Category Archimedes

I

All the papers in this collection are highly accomplished exercises in the history of technology but, of course, they represent history of somewhat different kinds. For example, they assume different positions along the so-called “internal-external” dimension. Profs. Smith and Mindell’s admirably detailed history of the turbo fan engine largely, but not wholly, falls into the “internal” category, as do the authoritative accounts by Prof. Vincenti of work done in supersonic wind tunnels, and Prof. Suppe on the instrumentation used in test flying. Of a more “external” character are papers dealing with the economic, institutional, and legal setting of aeronautical activity, such as Profs. Crouch and Roland’s discussion of patents or Prof. Douglas’ account of the different professional inputs to airport design. Again on the “externalist” side we have Prof. Jakab’s account of the hypocrisy and profiteering surrounding the emergence of McCook Field.

Apart from the “internal – external” divide we also have other kinds of variation. Some papers are clearly more descriptive, while some are more explanatory. I think it is true to say, however, that most contributors have chosen to keep a low profile when it comes to questions of methodology or theory. This is a wise strategy. Concrete examples often speak louder than explicit theorizing. Readers can put their own gloss on the material and see for themselves points of contact with other ideas. Nevertheless it is in the area of theory and method where I may have a role to play. I am not an historian of technology, indeed I am not an historian at all, and I have no expertise in aerodynamics. My comments will all be from the standpoint of sociology. I shall focus on what I take to be the more significant sociological themes that are explicit, or implicit, in what has been said. I shall try to suggest ways in which sociological considerations might be brought into the discussion or where they might help in taking the analysis further. Adopting a sociological approach biases me towards what I have called the “external” end of the scale, but there are some points of contact with the more technical, “internal” discussions, and I shall do my best to indicate these. Here I shall be struggling somewhat, but for me these points of contact between the social and the technical represent the greatest challenge, and the greatest interest, of the whole exercise.

II

Let me begin with Prof. Galison’s paper. His claim, and I think he makes it convincingly, is that aircraft accident reports are beset by an unavoidable

349

P. Galison and A. Roland (eds.), Atmospheric Flight in the Twentieth Century, 349-360 © 2000 Kluwer Academic Publishers.

indeterminacy. Investigators confront an enormously complicated web of circumstances and actions. How is this manifest complexity to be handled? There are two possible directions in which to go. One direction is to narrow down the explanation of the accident on to one or more particulars and highlight these as the cause. We will then hear about the Captain who chose to take-off in bad weather, or the co-pilot who was not assertive enough to over-rule him, or the quality control inspector who, years before, let pass a minute metallurgical flaw. The other direction leads to a broader view, embracing a wide range of facts which can all be found to have played a contributory role in the accident. Had the de-icing been done properly? Was the company code of practice wrong in discouraging the use of full – throttle? What is it that sets the style and tone of group interaction in the cockpit? Who designed the three hydraulic systems that could be disabled by one improbable but logically possible event? Accident reports, he argues, tend to oscillate between specific causes and the broad range of other relevant particulars.

Prof. Galison has put his finger on a logical point similar in structure to that made famous by the physicist Pierre Duhem in his Aim and Structure of Physical Theory (1914). No hypothesis can be tested in isolation, said Duhem. Every test depends on numerous background assumptions or auxiliary hypotheses, e. g. about how the test apparatus works, about the purity of materials used, and about the screening of the experiment. When a prediction goes wrong we therefore have a choice about where to lay the blame. We can, logically, either blame the hypothesis under test, or some feature, specified or unspecified, of the background knowledge. Prof. Galison is saying the same thing in a different context. No causal explanations can be given in isolation. Whether we realize it or not, the holistic character of understanding always gives us choice.

How are those in search of causes to make such choices? Prof. Galison tells us not to think in terms of absolute rights or wrongs in the exercise of this logical freedom. The unavoidable fact is that there are different “scales” on which to work. This, “undermines any attempt to fix a single scale as the single ‘right’ position from which to understand the history of these occurrences” (p.38). What, then, in practice inclines the accident report writer to settle on one or another strategy for sifting the facts and identifying a cause? Legal, economic and moral pressure, we are told, has tended to favour the sharp focus. This has the advantage that it offers a clear way to allocate responsibility: this person or that act was responsible; this defect, or that mechanical system, was the cause. The utility of the strategy is clear. Individuals can be shamed or punished; and a component can be replaced. Now, we may be encouraged to feel, the matter is closed and all is safe again.

Is this a universal strategy rooted in some common feature of human understanding, or near universal social need? Do accident reports from, say, India and China, have the same character as those from France and America? Or are some cultures more holistic and fatalistic and others more individualistic and litigious? It would be a fascinating, though enormous, undertaking to answer a comparative question of this kind. Prof. Galison gives us some intriguing hints to be going on with. In his experience there is not much difference between, say, an Indian and an

American report. What stand out strongly are not broad cultural differences but sharp conflicts of a more local kind. While the National Transportation Safety Board blames a dead pilot, authoritative and experienced pilots dismiss the report as pedantic and unrealistic. While official American investigators point to the questionable behavior of the control system of a French built aircraft under icy conditions, the equally official French investigators point the finger at the crew who ventured into the bad weather. Prof Galison speaks here of the “desires” of industry, of the “stake” of pilots, and of the “investment” of government regulators, as well as economic “pressures”. All of these locutions can be brought under the one simple rubric of ‘interests’. The phenomenon, as it has been described, is structured, strongly and in detail, by social interests. Let me state the underlying methodology of Prof. Galison’s paper clearly and boldly. It is relativist, because it denies any absolute right, and depends on what is sometimes called an ‘interest model’. As we shall see, these challenging themes weave their way through the rest of the book as well, and inform much that is said by the other contributors.

Ill

A number of papers take up a comparative stance – perhaps the most explicit being Prof. Crouch’s discussion of the differential enthusiasm with which governments encouraged aviation in Europe as compared to the United States. It wasn’t the Wright brother’s insistence on taking out patents that held back American aviation, it was lack of government awareness and involvement. Prof. Roland takes up the story of patents to show how the earlier shortcomings were later dramatically repaired. Once the government had got the message, expedient ways were found for circumventing legal issues. One of the themes of Prof. Roland’s paper is that of a culture of design, craft-skill and engineering expertise. His point is that if anything inhibited the flow of information and expertise it wasn’t patents, it was the localized character of engineering practices and the desire to keep successful techniques away from the gaze of competitors. So significant was this local character of engineering knowledge that when different firms came to produce aircraft or components according to the same specifications, they still turned out subtly different objects. Prof. Ferguson gives a stunning and well-documented example of this. He looks at the wartime collaboration of Boeing, Douglas and Vega in producing B-17 bombers. Despite the intelligent anticipation of such problems, and a determined attempt to overcome them, these three companies could not build identical aircraft. The ‘same’ tail sections, for example, proved not to be interchangeable. The lesson, and Prof. Fergusen draws it out clearly, is that practical knowledge does not reside in abstract specifications (not even in the engineer’s blue prints.) Knowledge is the possession of groups of people at all levels in the production process. It resides locally in their shared practices. The problem of producing the same aircraft remained insurmountable as long as the three local cultures survived intact and in separation from one another. It was only overcome by removing, or significantly weakening, the group boundaries. This was done by deliberately making it possible for people to meet and interact. In effect, it needed the three groups to be made into one group.

Others tell us of similar phenomena. Profs. Smith and Mindell speak of the different engineering cultures of Pratt and Whitney, on the one hand, and General Electric, on the other. General Electric’s rapid and innovative development of the CJ 805 by-pass turbofan engine was helped by their recruitment of a number of key engineers from NACA. Again, we see that knowledge and ideas move when people move. Smith and Mindell characterize the culture of Pratt and Whitney as “conservative” – and explain how, nevertheless, contingencies conspired to ensure that its incrementally developed JT8D finally triumphed over the product of the more radical group at General Electric. Their history of the triumph of the Pratt and Whitney engine shows how different groups produce different solutions to the same problem. Add to this their discussion of the Rolls Royce Conway, and the debatable question of which was the ‘first’ by-pass engine, and we have another demonstration of the complexity of historical phenomena and the diversity of legitimate descriptions that are possible. Finally, they address the general question of why turbo-fan engines emerged when they did – some twenty five years after Whittle and Griffith had suggested the idea? The concept became interesting, they argue, because of decline in commitment to speed and speed alone as the arbiter of commercial success in civil aviation. Turbo-fans could succeed economically in the high sub-sonic realm in a way that jets could not. We are dealing with new measures of success, “measures that embody social assumptions in machinery” (p.47). This pregnant sentence once again gives an interest-bound and relativist cast to the argument.

The theme of the local engineering culture, whether it be Pratt and Whitney versus General Electric, or Boeing versus Douglas, strikes me as sociologically fascinating. And it is, of course, something calling out for comparative analysis. How many kinds of local engineering culture are there? Are there as many cultures as there are firms? Ferguson’s work suggests that there are, and that we are dealing with a very localized and sharply differentiated effect. Nevertheless, even if local cultures are all different from one another, they might still fall into classes and kinds, perhaps even a small number of kinds. If they were to fall into a small number of kinds then comparison should enable us to see the same patterns repeating themselves over and over again. This might allow us to identify the underlying causes of such cultural styles. Calling Pratt and Whitney “conservative” might tempt us to think in terms of a dichotomy, such as conservative versus radical. Other dichotomies, such as open societies versus closed societies, or Gemeinschaft versus Gessellschaft, the organic versus the mechanical, or markets versus traditions, will then suggest themselves as models. But there are other options. Why work with a dichotomy? If we follow the suggestions of the eminent anthropologist Mary Douglas we arrive at the idea that there might be, not two, but four basic kinds of engineering culture (cf. Douglas,1973,1982).

Let me explain how this rather striking conclusion can be reached. The first step is to notice that the basic options open to anyone in organising their social affairs are limited. They can attend to the boundaries around their group and find them, or try to make them, either strong or weak. They can then structure the social space within the group in a hierarchical or an egalitarian manner. If we assume that extreme, or pure cases, are more stable, or self-reinforcing, than eclectic or mixed structures, this gives us essentially a two-by-two matrix yielding four ideal types.

The same conclusion can be arrived at by another route. Think of the strategies available to us for dealing with strangers. We can embrace strangers, we can exclude them, we can ignore them, or we can assimilate them to existing statuses or roles. To embrace strangers means sustaining a weak boundary around the group, while to exclude them requires a strong inside-outside boundary. Having a pre-existing slot according to which they can be understood and their role defined means having a complex social structure already in place which is sufficiently stable to stand elaboration and growth. Responses to strangers thus etch out the pattern of boundaries around and within the group and the options, of embracing, excluding, ignoring or assimilating, generate four basic structures – arguably the same four as identified above.

The second step in Mary Douglas’s argument is to suggest that our treatment of things is, at least in part, structured by their utility for responding to people. We have seen the principle at work in the large scale in the case of the by-pass engine as a response to market conditions. But the idea can be both deepened and generalized. Perhaps our use of objects, at whatever level of detail, will always be monitored for implications about the treatment of people. Do they leave existing patterns of interaction and deference in tact, or do they subvert them? Do they provide opportunities to control others, or opportunities to evade control? Take some simple examples. A place may be deemed “dangerous” if we don’t want people to go there. Or a thing is not to be moved or changed, because it embodies the rights of some person or group, say the right of ownership or use. Or the introduction of a new practice may render existing expertise irrelevant, and hence existing experts redundant. Put these two ideas together, social interaction as the necessary vehicle and unavoidable medium for the use of things, and the limited patterns of such interactions, and we may have a basis for a simple typology of cultures or styles in the employment of natural things and processes. We may even have a basis for a typology of styles of engineering.

