It always interests me to know when people talk of concept-formation in learning and intelligence just what sort of account of 'concept' they are using. Consider, for example, a 'cat'. Suppose we encounter one that is 25 feet tall. Is it still a cat? Is it a new type of cat? Or just an existing cat (a tiger, say) with extraordinary properties. Anyhow. This article (24 page PDF) takes these questions seriously, and in particular, examines Waismann's notion of open texture through the paradigm of the prototype theory of concepts, a theory that in turn evolved out of cognitive linguistics and can be contrasted with empirical and formal theories of concepts. In particular, it involves the idea that concepts can be open ended and vague, similar to Wittgenstein's 'family resemblances', such that (say) different entities can be more or less instances of a given concept (that is, being an instance of a concept is not an 'all or nothing' proposition). This paper is accessible and clearly written, and a good starting point for a serious inquiry into these ideas, if you're so inclined.
Today: Total: [Direct link] [Share]
