Analytic/Synthetic Distinction

by

Gillian Russell

• LAST REVIEWED: 01 September 2022
• LAST MODIFIED: 21 February 2023
• DOI: 10.1093/obo/9780195396577-0044

Introduction

“The analytic/synthetic distinction” refers to a distinction between two kinds of truth. Synthetic truths are true both because of what they mean and because of the way the world is, whereas analytic truths are true in virtue of meaning alone. “Snow is white,” for example, is synthetic, because it is true partly because of what it means and partly because snow has a certain color. “All bachelors are unmarried,” by contrast, is often claimed to be true regardless of the way the world is; it is “true in virtue of meaning,” or analytic. The existence of analytic truths is controversial. Philosophers who have thought they exist include Immanuel Kant, Gottlob Frege, and Rudolf Carnap. The philosopher most famous for thinking that they do not is Quine. Skeptics have sometimes argued that the idea of an analytic truth is incoherent, and they sometimes express this by denying the existence of the distinction. A related view is that there is a distinction but that it is trivial, since the class of analytic sentences is empty. A third kind of skeptic about analyticity questions its usefulness. It can be tempting to think that to defend analyticity one need only specify some paradigm cases and maintain that the analytic sentences are the ones like those. The use skeptic points out that analyticity is of interest only because it is thought to entail certain other features. One can define as many conceptions of analyticity as one likes, but none of them will do the work that analyticity has traditionally been expected to do. An analogy (due to Gilbert Harman) is with debates over the existence of witches. Someone might defend the claim that witches exist by pointing to the people who are taken to be paradigm cases of witches in their linguistic community (say, the people who have already been burned at the stake), claiming that “witch” properly applies to anyone who is like that. But a skeptic can argue that while one can define “witch” any way one likes, people have been burned at the stake because witches were thought to have certain salient features, such as having magical powers. The skeptic’s main point is that there is no person with those features—the features that justify the practice. Similarly, the skeptic about analyticity may allow that one can define some notions of analyticity while maintaining that there are no truths that will be useful to philosophers in the way analytic truths were supposed to be.

General Overviews

The literature on analyticity is vast. It encompasses work by and about important historical figures, such as Immanuel Kant, Gottlob Frege, and Rudolf Carnap; a prolific 20th-century debate spearheaded by Carnap and Quine; a more recent debate arising from Boghossian 1996 (cited under the Epistemology of Logic) and the extensive literature concerning applications of the distinction, especially in the foundations of mathematics, the epistemology of logic, and the methodology of philosophy. In addition, there is much work in philosophy of language—such as that on externalism, vagueness, and indexicality—that has important consequences for the distinction and should be read by anyone working in the area. Perhaps unsurprisingly, then, no single text provides a complete survey. The introductory texts listed here were selected both for accessibility and for influence on the course of the debate. Broader works are listed under Surveys. Ayer 1990 is extremely readable and does a good job of motivating interest in the analytic/synthetic distinction. Carnap 1958 is a shorter work but equally intoxicating. Quine 1951 is by far the most widely read paper objecting to the analytic/synthetic distinction (though it is best read in conjunction with Harman 1999 and chapter 16 of Soames 2003, cited under Useful Background). Grice and Strawson 1956 is a well-known response to Quine. Gellner 2005 is a popular book attacking the linguistic approach to philosophy associated with Oxford University in the 1950s. It includes a foreword by a sympathetic Bertrand Russell.

• Ayer, Alfred Jules. Language, Truth, and Logic. Harmondsworth, UK: Penguin, 1990.

  This short, opinionated book conveys some of the motivation for the distinction in an accessible way. First published in 1936.

• Carnap, Rudolf. “Empiricism, Semantics, and Ontology.” In Meaning and Necessity. 2d ed. By Rudolf Carnap, 205–221. Chicago: University of Chicago Press, 1958.

  A classic paper in which Carnap lays out his views about analyticity and linguistic frameworks in a nontechnical fashion. Ideal for undergraduate teaching.

• Gellner, Ernest. Words and Things. London: Routledge, 2005.

  A highly readable book expressing dissatisfaction with the linguistic turn in philosophy. Originally published in 1959.

• Grice, H. Paul, and Peter F. Strawson. “In Defence of a Dogma.” Philosophical Review 65 (1956): 141–158.

  DOI: 10.2307/2182828

  Grice and Strawson’s well-loved response to Quine.

• Quine, Willard V. “Two Dogmas of Empiricism.” Philosophical Review 60 (1951): 20–43.

  DOI: 10.2307/2181906

  The most famous of Quine’s attacks on analyticity; it is best supplemented with either Gilbert Harman’s “The Death of Meaning” or chapter 16 of Scott Soames’s Philosophical Analysis in the Twentieth Century, Volume 1, The Dawn of Analysis (Harman 1999 and Soames 2003, both cited under Useful Background).

• Quine, Willard V. “Carnap and Logical Truth.” In The Ways of Paradox and Other Essays. Rev ed. By Willard V. Quine, 107–132. Cambridge, MA: Harvard University Press, 1976.

  This paper contains more key Quinean arguments against analyticity, especially in set theory and the foundations of mathematics. Originally published in 1976.