Whatever your reaction to this idea, I hope you will agree that we badly need intellectual resources with which to think about such difficult themes. That is why I should like to hear a lot more about the sociology of Pratt and Whitney and General Electric, and a lot more from Prof. Suppe and Prof. Vincenti about the groups with whom they worked. How strong were the boundaries between the inside and the outside of the group? Did members of the group readily work with those outside it, trading ideas and information? How hierarchical or internally structured was the inside of the group? Did people swap roles or was the division of labor clear cut? Did they eat and relax together as well as work together? Who and how many were the isolates or outcasts -1 mean: those who were considered unreliable, incompetent or downright dangerous? These are the questions that Mary Douglas wants us to ask.

They may shed light on why some parts of a design are held stable while others are changed, or why a problem may be deemed solved by one group while another treats that solution as an unsatisfying expedient.

I say that these ideas may provide the basis of a typology. There are many ways of criticizing the line of thought I have just sketched. But I would ask you not to be too critical too soon. Give the idea time to settle. It may start generating connections that are not at first obvious to you. For example, Prof. Bilstein’s account of the international heritage of American aerospace technology connects with what I have just said about responses to strangers. He is telling us about a segment of culture that was eager to embrace strangers, where ‘stranger’ here means experts and outsiders from Europe. The European input is sometimes retrospectively edited out, but in practice and despite difficulties, it was seen as a source of opportunity rather than threat.

Or consider the “ways of thinking” alluded to by Prof. Vincenti. Here too we are dealing with questions of group culture and group boundaries. His first-hand account describes the problems of understanding an aerofoil in the transonic region. Two different groups of people had reason to address this issue: those designing wings, and those who worked on propellers. Those who work on wings routinely use the categories of ‘lift’, ‘center of lift’ and ‘drag’, but this is not exactly how the designers of axial turbines and compressors think: “because the airfoil-like blades of their machines operate in close proximity to one another, [they] think of the forces on them rather differently” (p.23). If NACA had structured its research sections differently and with different boundaries from those which actually obtained, might Prof. Vincenti’s group have had a different composition, and could that have enabled them to think about the forces on their wings “rather differently”? Or, to put the point another way, what would have happened if propeller and wing theorists had switched roles? Would the same understanding have emerged, or would they, like the teams at Boeing, Douglas and Vega, have made subtly different artifacts? These are enormously difficult questions that we may never be able to answer in a fully satisfactory way. The need to think counterfactually may defeat even the most well – informed analyst, but these, surely, are the ultimate questions that we cannot avoid addressing.

There is one misunderstanding that I should warn you against. Theories such as Douglas’s tend to be seen, by critics, in terms of stereotypes. The stereotype has them saying that “everything is social” or that “the only causes of belief are social.” Douglas’s approach is not of this kind at all. Society is not the only cause, just as it was not the only cause at work in the episode when the tail-sections from different factories would not fit together. We must also remember that certain large-scale pieces of equipment such as wind tunnels (or great radio telescopes or particle accelerators) have a way of so imposing themselves on their users that they actually generate a characteristic social structure. Given certain background conditions, the apparatus itself may be a cause of the social arrangements. This is compatible with what Douglas is saying. Her ideas would still come into play. The point would be this: suppose people have to queue up to use such things as wind tunnels, radio telescopes and accelerators. If so, there has to be a decision about who comes first, and how long they can have for their experiment. Things have to be planned in advance, timetables have to be drawn up and respected. These are all matters which presuppose a clear structure of authority. Someone must have the power to do this. These are not, perhaps, easy environments for the individualist. Having said that the physical form of the apparatus can be a cause, it would still be interesting to know what leeway there is here. To what extent is it possible to operate such things as wind tunnels, or huge telescopes, in different ways? And if it is possible, how many different ways are there?

IV

Two papers that I have not yet mentioned seem to me to provide a particularly good opportunity for opening up themes bearing on local cultures in engineering or science. The first of them is Prof. Schatzberg’s account of the manufacture of wooden aeroplanes during World War II. The British and Canadians were successful in their continuation of this technology, the Americans had ‘forgotten’ how to do it. Although not presented in these terms, this episode clearly bears on the question of local engineering culture and style. There must have been something about de Havilland which explains why they could and did build in wood and which differentiates them from the American manufacturers, such as Curtiss-Wright, who did not or could not. And presumably there was something equally special about the Canadian firms who were especially keen to get in on the act. Prof. Schatzberg grounds these facts about the culture of firms in the wider culture. He suspects that we need to go to the symbolic meaning of the material under discussion, namely wood. Wood for the Americans meant outmoded tradition; metal meant modernity. Wood for the Canadians meant the mythical heart of their nation, with its great forests. Wood for the British was more neutral, having neither strongly positive nor negative connotations. In that matter-of-fact spirit they built the formidable Mosquito – made of ply and balsa.

Symbolic meanings are not easy to deal with, so it is worth seeing if the argument can be recast in simpler terms. Would it suffice to appeal to, say, vested interests? Canada’s wood was a major economic resource – what better stance to take than to advocate a use and a market for it? Self-interest would dictate nothing less. The American manufacturers had moved on from wood, they had gone all out for metal so as not to be left behind in a competitive market. Who wants to go backwards in a manufacturing technique? It is expensive to re-tool and re-skill. What about de Havilland? They were still building the famous Tiger Moth trainer out of wood, and supplying it in great numbers to the RAF. Even more to the point, as Prof. Schatzberg points out, they could cite their experience with wooden airliners which had important similarities with the Mosquito (p.20). Perhaps economic contingencies such as these are basic, and explain the phenomenon attributed to symbolic meanings. But as Prof. Schatzberg argues, such contingencies often balance out. Canada may be rich in forests, but so was the United States. Canada was no more short of aluminum than the United States. Britain was short of both aluminum and timber and had to import both. As for the market, why did it lead manufacturers to metal in America but not in Britain? Why was demand differently structured in the two cases? That points to a difference of underlying disposition. Perhaps the presence and operation of such dispositions can be illuminated by the idea of symbolic meaning. But whatever our thoughts about the relation of symbolic meanings to interests, Prof. Schatzberg’s intriguing study raises important questions. How local are local engineering cultures, and to what extent do they depend on the wider society?

V

I now want to focus on Prof. Hashimoto’s paper. You will recall that this deals with the work of Leonard Bairstow and his team at the Natural Physical Laboratory in Teddington during and after World War I. We are told of their problematic relation both to PrandtTs school in Gottingen and the (mostly) Cambridge scientists who worked at the Royal Aircraft Establishment in Famborough. Bairstow worked with models in a wind tunnel. He had made his reputation with important studies of the stability of aircraft. Divergences then began to appear between the results of the wind tunnel work and those of the Famborough people testing the performance of real aircraft in the air. Bairstow took the view that his wind tunnel models gave the correct result and the full scale testing was producing flawed and erroneous claims.

Prof. Hashimoto’s account is a clear and convincing example of an interest explanation of the kind I have spoken about. Bairstow and his team had invested enormous effort into the wind tunnel work. Reputations depended on it. Is it any wonder, when it comes to the subtle matter of exercising judgement and assessing probabilities, that Bairstow should be influenced by the time and effort put into his particular approach? Notice I say “influenced” rather than “biased”. I don’t think that there is any evidence in the paper that Bairstow was operating on anything other than the rational plane. He seems to have been a forceful controversialist, but those arguing on the other side had their vested interests as well. Time, energy, skill, and not a little courage, had gone into their results too. Rational persons acting in good faith can and do differ, sometimes radically. To my mind the only questionable note in Prof. Hashimoto’s analysis is when a negative evaluation of Bairstow creeps into the historical description. I suspect that the analysis goes through as well without it.

Bairstow did not just argue in defense of his wind tunnel work on models. He also held out against Prandtl’s “boundary layer” concept. This stance seems to me even more intriguing than his opposition to scale effects. If I understand the matter rightly, Prandtl’s idea of a boundary layer functioned as a justification for certain processes of approximation and simplification which made the treatment of the flow round an aerofoil mathematically tractable. For example, it allowed the problem to be divided into two parts. Bairstow’s opposition seems to have centered on his suspicion of these mathematical expedients. His obituarist, in the Biographical

Memoirs of Fellows of the Royal Society (Temple, 1965 ), suggested that Bairstow didn’t want a simplified or approximate solution to the Navier-Stokes equation for viscous flow, but a full and rigorous solution for the whole mass of fluid. He tried to produce such a treatment, but with only limited success (p.25). So again we are dealing with a matter of scientific judgement on Bairstow’s part. His response to Prandtl was not foolish or dogmatic, but a rational expression of a particular set of priorities and intellectual values. I should like to know where these priorities came from, and if they can be explained.

At first sight it seems odd that a person such as Bairstow, who placed so much weight on the experimental observation of models in wind tunnels, should incline to a mathematically rigorous and complete solution to the equations of flow. We might expect an experimentalist to embrace any clever expedient that would get a plausible answer, even at the cost of a certain eclecticism. So there is a puzzle here. Of course, it may be that we are dealing with nothing but a psychological idiosyncrasy of Bairstow as an individual, but we should at least explore other possibilities. If his strategy commanded the respect of some portion of the scientific community then we are certainly dealing with more than idiosyncrasy. I have already suggested that large installations, like wind tunnels, may demand or encourage clear-cut social organisation and hierarchy. Could it be that living with the demand for rigorous and extensive social accountability lends credibility to the demand for a correspondingly rigorous and general form of scientific accountability? Or is the issue one of insiders and outsiders, of the boundary round the group? If there were a body of existing mathematical techniques which had always sufficed in the past then why import potentially disruptive novelty from the outside? Here we see the relevance of Mary Douglas’s observations about strategies for dealing with strangers. What she says about ways of dealing with novel persons applies on the cognitive plane too. Typically, dealing with new ideas means dealing with new people. A disinclination to do the latter might explain a disinclination to do the former. Clearly all this is speculation on my part. It is no more than a stab in the dark. But however wrong the suggestions might be, I think that this is the area where the argument can be developed.

Interest explanations, it seems, are ubiquitous, and beginning with Prof. Galison and ending with Prof. Hashimoto, we have seen how naturally the different contributors have had recourse to them here. Nevertheless, there is a great deal that we do not yet understand about the operation of interests. For example, the trade-off between long term and short term interests is little understood, though they often incline us to quite different courses of action. Prof. Deborah Douglas’s account, of the relation between engineers and architects, suggests that at first these two groups felt they were in competition with one another. In the event they reached a working relationship, seeing the advantages of co-operation rather than conflict. Prof. Jakab describes how the drying up of government contracts, in 1919, led to a change in the relation between the Engineering Division at McCook Field and the private manufacturers of aircraft. After a period of plenty, a divergence of interests came to the fore and set the stage for a decade of bickering. In offering an interest analysis, then, we are not dealing with a static and unchanging set of oppositions, but a potentially flexible and evolving system of self-understanding. Critics of interest explanations, and there are many of them, often assume they are static and forget that change and flexibility are actually inherent to this manner of explanation.

VI

A further theme I could have pursued is brought out with great clarity in Prof. Anderson’s paper (a paper which, for non-specialists such as myself, was invaluable for introducing some of the basic ideas in aerodynamics). Prof. Anderson’s theme was the coming together of academic or ‘pure’ science with the needs of the practical engineer, and the synthesis of a new form of knowledge that might be called ‘engineering science’. There is a temptation, amongst intellectuals, to treat theory as higher, or more profound, than practice. Practice is the application of theory so, (we might be tempted to conclude) behind every practice there must lurk a theory. In opposition to this, Prof. Anderson reminds us of the post hoc character of the circulation theory of lift. Similarly, with regard to drag, the theoretical explanation of just why engine cowling was so effective came well after the practical innovation itself. There is a whole epistemology to be built by insisting that practice can, and frequently does, have priority over theory. That epistemology will be the one needed to understand the history of aviation and the science of aeronautics. The theme of the priority of the empirical is also made explicit at the end of Prof. Suppe’s paper which tells us about the growth of instrumentation in the history of flight testing. He argues for the significance of this site as a source of new ideas to revivify the philosophy of science. That discipline has, arguably, been too closely linked with the academic laboratory, or at least, with the image of scientific work as theory testing. Prof. Suppe takes us out of the laboratory into the hanger, where the price of failure is far greater than the disappointment of a refuted theory. The aim is to discern useful patterns in the mass of data produced by probes and gauges, it is to root out systematic error, and to find the trick of making signals stand out from noise. Again, there are no rules and no absolute guarantees. Even, or perhaps especially, in these practical fields, reality cannot speak directly to those who want and need to act with the maximum of realism.

