Herbert F. Weisberg is Professor of Political Science at The Ohio State University. His research interests include voting behavior, Congress, and research methods. He has served as coeditor of the American Journal of Political Science and has coedited Controversies in Voting Behavior and Theory-Building and Data Analysis in the Social Sciences.  

If at one time we thought that the movement to science would yield unification of the discipline, it is now apparent that there are many roads to science. Still it is important for us to consider yet again what the appropriate goals are for our...

If at one time we thought that the movement to science would yield unification of the discipline, it is now apparent that there are many roads to science. Still it is important for us to consider yet again what the appropriate goals are for our scientific enterprise. What works in theory building; induction and deduction; prediction and control; the search for useful principles to guide us ' examining these questions, we can build a better science.Political science has come so far as a discipline that different schools and scholars have different interpretations of science in the study of politics, and that diversity is important to maintain. Advances made in the study of political institutions and behavior are described in twelve essays from the 1983 annual meeting of the American Political Science Association . Addressing they do not employ any single approach to the study of the science of politics. Taken as a whole, they illustrate the multiplicity of interpretations that are presently given to the common enterprise.

When American Political Science Association President-Elect William H. Riker appointed me Program Chairperson of the 1983 Annual Meeting of the Association, we chose as a theme for the meetings "The Science of Politics." For the 1983 meetings we...

When American Political Science Association President-Elect William H. Riker appointed me Program Chairperson of the 1983 Annual Meeting of the Association, we chose as a theme for the meetings "The Science of Politics." For the 1983 meetings we wanted to build on the very successful 1982 experience, while focusing attention more specifically on the scientific elements of the discipline and encouraging reflection on its scientific status. Not only did papers presented at the 1983 meetings extend the science of politics, but several panels were specifically devoted to assessing the status of science in the study of politics. In particular, the Lasswell Symposium on the second evening of the convention had "The Science of Politics"as its topic, and special "theme panels" were organized for each of the 23 regular sections of the official program on the science of politics as applied to that section of the discipline. The theme papers included in this volume originated at those special sessions. It seemed desirable to publish the theme papers from the 1983 meetings, given the excellent reception of the Finifter volume cited above. However, the Publications Committee of the Association wanted to further evaluate the success of that volume, so it was decided to seek a commercial publisher for a collection of the 1983 theme papers. Unfortunately, the large number of papers meant that the resulting volume would be too large. Rather than condense the many papers on all the topics into a single book, the decision was made to publish those papers which focus on political institutions and behavior. Thus the important scientific work being done in international relations, comparative politics, public policy, and the study of race, gender, and ethnicity issues cannot be considered here. While we regret that more comprehensive coverage was not possible, the in-depth analysis of the state of science in the areas included helps make up for this loss. The publication of this volume of papers inevitably owes much to the efforts of many people. As President of the Association, Bill Riker played an important role in the development of the 1983 Annual Meetings, the setting of the theme for the sessions, and the planning of the Lasswell Symposium and the program sections. The program committee found excellent people for writing theme papers on the science of politics in their areas of the discipline. The panel chairs and discussants at the theme panels helped the authors refine their views as presented in these papers, as did a special set of reviewers recruited for giving the authors final advice as to modifications of their papers. Regrettably it is impossible to list these reviewers for public gratitude while maintaining the commitment to anonymity, but at least they know the importance of their contribution to this effort. Tom Mann, Executive Director of the American Political Science Association was a valuable source of advice, solace, and encouragement throughout my service as program chair and my preparation of this volume. Terri Royed has helped prepare the manuscript. And finally, I should extend my own appreciation to the authors of these papers for sharing with us their views of the science of politics. All of these expressions of debt further make the larger point of this book: the study of the science of politics is now a collective enterprise in which a large number of people share the efforts. I hope that it is useful to summarize the status as of 1983 in this volume, fully realizing that this is only a prelude to our continuing development of the science of politics. Herbert E Weisberg

