Daniel Little looks at the role of theory in the social sciences:
Social theory and the empirical social world, by Daniel Little: How can general, high-level social theory help us to better understand particular historically situated social realities? Is it helpful or insightful to "bring Weber's theory of religion to bear on Islam in Java" or to "apply Marx's theory of capitalism to the U.S. factory system in the 1950s"? Is there any real knowledge to be gained by applying theory to a set of empirical circumstances?
In the natural sciences this intellectual method is certainly a valid and insightful one. We gain knowledge when we apply the theory of fluid dynamics to the situation of air flowing across a wing, or when we apply the principles of evolutionary theory to the problem of understanding butterfly coloration. But do we have the same possibility in the case of the social world?
My general inclination is to think that "applying" general social theories to specific social circumstances is not a valid way of creating new knowledge or understanding. This is because I believe that social ensembles reflect an enormous degree of plasticity and contingency; so general theories only "fit" them in the most impressionistic and non-explanatory way. We may have a pure structural theory of feudalism; but it is only the beginning of a genuinely knowledge-producing analysis of fourteenth-century French politics and economy or the Japanese samurai polity. At best the theory highlights certain issues as being salient -- the conditions of bonded labor, the nature of military dependency between lord and vassal. But the theory of feudalism does not permit us to "derive" particular features or institutions of French or Japanese society. "Feudalism" is an ideal type, a heuristic beginning for social analysis, rather than a general deductive and comprehensive theory of all feudal societies. And we certainly shouldn't expect that a general social theory will provide the template for understanding all of the empirical characteristics of a given instance of that theorized object.
Why is there this strong distinction between physical theory and social theory? Fundamentally, because natural phenomena really are governed by laws of nature, and natural systems are often simple enough that we can aggregate the effects of the relevant component processes into a composite description of the whole. (There are, of course, complex physical systems with non-linear composite processes that cannot be deductively represented.) So theories can be applied to complicated natural systems with real intellectual gain. The theory helps us to predict and explain the behavior of the natural system.
The social world lacks both properties. Component social mechanisms and processes are only loosely similar to each other in different instances; for example, "fealty" works somewhat differently in France, England, and Japan. And there is a very extensive degree of contingency in the ways that processes, mechanisms, agents, and current circumstances interact to produce social outcomes. So there is a great degree of path dependency and variation in social outcomes, even in cases where there are significant similarities in the starting points. So feudalism, capitalism, financial institutions, religions, ethnic conflicts, and revolutions can only be loosely theorized.
That is my starting point. But some social theorists take a radically different approach. A good example of a bad intellectual practice here is the work of Hindess and Hirst in Pre-Capitalist Modes Of Production, in which they attempt to deduce the characteristics of the concrete historical given from its place within the system of concepts involved in the theory of the mode of production.
Is this a legitimate and knowledge-enhancing effort? I don't think that it is. We really don't gain any real insight into this manor, or the Burgundian manor, or European feudalism, by mechanically subsuming it under a powerful and general theory -- whether Marx's, Weber's or Pareto's.
It should be said here that it isn't the categories or hypotheses themselves that are at fault. In fact, I think Marx's analysis and categories are genuinely helpful as we attempt to arrive at a sociology of the factory, and Durkheim's concept of anomie is helpful when we consider various features of modern communities. It is the effort at derivation and subsumption that I find misguided. The reality is larger and more varied than the theory, with greater contingency and surprise.
It is worthwhile looking closely at gifted social scientists who proceed differently. One of these is Michael Burawoy, a prolific and influential sociologist of the American labor process. His book, Manufacturing Consent: Changes in the Labor Process Under Monopoly Capitalism, is a detailed study of the American factory through the lens of his micro-study of a single small machine shop in the 1940s and 1970s, the Allied/Geer factory. Burawoy proceeds very self-consciously and deliberately within the framework of Marx's theory of the capitalist labor process. He lays out the fundamental assumptions of Marx's theory of the labor process -- wage labor, surplus labor, capitalist power relations within the factory -- and he then uses these categories to analyze, investigate, and explain the Allied/Geer phenomena. But he simultaneously examines the actual institutions, practices, and behaviors of this machine shop in great participant-observer detail. He is led to pose specific questions by the Marxist theory of the labor process that he brings with him -- most importantly, what accounts for the "consent" that he observes in the Allied workers? -- but he doesn't bring a prefabricated answer to the question. His interest in control of surplus labor and coercion and consent within the workforce is stimulated by his antecedent Marxist theory; but he is fully prepared to find novelty and surprise as he investigates these issues. His sociological imagination is not a blank slate -- he brings a schematic understanding of some of the main parameters that he expects to arise in the context of the capitalist labor process. But his research assumptions are open to new discovery and surprising inconsistencies between antecedent theory and observed behavior.
And in fact, the parts of Burawoy's book that I find most convincing are the many places where he allows his sociological imagination and his eye for empirical detail to break through the mechanism of the theory. His case study is an interesting and insightful one. And it is strengthened by the fact that Burawoy does not attempt to simply subsume or recast the findings within the theoretical structure of Marx's economics.
(Burawoy addresses some of these issues directly in an important article, "Two Methods in Search of Science" (link). He advocates for treating Marxist ideas as a research program for the social sciences in the sense articulated by Imre Lakatos. )
So my advice goes along these lines: allow Marxism, or Weber or Durkheim or Tilly, to function as a suggestive program of research for empirical investigation. Let it be a source of hypotheses, hunches, and avenues of inquiry. But be prepared as well for the discovery of surprising outcomes, and don't look at the theory as a prescription for the unfolding of the social reality. Most importantly, don't look to theory as a deductive basis for explaining and predicting social phenomena. (Here is an article on the role of Marxism as a method of research rather than a comprehensive theory; link.)