If we were looking for a philosophical standpoint that does justice to the practical character of knowledge, for an epistemology for the engineer or aviator, where might we look? I can think of two sources that would be worth exploring, though both suggestions may seem a little surprising. First we could follow the lead of Hyman Levy. Levy was professor of mathematics at Imperial College, London, and had written one of the early textbooks in aeronautics, Cowley and Levy (1918). His theoretical work in fluid mechanics linked him with Bairstow, and he had worked in this field at the National Physical Laboratory. Levy wanted to understand his science philosophically and to develop a philosophy appropriate for “modem man”. He was also a Marxist and belonged to that impressive band of left-wing scientists in Britain in the inter-war years which included J. D. Bernal, Lancelot Hogben, J. B.S. Haldane, and Joseph Needham [see Werskey (1988)]. They did much to spread the scientific attitude and to educate the public about the potential and importance of science. As a Marxist, Levy accepted the laws of dialectical materialism, one of which is the law of the unity of theory and practice. In his Modern Science. A Study of Physical Science in the World Today; published in 1939, Levy devoted a chapter to ‘The Unity of Theory and Experiment.’ He did not here talk in the language of philosophical Marxism, but about wind tunnels and the measurement of lift and drag. His exemplar of the unity of theory and practice was aeronautical research. This might commend itself to Prof. Anderson and Prof. Suppe. But I can also think of another source of philosophical inspiration, one coming from what, at first sight, appears to be the opposite ideological direction. For a philosophy which grounds theory in practice, and meaning in use, we could hardly do better than to consult Wittgenstein’s Philosophical Investigations (1953) and his Remarks on the Foundations of Mathematics (1956). The formalistic concern with theory testing that, as Prof. Suppe observes, limits so much academic philosophy, is wholly absent from Wittgenstein’s work. Perhaps it is no coincidence that Wittgenstein was trained as an engineer and did research on propeller blades. The young Wittgenstein once asked Bertrand Russell if he (Wittgenstein) was any good at philosophy. Wittgenstein’s reason for asking was that if he was no good he intended to become an aviator.1 [3] 1909 they set up the Advisory Committee for Aeronautics under the eminent Cambridge physicist Lord Rayleigh, and dedicated a section of the National Physical Laboratory to aeronautical questions. Prof. Schatzberg reminds us that such facts stand in opposition to many a political lament about Britain’s national decline and lack of industrial spirit. But can this rapid response really have come from a culture that was (as the stereotype has it) dominated by an anti-technological elite steeped in the classics?2

As the papers in this collection make amply clear, the history of aviation as a field of study is rich and many-faceted. There is material enough here to prompt major, and exciting, revisions in the history of culture – as well as in the theory of scientific and technical knowledge.

[1]

P. Galison and A. Roland (eds.), Atmospheric Flight in the Twentieth Century; vii-xvi © 2000 Kluwer Academic Publishers.

Predating the wartime cooperatives were a number of formal and informal means for sharing technological information. The Manufacturers Aircraft Association (MAA) had since 1917 served as a means to pool critical aircraft patents. The MAA’s reach did not cover manufacturing technologies though. Nor did it have much influence on military designs, since during the interwar years the US Navy and Army Air Corps could award production contracts to any manufacturer they chose – regardless of the design’s origin. Though not the original intention of the military’s policy, this was a powerful means for spreading leading-edge aircraft design knowledge across the nation’s industry. Manufacturing knowledge, however, depended on more informal channels, occurring routinely through personal communication, trade journals, worker mobility, tooling manufacturers and raw material suppliers, and the occasional act of industrial espionage. It is to manufacturing knowledge that we turn our attention specifically, because it enlightens our understanding of the design and production process.12

One of the more dramatic exchanges of manufacturing knowledge, brought about in part through mobilization, was the large-scale introduction of presses. Job shop production methods of the 1930s involved the manual bending and shaping of sheet metal using brakes, shears, and drop hammers.13 Low production runs did not justify large capital expenditures merely to increase throughput. When orders from England, France, and eventually the United States became significant, and the number of skilled workers proved insufficient to the task, the manufacturers adopted mechanical and hydraulic presses which had long been in use in the automobile industry. Though the automobile industry provided a powerful model for mass production, many of its techniques could not be readily transferred without some measure of adaptation. In this case, not only did aircraft stampings require greater accuracy, but they were in a different material – sheet aluminum versus sheet steel. Aluminum is considerably softer than steel, and exhibits different stretch and spring-back properties. In short, the same equipment could not be universally used in the two industries.

In the late 1930s, Henry Guerin, a department head at Douglas Aircraft, arrived at a way to use hydraulic presses with rubber as one side of a die (Figure 2). Gordon Ashmead, who chronicled aircraft manufacturing techniques in the 1950s, described this innovation:

[The Guerin Process] has been the greatest single contribution to the manufacture of all-metal airplanes. The almost universal adoption of this method of forming sheet metal was the factor that lifted the forming of sheet metal out of the hand-forming category and placed it into mass produc­tion. Most airplane manufacturers give it credit for winning our air war.

While the use of rubber in forming dates to at least the 19th century, Guerin developed techniques that made it appropriate for the aircraft industry. The process substituted layers of rubber for the female half of a die. After workers positioned a sheet of aluminum over the male die, the press lowered the rubber pad and compressed it against the aluminum to the point that the rubber began to flow. The aluminum sheet formed to the male die as the rubber applied equal pressure over the entire piece. Guerin’s process was distinct from earlier methods in that the rubber was restricted to a box that fit tightly around the male die, preventing the rubber from escaping and increasing pressure on the aluminum.14

The presswork championed by Detroit thus had to go through a translation before finding widespread application in the aircraft industry. The Guerin process would have been inappropriate for Detroit, where sheet steel would have quickly abraded the rubber. Not only was the Guerin process matched for aluminum, but it required only one metal die. At a time when tooling departments were working overtime to build jigs and fixtures, a fifty-percent reduction in diework was a welcome savings. Set-up time was reduced as well, since the rubber sheets did not have to be as accurately placed as a standard female die. Typically companies did not require many presses for the Guerin process, as a single large press with moving trays and 24-hour schedules could produce large quantities of stampings. In 1940, for example, Douglas operated a single 2,000-ton press with the Guerin process. As the war progressed, ever-larger presses came into use, including 5,000-pound presses with six work trays wherein a single shift could produce fifteen thousand parts. Ashmead indicates that the Guerin Process was most economical when quantities were between a few hundred and 50,000 parts – within the range typical of aircraft production. The Guerin process found widespread application throughout the airframe industry.15

The Guerin method is somewhat unusual in that most industrial processes were not patented at this time. Many processes were simply unpatentable, and so companies kept information proprietary through secrecy. Other processes, as noted, remained traditions of practice within a company’s tooling department. But even when a patent existed, as in the Guerin case, manufacturers quickly established distinct methods of using it, these too becoming part of local practice and forming a base of tacit knowledge within individual tooling departments. While Douglas’ patent on the innovation was sufficient to inform other manufacturers of the potential for such a process, it did not necessarily ensure good results. Surrounding the Guerin process arose a new body of engineering knowledge regarding its use, involving questions of press choice, die construction and arrangement, and appropriate pressures. The exchange of this accumulated tacit knowledge would eventually take place, but only after months and even years had elapsed, often traveling indirect routes between manufacturers. Typical of these indirect routes would be trade journals, tool manufacturers, and professional societies. In the case of the Guerin Process, representatives from Continental Rubber Works and Brewster Aeronautical (and later Eastern Aircraft – an aircraft division of General Motors) penned a series of articles on the many facets of the technique. Continental Rubber, though not strictly a tooling company, would obviously benefit from the adoption of the process. In this case Brewster and Continental Rubber surveyed different firms’ techniques and published them in Aero Digest, a leading trade journal of the time.16

It was not an uncommon practice for engineers and companies to champion particular techniques through articles and technical updates. With the outbreak of war, however, such indirect means of communication became difficult, as censorship restrictions limited the ability of trade journals to describe the latest technical advancements in detail. In some cases, the Army Air Force found itself reviewing material before publication; in many cases the resulting technical descriptions were so woefully vague that they could not have been of much assistance.17

Still, without other avenues, trade journals and user-producer relationships often provided the fastest pre-war means for accumulating and disseminating such esoteric information. Production and tooling information lacked the numerous channels of exchange that existed for other branches of aircraft technology. Tooling was not a subject for standardization within the Society of Automotive Engineers (SAE) or National Aircraft Standards Committee. As a practical art, tooling had no scientific component, and thus no society like the Institute for Aeronautical Sciences to pursue theoretical advancements. Not only were tooling patents excluded from the MAA, but the MAA agreement and military contracting policy intensified the importance of keeping production technologies proprietary. If significant aircraft design patents were to be shared under the MAA or transferred as part of a military design, it became all the more important to retain a competitive advantage in production techniques.

Aircraft have two main components: airframe and engines. To avoid disasters such as destruction of the second B-29 prototype and crew, February 18, 1943,3 a basic rule offlight test is that only one of these two main components should be unproven. In some cases this is accomplished by having a new airframe design use an established powerplant.4 This is not feasible if the new airframe also requires a new powerplant. Then existing airframes are modified to fit the new engine.

Thus, a B-29 Superfortress was used in 1943 as high-altitude jet-engine test bed for the General Electric 1-16 developed for the X-P59, America’s first turbo jet aircraft.5 A subsonic XF-4D swept wing fighter repeatedly was taken through the sound barrier in tests of GE turbines for supersonic aircraft. For the T-38 supersonic trainer, a modified FI02 carried the diminutive J-85 engine in its bomb bay and, when airborne, the tiny engine would be lowered hydraulically, air started, and then put through its paces. Flight testing of the gigantic J-93 engines for the X-B70 mach 3 bomber used a modified supersonic B-58 bomber with a J-93 engine pod slung in its underbelly (see Figure 12). Once engine development and testing had progressed

[3]

I have tried to relate the issues discussed in the preceding papers to some broader themes in sociology and the sociology of knowledge. I have tried to bring out the role of interests, the need for comparative study and the need for a typology of social forms that can help us see patterns in engineering styles. My aim has been to raise questions and problems of a somewhat more general kind than has been addressed in the papers themselves. The richness of the papers means that there are many possible lines of comment and inquiry that I could have pursued but did not. In particular I have neglected what might be called the political or institutional dimension. This means that I have rather short-changed those contributors who have addressed this side of things. There is much that could and should be said on such matters. For example, there is a whole nest of issues that emerges in these papers about the relation between the so-called ‘free market’ and various forms of government finance and subsidy. Time and again we see the importance of government research and state finance, whether it be tacit subsidy of civil aviation through military expenditure or the research effort of government laboratories. Prof. Roland quotes the figure of 85% of aerospace research deriving from government funding.

