ABSTRACT
Normally, social scientists are deprived of the opportunity to experiment with their ‘subjects’. Introducing some treatment, holding everything else constant, and measuring what happens is simply not feasible or would not be tolerated in the real world of complex, unpredictable, and costly human relations. Hence, most economists, sociologists, and political scientists must rely heavily upon comparison between ‘naturally’ occurring events or processes in their attempt to draw generally valid inferences about causality or to put forward compelling arguments about specific historical situations. Needless to say, getting away with this admittedly second-best mode of scientific analysis is even more difficult when those involved come from diverse academic disciplines and national backgrounds-as is the case with the authors in this volume.
