The controversial question of whether the millennium began on January 1, 2000 or on January 1, 2001 weighed heavily on my mind as I composed my talk. The dilemma is, I should think, evident: Should a historian of thought want to give the last address of the old millennium, or the first address of the new one? I couldn't make up my mind about this, so I decided to consult my friends, particularly those whom I felt had given the question their considered attention. I asked David Colander, and he assured me that my preferences didn't matter, since he in fact had certainly given the final presidential address of the last millennium when he spoke in Greensboro the previous June. I then asked John Davis, and he assured me that my preferences didn't matter since he, in fact, was certainly going to give the first one of the new millennium, in 2001. Now being not nearly so stupid as I look, it occurred to me that if I listened to both of these individuals, there was a chance that I was going to give neither the last address of the old nor the first address of the new millennium. Feeling that this was, to put it mildly, an unacceptable outcome, I decided to disregard the ill-considered advice of my so-called friends and take firm control of the situation. Ladies and gentleman, I am very proud to be before you here today, delivering both the last presidential address of the old millennium and the first of the new one.

According to a commonly held view, doctrinal history formed a largely uncontested part of the discipline of economics in the early years of the twentieth century. Economists like Edwin Cannan, Jacob Viner, and Joseph A. Schumpeter were at the same time respected economists and historians of economics. Contemporary historians of economics, on the other hand, tend to feel defensive about their field of study. The questions of why, how, and in which discipline one should pursue the history of economics is hotly debated among practitioners, while the number of universities and curricula still offering history courses is in steady decline. This is matched by a corresponding attitude among orthodox economists aptly summarized by Frank H. Hahn (1992, p. 165): “What the dead had to say, when of value, has long since been absorbed, and when we need to say it again we can generally say it much better.”

The place for the teaching of and research dealing with the history of economics continues to draw attention, at least from within our speciality. Recent examples are the History of Political Economy mini-symposium (1992) that was anchored to Margaret Schabas's paper and Donald Walker's affirming essay in the Journal of the History of Economic Thought (1999). The former asked whether the history of economics, given the neglect shown by mainstream economics, should be incorporated into history of science programs. The latter argued that this neglect is unfortunate because there is a place for the history of economics in the training of economists. (The neglect is documented by Aslanbegui and Naples (1997).) An important feature of Walker's essay is that the argument was based on structural aspects of the discipline. His point of departure was Donald Gordon's (1965) claim that there is little in the history of economic analysis that facilitates the understanding of current economics and the research problems with which it is concerned. Certainly the education of natural scientists, as Gordon noted, is consistent with this claim. George Stigler (1969) made a similar claim, and both more or less took the conclusion to be self-evident. Walker's response was that, at least among a variety of important sub-disciplines, there is not professional consensus and, therefore, there may not be a sharp distinction between current and past theory. This suggests that some form of historical analysis may be useful. Walker's paper moved the discussion away from arguments that begin with the presumption that every educated economist should be versed in the origins of economics. Instead, he focused on questions relating to history as a tool for understanding and for doing today's economics. In this respect, Walker returned to an issue addressed by Filippo Cesarano (1983)—they both focused on historical studies as a source of illumination for current economic analysis.

The individual's labor supply model, usually called leisure-income, is very well known in labor economics. It determines the time a person spends on the labor market by studying it as the outcome of a trade-off between leisure and income. The model is developed in almost every book on microeconomics. But strangely enough, although all economists know the model, no one seems to know precisely its origins.

Popular texts typically assert that guilds of craftsmen “monopolized” markets in medieval England. Norman Cantor's Medieval Reader declares “craft guilds' … main purpose and activity was narrow regulation of industrial productivity in order to restrain competition” (Cantor 1994, p. 278). Douglass North's Structure and Change in Economic History asserts “… guilds organized to protect local artisans … [and their strength] in preserving local monopolies against encroachment from outside competition was frequently reinforced by the coercive power of kings and great lords” (North 1981, p. 134). Henry Pirenne's Economic and Social History of Medieval Europe proclaims:

craft guilds, under a diversity of names, officium or ministerium in Latin, metier or jurande in French, arte in Italian, ambacht or neering in the Netherlands, amt, innung, aunft, or handwerk, in German, craft-gild or mistery in English, … fulfilled the need of economic protection. The pressing necessity to stand by one another, so as to resist competition from newcomers (Pirenne 1937, pp. 177–79).

