On the Very Idea of a Conceptual Scheme

The essay argues that the notion of a conceptual scheme—a framework that organizes experience and could be radically different from ours—cannot be made intelligible. Davidson rejects the scheme-content dualism that underlies this notion: the idea that there is a neutral something (experience, reality, the given) that different schemes organize or fit. His central move is to show that the metaphors of 'organizing' and 'fitting' cannot be cashed out without collapsing into the claim that a scheme is simply true of its content, which in turn requires translatability. Since a language we could not translate would not be recognizable as a language, the idea of a wholly alternative scheme dissolves.

Davidson's argument turns on the claim that we cannot divorce the notion of truth from that of translation. A Tarskian truth theory for a language—the kind of theory Davidson uses to model meaning—must be verifiable by an interpreter, and verification requires translating the language into one the interpreter already understands . If a conceptual scheme were truly incommensurable, its language would be untranslatable in principle. But then we could have no evidence that it was a language rather than mere noise. The very idea of recognizing something as an alternative scheme presupposes the possibility of interpretation, and interpretation presupposes translatability. The scheme-content dualism thus collapses under its own weight.

Davidson's targets were the conceptual relativisms he discerned in Kuhn, Quine, and Whorf . Kuhn's incommensurability thesis, Quine's indeterminacy of translation, and Whorf's linguistic relativity all seemed to Davidson to presuppose the scheme-content dualism he rejects. Responses varied. Quine largely accepted Davidson's criticisms while defending a modified position. Kuhnians argued that Davidson conflated strong incommensurability with weaker, more defensible forms of conceptual difference. Critics such as Greg Lynch have argued that Davidson's argument is invalid because it moves from the premise that schemes cannot be incommensurable to the conclusion that the idea of a scheme is incoherent; Lynch defends Davidson by arguing that the target position defines schemes solely in terms of incommensurability .

Some critics argue that Davidson considers only two possibilities: full intertranslatability or mutual untranslatability. They propose a third: asymmetrical failure, where a richer language can translate a poorer one but not vice versa . If such asymmetry is possible, the idea of a conceptual scheme might survive in a weaker form. Davidson acknowledges this possibility parenthetically but does not discuss it, saying he will 'neglect possible asymmetries' . Critics argue that this neglect is a significant gap, since asymmetrical failure might allow for a notion of conceptual difference that avoids the incoherence Davidson diagnoses. Defenders respond that even asymmetrical failure presupposes enough translatability to identify the languages as languages, preserving Davidson's core point.

A frequent misreading treats the essay as claiming that all languages are fully intertranslatable in practice, or that conceptual differences do not exist. Davidson's conclusion is more modest and more radical: he argues that the idea of a scheme that is in principle untranslatable is incoherent, not that translation is always easy or that all languages say the same thing. Conceptual differences are real and can be profound, but they are differences within a shared space of interpretability, not evidence of incommensurable frameworks. The essay's target is the metaphysical picture of schemes and content, not the empirical study of linguistic diversity. Davidson's point is that we cannot make sense of a scheme that is wholly alien, because recognizing it as a scheme requires interpreting it.