Historical Representation – Narratives, Metaphors and Models

Hayden White famously argued that historical discourse takes a narrative form. He also claimed that narratives are in some sense epistemically deficient. This worry is based on his assumption that successful representation in history depends on the existence of structural similarities between the historical narrative and the sequence of actual historical events. Since such similarities cannot be identified, White concludes that narratives do not adequately or objectively represent the past; they allude to historical reality merely in metaphorical ways. White’s critics tried to avoid this conclusion either by denying the essentially narrative character of historiography (Noël Carroll) or by arguing that there do in fact exist structural similarities between (historical) narratives and (past) actions (David Carr). My own approach to the problem of historical representation takes a different route. Unlike Carroll, I do accept that historiography is narrative in character, but I reject the similarity account of historical representation that informs both White’s and Carr’s approaches. Building on contemporary pragmatic and inferentialist accounts of scientific representation (Bas Van Fraassen, Ronald Giere, Mauricio Suárez), I argue that historical narratives represent the past in a similar manner as scientific models represent their target domain: they exploit similarities and dissimilarities in selective and context-dependent ways in order to allow a user (reader) to draw informative inferences about the past. This approach to narratives, I argue, can make sense of many of the constructive features of historical discourse that were highlighted by White without buying into the more radical claim that historical narratives are epistemically deficient.