Page images

scientific research directed toward the production of useful materials, devices, or methods, including design and construction of prototypes and demonstration of processes. (ref. 3.) This definition is broad enough to cover many of the categories used by other writers — and we mean it to be broad. Starting from a broad definition, we can avoid those subtle arguments which sprout like toadstools in the literature on science policy. For example: What are the differences among strategic research, product research, process research, and operations research? At what precise point does applied research become development? There would be nothing wrong with precise definitions, if they did not often lead to unproductive arguments about what an institution actually does — hardening of the categories, as it were. There have even been cases where a Federal laboratory would have reported no basic or applied research if it adhered strictly to definitions laid down by the National Science Foundation. * It is better to start with something comprehensive, refining our categories as we proceed.

Compared to the events that Whitehead discussed, the pace of technology development has accelerated immensely. (This is true both in absolute and relative terms, as any reader who owns a pocket calculator or personal computer or who plays video games can attest. Five years ago these items were either unavailable, or available only at prices beyond the reach of ordinary consumers.) In the case of major space and weapons systems, there must be simultaneous advances along a broad front: in electronics, materials, guidance and sensor systems, data processing, and the like. Of necessity, such research draws on many disciplines, since the problems to be solved are extremely complex; in some cases, new specialties combining several disciplines, like astrophysics and biochemistry, are created.

In this setting, the role of basic research becomes problematic. In

* There is something both amusing and depressing about arguments over what a research organization is doing. A study by the Congressional Research Service of the National Bureau of Standards (NBS) notes that different observers saw the Bureau doing different things. In 1970 and 1971, the Bureau responded to a survey by indicating that it was spending between $13 and $15 million annually on basic research. But at a 1971 congressional oversight hearing, the NBS Director “testified that only $3 million was being dedicated to basic research. Later he explained the apparent discrepancy by suggesting that the first figures reflected the judgments of individual scientists about their work rather than the, presumably, more accurate estimates of top management. As he subsequently told the Visiting Committee, 'if that's how the NBS staff view their work from a motivational view, that is fine.'” Congressional Research Service, The National Bureau of Standards: A Review of Its Organization and Operations, 1971-1980. A study prepared for the Subcommittee on Science, Research, and Technology, U.S. House of Representatives, Committee on Science and Technology, 97th Congress, 1st Session (May 1981), p. 110.

basic scientific research the purpose is to find out why things happen as they do in nature. It depends on experiment and theory to devise a structure for some finite element of the natural world. The emphasis is always on the word “why.” But for all the lip service paid to basic research, the proper relation between it and technology development is not susceptible to a once-for-all solution. The objectives of basic research are not always easy to define, and frequently the only quality control — whether a piece of basic research is good or not — is provided by the researcher's colleagues and collaborators. Moreover, research per se is seldom the objective of a product-oriented institution, or even of most publicly or privately sponsored research installations. The purpose of a space probe may be to expand our knowledge of the universe, but the goal of the installation managing its design is to produce an operating system within the time and budget allotted. Often, advances in basic research may be more a permissive than an active element in determining what kinds of technology development will be on the agenda. Indeed, the relation can and does run the other way: A technological breakthrough can be a stimulus to basic as well as product-oriented research, as producers and users try to understand, and thereby improve, the original innovation. A breakthrough, as Rosenberg has said, may signal "the beginning of a series of new developments of great importance, not their culmination ... the development of the transistor or the explosion of the first nuclear device or the first achievement of heavier-than-air flight is really the announcement of a new set of possibilities far more than their attainment.” (ref. 4.) There is nothing predetermined in deciding what kinds of development will be undertaken, especially where improvements in existing systems are to be more than incremental.

It can be argued that one of the most important changes in the way that technology is developed is that, over the past twenty years, we have come to understand the process so much better. If they are so inclined, research administrators are in a position to know that the lines between research and technology development run both ways; that the introduction of new technologies often marks only the beginning of a process of discovery; that the needs of government agencies have stimulated civilian industries, notably in electronics; and above all, that the decision to sponsor a major program of technology development always represents a political choice. Up to the eve of the Second World War, the rate of scientific advance was such that scientists, research managers, and government officials often perceived technology developments as flowing directly from scientific discoveries. Each new discovery was, in due course, developed into new technology and then into engineering projects; examples that come to mind include Roentgen's discovery of X-rays, James Clerk Maxwell's theory of electricity and magnetism, and John Dalton's atomic theory.

[ocr errors]

This is no longer true. Since technology development is generally very expensive compared to basic research, choices must be made. We simply do not have enough money to support all the possible developments that could be based on current knowledge. For example: Should we or should we not develop the technology to extract thermal energy from the oceans? Should we or should we not develop the technology of moving earth with nuclear explosives? Should we or should we not develop hypersonic passenger aircraft? All of these things probably could be done if the decisions to undertake the necessary technology developments were made, since the knowledge on which these developments would be based already exists. Yet for reasons of public policy, none of these programs has been undertaken. Our mechanisms for making choices regarding the initiation of new technology developments are still rudimentary. We have, as subsequent chapters will show, an established pattern, but it is not at all clear that this pattern is properly geared to the needs of our society.

