Generics and Typicality: A Bounded Rationality Approach
Although generics are studied mostly in formal semantics and philosophers of language, recently they attracted the attention of cognitive psychologists and social philosophers as well. The reason is that generics play a core role in the way we learn, represent and reason about groups in the world (cf. Leslie, 2008). One of the important— and worrying (Haslanger 2011)— findings in these areas is (i) that we accept generic statements of the form ‘Gs are P ’ typically if relatively many Gs have property P, but (ii) that if we are unfamiliar with a group and we learn a generic statement about it, we still interpret it in a much stronger way: (almost) all Gs are P (Cimpian et al 2010). This paper provides a defense and explanation of (i) and (ii) based on a new formal interpretation of Tverky & Kahneman’s (1974) ‘Heuristics and Biases’-program of bounded rationality, by saying that ‘Gs are f’ is true if f is a (stereo)typical feature of Gs, and by analyzing typicality in terms of how cognitive psychologists and philosophers of science have measured the causal strength of a cue to a result.
Although generics are studied mostly in formal semantics and philosophers of language, recently they attracted the attention of cognitive psychologists and social philosophers as well. The reason is that generics play a core role in the way we learn, represent and reason about groups in the world (cf. Leslie, 2008). One of the important— and worrying (Haslanger 2011)— findings in these areas is (i) that we accept generic statements of the form ‘Gs are P ’ typically if relatively many Gs have property P, but (ii) that if we are unfamiliar with a group and we learn a generic statement about it, we still interpret it in a much stronger way: (almost) all Gs are P (Cimpian et al 2010). This paper provides a defense and explanation of (i) and (ii) based on a new formal interpretation of Tverky & Kahneman’s (1974) ‘Heuristics and Biases’-program of bounded rationality, by saying that ‘Gs are f’ is true if f is a (stereo)typical feature of Gs, and by analyzing typicality in terms of how cognitive psychologists and philosophers of science have measured the causal strength of a cue to a result.