Trust and mistrust in the market place: statistics and clinical reasearch (1945-1960)

My paper will focus on the role of social mistrust in bringing about change s in experimental practice in medicine. Between 1940 and 1960, academic res earchers in the United States introduced a series of radically new experime ntal practices to evaluate new therapies: random allocation of patients to " experimental " or " control " treatments, " blind " assessment of therape utics, the increased use of objective measures of therapeutic outcomes, and the statistical analysis of results. These innovations were justified by r hetorical appeals about the need to remove therapeutic evaluation from the control of " untrustworthy " drug manufacturers and " unreliable " practiti oners. Social mistrust played a key role in establishing the new methodolog ical canons. Reformers relied on a prior tradition of social criticism of d rug marketing to convince colleagues to accept innovative research procedur es. I will argue that cultural changes in epistemological standards and bel iefs cannot by themselves explain the acceptance of new epistemological sta ndards. One must also look to whose evidence and practices are mistrusted.