Political scientists want to do good. They want to expand knowledge about political life, but also they wish to use knowledge for political reform. Usually this means desiring to promote “democratization.” Historically democracy and political science have tended to develop together. In modest ways political science can contribute to the emergence of democracy. Political reform succeeds best if it occurs incrementally, in the spirit of “one soul at a time.”

A central claim of both Marxist and neoclassical political theory is that under capitalism all governments must respect and protect the essential claims of those who own the productive wealth of society. This is the theory of “structural dependence of the state on capital.” Using a formal model, the internal logic and the robustness of the theory is examined. We conclude that in a static sense the theory is false: virtually any distribution of consumption between wage earners and owners of capital is compatible with continual private investment once an appropriate set of taxes and transfers is in place. Yet the state may be structurally dependent in a dynamic sense. Policies that, once in place, redistribute income without reducing investment do reduce investment during the period in which they are anticipated but not yet implemented.

Weber's commentators often accuse him of lacking a coherent political philosophy because his pluralist-elite theory of democracy seems indifferent to liberal-democratic values. I argue, however, that the core of Weber's political philosophy is a politicized neo-Kantian liberalism, one that produces an ethically significant and positive concept of politics. The problem is rather that Weber's pessimism about institutionalizing positive politics in bureaucratized societies left the ethical core of his political philosophy inexplicit. This introduced a conflict into his thought between his ethical commitments and his assessments of political possibilities. The conflict is compelling because it reflects the contemporary gap between the promise and performance of liberal democracies. At the same time, formulating Weber's problems in these terms helps identify democratic solutions that remain obscure in his assessment of conflicts between bureaucratization and democracy.

I provide a narrative of the emergence of an expressly articulated “political science” in the Scottish Enlightenment. Political science was designed by Hume, Smith, and others to advance both a Newtonian method for the study of politics and a politics of moderation whose tasks included a critique of enthusiasm. In this way, political science, moderation, and (anti)enthusiasm were conceptually connected. The emergence of political science, understood in this way, required a number of conceptual changes in a structure of argument shaped largely by Locke. These conceptual changes, in turn, fixed a rhetorical framework for persistent debates over the methodological and political identity of political science, even as ideology literally replaced enthusiasm. These persistent debates reveal the relevance of the history of political science as a forum for remembrance, reflection, and critique.

The contemporary estrangement of political theory from political science is in large measure the product of a quarrel that originated in the challenge to the values of U.S. political science initiated by émigré scholars during the 1940s. The behavioral revolution was in an important respect a conservative rebellion in defense of the values of liberalism and related notions of science, relativism, and historical progress that had traditionally informed the discipline. This controversy in the context of political science fundamentally structured the discourse of academic political theory and the contemporary constitution of the field both as a division of political science and as a wider interdisciplinary enterprise.

For the last quarter century public-law studies have been dominated by “political jurisprudence,” which tries to understand law as a product of political forces. Critics claim this outlook, as now articulated, has generated fragmented empirical work disconnected from larger normative issues. This essay uses recent “institutionalist” or “structuralist” perspectives, based on critiques of pluralist political science and both Marxist and functionalist sociology, to propose a political approach to public law studies that can avoid such criticisms. If both empirical and normative public-law scholars took as their central concern the “dialectic of meaningful actions and structural determinants” and recast their research in several specified ways, they might be better able to describe the role of normative ideas in law and to achieve a broader empirical agenda that could ground and inform normative debates.

Mancur Olson's model of economic growth has attracted great attention as a theoretical account of how interest groups influence the rate of economic growth over time. Moreover, the model appears to have received strong empirical support in Olson's tests employing U.S. state data. However, the specification of the Olson construct in these tests is insufficiently attentive to the complex causal chain implicit in Olson's argument, inadequately accounts for precisely how interest groups matter, and employs a static research design that obfuscates the cause-and-effect relationships posited by the model. We review these issues and develop a more complete specification of the Olson model. The respecified model is then tested using U.S. state data for the period of the late 1970s and early 1980s using new measures of interest-group influence.

