.

It would be ever so nice if there were a viable analytic/synthetic distinction. Though nobody knows for sure, there would seem to be several major philosophical projects that having one would advance. For example: analytic sentences2 are supposed to have their truth values solely in virtue of the meanings (together with the syntactic arrangement) of their constituents; i.e., their truth values are supposed to supervene on their linguistic properties alone.3 So they are true in every possible world where they mean what they mean here.4 So they are necessarily true. So if there were a viable analytic/synthetic distinction (‘a/s distinction’ often hereafter), we would understand the necessity of at least some necessary truths. If, in particular, it were to turn out that the logical and/or the mathematical truths are analytic, we would understand why they are necessary. It would be ever so nice to understand why the logical and/or mathematical truths are necessary (cf. Gibson 1998; Quine 1998).

Quine claims that holism (i.e., the Quine-Duhem thesis) prevents us from defining synonymy and analyticity (section 2). In Word and Object, he dismisses a notion of synonymy which works well even if holism is true. The notion goes back to a proposal from Grice and Strawson and runs thus: R and S are synonymous iff for all sentences T we have that the logical conjunction of R and T is stimulus-synonymous to that of S and T. Whereas Grice and Strawson did not attempt to defend this definition, I try to show that it indeed gives us a satisfactory account of synonymy. Contrary to Quine, the notion is tighter than stimulus-synonymy – particularly when applied to sentences with less than critical semantic mass (section 3). Now according to Quine, analyticity could be defined in terms of synonymy, if synonymy were to make sense: A sentence is analytic iff synonymous to self-conditionals. This leads us to the following notion of analyticity: S is analytic iff, for all sentences T, the logical conjunction of S and T is stimulus-synonymous to T; an analytic sentence does not change the semantic mass of any theory to which it may be conjoined (section 4). This notion is tighter than Quine's stimulus-analyticity; unlike stimulus-analyticity, it does not apply to those sentences from the very center of our theories which can be assented to come what may, even though they are not synthetic in the intuitive sense (section 5).

Quine claims that holism (i.e., the Quine-Duhem thesis) prevents us from defining synonymy and analyticity (section 2). In "Word and Object," he dismisses a notion of synonymy which works well even if holism is true. The notion goes back to a proposal from Grice and Strawson and runs thus: R and S are synonymous iff for all sentences T we have that the logical conjunction of R and T is stimulus-synonymous to that of S and T. Whereas Grice and Strawson did not attempt to defend this definition, I try to show that it indeed gives us a satisfactory account of synonymy. Contrary to Quine, the notion is tighter than stimulus-synonymy -- particularly when applied to sentences with less than critical semantic mass (section 3). Now according to Quine, analyticity could be defined in terms of synonymy, if synonymy were to make sense: A sentence is analytic iff synonymous to self-conditionals. This leads us to the following notion of analyticity: S is analytic iff, for all sentences T, the logical conjunction of S and T is stimulus-synonymous to T; an analytic sentence does not change the semantic mass of any theory to which it may be conjoined (section 4). This notion is tighter than Quine's stimulus-analyticity; unlike stimulus-analyticity, it does not apply to those sentences from the very center of our theories which can be assented to come what may, even though they are not synthetic in the intuitive sense (section 5).