ment agency or industrial corporation. On the other hand, the inquiry leading to the discovery of nylon by Wallace Carothers of the Du Pont research organization in the early 1930's is often cited by industrialists as an example of basic research conducted within industry. When Dr. Carothers joined the staff of Du Pont's fundamental research organization and began to delve into the new field of linear superpolymers, all that anyone reviewing the situation at the time could have anticipated was that any breakthrough in this direction would be of value, but Carothers had no particular objective in mind. Much theoretical knowledge and the very practical achievement of a new synthetic fiber actually emerged from the work. Yet, some might argue that this was "mission-oriented" rather than "uncommitted" basic research, since hardly any field of chemistry could be considered outside Du Pont's potential range of interest.

Operational definitions.-Efforts to overcome the problems raised by criteria of basic research which involve a high degree of human judgment have led to the suggestion that shifting to objective or operational indicators would make easier the task of sorting basic research from applied. None of the "objective" indicators which have been proposed has been free of serious limitations, however. For instance, since basic research results are generally regarded as being in the public domain, it has been suggested that projects which are published in the scholarly journals can be classified as basic. This would be largely correct. However, research which is not reported in the journals cannot, by the same token, be automatically defined as applied research or development. There is reason to believe, moreover, that the results of a significant amount of basic research performed in governmental and industrial laboratories reach publication at a much slower rate than similar work at the universities.

Difficulties of applying criteria

The foregoing summary has referred to some of the difficulties of employing, for definitional purposes, the criteria which have been noted. These difficulties make it impossible to apply any one of these criteria exclusively to large numbers of projects to distinguish basic research from applied without running into the types of inconsistencies which have been hinted at. Furthermore, two or more criteria used together in combination are often contradictory. For instance, applied research projects directed toward a very practical result can often be characterized by great uncertainty of outcome-as were the atom bomb projects carried on under the Manhattan District project in 1940-45-while they simultaneously yield findings of great importance.

Even were the criteria more generally applicable for purposes of classification, an additional problem would present itself-persons in different scientific fields tend to take somewhat different views on the basic nature of a given project. Mathematics, physics, and engineering, for instance, have interwoven areas of concern, yet it is probable that a highly complicated research project on stresses in suspension bridges might be classified as basic by the civil engineer; as applied, by the physicist; and, possibly, as development by the mathematician whose computing techniques are used.

Prospects for clarification

If we have no generally accepted and widely applicable criterion of basic research on hand, it is not, as the foregoing discussion implies, for lack of vigorous discussion and exploration of possibilities. Rather, the difficulty is that we have been looking for a universal definition to meet all needs when, probably, only a number of more limited approaches are possible. That is to say, we may have to think in terms of achieving a general understanding of the nature of basic and applied research and development, for purposes of policy discussions and program planning in government, universities, other nonprofit institutions, and industry. For more limited purposes, such as the conduct of statistical surveys of expenditures and personnel employed in research and development activity by various types of organization, some working compromises will have to be reached on concepts and definitions.

There is reason to conclude that there has been a gradual increase in broad public comprehension, not only of the more spectacular achievements of scientific research, but also of the complex nature of the research process. This growth of understanding can be likened, in general, to the wider public grasp, which gradually developed at earlier times in our history, of such key economic issues as unemployment and tariffs and trade.

It is necessary, however, to strive for fairly precise concepts and definitions of basic and applied research and development for the purpose of estimating the financial and manpower resources devoted to research and development. Two of the major purposes for which such estimates are required are (1) to comply with Executive Order No. 10521 which directs the National Science Foundation to make "comprehensive studies and recommendations regarding the Nation's scientific research effort and its resources for scientific activities" and (2) to provide data which will enable public and private officials to relate research and development activities to other phases of American life.

To meet these and other needs, the Office of Special Studies of the National Science Foundation has built up a regular program of recurring statistical surveys designed to yield annual estimates of the volume of funds devoted to basic research and other aspects of research and development by the major sectors of the economy. At present, these surveys employ the motivational definition of basic research cited by Dr. Waterman (see p. 5) because this is the characteristic of basic research on which there appears to be the greatest

consensus.

