Thoughts on "Operation research in war and peace" by various authors (1947), Chris Ryan
3. "Operation research in war and peace", Various authors, Nature, Vol. 160, (Nov. 15, 1947), pp. 660-662 [link]
My personal exploration:
Nature—the world’s premier natural science journal—devoted space to a discussion of operational research (OR) in 1947 at a symposium-style account of talks given at the British Association’s Dundee meeting. The notes features comments from luminaries of early British OR: Sir Robert Watson-Watt, O. H. Wansbrough-Jones, Solly Zuckerman, W. C. Wilson, F. A. E. Crew, and J. D. Bernal. Each was a scientist who had been intimately involved in wartime efforts, and each was grappling with what OR meant in retrospect and what it could become in peace.
Watson-Watt and the Origins of OR
Watson-Watt, who claimed responsibility for the term “operational research,” recalled that early efforts amounted to “doing arithmetic for Air Staff” and the “less naively offensive elements of motion study”. Motion study refers to the practice of recording the timing and movement of operators that finds its origin in the Scientific Management movement that gained prominence before World War I. I am not sure Watson-Watt's motivation for emphasizing the simplicity of OR efforts, but it was telling that did not seem to want to oversell OR's accomplishments. Before the war, “scientific management” had been circulating in Britain, but it lacked legitimacy (as detailed in Merckle's "Management and Ideology"), in part because the practitioners were not considered “real scientists”. Indeed, the scientific management movement sparked a large number of unqualified and undisciplined "efficiency engineers" looking to make a "quick buck" by over-promising and underdelivering the benfits of "science" to the management class leading up to the Great Depression. The prescence of this report in Nature suggests that OR acquired precisely this legitimacy through wartime mobilization and the involvement of "real scientists".
Watson-Watt went further: “Operational research…is beginning to look very like real scientific management, and…an important contribution to the still unsolved problem of the relationship between the administrator and the expert”. It is notable that during the war, scientists were involved in military operations, but were not asked to make executive decisions. OR practitioners wanted the power to advise but not the responsilibity of decision-making. Scientific management, by contrast, was associated with a "take over" of of managerial oversight, with a rhetoric that management decisions should be in the hands of well-educated and objective professionals. I believe the "problem of the relationship between the administrator and the expert" remains unsolved until this day, and arguably one of the core challenges facing the future of operations research. We will have occassion to return to this theme often in this annotated bibliography.
Wansbrough-Jones: Freedom, Context, and Limits
Dr. Wansbrough-Jones emphasized that OR’s effectiveness came not just from training but from the conditions of practice: the freedom to observe directly and to escape daily routine. As he put it, OR was about “giving greater freedom both in opportunities of direct observation and of freedom from daily routine.” This resonates with later movements in organizational learning (Argyris, Senge, Liker), which stress that repetition need not become mindlessness. OR worked when scientists were allowed to think freely about recurring problems.
Yet Wansbrough-Jones was clear that training and objectivity were not enough. Success required receptive leadership, scientists with military knowledge, and above all, the stimulus of national need. Without these, “this type of work would not be useful”. In this sense, OR’s wartime potency already contained the seeds of its postwar weakness. Just as “scientific management” failed as a movement without institutional receptivity, so too OR risked irrelevance if applied in isolation from supportive contexts. In the United States, post-war OR had the support of the so-called "military industrial complex", as detailed at length in William Thomas's Rational Action. In Britain, OR made its way into a tool for government bureaucracy in the Labour governments that followed World War II.
Wansbrough-Jones recommmended that young scientists (as in physicists and biologists) take up OR as a “period of service,” not a career. For young scientists, a diversion into operational research might be “not only justifiable but desirable”—but not too long. The impression is that OR was valuable public service but not the substance of an academic career. This helps explain why OR took several years after the war to crystallize as an academic field.
Zuckerman: The Problem of Objectivity
Zuckerman averred that OR’s main contributions lay in “objectivity” and “statistical method.” As forcefully detailed in Porter's Trust in Numbers, objectivity as delivered by quantification was a growing pregortative of governments aiming for increased centralization of power through government oversight during the mid-20th century. "Experts", such as military strategists, often based their decision on experience and gut feeling, which worried scientists getting invovled with social enterprised concerned about how thinking can be biased when divorced from "evidence". Quantification offered objectivity because local concerns that have the potential to bias can be "averaged out" by truth-revealing summary statistics.
Zuckerman felt that opportunities for OR to blossom after the war were abundant. He wrote:
"The Government has added to the normal function of administration executive duties which make it necessary either that officials should lean more and more upon scientific advice or become more and more scientific themselves."
While the scientists of OR could offer a shoulder to "lean more and more" on, the possibility that executive decision makers "more and more scientific themselves" suggests a threat to the "secret sauce" of OR at this very early stage. If, as Watson-Watt said, OR is just "arithmetic", "becoming scientific" should be relatively easily. But this more profound notion of "statistical reasoning" was maybe harder to grasp, especially given at that time, statistical ideas were less well-known and less taught in schools and colleges like they are today. But one can easily see how "scientific advice" of the kind that appears to have been given during the war could eventually be learned by decision-makers themselves.
Of course, this could have also inspired an initial "spark" for what operations research academics might do: teach decision-makers to "be more scientific" and research the ways in which others might become more "scientific" in their approach. Indeed, this style of education remains standard in operations research education in business and engineering schools to this day, although now there are many more places in schools and industry where one can learn these ideas. High schools now even offer courses emphasizing statistical concepts and common fallacies. One notable area of training is "Total Quality Management", that champions the idea of welcoming even the most humble front line worker into a "community of scientists" (see, for examople, Spear and Bowen's paper "Decoding the DNA of the Toyota Production System").
Of course, academic operations research also endeavored to offer more advanced "scientific advice" based in ever-more complex mathematics, continually expanding the apparent need for operations research expertise. Maybe this drive for "scientific novelty" far beyond what is taught in the classroom was partially driven by the possibility Zuckerman expresses here, that less than cutting edge tools might eventually be absorbed by managers themselves.
Bernal: OR as Social Science
Finally, J. D. Bernal argues passionately for extending OR into peacetime reconstruction, arguing that “it would be tragic ... if this one organisational development of the War were not given the chance to show its capacity in the peace”. Bernal’s vision was expansive: OR as “trained scientists doing useful work for society.” Bernal's view of OR was that of a social science, but embracing the best practices of natural scientists, applying their craft to economic planning, production, and reconstruction. This echoes Blackett informal defintion of OR stated in his 1950 paper "Operational Research": natural scientists doing social science. This is not a surprise, Bernal and Blackett were compatriots and both heavily aligned with technocratic utopian socialist ideas. Bernal himself was an open member of the Communist Part of Great Britain and both Bernal and Blackett getting the label of "red scientists" during the Cold War.
Reflections
What emerges from this discussion is both the promise and the ambiguity of OR in 1947. On the one hand, it was seen as scientific management reborn, legitimized by scientists and empowered by state-scale problems. On the other hand, its authority rested on fragile cultural concepts like “objectivity,” on temporary conditions like wartime urgency, and on the willingness of administrators to listen. Bernal’s vision of OR as a vehicle for social planning was bold, but it also pointed toward the eventual tension between OR as service and OR as an academic discipline. That tension may help explain why dedicated OR programs were not institutionalized until several years later.