A second example, to which I cannot resist drawing attention, is provided by a significant observation made in three of the papers. Profs. Bilstein, Crouch and Hashimoto, respectively, all note how rapidly, in the early years of this century, the British government responded to the emerging phenomenon of the aeroplane. In

The Boeing-Douglas-Vega (BDV) pool was among the first of the major wartime aircraft cooperatives. As a product-oriented organization it illustrates the manner in which engineering knowledge is embedded within a firm. Before the attack on Pearl Harbor, the US government made arrangements for a number of standard aircraft types to be produced jointly by different manufacturers. Boeing, Douglas and Vega were to produce the B-17 bomber, a four-engine Boeing design begun in 1934. Implicit in such an arrangement was that Douglas and Vega would gain access to Boeing’s system of manufacture. Within this system Boeing had many individual production techniques unique to their company, some proprietary, others matters of Boeing’s manufacturing culture. Among the differences noted in the trade journals were Boeing’s drawing system, factory methods, and unique tools.18 To coordinate the three companies’ manufacturing, procurement, design changes, and scheduling issues, they formed the BDV Committee. Begun in May of 1941, the Committee was located in Seattle and included representatives from all three manufacturers as well as the Army Air Corps (later the Army Air Forces after its 1942 reorganization). In addition to a powerful executive committee, three sub­committees operated, including Tooling, Technical, and Procurement, all based in Seattle. Through these committees, Douglas and Vega were given “ready access to all phases of Boeing production.”19

The BDV Committee contemplated a wide-ranging transfer of information, extending beyond aircraft blueprints to include production information as well. Douglas and Vega received all the master tooling and jigs, as well as one complete aircraft and numerous sets of parts for initial production. From their factories in southern California, Douglas and Vega sent scores of workers to Boeing’s Seattle facilities to personally observe tooling construction and manufacturing methods. The extent of cooperation was so great that workers from all three companies joined in building assembly and sub-assembly jigs. Highlighting the relationship between production and organization, an integral part of the technology transfer to Douglas and Vega was managerial information, including major assembly man-hour lists, percentage breakdown charts, schematic drawings with man-hour indications, assembly books, parts cards, and parts list. These were all means for controlling and auditing production flow. Finally, Boeing learned to adopt some of the methods it found in Southern California, such as Douglas’ “production illustration” method.20

The members of the BDV Committee were not entirely unprepared for the obstacles posed by technology transfer. Indeed, they seemed quite aware of the fact that traditions differed dramatically between firms. Regarding tooling, Mac Short, the Vega Vice President of Engineering wrote:

A survey of existing tooling by the Tooling Subcommittee revealed that there was in existence in the Boeing plant a number of small tools for which there were no available drawings. These tools had been constructed in many cases by the workmen on the job during the preceding years of building B-17s. It was the decision of the BDV Committee that these tools represented “know­how” and would be of aid to the Douglas and Vega companies if the information could be transferred to them.

The exchange of such “know-how” was a necessary component of bridging these different engineering cultures. Indeed, the BDV Committee had to create a standard nomenclature for design and tooling engineers since all three companies had, essentially, different languages.21

Still, there was great confidence that the three manufacturers would be able to build identical aircraft. Boeing’s own press releases boasted that “[Boeing, Douglas and Vega] will turn out completely assembled four-engine bombers, identical even to the point of interchangeability of parts.”22 What members of the BDV Committee did not anticipate was the extent to which differing traditions of manufacture as well as conflicting production philosophies would preclude the complete adoption of Boeing’s production system. So long as differences remained, there were bound to be discrepancies between the aircraft produced.

Boeing used what it called a production-density system (later termed multi-line). The process brought nearly complete assemblies together at the very last stage of manufacture.23 Rather than a dominant single assembly line stretching from jigs to the hangar doors, each assembly had its own line terminating at final assembly (figure 3). Most of the installation work on the assemblies was completed before final assembly, meaning that aircraft spent the least amount of time in the stage that consumed the most space. Production-density was considered, “the antithesis of the elongated single production line.”24 The logic behind concentrating on assemblies and sub-assemblies as opposed to final assembly was ably described by Sidney Swirsky, writing with a touch of vitriol in 1943:

That the ideal straight-assembly line for aircraft is, after the fashion of the auto-makers, the final assembly line, is in itself strictly mythical. Actually, the aviation production line must start when parts are received in the form of raw stock at one end of the plant. In order to get more sub-assem­blies to the final line faster, the plant’s board of strategy, including the vice president of manufacturing, the factory manager and the plant layout engineer, must straighten out as much as possible the lines by which fabricated parts and sub-assemblies are to reach the final line. The final

line itself in an aircraft plant is much less a problem than are the sub-assembly lines; the complex­ity of routing and fabricating airplane sub-assemblies is the cause of the ever-threatening bottleneck which can stop production – not straightness or lack of it in the final line, [his italics]25

For Boeing, eliminating the long final assembly line and opting for numerous lines of shorter length had the advantage of requiring less expansive production facilities – one of the original reasons for its adoption in the late 1930s. But as prime contractors for production, not subcontractors, Douglas and Vega were not obligated to implement Boeing’s production-density system. There is no doubt that they benefited from Boeing’s experience, but they were forced to blend Boeing’s aircraft into their own companies’ manufacturing culture. Engineers at Douglas decided that straight-line final assembly was just the thing (Figure 4). Cradled by overhead carriers, workers mated fuselage sections which progressed from station to station down the line. At the first station of final assembly, the fuselage was already in one piece while workers attached the wing sections. Installation of parts and components continued through final assembly. Douglas would later call its straight-line assembly system “Flow-line” and apply it to the A-20 and A-26 aircraft lines. Vega opted for a third system, which it vaguely described as a
“breakdown system.” The focus here was on the fabrication of numerous small pre­assembled components that came together on master jigs. Vega, not unlike the other manufacturers, explained that its system attempted to make the best use of very limited floorspace.26

The result was three different ways of producing the B-17, and an indeterminacy about the best way to mass produce aircraft. An engineer writing on the Douglas system was careful to avoid making comparisons between different production ideologies, arguing that, “Among the many abilities of Americans… is the ingenuity by which different groups achieve similar or identical results, but with widely varying, sometimes seemingly opposite methods.” Unfortunately, the resulting aircraft from these methods were merely similar, and never identical. In fact, in October of 1943 General “Hap” Arnold specifically contacted the National Aircraft War Production Council (NAWPC) about the “non-interchangeability” of assemblies from different companies. He complained that “the different processes employed by the various manufacturers preclude the replacement of a Vega tail assembly by a Boeing or a Douglas tail assembly.” Despite rigid adherence to a single set of master gages and templates supplied by Boeing, differing production ideologies and traditions of practice brought about three different products.27

Figure 4. Douglas B-17 Major and Final Assembly (Building 12 and 13 of the Long Beach Plant). The assemblies and sub-assemblies move from left to right in the bottom building, and then from right to left in the upper building. Source: US Army Air Force, Air Materiel Command, Industrial Planning Branch, “Construction and Production Analysis: Douglas – Long Beach, B-17” (August 1946), exhibits 8 and 9.

While writers lauded the identical qualities of the aircraft in public, the manufacturers privately expressed their concerns about interchangeability within the Aircraft War Production Council. Even companies that were not part of the larger co-production schemes had great difficulty maintaining interchangeability. They attributed the problem to a number of causes, including sub-contractors who, despite having master jigs, often produced parts differently. Within the main airframe firms, they noted the following:

Here a distinct difference in manufacturing processes, tooling, model design, equipment and some small parts exists. The introduction of changes varies. The character of tooling is as diverse as the date, volume, and production rate of each contract, the plant lay-out, type and skill of personnel, climatic conditions, local fabricating habits, and material substitutions and improvisations.

The problems outlined by Hap Arnold were eventually addressed in military specification AN-I-21, which called for various classes of interchangeability. But while the military wanted interchangeable assemblies, sub-assemblies, and parts, such requirements would entail that engineering cultures attain some level of interchangeability as well. Ultimately the Aircraft War Production Council informed the government that the AN-I-21 specification would be impossible to implement across the industry. They argued that because “each company and product presents a specific design and manufacturing problem,” that solutions would need to be equally individualized.28

Despite the lessons of the BDV experience, technology transfer and interchangeability issues remained, even for subsequent cooperative efforts such as the B-29 pool. The Army had chosen the Boeing B-17 and Consolidated B-24 to be the standard large bomber types for the war, but they also sought to place a more modem and capable aircraft into production. This was the Boeing B-29. Like many other aircraft, it would be produced by a group of manufacturers, including Boeing – Wichita, General Motor’s Fisher Division, and North American Aviation. Eventually GM and North American were replaced by Bell (Marietta, Georgia) and Boeing-Renton, with the Fisher Division remaining as a subcontractor. In time, the Omaha plant of the Glenn L. Martin Company would also join the pool. As with the B-17 pool, there was a B-29 Committee established to handle the same coordination problems that arose on the earlier aircraft. Specific details from the B-29 pool give us additional evidence in understanding the difficulties of engineering exchange.

Engineering drawings, when properly rendered, are supposed to communicate all the essential aspects of a design, such that trained workers with appropriate tools can recreate the artifact to specification.29 While Boeing did supply thousands of drawings to the other contractors, all participants understood that these were insufficient to achieve complete interchangeability. Just as a skilled craftsperson uses the same measuring tape for a single project, so did Boeing. The company built master control gages in Seattle which served as the final word on all critical dimensions. These gages remained in Seattle, and were used not to construct aircraft, but to produce other master gages, which in turn would be used to check the accuracy of jigs and completed parts (see figures 5 and 6). Usually the gages bore only slight resemblance, if any, to the assemblies or sub-assemblies to which they corresponded.

Made of thick pieces of steel, the gages were extremely rigid; they conformed to actual aircraft parts at specific reference points – typically locations where a part or assembly would meet another. The B-29 Committee directed the following:

Master gages and master templates shall be used by all four prime contractors and their subcon­tractors to insure interchangeability of all parts which the Air Corps requires to be interchangeable…. Tooling for fabrication of parts shall be such as to insure that all parts are struc­turally and aerodynamically identical. It is not contemplated that tooling be identical. Such prime contractor and subcontractor may construct parts by the method or methods best suited to his shop equipment and practices, consistent with the above; however, the Committee must be assured by the contractor concerned that such tooling will produce satisfactory parts.

In theory, exact master gages distributed to all the manufacturers should have rendered the same parts, sub-assemblies, and assemblies.30

Two situations show the difficulty of using master gages as a way to communicate engineering information. In November of 1942, the B-29 Committee wrote: “Trial indicated that the extremely close angular tolerance required could not be successfully transferred from the Boeing furnished Master Gages.” While the

information about angular tolerances was encoded into the gage, it could not be used to ensure precision in the produced parts. In effect, the engineering information could not be extracted from the gage without the production of additional gages. A more interesting situation developed a year later when the Committee learned of discrepancies between the nacelles produced by GM’s Fisher plant, and the aircraft wings to which they were attached. The nacelles, when placed on aircraft from the Boeing-Renton facility, had attachment holes that were as far as ft a hole diameter off (a very large distance in a precision product). Boeing-Renton checked its own master control gages and found that their own parts were correct, but that the Fischer parts were not. Bell in Marietta experienced the same problems, but oddly, Boeing’s main B-29 plant in Wichita did not. They fit perfectly there and the plant had already turned out nearly 50 aircraft. The conclusion of the Committee was that,

.. .whether Fisher and Wichita are right or wrong in contrast to Boeing-Renton and Bell, it appears that the only practical answer is for Boeing-Renton and Bell to change their hole locations to coin­cide with the nacelles as produced by Fisher. This of course will mean the changing of the master gages at Boeing-Renton and Bell and the changing of the master control gage at Boeing-Seattle to agree with the masters at Wichita and Fisher.