The National Science Foundation's experience with surveys of research and development expenditures and personnel is the first centralized sustained experience of this type, and it is yielding valuable insights into the types of conceptual problems which have been touched on in this section. The Office of Special Studies is now capitalizing on this experience by planning a broad review of problems of concept and definition in the field of research and development surveys. Not only the difficulties of distinguishing between basic and applied research will be dealt with in this review, but also such equally challenging problems as differentiating between applied

research and development, and between research and development as a whole and such closely related activities as industrial production, medical care, and scientific training. Drawing on widespread discussion and consultation with persons throughout the scientific community and other related organizations, this review is expected to lead to a report which will present an interrelated group of concepts and definitions that take account, as far as this is possible, of the difficulties which will always beset the effort to pin down and classify many projects in scientific research.

Part 3

THE BASIC MISSION OF THE NATIONAL SCIENCE

FOUNDATION

Biological and medical research in the United States has increased substantially during the last 10 years, requiring larger expenditures of funds and an increasing number of biological and medical scientists. These increases in biological and medical research, of course, are characteristic of the trends exhibited generally by scientific research and development activities in this country.

Also characteristic of the trends in scientific research, including biological and medical research, is the enlarged support given by the Federal Government. The areas of common interest among all Federal agencies involved in scientific research and development require that effective channels be developed for the interchange of scientific information and that unnecessary duplication be avoided. The special interests imposed by assigned missions require, furthermore, that some overall planning exist to insure a balanced Federal research program. For such purposes, it seems wise that an agency independent of specialized missions, yet having an interest in common with other agencies in the progress of science and the national welfare, serve as a focal point for studying the Federal participation in research and development. Such an agency is the National Science Foundation.

Established in 1950 as an independent agency in the executive branch of the Government, the Foundation was authorized and directed by the Congress to fulfill the following basic missions: 1. Develop and encourage the pursuit of a national policy for the promotion of basic research and education in the sciences.

2. Initiate and support basic scientific research in the mathematical, physical, medical, biological, engineering, and other sciences.

3. At the request of the Secretary of Defense, initiate and support specific scientific research activities in connection with matters relating to the national defense.

4. Award scholarships and graduate fellowships in the mathematical, physical, medical, biological, engineering, and other sciences. 5. Foster the interchange of scientific information among scientists in the United States and foreign countries.

6. Evaluate scientific research programs undertaken by agencies of the Federal Government and correlate the Foundation's scientific research programs with those undertaken by individuals and by public and private research groups.

7. Establish such special commissions as the National Science Board may from time to time deem necessary.

8. Maintain a register of scientific and technical personnel.

9. To appraise the impact of research upon industrial development and upon the general welfare.

These are broad responsibilities. Principal functions of the Foundation may be divided into three categories: (a) the encouragement and support of basic research, (b) the training of future scientists, and (c) the development of national science policy recommendations, that is, indicating the role the Federal Government should play with respect to science. Although each of these functions is related to the others, the first two may be thought of as the operating programs, and the last may be considered as the policy function of the Foundation. The cornerstone of the Foundation's policies with reference to the biological and medical sciences is contained in the outline of the functions of the Foundation in Public Law 507, section 3, 81st Congress, which states in paragraph 2, that the Foundation is authorized and directed:

to initiate and support basic scientific research in the *** medical, biological *** and other sciences *

Further along in the same section, the NSF Act states that it is a function of the Foundation:

to award *** scholarships and graduate fellowships in the ** medical, biological *** and other sciences ***.

In the legislative history of the National Science Foundation Act, there are several references to the Foundation and the biological and medical sciences. Perhaps the outstanding impressions from the legislative history are contained in Conference Report No. 1958 of the National Science Foundation Act of 1950 and in the Congressional Record of February 27, 1950.