In this post, I will quote from ‘Paths of Fire: An Anthropologist’s Inquiry into Western Technology’ (1996) by Robert McCormick Adams. “The proportion of doctorates in science and engineering fields earned by foreign students has also increased rapidly. No doubt this once again reflects the complex interaction of many factors. With the great expansion of U.S. research universities, they have become a mecca for aspiring professionals from all over the world. In fields like engineering, virtually half of the new doctorates go to foreign students. It should be recalled, of course, that only approximately half of these graduates reportedly remain in the United States and become part of its trained labor force. In some ways even more relevant for technology than the numbers of doctorates are the numbers of masters degrees in science and engineering. Recipients of doctoral degrees may function exclusively in teaching and basic research programs, while the great majority of masters degree recipients seek employment in industry. To hazard a general evaluation of the differences, computer science has grown strikingly rapidly, but to a much greater extent as an industrial and technological field than as an academic one. In the physical and biological sciences the doubling in the number of doctorates while the numbers of masters recipients have stagnated either suggests the reverse or indicates that there is little industrial demand for individuals with less training than the doctorate represents. Engineering, its practical role unquestioned, appears to have strengthened itself academically. As described in chapters 5 and 6, modest federal support for technologically relevant R & D may in a sense be said to have originated almost at the outset of the Republic with early patent legislation. More actively, the nineteenth century saw gradually growing levels of budgetary support for the protection and promotion of commerce like the Coast Survey, and for the encouragement of agricultural research through the establishment of land-grant colleges and research stations. More affirmative and substantial steps were taken soon after the end of the century, including the creation of the National Bureau of Standards, regulatory agencies concerned with threats to public health and well-being, and presently the National Advisory Committee on Aviation. Evident in retrospect, if not so clearly at the time, was the sense that growing national integration required a wider exercise of governmental responsibilities. Actions that were largely argued for and taken on an individual, ad hoc basis can collectively be seen as an opening wedge of commitment to sustain a favorable infrastructure of research as well as regulation in emergent areas of potential risk and economic importance. Although the impact was relatively brief, World War I was a pioneering exercise in the superimposition on this still only loosely articulated structure of new federal agencies directly committed to mobilizing academic scientists as well as engineers in support of military R & D. As we have seen, that theme re-emerged with much greater strength and more lasting effect in World War II. Established in a decision process that began already during the war and was consummated only a few years afterward was the general shape of national priorities for science as well as technology that has persisted ever since. The cluster of emphases is well recognized. Their impacts on technology can perhaps best be identified in the total basic and applied research budgets of the major mission agencies. Setting aside the role of agencies like NIH and NSF with mandated responsibilities for the support of academic research, at the head of the list of priorities is a primary commitment of large-scale support for military R & D under the direct administration of the Department of Defense. For the last decade or more, the Department’s share of total federal R & D has substantially exceeded 50 percent, falling in the range of $21 to $35 billion (constant 1987) dollars annually. The share set aside for technology development has consistently been about one fifth of the total, with $3.6 billion of the 1994 allocation designated specifically for advanced technology development. Less direct, but no less fundamental as at least an initial stimulus, was the importance of technological rivalry with military overtones in the launching of the Apollo Project and the National Aeronautics and Space Administration. With a monopoly on manned launching capability through the space shuttle, NASA missions include placing unclassified scientific instruments as well as military surveillance satellites in orbit. The Apollo mission to the moon was itself, of course, an extraordinary, unequaled achievement that must be seen primarily as a technological rather than scientific triumph. Because it is focused, the high priority given to defense-related R & D, and more especially to the advancement of technology with perceived military applications, deserves further attention. To be sure, it amounts to only one-third or so of industrial R & D in the private sector, and to slightly less than half of that in the high-tech private sector. But the greater part of these larger amounts is devoted to the development of a very broad spectrum of consumer products under what is arguably a different set of cost constraints. If so, military R & D, beyond the acknowledged superiority of a high proportion of advanced American weapons to which it has no doubt contributed, may have had some more questionable side effects. The coherence and specificity of military requirements, in particular, tends to stand in sharp contrast to the heterogeneity of civilian ones. Faced with an “underdetermination” of the technical solutions needed to meet the demands of the market, it is the military ones that prevailingly have had the greater impact. A case can be made for doubting that military R & D has had a preponderantly positive impact on the competitive standing of other U.S. hightech industries in world markets, or on our R & D capabilities more generally. Making this judgment with any accuracy would be very difficult since many politico-military as well as economic considerations are involved, but David Mowery and Nathan Rosenberg have thoughtfully outlined some of at least its economic parameters. Well-financed federal R & D, by increasing the demand for scientists and engineers in certain fields, has raised their rates of remuneration (and of course also, but more slowly, the supply) and hence the costs of private R & D as well. That effect is likely to have been particularly large in increasingly competitive, high-technology fields like microelectronic equipment, instrumentation, and aircraft, and to have led to a displacement of research activity away from fields such as chemicals and petroleum that receive few federal funds. These distortions inevitably involve costs, even if the costs remain largely hidden. More tentatively, Mowery and Rosenberg touch on the issue of differences between federal and private cost constraints that was mentioned earlier: “Were engineers who had worked on programs where small performance improvements were sought almost regardless of cost effective designers of products for civilian markets where cost considerations and sensitivity to nuances of consumer preferences were likely to be far more significant? Have large federally supported “crash” programs shaped the approach and influenced the (perhaps implicit) trade-offs of U.S. engineers and product designers in ways that are dysfunctional for highly competitive consumer markets, such as consumer electronics?” These are complex, subtle questions, not permitting easy, unambiguous answers. They center on the likely performance of individuals under changed conditions, while the only