All attempts to define American institutionalism, whether in terms of a set of key methodological or theoretical principles or in terms of the contributions of the three generally accepted “founding” figures of Thorstein Veblen, Wesley Mitchell, and John R. Commons, have run into a problem with the apparent disparities within the movement. In terms of the three “founders” there are obvious and quite dramatic differences between the methodologies and theoretical directions of the three men. Veblen is associated with an evolutionary approach, a key distinction between pecuniary institutions and technological or industrial requirements, and a biting critique of orthodox theory and business practices; Mitchell with quantitative methods and detailed research on business cycles, an approach he established at the National Bureau of Economic Research (NBER); Commons with documentary histories, work on labor issues and public utility regulation, and an analytical scheme emphasizing the evolution of legal institutions and processes of dispute resolution. The same problem shows up with more explicit types of definition that often seem to capture only some parts or aspects of the movement and not others, or are so broad as to lack much specific content. Institutionalism easily appears as incoherent, as little more than a set of individual research programs with nothing in common other than a questioning of more orthodox theory and method. Thus, Mark Blaug has stated that institutionalism “was never more than a tenuous inclination to dissent from orthodox economics” (Blaug 1978, p. 712), and George Stigler has claimed that institutionalism had “no positive agenda of research,” “no set of problems or new methods,” nothing but “a stance of hostility to the standard theoretical tradition” (quoted in Kitch 1983, p. 170). This view still finds wide currency— for example Oliver Williamson has recently argued that “unable or unwilling to offer a rival research agenda, the older institutional economics was given over to methodological objections to orthodoxy” (Williamson 1998b, p. 24; see also 1998a).

A central message of Allyn Abbott Young's (1928) seminal paper on increasing returns and economic progress was the seeming tautology that “the division of labor depends in large part upon the division of labor.” From this he deduced that a major source of growth is growth itself. In this fundamental sense growth is endogenous and cumulative. He spoke of the “togetherness” of economic phenomena and doubted that the apparatus of supply and demand and marginal productivity theory could be integrated to give the social picture or explain why growth tends to be self-sustaining rather than self-exhausting.

Frank Steindl poses a surprising question in the title of his 1997 article, “Was Fisher a Practicing Quantity Theorist?” and reaches the conclusion that, “Clearly, with the decade of the Great Depression, Fisher was no longer a practicing quantity theorist” (Steindl 1997, p. 259). Such a change in Fisher's monetary economics would sharply revise the view of Irving Fisher generally prevailing in the history of monetary economics, which is based primarily on The Purchasing Power of Money (Fisher with Brown 1911). Fisher's photograph (along with photographs of Marshall and Wicksell) appears on the cover of The Golden Age of the Quantity Theory (Laidler 1991). As Mark Blaug (1995, p. 3) put it, “isn't Irving Fisher the quintessential quantity theorist if there ever was one [?]” Perhaps the most striking tribute to Fisher in the quantity theory tradition is from Milton Friedman, who, addressing the American Economic Association on the question “Have Monetary Policies Failed?” and having quoted from Fisher's 1910 exchange with J. L. Laughlin, remarked “And now I must cease quoting from Fisher, with whom I am in full agreement, and proceed instead to plagiarize him—albeit with modifications to bring him down to date” (Friedman 1972, p. 12).

I do not think that it would serve much of a purpose to answer Hollander and Peart (1999) point by point, to repeat old arguments, and to once again cite long quotations from Mill. Furthermore, I doubt that the Editor of this Journal would give me the space and time to do so. Instead I propose to ask a number of basic questions about Mill's methodology, give my answers to them and show why I find fault with what I take to be Hollander and Peart's answers. I doubt that this will convince my adversaries. But I hope to show our readers what the major issues are in the disagreement and, perhaps, even to get Hollander and Peart to agree that these are the issues that separate us. If I fail in this latter task they will have the opportunity to point this out in their Rejoinder.

The dispute between Hollander and Peart, and Hirsch, turns on the nature and role of verification in Mill's perception of the appropriate method for Political Economy. Professor Hirsch maintains against us that, for Mill, the models constructed by political economists are insulated from verification. His case is based on two counterclaims. First, that when Mill writes of “verification” in Book III of the Logic, he has in mind a procedure differing from that appropriate for Political Economy, which allows only “indirect verification” (outlined in Book VI). Hirsch finds that Hollander and Peart confuse the two. Secondly, since the contexts of our case studies often relate to policy formulation, Hirsch finds our elucidations of an appeal to experience of a more basic order to be unconvincing.

Rod O'Donnell's (1999a, 1999b) and James Ahiakpor's (1999) contributions on the origin of the only diagram in J. M. Keynes's General Theory raise some interesting points regarding the use of documentary evidence for interpreting authors' ideas or specific passages in their writings.