On Time and Comparative Research

Abstract

The essay argues why the comparative method has been amply debated in reference to cross-cultural synchronic research, while less attention has been devoted to the problems of its application to temporal variance. Three central method-issues are discussed: (1) the definition of `temporal units' of analysis; (2) the special character of cross-time generalizations versus cross-spatial ones; and (3) the problem of historical multi-collinearity. The core arguments are that: (1) the search for a specific `historical method' is vitiated by the confusion between the peculiarity of historical `rules of method' - which cannot be identified - and the peculiarity of `cross-time variance' - which, on the contrary, is so momentous; (2) in controlling the validity of causal generalizations, resorting to temporal variance only presents methodological pitfalls not inferior to those stemming from the resort to the sole spatial synchronic variance; (3) consequently, control-maximization is best achieved by research designs which combine both types of variance and permit `slides of synchronic comparisons through time'.