Published online by Cambridge University Press: 01 February 2004
A central pillar of Carpendale & Lewis's (C&L's) argument is Wittgenstein's later work on language. I suggest that this support is not as strong as might be wished, and offer an alternative approach to their conclusion that language learning, especially of folk psychology, involves a socially embedded constructivism.
1. Normativity” is used here in the philosophical sense of involving the possibility of being bad or wrong. It includes such oppositions as functional-dysfunctional, true-false, correct-incorrect, and so on. Regarding Wittgenstein and language, see, for example, Glock (1996) and Shankar (1996).
2. It is an empiricist assumption that it must be so constituted. In contrast, any action-based model of representation forces learning to be a constructivist process: The environment cannot impress successful action systems into an otherwise passive mind (Bickhard & Campbell 1989).