My argument builds on the standard rationalist arguments by showing why states fail to use costly signals, limited force, and enforcement mechanisms when they could. I draw from a small and often overlooked set of rationalist models that illuminate the paradox that threatening to use force can be more costly than using it. These “costly peace” models show that coercion can be expensive when states confront rivals who are either highly resolved to resist30 or must be coerced indefinitely to comply.31 The costs of peace can cause policymakers to prefer war.
What these models do not clarify, however, is when the strong will see coercing the weak as more costly than war. Weak leaders may be relatively cheap to defeat, but because of this weakness, they should also be relatively cheap to coerce, even indefinitely. At the same time, leaders who are highly resolved to fight may be costly to coerce, but they should also be costly to defeat because their resolve makes them willing to fight longer and harder.32 What we lack is a theory that can explain why a leader would be committed to resist (and thus costly to coerce) but, at the same time, comparatively cheap to defeat. In this book, I show that a leader’s domestic opposition can make the difference, making the leader at once more resolved to resist and more vulnerable to regime change.
Defining Regime Change
FIRC can take a variety of forms. Some cases involve large-scale military invasions, such as those undertaken by the Allies in Germany and Japan during World War II and by the Soviet Union in Afghanistan in 1979. Others entail support for insurgents, much like the 1909 American intervention in Nicaragua, American support for the Nicaraguan Contras during President Ronald Reagan’s administration, or NATO’s overthrow of Qaddafi in 2011. FIRC can also include foreign-orchestrated coups such as the Soviet-sponsored communist takeovers in Czechoslovakia and Poland or US covert operations in Albania in 1949, Iran in 1953, and Chile in 1973. Paradoxically, FIRC can also occur through a negotiated agreement. In 1994, for example, President Bill Clinton threatened an invasion to convince Haiti’s military junta to step down.
These cases may not appear on the surface to be alike, but they share at least three important traits. I use these traits as conditions to identify cases of FIRC for this study. First, policymakers in the imposing state have explicitly abandoned negotiations that would allow the target state’s leader(s) to remain in power. Second, policymakers in the imposing state have implemented a specific plan to replace the target state’s leaders and/or political institutions. Third, policymakers target a sovereign state and intend to restore sovereignty to that state. I explain each of these conditions below.
First, to qualify as an instance of FIRC, policymakers in the imposing state must reject any bargaining agreement with the targeted head of state that would leave that leader in power. This distinction makes it possible to distinguish between cases in which the stronger power’s primary aim is to change the opposing side’s policies and those in which the goal is to change its policymakers. This is an important distinction because countries can threaten regime change when it is not their primary goal. In the 1980s, for example, the Central Intelligence Agency (CIA) made plans to topple Qaddafi, but President Reagan made clear that if Qaddafi abandoned his support for terrorism, the United States would abandon regime change.33 Reagan’s main goal was a change in Libya’s policies, not necessarily in its top policymaker. States may also sponsor rebels to coerce policy change. The United States and Iran backed a Kurdish rebellion in Iraq to pressure Saddam Hussein into signing the 1975 Algiers Agreement. To identify sincere threats to impose regime change, I look for evidence that the imposing state refused to abandon regime change in exchange for policy change.
States may also demand regime change as part of a negotiated settlement, as the Clinton administration did in Haiti. Although states employ coercive bargaining tactics to attain these settlements, these cases qualify as FIRC. The foreign power is determined to oust the leader rather than accept an agreement that would let the leader remain in power. To exclude these cases would be to eliminate from the data instances in which the foreign power shares the same goal as other states seeking regime change but adopts different methods. I distinguish instances of coerced regime change by looking for evidence that the imposing state explicitly demanded that the targeted leader step down. The settlement ending the Franco-Prussian War, for example, obligated France to hold new elections. In contrast, Germany’s 1940 invasion of France, which coincided with a change in French leadership, is not included because France’s Vichy regime assumed power before seeking an armistice.34
In addition to the methods used to impose regime change, the extent of political change can vary as well. States may either remove the target state’s top leaders, or they may transform its political institutions. Some studies exclude leadership change from the definition of regime change because the target state’s institutions are left intact and, therefore, its regime, traditionally defined as a set of political institutions, does not change. However, this distinction conflates how regime change ends with how it is pursued. Foreign powers sometimes depose leaders to produce institutional change. The United States, for example, helped facilitate Iran’s transition from a constitutional monarchy to an authoritarian state through leadership change in 1953. By removing that country’s popular prime minister, Mohammad Mosaddeq, and supporting the Shah, the United States obtained institutional change, but it did not impose institutions. In Chapter 2, I discuss the various ways in which states effect regime change, including what level of force they use and whether they transform the target state’s institutions or simply depose its leaders. The data used in Chapter 3 include both types of FIRC.
The second criterion for regime change requires that policymakers implement a plan explicitly aimed at replacing the target state’s leaders or political institutions. By requiring that the initiator implements a plan of action, I can exclude cases in which policymakers call for regime change largely for symbolic or domestic political reasons. After the 1979 revolution in Iran, Ayatollah Ruhollah Khomeini called for regime change in Iraq and encouraged Iraqi Shi’a to revolt, but the Iranian government did not pursue a direct course of action aimed at removing Saddam Hussein from power. After Iraq invaded Iran in 1980, however, Iran counterinvaded and actively pursued regime change in Iraq.35 As this case also shows, attempts to pursue regime change need not be successful. In my data, I include cases in which states eschewed bargaining and pursued regime change but ultimately abandoned that aim. These cases are important to include because the conditions that cause states to pursue regime change should apply regardless of whether their attempts succeed or fail.
Finally, for a case to qualify as an instance of FIRC, there must be evidence that the attempt to overthrow the leader or regime was driven at least as much by foreign actors as by domestic ones. To make this distinction, I use evidence that the foreign power sent its own personnel to the target state, provided military aid that the opposition needed in order to act, or assumed a controlling stake in the operation.36 For example, the United States’ effort to assist the White movement in Russia following the communist revolution involved thousands of American troops and thus is included as a regime-change attempt.37
The third and final condition requires that the initiator intends to restore sovereignty to the target state. In contrast to annexation or colonization, in which the target state or territory remains under the direct control of the intervening state, regime change involves the installation of a new or restored sovereign government. I identify cases of regime change by looking for evidence that the foreign power was not planning to exercise permanent control over the target state. The foreign power may rule its target temporarily, but it must at least plan to restore sovereignty. Nazi Germany, for example, ruled the Netherlands, Belgium, and Denmark (after 1943) through military governors but did not plan to annex or dissolve these states, as it did Czechoslovakia, Austria, and Poland.38 The former qualify as regime-change events, the latter do not. In some instances, a third state may help an ally annex another. The United States, for example, attempted to overthrow the North Korean government to bring it under the jurisdiction of the South. In these cases, I code