Most discussions of how one should write the history of economics are conducted at a very abstract level. They debate the merits of alternative approaches without relating them to specific problem-situations or specific periods. The literature uses terms such as “absolutism,” “relativism,” “Whig history,” “thick history,” “historical reconstructions,” “rational reconstructions,” and “presentism,” frequently arguing that one of these approaches is better than one or more of the others. In warning historians of economics against sin, they typically define sin in absolute terms. This is particularly true of those who argue for “relativism,” “thick history,” “historical reconstructions” and the like. There is a simple reason for this. Terms such as “absolutism,” “Whig history,” and “presentism” all carry negative connotations, with the result that their advocates usually feel impelled to qualify them. On the other hand, supporters of “historical reconstruction,” “thick history” and so on feel much less pressure to offer such qualifications—the words used have the sound of “real” history. The terminology itself is value-laden.

About three lines into his book Making Natural Knowledge, Jan Golinski defines the “constructivist” view of science as “that which regards scientific knowledge primarily as a human product, made with locally situated cultural and material resources, rather than as simply the revelation of a pre-given order of nature” (Golinski 1998, p. ix). Having a pretty good idea of what Professor Golinski's book was going to be about, this stopped me for a moment. Is that all there is to it? Is that what the work of the Strong Programme, Latour, Pickering, Knorr-Cetina; the talk of mangle, black boxes and resistances, is all about? Described this way, it seems quite reasonable; a way of thinking about scientific activity that should at least be given a hearing. Admittedly, I have seen other, usually more prolix, ways of characterizing what Golinski has called the constructivist view of science. But I like this one. It sends what I believe to be the correct message that this “constructivism” is, at its heart, better characterized as plausible and intriguing than outrageous, and that it is not necessarily threatening to any core values held by historians.

Eleven years ago, at my first HES conference, I attended a session on the historiography of economics. In my naiveté and brashness (not a good combination!) as a grad student, I asked why so much of the conversation about how to do historical work in economics was dominated by reference to methodological categories provided by philosophers of science like Lakatos, Popper, and Kuhn. I enjoyed my own investigations in the philosophy of economics, and was starting to write about Frank Knight, who was certainly not naive philosophically, but I wanted a different reference point as a historian of economics. I remember one member of the panel looking at me somewhat puzzled and asking, “Where else would you look?”

Today, we gather to reflect, once again, on the historiography of economics. This time, we are asked to reflect on a new reference point for our work: the sociology of scientific knowledge or science studies. Jan Golinski has provided us with an excellent survey of work in this field, and a synopsis of its relevance over the past fifteen years or so to the history of the natural sciences. His book is not a rallying cry for constructivism, but rather a careful analysis of the benefits and costs to historians of science if they choose to adopt constructivist techniques.

When I went through my graduate training in economics at Stanford University, I learned that economics consists of ideas. These are often expressed in mathematical terms and can be found in books and articles. To become a successful economist, you have to understand these ideas and come up with variations on them. So, during my first year of graduate studies, I spent the majority of my time working my way through the many books and articles assigned for my classes in microeconomics, macroeconomics, and econometrics. During their lectures, our instructors would walk us through any difficult mathematical manipulations that we encountered in our readings. Surely, I thought, I was on my way to becoming a respected economist. Still, I needed to specialize in a few fields, which I did in my second year.

While most scientists and philosophers of science privilege scientific knowledge, and have sought demarcations of science from non-science to justify the privilege, sociologists of science, small numbers of philosophers of science, anthropologists, and some scientists themselves have been attracted to a new way of talking about science. Prefigured by Ludwik Fleck (1935/1979) and Gaston Bachelard (1934/1984), nurtured by the controversies over Thomas Kuhn's work, and instantiated in the Edinburgh School's Strong Program, the naturalistic turn portrays science as a human activity, part of the woof and warp of culture itself. Yet curiously historians of science have been less involved in this recent reconceptualization of both science and scientific knowledge.