Later, Fisher changed the process after the 210th set of nacelles – such that the discrepancies were removed. Why these particular nacelles fit at Boeing – Wichita was not explained, but it begs a question. If two of the most prominent engineering organizations of the 1940s, the Fisher Division of General Motors and the Boeing Company, could not transfer engineering information via accepted methods (drawings and master gages), what engineering organizations in the US were competent to produce and interpret such materials? Obviously the engineering precision that is supposedly built-in to engineering drawings and master gages must be complemented by not only technical skill, but additional forms of communication as well in order to bridge different engineering cultures. This explains, in part, Bell’s insistence that they be furnished a large quantity of parts so as to hand-assemble their first five aircraft. In addition, they asked for “two fuselages in sections, structure only… one fuselage in detail parts,” and finally a “sample airplane complete with installations.” All of these structures served to educate engineers and workers in a way that drawings and gages could not.31

Where gages were not used, drawings were supposed to suffice. One last incident concerning the B-29 gives us additional insight into the process of technological design. In August of 1942 the Fisher Division complained that it had yet to receive the drawings for the electrical wiring. Boeing responded, saying that such drawings were not produced in final form until after the first few aircraft were built. Boeing suggested that someone be sent from the Boeing Electrical Shop to personally advise Fisher, the implication being that this aspect of aircraft design routinely took place closer to the factory floor. Unlike the aforementioned cases where design and tooling information was somehow lost or changed in the transfer process, here we find that design information accrued as the product took shape. This is not surprising, but it suggests further that design trajectories are not fully determined on the engineer’s drafting table. Technicians, mechanics, and laborers not only misinterpret or lose initial design information, they create it as well.32

The BDV and B-29 Committees were product-specific cooperative organizations. Creating broader industry-wide cooperation fell to a number of groups, the most prominent of which was the Aircraft War Production Council (AWPC). Begun in April of 1942 by eight of the largest Southern California aircraft manufacturers, the AWPC was an unusual departure from normal peacetime competitive operations. It involved the transfer of knowledge, materials, labor, and facilities among former rivals.33 One of the intriguing characteristics of the AWPC was the wide-ranging mandate to share any information that could aid the prosecution of the war. The Board of Directors, composed primarily of the presidents of the West Coast Companies, strongly supported exchange and set a clear policy for Council activities:

It is the instruction of the Board of Directors that members of the Advisory Committees shall be free to interchange all information which will help the war production of the company receiving the information, without regard to protection of manufacturing processes which in normal peace times might be regarded as a secret.34

The sentiment was genuine and followed resolutely by the committees. Like the BDV and B-29 Committees, many of the market-based obstacles normally hindering technology transfer among competitors in peacetime were thus removed. Similarly, the AWPC did not simply lower corporate barriers; it established direct channels for exchange. AWPC Committee workers had strong incentives to cooperate with each other since their immediate superiors in the Council were also likely to be their superiors at their respective companies (see figure 7).35

While never completely isolated from each other before the war, the manufacturers generated much of their engineering information, especially production information, in-house. The principal exception to this was wind tunnel data, the province of affluent institutes and organizations. The challenge for the Council was to establish means not only for sharing information, but generating it cooperatively. Encouraging disparate groups to exploit outside information led to special efforts to tailor information retrieval to the likes of engineers. Mutual engineering research, like cooperative aircraft production schemes, would eventually take on the methodology of standardization in order to make research results portable. The regularity of exchange, while not a direct indicator of

successful technology transfer, indicates that at a minimum there was significant demand for outside information.

As early as 1936, the southern California manufacturers had begun their own in­house technical libraries. With the outbreak of war, it was to the Pacific Aeronautical Library (PAL), begun in 1941, that they turned as the locus for technical exchange. Operated by the Institute of Aeronautical Sciences, the PAL quickly became a natural partner to the AWPC and was ultimately subsumed into the Council itself in 1943. Collection, indexing, and exchange were the library’s three functions, and it became the principal means for publicizing and transferring written information between companies.36

While the PAL did serve as a simple repository, its real strength lay in its ability to make a wide range of materials available to professionals. In October, 1942, the librarians began an indexing project oriented towards engineers and comprising materials well beyond the PAL’s own collection. Finding traditional subject headings inappropriate for aircraft design and production, the librarians created their own. The standard indexes, including the Library of Congress subject headings, the Engineering Index, and the Industrial Arts Index, approached aviation in far too general terms. In consultation with workers from local companies, the librarians exploited the engineer’s lexicon. For example, they classified materials by vendors or trade names rather than under broadly scientific material categories. To satisfy the needs of both the design and production aspects of aircraft manufacture, topics were given multiple entries (for example, one applying to a particular material, another applying to the processes surrounding that material). Furthermore, the PAL attempted to be all inclusive, a single reference point for all matters. The index comprised articles from all the relevant engineering and trade journals, including NACA and SAE publications.37 Eventually the PAL index came to include the collections of member companies, creating a meta-collection available to everyone. Engineers could thus easily order information and have it quickly copied and delivered. Interchange among companies, institutes such as Caltech, and the PAL grew large enough to require a messenger service that traversed a 120-mile route between libraries and companies twice a week.38 The PAL’s indexes were reprinted many times over, being sent to manufacturers across the US and the NACA.39

As table 1 indicates, the PAL served customers throughout the local aviation community. Reinforcing the idea that the PAL acted as an information service, as opposed to a simple repository, is the high percentage of research questions asked relative to other services; they account for over 36% of the total. It is also interesting to note that Vega exploited the service the most, even though it was a smaller company compared to Douglas, Lockheed Aircraft Corporation, and North American Aviation. One probable explanation is that the larger companies were much more self-sufficient, and that the PAL was relatively more important to smaller firms, including subcontractors like Adel Precision Products Corporation and AiResearch. As this was a formative period in aircraft industry subcontracting, technology transfer and the establishment of basic aeronautical expertise would have been critical to these smaller companies.

The operation of a central library fulfilled only a small portion of the Council’s technology transfer goals. A principal objective – eliminating duplicated research among manufacturers – remained for the Council’s committees. The airframe firms continually carried out research in many areas, not just product design, but tests on materials and production processes too. These activities are not well known since the reports were normally proprietary, and tended to be destroyed as newer information superceded the old. They were directed at very specific design and production issues, giving information on how to do something, as opposed to why something occurs. In this, they were quintessential^ engineering, not scientific reports. Table 2 gives the results of the October 1943 survey in which the Engineering Committee found a total of 321 different research projects at all member companies (for an average of 49 projects per company, per month).40 Aside from the sheer volume of research, it should be observed that little of it was in the area traditionally associated with aircraft research – aerodynamics, which had the least number of projects of any other category. Not only was theoretical research a small fraction of the overall research scheme, but studies on manufacturing problems were as numerous as those on design-oriented research. Such research differed from traditional corporate research and development in that it was not restricted to a laboratory setting. What we might call production-design research took place at many levels throughout the factory and

involved a wide variety of personnel, including engineers, technicians, tooling designers, and machinists. The research represented a systematic effort to quickly produce locally valuable design data, including design data for production.

The extent to which companies shared this information with each other can be gleaned from AWPC Engineering Committee statistics. Unfortunately only a partial record of these statistics remain, as indicated by table 3. Many different things

Source: Meeting Report, Engineering Committee, 9 Oct. 1943, Committee Reports on Production Division (September 15 – October 15, 1943), box 24, NAWPC.

Sources: Meeting Report, Advisory Committee on Engineering, 2 Jan. 1943, Committee Reports on Production Division (December 1942), box 22, NAWPC; Meeting Report, Advisory Committee on Engineering, Committee Reports on Production Division (January 1943), box 22, NAWPC; Meeting Report, Advisory Committee on Engineering, 3 April 1943, Committee Reports on Production Division (March 15 – April 15, 1943), box 22, NAWPC; Meeting Report, 7 August 1943, Committee Reports July – Aug. 1943, box 23, NAWPC; Meeting Report, 4 Sept. 1943, Committee Reports Aug. – Sept. 1943, box 23, NAWPC; Meeting Report, 8 January, 1944, Reports on Committee Activities August 1944, box 23, NAWPC; Meeting Report, 6 May 1944, Reports on Committee Activities June 1944, box 25, NAWPC; Meeting Report, 8 July 1944, Reports on Committee Activities, box 25, NAWPC; Meeting Report, 9 Dec. 1944, Reports on Committee Activities January 1945, box 25, NAWPC; Warplane Production 2, no. 3-4 (March – April 1944).

a The AWPC EC number presumably is for information sent to the AWPC EC, since the Engineering Committee would not have been interested in how other organizations were benefiting the AWPC.

b The cumulative figures are those given by the AWPC. This includes some material that is not reflected in the given monthly numbers.

qualified as an incident of engineering exchange, though all of the data listed falls in the information category, as opposed to tooling or material exchanges. For the given sample, exchange between member companies accounts for 60 percent, while exchange with the East Coast accounts for 12 percent, and non-members for 5 percent. Because the Council was specifically interested in the amount of time saved through cooperation, Table 3 does not include duplicated reports. Additional statistics show that many of these engineering reports were reprinted many times over. For example, in August 1943, the AWPC mailed a total of 22,582 articles, 10,193 engineering reports, and 319 miscellaneous publications. Respective totals from the beginning of AWPC operations in April 1942 to the first of September, 1943, 17 months, are: 179, 241; 91,284; and 319.41

Underscoring the inability of written information to communicate all engineering knowledge was the role played by interplant visits. For the months in which they were included in Council statistics, they count for about ten to twenty percent of total exchange. Sometimes plant visits were incorporated into a panel’s meetings. For example the May meeting of the Subcommittee on Tooling Coordination took a tour of the Boeing plants in Renton, Washington. Minutes from the meeting note the following:

The Committee met at 10:00 a. m. and a tour was conducted through both Plant #1 and Plant #2 where the manufacture of roll forming, curving and stretching of sheet metal was in process. The committee was very interested in our methods of manufacture, and special interest was shown in the movable stretching tables built by Consolidated. The tour came to an end at 12:30 p. m. Discussion was then opened on the subject, and Mr. Englehardt, General Foreman of the Metal Bench Department, answered all questions asked pertaining to rolling, curving and stretching of sheet metal parts.42

This kind of interaction represents the height of tacit knowledge transfer. These processes existed in physical form on the plant floor and in the activities of the laborers. Such practices rarely warranted internal reports or studies, and in some cases, might only be known to shop workers. For example, it came to the attention of the Testing and Research Panel that “it is apparently common practice for draw press operators to heat aluminum alloy in warm water just before drawing.”43 Meetings between tooling engineers would have been insufficient to communicate all such facets of a production system. Instead, interplant visits were a crucial method of communicating differences in engineering practice and shop culture. Even in cases where there was published information, plant tours could be highly educational. In one instance, members of the Project Group of the Methods Improvement Panel touring the Northrop factory were able to each try their hand at Northrop’s Heliarc welding process44

On the occasions where the Council found common problems among the manufacturers, it often initiated cooperative research projects, or at least brought existing research groups together. Here again, the critical exchange of engineering information was impeded by differing traditions. In many cases, Council committees and subcommittees resorted to methods of standardization in order to overcome these divides. As such, standardization can be considered a means of achieving technology transfer. For the AWPC, standardization was unlike the SAE’s efforts in the automobile industry, where common parts, materials, and measures permitted external economies. Rather, the standardization of things like test procedures permitted the exchange and comparison of engineering knowledge. It permitted one company’s test results to be compared against another’s. This produced an external economy of a different kind, one based on abstract technical data rather than physical parts or materials. While the AWPC had removed proprietary restrictions on the exchange of data in 1942, it would require standardized company practices before data were truly interchangeable. Standardization efforts did not end there though. AWPC committees standardized language, defects, procedures, and training.

