History of Economics Playground

Disaggregate, disaggregate!

Last June at a History of Social Science workshop , David Engerman presented a paper on the Harvard's Refugee Interview Project (1950-1954). He told the subtle story of a project which was set up by some sociologist and anthropologists with no specific soviet studies background, which was funded by Air Force officials whose academic background entrusted them with intellectual agenda different from the broader -and changing- concerns of their army. He did not only document the tensions between academic freedom and war needs for intelligence, but he also touched upon the consequences of very concrete hurdles (difficulty to interview Russian refugees in Germany because of clearance issues) and details the unintended offsprings of this failed project (the training of a set of graduate students who later became true sovietologists).

Unsurprisingly, subsequent discussions developed along the lines discussed in the 2010 ISIS issue on Cold War scholarship, to which Engerman is a contributor (see Will Thomas's review of the issue). The audience was divided between those fed up with the “everything cold war” interpretation of postwar science and those who feared that the kind of detailed case study analysis performed by Engerman would make impossible any generalization regarding the influence of the Cold War context.

For Engerman's case study is indeed embedded in a wider vision. During his talk, he repeated on several occasions one central idea: “disaggregate, you have to disaggregate.” (funding, in the present case). A line of thought he elaborates in his ISIS piece :

Social science in the Cold War, in this formulation, equals Cold War social science. Yet an increasing number of historians are challenging that blithe equation by examining social scientists' complex relationship to the means of scholarly production; this essay will examine at least three trends in this new scholarship. First, some scholars are broadening their historical perspective by showing how social scientists in the Cold War expanded upon the ideas of their scholarly predecessors. Second, scholars are detailing the dramatic institutional changes of World War II to show how social scientists' service in the “Good War” shaped structures and assumptions of postwar intellectual life. And finally, more historians are following sponsors' money not just into scholars' bank accounts (or summer homes) but into their publications. All three approaches give more agency to social scientists, depicting them as scholars responding to older ideas while working in a new institutional environment and navigating complex relationships with their patrons.

This idea struck a chord with me. First, because of my growing discomfort toward the literature on economics and neoliberalism in the postwar era, in which it is often assumed too easily that such and such research had ideological underpinnings on the sole basis that it was funded by a conservative patron or that its author was a member of a neoliberal organization, such as the Mont Pelerin Society. More fundamentally, the more one dives into the structure of the patrons of postwar economics (from foundations to military agencies to public bodies, NSF etc.), the less it seems that generalizations can be derived from funding patters. For instance, the statement that “economic project X was founded by Ford Foundation” conveys little information in the end. Depending whether the money was granted by Ford Area III or Ford Area V, by MIT economist and CIA alumni Richard Bissell in 1951, by Chicago political scientist Bernard Berelson in 1953, or by Yale labor economist Lloyd Reynolds in 1955, the underlying agenda had to do with Cold War intelligence, the construction of a broad interdisciplinary social science, or the advancement of theoretical economic programs.

We have to “begin to deal more seriously with the fact that, while [scientific] work certainly existed within the Cold War (who would argue otherwise?), it cannot be satisfactorily characterized as a Cold War phenomenon.  Other illuminating contexts must be found.” Will Thomas argues in his review of the ISIS issue. True, but it's not only a question of contexts. More compelling ways to tie old and new contexts to the personal histories and worldviews of the researchers involved, to their commitments, ethical and political leanings, religion, culture, understanding of the role of scientists within society are needed. To chasten the “Everything Cold War” history, one also need to debunk “Puppet history,” that is, narrative in which a wide set of scientists straightforwardly endorse the vision of a “leader” (Simon, Von Neumann, Hayek, Samuelson and Keynes have been fashionable in the past years) or the agenda of a patron.

If such ideas are adopted by historians, then the tricky issue of course  becomes “How do you (re)aggregate?” How do you evolve from a maze of microhistories into an overarching narrative of the evolution of economics as a discipline? Will Thomas ties his review to a former post on historiagraphy in which he explains that historians should strive to make the history of the era more “navigable,” by writing papers they are able to “hook” to narratives on comparable institutions, milieu, research problems, communities. Such venture requires a clear picture of how we historians construct our sets of narratives. And our historiography is still fragmentary.  It seems, for instance, that the research on the Chicago School of Economics evolved from a focus on a few keys individuals (Friedman, Stigler, Hayek, Knight, Viner, Becker) to a larger perspective that included new figures, extra-academic institutions, extra-economic bodies, and subdiciplines. Yet, I still have little idea whether they was any early concerted effort to write such history. The scholarship on MIT, Samuelson and neokeynesians seems to develop differently, although it's a bit early to say. Any instance of a successful bottom up approach of the history of a group/subdicipline?




