Abstract

A history of cohort analysis has been given, and it has been pointed out that the bulk of the literature on the subject has dealt with concepts that assume either that graduation and possibly extrapolation are desirable, preferably in conformity with some formula expressing what is implicitly accepted as a "law of mortality", or that, whether or not a fixed pattern of mortality exists, the intensity of mortality risk is largely determined in early life. The view is advanced that either concept alone, is, or both concepts together are, inadequate and may lead to an improper assessment of the nature-nurture complex, since environment and therapeutic measures are constantly changing. The plea is made for the technique of cohort analysis to be used as a narrative or historical technique, and for a synthesis of knowledge derived from social history, medical history, and cohort analysis to be made to interpret the narrative. The effects of a changing rate on the mortality pattern as depicted by date-contour analysis and cohort analysis are outlined and the danger of basing biological theories on the mortality pattern revealed by date-contour analysis alone is indicated. This point is illustrated by reference to a recent theory that has been based on such an analysis.