The 1934 Reciprocal Trade Agreements Act (RTAA) changed the structure of the making of U.S. trade policy and made possible a dramatic reduction in tariffs. The authors demonstrate that the key institutional innovation in the RTAA was its mandate to lower tariffs through reciprocal agreements with foreign nations. The expansion of exports under the RTAA enhanced political support for increasingly lower U.S. tariffs. Evidence that export interests were positively associated with congressional votes for free trade supports this view.

Governing parties face two fundamental tasks: they must pursue policies effectively, and they must win elections. Their national coalitions, therefore, generally include two types of constituencies—those that are important for policy-making and those that make it possible to win elections. In effect, governing parties must bring together a policy coalition and an electoral coalition. This distinction sheds light on how the transitional costs of major economic policy shifts can be made sustainable in electoral terms. It also provides a starting point for analysis of how two of Latin America's most important labor-based parties, the Peronist party in Argentina and the Partido Revolucionario Institucional (PRI) in Mexico, maintained electoral dominance while pursuing free-market reforms that adversely affected key social constituencies. Peronism and the PRI are conceived of as having encompassed historically two distinctive and regionally based subcoalitions: a metropolitan coalition that gave support to the parties' development strategies and a peripheral coalition that carried the burden of generating electoral majorities. This framework permits a reconceptualization of the historic coalitional dynamics of Peronism and the PRI and sheds light on the current process of coalitional change and economic reform.

When one state engages in a military buildup, other states sometimes take this as a sign that it is more aggressive or expansionist than they previously thought. Some argue that such increases in mutual suspicion can drive arms races and even lead to war. Psychological bias is often invoked to explain this pattern of growing suspicions leading to hostility. This article presents an incomplete information model of an arms race and investigates when escalations should rationally generate increased fears and when, in order to reduce such fears, security seekers can refrain from building. It shows that escalations rationally provoke fear even in the absence of bias and that weak states and states facing high costs of arms racing and war will be especially likely to refrain from building as a way of signaling benign intentions.

Practically everywhere one looks these days the concept of “civil society” is in vogue. Neo-Tocquevillean scholars argue that civil society plays a role in driving political, social, and even economic outcomes. This new conventional wisdom, however, is flawed. It is simply not true that democratic government is always strengthened, not weakened, when it faces a vigorous civil society. This essay shows how a robust civil society helped scuttle the twentieth century's most critical democratic experiment, Weimar Germany. An important implication of this analysis is that under certain circumstances associationism and the prospects for democratic stability can actually be inversely related. To know when civil society activity will take on oppositional or even antidemocratic tendencies, one needs to ground one's analyses in concrete examinations of political reality. Political scientists should remember that Tocqueville considered Americans' political associations to be as important as their nonpolitical ones, and they should therefore examine more closely the connections between the two under various conditions.

The recent trend toward democratization in countries across the globe has challenged scholars to pursue two potentially contradictory goals. On the one hand, they seek to increase analytic differentiation in order to capture the diverse forms of democracy that have emerged. On the other hand, they are concerned with conceptual validity. Specifically, they seek to avoid the problem of conceptual stretching that arises when the concept of democracy is applied to cases for which, by relevant scholarly standards, it is not appropriate. This article argues that the pursuit of these two goals has led to a proliferation of conceptual innovations, including numerous subtypes of democracy—that is to say, democracy “with adjectives.” The article explores the strengths and weaknesses of alternative strategies of conceptual innovation that have emerged: descending and climbing Sartori's ladder of generality, generating “diminished” subtypes of democracy, “precising” the definition of democracy by adding defining attributes, and shifting the overarching concept with which democracy is associated. The goal of the analysis is to make more comprehensible the complex structure of these strategies, as well as to explore trade-offs among the strategies. Even when scholars proceed intuitively, rather than self-consciously, they tend to operate within this structure. Yet it is far more desirable for them to do so selfconsciously, with a full awareness of these trade-offs.