In a forthcoming paper ('A history better served cold' to be published in The Cold war in Pieces, Joel Isaac and Duncan Bell eds), Phil Mirowski, author of some of the broadest and most famous narrative on Cold War science, responds to historians' recent attempts to reevaluate the concept of “Cold war” as an organizing principle to write postwar history.  He challenges the “subculture performativity” approach he says characterizes Joël Isaac's writings. He recognizes that contemporary historians are now able to discern distinctive feature of the “Cold War” science, but ties it to a profound change in the historians' knowledge regime, a "regime of globalized privatized science.” I have to read more from Isaac and I'm still struggling to understand what is at stake in Mirowski's claim. The core of Mirowski's paper, namely its “don't throw the baby with the bathwater” admonition, nevertheless calls for further consideration. In his effort to show how "frosty" the atmosphere of the forties to eighties was, he outlines three ideas he thinks are worth retaining in future approaches of the cold war intellectual life:

(1) the cold war produced a specific time frame, common to various hard and social sciences, in which it was believed technological and intellectual progress could be planned (a belief to be replaced with the “market of ideas” philosophy in the eighties.)

(2) the interdisciplinarity paradox: the Cold War military patrons are responsible both for the highly disciplinary organization of science within universities, Mirowski argues, and for the foundation of several transdiciplinary independent bodies aimed at fostering creativity in emerging areas.

(3) the intellectual structure of cold war science was grounded in “closed worlds” representations (eg in terms of systems sufficient upon themselves).



I dot not know whether it is going to add much to your inquiry, but this made me think of a number of questions that arose to me while reading Didier Eribon's excellent biography of Foucault (of which a Second edition has recently been published in French). At some point, Eribon evokes Jean-Luc Godard's criticism of Foucault: in his movie La Chinoise, you see one character throwing tomatoes at The Order of Things. When asked about the meaning of this scene in the movie in an interview, Godard said he did not like the kind of thinker who writes things such as: "at that time, people were thinking like this, whereas at some other point, they were thinking like that". "How can anybody know how people as a whole think?", Godard asks. And he adds that it is to prevent people like Foucault to assert such things that he wants to make movies. This is kind of funny, but it is also meaningful and interesting. I am sure that when Foucault writes that before the Classical Age, madness was something that was tolerated in the Society and which was not separated from reason (in Erasmus, for instance) and that the separation occured with Descartes, testifying to a change in the way madness is seen in the Western society, then many micro-histories would prove him wrong. Historians of literature and of the arts can certainly locate books and paintings of the so-called "classical" age in which madness is addressed in a way that is different from that which Foucault depicted in his generalization. Is Foucault wrong or can we just say that the exceptions are only derivations from the rule? And can we really say that Foucault's petty inaccuracies preclude his larger depiction of the zeitgeist of 17th century thought? Is it not true that in some ways these errors are not only forgivable but even necessary if one wants to think in the large? I remember a conversation which I had with Loic a few months (years?) ago. Loic argued that there is a division of labor between disciplinary historians (such as historians of economics and sciences like ourselves), who are expected to write micro-histories and to think mostly about individuals and case-studies, therefore participating in an effort of disaggregation, as Engerman would put it, and general historians who tend to write mostly generalizations. The thing is that generalization are always false when we look closely at the facts, while micro-histories, though accurate and interesting, are often less useful. The usefulness-accuracy dichotomy, that Paul Valéry has referred to in a famous sentence which I could not quote at lenght in English, holds pretty well here. As a historian of economics, I really would like to assert some Foucault-like generalizations sometimes, as it would look quite brillian, but sticking to the case-study framework, I think, is safer for publication purposes! On the other hand, think about the difference between Sonja Amadae's Rationalizing Capitalist Democracy and Robert Leonard's Von Neumann, Morgenstern and The Creation of Game Theory. While the latter constitutes a satisfying aggregation of accurate micro-histories without an overarching thesis, the former is more likely to be cited because it has a strong reading of sone of the same events. The sentence "things were more complex than what they seem" signals good research but doesn't ensure success. Pretty depressing, right?


