A Counterfactual Committee
If Haldane had not followed the advice of a Trinity mathematician when he set up the Advisory Committee for Aeronautics but had, say, recruited Cambridge engineers rather than mathematical physicists, he might have got a very different committee. In principle he could have done this because Cambridge had a distinguished school of engineering.25 Predictably, there had always been a tension between the demands of a practical engineering education and the demands of the traditional, Cambridge mathematical curriculum. There was not time in the day to succeed at both, except for the outstanding few. It was not until 1906 that a satisfactory accommodation was reached when Bertram Hopkinson, the professor of mechanism and applied mechanics, gave a viable and independent structure to the Mechanical Sciences Tripos. This took the engineers out of the competitive hothouse, though inevitably it meant they operated at a somewhat less sophisticated mathematical level. The products of Hopkinson’s department played a distinguished role in the development of British aeronautics. Busk (who made the BE2 stable), Farren (who championed full-scale research at Farnborough), Melvill Jones (who worked on low-speed control and gunnery), Southwell (who worked on airship structures), and McKinnon Wood (the experimentalist who went with Glauert to see Prandtl) were all products of the Cambridge school of engineering. Hopkinson himself, though of an older generation, learned to fly during the Great War and did important work on aircraft testing at Martle – sham Heath. He met his death at the controls of a Bristol Fighter in a flying accident in August 1918.26
Perhaps a counterfactual Advisory Committee, made up of men like Bertram Hopkinson, would have embraced Lanchester and the circulatory theory. This is consistent with my analysis, although the conclusion would only follow if Cambridge engineers were significantly different in their judgments and orientation from their Mathematical Tripos colleagues and significantly similar to the German engineers from Gottingen and Aachen. Such a premise is plausible but it cannot be taken for granted, and there is some evidence that calls it into question. In certain respects Cambridge engineers adopted attitudes that were similar to those of the more traditionally trained mathematicians. This is not surprising in the case of the older products of the Cambridge school of engineering because they too were steeped in the earlier Mathematical Tripos tradition. Bertram Hopkinson’s father, John Hopkinson, had been both an engineer and a senior wrangler. After a fellowship at Trinity he become a professional, consulting engineer in London but retained a strongly mathematical bent. He developed a mathematical analysis of the alternating current generator and predicted that it should be possible to run such generators in parallel. Practical engineers knew that, given the available machines, they could not be run in parallel, but none of this blunted the confidence of the mathematically able Hopkinson.27 Bertram Hopkinson himself had also read for the Mathematical Tripos and had been a highly placed wrangler. Although more practically inclined than his father, he had worked on topics in hydrodynamics and had written a paper on the theory of discontinuous flow.
He extended the work of A. E. H. Love by making allowance for the presence of sources and vortices.28 None of those who contributed to the mathematical development of the theory of discontinuous flow ever went on to work on the mathematics of the circulation theory of lift.
What about the somewhat younger generation of Cambridge engineers such as Busk, Farren, Melvill Jones, McKinnon Wood, and Southwell? Here, too, the evidence indicates that, although they had some sympathy with Lanchester, that sympathy had its limits. And those limits were characteristic of the milieu of Cambridge mathematical physics. Southwell’s role as a lecturer in mathematics at Cambridge (despite his engineering background), and his commitment to a fundamental physics of lift, have already been described. McKinnon Wood was perhaps different. He had come round quickly to the side of the Lanchester-Prandtl theory of circulation, though exactly why is unclear. It is possible that he was taking his lead from Glauert. Writing retrospectively, McKinnon Wood recalled that, as a student, he had borrowed Lanchester’s Aerodynamics from the Union library but then, after reading it, had thought no more about it.29 Looking back, all that he could report was regret at this “mysterious blindness.” Farren appears to have accepted a version of Glazebrook’s excuse and recalled Lanchester as an almost a tragic figure. He was a man who had clear insights but who could not give them expression. “The vision of Lanchester brooding over the irony of a world which at last understood what he had seen so clearly, but had been unable to explain, will not fade from the minds of those who knew him.”30
Bennett Melvill Jones sought to shed light on these mysteries in a talk he gave at the Royal Aircraft Establishment at Farnborough in 1957.31 He described how, as a young man, his friend and fellow engineering student at Cambridge, Edward Busk, had introduced him to Lanchester’s work. It had deeply impressed them. After studying Lanchester’s treatise, “Ted and I solemnly decided to spend the rest of our lives on aeronautics.” But, Melvill Jones went on, aeronautics in Britain had then developed in an almost entirely empirical fashion, that is, without the guidance of Lanchester’s theory. This may seem surprising, he told his audience, “because the theory of ideal fluid, which now forms the basis of all your work, had already been very fully developed by Lamb and others. But in this theory, no body which starts from rest can, when in steady motion, experience any reaction whatever. And to us, whose business it was to study this reaction, this did not seem very helpful.” The source of the trouble, he explained, was the impossibility of accounting for the origin of circulation. This was the very same objection that the young G. I. Taylor had used against Lanchester in 1914. The origin of the circulation and hence the lift could not be deduced within the terms of the theory. If
Melvill Jones’ recollections were correct, then the products of the Mechanical Sciences Tripos had been at one on this question with the products of the Mathematical Tripos. Both saw its dependence on the classical hydrodynamics of potential flow as a fatal objection to Lanchester’s approach.
Melvill Jones went on to say that now (that is, by 1957) it may seem obvious how this difficulty should have been resolved. He added, ruefully, that perhaps he and his contemporaries had not been very clever in making the response they did, but, he insisted, “our view was, at that time, shared by all engineers.” Here Melvill Jones was making a historical mistake. Even if the negative reaction to ideal fluid theory was shared by all Cambridge aeronautical engineers of his generation, or even by all British engineers, the response was not shared by the German engineers who developed the circulatory theory. In an attempt to justify the negative reaction to Lanchester, Melvill Jones said that the theory of ideal fluids, then known as hydrodynamics, “was regarded purely as an exercise for the amusement of students.” Again, while this may have been true in Cambridge it was not how things were seen in Gottingen or Aachen. This is not how ideal fluid theory was regarded by Betz, Blumenthal, Foppl, Fuhrmann, von Karman, Kutta, von Mises, Prandtl, or Trefftz. For them it was more than a source of ingenious examination questions; it was a serious instrument of research and a serious technique with which to gain some purchase on reality.
The Cambridge ethos seems to have enveloped its engineers as well as its mathematicians. It therefore cannot be assumed that a counterfactual committee of Cambridge engineers would have responded as the German engineers did and welcomed the circulatory theory.32 This answer obviously depends on the range of factors that are allowed into the imaginary picture. If engineers had been given a dominant position on the ACA, they might have gained greater independence of mind. They might then have been less influenced than they actually were by the older Tripos tradition of mathematical physics. No one can know. The fact remains that these were not the men who initiated the research program into the theory of lift, and they did not have a dominant role on the committee that Haldane actually created.33