Complexity and accountability: the witches' brew of psychiatric genetics.

This paper examines the role of complexity in descriptions of the aetiology of common psychiatric disorders. While scientists attest to the discovery of an underlying reality of complex inheritance--the so-called 'witches' brew' of genetic and non-genetic factors--we argue that 'complexity' also performs rhetorical work. In our analysis of scientific review papers (1999-2008), we find a relatively stable genre of accountability in which descriptions of complexity appear to neutralize past failures by incorporating different and sometimes competing methodological perspectives. We identify two temporal strategies: retrospective accounting, which reconstructs a history of psychiatric genetics that deals with the recent failures, citing earlier twin studies as proof of the heritability of common psychiatric disorders; and prospective accounting, which engages in the careful reconstruction of expectations by balancing methodological limitations with moderated optimism. Together, these strategies produce a simple-to-complex narrative that belies the ambivalent nature of complexity. We show that the rhetorical construction of complexity in scientific review papers is oriented to bridging disciplinary boundaries, marshalling new resources and reconstructing expectations that justify delays in gene discovery and risk prediction.