Social science history and historical social science

I talked recently with Tom Sugrue about his approach to historical research, and he had quite a few interesting things to say. (Tom is professor of history and sociology at the University of Pennsylvania and author of The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit. The interview is posted on YouTube.) One topic we discussed was the relationship between historical research and social science research -- especially those areas of social science research that take history seriously. Tom's central observation is that historians pay very close attention to the empirical and historical data that they work with -- the surprises and gritty texture that will be encountered in the archives. But historians sometimes lose track of the larger questions that ought to give focus to their work. This is where historical social science can be helpful; the social scientists are interested in large questions such as power, class, race, or economic structure. But the social scientists have their own symmetrical weakness -- they often give too little attention to the empirical details of the cases or events they include in their analysis. So Tom seems to be saying that history and social science can both contribute most strongly when the macro-disciplines work together, bringing theoretical vision and factual specificity into play.

This contrast has come up quite a bit in the past twenty-five years. Barrington Moore and Theda Skocpol offered social science theories of the causes of large political upheavals such as dictatorship and revolution. Moore's Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World was pathbreaking in the way it defined the intellectual challenge of explaining fascism and dictatorship. In Skocpol's important book States and Social Revolutions: A Comparative Analysis of France, Russia and China she offered a comparative treatment of the Russian, French, and Chinese Revolutions, in an effort to tease out the social causes that brought about successful revolutions in these cases. This is an important and compelling question for social scientists, and Skocpol's analysis has been highly influential. But historians of each of those revolutions often complained that her treatment wasn't historical enough: it wasn't based on her own archival work, it was more abstract than a good history of the French Revolution would be, and it was offered as an effort to arrive a some causal generalizations -- rather than an account of this one specific messy historical complexity. So there was a macro-disciplinary difference of perspective between the comparative historical sociologists (Moore, Skocpol, Goldstone) and the historians.

I am inclined to think that the tension between the two disciplines is inherent and healthy; the historians and the comparative sociologists are trying to accomplish fairly different intellectual tasks. The comparative sociologist is looking for some sort of causal or structural similarity across cases -- instances of dictatorship, revolution, or labor union -- and necessarily reduces the historical complexity of the cases to an analytical framework. This means putting aside much of the messy complexity of the actual cases -- the particular strategies used by the Chinese Communist Party in a base area, the rhetoric of competition between the Parisian parties in 1790, the accidents of history that intruded into the particular cases. The historian, on the other hand, is primarily interested in the particulars. How did the struggle for civil rights for African-Americans proceed in this city in this time period?

We might push the question a little bit deeper and ask, what resources can the social sciences offer working historians? One part of the answer is conceptual: social scientists have framed a number of conceptual frameworks in terms of which to characterize and interrogate historical reality. Marx's theory of class, Durkheim's theory of anomie, Tocqueville's highlighting of civic associations -- these are all instances of an effort by a social theorist to formulate a concept and a set of correlative theories in terms of which to analyze the historically given. Second, social theorists devote much of their intellectual energy to discovering and analyzing common social mechanisms -- free-rider problems, class conflict, collective action, ethnic violence. Historians can benefit by borrowing from each of these areas of knowledge. And third, some people think that social scientists aim to discover laws or regularities that govern social phenomena. If this were so, then the task of the historian would be very simple: go through the relevant social science literature, dredge up the pertinent laws, and explain the particular in terms of the workings of the laws. Unfortunately, no such laws exist; there are no "laws of motion" of modern society. So the historian's intuition -- that every historical event has its own individuality -- is born out. At the same time, the social sciences can provide concepts and mechanisms on the basis of which the historian can do a better job of formulating and explaining the historical event of interest.

As Tom Sugrue says in the interview, both these perspectives benefit from deep immersion in the findings and research efforts of the other. But I'm inclined to think that they are not simply different ends of a spectrum. Rather, they are different kinds of intellectual activity, and the criteria of success are different as well.