Building on suggestions by Isaiah Berlin, a number of thinkers have elaborated a moral theory of value pluralism. Berlin himself believed that value pluralism was consistent with liberalism, which he understood as a political theory giving great weight to the value of negative liberty. Theorists led by John Gray have argued, however, that Berlin's pluralist stance toward values is not consistent with his commitment to liberalism. Gray's critique has triggered a wide-ranging theoretical debate, and the purpose of this article is to assess that debate. I sketch the essentials of value pluralism and explore its implications for liberal political theory. I conclude that Gray's argument modifies our understanding of liberal democracy but does not drive a wedge between value pluralism and liberal democratic theory. Nevertheless, it underscores the importance of prudence and attention to specific circumstances when using liberal democratic norms to guide the reform of illiberal institutions.

What is the effect of regime type on public expenditures for social programs? We investigate the relationship between democracy and the change in social spending—controlling for GDP, the debt, inflation, and age structure of the population—through a time-series cross-sectional panel data set for 17 Latin American countries from 1980 to 1992. The results show that, especially in poor countries during economic crisis, democracies increase the allocation of resources to social programs relative to authoritarian regimes. This suggests that the latter are more constrained by economic forces, whereas democracies are more constrained by popular demands. Hence, calls to abandon broad categorizations of regime type appear to be premature: Democracy can matter in systematic and substantial ways.

We examine formally the link between domestic political institutions and policy choices in the context of eight empirical regularities that constitute the democratic peace. We demonstrate that democratic leaders, when faced with war, are more inclined to shift extra resources into the war effort than are autocrats. This follows because the survival of political leaders with larger winning coalitions hinges on successful policy. The extra effort made by democrats provides a military advantage over autocrats. This makes democrats unattractive targets, since their institutional constraints cause them to mobilize resources for the war effort. In addition to trying harder, democrats are more selective in their choice of targets. Because defeat is more likely to lead to domestic replacement for democrats than for autocrats, democrats only initiate wars they expect to win. These two factors lead to the interaction between polities that is often referred to as the democratic peace.

I propose a new majoritarian bargaining model in which more than one implicit proposal can be on the table at the same time. Institutional differences from system to system affect the order of play, the equilibrium majorities, and the policy outcome. The ex post distribution of payoffs within a winning coalition is, however, invariant to fundamental institutional differences, and it is always proportional to the distribution of relative ex ante bargaining power. The bargaining process is modeled as a sequential demand game, in which players are called to propose a policy and to specify their desired share of the private benefits related to being in office. The order of play is endogenous, and the distribution of payoffs within an equilibrium majority usually does not depend on who is the proposal maker. The role of the head of state and the advantages to center parties are also studied.

Revisionist accounts conclude that divided and unified government do not differ significantly in the production of “important” public policy. I argue instead that when one theoretically reclaims the concerns about party responsiveness and institutional features of American politics that have animated party government scholars, unified government is significantly more productive than divided government. Employing a range of measures of important legislative enactments in the postwar period, I find that unified government produces greater quantities of significant enactments and is more responsive to the public mood than is divided government. The evidence suggests that parties do, as party government theorists maintain, generate incentives to cooperation that help transcend some of the policymaking gaps created by the Constitution.

American constitutional thought has long held that short, framework-oriented constitutions last longer than lengthy, statute-oriented constitutions. The longevity of the U.S. Constitution contributes heavily to this assumption. Not surprisingly, political scientists criticize state constitutions for their greater length and tendency to address issues better dealt with through ordinary statute law. These “defects” are frequently cited as responsible for the shorter lifespan of state constitutions. An examination of the 145 constitutions used by the American states since 1776, however, reveals a relationship among content, length, and durability that refutes the assumption that the design of the national constitution is necessarily superior. To the contrary, the analysis here reveals that longer and more detailed design of state constitutions actually enhances rather than reduces their longevity.

The conventional wisdom about negative political advertisements holds that no one likes them, but they work, that is, they have the consequences their sponsors intend. Moreover, many analysts have expressed concern over the detrimental effects of such negativism on the American political system. We examine the accuracy of the conventional wisdom and the legitimacy of the fears about the consequences for the political system via meta-analysis, a systematic, quantitative review of the literature. The data do not support either contention. Negative political ads appear to be no more effective than positive ads and do not seem to have especially detrimental effects on the political system. Eleven subsidiary hypotheses about particular circumstances in which significant effects are likely to be found are tested and rejected. Discussion focuses on why negative political advertisements have become so popular in practice when there is so little evidence that they work especially well.

