Page images
PDF
EPUB

CHAPTER XI

Conclusions

In this book we have tried to explain the workings of a special kind of research institution, rather than research institutions in general. The technology development laboratories we have dealt with, whether federally funded or the "corporate" laboratories of such large corporations as DuPont, General Electric, or General Motors, require that three basic irreducible conditions be fulfilled: There must be a "critical mass" of personnel and facilities; the laboratories must have the multidisciplinary capabilities needed to move from one mission to the next; and there must be a promise of relatively stable financial support extending over several years or even decades.

The missions of the laboratories range from developing new spacecraft or non-conventional hard materials to devising new measurement techniques. All of these entail the development of some new process, procedure, technique, or capability, and this is essentially what the term "technology development" means. The missions of these laboratories differ from that of a university, which is the creation of new knowledge in a series of well-defined disciplines. This distinction is easy to draw. It is not quite as easy to define the difference between what we have called technology development and product development, as it is usually understood in the industrial sense. Very often, the precise function of technology development laboratories is determined by the attitude of their sponsoring agencies or corporations toward product development. Some, particularly the National Bureau of Standards, elect to stop well short of final products; while others, like the multiprogram laboratories under the Atomic Energy Commission and later the Energy Department, have preferred to push a given technology say, a fast breeder reactor all the way to a demonstration prototype. In the private sector, General Electric's corporate laboratory at Schenectady does no product development, but in smaller corporations the process of technology development and product development may be combined.

Depending on the agency's philosophy, each laboratory will provide a different answer to the question of when the mission, or the project within the mission, has been accomplished. That point may come at any one of a number of stages: when a prototype of flight hardware has been

successfully tested; when a component of a new weapons system is delivered to a contractor for systems integration; when research into the structure of composite materials leads to improvements in the design of helicopter blades; or when a deep-space probe returns new knowledge about the structure of Saturn's rings. The point to be emphasized is that, in a good laboratory, there is something open-ended about its mission. Research tasks and projects will have discrete beginnings and ends, but the mission itself will continue in some other manifestation.

It may go without saying, but let it be said anyway: This book is about successful technology development laboratories. In the case material in earlier chapters, we have postulated certain features of the successful mission-oriented laboratory. We find a small but significant segment of laboratory resources devoted to basic research and, along with it, considerable discretion in the allocation of funds to promising new areas; good communications between groups involved in basic research and those in applied research and subsequent development; an ability to diversify creatively within the organization's primary mission; strong overlap of all phases of research and development; a policy of early identification of future managers among professional staff; and emphasis on a lateral, rather than a vertical, direction of communications (ref. 232). The great research executives of the last thirty years have had an uncanny insight into the relation of the work done at the division and directorate level to their laboratories' missions. What is more, they have often been able to move their organizations into new areas, even should we say especially? - in the absence of a guiding national policy. It was not foreordained that the next step after Apollo would be the Space Shuttle or that ship nuclear propulsion might become the basis for a new commercial technology.

[ocr errors]

or

Thus, the tendency of a successful technology development laboratory is to apply the word "mission" in both a very narrow and a very broad sense. In the former case, the emphasis is on getting the immediate job done; in the latter, it is more a matter of maintaining the organization's technical competence over the long run. Why is this competence building so important? Because it positions the organization to exploit its own discoveries and, as desirable, to move into new areas. As Harvey Brooks has observed, "A company or for that matter, a nation that has a broad technical capability can quickly exploit the ideas of others, and can catch up on the bets that it misses provided it has the technical sophistication to identify promising ideas at a sufficiently early stage... Just as a company or a nation cannot expect to exploit every promising scientific discovery, so every discovery that it exploits need not be its own." (ref. 233.) Viewed in this light, mission and capability tend to blend into one another. It cannot be emphasized too strongly that, while most research organizations live to an extent from an

inherited intellectual capital, this can lead to stagnation and decline without the stimulus of new ideas. Building on an organization's scientific competence then becomes almost as important as any programmatic mission, because it is the principal way of assuring that such missions can be accomplished.

