In examining the contributions that the National Science Foundation has made to the evolution of major engineering innovations, it is essential to understand the several contexts within which such contributions originated. The relevant contexts that shaped the nature and extent of NSF contributions include the political and organizational environments in which engineering programs were initiated and operated within the Foundation; the financial support provided to engineering relative to other fields within the Foundation and relative to the support provided to engineering by other federal agencies; and the managerial strategies employed by NSF program directors and managers in planning, setting priorities, and selecting engineering projects to support. The purpose of this overview is to set the stage for our subsequent analyses of specific cases of engineering innovation so that NSF's potential role can be anticipated, and its actual contribution fully identified and accurately characterized. Thus, we review the engineering field's effort to define its proper niche within the Foundation, establish a stable organizational base, and develop managerial strategies that were effective in achieving the Foundation's broader objectives while accounting for the engineering field's unique place in research, education, and technological innovation.
 

II. Defining Engineering's Role in Research and in NSF

Between the 1950s and the 1970s, engineering research in U.S. academic institutions underwent a fundamental transformation. NSF's Third Annual Report (FY 1953) observed:

NSF's first budget for research, in 1951, was for $1.1 million, of which $42,000 (3.8%) was for engineering (Belanger, 1-21) When the first grants for research were awarded in June 1952, 3 of the 96 grants were in the engineering sciences. During most of the 1950s, engineering grants were investigator initiated and, because they were intended to increase understanding of basic phenomena, used basic science approaches (Belanger, 1-28). The Sixth Annual Report (FY 1956) was the first to describe, in a short paragraph, the activities of the engineering sciences program within the Division of Mathematical, Physical, and Engineering Sciences. By FY 1957, on the eve of Sputnik, 103 grantees received $1.4 million for engineering research. This represented 8.5% of NSF's research budget of $40 million, a proportion that has remained relatively constant over the years (Belanger, 1-38). The shock of Sputnik drove NSF 's budget from $50 million to $133 million in one year; engineering grew proportionately from $1.5 million to $4.2 million (Belanger 1-40).

As perhaps befitting NSF's characterization of its role in supporting basic engineering research during these initial years, the projects that the Foundation supported typically did not match traditional engineering disciplines such as electrical or civil engineering. Instead, projects could better be described as falling within fluid mechanics or thermodynamics, and frequently were interdisciplinary. Belanger observes that "Use of such 'science' terms, of course, would have political value in NSF's basic science milieu" (Belanger, 2-5). In 1961, the Director of the Division of Mathematical, Physical, and Engineering Sciences defined basic research in engineering as "any scientific activity that strengthen[ed] engineering practice" (Belanger, 2-9).

In 1963 Leland Haworth became Director of NSF, and this change of leadership occasioned pleas for greater recognition for engineering. Eric Walker, the sole engineer on the National Science Board and who had pushed for greater status for engineering, became chair of the Board in 1964. During FY 1964 NSF provided about $13 million for basic research in engineering. The following year, the forceful Assistant Secretary of Commerce for Science and Technology, Herbert Hollomon, pressed NSF to support research in areas of greater relevance to societal needs and problems. In FY 1965, NSF supported 541 engineering grants totaling $15.2 million (Belanger, 2-26-27). The following description, from NSF's tenth annual report (1961) typifies NSF's view of the engineering sciences during the first decade of the Foundation's existence:

NSF's 1972 Annual Report stated that "the Engineering Division supports research programs, primarily at universities, that are aimed at the generation of engineering information, development of methodologies for analyses and design, assessment of the effects of engineering projects, and the translation of basic technology into a form which can be used in practice" (Annual Report FY 1972: 26). The 1975 Annual Report states that "The engineering program especially seeks opportunities to hasten technological innovation by shortening development time and reducing the costs to the technological entrepreneur. By selective funding in areas that show promise . . . , NSF hastens the development of the technology needed to evaluate the potential of an area." This description of focus was followed, on the same page, by this statement: "While industrial technology remains an important field, some of the work [formerly supported by NSF] did not contain the mixture of basic research that the program should support, while other aspects of industrial technology could be dealt with in other existing programs" (Annual Report FY 1975: 32). To the extent that these authoritative statements reflected different or changing views of what should be supported under the rubric of "engineering science," potential grantees faced a difficult problem of identifying just the right niche.

In 1973 a task force under John Whinnery, Professor of Electrical Engineering and Computer Sciences at Berkeley, examined the role of applied research at NSF. Their report argued that in its support of engineering and applied research the Foundation should focus on "highly advanced, long term, risky projects." Notably, the task force made a distinction between applied research and research applications. It agreed on three principles: (1) that basic research and NSF's role in it were unique and absolutely essential, (2) that basic research, applied research (to meet longer range, generalizable applications objectives), and research applications (user-focused, short-term activities) were interdependent and mutually supportive; (3) that a pluralistic approach to problem solving was commendable if it was first rate and properly coordinated (Belanger, 5-9).

