I’d like to test drive my new critical tool (“discipline & ontology” vs. “projects”) on my new article, “Operations Research vis-à-vis Management”. I think it’s a useful alternative analysis, which would never have made any final, published version of the article, but which nicely brings out the intricacy, subtlety, and importance of the issues at play.
I would argue that the historiography of OR has been dominated by the notion that OR was, essentially, an attempt (in the footsteps of Taylorism) to transform the ontology of military planning and industrial management from one of seasoned leadership into one of “science”. This shows up in the historiography of wartime OR, but especially in treatments of OR’s postwar adoption of mathematical formalism as its intellectual core. This last turn has been regarded as a clear departure from any sensible conception of management, and it can therefore only be explained as a kind of fetishization of science.
As I put it in my paper:
Prior accounts of OR’s turn to mathematical specialization have … assumed that the development of a mathematical canon represented a sort of pathology of professionalization, which detached it from the generalist investigations touted by its wartime practitioners. Andrew Abbott [The System of Professions (1988)] has suggested that ‘mathematical preeminence’ was a ‘professional regression’ resulting from a turn toward self-regarding academic virtuosity in OR. Thomas Hughes [Rescuing Prometheus (1998)] has grouped OR with systems engineering as a technical form of expertise that became subjected to typical criticisms of technocratic management and had to be supplemented by more humanistic and democratically inclusive ‘postmodern’ methods. Such accounts … suppose a chronological process of neglect or attainment of some general nontechnical conception of management, which might have granted OR wider and more legitimate authority.
In a footnote, I also observe how recently authors have transplanted this ontological rift from chronology into geography. Both Philip Mirowski in his Machine Dreams (2002) and John Krige in his American Hegemony and the Postwar Reconstruction of Science in Europe (2006) have reckoned that the movement to institutionalize and formalize OR was a distinctively “American” phenomenon that threatened a more demure British or European model of it.
All of these accounts take OR’s adoption of a mathematical canon as a logical technocratic culmination of a process of rendering governance scientific that began during World War II (or perhaps with Frederick Taylor). My argument is that this reading of OR as attempting to transform the ontology of management derives from a very particular reading of OR scientists’ own narratives about themselves.
Wartime advocates of OR certainly described what they were doing as the application of “scientific method” to problems of military decision. And some of their statements, such as physicist Patrick Blackett’s claim in a wartime memorandum that scientists could “help to avoid running the war by gusts of emotion,” would seem to reinforce this sort of ontological rift. However, as I argued in my 2007 BJHS piece, “The Heuristics of War: Scientific Method and the Founders of Operations Research,” wartime advocates of OR did not envision “science” as replacing some alternative “military” way of making decisions. Rather, it was merely an augmentation and routinization of existing planning methods with the aim of meeting already-accepted standards of good decision. In Blackett’s quote, attention should be directed to the word “help”.
The project (Project I) of those who sought to establish OR as its own discipline in the postwar period was to transpose this wartime notion of OR into industrial settings. The canonization of a body of formalism was actually antithetical to that project, because it was only applicable to a restricted set of managerial decisions and engineering tasks. Therefore, that canonization constitutes a separate project (Project II), the ascendancy of which requires explanation.
For those OR practitioners who adhered to the original project, the canonization of mathematics represented a threat to their aspirations, because the new canon was clearly insufficient to stake a claim to general influence over managerial thought. In response, by the 1960s and ’70s, some among this subset of Project I advocates responded by vocally developing a third project (Project III) to expand upon (or escape) the boundaries of mathematized OR by emphasizing things like “soft systems” and “wicked problems”, which were meant to supplement OR’s restrictive mathematical canon with a more humanistic methodology.
I believe it is from this sub-subset of OR advocates that we obtain the historiographical notion that the bounds of OR were coterminous with the ontology of management itself, and thus that mathematization could be understood as simply a fetishistic pathology in the history of OR.
In the hands of historians adhering to a “discipline & ontology” model, Project III’s narrative becomes an irresistible Whig narrative — a movement of the ontology of management from a technocratic misconception to a more mature, culturalized perception. In effect, the Project III narrative jibes with our own professional narrative, and is therefore accepted unquestioned. And, actually, there is even some evidence of disciplinary entanglement between us and Project III proponents in the ’60s and ’70s!
