How Public Opinion Constrains the Use of Force: The Case of Operation Restore Hope

Article excerpt

At some level, the political decisions preempt in these kinds of operations.... Everything that's done is political.--General Joseph Hoar, USMC (ret.) Former CINCCENT (1)

In early 1991, Somalia fell into a state of civil war, precipitating a catastrophic famine. (2) Almost two years later, in late November 1992, following a year in which the United States first resisted intervening altogether and then did so only relatively modestly, President George H. W. Bush decided to launch a large-scale, American-led military intervention termed Operation Restore Hope. Less than a year later, amid rapidly deteriorating public and congressional support for the mission, President Bill Clinton announced his intent to end U.S. involvement in Somalia.

A review of the myriad studies of the U.S.-led humanitarian intervention reveals numerous proposed explanations for the decision making of Presidents Bush and Clinton, including the rapid turnabout in the Bush administration's attitude toward intervention. While typically presented in the unique context of the Somalia case, most such explanations are merely case-specific variants of several of the most widely employed variables in the theoretical literature on the domestic sources of foreign policy. Such factors include bureaucratic politics and organizational routines (Allison 1969; Sagan 1993; Zegart 1999), the individual preferences, values, and characteristics of leaders (Larson 1983; March and Olsen 1998), electoral incentives (Gaubatz 1991; Smith 1996), and public preferences (Zaller 1994; Ostrom and Job 1986; Powlick 1995; Sobel 2001), including, in the last instance, the constraining effect of public opinion (Holsti 1996; Sobel 1993, 2001; Rosenau 1961; Powlick 1995). I argue that with one exception, these explanations provide only partial insight into the full range of U.S. policy decision making with respect to Somalia between 1992 and 1994. The exception is the constraining role of public opinion.

In this respect, the Somalia intervention exemplifies a more general pattern in presidential decision making. In this study, I develop a theory to explain the circumstances under which public scrutiny is likely to influence presidents in their foreign policy decision making. My argument refines previous theories concerning the influence of domestic political risks on crisis outcomes, allowing more nuanced predictions concerning conditions under which presidents are likely to be willing to use military force abroad. I argue that, unless a president is highly confident of success, an attentive public can, when the strategic stakes are relatively modest, inhibit him from undertaking risky foreign policy initiatives, including using military force. The political costs of failing are greater if the public is attentive because it will be more inclined to punish a president if it is watching closely when he falters than if it is uninterested or distracted. Moreover, this pattern is self-reinforcing. All else equal, a policy fashioned under intense public scrutiny is more likely to fail, because public scrutiny reduces the president's freedom to develop an optimal foreign policy, free from domestic pressure to compromise. This further raises the political risks associated with an attentive public. And when the strategic stakes in a foreign crisis are modest, a president is likely to weigh more heavily the potential political risks associated with a given policy, thereby making public attentiveness a potential constraining factor. Hence, unless they are quite confident of success, presidents are more likely to use force in low-stakes foreign crises when the public is inattentive. (See Baum n.d. for a game theoretic extension of the theory and large-N statistical testing.)

After explicating the theory, I illustrate some of its key elements through a case study of Operation Restore Hope, the U.S.-led multinational humanitarian relief operation in Somalia that spanned the G. …


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.