When political science began to be "scientific," this generally meant that political scientists were becoming concerned with objective description and generalization. Induction was the dominant mode of theory building, with the goal of explanation being paramount. But we have come far enough along in our scientific endeavor that some would now demand more for the science of political science. An increasingly common view of science is the deductive approach to theory, as emphasized by Gerald Kramer in chapter 2 of this book. For Kramer, science is theory building. He speaks of prediction and control as "useful byproducts," but the "central object" to him is understanding, by which he means explanation in terms of a simpler set of principles. He finds the formal theory endeavors closest to this approach. While admitting that too much of the early formal work was devoted to impossibility theorems, he is heartened by the current trend toward positive models of processes. The lack of empirical testing is described as in large part due to poor measurement in empirical work, with insufficient attention to error in data and overuse of inadequate measuring instruments. Finally, he expresses his concern that we need less complexity in our theory combined with less simplicity in our measurement. By contrast, others would consider successful prediction to be the ultimate goal of our scientific inquiry. Duncan MacRae makes this case in chapter 3. Certainly such prediction must be based on theory, but his test of success is prediction. His enjoyable idiom for this test is our inability to predict the next election when asked by friends from the natural sciences at cocktail parties. In part this is like looking at the most recent voting behavior articles and asking "Where's the politics?" After all, what could there be to the study of voting if we can't predict the results of elections? But it is also useful in forcing us to remember the questions of consequences and uses of our work. MacRae is arguing that discovery is not enough in political science, that we must be concerned with the use of our results. Because of the importance of that "practical action," political science to MacRae is more than just a science. A further criterion for science in political science that goes beyond the debate between Kramer and MacRae is a capability for incorporating notions of political change. If our models are to be truly explanatory, they must be capable of explaining change as well as constancy and must be able to cope with change in the system. After all, change is inherent in politics, so a theory of politics should not be time bound. Ideally a theory of institutions would include a theory of institutional change while a theory of political behavior would incorporate behavioral change. Our first cut at theory development can be static, but as scientific observations accumulate, it becomes more important to be able to understand the over-time changes in those observations. Unfortunately, a science of political change can be even more challenging to construct than is a science of politics. The existence of this multiplicity of criteria for science aptly points to a dilemma in our current development of science in political science. The pioneers in the scientific treatment of politics expected that the scientific revolution would lead to unity in the understanding of political science. That has not been achieved even if our means of data collection and analysis have become more scientific. In part, this is because we do not agree on what "theory" is. Thus there is still a debate between the "empirical theory," which has become common in some areas of the discipline, and the "normal theory " which Kramer supports, to which we might add the "predictive theory" that MacRae desires and the "dynamic theory" advocated in the preceding paragraph. True believers may see value in only one of these approaches, but many political scientists recognize the contribution of each and do not wish to choose a single road toward science. The dialogue as to the proper criteria for science and the proper role for theory in science is continued in the assessments of the state of science in the study of political institutions and behavior in the later chapters of this book, with different authors advancing the different approaches discussed in this section. THE SCIENCE OF POLITICAL INSTITUTIONS The second section of this book examines the familiar topic of political institutions, but often from new perspectives. At times it has seemed more difficult to establish science in the institutional realm than in the behavioral realm, as if there were a discontinuity between science and the study of institutions. These essays show that there has been real progress in developing the science of institutions, even if the enterprise is not completed. The reexamination of the role of institutions is illustrated in chapter 4 by Kenneth Shepsle, who views institutions as providing an element of equilibrium into a political system where individual preferences might not otherwise lead to equilibria. Shepsle's chapter directly challenges the claim of discontinuity between science and the study of institutions. At one level, he examines the role of institutions in the policy process. Institutions are intermediary between voters and policy, and he focuses attention on that role. At another level, Shepsle considers science in the legislative area, showing how a formal model perspective can be useful in the study of legislative institutions. He is not content with an overly simplistic model of a legislature, but instead tries to incorporate the institutional characteristics that make legislatures special.�  Legislatures are also Lawrence Dodd's topic in chapter 5. Dodd suggests a broadbased theory of legislative change which relates change in the legislature to change in the public. The vastness of the area of legislative politics is such that Dodd's essay just covers one of many possible topics; it does not review science in the study of legislative committees, science in the study of voting in legislatures, or science in the study of political representation. The development of science is probably further along in each of these areas, whereas the topic of change (whether in the legislative or other arenas) has proved to be more difficult for scientific study. Dodd finds an absence of theory on legislative change, and so he builds one. His approach is not mathematical, but it is based on an understanding of the goals of political actors within an institutional setting. The work is exciting in terms of building a theory where one did not previously exist. Dodd does a nice job of integrating diverse strands of insights in the literature....

CONTENTS Part I. The Science of Political Science 1. Introduction: The Science of Politics and Political Change (Herbert F. Weisberg) 2. Political Science as Science (Gerald H.

CONTENTS Part I. The Science of Political Science 1. Introduction: The Science of Politics and Political Change (Herbert F. Weisberg) 2. Political Science as Science (Gerald H. Kramer) 3. The Science of Politics and Its Limits (Duncan MacRae, Jr., University of North Carolina, Chapel Hill) Part II. The Science of Political Institutions 4. Institutional Equilibrium and Equilibrium Institutions (Kenneth H. Shepsle, Washington University, St. Louis) 5. The Cycles of Legislative Change: Building a Dynamic Theory (Lawrence C. Dodd, Indiana University) 6. Presidential and Executive Studies: The One, The Few, and the Many (Bert A. Rockman, University of Pittsburgh) 7. The Social Science of Judicial Politics (James L. Gibson, University of Houston--University Park) 8. The Positive Theory of Hierarchies (Gary J. Miller, Michigan State University, and Terry M. Moe, Stanford University) Part III. The Science of Political Behavior 9. Structural Estimation with Limited Variables (Charles H. Franklin, Washington University, St. Louis, and John E. Jackson, University of Michigan) 10. The Dynamics of Public Opinion (Richard G. Niemi, University of Rochester) 11. Choice, Context and Consequence: Beaten and Unbeaten Paths Toward a Science of Electoral Behavior (Paul Allen Beck, Florida State University) 12. Model Choice in Political Science: The Case of Voting Behavior Research, 1946-1975 (Herbert F. Weisberg) Author Index

Paul Allen Beck is Professor of Political Science at Florida State University and Chairman of the department. He has written extensively on electoral behavior and public opinion, with recent articles appearing in The American Political Science Review, The Journal of Politics, and Political Behavior. His recent coedited book, Electoral Change in Advanced Industrial Democracies (Princeton University Press), focusses on the realignment and dealignment of contemporary electorates. Currently he is...