Thus technology development mediates between basic research and engineering that is, the application of the mathematical and natural sciences to develop ways to utilize the forces of nature for human benefit. Where a particular field of study attracts a sufficiently large group of workers under an established name, they tend to form professional societies, start their own specialized publications, and organize

, departments within the university, leading to recognition by the scholarly community. Sometimes the origin of such a field lies in the recognition of a need (the splitting of engineering into electrical engineering, civil engineering, etc., and more recently, into systems engineering and biomedical engineering), and sometimes in combining two previously separate disciplines. In contrast to basic research, technology development tends to be a group activity rather than an individual enterprise. * In contrast to engineering, the time scale is longer and the costs anywhere from ten million to billions of dollars — greater than for all but the largest engineering projects. The reasons for undertaking technology development have included, for example, a perceived crisis such as that leading to the development of radar and nuclear weapons in the Second World War, or some large perceived profit if the technology succeeds, as in the case of Polaroid-Land cameras and very large scale integrated circuits.

* The difference is much more one of degree than of kind. Basic research in such disciplines as high-energy physics and astrophysics is a group activity demanding access to sophisticated facilities and a large supporting staff. By comparison, companies such as Hewlett-Packard and 3M have deliberately kept their engineering and product development teams small. Yet there is still a difference in scale between a weapons development program and an experiment to detect the presence of neutrinos.

[ocr errors]

To summarize the argument to this point: Technology development is the process by which newly developed scientific principles are brought to the point where they can be applied in an engineering sense. Typically, the time between the beginning of a project and its completion (lead time) is on the order of five to ten years; a really large program, like the lunar · landing mission, may cost billions of dollars. It has become increasingly clear since the 1960s that such programs — some of whose features will be discussed in the next section — cannot be managed successfully in terms of a classical hierarchical structure. What we are dealing with here is the “large-scale endeavor,” a concept applied to his agency by the former Administrator of the National Aeronautics and Space Administration (NASA), James Webb. The endeavor characteristically results from a new and urgent need or a new opportunity created by social, political, technological, or military changes in the environment. Most often, it requires “doing something for the first time and has a high degree of uncertainty as to precise results” and it will have second-and third-order consequences, often unintended, beyond the main objective (ref. 5). A large-scale endeavor is so complex that senior executives in the sponsoring organization cannot be expert in all facets of the operation. “They must delegate important responsibilities to lower echelons and then find ways to make sure the delegations accomplish their purposes without harmful compartmentation.” (ref. 6.) The organization must be adaptive; “no longer can you have a grand idea and then go to work and cut and fit and try.” (ref. 7.)

Webb's description can, of course, apply to many endeavors besides the space program. Recent examples include the attempt to build and operate a national rail passenger network, to develop a strategic petroleum reserve, to build the Alaska pipeline, and many programs related to the national defense — all share many of the features Webb enumerates. But space and comparable programs have had certain advantages, stemming from the nature of their missions, in attaining their goals which most of the endeavors named above lacked.

Consider the American civilian space program of the 1960s. Goals could be stated in precise, operational terms. NASA would describe a goal within the broader mission: Put a communications satellite in synchronous Earth orbit; or, develop an unmanned spacecraft to soft-land on the Moon and a vehicle with a liquid-hydrogen upper stage to launch it. Such precision may be contrasted with those Federal agencies charged with improving the quality of education, fighting alcoholism and drug abuse, or finding permanent jobs for the hard-core unemployed. As Lindblom and Cohen have noted, "government agencies are again and again assigned . . . responsibilities beyond any person's or organization's known competence. They do not typically resist these assignments because they are funded and maintained for their efforts, not for their


results.” (ref. 8.) However this contrast developed, technology development managers have the tools and resources to deal with many technical unknowns and overcome enormous problems of time and budget. This book is about the logic of this process as it applies to Federally sponsored institutions.

The Federal Technology Development Laboratory

So far we have discussed technology development as a general category, without much regard to the sponsoring organization. At this point, it would be well to define those features which distinguish Federal from privately-sponsored research and technology development. However one defines technology development, the Federal Government is doing a lot of it. In 1981, Federal spending on research and technology development amounted to roughly $40 billion, compared to $34 billion in the private sector (ref. 9).* While some two-thirds of Federal research and technology development obligations go to industrial firms or for basic ! research carried on by universities, the remainder is done under direct Federal supervision, whether through field installations run by government employees, non-profit contract research centers, or governmentowned, contractor-operated facilities. ** The impact of these programs alone would be sufficient reason to discuss them; a really large development program like Apollo at its height employed over 400 000 persons and generated $24 billion in expenditures, all of which NASA officials liked to point out were spent on Earth. What makes Federal technology development distinctive?

First, while there is no "typical” Federal research installation, many of the larger ones combine basic research with engineering and technology development. NASA's Ames Research Center, for example, has engaged in basic research, notably in the life sciences; it has been the systems manager for the Pioneer series of interplanetary probes, and its


* But according to the National Science Foundation, 1983 was the first year in which corporate expenditures, at $41.7 billion, exceeded government research spending. Mark Potts, “U.S. Companies Probe Technology's Frontiers,” Washington Post (January 8. 1984), G14.

** According to the National Science Foundation (NSF), about 24 percent of Federally-supported Research and Technology Development is carried on in some 700 laboratories directly operated by government personnel. Another 9 percent is carried on in so-called Federally-funded research and development centers, which normally work exclusively or mainly for a single sponsoring agency. For the purposes of this book, both kinds of organization are considered to be engaged in intramural research. See Chapters VII and VIII for a discussion of the role of the research and technology development laboratory working under contract to a single sponsor. On trends in Federal research and technology development obligations, see NSF, Federal Funds for Research and Development, Fiscal Years 1981, 1982, and 1983 (NSF 83-320), Section 2.

« PreviousContinue »