Totalitarianism, Model Mania and Learning from Error

This article is conceived as a learning from error exercise under the assumption that unless we take stock of our past mistakes we are bound to repeat them. The notion of totalitarianism appears to serve this purpose well. It is shown, (i) that totalitarianism was dismissed prematurely for wrong reasons, (ii) that its successor concepts have been wrongly conceived as `models', and (iii) that model mania had developed to a point of harming our understanding. The exercise is retrospective in order to be prospective. Thus, totalitarianism is revisited to illustrate the analytic handling of typological constructs and to buttress the general point that political science has backed itself into a number of blind alleys.