Representation reconsidered: Book review

TLDR: Ramsey as mentioned in this paper argues that the notion of representation has the same explanatory value as knowledge, and that it cannot be used as a theoretical posit in certain branches of cognitive science, e.g., psychology and neuroscience.

Abstract: Some books are just begging to be written. This is one such. It gives a long overdue critical look at the nearly universal tendency to invoke the notion of ‘‘representation’’ as a theoretical posit in certain branches of the cognitive sciences, e.g., psychology and neuroscience. As the preface and opening chapter make clear, Ramsey’s project is to ask, from the vantage point of philosophy of science, whether positing representations has the sort of explanatory value it is generally imagined to have. His principal focus is to determine if the explanatory posits that are in fact employed by these sciences meet the minimal criteria for doing bona fide representational work. As he puts it, the question is whether or not such proposals meet the ‘‘job description challenge.’’ Adequately meeting that challenge requires saying not only what determines the content of a state or structure but also, critically, saying how that state or structure serves or functions as a representation in a larger system. Ramsey’s assessment is that when the notion of representation is invoked in an important class of cases this challenge cannot be met. However, he claims (chapter 3) that there are, at least, two prominent uses of the notion in the classical framework of cognitive science that are exceptions to this rule. Nevertheless, even these uses—so he argues— are at odds in important ways with the standard (folk psychological) interpretation of what being a representation amounts to (chapter 2). Against this backdrop, in making his core argument he is critical of the popular tendency to regard representations as states (or ensembles of states) that only are reliably caused by (or nomically depend upon) the occurrence of certain external features (chapter 4) or those that dispositionally produce certain effects under specific conditions (chapter 5). He classifies these sorts of theory as subscribing to what he designates, respectively, ‘‘receptor’’ and ‘‘tacit’’ notions of representation. These ideas come into play, on the one hand, when scientists speak of states or processes that serve as ‘‘detectors’’ or ‘‘indicators’’ of some external feature or other, or on the other hand, when they talk of a system or organism’s implicit or embodied ‘‘know how’’ as being responsible for generating reliable effects. Subjecting these accounts to a detailed analysis, Ramsey demonstrates that both the receptor and tacit notions of representation—those favored by today’s