Teleosemantics and the Causal Constraint on Perceptual Content

Teleosemantics offers an account of representational content in terms of biological functions. Teleosemantic theories must impose a range of constraints on content determination so as to solve the problem of content indeterminacy. When a frog’s fly-catching mechanism detects a fly, does it represent a fly? a small dark moving dot? frog food? This talk makes the case that teleosemantic accounts of perceptual representation should incorporate a causal constraint: a perceptual representation type PR cannot represent a property P unless instances of P cause PR-tokens in cases of successful perception. I motivate the constraint by appeal to common sense and explanatory practice in vision science. I then consider two potential objections. The first concerns cases of successful visual perception of a property from cues that are correlated with but not caused by the property (Ganson, 2021). The second concerns motion perception, where the causal constraint seems to yield contents that diverge from those ascribed in vision science (Green, 2017). I defend the constraint against both challenges. Adopting the constraint has implications for ongoing debates in teleosemantics, favoring causal-informational teleosemantics (Neander, 2017) over varitel semantics (Shea, 2018).
 

Please reach out to valentina.buccheri@ehess.fr or tong.yi@ens.psl.eu for any question or need further information.