In 1978, NSF Engineering Director Henry Bourne eloquently described the apparent paradox in the term "basic engineering research," which was the niche NSF had defined for itself for supporting engineering research:

• Study of principles and phenomena making possible design of physical devices, systems, and processes
• Study of these principles in the design of mathematical models of more abstract systems
• Ranges from exploring physical phenomena to general design methodologies
• Impacts of design of large classes of devices, systems, and processes
• Long-range effects although not to exclusion of short-range effects (NSF Program Report, Division of Engineering, Vol. 2, No. 1, May 1978: 1-2)

III. Reorganization Blues: the Tortured Path to Directorate Status

When the National Science Foundation commenced operation in 1951, engineering was placed in the Division of Mathematical, Physical, and Engineering Sciences (MPE). By 1953, engineering comprised about 10% of total NSF obligations for research, where it has remained through most of the Foundation's history. In 1958, partially in response to the perception in the engineering community that NSF support for engineering was too low and received too little attention, an ad hoc committee of the American Society for Engineering Education proposed that a Division of Engineering Science be created (Belanger, 1-37).

In 1961 the MPE Division was organized into seven sections, previously called program offices, one of which was Engineering. As noted above, the appointment of Leland Haworth as the second Director of NSF in 1963 increased the level of lobbying for greater recognition for engineering. Eric Walker's elevation to Chair of the National Science Board in 1964 was certainly related to the decision to create an Engineering Division in that year (Belanger, 2-21). Also in 1964, NSF Graduate Traineeships were initiated, and all $6 million went to engineering. A new Engineering Division Committee composed of academic and corporate leaders, influenced by the Engineers Joint Council report The Nation's Engineering Research Needs 1965-1985, identified areas ripe for research because of their economic significance (Belanger, 2-22). The following year the Engineering Division Committee was renamed the Advisory Committee for Engineering. The Committee continued to study program direction and resource allocation; it was the target of the pressure from Herbert Hollomon, mentioned earlier, to recommend greater support for research that had more immediate relevance to societal problems.

The continuing pressure for "relevance" for NSF research and to increase the payoff from the nation's investment in basic research culminated in the Daddario amendments to NSF's authorization, explicitly authorizing NSF to support applied research. IRRPOS and RANN were the Foundation's response. While these highly visible organizational entities represented NSF's answer to the call for relevance, engineering as a coherent discipline suffered in the sense that its activities were spread among several organizational units that themselves were subject to internal shuffling.

In 1975 the Foundation's basic structure was changed to include three Directorates: Mathematical, Physical, and Engineering Sciences; Atmospheric, Earth, and Oceanographic Sciences, and Biological and Behavioral Sciences. In 1977 RANN was abolished, and in 1978 a Directorate for Applied Science and Research Applications was created; but the Division of Engineering remained in the MPE Directorate. In 1979 a Directorate for Engineering and Applied Sciences was created, an amalgamation of the ASRA Directorate and the MPE Division of Engineering. Finally, partially in response to George Brown's call for a National Technology Foundation, the Directorate for Engineering was created in 1981. It had four divisions:

• Electrical Engineering
• Computer and Systems Engineering
• Mechanical Engineering and Applied Mechanics
• Problem Analysis Group.

• Mechanics, Biochemical, and Thermal Engineering
• Mechanics, Structures, and Materials Engineering
• Electrical, Communications, and Systems Engineering
• Science Base Development in Design, Manufacturing, and Computer Engineering
• Fundamental Research in Emerging and Critical Engineering Systems
• Office of Cross-Disciplinary Research (oversight of the Engineering Research Centers Program).

• Chemical and Thermal Systems
• Electrical and Communications Systems
• Mechanical and Structural Systems
• Design and Manufacturing Systems
• Biological and Critical Systems
• Engineering Centers
• Office of Educational Infrastructure.

IV. NSF Support for Basic Engineering Research in the Federal R&D Context

Since the Foundation's beginnings, the lead agencies providing federal support to U.S. colleges and universities, its primary constituency, have been the National Institutes of Health (NIH) in medical research, well out in front of other contenders, followed by the Department of Defense (DOD) and NSF, essentially running neck-and-neck throughout the forty years since 1953. During this period, NIH has provided approximately half of total federal support for science and engineering, with NSF and DOD combined providing another 25-30%. Considering only basic research support to colleges and universities, NSF's constituency, the Foundation's role looms considerably larger. Although NIH again dominates among federal agencies, NSF has consistently been the second largest provider of basic science funds. Ahead of DOD for most of its history, in 1972 the Foundation began providing more than twice the support for basic research in colleges and universities provided by DOD.