Standardization was not only a means of achieving internal and external efficiencies, it was a familiar method of establishing order and creating novel systems of engineering knowledge. It implied the exchange of best-practice technology as well as the consensual definition of what was appropriate behavior for a worker, an engineer, or a manufacturing firm. The pervasive and continued recourse to standards activities by engineers was due in large part to the fact that this was an extremely familiar form of engineering exchange within the context of industrial coordination. The AWPC committees themselves had been established in a manner very similar to the SAE’s committees, both characterized by technical expertise, small groups with a tight focus, company representation, and industrial coordination.45

CONCLUSION

In the early days of mobilization, it was commonly assumed that America’s industrial strength could be redirected towards the mass production of armaments. In an atmosphere of hopeful naivete, planners believed in a kind of design – production modularity, where one company’s high-performance designs could be transferred to another company’s high-performance production system. Ultimately the country did convert to wartime production and achieve remarkable success in the mass production of aircraft. Yet it required a painful education, one that exposed the extent to which design and production were embedded within a company’s engineering culture and traditions of practice. One might be able to explain away Ford’s well known difficulties in producing the B-24 as simply a case of inter­industry technology transfer, but just as many problems appeared among longtime members of the aircraft industry.

In ways that are often invisible to engineers and managers within a company, the mixing and matching of design and production systems and widespread technology transfer illuminates the nature of engineering information. Engineering knowledge is not solely the province of the research engineer or the product designer. Nor is it embodied succinctly and completely in numerical data or technical drawings. On the way to the final product, information is both lost and accrued. In an ironic turn to the historiography of technology as knowledge, we find engineering knowledge

embodied in physical artifacts, not simply in the final product as we would expect, but in design and production. In tools, machinery, processes, and plant layout, information is perpetuated within a company, reinforcing workplace traditions while simultaneously reflecting workers’ traditions of practice.

The process of production, hidden by scholarly emphasis on product design and performance, complicates our understanding of engineering knowledge. For the manufacturing firm, the ability to design and produce is not the function of a single engineer, but of an organization. We can neither presuppose that design exists outside of production, nor model the factory as a linear process from idea to finished product. Underlying this argument are the different factory actors who execute some manner of design: aeronautical engineer, structural engineer, production engineer, tooling engineer, machinist, foreman, worker, and occasionally, manager. The engineering department may arrive at what it believes to be a complete design, but what rolls out the factory doors will in all likelihood be a different product.

The historical circumstance of wartime cooperation is itself remarkable. The quantity of exchange, where it is documented, is staggering. Both the product – oriented and industry-wide organizations found ways to transfer information, however ungainly they might seem to those who believe in the power of the technical drawing. Drawings were complemented by thousands of master gages, jigs, fixtures, tools, and the occasional example aircraft. Where documents and physical devices could not guarantee precision, these organizations resorted to the exchange of personnel. Interchangeable airplane assemblies were as much a result of practice as they were of accurate measurement. So too, the AWPC promulgated standardized procedures in order to make engineering information interchangeable. Standardization served to expunge critical differences in culture and practice.

What this history tells us is that for many years prior to World War II, firms developed their own ways of doing things. They were content with their handling of engineering information so long as it produced the desired artifact. In manufacturing, companies are held to the success of their product, not the outside reproducibility of their engineering information, as in a scientific laboratory. World War II subjected the production process to new scrutiny as manufacturers attempted to understand why engineering information was not truly portable. In the end they found that production systems were as varied and idiosyncratic as the aircraft they produced.

INTRODUCTION

The “modem” American commercial airport was invented during the two decades before the start of the Second World War. This opening statement has to be written in the passive voice because while it is easy to speak of the invention of airports, there are no inventors – no “Wright Brothers” – responsible for this remarkable technological system which makes commercial air transportation possible. This paper explores a major aspect of the “invention” of airports which was the process of standardization of airport design. In particular, it emphasizes the interaction between and among three separate groups of professionals – engineers, architects and city planners.

The idea that airports were technologies which needed to be designed, let alone something which required the services of “experts” to undertake, did not always exist. Orville Wright argued in a special editorial for Aviation magazine in 1919 that: “The airplane has already been made abundantly safe for flight. The problem before the engineer today is that of providing for safe landing.”1 Few paid much attention to Wright’s pronouncement though. Commercial air transportation entrepreneurs and other aviation enthusiasts wanted to emphasize the image of an airplane whizzing freely through the sky unhindered by all earthly obstacles. Above all else, this preoccupation with unfettered speed worked to eclipse a nascent discussion about the infrastructure needed to transform civilian aviation from mainly a random and recreational activity to a regular and commercial operation.

As various enterprises got underway, most notably the U. S. Post Office’s Air Mail Service, attitudes changed. Whether pilot or passenger (or even just the shipper of a letter or parcel), only the knowledge that it was possible to depart safely and then later return to earth made flight desirable. Consequently, interest in the details – the technical problems associated with safe landing – became paramount. Subsequent sections of this paper describe how first engineers, then architects, and finally city planners came to form a partnership of experts whose coordinated services were required for the design of airports. The creation of this specialized technical community helped fix the fundamental design, shape and purpose of the commercial airport, as well as the technologies and techniques to build and operate them.

301

P. Galison and A. Roland, Atmospheric Flight in the Twentieth Century, 301-322 © 2000 Kluwer Academic Publishers.

The few historical accounts of airport development assume (either explicitly or implicitly) a deterministic relationship between airplane and airport. In other words, the size, shape, weight and speed of the airplane determined the physical characteristics of the place for taking-off and landing. Standardization of airport design then, it is argued, is a consequence of the emergence in the 1930s of the “modem” airliner, most especially the Douglas DC-3.2

Closer examination reveals a different story. Powerful political, economic, and cultural forces profoundly influenced airport design; even more than the performance characteristics of airplanes (which for commercial transports only rarely were designed to exceed the capacity of the existing infrastructure). Especially important to understanding this history is the story of the formation of the unusual interdisciplinary technical community responsible for airport design. It would be an alliance among professionals more typically noted for their extreme professional rivalry. Yet, there was more cooperation than conflict and that fact is critical to understanding how airport design was standardized.

This is not to say that there was no professional rivalry but the “struggle” among engineers, architects and city planners was limited.3 The relative harmony came through the steady expansion of the definition of what constituted the technological functions of an airport. Increasing consumer demand for air transportation profoundly shaped the technology of airports. So too did the participation of the federal government. The partitioning of a technological design problem does not necessarily result in positive relations among the participants but in this instance these factors served to mitigate conflicts. At the start of World War Two, this cooperative working arrangement among engineers, architects, and city planners mediated by government officials was manifest in an airport system design that had become a “normal technology.”4

Archibald Black was a budding entrepreneur who found his niche as an airport engineering expert in the early 1920s. Bom in Scotland in 1888, Black emigrated to the United States in 1906 and became a citizen in 1913. His education was something of a hodge-podge, including three years at New York City’s Cooper Institute, plus courses at the Cass Technical Institute in Detroit and Columbia University’s extension school. Black became an electrical engineer and worked on a variety of electrical and construction jobs until discovering aviation in 1910. In 1915 he began working for Curtiss Aeroplane Company in 1915, transforming a hobby into a full-time vocation. Within a year, he had become an airplane designer for Curtiss. In 1917, he took a new job as chief engineer for L. W.F. Engineering Company where he designed the first airplane to incorporate the famous Liberty engine. A wartime stint in the Navy brought him to Washington where he was put in charge of preparing all of the Navy’s aeronautical specifications.5

After the war, he began a long and productive career as a consulting engineer. Over the decade he would emerge as the most influential figure in the design of

American airports. But it was with much bravado that the 33-year old would assert in a letter to Secretary of Commerce Herbert Hoover that: “in addition to being the most experienced consulting engineer in this field in the country, the recent dissolution of a competing firm now makes ours the oldest also.”6 It was a bold claim that said more about the newness of aviation than it did about Black’s engineering expertise.

Still, within a year, Black had managed to gain quite a bit of attention for himself. His article, “How to Lay Out and Build an Airplane Landing Field,” appearing in Engineering News-Record, was selected for reprinting by the National Advisory Committee for Aeronautics in its Technical Memorandum Series. Aerial Age said the paper was “probably the first really constructive discussion of the subject published.”7 Black, who had spent much energy beating the drums in a generic pitch for the contributions engineers could make to aviation, suddenly found a rhythm which resonated. While he never lost his broad interests in advancing commercial air transport, Black became a champion for “scientifically arranged landing fields in America.”8

In 1922, landing fields (the word airport was not in common usage until the late 1920s) for airplanes did not seem like much. In comparison with the stupendously complex civil engineering design projects for bridges, tunnels, canals, railroads, and hydroelectric dams, the small square (usually) sod fields with a wooden hanger, a gas tank, and a wind cone hardly seemed to demand the services of an expert, let alone a specially-trained engineer. Appearances were deceiving according to Black. Even in his earliest articles, he argues for careful planning based on significant technical knowledge of civil and mechanical engineering, meteorology, and construction technology. The term “scientifically arranged” referred only to the design of a facility which enabled aircraft to get in and out of the air.

Thus, Black’s early articles emphasize that the shape and size of the plot required the detailed scrutiny of meteorological data. The sky was not benign and aircraft were fragile. Soil samples were required to design the surfaces for take-off and landings. Regular use required a prepared surface able to bear the loads exerted by an airplane. There were formulas for calculating the impact of obstacles such as trees or tall buildings, landing gear stresses, and how much grass seed should be sown. The type and arrangement of buildings needed to be optimized for safety and functionality. And finally, it was necessary to develop various communications systems that enabled ground personnel to relay vital information to airborne pilots.9

Black was not the first to suggest these things, however. Desperately casting about for a postwar mission, aviation officers in the Army embraced the idea of sharing with the public the knowledge gained during the war. Right after the war the Army had decided to plan its own national airway system and had begun to survey potential sites for airports. That work was endorsed by President Harding who stated during a special message to Congress in April 1921 that the Army Air Service should take the lead and aid the “establishment of national transcontinental airways, and, in co-operation with the states, in the establishment of local airdromes and landing fields.”10

Harding’s pronouncement led to the establishment of the Airways Section within the Training and War Plans Division of the Air Service the following December and the creation of a “Model Airways” program in June 1922. Fulfilling its obligation for the “promulgation of information pertaining to airdromes and airways,” the Air Service issued “Airways and Landing Fields,” a new Information Circular in March 1923.11

Like its earlier “Specifications for Municipal Airports,”12 this Army circular includes details about what constituted an ideal landing field. The Army “ideal” was developed in the context of an expanded model airway that served, first and foremost, military objectives which were to provide for the aerial defense of the nation. The Army had included in the specifications for its very first airplane (ordered from the Wrights in 1908) that the airplane be easily disassembled and, preferably, easily shipped in a standard Army wagon crate 13 After the war, however, the Army expected to be able to fly from base to base. The typical range of aircraft in the early twenties required frequent stops (indeed, the working assumption of the time was that safety required an airfield every 10-25 miles). Lieutenant F. O. Carroll, a Landing Field Officer for the Army Air Service, wrote that the Army would follow along the same routes of the Conestoga pioneers and continental railroads. His challenge was to “arouse the interest and secure the co-operation of the cities in this important national enterprise, and make them see that their future and the future of the country depends on the establishment of landing fields.”14

What the Army wanted most of all was air fields that conformed to a set of standards. Recognizing that all communities could not afford to build the same sort of facility, the specifications proposed four classes of landing fields. A fourth class field was for emergency use only; its landing surface might be just a narrow strip of land about 600 feet in length. The very best, the first class field, would be a square field at least 2,700 feet in any direction. The surface for landing and taking-off would be improved – level, smooth, sodded and well-drained. All approaches to the field would be free from obstacles such as telephone poles, electrical lines, trees and tall buildings. The field would be well marked with a landing circle (a large white circle, 100 feet in diameter, set into the ground), wind direction indicators, and the name of the facility in 15 foot high letters. Finally, a well-developed field had hangars, shops, gasoline, oil, telephones and transportation to the nearest town. Second and third class fields had slightly smaller dimensions and fewer amenities although standards for marking the field remained the same. There was little change in the basic requirements for air fields between 1920 and 1922, although the “square” field design was no longer the only preferred landing field shape – An “L” shaped field was considered almost as good as the square-shaped field.15

