Print

A priori

DOI
10.4324/9780415249126-P001-1
Versions
DOI: 10.4324/9780415249126-P001-1
Version: v1,  Published online: 1998
Retrieved March 29, 2024, from https://www.rep.routledge.com/articles/thematic/a-priori/v-1

1. Necessity, analyticity and the a priori

Contemporary understanding of the distinction between the a posteriori and the a priori, as the distinction between the empirical and the non-empirical, derives mainly from Kant’s Critique of Pure Reason (1781/1787), although versions of it precede Kant in the writings of Leibniz and Hume (see Kant, I. §4). The epistemological distinction between a priori and a posteriori knowledge differs from the logical or metaphysical distinction between necessary and contingent truth, and from the semantical distinction between analytic and synthetic truth (see Analyticity). In particular, the concept of a priori knowledge is not the same as either the concept of what is (logically or metaphysically) necessarily true or the concept of what is true analytically, just in virtue of the meanings of a proposition’s constituent terms. Kant’s talk of a priori ‘modes of knowledge’ suggests an epistemological, knowledge-oriented characterization of what is a priori. As standardly characterized, a priori knowledge is knowledge that does not depend on evidence from sensory experience. The previous considerations do not, however, settle the issue of whether every proposition knowable a priori is either necessarily true or analytically true.

A necessarily true proposition is not possibly false, or in Leibniz’s words, is true in ‘all possible worlds’. Contingently true propositions are possibly false, that is, false in some possible worlds. Traditionally, many philosophers have assumed that a proposition is knowable a priori only if it is necessarily true, presumably on the ground that if a proposition is possibly false, then it requires for its justification supporting evidence from sensory experience. Contingent truths, according to this traditional view, are not candidates for a priori knowledge.

Saul Kripke (1980) has argued that some contingently true propositions are knowable a priori. He cites the knowledge that stick S is one metre long at a certain time, where stick S is the standard metre-bar in Paris. If one uses stick S to ‘fix the reference’ of the term ‘one metre’, then, according to Kripke, one can know a priori that stick S is one metre long. The truth that stick S is one metre long is contingent rather than necessary; for S might not have been one metre long. (Application of sufficient heat to S, for instance, would have changed its length.) It seems arguable, then, that some contingent truths are knowable a priori, contrary to what many philosophers have assumed. This matter has prompted considerable discussion among contemporary philosophers, with some still contending that no contingently true proposition is knowable a priori. Some of the latter philosophers have noted, with regard to Kripke’s example, that ‘one metre’ can be used either as (1) the name of the length of S whatever that length may be, or as (2) the name of a particular length singled out by a speaker. Given option (1), these philosophers hold, the claim that stick S is one metre long will be necessary and knowable a priori, and given option (2), the claim that stick S is one metre long will be contingent and knowable only a posteriori.

Many philosophers have held that a priori knowledge is restricted to such analytic truths as ‘All rectangles have four sides’ and ‘All bodies are extended’. If such truths are analytic, they are true just in virtue of the meanings of their constituent terms. Such truths differ from synthetic truths, which are true in virtue of something other than just the meanings of their constituent terms (for example, in virtue of observable situations in the world). Synthetic judgments, according to Kant, are ‘ampliative’ in that they ‘add to the concept of the subject a predicate which has not been in any wise thought in it, and which no analysis could possibly extract from it’ (1781/1787: A7/B11). Some philosophers, notably W.V. Quine, have contested the viability of any philosophically important distinction between analytic and synthetic truths (see Quine, W.V. §3).

One issue of philosophical controversy is whether any synthetic truth is knowable a priori. Kripke’s aforementioned metre example offers, according to some philosophers, a synthetic truth knowable a priori. Kant held that some synthetic truths, for example, those of geometry, have a kind of necessity that cannot be derived from experience, and can be known a priori. Such synthetic truths, Kant argued, can be known independently of evidence from sensory experience. Kant’s doctrine of synthetic a priori truths still generates controversy among philosophers, specifically in connection with such apparently synthetic propositions as ‘Nothing can be green and red all over’ and ‘A straight line is the shortest path between two points’. The later Wittgenstein, for example, proposed that propositions of the latter sort are actually conventional ‘rules of grammar’, or non-synthetic normative standards for representation (see Necessary truth and convention).

Print
Citing this article:
Moser, Paul K.. Necessity, analyticity and the a priori. A priori, 1998, doi:10.4324/9780415249126-P001-1. Routledge Encyclopedia of Philosophy, Taylor and Francis, https://www.rep.routledge.com/articles/thematic/a-priori/v-1/sections/necessity-analyticity-and-the-a-priori.
Copyright © 1998-2024 Routledge.

Related Articles