It seems to me that the historical discipline is divided between those who underestimate and those who overestimate the limits and potentialities of quantitative methods. This is a result, in part, of an inefficient application of quantitative techniques and the related computer technology to historical data. The most cautious quantifiers use the computer as an adding machine; the most zealous assume that quantitative methods can answer all questions. Statistical and computer methods have been

more »

... ost successfully used in systems analysis where the analyst realizes that some portions of a system are best understood in non-quantitative terms and has no hesitation about using a variety of methods. Simple counting, which is perfectly legitimate method of quantifying, has led to some interesting historical insights. Both Alan Simpson in British history and Stephan Thernstrom in American history have provided the profession with insightful scholarship using methods available to a high school student. Their ideas and conceptual frame of reference, however, have given their rather simple methods grandeur. Far more sophisticated methods are available, however. Robert Fogel, for example, developed a model for nineteenth-century American economic structure in order to test the assumption that the railroad was the primary agent of economic change. By using counter-factual assumptions, he removed the railroad from the model. I would not suggest for British history that the gentry be removed from the sixteenth and seventeenth centuries, but some application of the counter-factual assumption to the economic structure of the empire might prove rewarding. In my own work, I have found the use of simulation models worthwhile in analyzing the structure of cities and the effects of technological innovation on society. Obviously I am addressing myself to the need for a more rigorous style in historical analysis rather than quantitative applications. Many non-quantifiers have provided excellent examples of analytical history that quantifiers should study more carefully (e.g. Marc Bloch and Perry Miller). We should use those methods that make sense in the context of the to'al system we study.