As the U.S. states develop their political institutions and take greater responsibility for their economic well-being, two concerns that have long driven research on national elections—electoral insulation and economic accountability—should become central in research on state elections. I investigate institutionalization's effects on the vulnerability of state elections to major periodic forces—coattails, turnout, and economic conditions—and how political responsibility for economic growth is apportioned between presidents and governors in state elections. The investigation relies upon dynamic models of state legislative and gubernatorial outcomes estimated with a pooled data set comprised of most states and elections in the years 1940–82. The results, which have important implications for state government more broadly, indicate that institutionalization has substantially insulated legislative elections against major threats and that state legislators and governors have less to fear from their state economies than is often thought, but also that state elections are becoming more susceptible to swings in the national economy.

A brinkmanship crisis with two-sided incomplete information is modeled as a game of sequential bargaining in which each state is uncertain of its adversary's resolve. The sequential crisis equilibria are characterized explicitly and used to analyze the influences of resolve, misperception, and the status quo on escalation and crisis stability. The description of brinkmanship as a contest of resolve is found to be misleading: the state with the greatest resolve may not prevail in the crisis; a state may be less, not more, likely to prevail the greater its resolve; and a states' expected payoff may be less, not more, the greater its resolve. Moreover, reducing misperception may destabilize a crisis. Surprisingly, increasing the stake a potential challenger has in the status quo may not make a challenge less likely. Finally, crises involving severe conflicts of interest are shown to be less likely than crises not entailing a severe conflict of interest.

Findings from a data-based study of bargaining in recurrent crises between evenly matched states provide the foundation for the construction of four crisis-learning games. Symmetrical and asymmetrical nuclear and nonnuclear sequential three-by-three games assuming complete information and nonmyopic play are presented and analyzed. The empirical study indicated that states that were unsuccessful in one crisis were likely to move to more coercive bargaining strategies in the next crisis. The four sequential games offer insights as to why this is likely to produce unwanted consequences, while demonstrating the importance of the participants' initial strategy choices. With the realpolitik lessons suggested by the earlier study removed, the dynamics of the games present a case for beginning with a cooperative strategy and moving to reciprocating, or tit-for-tat, bargaining.

The core of a voting game is the set of undominated outcomes, that is, those that once in place cannot be overturned. For spatial voting games, a core is structurally stable if it remains in existence even if there are small perturbations in the location of voter ideal points. While for simple majority rule a core will exist in games with more than one dimension only under extremely restrictive symmetry conditions, we show that, for certain supramajorities, a core must exist. We also provide conditions under which it is possible to construct a structurally stable core. If there are only a few dimensions, our results demonstrate the stability properties of such frequently used rules as two-thirds and three-fourths. We further explore the implications of our results for the nature of political stability by looking at outcomes in experimental spatial voting games and at Belgian cabinet formation in the late 1970s.

A principal-agent perspective has been employed in recent studies to rediscover the importance of democratic hierarchies in shaping public bureaucratic outputs. I test the robustness of the hierarchy model for explaining outputs from an agency that has often been cast in the image of bureaucratic independence, the Environmental Protection Agency. Examining the effect of the Reagan presidency on EPA outputs for clean air, Box-Tiao models are constructed to explain shifts in the vigor of air pollution enforcements between 1977 and 1985. The analysis shows that the influence of elected institutions is limited when an agency has substantial bureaucratic resources and a zeal for their use. Moreover, under these conditions, bureaucracy can even move outputs in directions completely opposite from what a model of hierarchy would predict. The implication is that for some agencies it is necessary to give greater consideration to the agent in explaining implementation outcomes through time.

Do U.S. senators adjust their policy positions or voting behavior—engage in “strategic moderation”—in their quest for reelection? In the June 1986 issue of this Review, Gerald Wright and Michael Berkman sought to demonstrate that Senate incumbents moderate their ideological positions as elections near. This endeavor was part of their larger effort to show the importance of policy issues in the selection of members of Congress. Robert Bernstein takes the view that the claims about strategic moderation rest on methodological flaws. But Wright and Berkman argue that most investigators agree on the general direction of senatorial candidate behavior. The controversy turns on conception and interpretation of analytical results.