Literature and Medicine as a Critical Discourse

jjl Literature and Medicine as a Critical Discourse* Bruce M. Psaty ... die [modern medical] examination opened up two correlative possibilities : firsdy, the constitution of the individual as a describable, analysable object . . . and, secondly, the constitution of a comparative system that made possible the measurement of overall phenomena, the description of groups, die characterization of collective facts, die calculation of gaps between individuals, their distribution in a given 'population '. — Michel Foucault, Discipline and Punish: The Birth of the Prison, p. 190 Modelling in science remains, partly at least, an art. Some principles do exist, however, to guide die modeller. The first is diat all models are wrong; some, though, are better than others and we can search for the better ones. - P. McCullagh and J. A. Neider, Generalized Linear Models, p. 6 Biostatistics has become for medicine the arbiter of trudi. Embracing the models and values of science, the medical community has declared a set of criteria diat permit a certain class of propositions to enter the realm of die verified. What ultimately invests data with meaning or significance is statistical analysis. Each study we judge first of all as an isolated event even diough its hypodiesis originates from the configuration of the known and, dierefore, depends on an accumulation of concepts * Preparation of this article was assisted by a grant from the Robert Wood Johnson Foundation, Princeton, N.J., and by the Health Services Research and Development Program , Veterans Administration Medical Center, Seattle, Wash. The opinions, conclusions, and proposals in the text are mine and do not necessarily represent the views of the Robert Wood Johnson Foundation or the Veterans Administration Medical Center. I also wish to thank Thomas S. Inui, Jerry Herron, Halsted R. Holman, Eugene Kintgen, Sue Psaty, and Charles E. Odegaard, for their helpful comments and suggestions. Literature and Medicine 6 (1987) 13-34 © 1987 by The Johns Hopkins University Press 14 A CRITICAL DISCOURSE already present and, sometimes, almost invisible. The notion of biological sense is a secondary activity; it refers to making explicit, in the discussion, some of die connections between the known and die results of a study as statistical analysis has declared their trudi value. Each hypothesis thus operates under two constraints: in addition to making sense, it must be testable. In order to become true, an hypothesis must assume the form of a conventional study, manifest itself in terms of quantifiable categories, and imitate the structure of a statistical test capable of yielding a p-value. Biostatistics defines, in short, the syntax of medical research — the ways in which we can legitimately compare, combine, and evaluate concepts. At the same time that medicine has come to speak increasingly the language of biostatistics, the journals of medical science and committees on medical education have also expressed a renewed interest in the art of medicine and the place of the humanities.11 intend to claim no causal connection between these two historical trends, which have merged in part in the new genre of studies on patient-physician communication. These studies have encouraged me to articulate my own views about how literature and medicine may be related. Subjecting the interaction between patient and physician to the scrutiny of science, in an effort to study patientphysician communication, produces some sort of conclusion: the results answer yea or nay to an initial hypothesis. But to take as the object of study an area traditionally regarded the province of the humanities is, I maintain, to incorporate the humanities into medicine in, at most, a nominal way. If die humanities are to enrich the process of medical care, they can do so, not by becoming the object of scientific studies, but only by offering new models, alternative ways of perceiving or understanding these human relationships. This essay, then, outlines crucial, perhaps also obvious, distinctions between literature and medicine, between the models of the sciences and those of the humanities. The first part focuses on the structural correspondence between scientific studies and statistical tests, which together represent historical events as a set of mathematical relationships taking place outside of time. In the second section, I address several reports on patient-physician communication. While their contribution has been to introduce the...