The comparative method can be used to find explanatory relationships between phenomena, to discover the uniqueness of different societies and to formulate historical problems. The same logic is always used; if phenomenon A is said to exist because of the existence of condition B, we look for other social units where A occurs without B. If we find none, confidence in our hypothesis increases. Units of comparison vary not only with the aspect of social life being studied, but also with (...) the experimental hypothesis used for comparison. Units need not be geographical, but may be any social systems. Spatial and temporal proximity of units is helpful, but not necessary. Comparative logic offers only a set of rules for gathering evidence for tests, and hypotheses must be supplied by historical imagination. (shrink)