But this perspective also transposes the agenda of Project III onto the other projects that co-existed with it. My suggestion in my new paper is that proponents of Project I were highly unrealistic in their expectations, not because they misconceived management as something that could be made scientific, but because their project was actually in competition with a host of other Existing Projects to improve management (like management consulting and consumer research), which they either felt they could subsume, or to which they simply paid little heed. Because they ignored those projects’ importance, so have their historians, who have assumed that whatever failures OR experienced came not from competition, but from resistance by those representing a more traditionalist ontology (either conservatively or wisely).
My argument here accords with that made by Imperial College PhD student Michael Weatherburn, who is studying various managerial modernization “systems” in interwar Britain. According to Michael, the British historiography of industry assumes that British industrialists were simply conservative-minded, and thus slow to modernize. In fact, his assiduous research into industrial archives and various modernization systems shows that most industrial modernization efforts have simply been ignored because they were not publicized, and they did not bear the imprint of whatever “system” critics and historians have assumed constituted the vanguard of modernization.
By my argument, then, Project I proponents — failing to distinguish themselves amid this rich, but largely uncharted array of existing modernization projects — were not undermined by mathematical fetishists in their ranks or by traditionalist management; they were saved by a new association with Project II proponents, whose techniques were novel and valuable (if restricted in their applicability). In business-speak, Project I had the (war-tested) “OR” brand, Project II had a product that could be sold in a crowded market.
I would like to stress that we should not think of any of the projects discussed here as in existential tension with some “traditional” ontology of management (a chimerical Primordial Project?). If Project III did seek to develop a body of intellectual knowledge coterminous with management, Projects I and II relied not on their authority over the intellectual contents of management, but on institutional arrangements that integrated their work legitimately and productively with existing managerial work.
This becomes clear if you take proponents’ meta-commentary on their work seriously, or look at their pedagogical practices. For example, in my new paper, I point to the report on the first course in OR taught at MIT from 1950 (during the early ascendancy of Project II, but before it had supplanted Project I):
‘Real problems,’ the report stressed, ‘do not fall neatly into one or another of the usual categories of knowledge, but cut widely across boundaries. Successful O/R’ made sure that the ‘wholeness of the problem’ was ‘not artificially suppressed.’ The report pointed out that qualitative understanding of a problem was a ‘prerequisite to useful quantitative work. A crude solution of the real problem’ could have great benefits, ‘whereas a precise solution of an unjustified idealization of the problem is at best worthless and, if its misleading conclusions are acted on, may do great harm’ (110).
The report later went on to describe how students were specifically disciplined not to see management in purely formalized terms, noting the
‘interesting tendency of many of the students, especially those majoring in mathematics … ; they appeared to be disturbed emotionally by the uncertainty and large scope of some of these problems, and too quick to seize on some neat set of assumptions and then quickly reduce the discussion to a problem of pure mathematics.’ These students tended to spend only a few lines formulating the problem and then working out pages worth of analytical solution. The report went on: ‘It was necessary to combat this tendency vigorously by emphasizing the importance of qualitative considerations and crude approximate analysis by showing that in many cases the essential features of such analysis are understandable in very simple terms and, where not, are not reliable anyway.’ In the end, apparently, ‘all but two or three fell into the spirit of the assignments and produced very creditable reports’ (111-112).
Certainly not the words of people susceptible to mathematical fetishism! — but also not anyone likely to stick out in the crowded world of managerial improvement.
What, then, of the decision and management theorists who explicitly attempted to develop formalized theories of managerial decision-making (as at Carnegie Tech)? I would group them as a fourth project (Project IV!), which was an attempt to develop an academic vocabulary, which could bring intellectual order to a stew of managerial thought, but which was not readily translatable into actual suggestions for managerial improvement. In other words, I would stress that the historical record shows no strong evidence that Project IV work gave intellectual authority to Project II proponents attempting to realize the aspirations of Project I. That would be a story that would only make sense to proponents of Project III, who wanted to privilege their own approach to management, while characterizing others’ approaches as fetishistic and infeasible.
Nobody ever said history was simple! All these projects were, of course, interrelated, but they are, in fact, distinguishable if you bother to parse them out of the record. In many ways, the ways these projects played off each other actually relied on the fact that they were not easily distinguishable. However, the only way the history makes sense without invoking things like scientific fetishism is if you make the distinctions.