Considering only federal support of basic engineering research at colleges and universities, NSF and DOD have consistently been the largest providers, with other agencies well behind. In the early 1980s, DOD edged ahead of NSF as the primary federal source of funds. Although time series data on the relative support for different fields of engineering by federal agencies are unavailable prior to 1975, the data show striking differences in relative support among fields: NSF has consistently been by far the dominant provider of support in chemical and civil engineering (the latter dominated by earthquake research); in electrical and mechanical engineering, NSF and DOD have together provided the great bulk of federal support, with DOD consistently providing slightly more than NSF over the years. In metallurgy and materials engineering the picture is somewhat more complex. Until 1985, NSF and DOD were the largest providers of federal support (for basic engineering research at colleges and universities), with the Department of Energy (DOE) a close third. In 1985 NSF support dropped precipitously and has not recovered, while DOD support increased roughly proportionately.
 

V. Managerial Strategies Employed by NSF for Support of Engineering Research

The standard set of functions carried out by managers of research programs includes planning, priority setting/resource allocation, project selection, monitoring, and evaluation. In the National Science Foundation, project selection procedures have remained constant, based in peer review, now slightly modified and labeled merit review; monitoring and evaluation, at least at the project and program level, have for the most part been minimal. However, managerial strategies for planning and priority setting have changed over the Foundation's history, both at the level of the entire organization and with respect to the support of engineering research. We focus here on the latter.

During the Foundation's first decade of operations, engineering program managers responded to unsolicited proposals developed by individual researchers. Topics of proposals accepted for review were shaped very broadly by the niche NSF sought to establish as "basic engineering science," but the specific subjects of proposals were determined by principal investigators' views about what research problems were worth investigating. The 1952 Annual Report described the broad areas in which NSF would support engineering research:

The first evidence of a more active strategy appeared in 1962, when the Engineering Section of MPE began a fire research program. This was followed in 1965 by an earthquake engineering research program. By 1970 the list of what might be called "problem-oriented" research on topics of public concern included biomedical engineering and enzyme engineering (Director's Program Review of Engineering, Feb. 24, 1970, pp. 10; 24-43). Anticipating the forthcoming change in the NSF authorizing legislation that would generate increased emphasis on applied research, the FY 1967 Annual Report observed:

• potential for impact
• contribution to U.S. leadership in technology
• ripeness of scientific background
• relation to missions of other agencies (p. 23).

By the early 1970s, evidence of another managerial strategy began to appear in official documents: workshops involving both academic and university researchers. To be sure, workshops involving academic researchers had been a part of the Foundation's activities since its inception. NSF Annual Reports listed, for the first decade or so, all the conferences and workshops the Foundation had fully or partially supported. The purpose of these conferences was to facilitate communication among academic researchers working in the same or related fields. In some cases a product of the workshop was a research agenda to which academic researchers could respond. What was different in the 1970s was involvement of industry. Thus the research agendas that resulted from some of these workshops were not the product of academic interests alone, but reflected industry's concerns as well. For example, NSF held a workshop on glass processing in 1971. The impetus was the economic significance of the industry ($5 billion in annual sales), the continual threat of foreign competition, and the very low level of federal investment in glass research (Director's Program Review of Engineering, 6/27/72: 36). A half-million dollar program of research was proposed, which included provisions for semi-annual meetings of grantees and industrial representatives to discuss the progress of the research. Similar workshops were conducted in the fields of optical computing and in the electromagnetics of continuous media. By 1978, the Engineering Division could note that, to promote industry/university interaction in research, it worked primarily through workshops numbering between 28 and 41 annually for the period 1974-77 (Director's Program Review of Engineering, May 1978: 55).

The Division went even further: it encouraged the initiation and development of cooperative research programs not only among universities, but also between universities and industry (Annual Report, FY 1973: 28). Other evidence of interest in incorporating industrial research problems into NSF programs was manifested in program director visits to academic and industrial labs to discuss program priorities. These were limited by time and budgetary considerations, but NSF clearly regarded them as having significant value: "It is realistic to say now that this initial interaction [with Bell Labs] made a critical contribution to the current strong foundation of university-industry cooperation that is a vital part of this [optical communications] program" (Director's Program Review of Engineering, 1/28/75: 10). Still, Engineering's primary managerial strategy continued to be responding to unsolicited proposals; the difference was that those who generated the proposals were aware of specific priorities being favored in the Foundation, which in turn were influenced by parties other than those who would perform the research.