available data are almost certain to be aggregated at the level of firms. But it is worth noting that, according to a recent survey, the defense industrial base has been found to be “substantially ‘dual-use’.” A “vast majority” of defense-involved firms simultaneously meet commercial customers’ requirements with the same equipment and work force, and in so doing are faced with seemingly comparable competitive pressures in both. The doubts of Mowery and Rosenberg on this score are rendered somewhat less plausible, although not entirely dispelled, by these findings. Another of their critical arguments is based on comparisons with our principal industrial competitors. Ratios of civilian R & D to GNP have been substantially higher in Japan and Germany, with relatively much smaller military budgets, than in the United States. To Mowery and Rosenberg, this suggests that “the true opportunity costs to the U.S. economy of high levels of defense R & D have been very high.” Such comparisons are extremely difficult, they concede, but at least on purely economic grounds their case clearly has some substance. Providing some additional support for this position is a fuller account of individual technologies that currently are advancing rapidly and are of critical importance to the United States for their economic significance and numerous applications. Twenty-two technological sectors have been termed “essential to satisfy such national needs as defense, economic competitiveness, public health, and energy independence” by the U .S. National Critical Technologies Panel, and described as representing “the lion’s share of the future growth of the nation’s economy.” As a prescription for federal policy the listing perhaps has deficiencies, some of its details possibly reflecting corporate self-interest or protectionist sentiments in a volatile atmosphere of international competition. But it provides a useful basis for briefly scanning what are today the key areas of technological advance. Five of the twenty-two fall within a broader category concerned with the synthesis and processing of advanced materials, to which the dominance of defense-related considerations seems clear. The diversity and potential capabilities of many of them are most impressive. Polymer and metal matrix composites, reinforced with high-strength fibers or particles, can achieve several times greater strength and greater stiffness than traditional metals or even superalloys, with a 20 to 30 percent weight reduction as an additional advantage. “Intermetallics,” involving many combinations of nickel, cobalt, iron, lithium, and titanium with aluminum, have been found to promise the special advantage of greater strength at high temperatures. Carbon-based and ceramic matrix composites tolerate higher temperatures than any metal alloys, making them especially relevant for turbine engines, rocket nozzles, and space re-entry vehicles. High-strength, reinforced ceramics, in spite of persistent tendencies toward brittleness, are even being developed as the major engine component in missiles, drones, and other short-life applications. High cost is, however, a regular trade-off for high performance. Not only expensive in themselves, the new materials generally require new and expensive processing methods: additional refining for ultrapurity, “near net shape processing” of alloys and ceramics through “hot isostatic pressing”; laser surface hardening; superplastic forming (slow deformation under high temperature and pressure to reproduce exact, intricate shapes); and very rapid cooling and solidification of metals in order to reduce cracking and weakness along grain boundaries. Another broad category involves sophisticated, complex materials that are “sometimes engineered literally an atomic layer at a time through chemical vapor deposition or molecular beam epitaxy, and the integration of electronic and photonic materials to form single circuits. The integrated circuits are an exception, but few of the other examples cited are likely to be of near-term commercial utility. Most of these materials have been developed for specialized uses involving the need to operate at very high temperatures in order to achieve high performance, as well as exceptionally high strength-to-weight ratios and resistance to wear or corrosion under almost any conditions. Cost has been at best a secondary factor in their introduction, and so long as military orders are the source of funding is not likely to become a primary consideration. Hence, a commercialization phase of further development seems more likely to occur in Japan, and in a number of cases is reported to have already begun there. Aeronautics is a somewhat countervailing example, with commercial developments clearly the net beneficiaries of R & D funding provided by NASA as well as the Department of Defense. Cost constraints are obviously different in the private sector, but there are many areas of convergent interest where federal support can assure more rapid progress. These include improved, computer-aided airframe (and especially integrated engine/airframe) design, more powerful engines with better fuel economy, avionics that enhance safety and improve man/machine interfaces, reducing and monitoring metal fatigue and corrosion, as well as many others. Partly because of the military stimulus, the U.S. commercial lead is still substantial. But with major, heavily subsidized European entries it is shrinking. A broader category of “critical technologies” involves advanced manufacturing processes. By definition, initiatives for it lie primarily within the private industrial sector. Included under this heading are diverse approaches to flexible computer-aided design, inspection, and assembly; intelligent processing equipment and robotics; devices for micro and nanofabrication and nanolithography of integrated circuits; thin films and other forms of surface treatment; and systems information and management technologies. Japan is once again a major, in many cases dominant, competitor, devoting, as noted earlier, twice as large a proportion of its industrial R & D to manufacturing processes as the United States does. Meanwhile, leadership in the race for important robotics patents is vigorously contested between Japan, the United States, and Western Europe. As the National Critical Technologies Panel observes, “the culture and management practices of many U.S. companies must change if the United States is to remain strong in today’s manufacturing environment. The information industry comprises seven of the twenty-two “critical technologies,” the largest and perhaps most significant grouping. The whole ensemble may well be undergoing such rapid and far-reaching changes as to justify speaking of a full-fledged Information Revolution, distinct from the Industrial Revolution but at least potentially comparable to the latter in importance. It has widely been observed, for example, that there has been well over a 20 percent annual decline in the price of a given amount of computing power over the last three decades, “a steady rate that eclipses any sustained price decline in recorded history. But impressive as this achievement is, it would misrepresent the advances made in different but complementary sectors of the industry – involving software as well as hardware, and concerned with the assembly, storage, transfer, and processing of unprecedented masses of data – to think of their emergence and consolidation as an internally energized process. “For many years, these breakthrough computers were the carefully nourished offspring of government encouragement.” Software is the most dynamic frontier of the information industry, as the meteoric rise of Microsoft and relative decline of IBM and other corporate giants associated primarily with hardware attests.”
Igor's Blog 