Exemplar semantics through parallel corpora Something about indefinite pronouns

We can determine ‘similarity’ of meaning typologically. If two particular meanings are often expressed by the same surface form (across a random sample of languages), then we can assume that the two meanings are ‘similar’ to the human mind. [. . . ] From ‘similarities’ it is a short step to maps of grammar/meaning space. We arrange different meanings on a map so that ‘similar’ meanings are close together, non-similar meanings farther apart. [. . . ] If we have successfully constructed such a universal map, most grammatical categories or words will have a single range of uses . . .That range will be a compact contiguous area on the map. (Anderson, 1980, 227-228) Later applications: Kemmer (1993); van der Auwera and Plungian (1998); Haspelmath (1997, 2003); Levinson, Meira, and The Language and Cognition Group (2003); Cysouw and Wälchli (2007); Majid, Boster, and Bowerman (2008); Croft and Poole (2008); Hartmann, Haspelmath, and Cysouw (2014); special issues of Linguistic Discovery, Theoretical Linguistics. Relation between typology and cognition directly: Bowerman (1993); Gentner and Bowerman (2009). But why? Argument from cultural evolution (Silvey, Kirby, & Smith, 2015): ‘Word Meanings Evolve to Selectively Preserve Distinctions on Salient Dimensions’. So: inferring from many evolutionary outcomes (languages) what the salient dimensions (of the map) are.