Department of Theoretical Philosophy
The Perspectivity of Representation
Cognitive science often posits the existence of internal representations in the brain. However, as William Ramsey (2007) points out, it is still unclear how these representations are supposed to be cashed out in naturalistic terms. He argues that the most important open issue is articulating how a neurological or computational state functions as a representation in a biological or computational system. Ramsey’s own suggestion for separating representational chaff from wheat is arguing that X only represents Y when X has the role of ‘standing (in) for Y’ in a system. In this paper I evaluate this proposal and offer an alternative.
The evaluation and alternative are inspired by Bas van Fraassen (2008), who argues that there is only representation to the extent some things are taken, made, or used to represent some other thing. I distill some necessary conditions for such a use, and argue that ordinary representation needs a user that relates X to Y in order to use X as a representation of Y. This, however, presupposes that the user has at least some access, either directly or indirectly, to Y, otherwise X cannot represent Y to him. Because there supposedly is no user capable of connecting X to Y in Ramsey’s ‘naturalized’ representation, it is unclear how representation exactly comes about here. I suggest that ordinary representation might be a good alternative. While a particular system itself might indeed not be able to connect X to Y, there is always a capable user implicitly present: we, in our roles as, for example, explainers of that system. Through the use of representations located internal to the system under consideration, the user (which is the explainer, to whom it represents) aims to explain a system’s functioning by highlighting relevant underlying patterns and relations obtaining between X and Y. This instrumental approach to representation can account for when and how representation occurs. As a further advantage it provides a third way out of the debate on the existence of internal representations: they are only ‘there’ to the extent they are used as such.
|Last modified:||23 July 2015 5.01 p.m.|