by Daniel Bell
I have argued that the university increasingly becomes the primary institution of the post-industrial society.90 In the past twenty years, the university—and by “the university” I have in mind the elite group—has taken on a vast number of enlarged functions: in basic research, as a service institution, and in the expansion of general education. In one sense, none of the specific functions is new, since the university, when it first undertook the organization of graduate schools, going back to Johns Hopkins and Chicago, had these functions in mind. What is new is the vast change of scale. The majority of research scientists in basic research today are in the universities; the university serves as the source for the specialized intellectual personnel needed in government and public organizations; even the majority of critics and writers today are employed in the university. The university has become the center of establishment culture. The unrest of the students in the mid-1960s was itself a significant sign of protest against the neglect of traditional teaching functions and the inattention to the student. But the singular fact is that, lacking any organized academy, system, the government has forced on the university, willy nilly, a huge array of tasks that in other countries are performed outside the university system. It is not the traditional dependence of the educational system on the polity which is the important dimension of the postwar society but the “scientific-administrative complex” which represents an intermingling of government, science and the university unprecedented in American history. While it is often mentioned that in his “farewell speech” President Eisenhower warned against the military-industrial complex, it is seldom recalled that in the same speech President Eisenhower balanced his sentence with an equal warning against the scientific-administrative complex which he felt also represented an undue concentration of influence.
If, as is projected to the end of the century, we may see a doubling of student enrollments in higher education, there is a very significant question whether the existing concentration of elite universities will continue. Much depends upon the source of the student body. If a significant proportion is recruited from the children of working-class parents, it is likely that the greater number will go on to junior colleges. The junior colleges have been the fastest-growing segment of the American education scene. In 1930, there were 217 junior colleges; by 1950 the number had jumped to 528; by 1968 there were 802 junior colleges in the country. The elite private schools, universities, and colleges have begun to limit enrollment, in contrast to the major state universities, such as Wisconsin, Ohio, Minnesota, Michigan, and the California schools, which have expanded enormously. As Jencks and Riesman have noted: “The private sector’s share of the market, which had hovered around 50 percent from 1910 to 1950 started falling about 1 percent annually. It was 36 percent in 1964 and is expected to be about 20 percent in 1980. Limiting enrollment had two consequences. One, it raised the ability of the average student, making private colleges relatively more attractive to both students and faculty, and probably indirectly raising the cash value of their degrees. Two, it meant that the philanthropic income did not have to be spread so thin.” 91
Over the years, the number of elite schools has remained comparatively stable (though there have been changes of standing within the group). Whether this will continue is an open question.
Despite the enormous sums spent by the United States government on research (for details, see the next section), there is no central science or research budget in the government, no set of priorities or objectives, no evaluation, no long-range planning as to fields of necessary interest or kinds of manpower to be encouraged. Beginning with the Manhattan District, which produced the atomic bomb, American research policy has been overwhelmingly “mission-oriented,” and each sector of the government—defense, health, atomic energy, space—determines its own missions, the monies being subject to some review by the Budget Bureau and the allocation by Congress. Because of this mission-orientation, there is no system in which existing laboratories or resources belonging to one agency or department are able to put their resources, manpower, and facilities at the service of another. As fresh needs, urgencies, and priorities develop, new research facilities, organizations, and laboratories, and new arrangements with universities are created to meet these fresh tasks. Because needs were often urgently defined, and capacities were unavailable within government, a whole host of “federal contract” devices, with newly established non-profit corporations and universities, was designed in which these tasks were performed outside of government. So scattered and dispersed are the institutional structures of science and research activities of the government that there is no single description extant of its range and structure!
Within the Executive Office of the President, there was the President’s Special Assistant for Science and Technology, created in 1957. He served as chairman of the Office of Science and Technology (OST, created in 1962), the President’s Science Advisory Committee (PSAC, 1957), the Federal Council for Science and Technology (FCST, 1959) and as a member of the Defense Science Board of the Department of Defense.
The President’s science advisor is supposed to have an overall view of federal science policy, but his position is weakened by the fact that almost 90 percent of the expenditures for research and development are expended by four agencies—defense, atomic energy, space, and health—and the science advisor has little say in their activities. The President’s Science Advisory Committee is a government agency whose membership is drawn from outside the government. It is a policy advisory body charged with defining necessary new areas of science expenditures, and assessing the balance of science resources between science and technology and military and non-military uses. But on actual political issues, it has had little influence.
In the Nixon administration, the role of the science advisor was reduced considerably and in January 1973, the Office of Science and Technology was abolished; some of its functions were distributed among other agencies, principally the National Science Foundation. What it meant concretely was that the scientific community no longer had direct access to the President as it had from 1945 to 1968.