Yann; When I'm depressed about the impossibility to produce a general narrative about a science, a country or an era which would encompass rich and undistorted individual histories, I'd rather think of the sentence “the unintended collective consequences of,” or the ideas of “meeting of mind” and zeitgeist, or the idea that magically, these ideas and practices “coalesced,” or the actors "compromised" although then I'd like to know more about coalescence and compromise processes. What I retain from the Foucault exemple, aside from the Eribon reference which is now sitting in my “great books to read” list, is the idea that you don't write a broad narrative in isolation from the knowledge of some case studies from the relevant period (be it a millenium) or on the relevant subject, and, in the same way, you don't start from scratch on a new subject by focusing only a few case studies: without a broader narrative somewhere in progress, or a broader narrative taken from adjacent subjects, there is no way you can make sense from your own case study. You don't know, as you say, whether such and such phenomenon, project or else is archetypal of an era, or is an exception, or is embedded in a very specific cultural framework but nevertheless bear characteristics of a trend, etc. You can't make sense of it. Therefore I'm not sure about that “division of labour” thing. At least there must be some serious and sustained conservation between microhistorians and broad thinkers.  

Clement: the quote illustrates my own trouble: “case studies are good for deconstructing claims of universality”, but what can we construct from them? Can I get the full paper? What do they propose next?

Also, I don't believe that the analogy with disaggregation issues in data analysis, or even with macroeconomics to be irrelevant. I even indulged in a thought experiment when I wrote that post and realized that its bottom line was about the concealed dynamics of our profession. Imagine that for various reasons (new archives, interests coalescing because of the current history, the direction of past research, personalities, hazards), several research are beginning to work on a new subject (the history of MIT, neuroeconomics, economics in the media or anything else). Imagine that instead of pursuing individual research agenda, asking for grants, organizing various conference, launching competing and sometimes redundant projects,focusing on a few themes, a few key individuals only, we would wait until a network analysis maps out the people, institutions, research areas relating to our subject (which implies a methodology, which implies a preliminary picture of the new area of research, yes, but forget about this). We would wait until the next HES meeting, gather in a room and try to divide the work between us. Would we take into account a wider range of puzzles and struggles, would we end up with a richer picture of our subject, would we be able to “hook” our case studies to one another in a better way? And anyway, would we even know from a network representation what to choose as meaningful case studies? Does recurring links or the biggest nodes necessary represent meaningful historical object. Or do you ex ante need a set of a priori meaningful objects to build your representation on? And how do you built it (historical analysis, semantic analysis ?) (yes, this is a incentive to make up for your fellow bloggers' distressing ignorance of network analysis methodology by writing an introductory post on the subject). 


Beatrice, thanks for the very a propos links to EWP. The struggle between lumpers and splitters seems to show no signs of letting up, however buried the debates over externalism/internalism and social constructionism may be declared to be. I have nothing against lumping, in principle, but am of the opinion that the organization of the historiography is such that through the pervasive Marxism-derived interest in linking history to overarching ideologies (capitalist, militarist, technocratic, etc.), the practical limit of lumping has long since been reached.

The ideology-linking program, I believe, is mainly sustained by imagining that connecting finely-split case studies to lumped ideologies is still a productive and (to outsiders, anyway) surprising activity. This, I believe, is the particular comfort of the historiographical status quo that needs to be questioned vigorously and repeatedly, though I realize that the strategy may still be considered salubrious in a historiography of economic thought where a rote exploration of an established canon still partially prevails.

Thanks very much for taking to heart my embryonic ideas about "hooks" and "navigability" as a strategy for creating an alternative historiography of what Robert K. Merton, and (much more recently) Joel Isaac have called the "middle range". I believe that the chief obstacle to splitting away from lumping's practical limit is not the impossible or unappealing wonkishness of even the most modest splits, but the difficulty of retaining historiographical gains in a scholarly environment that is ill-equipped for the task. Better technologies of navigation are crucial to such retention. Designing and putting in the leg work into creating these literary technologies is, of course, the hard part. Believing that we will alienate a chimerical "broad audience" by splitting is, in my mind, simply an excuse not to get to work. It is also a step toward condescending to the whatever audience we wish to attract, and might attract if our work was more approachable and navigable (rather than candy-coated).