By the late 1970s, an additional strategy was emerging: the stimulation of new, cooperative organizational forms to promote industry-relevant research in universities. Encouraged by the positive experience of the Industry/University Cooperative Research Centers, dating from the early 1970s, NSF expanded the idea of university-based research consortia focused on industry-defined research problems by launching the Engineering Research Centers program in 1984. To be eligible for support, proposals for ERCs had to demonstrate that the research focus held promise of contributing to the competitiveness of U.S. industry and that a significant number of industrial firms were willing to support the activity financially.

By the late 1980s, therefore, engineering research at NSF employed a wide range of managerial strategies simultaneously. The following list is ordered roughly in the chronological order in which each appeared. As it turns out, the list also is ranked roughly in order from relatively reactive to relatively active managerial strategies. It is important to remember that the dominant strategy remains reactive; but increasingly active strategies were employed beginning in the mid-1960s, so that now a significant minority of engineering research is supported by an (unknown) mix of the more active strategies.

Chronology of Management Strategies: NSF Support of Engineering Research

Explication of these management strategies raises a number of questions:

• How much problem-oriented research should be supported?
• What is the proper balance between fundamental and problem-oriented research?
• What are the most effective mechanisms for introducing industrial research priorities into NSF's research priorities?
• What are the relative payoffs from industry-driven research versus research driven by the desire to push the state-of-the-art of fundamental knowledge?

VI. Summary and Conclusions

Through the 1950s and 60s, engineering in NSF sought recognition internally while attempting to define a narrow range of research to support that was recognizably engineering rather than "applied science," yet was sufficiently "basic" to match the intent of NSF's charter. The term "basic engineering science" meant "any scientific activity that strengthened engineering practice;" work of "a true scientific nature and not routine engineering practice;" research that "provides the essential information and methods with which existing problems may be solved and new opportunities . . . recognized." Engineering program directors quickly concluded that supportable projects would not match traditional engineering disciplines but instead would be categorized according to underlying fundamental phenomena such as thermodynamics or fluid mechanics. Supportable projects would also tend to be interdisciplinary, frequently involving two or more engineering disciplines as well as science.

As this delicate balance became codified, understood, and accepted by the academic engineering community, the system was shaken in the late 1960s and throughout the 70s by the new mandate in NSF's authorizing legislation. This mandate introduced new terms, "applied research" and "research applications," that seemed to imply engineering, and new organizational forms designed to respond to it. NSF's response to the mandate also separated its support of engineering into multiple, changing loci. Even before the Daddario amendments, however, engineering at NSF was beginning to respond to larger societal problems by creating programs in problem-oriented areas such as earthquake research. During the organizationally tumultuous 1970s, engineering research at the Foundation was supported from multiple and shifting locations, no doubt making strategic planning and priority setting a difficult task. It was not until RANN was abolished and the Engineering Directorate created in 1981 that engineering became focused and stabilized organizationally, achieving equal status with the sciences. Interestingly, the names of the divisions of the new Directorate closely matched those of the traditional engineering disciplines. The current divisional labels replace the term "engineering" with "systems," but otherwise largely parallel traditional disciplines.

From a first, small effort in 1962 to respond to larger social problems, engineering programs became increasingly proactive in the ways they set research priorities. In the early days, "proposal pressure" defined areas for emphasis. Then engineering program managers began to hold workshops, at first including only grantees and other academics but later including industry representatives. These workshops' purposes were to build stronger research communities, identify promising areas for research, and eventually bring industry's priorities into the priority-setting equation. By the time the Engineering Directorate was created, approximately 30% of engineering research was problem-oriented, focusing on topics identified through numerous channels such as these workshops and outside advisory groups. Another significant management strategy, introduced in the 1970s but not becoming prominent until the 1980s, was to encourage industry-university research consortia to form around subjects of interest to industry. Centers--IUCRCs, ERCs, MRLs--now constitute a substantial element in the Foundation's portfolio of strategies for managing research programs. The relative effectiveness of these strategies continues to be a subject of importance for NSF policymakers and program directors.

For over a generation NSF and DOD have been the major sources of support for academic research in the fields of fundamental science and engineering other than biomedicine. Thus, the nation's current generation of Ph.D. scientists and engineers and the contributions they have made to knowledge and technological innovation can be attributed in no small degree to support from these two agencies of the federal government. In some fundamental engineering fields such as chemical and civil engineering NSF's relative contribution dominates. Tracing recent engineering innovations to their roots in knowledge and technology, and to the scientists and engineers who produced them, reveals the prominent influence of these two agencies and the mechanisms through which that influence is manifested.
 

VII. References
 

• Belanger, Diane. Engineering: An American Tradition. unpublished manuscript, 1997.
• U.S. National Science Foundation, Annual Reports, FY 1953-90.
• U.S. National Science Foundation, Director's Program Reviews, various years.