Testimony by Warren E. Miller Before the N.S.F. Biological and Behavioral Science Task Force on Reorganization, November 29, 1990

Abstract

Political science and other closely related social science disciplines could certainly benefit from the creation of a Directorate for the Social and Behavioral Sciences within the National Science Foundation. I present the case for such organizational restructuring on behalf of the American Political Science Association and the Western Political Science Association, and as a charter member and former President of the Social Science History Association. That a benefit would accrue from a reorganization would seem likely in the face of two organizational imperatives. First, political science and its sister disciplines need direct representation by senior officers of their own directorate in the policy making and resource allocation of at least three additional existing directorates: the Directorate for Computer and Information Science and Engineering, the Directorate for Education and Human Resources, and the Directorate for Scientific, Technological and International Affairs. The needs of political science in these three domains are similar to those of the other social sciences and are distinctly different from the needs of either the life sciences, the geosciences or the mathematical and physical sciences. Social science needs will not, and most likely cannot, be articulated by Foundation officers whose organizational responsibilities are overwhelmingly defined by the needs -and current resources-of the biosciences and whose professional backgrounds lie in one of the biosciences. We believe that the manifold resources of the Foundationprofessional and technical as well as budgetary-have not successfully addressed the needs of the social sciences in large part because the social sciences are not directly represented at the appropriate organizational level within the Foundation. The second organizational imperative stems from the need for greater organizational differentiation within the social sciences. Even though few of the social and behavioral science disciplines are as diverse as the array of subfields in chemistry or its sister disciplines, the full panoply of research specialties across the several social sciences is on a par with the diversity represented in the other substantive directorates. Many of the existing activities of the present Division of Social and Economic Science could be relocated as divisions of the new directorate. For example, without attempting to provide an organizational blueprint for the future, it may be suggested that, as with the other substantive directorates, each of the present disciplinary programs in S.E.S. might well be a division within a Social and Behavioral Science Directorate. They might be joined by a Division of Methods, Measurement and Instrumentation needed to address those problems of data generation and analysis that the disciplinary divisions have in common. Similarly, there should also be a separate division for large-scale multi-purpose data collection and resource development. A quite new division might also be established for activities centered on increasing the scientific usefulness of data generated by governmental agencies. Finally, and still illustratively, a separate division might be created for multi-disciplinary or multi-institutional projects or programs. To give greater clarity to the foregoing prescriptions, consider the following. First, with regard to representing social science needs in other directorates, the computer has become as central-and totally indispensable-to the work ways of social science as to the other sciences. And yet the central tasks for the computer are somewhat different. Certainly in contrast to mathematics, social science does much more data management of numeric data, more archiving and retrieval of non-quantitative materials, and much less sheer computation. On a quite different dimension, social science has its own version of the adaptation of the computer to data generation. In the harnessing of the computer in Computer Assisted Telephone Interviewing, in improving methods of textual analysis, and in the use of the lap top computer for data collection in the field (and quite apart from use in simulation exercises), we are only beginning to exploit fully this technological wonder. As a third illustration, it can be noted that in the absence of large laboratories or research centers which bring together scientists working on common problems, the computer network is becoming an essential feature of the social scientist's life. Both the transmission of data by computer nets and inter-personal exchanges among scientists are probably more crucial for the maturing social sciences than for the more developed disciplines. Many of these and other needs of the social scientist are served indirectly and inadvertently by computer developments in other realms. However, without a new directorate in the Foundation, it seems unrealistic if not unreasonable to expect strong and direct representation of social science computing needs that should affect future Foundation policy and resource allocation. A separate Directorate for the Social and Behavioral Sciences is a necessary, if not sufficient, condition to have an impact on Foundation decisions concerning the development of computer and information science.