No label defined (Q4488093)

A remarkable analysis of concept formation based on non-mentalistic construal of concepts: ``\dots{} concepts are treated as stabilising classificatory, contiguity, and ordering structures on growing sets of data'' (p. 5). The author characterises her method as being philosophical and logico-semantical (as opposed to mentalistic or purely linguistic methods). In the Introduction she articulates some main starting points of her theory, (only) one of which is that the `quasi-concepts' (which evolve to genuine concepts) arise as equivalence sets induced by stabilisation of sequences of finite sets of data. The topics in a still very general chapter, Preliminaries, are: regularities of various kinds, one of the key notions, viz. `satisfaction situation', and explications of polysemy/homonymy and semantic correctness. Here also the first level and second level concept formation are distinguished. The following chapters analyse particular problems exploiting the general princples which have been adumbrated in the introductory chapters above. Chapter 1, ``Concept formation: its basis and structure'', deals with first level and second level concept formation, where one of the notions important w.r.t. the former is the notion of perspectives (under which identities and differences can be discerned), and theoretical and formal concepts are defined for the latter. Chapter 2, ``Concept formation: the construction of new concepts'', connects constructing new concepts with generating polysemies due to using metaphors/metonymies. The notion of understanding is here explicated in connection with concept formation and creating new concepts. Chapter 3, ``Knowledge and understanding'', explains some rather complex problems like (among others) identity of individual concepts, identity of concepts across possibilities and counterfactual possibilities and across changing knowledge; besides, it brings an explication of some epistemic attitudes in terms of might, must, might have been. An interesting problem of `correcting concepts' is analysed in Chapter 4, ``Correction in concept formation'', together with a well-illustrated classification of such corrections. Some new views concerning the great theme `propositional attitudes' can be found in Chapter 5, ``Conceptual semantics and propositional attitudes''. The author tries to articulate some specific features of her theory as compared to the theory of structured propositions or with Fodor's theory of ``language of thought'. Finally, in Chapter 6, ``Concept formation and connectionist models'', it is shown that there are some `instructive parallels' between Dynamic Conceptual Semantics (Bartsch) and connectionists. On the other hand, the latter have created some problems that they are hardly able to solve. A summarising and concluding last chapter, a rich bibliography and a subject index close this most interesting and inspiring book. The reviewer misses only references to Bolzano, Bealer, Kuhn and perhaps Tuomela.