Search

The long-debated issue whether social scientists should be concerned with improving the human condition, and if so how they should attempt this task, has been raised afresh in the exchange of views between Lee Benson (1978) and Warren Miller (1978) on “Changing Social Science to Change the World.” Before plunging into this debate, however, I should like to suggest that all of us who participate in it are viewing the world from a particular occupational perspective. Some years ago, when George Lundberg’s Can Science Save Us? (1947) was current, a friend suggested to me that a prior question was, “Should science save us?” He may have meant, as Miller suggests, that saving or changing the world is not the principal task of social scientists; or in another meaning, he may have been anticipating more recent critics in noting that a world “saved” by social science might not be one in which everyone would prefer to live. I hope that our efforts will not be counter-productive.

Since the theory of games was first made widely available, with application to economic behavior, its use has been suggested in many other areas, from the global to the individual. Several correspondences between game theory and certain aspects of political process have been noted.The contribution of game theory to substantive knowledge in the empirical sciences, however, has been modest; Luce and Raiffa judge that its impact has been greater in applied mathematics. The area of political behavior—despite the apparent applicability of the notion of conflict of interest—is similarly lacking in studies, although a few notable exceptions exist.

Among the causes of the demise of the Fourth Republic, one prominently mentioned has been the incapacity of governments to govern. Not merely cabinet instability as such, but the association of premiers' popularity in the Assembly with their disinclination toward bold policy innovations, is the charge. Stability indeed there was: continuity of ministers under changing cabinets, relative persistence of cabinets (such as those of Queuille and Mollet) that maintained the existing political balance or even defended the Assembly against innovation. Stability, too, of policy: the continuation of the Indo-Chinese war and the Algerian war, for example, long after a policy change might have been considered. Underlying this sometimes artificial stability was the threat that an innovating premier or minister might soon find himself returned to the status of an ordinary denuty.

In the past seventy years, lasting reorientations of the national electorate have taken place in two periods, centering about the presidential elections of 1896 and 1918. Most other presidential elections have involved relatively uniform swings of states or counties toward one party or the other; Louis Bean summarized this phenomenon in his chapter title, "As Your State Goes, So Goes the Nation."' But the occasions when this uniform swing does not occur are of special interest, because if the reorientations persist they can mark the injection of new issues into national and state politics for a generation. Lubell noted the importance of the "Al Smith revolution" which preceded the "Roosevelt revolution" ;2 and Key, naming these phenomena "critical elections," went on to show that Bryan's candidacy in 1896 marked an earlier major reorientation of the electorate.3 He defined a critical election as one in which "the depth and intensity of electoral involvement are high, in which more or less profound readjustments occur in the relations of power within the community, and in which new and durable electoral groupings are formed."4 Since no sample survey data are available for either 1896 or 1928, the systematic study of voting in these elections must rest largely on analysis of voting statistics. These statistics can be interpreted more fully with the aid of

Roll-call votes are being used increasingly to throw light on various aspects of the legislative process. As long as these votes are neither simply unanimous nor cast purely on party lines, they contain information that can often be rendered more intelligible by the simplification or condensation of many votes into fewer variables or dimensions. The researcher interested in a particular legislative decision can thus profit by seeing whether it exemplifies a more general and repeated type of occurrence. The techniques of analysis used in studying legislative votes are broadly applicable to collegial bodies of many sorts, including municipal, state, and national legislative bodies; party congresses and conventions; the U.S. Supreme Court; and the United Nations General Assembly.Two major questions have been asked which lead to the search for different kinds of simplifying variables in this analysis. One concerns theissuesthat divide a given group of legislators at a given time,i.e., what general matters are being argued about? The second concerns thesubgroupsof legislators within the group selected for study: what are the blocs, factions, cliques, and the like, whose more persistent existence is reflected by the division on a given vote?