Does negative campaigning influence the likelihood of voting in elections? Our study of U.S. Senate campaigns indicates the answer is “yes.” We find that people distinguish between useful negative information presented in an appropriate manner and irrelevant and harsh mudslinging. As the proportion of legitimate criticisms increases in campaigns, citizens become more likely to cast ballots. When campaigns degenerate into unsubstantiated and shrill attacks, voters tend to stay home. Finally, we find that individuals vary in their sensitivity to the tenor of campaigns. In particular, the tone is more consequential for independents, for those with less interest in politics, and for those with less knowledge about politics.

As political campaigns become increasingly adversarial, scholars are giving some much-needed attention to the effect of negative advertising on turnout. In a widely recognized Review article and subsequent book, Ansolabehere and his colleagues (1994, 1995) contend that attack advertising drives potential voters away from the polls. We dispute the generalizability of this claim outside the experimental setting. Using NES survey data as well as aggregate sources, we subject their research to rigorous real-world testing. The survey data directly contradict their findings, yielding no evidence of a turnout disadvantage for those who recollected negative presidential campaign advertising. In attempting to replicate Ansolabehere et al.'s earlier aggregate results we uncover quite substantial discrepancies and inconsistencies in their data set. We conclude that their aggregate study is deeply flawed and that Ansolabehere et al. exaggerated the demobilization dangers posed by attack advertising, at least in voters' own context.

Experiments show significant demobilizing and alienating effects of negative advertising. Although internally valid, experiments may have limited external validity. Aggregate and survey data offer two ways of providing external validation for experiments. We show that survey recall measures of advertising exposure suffer from problems of internal validity due to simultaneity and measurement error, which bias estimated effects of ad exposure. We provide valid estimates of the causal effects of ad exposure for the NES surveys using instrumental variables and find that negative advertising causes lower turnout in the NES data. We also provide a careful statistical analysis of aggregate turnout data from the 1992 Senate elections that Wattenberg and Brians (1999) recommend. These aggregate data confirm our original findings. Experiments, surveys, and aggregate data all point to the same conclusion: Negative advertising demobilizes voters.

Even though they contain one of the most forceful critiques of liberalism in contemporary political thought, the political writings of Stanley Fish have been neglected by political theorists. Fish's critique of liberal claims of moral and religious neutrality points to the conclusion that the liberal separation of church and state lacks a coherent justification. I offer a qualified defense of liberalism by arguing that while Fish's critique of liberal neutrality is sound, he fails to do justice to liberalism's substantive basis. Moreover, by simply negating liberalism, Fish's thinking remains within the liberal horizon in a way he fails to recognize.

Although J. Judd Owen's account of my work is on target in many ways, on some points he mischaracterizes my argument by making it claim or do too much. His misunderstandings flow in part from a conflating of two assertions: (1) that our convictions cannot be grounded in any independent source of authority and (2) that our convictions are ungrounded. I certainly assert the first but never the second. Rather, it is my contention that while we have no independent grounds—grounds implicated in no particular vision of life or comprehensive doctrine—we have, because we live within them, the grounds that are constitutive of our everyday lives, their practices and routines. What are the consequences of this argument? My answer is none whatsoever. If you are persuaded that no independent grounds are available, but that the grounds of your everyday practice are sufficient, you will feel neither disabled by what you do not need nor enabled by recognizing the nonindependent foundations you have always rested on and will continue to rest on. In the end, the most salient characteristic of my argument, a characteristic Owen resists, is its minimalism.

Efforts to replicate our study of the effects of politics on trade flows between the major powers have revealed that the computer program written to calculate the estimates produced errors in both the coefficients and the standard errors. Furthermore, these errors have some consequences for the results. In this brief corrigendum, we present corrected tables for our research note and a slightly modified interpretation of the results.

Our study examined evidence to assess three arguments about how politics may affect trade. The first contends that positive political relations in a dyad result in higher levels of trade. This argument leads to two hypotheses: States in conflict will have a lower level of trade than those not in conflict, and trade will increase with the degree of common interests between a pair of states. The second argument concludes that democratic dyads will have a higher level of trade than other dyads. The third argument focuses on the security concerns of states and leads to the hypotheses that alliances will increase trade in a bipolar system but will have no effect in a multipolar system. We refer readers to the original article for a fuller statement of the logic of these arguments, their respective hypotheses, and a complete statement of our research design and operationalization of the variables (Morrow, Siverson, and Tabares 1998).