The Need for Charity in Semantics

I nterpretation of what other people say begins with the identification of some set of true affirmations by informants. The interpreter seeks to match true sentences to the world in such a way as to make them come out true under the interpretation. The difficult part is that we must somehow tell which sentences are true before we are sure what any of the sentences mean. The solution, by most accounts, is to appeal to a broad-scope, methodologically grounded assumption of charity. We simply assume that most of the sentences which our informants hold true are in fact true, or at least that informants are rational in how they decide which sentences to hold true. The justification for such an across-the-board assumption of charity is methodological: without it, we are told, there is no possibility of interpretation and hence no prospect of regarding informants as believers. However, such sweeping assumptions of charity are at once controversial and sketchy in terms of how they are to be applied. First, we must wrestle with the worry that a blanket assumption may be altogether unreasonable. Our informants might be fundamentally and massively wrong in their beliefs. Second, even if we agree with such an assumption, the details of its use remain obscure. Informants are fallible and thus must be assumed to have at least some false beliefs. Yet it is unclear how the assumption that they are mostly right helps us to distinguish individual true beliefs from false beliefs. The principal objective of this paper will be to develop an alternative methodology for testing semantic theories, a methodology which does not require any assumption that informants are mostly right in their beliefs or even mostly rational in their reasoning processes. The method of testing to be developed here does indeed depend on judicious, selectively charitable hypotheses. However, I shall argue that the needed charitable assumptions are not broad in scope but narrow in scope: They concern the reliability of individual speakers with respect to specific sentences under particular empirical circumstances. More important, the grounds for such assumptions, for the most part, are not methodological but

Analytic philosophy · Contemporary philosophy · Epistemology · General interest · Programming language · Semantics (computer science) · Computer Science · Epistemology, Ethics, and Metaphysics · Philosophy

• Inconsistent Languages

  Open Access•Matti Eklund•Philosophy and Phenomenological…•2002

  Add to list