The paper argues that ethical discourse is intimately involved in the research literature of social science, and especially of political science, but is relegated to a subsidiary position. It therefore shares some of the vagueness, flexibility, and potential self-contradiction of the discourse of everyday life, rather than being sharpened by rational criticism. The norms governing scientific discourse provide not only for empirical testing, but also for rational criticism in the formulation of theories; an analogous type of criticism is shared by legal discourse. Some of the norms of scientific communication may be transferred to ethics by the specification of rules for ethical argument, requiring that arguments derive from previously specified, clear and consistent “ethical hypotheses.” Within such rules, ethical systems formulated by social scientists and philosophers may be compared critically. Systems amenable to such comparison include those of welfare economics, cost-benefit analysis, and formal democratic theory. The discourse embodying this argument and criticism is particularly appropriate within the normative tradition of political science. Its possible benefits include clarity about our valuations; communication among disciplines that enlarges the perspectives of each discipline; and a more independent, self-conscious examination within the university of the criteria for policy formation.

“… [T]here is no necessary conflict among these three desires of the American social scientist: to be a scientist like physical and biological scientists, to provide useful technical services, and to be significant at the level of policy. The chapters of this symposium are intended to illustrate their compatibility.”This statement indicates a major theme of The Policy Sciences – a volume that marked, as of 1951, the aspirations of a group of leading American social scientists for the policy applications of their disciplines. The harmony of goals that it suggests is no longer evident today.The possible incompatibilities among the goals of pure science, applied science, and policy can be seen by examining The Policy Sciences in two decades' perspective. They are of three major kinds:1. To provide intelligent advice on practical problems, the social science disciplines need to include systematic valuative discourse in a way that natural science does not.2. Applied social science (like applied science generally) differs from pure natural science in stressing valuative dependent variables that may not be closely related to the conceptual schemes of pure science, and independent variables related to alternative choices open to the actor.3. Different roles and channels of influence are appropriate for pure and applied science; and for applied social science in democratic regimes, participation and consent on the part of those influenced are of vital significance.

A major concern of political theorists has been the definition of the proper role of legislators in relation to their constituencies. Yet relatively little analysis has been made of the uniformities of behavior that actually prevail in these relations. Such uniformities, if they could be found, would bear directly on the theory of the party system, on speculation about the nature of representative government, and on the feasibility of proposals for a reordering of party practices.It has been shown that the tendencies of Congressmen to vote with their party or to cross party lines are associated with the similarity or dissimilarity between party policy and presumed interest of constituency. One aim of this study is to test the applicability of this proposition to the Massachusetts House of Representatives, by examination of the relation between roll-call votes and constituency characteristics. The evidence indicates that a similar relation exists in this body, and has been present consistently throughout the last two decades.

The social indicator movement has always faced in two directions—toward academic disciplines that provide quality control and estimate causal relations, and toward the political system that chooses and uses indicator statistics. At worst, the movement has risked appearing to be peripheral to both theoretical social science and policy choice; such perceptions may have contributed to the movement's weakening. The use of noneconomic time series of data to guide the definition of public problems, however, did not and will not die away.

The draft constitution for the Association, presented to us by the Constitutional Revision Committee, responds to a widely expressed desire for assurance that our officers are representative of our membership. We live in a time of increasing demands on all of our institutions, including those of the APSA, An important purpose in devising workable democratic forms is therefore to insure that those who make these demands do not feel that they have been unheard because of biases in their institutions of government. Although this is not the only purpose of our constitution, it is an important one, and the central one that the framers of the proposed draft seem to have had in mind. I do not believe that this purpose can easily be attained by other alternatives before us, including our present constitution.

Twenty years ago the publication of Toward a Social Report by the United States Department of Health, Education and Welfare was hailed as a major forward step in developing indicators of conditions in society into a national system of social accounting of relevance to public policy. The resulting social indicators movement quickly mobilized able social scientists to produce a variety of indicators monitoring trends in their society, and internationally. National governments too began to sponsor new types of social reports. The years since have seen an apparent decline in the momentum of the social indicators movement. Hence, to evaluate developments, the Journal of Public Policy invited a number of distinguished pioneers in the movement in Europe and America to give their individual assessment of what has happened to social indicators.