Because research budgets are primarily in the hands of the different federal agencies, a multifarious system has developed which varies from agency to agency. The National Aeronautics and Space Agency (NASA) built a large in-house technical capacity, but much of its development work was contracted with private industry. The Atomic Energy Commission (AEC) has created a large number of national laboratories, but in almost all instances these are managed, under contract, by universities (e.g. the Argonne laboratory at Chicago by the University of Chicago), a consortium of universities (e.g. the Brookhaven laboratory on Long Island) or a private corporation (Oak Ridge, managed by Union Carbide, or Sandia by Western Electric). The Defense Department has a wide variety of devices. Applied research and development may be evaluated by nonprofit corporations, such as Rand, or the Institute of Defense Analysis; exploratory research may be handled on contract with universities, such as the Lincoln Lab at MIT; design work may be handled by nonprofit corporations which had been created by universities, such as MITRE from MIT or the Riverside Institute from Columbia; development work would be handled by nonprofit corporations such as the Aerospace Corporation, and production by major corporations such as Lockheed, Boeing, etc. In the health field, there has been a move toward the setting up of government institutes, and the National Institutes of Health, created in 1948, today comprise nine institutes. The National Institutes of Health are responsible for nearly 40 percent of the total American expenditure on medical research. From the start, NIH was empowered to make research grants as well as to operate its own research facilities. At the start, these activities were in equal balance. Since then, the weight of activities has swung largely to research grants; and in 1966, about $912 million was disbursed in contracts and $218 million for in-house operations. (As a measure of the expansion of these activities, in 1950 some $30 million was spent in research and $15 million in dir
ect operations.)
In general, the institutional structure of U.S. science policy until now has been marked by two features: where special tasks are defined, particularly in new fields, applied research and development, new institutional groupings and forms have been created ad hoc to meet these missions; in pure and basic research, money has been given, on a project basis, to individuals who have been able to convince juries or research panels of the worthiness of the project or their competence as researchers.
This double feature of mission-orientation and project grants has had the unique quality of encouraging a high degree of success, by concentrating on the specific mission and mobilizing large resources for the tasks, and by stimulating a high research productivity by individuals who can prove themselves very quickly (compared to the European pattern, where a research man may be “indentured” for a long time to a specific professor). The drawbacks are equally obvious: there is a loss of sustained institution-building, either as an in-house capacity of government, or even in a university, since in most cases research facilities are provided largely for individuals or small teams, not for the institution. (The university, Clark Kerr has remarked wryly, has as often as not simply been a hotel.) Nor is there the possibility for sustained, long-run research since the project system tends to emphasize specific and identifiable bits of research that can be completed in two or three years.
In the larger, political context, the lack of a unified science policy, or a major academy or ministerial system, has meant that the “technocratic potential” inherent in the growing influence of science and the nature of technical decision-making is minimized in the American system. Science itself has simply become a constituency, but with no inherent unity other than some major professional associations and the political role of older clique groups who had played influential roles during World War II and shortly after. As a constituency, it is one more claimant on the national resources, like industry, labor, farmers, or the poor, although much of its “business” is done with the executive agencies, rather than with Congress. But power, in science policy, has rested with the political and bureaucratic interests of the major agencies—Defense, Atomic Energy, and Space—rather than with the scientific community, or even an overall political policy body for science.
THE ALLOCATION OF RESOURCES
By common agreement, the “financial” measure of the growth of science and technology has become the expenditures on research and development (R & D). Efforts have been made to relate the expenditures of R & D to economic growth, to scientific productivity, to the acceleration of the pace of invention, to the shortening of time between invention and production, and the like. There are analytical problems in each of these alleged relationships. What we can take as the simple indicator, however, is the commitment of a country to its scientific and technological potential by the expenditures on R & D, and, to a secondary extent, on education.
The United States, by devoting 3 percent of the GNP to research and development, in the words of the OECD report on science in the United States, became “a symbol for other countries which now regard this as a target to be reached.”92 From the end of the war and for the next two decades R&D expenditures in America multiplied by 15 times, and the total expenditure on education by six, whereas GNP itself has only tripled. In 1965, the United States was spending more than 9 percent of the total GNP on R & D and education.93
While international comparisons in this area are quite risky, a comparison between the American effort and those of Western Europe, Canada, and Japan reveals a very large gap, indeed. As a percentage of national product, R&D expenditure amounted to 2.3 percent in the United Kingdom, which is the country nearest to the magic 3 percent mark, and about 1.5 percent for the other large industrialized nations. In comparing the number of researchers with population, the United States had about four times as many as Germany, France, Belgium, or Canada, and more than twice as many as the United Kingdom or Japan (Table 3-22).
What is striking about the pattern of R & D expenditures is that the federal government supplied most of the funds while the work was performed principally by industry, universities, and the nonprofit organizations. Without the lead of the federal government, there probably would have been little expansion in R & D in the United States. Federal expenditures on R & D between 1940 and 1964 grew at the average annual rate of 24.9 percent. In 1965, a total of $20.5 billion was spent on R & D, of which the federal government financed 64 percent of the total; industry contributed 32 percent, universities spent 3.1 percent, and nonprofit institutions 1 percent.94 Yet only 15 percent of the work was done by the federal government; 70 percent was performed by industry, 12 percent by universities (including 3 percent at federal contract research centers) and 3 percent by nonprofit institutions. For fundamental research, the federal government provided about 64 percent of the funds, but the universities were the principal performers. Of almost $3 billion spent for fundamental research in 1965, 58 percent was used by universities, 21 percent by industry, and 7 percent by non-profit institutions (Table 3-23).