(Of course, despite how widely known he is, no one could ever accuse Foucault of writing to a broad audience. I would agree with criticisms of Foucault as a reckless lumper, but, at the same time, Order of Things is probably my favorite work of his, given that it does have some useful general insights into how the argumentative structure of certain scholarly genres evolved from the 1500s to the 1800s. I believe it is important to think of Foucault as an intellectual historian, whatever populist implications his criticism is deemed to have.)

I'm particularly glad to have a discussion about this with respect to economic thought. Since economics is both a) very important to my work on the history of operations research, and b) not the direction from which I initially approached the topic, I have often felt a very strong, but unfulfilled need for an accessible and nuanced historiography of economics of the interwar and postwar periods. I am excited to see the fruits of the new historiography of MIT economics.

Also on my wishlist: I have long wanted to find a more general appreciation of Kenneth Arrow's economics (i.e., one that does not revolve solely around the impossibility theorem and Arrow-Debreu). I would be very interested to know more about thinking about "insurance" in these periods, which seems crucial, but is not much remarked upon. I have a very strong interest in thinking about inventories as well. Just the other day for the first time I ran into UCLA/Chicago School economist Armen Alchian's 1950 argument for linking evolution to the assumption of rational actors in economics, which I knew had to be somewhere in that milieu, and I was happy to find it right around where I expected it. I would like to know more. It's a bit of a side point for me, but I'd love to know more about the Harvard Keynesians (Hansen, Metzler, Goodwin...). How all of this related or does not relate to linear programming, Leontief input-output analysis, Walrasian equilibrua, and the thinking of Schumpeter would be of great interest as well.

I'm open to the possibility that I am bibliographically ignorant on these questions, but I suspect that the historiography simply doesn't operate in a way that would satisfy my needs. All of these are questions well-suited to a split historiography, but are not well-answered by studying economic work's relationship to some overarching ideology and intellectual program, or, for that matter, an established canon.

I also think such pictures might seem more familiar and sympathetic to working economists, who might be more prone to paying attention to work that situates work within economic, institutional, and ideological contexts, while nevertheless retaining an interest in the intellectual appeal and origins of not only great works, but a whole variety of intellectual programs. This would surely help in building receptive audiences, and providing some order to a truly convoluted history.


Will Thomas wrote:
Also on my wishlist: I have long wanted to find a more general appreciation of Kenneth Arrow's economics (i.e., one that does not revolve solely around the impossibility theorem and Arrow-Debreu). I would be very interested to know more about thinking about "insurance" in these periods, which seems crucial, but is not much remarked upon. I have a very strong interest in thinking about inventories as well. Just the other day for the first time I ran into UCLA/Chicago School economist Armen Alchian's 1950 argument for linking evolution to the assumption of rational actors in economics, which I knew had to be somewhere in that milieu, and I was happy to find it right around where I expected it. I would like to know more. It's a bit of a side point for me, but I'd love to know more about the Harvard Keynesians (Hansen, Metzler, Goodwin...). How all of this related or does not relate to linear programming, Leontief input-output analysis, Walrasian equilibrua, and the thinking of Schumpeter would be of great interest as well.

This is pretty much the book-length project Till Düppe and I have underway. His piece soon to appear in JHET will answer some of these questions, and our book will deal with many of the other points you raise.

Post new comment

The content of this field is kept private and will not be shown publicly.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.
  • Allowed HTML tags: <a> <blockquote> <br> <cite> <code> <dd> <div> <dl> <dt> <em> <h2> <h3> <h4> <img> <li> <ol> <p> <span> <strong> <sub> <sup> <table> <tbody> <td> <tr> <ul>
    Allowed Style properties: display, float, height, margin, margin-bottom, margin-left, margin-right, margin-top, width
  • You may insert videos using embed codes like these:
    • [video_large:KoqLu5CKx-o]
    • [popupvideo_mini:KoqLu5CKx-o right]
    • [lightboxvideo_mini:KoqLu5CKx-o]
    • [text_popupvideo:KoqLu5CKx-o nostart noicon|Click here to open the video.]
    • [text_lightboxvideo:KoqLu5CKx-o|Open this video in a lightbox.]
    To learn more, please click on the "More information..." link below.

More information about formatting options