Do democracies make more effective coercive threats? An influential literature in international relations argues that democratic institutions allow leaders to credibly signal their resolve in crises, thereby making their threats more likely to work than threats by nondemocracies+ This article revisits the quantitative evidence for this proposition, which we call the "democratic credibility hypothesis," and finds that it is surprisingly weak+ Close examination of the data sets most commonly used to test this hypothesis reveals that they contain few successful democratic threats, or indeed threats of any kind+ Moreover, these data sets' outcome variables do not properly measure the effectiveness of threats, and therefore yield misleading results+ The article then reassesses the democratic credibility hypothesis using the Militarized Compellent Threats data set, a new data set designed specifically to test hypotheses about the effectiveness of coercive threats+ The analysis indicates that threats from democracies are no more successful than threats from other states+ It is commonly argued in international relations scholarship that coercive threats issued by democracies are more successful than those issued by nondemocracies+ 1 According to this view, democratic leaders who issue threats in crises and thenThe order of the authors is alphabetical and reflects equal contributions+ For helpful comments on previous drafts we thank Dale

Recent research suggests that a crucial factor in understanding the outcomes of military conflicts is the extent to which militaries are mechanized-that is, their relative dependence on tanks and armored vehicles compared to manpower. Since World War II, militaries have become increasingly mechanized both among the great powers and in unexpected quarters of the developing world. Yet the extent of military mechanization varies widely across states. Why have some states adopted highly mechanized force structures, while others have not? This paper tests several hypotheses about the determinants of military mechanization. One perspective suggests that strategic factors-including the force structures of adversaries and neighbors, recent combat lessons, and internal insurgency threats-shape a military's mix of manpower and vehicles. A second set of hypotheses points instead to domestic institutions such as democracy and civilian control of the military. Still other theories emphasize economic forces and international norms. To test these and other hypotheses, we construct a new data set containing mechanization rates for more than 150 militaries from 1979 to 2001. Broadly, we find significant support for the strategic perspective and little support for domestic institutional explanations. In addition, the results suggest systematic and predictable differences in the ways states structure their militaries in response to security pressures.Often overlooked in the political debate surrounding the 2007 ''surge'' of US troops into Iraq was the shift in military doctrine that accompanied it. After nearly four years of struggle in Iraq, the US military began to abandon its Authors' notes: The order of the authors is random and reflects equal contributions. For helpful comments on previous drafts, the authors thank

How can states signal their alliance commitments? Although scholars have developed sophisticated theoretical models of costly signaling in international relations, we know little about which specific policies leaders can implement to signal their commitments. This article addresses this question with respect to the extended deterrent effects of nuclear weapons. Can nuclear states deter attacks against their friends by simply announcing their defense commitments? Or must they deploy nuclear weapons on a protégé's territory before an alliance is seen as credible? Using a new dataset on foreign nuclear deployments from 1950 to 2000, our analysis reveals two main findings. First, formal alliances with nuclear states appear to carry significant deterrence benefits. Second, however, stationing nuclear weapons on a protégé's territory does not bolster these effects. The analysis yields new insights about the dynamics of "hand-tying" and "sunk cost" signals in international politics.

States typically issue compellent threats against considerably weaker adversaries, yet their threats often fail. Why? Expanding on a standard model of international crisis bargaining, I argue that a theory of reputation-building can help shed light on this puzzle. The model casts reputation as a strategic problem, showing that challengers issuing compellent threats have incentives to anticipate the reputation costs that target states incur when appeasing aggressors. If challengers can recognize these costs and offset them with side payments or smaller demands, then even reputation-conscious targets will acquiesce. I argue, however, that military strength contributes to information problems that make challengers more likely to underestimate their targets' reputation costs and insufficiently compensate them. In this way, military power can undermine the effectiveness of compellent threats. The logic is illustrated by the 1939 Russo-Finnish crisis, and the argument's implications for the study of coercive diplomacy are explored.

The dominant (though contested) wisdom among international relations scholars is that military officers tend to be more cautious than their civilian counterparts about initiating the use of force. Sobered by the experience of combat, the theory holds, soldiers are hesitant to recommend military action except under the most favorable of circumstances. It might be the case, however, that military conservatism is simply a product of strong civilian oversight. Indeed, scholars have suggested that military officers actually have powerful incentives to promote the use of force, but these predilections may be muted when civilian leaders can punish officers for botched military adventures. In this article, the author details a quantitative, competitive test of these propositions, showing that states with strong civilian control are on average less prone to initiate military action than states without it. The results suggest that civilian control should play a central role in future models of conflict initiation.