(2) summer fellowships for graduate teaching assistants, and (3) cooperative graduate fellowships.

The Foundation, through its Science Information Service, fosters cooperation and coordination of scientific information activities among agencies of the Federal Government and among non-Government organizations engaged in activities in this field. A major goal of the Service is to insure the ready availability to all U.S. scientists of the world's current and past output of significant scientific information. Its four major programs during the past year continued to be concerned with research in the scientific information field, with the support of scientific publications, with the unpublished results of Government-sponsored and other research, and with foreign scientific information. Information programs dealing with the life sciences that were supported by the Foundation during fiscal year 1958 are described in this report.

If the Foundation is to play its part as an instrument of national policy, it must be responsive to the growth and changes in science. Although its study programs are designed to contribute to making the Foundation sensitive to the needs of science, it must in the long run depend upon the scientists and their respective societies and the cooperation of other Federal agencies in order to help it fulfill its responsibility; namely, the encouragement of national and international programs for the programs of scientific research.

The National Science Foundation's budgets are summarized in appendix 3, tables A and B. These tables include the funds authorized and obligated in the 1957 and 1958 fiscal years, together with estimated funds for the 1959 fiscal year.

44271-59

Part 2

PROBLEMS OF CONCEPT AND DEFINITION WITH RESPECT TO BASIC RESEARCH

Introduction

In a report such as this, which reflects the wide support and great achievements characterizing basic research in the life sciences and in other contributing fields, it may seem rather surprising to devote a section to problems of concept and definition. Basic research as an expression of man's unquenchable thirst for knowledge needs no definition in order to thrive. Yet, it is because of this thriving that definitions have become more important in the past decade or so, for as basic research has become increasingly a factor in public policy the need to describe it more clearly has grown. Universal definitions of basic research and applied research do not exist today. No attempt will be made here to reconcile the divergent lines of thought which do exist but rather to summarize these differences of opinion and indicate the present outlook.

Emergence of definitional problem

The problem of achieving satisfactory concepts and definitions with respect to basic and applied research and development is a relatively recent one in the long and fruitful history of research. Yet, it does not come from any newly arisen difference of opinion over the nature of research or over the general distinction between basic and applied research.

Indeed, it should be emphasized that most persons concerned with scientific research-even when drawn from greatly different environments and types of position-find themselves in considerable accord on a broad general description of basic research. Few would take exception to the following characterization which physicist Merle Tuve of the Carnegie Institution made to the forum of scientists who forgathered in May 1959 to consider means of improving the Nation's basic research effort:

We all know what we mean by truly basic research. We mean a devoted and almost passionate personal activity in search of new knowledge, not just factual information, but knowledge of the kind which can enlarge our understanding, knowledge which is not facts in isolation but facts related to guiding hypotheses or principles, knowledge which relates to natural law. This kind of truly basic research is a creative activity, an expression of wonder ***. It is a quest, not a job to be done. The measure of success is the quality of the effort and the character of the critical selection of goals to be sought, not the quantity of the output of scientific results. These men serve the conviction that greater knowledge and deeper understanding are undeniably good.1

Dr. Alan T. Waterman, Director of the National Science Foundation, in addressing the same group, cited the definition used by the Foundation in its fact-gathering surveys and went on to interpret it:

"Basic research is that type of research which is directed toward increase of knowledge in science. It is research where the primary aim of the investigator is a 1 Merle A. Tuve. "Is Science Too Big for the Scientist?" Saturday Review, June 6, 1959, p. 49.

fuller knowledge or understanding of the subject under study, rather than a practical application thereof."

This definition [continued Dr. Waterman] has stood the test of time fairly well. One objection commonly heard, however, is: How can one determine the investigator's motives? The simplest answer is that if one feels he must make a psychiatric test of an individual to determine why he wants to do a piece of research, then it is undoubtedly basic! 2

Most persons concerned with research and development, too, accept, for purposes of formal clarification, the concept of a spectrum of research and development activity. According to this analogy, basic research can be said to begin at the point where fundamental questions and speculations concerning the world about us are made the subject of systematic exploration and inquiry simply, as Dr. Tuve has indicated, for the sake of adding to our store of knowledge and understanding. Basic research includes a wide variety of activities and often contributes directly and immediately to applied research which seeks and uses new knowledge for practical purposes. These latter purposes may range from the conservation of human life to the improvement of a firm's product line. As applied research projects or programs progress and their immediate objectives are realized, the results may be employed in the development of new processes, products, materials or systems, or the improvement of existing ones.