It is time to consider what conclusions can be derived from the analyses of the preceding chapters. It is not to be expected that we can here offer specific recommendations about what should be done in space, defense, or energy. Nor can we be certain that the structure of the U.S. science and technology base will resemble the present state five or ten years hence. But if we assume, not unreasonably, that the Federal Government and our large industries will have a continuing stake in promoting scientific research and technology development, then the following conclusions should prove useful.

1. The particular management arrangement under which a laboratory is operated is, and should be, primarily a matter of administrative convenience.

Technology development has occurred in a large variety of organizational arrangements. Laboratories may be operated by government employees or by contractors; they may be Federal contract research centers, like RAND, working primarily for one sponsor; or they may have a variety of sponsors, as in the case of SRI International. One mission may be perfectly compatible with different arrangements, while a changing mission may provoke the sponsoring agency into changing the structure of its laboratories. There is nothing final about the structure of a research organization. To take one example: Within the past decade the Air Force has converted its Cambridge Research Laboratories from basic research to an "exploratory development" institution, closed the Aerospace Research Laboratory and transferred part of its staff to other Air Force installations, converted one Federal contract research center, ANSER, into a conventional contractor, and authorized other contract centers to create units separate from their work for the Air Force (ref. 234). Since a laboratory may well outlive its original functions, there is no reason to expect its structure to remain unchanged.

As a corollary, the distinction between government-owned, government-operated laboratories and those run by contractors is losing much of its sharpness. One reason was given in Chapter VII: No Federal agency can abdicate responsibility for seeing that congressional appropriations are used properly. But the nature of Federally-sponsored research and development is itself changing, to the point where the distinctions we are talking about have become, if not meaningless, then at least less relevant than they might have been in the 1960s. These changes are of several kinds. First, the distinction between research and development functions, which are reserved to government employees,

and support services, which may be contracted out, is eroding. It is difficult to draw an absolute boundary between a support service, such as writing a complex computer program, from ostensible research and development functions like designing the spacecraft for which the program is being written. Both may be — indeed, they usually are — performed under contract. Second, the kinds of technology that support large mission-oriented projects are also changing. No Federal laboratory any longer has the capability for designing and building all the major components of systems as complex as the Space Shuttle or a ship-launched cruise missile. What the laboratory must retain is the ability to specify the research or systems concepts, select its prime contractors, and certify at any point along the line that the system meets all requirements. At an agency like NASA, these functions would be performed by government employees, with guidance from Headquarters: at the multiprogram laboratories of the Energy Department, the ultimate authority to specify the work to be done would reside at the Washington headquarters, with field offices monitoring work in progress. But the point to bear in mind is that the structure of the NASA centers, on the one hand and of the Energy laboratories, on the other, reveals very little about what these institutions actually do. Even within NASA, the work done at the Jet Propulsion Laboratory is not easily to be distinguished in kind from work carried on at the centers that are staffed largely by civil service employees.

What emerges from this analysis is the conclusion that the organization of most laboratories owes as much to the origin of the institution as to the present purpose. We believe that if a mission is of sufficient importance to justify establishing an installation to carry it out, then that mission can be carried out even as the institution changes. What is remarkable about older laboratories like the National Bureau of Standards and the Langley Research Center is how closely they have adhered to their original mission to maintain a national measurement system in one case, to conduct advanced aeronautical research in the other despite all the changes that have intervened. With some slight exaggeration, we would assert that any institutional arrangement that sustains the laboratory's mission is justified. The right arrangement, under any regime, means that laboratory executives have the discretion to assign work and to start research in new areas without the need to seek prior approval; that professional staff are involved in defining the design of major systems; and that the laboratory is something more than a funnel through which funds from the sponsoring agency pass through to

contractors.

2. A capability for performing basic research is essential to the work of a technology development laboratory.

Mention has been made of the importance of basic research in a

« PreviousContinue »