Paul Allen Beck is Professor of Political Science at Florida State University and Chairman of the department. He has written extensively on electoral behavior and public opinion, with recent articles appearing in The American Political Science Review, The Journal of Politics, and Political Behavior. His recent coedited book, Electoral Change in Advanced Industrial Democracies (Princeton University Press), focusses on the realignment and dealignment of contemporary electorates. Currently he is conducting research designed to explain the often contradictory attitudes of citizens towards local taxes and spending. Lawrence C. Dodd is Professor of Political Science at Indiana University. He has written extensively on legislative politics and legislative-executive relations, and is concerned particularly with the conditions that foster institutional change and policy responsiveness. He is the author of Coalitions in Parliamentary Government, the coauthor, with Richard Schott, of Congress and the Administrative State, and the coeditor, with Bruce Oppenheimer, of Congress Reconsidered. Charles H. Franklin is Assistant Professor of Political Science at Washington University in St. Louis. His research on the dynamics of party identification has appeared in the American Journal of Political Science and (with John E. Jackson) the American Political Science Review. James L. Gibson is Associate Professor of Political Science at the University of Houston-University Park. His research interests are generally within American politics, with specific foci on judicial process and behavior, political parties, and public opinion (political tolerance). His articles have appeared in the American Political Science Review, the American Journal of Political Science, the Journal of Politics, Law and Society Review, and in several other journals. Recently he coauthored Party Organizations in American Politics and Civil Liberties and Nazis (both by Praeger). Currently, he is involved in additional research on the role of party organizations in the electoral process, and the implications of political intolerance for freedom, political repression, and democracy. John E. Jackson is Professor of Political Science, Director of the Program in American Institutions, and a Program Director in the Institute for Social Research at the University of Michigan. His research interests include empirical methodology, political behavior, and political economy and institutions. He is the coauthor, with Eric A. Hanushek, of Statistical Methods for Social Scientists and of articles on Congress, public opinion, and political economy. He is currently working on a book on institutions and public policy. Gerald Kramer received his Ph.D. at the Massachusetts Institute of Technology. He has taught at the California Institute of Technology, the University of Rochester, and Yale University. He has been a Ford Fellow and has been at the Center for Advanced Study in the Social and Behavioral Sciences. He has published extensively on positive political theory, including seminal articles on economics and voting behavior and on equilibrium in multidimensional voting. Duncan MacRae, Jr. is William Rand Kenan, Jr., Professor of Political Science and Sociology at the University of North Carolina at Chapel Hill. He has written extensively on legislative behavior and on the foundations of public policy analysis. His most recent book is Policy Indicators: Links between Social Science and Public Debate. Gary J. Miller is Associate Professor of Political Science at Michigan State University. His principal research interests are formal theories of bureaucracy and small group experimentation. He is author of Cities by Contract: The Politics of Municipal Incorporation and coauthor, with John H. Aldrich, Charles Ostrom, Jr., and David W. Rohde of American National Government. Terry M. Moe is Associate Professor of Political Science at Stanford University and Senior Fellow at the Brookings Institution. His research interests include bureaucratic politics, organization theory, regulatory policy, and interest groups. He is author of The Organization of Interests and various articles, the most recent of which are Ã??Ã??Ã??'Ã??'Ã??'An Adaptive Model of Bureaucratic PoliticsÃ??Ã??Ã??'Ã??'Ã??'� (with Jonathan Bendor) and Ã??Ã??Ã??'Ã??'Ã??'Control and Feedback in Economic Regulation: The Case of the NLRB,Ã??Ã??Ã??'Ã??'Ã??'� both in the American Political Science Review. Richard G. Niemi is Professor of Political Science and Distinguished Professor of Graduate Teaching at the University of Rochester. He has written extensively on political socialization, voting behavior, and formal models of voting. He is currently engaged in work in all three areas along with studies of political districting. Bert A. Rockman is Professor of Political Science at the University of Pittsburgh. His research interests encompass the American presidency and the comparative study of leadership, the comparative analysis of bureaucracy and civil servants, and foreign policy making. Among other works, he is the author of The Leadership Question: The Presidency and the American System and the coauthor of Bureaucrats and Politicians in Western Democracies. Kenneth A. Shepsle is a Professor of Political Science and a Fellow in the Center for Political Economy at Washington University, St. Louis. His research interests include positive political theory and American politics, and currently focus on analytical models of political institutions. He is the author of The Giant Jigsaw Puzzle: Democratic Committee Assignments in the Modern House and, among his articles, are a series of papers on institutional equilibrium. Herbert F. Weisberg is Professor of Political Science at The Ohio State University. His research interests include voting behavior, Congress, and research methods. He has served as coeditor of the American Journal of Political Science and has coedited Controversies in Voting Behavior and Theory-Building and Data Analysis in the Social Sciences.