“Let us remember Mr. Justice Holmes’s dictum: ‘I always say that the chief end of man is to form general propositions,’ and let us not altogether forget what he added: ‘And no generalization is worth a damn.’”
-- George C. Homans, The Nature of Social Science (Harcourt Brace & World, 1967), pp. 9-10.
“Reification” (see this earlier post and comments) features in an article published last fall which I recently read: Mark Bevir and Asaf Kedar, “Concept Formation in Political Science: An Anti-Naturalist Critique of Qualitative Methodology,” Perspectives on Politics 6:3 (September 2008), pp.503-517. (Here is a link to a PDF version of the article. Note, however, that the page numbers in this post refer to the article as published in Perspectives, not the PDF version.)
Although the public may not generally realize it, social scientists (as well as philosophers) have been quarelling for over a century about what social science is and how it should be done. Are the social sciences immature cousins of the natural sciences, disciplines that can be expected eventually to produce a comparable body of cumulative knowledge? Or does the subject matter of the social sciences, namely human action, make them different in kind from the natural sciences, orienting them toward “understanding” rather than prediction, causal explanation, and control? Or is there a viable middle ground between these two views, one that sees very significant differences between the social and the natural sciences but holds that “general propositions” are not entirely beyond the reach of social science?
The authors of the article in question have a clear position on these issues. Bevir, a political science professor at Berkeley, and Kedar, a graduate student there, are critics of “naturalism,” the notion that the social sciences “should strive to develop predictive and causal explanations akin to those found in the natural sciences” (p.504). In their view, this aim ignores that human action “is meaningful and historically contingent” (p.505), and always rooted in particular contexts. Human conduct is inherently subject to choice and chance, hence inherently unpredictable by general laws – thus the argument runs.
So where does reification come in? Bevir and Kedar (hereafter B&K) argue that the work of political scientists Giovanni Sartori and David Collier on concept formation reflects flawed “naturalist” premises. Specifically, this work, according to the authors, is marred by “reification, essentialism, and an instrumentalist view of language” (p.507). For B&K, “reification” equals insufficient attention to the “meaningfulness” of human action. Here’s the nub of what they say on it:
“Anti-naturalism implies that many – perhaps even all – social science concepts denote objects that are composed at least in part of meanings or intentional states. Reification occurs whenever these concepts are defined either in ways that neglect relevant meanings entirely or in ways that neglect the holistic character of meanings, thereby likening human action to meaning-less ‘things’…. Reification occurs whenever the attributes of a social science concept are regarded as reducible to causal laws, probabilities, or fixed norms. For example, the concept of ‘social class’ is reified insofar as it is understood in terms of supposedly objective socio-economic criteria such as relation to the means of production or income level, without taking into account how the members of a given social class themselves construe and experience their social situation.” (p.507)
As it turns out, what bothers me about this article is not so much the way the authors define “reification” as their conviction that reification (as they define it) is always bad or unwarranted. Take the notion of social class. As noted above, B&K condemn any treatment of class that defines it ‘objectively’ (e.g., in terms of income level or relation to the means of production) without also considering “how the members of a given social class themselves construe and experience their social situation.” However, there are times when a “reified” definition of social class is defensible, especially if one accepts (as B&K do not) the possibility of trans-historical generalizations. If, for example, you're interested in the role played by peasant revolts in social revolutions (or in the conditions under which such revolts have occurred), you may be more concerned with the institutional character of agriculture – e.g., are there mostly large estates or mostly small holdings? – than with the ways in which a particular group of poor farmers has construed its own situation. Moreover, the latter type of information may be difficult or impossible to assemble, since peasants have often been largely or entirely illiterate (thus leaving behind few written records in their own words), and since you can’t interview people who are dead. It’s not always possible, in short, to arrive at concepts “through a kind of dialogue with the social actors being studied” (p.508), desirable as that might often be. Although they do not say this explicitly, B&K’s position would discard such modern classics of historical sociology as Skocpol’s States and Social Revolutions, Wallerstein’s The Modern World-System, Tilly’s Coercion, Capital, and
This all points to a more basic issue: Is there only one correct, philosophically defensible way to do social science? Some scholars believe that only an approach aimed at causal explanation is valid. B&K take the opposite side but adhere to an equivalent exclusiveness. The implication of their position seems quite clear: only one kind of social science will pass muster.
B&K are opposed to, or at least highly skeptical of, any effort at explanatory generalization. They admit the possibility of generalization but deny that generalizations explain anything: “[W]e can say that X, Y, and Z are all democracies but that does not explain any other feature they might have in common” (p.506). Not only do B&K posit, as some other writers have, a sharp dichotomy between “understanding” and “explaining”; they hold that only understanding is legitimate, since explaining is tied to a “discredited” naturalist perspective. Although their article is subtitled “a critique of qualitative methodology,” it is actually a broader critique. While I’m sympathetic to certain of their points and while I certainly do not believe that all good social science must involve causal explanation, B&K have not persuaded me of the illegitimacy of any social science that aspires, however tentatively and imperfectly, to discern causes of social phenomena.
1. B&K observe that Collier defines reification as neglect of contingency and historical flux, whereas they define reification as neglect of “the meaningful or intentional nature of action” (p.511). They argue that their definition “better reflects the source of the concept in Hegelian and Marxist contexts where reification is the process whereby external objects are detached from their relation to (and origin in) human consciousness” (note 82, p.515).
2. There is disagreement in the literature about what counts as “explanation.” I won’t go into this here. See B&K’s note 24, p.514.