There were some who felt the Army’s vision for airways and airports left something to be desired. It was not the generic vision of a well-functioning system of airports and airways that was problematic to these individuals. Rather it was the belief that the Army’s own airfields had little to commend about them and that most Army officers were unable to render good advice on how to design and build an airport. The leading critic was the engineer, Archibald Black. Black’s main purpose in making his complaint was to lend support to his conviction that the best possible training for a would-be airport designer was engineering. Pilots, of course, had a great interest in airports, but Black believed that merely possessing the skill to pilot an airplane was not adequate qualification to design an airfield.16

Black actually had three related concerns. His first concern related to the quality of existing Army air fields. The tremendous urgency to build military air fields during the war had meant relying on firms which had neither experience in designing and building such facilities nor the time to make a thorough study of the problems that were unique to aviation. This observation was not intended to be an indictment of either the Army or the construction companies. Rather, Black believed that given the poor result there was “no reason for ignoring past experience and common sense, and arranging fields without careful consideration of the entire subject beforehand.”17

Black’s second concern about relying on the Army example was that subsequent to the war, some of the fields the Army had completed were of very high standard – too high to be meaningful for the average municipality. “Army equipment,” he wrote, “is considerably more elaborate than is likely to become necessary at municipal fields for some time.”18 In one article scrutinizing various airport designs, Black was equally critical of fields that were too large or elaborate as he was of small or “carelessly” organized layouts. The former represented unforgivable profligacy with public monies; the latter were safety hazards.19

Black’s third and most doggedly asserted criticism was reserved for “the customary practice of consulting some local hero, in the person of a former Army pilot.. ..”20 The veteran pilot while worthy of veneration was unlikely to be familiar with both air field design and current commercial aircraft technology. Black repeated this assertion at every opportunity. In American City he wrote, for example, “it is more advisable either to obtain the services of a specialist or to appoint, as a substitute, a well-rounded committee to do the planning. Any other policy may prove surprisingly expensive at some later date and the cost may be counted in human lives as well as in dollars.”21

The resolution to all these concerns, according to Black was to establish airport design as an engineering function. Thus, the essence of good design was one which placed safety paramount, followed by the careful expenditure of funds.22 His fullest statement on this subject was in an article written for Landscape Architecture in 1923. The piece, “Air Terminal Engineering,” began: “The selection, arrangement and construction of aircraft landing fields and other types of air terminals represent an entirely new phase of engineering which is yet very much in the paper state.”23 The benefits to relying on an “airport engineer” were three-fold. First and foremost, while a bad design would absolutely result in crashes, a good design would prevent costly accidents. Second, good design would result in the best overall construction price. And third, Black believed an engineer was best suited to determine the airport location, as well as to plan the buildings at the facility to take advantage of existing ground transportation and power, water, and telephone utilities.24

Black had recently returned from a European study trip and his head was swimming with details related to airport construction. It was in Europe that he first formed the belief that airports ought to be engineered. He was especially enthusiastic about the developments at Croydon Airdrome in London. There were three airports in London but Croydon had become the city’s main commercial facility after the war. Black was undeniably impressed by what he saw but two things seem to have been especially influential – Croydon’s airport lighting system and its carefully designed landing area surface.25

In his influential Engineering News-Record article, it becomes especially clear why Black believed airport design should be done by engineers. In this article, Black makes a strong case for special analytical skills an engineer might bring to the problem of design. Existing grass fields were fast proving inadequate. As the number of airplanes taking off and landing at any given airport increased beyond a small handful, the sod-covered surfaces rapidly disintegrated. Airplanes of this period were all “tail-draggers” and their skids carved deep gouges into rain-softened surfaces. The Army had experimented with concrete landing and take-off crosses. Black urgently warned readers about the pitfalls of concrete as a landing surface (it was fine for taking off). In addition to the great expense, it was too smooth for an era when airplanes did not have brakes and pilots relied on a rougher surface to slow down the vehicle.26

There was a prevalent misconception that turf was “softer” than concrete and therefore safer. Black dismissed such ideas. The landing gear and tires of an airplane were designed to bear specific loads; one surface or another made little difference to the airplane (assuming the initial design was adequate). Based upon Black’s analysis of airplane tire sizes and their normal loads, he readily concluded that most landing surfaces were not adequate. He became an early advocate for “runways,” special strips 75 feet wide and 1000 feet long. Runways were part of the larger landing surface area but they would be specially designed and constructed to carry the full load of an airplane during take-off and landing. This necessitated careful preparation of the subgrade as well as some kind of special surfacing. It also required renewed attention to drainage matters. Conventional roads relied on crowning as the central element of drainage design but Black found that such curved surfaces were very hazardous.27

Black pursued his studies of runways, consulting with highway engineers as well as experts at the Bureau of Public Roads. In later articles (and books) he provided much more substantial discussion of runway surfaces as well as providing specifications and cost analysis information.28 These articles were of noticeably different character from the literature being supplied by the Army which continued in the “do-it-yourself’ manner. What Black wanted to demonstrate was that airport design would only mature when the designers were able to comprehend fully the assumptions and methods employed by aircraft engineers in the design of airplanes. Black, who had spent several years designing aircraft, believed that the symbiotic relationship between airplane and airport could be best characterized in engineering terms – in a language of speed, power, weight, size, wing loading, power loading, lift, drag, and aspect ratios. The development of airports and airplanes needed to proceed in a coordinated fashion or commercial aviation would falter. These lessons became especially ingrained during his tenure with the group of engineers writing the Aeronautic Safety Code. By 1925, it was clear that Black had made his point, although the full impact of it would not be perceived until the passage of the Air Commerce Act in May 1926.29

In the NTSB’s final judgment of probable cause was an explicit reference to the fact that the captain failed to reject the takeoff “when his attention was called to anomalous engine instrument readings.” Though not formalized in the probable cause assessment, the investigative team did comment elsewhere in the report that the Safety Board strongly believed in the training program of command decision, resource management, role performance, and assertiveness. As the NTSB pointed out, it had already, in June of 1979 (A-79-47), recommended flightdeck resource management, boosting the merits of participative management and assertiveness training for other cockpit crewmembers. Here a new analytical framework entered, in which causal agency fell not to individual but to group (social) psychology. That framework (dubbed Cockpit Resource Management or CRM) was fairly recent and came in the wake of what became a set of canonical accidents. The NTSB – interpreted record of Air Florida flight 90 became a book in that canon.

For United Airlines, the transformation in their view of CRM came following the December 28, 1978 loss of their flight UA 173. Departing Denver with 46,700 pounds of fuel, with 31,900 predicted necessary for the leg to Portland, the DC-8 came in for final approach. When the gear lowered, those in the body of the plane heard a loud noise and sharp jolt. The captain felt that the gear had descended too rapidly, and noted that the gear lights did not illuminate. Asking his second officer to “give us a current card on weight, figure about another fifteen minutes,” he received a query in reply, “fifteen minutes?” To this, the captain responded “Yeah, give us three or four thousand lbs. on top of zero fuel weight.” Second officer: “not enough. Fifteen minutes is really gonna really run us low on fuel here,” then later: “we got about three on the fuel and that’s it.” When the first officer urged, “We’re going to lose an engine,” the captain responded “why?” To which the first officer responded “Fuel!” Within eight minutes the plane was down in the woods outside the city, with a loss of ten lives.19 The canonical interpretation read the accident in terms of a failure of communication: Why, United Airlines personnel wanted to know, was the captain not listening to his officers?

According to United Airlines’ CRM curriculum of the mid 1990s, the conversion of Delta Airlines to CRM came seven years after the United 173 crash, in the aftermath of its own disastrous flight 191. Approaching Dallas Fort Worth airport on August 2, 1985, Delta’s L-1011 hit a microburst, descended into the ground, and disintegrated. The question raised by investigators was why the otherwise prudent captain had entered an area of known lightning – that is to say a thunderstorm – close to the ground and in a shaft of pounding rain. “Probable cause” included the decision to enter the cumulonimbus area, a lack of training in escape from windshear, and lack of timely windshear warning. Unlike the captain of United 173 or Air Florida 90, no one suggested here that the Delta captain was not listening to the flightcrew. Instead, “given the fact that the captain was described as one who willingly accepted suggestions from flightcrew members,” the Board did not infer that they were intimidated by him. But because neither first nor second officer dissented from the continued approach, the NTSB held the flightcrew responsible for the decision to continue. “Suggestions were not forthcoming,” concluded the investigation, on the basis of which the NTSB argued that air carriers should provide formal cockpit resource management and assertiveness training for their crews.20

When, in the mid 1980s, the airlines began to develop their CRM courses, they invariably turned back to the by-then widely-discussed proceedings of a meeting held

under NASA’s auspices in San Francisco over 26-28 June 1979. In various ways, that conference set out the outline for hundreds of courses, books, and pamphlets designed to characterize and cure the “dangerous” failures of communication on the flightdeck. Most prominent among the speakers was Robert Helmreich, a social psychologist from the University of Texas at Austin, who came to the problem through his work on Navy and NASA crew training efforts for the space program. Psychology (Helmreich declared at the San Francisco meeting) had so far failed those in the cockpit. On one side, he noted, there was personality psychology which had concentrated solely on exclusion of unacceptable candidates, failing utterly to capture the positive individual qualities needed for successful flight. On the other side, Helmreich contended, social psychologists had so far ignored personality and focused on rigorous laboratory experiments only loosely tied to real-life situations. Needed was an approach that joined personality to social interaction. To this end he advocated the representation of an individual’s traits by a point on a two-dimensional graph with instrumentality on one axis and expressivity on the other. At the far end of instrumentality lay the absolutely focused goal-oriented pilot, on the extreme end of expressivity lay the pilot most adept at establishing “warmer” and more effective personal relationships. In a crisis, (argued the authors of United’s CRM course) being at the high end of both was crucial, and likely to conflict with the “macho pilot” who is high in instrumentality and low in expressivity.21

In various forms, this two-dimensional representation of expressivity and instrumentality crops up in every presentation of CRM that I have seen. Perhaps the most sophisticated reading of the problem came in another plenary session of the 1979 meeting, in the presentation by Lee Bolman from the Harvard Graduate School of Education. Bolman’s idea was to pursue the mutual relations of three different “theories”: first, there was the principals’ short-term “theory of the situation” which captured their momentary understanding of what was happening, here the pilots’ own view of the local condition of their flight. Second, Bolman considered the individual’s longer-term “theory of practice,” that collection of skills and procedures accumulated over a period of years. Finally, at the most general level, there was a meta-theory, the “theory-in-use” that contained the general rules by which information was selected, and by which causal relationships could be anticipated. In short, the meta-theory provided “core values,” “beliefs,” “skills,” and “expected outcomes.” Deduced from observation, the “theory in use” was the predictively successful account of what the subject will actually do in specific situations. But Bolman noted that this “theory-in-use” only partially overlapped with views that the subject may explicitly claim to have (“the espoused theory”). Espoused knowledge was important, Bolman argued, principally insofar as it highlighted errors or gaps in the “theory in use”:

Knowledge is “intellectual” when it exists in the espoused theory but not in the theory-in-use: the individual can think about it and talk about it, but cannot do it. Knowledge is “tacit” when it exists in the theory-in-use but not the espoused theory; the person can do it, but cannot explain how it is done. Knowledge is “integrated” when there is synchrony between espoused theory and theory-in-use: the person can both think it and do it.22

Bottom line: Bolman took the highest level theory (“theory-in-use”) to be extremely hard to revise as it involved fundamental features of self-image and lifelong habits. The lowest level theory (“theory of the situation”) might be revised given specific technical inputs (one gauge corrected by the reading of two others) but frequently will only actually be revised through an alteration in the “theory of practice.” It was therefore at the level of a “theory of practice” that training was most needed. Situations were too diverse and patterns of learning too ingrained to be subject to easy alteration. At this level of practice could be found the leamable skills of advocacy, inquiry, management, and role modification. And these, Bolman and the airlines hoped, would contribute to a quicker revision of a faulty “theory of the situation” when one arose. CRM promised to be that panacea.