Interestingly enough, the general consensus which prevails among research scientists, research administrators, students of the research. process, and others on the broad difference between basic and applied research and between applied research and development often fades into sharp but honest disagreement when individuals in these groups are faced with such tasks as developing separate budget categories for basic and applied research or summarizing and analyzing a diversified research program. This difference of interpretation arises from the obvious fact that no one individual or organization, however strategically placed in the research community, can see all the facets of the basic and applied research picture. Rather, the view varies from university to industry to government; from projects in mathematics to those in physiology or geology; from the office of a Cabinet Secretary to that of a research director or laboratory scientist. Criteria for identifying basic research

The discussion of criteria for distinguishing between basic or fundamental research and applied research has touched on a wide range of factors. Some commentators have felt that basic research was characterized by a combination of these factors. Others have felt that one aspect was salient enough to clinch the classification. Still others have pointed out inconsistencies among the various suggested criteria. Here we will touch only on the major lines of thought.

3

Motivation. One of the best known and most widely employed criteria of basic research is the requirement that the investigator be motivated primarily or exclusively by intellectual curiosity and an interest in the study of the laws of nature for their own sake, without regard for the immediate applicability of any findings he may reach. The National Science Foundation definition cited by Dr. Waterman makes this point.

Dr. Alan T. Waterman. "Basic Research in the United States." Remarks at the Symposium on Basic Research, New York City, May 14, 1959.

A discussion along these lines appears in Kidd, Charles V., "Basic Research-Description versus Definition," Science, Feb. 13, 1959, pp. 368-371.

A classical example of such research is the work of Gregor Mendel, the Austrian abbot and biologist, who found, in the course of 8 years of painstaking experimentation with pea plants possessing contrasting characteristics, that an underlying principle determined the manner in which these characteristics were inherited by successive generations of peas. Mendel published his findings in 1865 but they remained unnoticed until 1900 when his paper was rediscovered and his interpretations (checked and verified) became the foundation of the modern theory of heredity.

Yet, striking as this and other similar examples are, the motivational criterion of basic research is difficult to apply in an analysis of an agency's research program because of its great subjectivity. A given project may be classified differently by two different organizations which are interested in it. For instance, the university scientist who is conducting the project because of its promise as a contribution to knowledge will regard it as basic, while the fiscal officer of a Federal agency which is financing the work with some ultimate practical goal in view may allocate it in an applied budget category. Where more than one scientist is working on a project, moreover, motivations may differ. And, with respect to a given project, persons at different levels the research scientist, his fiscal officer, and the vice president in charge of research-may all have a different point of view.

Uncertainty. While uncertainty of results characterizes all research and development work, basic research is regarded as involving the highest degree of uncertainty, with even negative findings constituting new information. Associated with this uncertainty is the striking element of "serendipity"-the possibility of coming upon valuable findings which were not anticipated. Dr. Vannevar Bush has summarized the element of uncertainty as follows:

One of the peculiarities of pure science is the variety of paths which lead to productive advance. Many of the most important discoveries have come as a result of experiments undertaken with quite different purposes in mind. Statistically it is certain that important and highly useful discoveries will result from some fraction of the work undertaken; but the results of any one particular investigation cannot be predicted with accuracy.

Degree of freedom.-Closely related to the two foregoing criteria is the requirement of freedom of mind and lack of commitment to a particular line of inquiry. As Dr. Bush has said:

The unpredictable nature of pure science makes desirable the provision of rather special circumstances for its pursuit. Pure research demands from its followers the freedom of mind to look at familiar facts from unfamiliar points of view. It does not always lend itself to organized efforts and is refractory to direction from above. In fact, nowhere else is the principle of freedom more important for significant achievement."

Such freedom is also described as involving detachment from pressing time and financial commitments, although obviously this must be a relative matter, and the freedom to abandon as well as undertake a line of inquiry.

Those who feel that freedom from a program commitment constitutes a decisive element in distinguishing between basic and applied research question whether "truly basic research" can take place within the larger framework of the operating mission or goal of a Govern

Vannevar Bush. Science, The Endless Frontier. A report to the President. Washington, D.C., 1945, p. 75. Ibid., pp. 75-76.