Comment on Geoffrey M. Hodgson's "What Are Institutions?"
Wilson, Matthew, Journal of Economic Issues
This comment argues a point that may seem obvious to some readers. Hodgson's definition of "institutions" is best understood by situating his theory within the larger intellectual context of the ontological movement in philosophy and social science, which has occurred over the past quarter century. The critical realist movement is perhaps the best-known expression in economics of the movement to restore ontology. Hodgson has much in common with critical realism, although he does not appear uncritically to accept any particular formulation of that doctrine (e.g. 2004a; 2004b).
While it is not my intention to label Hodgson or attribute anyone else's views to him, I want to suggest that his definition of an institution is certainly consonant with, and should be understood as being positioned broadly within, the ontological movement in social theory. I myself advocate a renewed emphasis on ontology, although I have expressed some criticisms regarding how the ontological turn has been carried out in economics (Wilson 2005a; 2005b).
Institutions Defined Without Deference to Behaviorism
Hodgson's (2006) article should be applauded for revisiting one of the most basic conceptual problems that has always faced institutionalism, namely how to define institutions. He proposes the following definition: Institutions are "durable systems of established and embedded social rules that structure social interactions.... In short, institutions are established social rule systems, not simply rules" (13, emphasis in original).
His article does not beat around the bush when it comes to criticizing earlier definitions of institutions. For instance, he writes: "some institutionalists such as John Fagg Foster (1981,908) have misleadingly defined institutions as 'prescribed patterns of correlated behavior'" (Hodgson 2006, 2). Similarly, in a footnote, he comments: "Tony Lawson (2003a, 189-94) listed several behavioral definitions in the institutionalist literature and rightly criticized them" (Hodgson 2006, 21).
But what is wrong with defining institutions in terms of behavior? In order to understand the objection, it is helpful to locate Hodgson's theory within the larger context of the late twentieth century ontological movement in philosophy and social science (e.g. Taylor 1985; Bhaskar 1975; 1989; Archer 1995; Lawson 1997; 2003b; among others). In particular, this movement involves the rejection of behaviorism and its positivist mooring.
In the 1920s, as a way of separating institutionalism from Veblen's instinct-habit psychology, Clarence Ayres and Morris Copeland initiated efforts to bring behaviorism into institutionalism (Hodgson 2004a, 263-69, 394). Behaviorism, is positivist philosophy pressed into play as a research methodology in social science (Taylor 1985; Hodgson 2004a). Given that so many institutionalists have advocated pragmatism, it seems likely that this connection between behaviorism and positivism has not always been well understood.
Pragmatism characteristically involves an anti-foundationalist position, in the sense that all knowledge is viewed as fallible. However, that position should be distinguished from the anti-metaphysical stance of the positivists. For example, the American pragmatist John Dewey, who has inspired many institutionalists, criticized what he called spectator theories of knowledge. He wrote that: "Logic provided the patterns to which ultimately real objects had to conform, while physical science was possible in the degree in which the natural world, even in its mutabilities, exhibited exemplification of ultimate immutable rational objects" (Dewey 1929, 16). However, Dewey's rejection of this doctrine of eternal forms should not be mistaken for an antimetaphysical stance. Clearly, it involves an anti-foundationalist position, but this position does not involve a rejection of metaphysics per se.
The positivists, in contrast, did pursue an anti-metaphysical agenda. …