Textbooks and airlines leaped at the new vocabulary of CRM. Stanley Trollip and Richard Jensen’s widely distributed Human Factors for General Aviation (1991) graphed “relationship orientation” on the у-axis against “task orientation” on the abscissa. High task orientation with low relationship orientation yields the dreadful amalgam: a style that would be “overbearing, autocratic, dictatorial, tyrannical, ruthless, and intimidating.”

According to Trollip and Jensen, who took United 173, Delta 191, and Air Florida 90 as principal examples, the co-pilot of Air Florida 90 was earnestly asking after take-off procedures when he asked about the slushy runway departure, and was (according to the authors) being mocked by captain Wheaton in his response “unless you got something special you’d like to do,” a mockery that continued in the silences with which the captain greeted every subsequent intervention by the co­pilot.23 Such a gloss assumed that copilot Pettit understood that the EPR was faulty and defined the catastrophe as a failure of his advocacy and the captain’s inquiry. Once again agency and cause were condensed, this time to a social, rather than, or in addition to, an individual failure. Now this CRM reading may be a way of glossing the evidence, but it is certainly not the only way; Pettit may have noted the discrepancy between the EPR and N1, for example, noted too that both engines were reading identically, and over those few seconds not known what to make of this circumstance. I want here not to correct the NTSB report, but to underline the fragility of these interpretive moments. Play the tape again:

F. O. Pettit (CAM-1): “That’s not right… well …”

Captain Wheaton (CAM-1): “Yes it is, there’s eighty”

Pettit (CAM-2): “Naw, I don’t think that’s right …. Ah, maybe it is.”

Wheaton (CAM-1): “One hundred twenty”

Pettit (CAM-2): “I don’t know.”

Now it might be that in these hesitant, contradictory remarks Pettit is best understood to be advocating a rejected takeoff. But it seems to be at least worth considering that when Pettit said, “I don’t know,” that he meant, in fact, that he did not know.

United Airlines put it only slightly differently than Trollip and Jensen when the company used its instructional materials to tell new captains to analyze themselves

The Grid Approach To Job Performance A study of how the Grid framework applies to the cockpit can aid individuals in exploring alternative possibilities of behaviour which may have been unclear. Understanding these concepts can enable a person to sort out unsound or ineffective behavior and replace it with more effective behaviors. The Grid below can be used as a frame of reference to study how each crewmember approaches a job.

High

0 a. о £ w a E 0 о e 8

Low

Figure 4. United CRM Grid. Source: United Airlines training manual, “Introduction to Command/Leadership/Resource Management,” MN-94, 10/95, p. 9.

and others on the Grid, a matrix putting “concern for people” against “concern for performance.” Each of several decision-making elements then get graphed to the Grid: inquiry, advocacy, conflict resolution, and critique. Inquiry, for example, comes out this way in the (1,9) quadrant: “I look for facts, decisions, and beliefs that suggest all is well; I am not inclined to challenge other crewmembers” and in the (9,1) quadrant as “I investigate my own and others’ facts, decisions, and beliefs in depth in order to be on top of any situation and to reassure myself that others are not making mistakes.”24 United’s gloss on Flight 90’s demise is not much different from that of Trollip and Jensen: the first officer made various non-assertive comments “but he never used the term, ‘Abort!’ The Captain failed to respond to the inquiry and advocacy of the First Officer.”25

Not surprisingly, the 747 pilot I quoted before, Robert Buck, registered, in print, a strenuous disagreement. After lampooning the psychologists who were intruding on his cockpit, Buck dismissed the CRM claim that the accident was a failure of assertiveness. “Almost any pilot listening to the tape would say that was not the case but rather that the crew members were trying to analyze what was going on. To further substantiate this is the fact the copilot was well-known to be an assertive individual who would have said loud and clear if he’d thought they should abort.”26 With snow falling, a following plane on their tail, АТС telling them to hurry, and the raging controversy over VI still in the air, Buck was not at all surprised that neither pilot aborted the launch.

Again and again we have within the investigation a localized cause in unstable suspension over a sea of diffuse necessary causes.27 We find agency personalized even where the ability to act lies far outside any individual’s control. And finally, we find a strict and yet unstable commitment to protocol even when, in other circumstances, maintenance of that protocol would be equally condemned. In flight 90 the final condemnation fell squarely on the shoulders of the captain. According to the NTSB, Wheaton’s multiple errors of failing to deice properly, failing to abort, and failing to immediately engage full power doomed him and scores of others.

I now want to turn to a very different accident, one in which the captain’s handling of a crippled airliner left him not condemned but celebrated by the NTSB. As we will see even then, the instabilities of localized cause, protocol, and the human/technological boundary pull the narrative into a singular point in space, time, and action, but always against the contrary

The first contracts for commercial jet transports in the United States were signed in 1955.24 The Boeing 707 and the initial version of the Douglas DC-8 were to be powered by a commercial version of P&W’s J-57, designated the JT3C-6. The intercontinental Boeing 720 and an intercontinental version of the DC-8 were to be powered by a commercial version of P&W’s J-75, designated the JT4A. The first contract for General Dynamics’ Convair 880 was signed in 1956. It was to be powered by a commercial version of GE’s J-79, designated the CJ805-3. All three of these commercial engines were slightly modified versions of their military counterparts, sans afterburner.25 These changes were minor, however. In effect, the military had borne virtually all the cost of developing the high performance engines that powered the first commercial U. S. transports. Two of these engines, the CJ805 and the JT3C, ended up providing the requisite high performance gas generators of the first successful turbofan engines.

Theory – To describe the shape of a wing, engineers distinguish between planform (outline of the wing viewed from above) and airfoil (shape of a fore-and-aft section). Though taking account of planform at supersonic speeds was just beginning in the mid-1940s, methods for calculating two-dimensional (i. e., planar) supersonic flow over airfoils seen as sections of a constant-chord wing of infinite span had been available for some time. The physical and mathematical principles went back to the nineteenth century, and the groundwork for aerodynamic applications had been set down in papers by Ludwig Prandtl, Theodore von Karman, and Adolf Busemann at the landmark Volta conference on “High Speed in Aviation” at Rome in 1935.3 Even at that early stage, results for the supersonic flow around a sharp-nosed airfoil could be obtained with a degree of rigor unusual for the nonlinear equations of gas dynamics.

The method has its basis in the special properties of supersonic flow. In such flows generally, a pressure signal moves past a point at the speed of sound relative to the local flow at that point. As a consequence, and in contrast to the situation in subsonic flow, a signal cannot propagate upstream, and the flow at a point on an airfoil surface cannot be affected by the shape of the airfoil aft of that point. Flow along the surface can therefore be calculated stepwise from the leading to the trailing edge, taking into consideration only the flow ahead of the point in question. With only very weak approximation, the method reduces in practice to sequential application of known nonlinear relationships for two flow situations (fig. 1): (a) discontinuous compression through a shock wave, used to find conditions at the point immediately behind the sharp concave turn at the leading edge, and (b) continuous expansion through a distributed fan-like field, to calculate flow properties along the convex surface of the airfoil. The latter relationship exists by

virtue of the simplicity of planar flow. (The Mach lines in the expansion fan of figure 1 show the limited, rearward-growing region of influence of representative points on the airfoil’s surface.) For more rapid calculation, this nonlinear “shock – expansion” method can be approximated by a linear (or first-order) theory initiated by Jacob Ackeret in 19254 or by a more accurate second-order theory put forward by Busemann in his paper at the Volta conference. By 1946, the various methods had been used to calculate the performance of a variety of airfoils.

The foregoing theories all depend on the assumption of a fictitious inviscid gas, that is, a gas lacking the viscosity present in real gases. They thus omit viscous forces and deal only with pressure forces. Such theories had long been supplemented at subsonic speeds by Prandtl’s boundary-layer theory of 1903, which deals specifically with the ffictionally retarded viscous layer that forms close to a surface in a real gas.5 On this basis, a large body of subsonic experience had been accumulated with both quantitative calculation and qualitative thinking regarding viscous effects. At supersonic speeds, little such experience was available, though indications existed that the presence of shock waves might lead to new kinds of viscous phenomena.

Experiment – Supersonic wind tunnels large enough for experiments in flight aerodynamics came into being in the 1930s. In a paper at the Volta conference, Jacob Ackeret described at length an impressive tunnel recently completed at the Federal Technical University in Zurich, and, in comments following the talk, Mario Gaspari of the University of Rome added details of a near copy then under construction at Guidonia, a short distance from Rome.6 (The most advanced tunnels, however, came into operation in 1939 at the German army’s laboratories at Peenemtlnde following small-scale development at the Technical University of Aachen. These tunnels were used in the design of ballistic missiles. Their existence did not become known in the United States until the end of World War II.)7 It was in the Guidonia tunnel that Antonio Ferri conducted in the late 1930s the first extensive experiments on airfoils at supersonic speeds. These tests, made on a constant – section model spanning the rectangular test section of the tunnel and thus simulating infinite span, supplied a wealth of pressure-distribution and other data on an assortment of airfoil shapes. Except for a few overall-force tests in the late 1920s and during World War II in small tunnels at the National Physical Laboratory in England,8 Ferri’s results provided the only experimental assessment of airfoil theory available at the time of the wing studies to be discussed here.

Comparison – Ferri’s findings, which were to prove useful for our Ames work, can be characterized by a figure reproduced from the latter work (fig. 2). This shows the theoretical and experimental distribution of pressure along the surface of an uncambered 10-percent-thick biconvex airfoil at a free-stream Mach number M0 of 2.13 and an angle of attack a of 10 degrees. (The Mach number M at a point in a flow is the ratio of the speed of flow to the speed of sound, both at that point.) The vertical scale in figure 2 is a dimensionless measure of the difference between the surface pressure p and the free-stream pressure p°. As is customary for airfoil work, negative values are plotted upward, positive downward, so that the area

between the upper – and lower-surface values can be seen as a close measure of the overall lift. The experimental points and the shock-expansion curve are taken from Ferri’s report; linear theory was added as part of the research to be described here.

As can be seen, Ferri’s measurements for the biconvex profile showed near agreement with the shock-expansion theory over most of the airfoil, a typical finding. The higher-than-theoretical pressures over the rear 40 percent of the upper surface he attributed to interaction between the viscous boundary later and the shock wave at the airfoil’s trailing edge. The retarded air in the boundary layer apparently lacked the kinetic energy necessary to negotiate the pressure rise through the shock wave. As revealed by optical studies, the resulting readjustment of the flow found the boundary layer separating from the surface ahead of the trailing edge, with a shock wave forming a short distance above the surface at the location of the separation and a more or less constant surface pressure from there to the trailing edge; a second shock wave formed outside the separated region at about the latter location. Because of the unpredictedly high pressures on the upper surface in the separated region, measured overall lift on the airfoil was less than calculated from the theory. Fern’s results thus brought to light one of the new viscous phenomena characteristic of supersonic flow. They also illustrate how theory and experiment are frequently used together to understand phenomena that are (a) ruled out of the theory by the assumptions that make it mathematically feasible but (b) would be difficult to comprehend without the theory for comparison. The relevance of the linear theory, which Ferri did not concern himself with, will become apparent later.