TABLE 3-22
Comparison of the R & D Effort of the United States with That
of Other Western States and Japan
SOURCE: OECD, Reviews of National Science Policy: United States, p. 32.
a Full-time equivalent.
bEstimated according to OECD standards and not according to those of the NSF.
TABLE 3-23
Expenditures of Fundamental Research, 1965
SOURCE: OECD, Reviews of National Science Policy: United States,p.34.
a Federal contract research centers.
If we think of R & D not just as contributing to economic growth, or being the engine of science and technology, but in political terms, then a very different picture of the American effort in that period emerges. The largest proportion of total R&D expenditures was spent for defense purposes. These direct expenditures (Department of Defense and certain Atomic Energy Commission programs) have fluctuated around 50 percent from 1953 to 1961 but’ according to the NSF, decreased to 32 percent in 1965. But much of this offset in proportions was due to relatively increased spending for space, rather than domestic needs, and if, with the OECD report on science policy in the United States, we consider “as a single category” all expenditures connected with external challenge, it appears, from Table 3-24, that this political reason dictated more than 80 percent of all federal expenditures and more than 60 percent of the total R & D expenditure. (Since a large proportion of the privately financed industrial R&D was also connected with defense, the proportion of the total R&D related to the political response to external challenge is undoubtedly higher than 60 percent.)
Given this pattern, the considerable lead of the United States over other countries in R & D assumes a different proportion. For the United Kingdom devoted about 33 percent of its R & D expenditure to military research and defense (including military atomic research); Germany, 17 percent to atomic, space, and military research; Italy, 21 percent; Canada, 23 percent; japan, 3 percent; and France, 45 percent (of whjch 22 percent is devoted to atomic research). To this extent it is clear that the driving force of the American government in financing R & D is primarily related to political objectives, as is, in fact, the proportions spent by the state in any country.
TABLE 3-24
Research and Development Linked with External Challenge, 1954-1967
SOURCE: OECD, Reviews of National Science Policy: United States, p. 38.
NOTE: The figures in column 2 are obtained by adding the expenditure of the Departments of Defense, NASA and about 50 percent of the expenditure of the Atomic Energy Commission which can, in the view of most experts, be regarded as “defense-oriented.”
What of the future? Research and development expenditures rose at a compound annual rate of 12.1 percent, from $5.2 billion in 1953 to about $20.5 billion in 1965. Over the same period, GNP rose by a compounded rate of 5.3 percent. But the average growth rate in R&D, m
easured from the survey base year of 1953, has been falling since it peaked at 17.6 percent in 1953-1956. In the 1964-1965 period, while GNP moved upward to 7.8 percent, R&D slowed down to a 6.7 percent increase, the first period in which the percentage increase in research and development was less than that for the economy as a whole.
Research and development manpower, the most critical component of research, grew faster than the country’s civilian labor force during the decade of 1954-1965, advancing from 237,000 to 504,000 persons, an annual rate of 7.1 percent compared with the 1.5 percent for the labor force as a whole. As a percentage of the labor force, the number of R & D scientists moved from 0.37 to 0.68 percent in that same period. Industry, in 1965, as in the past, was the largest employer of R & D scientists and engineers, reporting 351,200 in full-time equivalent numbers or around 70 percent of the 503,600 total. The federal government employed 69,000 professional scientific and engineering personnel or about 14 percent of the total. The universities and colleges employed 66,000 R&D scientists and engineers for 13 percent of the total, 54,900 of them in universities and colleges proper. Three percent of the R&D scientists worked in the other non-profit sector. This percentage distribution was close to the pattern of 1958 (Table 3-25).
Research and development data are classified by the National Science Foundation into funds allocated for development, applied research, and basic research. (Federal obligations for total research are shown in Table 3-26.) Development is defined as the design and testing of specific prototypes and processes to meet a specific functional (e.g. defense) or economic requirement. Applied research is defined as the first pilot steps in translating existing knowledge into applications. And basic research is defined “as primarily motivated by the desire to pursue knowledge for its own sake ... free from the need to meet immediate objectives and ... undertake to increase the understanding of natural laws.” Whether these distinctions, particularly between basic and applied research, are meaningful is an important theoretical question that needs to be pursued.95 Inasmuch as these distinctions, however, are used by the National Science Foundation, one can follow certain trend lines and establish future baselines from their data. While the proportion of money spent for development and applied and basic research has remained relatively constant—about two-thirds of the money has gone for development and one-third for research—the balance between money spent for applied and basic research has changed somewhat. In 1965, basic research outlays amounted to 14 percent of total R&D and applied research about 22 percent; in the period between 1953 and 1958, the money spent for basic research was about 9 percent of the total.
