In this article, I argue that the problem of representation stands at the center of the debate concerning the legitimacy of cognitivism as a research strategy for the humanities. Yet, curiously, very few commentators in this debate see representation to be a problem at all. Questions about the anthropological origin and function of representation tend to be regarded as at best supplemental, and at worst simply irrelevant, to the synchronic question of the causal mechanisms involved in the production of representation in the brain. I argue that this view is fundamentally mistaken and,furthermore, that we won't get clear about the central issues in the debate over cognitivism in literary studies until we get clear about the problem of representation. The problem is essentially one of how to define the human in terms of its most unique trait: the capacity for symbolic representation. After a review of how cognitivism misinterprets this question as a gradually evolved genetic adaptation of the nervous system, I turn to the theory of representation proposed by the neuroscientist and anthropological researcher Terrence Deacon in The Symbolic Species: The Co-Evolution of Language and the Brain (1997). Deacon stands out among cognitive scientists for his conclusion that language—symbolic reference—is an“evolutionary anomaly,” that is, inassimilable to the mechanism of gene replication. By understanding the exact nature of this evolutionary anomaly, we are in a much better position to assess the skepticism that is routinely directed toward those who use cognitive science to interpret literature. More precisely, I argue that the originary function of the symbolic sign is the deferral of lower-level indexical reference strategies. This originary anthropological function is most clearly evident in literature,art, and religion.
Skip Nav Destination
Richard van Oort; Cognitive Science and the Problem of Representation. Poetics Today 1 June 2003; 24 (2): 237–295. doi: https://doi.org/10.1215/03335372-24-2-237
Download citation file: