Reining in a Rogue State: Instrument Selection in the U.S. Campaign to Confront Libya

Reining in a Rogue State: Instrument Selection in the U.S. Campaign to Confront Libya Paper prepared for the ISA Annual Convention 16 March 2011 Montreal, Quebec, Canada by Dianne R. Pfundstein PhD Candidate, Columbia University drp2109@columbia.edu PLEASE DO NOT CITE WITHOUT AUTHOR S PERMISSION Abstract: This paper examines why U.S. policymakers selected the instruments that they did to coerce Libya from 1972 through 2011. The United States employed diplomacy, economic sanctions, military strikes, and positive incentives in efforts to compel Libya to change its behavior during this period. Current studies of coercion in international politics tend to focus either on military coercion or economic sanctions in isolation, but fail to consider the fact that a state hoping to coerce a target must always weigh the costs and benefits of the range of instruments at its disposal, rather than evaluate a single tool in isolation. I argue that U.S. policymakers selected instruments to exert greater pressure on Libya as its violations of norms of international behavior increased over time, until they moderated U.S. goals in the early 1990s and began to offer Qaddafi positive incentives by 1999. I demonstrate that the ability to employ different instruments is also a function of the past relationship with the target. Finally, I demonstrate why studies that evaluate the conditions under which a state chooses a single instrument may lead to incorrect inferences about the relationship between coercer and target. D. Pfundstein 1 ISA 2011

Introduction: Why This Instrument? The demise of the Soviet Union and the rise of the United States as the world s sole superpower inaugurated an era in which the possibility of war between major states no longer dominates the international agenda. This does not mean that there has been no conflict in the post-cold War world. Instead, many of the most attention-grabbing actors have been so-called rogue states, marginalized by the major players in international politics but still capable of disrupting the international order. The United States has frequently taken the lead in confronting such badly behaving states, and can choose from an unmatched range of instruments with which to coerce its targets into changing their behavior. Some of the most active debates within the literature on international coercion center on the effectiveness of these instruments. The question of whether economic sanctions are effective in altering the behavior of target states, for example, has a great bearing both on our understanding of the exercise of power in international politics and on the choices that policy makers face in attempting to enforce order in international politics. Many studies of coercive effectiveness, and popular media accounts, suggest that coercive instruments rarely succeed, raising the question of why they are employed at all if their success rates are so uncertain. Just as important (and far less studied) as the question of whether an instrument is effective is the question of why one coercive instrument is chosen over another. Indeed, modeling why sanctions are chosen over diplomacy, or why bombing might be chosen over sanctions, is a necessary precondition for comparing the relative effectiveness of the various instruments of influence. Current studies of instrument selection tend to focus either on military coercion or economic D. Pfundstein 2 ISA 2011

sanctions in isolation, but fail to consider the fact that a state hoping to coerce a target must always weigh the costs and benefits of a range of instruments at its disposal, rather than evaluate a single tool in isolation. An examination of the coercive instruments that the United States employed against Libya over the past four decades yields many insights into the factors that affect instrument selection. Over the course of several presidential administrations, the United States employed diplomatic measures, economic sanctions, military strikes, and positive incentives to coerce Libya into changing its policies. I argue that U.S. policymakers chose instruments to exert greater pressure on Libya as its violations of norms of international behavior were perceived to increase over time, until they ran out of punitive options short of war, scaled back U.S. demands, and began to offer Qaddafi positive inducements by late 1999. I will also address the United States coercive options in light of the rebellion currently underway in Libya. Finally, I will demonstrate why studies that evaluate the conditions under which a state chooses a single instrument may lead to incorrect inferences about the relationship between coercer and target. The Theory of Coercion Coercion is fundamentally an exercise in persuasion. A coercer succeeds by convincing the adversary that she will suffer pain, or an increase in the current level of pain, if she chooses not to comply with the coercer s demands. As Schelling asserted, It is the threat of damage, or of more damage to come, that can make someone yield or comply. It is latent violence that can influence someone s choice. 1 Schelling distinguishes between deterrence, to turn aside or discourage through fear; hence, to 1 Thomas Schelling, Arms and Influence (New Haven: Yale University Press, 1966), 3. D. Pfundstein 3 ISA 2011

prevent from action by fear of consequences, 2 and compellence, which relies on a threat intended to make an adversary do something. 3 Brute force methods succeed by physically forcing the adversary to change her behavior, but the goal of interstate coercion is to persuade the opponent to stop or to undo encroachment instead of bludgeoning him into doing so or physically preventing him from continuing. 4 There are many different tools available to a state attempting to coerce a target into changing its behavior. David Baldwin describes a four-category taxonomy of instruments available to policy makers. The hierarchy begins with propaganda measures at the bottom, and then ascends to diplomacy, and then to economic statecraft, and finally to military statecraft. 5 Some of these instruments may coerce through the direct application of pain to the target state, while others may simply be threatened in order to induce fear of future pain in the target. In the case of economic sanctions, for example, the pain may be inflicted directly, and the coercer hopes that the specter of continued suffering will convince the target to accede to his demands. In the case of a threatened military strike, on the other hand, the pain is threatened but not necessarily inflicted. Both methods influence the target by generating an expectation of the future pain she will suffer for noncompliance. Baldwin argues that evaluating the use of a single instrument in isolation may provide misleading guidance for policymakers, since Policy making involves making decisions, and decision making involves choosing among alternative courses of action. 2 Ibid., 71. 3 Ibid., 69. 4 Alexander L. George, "Coercive Diplomacy: Definition and Characteristics," in The Limits of Coercive Diplomacy, 2 nd Edition, ed. Alexander L. George and William E. Simons, 7-11 (Boulder: Westview Press, 1994), 11. 5 David A. Baldwin, Economic Statecraft (Princeton: Princeton University Press, 1985) 13-14. D. Pfundstein 4 ISA 2011

The advantages and disadvantages of various policy options acquire significance primarily by comparison with other policy options. 6 Most studies of the effectiveness of individual coercive instruments, and the few studies that directly address the issue of instrument selection, ignore this crucial insight, but I will return to this point in the analysis below. Coercive Success There are many studies that identify factors contributing to successful coercion, but a full review of these studies is beyond the scope of this analysis. 7 The purpose of this study is to identify why U.S. policy makers chose the instruments that they did to confront Libya at various points in time. It would be natural to assume that policymakers would take into account the perceived effectiveness of different instruments when making their selection. But the effectiveness of an instrument alone is not enough: the relative costs of the various instruments should also influence the choice. Baldwin argues that, a successful policy choice is one that maximizes the utility of the policymaker in a given situation. From this standpoint, identification of a successful policy choice requires consideration not only of the costs and benefits of using economic sanctions, but also the costs and benefits of using alternative policy options. 8 In other words, an 6 Ibid., 15. 7 Some of the more comprehensive studies of coercion include: Schelling, Arms and Influence; Alexander L. George, David K. Hall and William E. Simons, The Limits of Coercive Diplomacy: Laos, Cuba, Vietnam (Boston: Little, Brown and Company, 1971); Barry M. Blechman and Stephen S. Kaplan, Force Without War (Washington, DC: Brookings Institution, 1978); Alexander L. George and Williams E. Simons, The Limits of Coercive Diplomacy, 2nd Edition (Boulder: Westview Press, 1994); Barry M. Blechman and Tamara Cofman Wittes, "Defining Moment: The Threat and Use of Force in American Foreign Policy," Political Science Quarterly 114, no. 1 (1999): 1-30; Daniel Byman and Matthew Waxman, The Dynamics of Coercion (Cambridge: Cambridge University Press, 2002); Robert J. Art and Patrick M. Cronin, The United States and Coercive Diplomacy (Washington, DC: United States Institute of Peace Press, 2003). 8 David A. Baldwin, "The Sanctions Debate and the Logic of Choice," International Security 24, no. 3 (Winter 1999-2000): 85. D. Pfundstein 5 ISA 2011

instrument is never chosen or evaluated in isolation, but instead weighed against the relative benefits and costs of alternative approaches. There are active debates on the effectiveness of both economic sanctions and aerial bombing, the instruments that generally receive the most attention in a study of the United States relationship with Libya. Because their perceived effectiveness influences policymakers use of these instruments, it is important to highlight the main points in these debates. Coercive Success: Economic Sanctions The most comprehensive work on economic sanctions analyzes 174 cases from 1914 to 2000 and finds that sanctions [are] at least partially successful in 34 percent of the cases that [they] documented. 9 The authors note, however, that the likelihood of success varies according to the goals sought by the imposer of sanctions: sanctions are more likely to be effective in cases in which the goals sought by the coercer are relatively modest. 10 The authors admit that, sanctions are of limited utility in achieving foreign policy goals that depend on compelling the target country to take actions it stoutly resists. 11 Finally, they note a decline over time in the effectiveness of sanctions imposed by the United States: U.S. sanctions were more effective in the early post-wwii period than in the final decades of the twentieth century. 12 A full review of the literature on the effectiveness of economic sanctions is beyond the scope of this inquiry. Some of the debate hinges on how we should measure 9 Gary Clyde Hufbauer, Jeffrey J. Schott, Kimberly Ann Elliott and Barbara Oegg, Economic Sanctions Reconsidered, Third Edition (Washington, DC: Peter G. Peterson Institute for International Economics, 2007), 158. The authors also describe several cases that occurred from 2000-2006, but do not include these cases in their statistical analysis (33). 10 Ibid., 163. 11 Ibid., 159. 12 Ibid., 128-129. D. Pfundstein 6 ISA 2011

and define successful economic sanctions. 13 Some studies seem to support the conclusion reached by Hufbauer et al, that sanctions can sometimes be effective in coercing a target state. 14 The impact of international cooperation is a factor commonly cited as having a large impact on the successful implementation of sanctions. Hufbauer et al find that having a large number of countries on board with a sanctions regime does not necessarily improve the chances for success, as securing an agreement among the sanctioning states can hurt the chances of success by diluting the scope and impact of the sanctions. 15 Daniel Drezner evaluates the mechanisms by which multilateral enforcement of a sanctions regime may fail, but he also finds that, multilateral sanctions that have the support of an international organization are significantly more effective than unilateral efforts. 16 Much of the literature is more skeptical of the effectiveness of economic sanctions as a coercive instrument. For example, Robert Pape reexamines the data from the second (1990) edition of Hufbauer et al and argues that only five of the 115 cases should count as successes for economic sanctions, rather than the forty that the authors identified in 13 For a good review of the debate on definitions of sanctions and measuring effectiveness, see: David A. Baldwin and Robert A. Pape, "Evaluating Economic Sanctions," International Security 23, no. 2 (1998): 189-198. 14 Some other works that suggest economic sanctions can be effective tools for achieving policy change include Kimberly Ann Elliott, "The Sanctions Glass: Half Full or Completely Empty?"," International Security 23, no. 1 (1998): 50-65; Nikolay Marinov, "Do Economic Sanctions Destabilize Country Leaders?," American Journal of Political Science 49, no. 3 (2005): 564-576: Marinov finds that a leader of a state experiencing economic sanctions is more likely to lose office in the following year than a leader not facing economic sanctions (565); David Lektzian and Mark Souva, "An Institutional Theory of Sanctions Onset and Success," The Journal of Conflict Resolution 51, no. 6 (2001): 848-871: the authors find that sanctions are more likely to succeed against a democratic target than against a nondemocratic target (849). 15 Hufbauer et al, Economic Sanctions Reconsidered, 172. It is also possible that multilateralism is not associated with success because multilateral sanctions are more likely to be employed in the more difficult cases of target compliance. 16 Daniel W. Drezner, "Bargaining, Enforcement, and Multilateral Sanctions: When Is Cooperation Counterproductive?," International Organization 54, no. 1 (2000): 97-98. D. Pfundstein 7 ISA 2011

1990. 17 Pape is most troubled by the cases that were counted as successes by HSE and in which military force was also employed since, as he argues, military conquest, when it occurs, is always a more credible explanation than economic sanctions because the target state s failure to concede before military defeat is in itself evidence of the failure of coercion. 18 He concludes that, there is little valid social science support for claims that economic sanctions can achieve major foreign policy goals. 19 Coercive Success: Bombing The United States also turned to military instruments in its attempts to coerce Libya. The debate on the effectiveness of air power as an independent instrument of coercion is even more heated than that surrounding the effectiveness of economic sanctions. In his comprehensive study of coercive air campaigns from 1917 through 1991, Robert Pape finds that strategies designed to inflict suffering on civilian populations (punishment and risk strategies in Pape s terms) are ineffective in coercing target states, as are decapitation campaigns designed to directly target a state s leadership and/or instruments of governmental control. Only denial strategies that directly target an opponent s military capacity and battlefield strategy can be effective, under some circumstances. 20 17 Robert A. Pape, "Why Economic Sanctions Do Not Work," International Security 22, no. 2 (1997): 93. 18 Ibid., 97. 19 Ibid., 106. Other works skeptical of economic sanctions include, for example, Richard Haass, "Economic Sanctions: Too Much of a Bad Thing," Brookings Policy Brief Series # 34 (Washington, DC: The Brookings Institution, 1998), http://www.brookings.edu/papers/ 1998/06sanctions_haass.aspx; Arne Tostensen and Beate Bull, "Are Smart Sanctions Feasible?," World Politics 54, no. 3 (2002): 373-403. 20 Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca: Cornell University Press, 1996), 314-315. D. Pfundstein 8 ISA 2011

Several other studies agree with Pape s conclusions about the limited effectiveness of strategic bombing. 21 Others suggest that air power and strategic bombing can be very effective tools in coercing target states. For example, in a response to Pape s work in Bombing to Win, John Warden III argues that the rise of precision munitions has had an immense impact on the way in which military operations are conducted, and that measures of the sheer physical damage inflicted by a bombing campaign fail to capture the interruption in target function that bombing can cause. 22 The use of unmanned aerial vehicles in the wars in Afghanistan and Iraq continues to raise questions about the utility of precision-strike air power for coercing target states. 23 Coercive Success: Carrots Although there are many studies evaluating the effectiveness of punitive instruments, there is relatively little work directly examining the use of positive incentives to coerce. 24 Schelling s work on coercion focuses on the threat of violence in reserve 25 to coerce targets, but George suggests that the use of positive inducements may be an effective means of inducing target compliance: What the threatened stick cannot 21 See, for example, Robert A. Pape, "The Limits of Precision-Guided Air Power," Security Studies 7, no. 2 (1997): 93-114; Daniel L. Byman and Matthew C. Waxman, "Kosovo and the Great Air Power Debate," International Security 24, no. 4 (2000): 5-38; Conrad C. Crane, "Sky High: Illusions of Air Power," The National Interest, no. 65 (Fall 2001): 116-122; Michael Horowitz and Dan Reiter, "When Does Aerial Bombing Work?: Quantitative Empirical Tests, 1917-1999," The Journal of Conflict Resolution 45, no. 2 (2001): 147-173; Daryl G. Press, "The Myth of Air Power in the Persian Gulf War and the Future of Warfare," International Security 26, no. 2 (2001): 5-44. 22 John A. III Warden, "Success in Modern War: A Response to Robert Pape's Bombing to Win," Security Studies 7, no. 2 (1997/98): 172-190. Other works arguing that air power is a potent coercive tool include Eliot A. Cohen, "The Mystique of U.S. Air Power," Foreign Affairs 73, no. 1 (1994): 109-124; Andrew L. Stigler, "A Clear Victory for Air Power," International Security 27, no. 3 (2002-2003): 124-157. 23 On the use of unmanned aerial vehicles (UAVs) and robots in modern warfare, see P. W. Singer, Wired For War: The Robotics Revolution and Conflict in the Twenty-first Century (New York: Penguin Press, 2009). 24 For a good review of the conceptual distinctions between positive and negative sanctions and the challenges of assessing the effectiveness of each, see: David A. Baldwin, "The Power of Positive Sanctions," World Politics 24, no. 1 (1971): 19-38. 25 Schelling, Arms and Influence, 143. D. Pfundstein 9 ISA 2011

achieve by itself, unless it is formidable, can possibly be achieved by combining it with a carrot. 26 Empirical work on positive incentives is limited. Wang finds that foreign aid can be used to induce states to vote according to U.S. preferences on major votes at the U.N. 27 In a review essay, Dorussen notes that several studies find considerable empirical support for the position that incentives can be important policy instruments whose effectiveness and efficiency are generally underrated. 28 Drennan s study of efforts to address North Korea s nuclear program suggests the Agreed Framework constituted a successful use of positive incentives in dealing with a troubling target state; 29 however, North Korea s subsequent withdrawal from the NPT and testing of a nuclear device suggest that it may have been closer to an example of blackmail than a successful use of positive incentives. Regardless of how one codes this particular case, the use of positive incentives for effecting policy change in target states remains an understudied phenomenon in international politics. Coercive Success: Why This Instrument? The United States employed a variety of instruments, including diplomacy, economic sanctions, aerial bombing, and even positive incentives, to try to coerce Libya over the past three decades. Empirical analyses of the effectiveness of these instruments suggest that they are effective only under narrow circumstances even Hufbauer et al, who are comparatively optimistic about the effectiveness of economic sanctions, find that 26 Alexander L. George, "Theory and Practice," in The Limits of Coercive Dipomacy, ed. Alexander L. George and William E. Simons, 13-21 (Boulder: Westview Press, 1994), 17. 27 T. Y. Wang, "U.S. Foreign Aid and UN Voting: An Analysis of Important Issues," International Studies Quarterly 43, no. 1 (1999): 208. 28 Han Dorussen, "Mixing Carrots with Sticks: Evaluating the Effectiveness of Positive Incentives," Journal of Peace Research 38, no. 2 (2001): 255. 29 William M. Drennan, "Nuclear Weapons and North Korea: Who's Coercing Whom?," in The United States and Coercive Diplomacy, ed. Robert J. Art and Patrick M. Cronin, 157-223 (Washington, DC: United States Institute of Peace Press, 2003), 196. D. Pfundstein 10 ISA 2011

sanctions succeed in only one third of the cases they examine. A survey of the literature on coercive instruments leaves one with the following question: if the effectiveness of these instruments is so questionable, why are they employed so frequently? Part of the answer to this question certainly lies in the general allure of coercion. Coercive diplomacy is an attractive strategy because it offers the defender a chance to achieve reasonable objectives in a crisis with less cost, with much less if any bloodshed, with fewer political and psychological costs, and often with less risk of unwanted escalation than is true with traditional military strategy. 30 Coercion is attractive because it provides the coercer with the chance to achieve its objectives more quickly, easily, and cheaply than seizing them by force. Yet this is true for coercive instruments in general, and cannot explain why or under what conditions policy makers would choose one instrument rather than another. In other words, why do policymakers sometimes employ economic sanctions when dealing with a misbehaving target state, but a military strike to deal with another? Why would a coercer employ different instruments against a single state at different points in time? There is much less work on the question of why states choose to implement coercive instruments than there is on the questions of the instruments effectiveness. Kaempfer and Lowenberg argue that the decision to impose economic sanctions against a target is a function of competition among interest groups within the coercing state, and that their purpose may be expressive rather than instrumental. 31 Michael Hiscox argues that voting on U.S. trade legislation is a function of the degree of factor mobility in the 30 George, Coercive Diplomacy: Definition and Characteristics, 9. 31 William H. Kaempfer and Anton D. Lowenberg, "The Theory of International Economic Sanctions: A Public Choice Approach," The American Economic Review 78, no. 4 (1988): 786. D. Pfundstein 11 ISA 2011

coercer s domestic economy. 32 These approaches suggest that the domestic politics of the sanctioning state determine whether and when sanctions are imposed, without reference to the goals the state is trying to achieve or to characteristics of the target state. Martin argues that, since unilateral sanctions should be largely ineffective in changing a target state s behavior, the key to understanding the implementation of economic sanctions lies in the mechanisms by which state are able to cooperate on a sanctions regime. 33 Drezner s rational choice model suggests that sanctions implementation is a function of expectations of future conflict between the coercer and the target state. 34 None of these studies addresses the question of interest, i.e., why a state would choose to implement one instrument rather than another rather, they all start from the assumption that the coercer has already zeroed in on sanctions as the preferred instrument and eliminated all other options for coercing the target state. Other studies focus on regime type as a key factor determining the decision to implement economic sanctions. The separate peace among democracies identified by Michael Doyle 35 suggests that democracies do not attack each other, but may be very war-like in their relations with non-democracies. The separate peace also seems to extend to economic attacks: in a multivariate analysis, Lektzian and Souva find that democracies are significantly less likely to use trade sanctions against another democracy. 36 In a study of 115 cases of sanctions from 1978-2000, Cox and Drury find that a democracy is more than five times more likely to sanction a non-democracy than to 32 Michael J. Hiscox, International Trade and Political Conflict (Princeton: Princeton University Press, 2002). 33 Lisa L. Martin, Coercive Cooperation: Explaining Multilateral Economic Sanctions (Princeton: Princeton University Press, 1992). 34 Daniel Drezner, The Sanctions Paradox (Cambridge: Cambridge University Press, 1999). 35 Michael W. Doyle, "Liberalism and World Politics," The American Political Science Review 80, no. 4 (1986): 1151-1169. 36 Lektzian and Souva, An Institutional Theory of Sanctions Onset and Success," 862. D. Pfundstein 12 ISA 2011

sanction another democracy. They also find that a democracy is twenty-four times more likely to implement economic sanctions than a non-democracy. 37 Letkzian and Sprecher find that the imposition of economic sanctions increases the likelihood of a militarized dispute in the future: sanctions tend to increase the probability of militarized conflict, particularly when used by democratic countries, because of their propensity to signal weakness while simultaneously tying the hands of democratic leaders. 38 These studies still do not answer the question of interest in the present study: why, and under what conditions, do policymakers choose one instrument over another? I am aware of only one large-n study that directly evaluates the decision to initiate economic sanctions. Drury examines sanctions that the United States implemented between 1966 and 1992, and he finds that the President is more likely to initiate sanctions as the intensity of an interstate dispute increases. He finds no relationship between the electoral cycle and sanctions initiation, but does find that an increase in the President s approval rating is associated with a marginal increase in the probability that the President will initiate sanctions. 39 Drury s analysis also yields a rather counterintuitive finding that reveals a larger flaw in his research design. He argues that, when the target state acts belligerently enough, it sends a signal to the U.S. that not only is the target unwilling to back down but also may be in the process of garnering public support for its resistance to the 37 Dan G. Cox and A. Cooper Drury, "Democratic Sanctions: Connecting the Democratic Peace and Economic Sanctions," Journal of Peace Research 43, no. 6 (2006): 716. Cox and Drury do not examine the possibility that this observation could be driven by the fact that democracies are generally wealthier than non-democracies, and thus many non-democracies may simply lack the capability to implement economic sanctions. 38 David J. Lektzian and Christopher M. Sprecher, "Sanctions, Signals, and Militarized Conflict," American Journal of Political Science 51, no. 2 (2007): 415. 39 A. Cooper Drury, "Sanctions as Coercive Diplomacy: The U.S. President's Decision to Initiate Economic Sanctions," Political Research Quarterly 54, no. 3 (2001): 500-503. D. Pfundstein 13 ISA 2011

sanctions. 40 Accordingly, Drury asserts that the president is deterred from using sanctions 41 when the target state is able to put on a show of belligerence; however, it is possible that the President does not impose economic sanctions in the face of an increasingly belligerent target because he decides to employ a more forceful means of influencing the target state. Rather than being deterred from taking further action, the President may decide to ascend to the fourth level of instruments described by Baldwin (1985): military action. Because he employs a dichotomous dependent variable, Drury s data set establishes a false choice between imposing economic sanctions and taking no action against the target state. It would be consistent with Drury s data for a state to escalate the dispute beyond economic sanctions when the target becomes increasingly belligerent, but such an effect could not be captured with this data structure. 42 Most studies of both military strikes and economic sanctions focus on whether they are effective, and do not evaluate the conditions under which a policymaker chooses one instrument over another. Those studies that do examine the decision to implement coercion consider the choice of a single instrument in isolation from the other options available to policymakers, and may therefore draw incorrect conclusions about the relationship between the coercer and the target state. When a state wishes to coerce a target, the choice is not necessarily between strategic bombing and no response, nor is it necessarily between economic sanctions and no response. Rather than evaluate a leader s decision as a choice between any single method of coercion and inaction, a more appropriate analysis should consider the range of available policy choices. Only by 40 Ibid., 501. 41 Ibid., emphasis added. 42 I am aware of no comparable large-n studies that evaluate the decision to employ aerial bombing against a target state. D. Pfundstein 14 ISA 2011

starting from the assumption that the whole range of coercive instruments would be available to a policymaker, and then determining the conditions that led her to eliminate other options in favor of the one chosen, can we have an accurate picture of the factors that contribute to instrument selection. Especially when we evaluate a coercive campaign that proceeded over several stages, it is essential to consider the range of options available rather than focusing on one instrument in isolation, or we risk missing key developments in the relationship. The U.S.-Libya Relationship: Selecting the Tools of Persuasion I will employ such an approach in my analysis of the U.S. campaign to coerce Libya over the past four decades. The goal of this study is to determine why U.S. policymakers chose to employ the specific instrument that they did at a given time. Contrary to formal models of sanctions initiation, which locate the decision solely in the domestic politics of the coercing state, I hypothesize that the choice of instrument will be driven by the goals that the United States wishes to achieve, and the perceived difficulty of attaining these goals. Furthermore, I hypothesize that policymakers will ascend through the four levels of instruments developed by Baldwin (1985) to choose ever-moreforceful instruments of coercion as time passes. I will begin with a review of the United States relationship with Libya since the 1970s, cataloguing the points at which U.S. policymakers chose to implement coercive instruments against Libya, and in particular, when they decided to switch from one instrument to another. I will then focus on a few of the major episodes of coercion and evaluate the justifications made by the policymakers themselves for the choice of policy. I will also consider the evolution of the relationship between the two countries and the D. Pfundstein 15 ISA 2011

strategic situation to offer some hypotheses about the ways in which U.S. policy evolved over time. Finally, I will demonstrate why this approach is superior to one that considers the question of implementing only a single coercive tool in isolation, and suggest that studies of this kind are a necessary building block for studies that could compare the relative effectiveness of different instruments. The 1970s: Diplomatic Disapproval, but Increasing Economic Ties The United States relationship with Libya has been complex since the state first gained independence by UN mandate in December 1951. 43 The United States and Great Britain both initially supported King Idris I, who ruled Libya from its independence until Muammar Qaddafi seized power in a 1969 coup. Qaddafi was a great admirer of Egyptian president Gamal Abdel Nasser, and when he assumed control of Libya, he did so with a doctrine combining pan-arabism with Islamic principles. As far as the United States and Soviet Union were concerned, Libya at this time embraced an official policy of positive neutrality, 44 and preferred to focus on unity among the states of the developing world rather than the superpower competition. Despite Qaddafi s criticism of both Cold War superpowers, Libya retained and further developed commercial ties with the West during this period, using the money from sales of oil to Europe and the United States to import consumer goods and technology. Libya would also increasingly purchase technology and arms from the Soviet regime. 45 At the beginning of the 1970s, the United States primary goal in its relationship with Libya was to block Soviet influence in North Africa and the Middle East. 43 For a more complete review of the United States relationship with Libya, see: Ronald Bruce St. John, Libya and the United States: Two Centuries of Strife (Philadelphia: University of Pennsylvania Press, 2002). 44 St. John, Libya and the United States, 101. 45 Ibid., 102-105. D. Pfundstein 16 ISA 2011

Consequently, the United States initially viewed the revolutionary government under Qaddafi as a means of thwarting Soviet interest in the region and as a way of ensuring continued access to Libyan oil, despite Qaddafi s termination of the lease for the Wheelus Air Force base soon after he seized power. Although Henry Kissinger raised the option of covert action against Qaddafi in response to the closure of the base, the Nixon administration concluded that it was more in the United States interest to refrain from taking hostile actions that could drive Qaddafi into the arms of the Soviets. 46 In fact, on three separate occasions from 1970-1972, the CIA aided Qaddafi in thwarting attempts to overthrow his government. 47 The U.S.-Libyan relationship began to sour in 1973 when Libya took an increasingly active role in the Arab-Israeli conflict, particularly with its participation in the planning of the 1973 War. Qaddafi also nationalized four American oil companies at this time, despite assurances when he came to power that American investments in Libyan oil were safe. Libya s role as a potential strategic partner in the Middle East became increasingly uncertain when Libya made its first major purchase of Soviet arms in mid-1974. 48 Although the Libya case does not appear in Hufbauer et al s economic sanctions dataset until 1978, when the United States officially banned the sale of military equipment to Libya, 49 the United States was already beginning to express its displeasure in the mid-1970s. Although these acts were probably not designed to coerce Libya into 46 Robert S. Litwak, Regime Change: U. S. Strategy through the Prism of 9/11 (Washington, D.C.: Woodrow Wilson Center Press, 2007), 171. 47 St. John, Libya and the United States, 104-105. 48 Ibid., 104-107. 49 Hufbauer et al, Economic Sanctions Reconsidered, 25. The history for case 78-8 appears on the supplementary CD-ROM, Case Histories and Data. The full sanctions regime is coded as lasting from 1978-2004, and a separate case is listed for UN sanctions against Libya (1992-2003) in response to the Lockerbie bombing (30). D. Pfundstein 17 ISA 2011

making any many policy changes, they were intended to signal the United States growing disapproval of Libyan behavior. 50 These demonstrations of the United State s disapproval during the early 1970s consisted of diplomatic slaps on the wrist and minor interference in the sale of military equipment to Libya. When Ambassador Palmer resigned from his post in Tripoli in late 1972, Washington chose not to appoint a successor, and instead conducted negotiations at the chargé d affaires level until the embassy was closed in 1980. In 1973, the United States blocked delivery to Libya of eight airplanes for which it had paid, and it denied Libya additional purchases of weapons and other military systems in 1974 and 1975. 51 Relations between the United States and Libya further soured in 1976 when Qaddafi was reported to have solicited attacks on the 1976 Republican and Democratic national conventions; in 1976 the Ford administration then promised Cairo to deter Soviet intervention in the event Egypt launched a military attack on Libya. 52 In mid- 1976 the administration publicly announced that it believed Libya was supporting international terrorism, and a Department of Defense report in January of the following year named Libya as the United States fourth-leading enemy. Despite Libya s hopes that a new administration would adopt a more friendly policy, in 1977 the Carter administration blocked the transfer of Italian planes to Libya, sided with Egypt in a dispute with Libya in July, and continued to refuse to sell military equipment to Libya while announcing a $200 million sale to Egypt. Qaddafi s signing of three UN 50 St. John, Libya and the United States, 108. 51 Ibid. 52 Ibid., 109. D. Pfundstein 18 ISA 2011

conventions on terrorism in 1978 did little to alter the U.S. administration s perception of Qaddafi as an international menace and sponsor of terrorism. 53 The United States views of the Libyan regime further crystallized in 1979 when Libyan demonstrators burned the U.S. embassy in Tripoli in a show of support for the hostage takers in Tehran, an attack that could have occurred only with the tacit approval of the Qaddafi regime. After the French embassy was similarly attacked on 4 February 1980, the United States officially closed its embassy. 54 Yet, despite this hardening of diplomatic relations with Libya throughout the 1970s and a growing belief that Qaddafi was sponsoring international acts of terrorism, economic ties were not yet severed. Although the sale of some planes and military equipment to Libya was blocked by 1978, Libyan exports to the United States actually increased by a factor of ten between 1973 and 1976, and American exports to Libya more than doubled during the period. From 1977 until 1980, the United States was also the single largest purchaser of Libyan oil. Although the Carter administration did not release the planes that had been blocked from transfer in 1973, the State Department permitted the sale of two Boeing 727s to Libyan Arab Airlines in November 1978, under the condition that they be used solely for civilian or commercial operations. 55 St. John argues that the encouragement of U.S.-Libyan trade through the 1970s was undertaken in the hope that, the promotion of commercial relations with Libya would encourage a more constructive dialogue on political issues. 56 53 Ibid., 108-112. 54 Ibid., 110-114. In May the United States expelled individuals from the Libyan People s Bureau in Washington who were believed to be harassing anti-qaddafi Libyan students in the United States. The United States would later link these individuals to schemes that killed Libyan dissidents across Europe in the summer of 1980. 55 Ibid., 109-111. In March 1979 the US administration approved the sale of three Boeing 747s, but later reversed this decision when Qaddafi used one of the 727 s to evacuate his troops from Uganda. 56 Ibid., 111. D. Pfundstein 19 ISA 2011

The fact that the United States became the single largest buyer of Libyan oil during the 1970s suggests that policymakers were unwilling to impose trade restrictions that would have threatened the U.S. economy, particularly in the wake of the oil crisis earlier in the decade. Refusing to reappoint an official ambassador, blocking the sale of airplanes to Libya, and closing the U.S. embassy were not acts that could have compelled Libya to change its political behavior, and nor could the refusal to sell military equipment have denied Libya the ability to extend its military power, since Libya was able to purchase arms from the Soviet Union. These actions were instead intended to convey disapproval, but the United States also dangled the carrot of economic relations in front of Qaddafi in the hope that it would draw the unpredictable leader into a better relationship with the United States. This carrot would prove insufficient to alter Libya s behavior. It was not until the U.S. and Libyan militaries clashed in the early 1980s and Qaddafi s support of terrorism seemed to escalate that the United States abandoned carrots in favor of an approach that relied mainly on sticks. 1980-1985: Unilateral Economic Sanctions In 1979, Qaddafi declared that the entire Gulf of Sidra was part of Libya s territorial waters, even though the gulf extended far deeper into Libya s landmass than twelve miles, the limit on territorial waters set by international law. The gulf s location far from shipping lanes had made it one of the few locations in the region in which the U.S. Sixth Fleet conducted live-fire exercises, and thus Qaddafi s claim to the gulf was perceived as an affront to U.S. military power. Qaddafi threatened to use force to defend the waters, but the United States did not immediately challenge his claim, and prohibited the Sixth Fleet from crossing into the Gulf as marked out by Qaddafi. Libyan MiG D. Pfundstein 20 ISA 2011

fighters harassed and fired on an unarmed U.S. reconnaissance flight later in 1979, but the shots missed their targets and the United States did not respond. 57 When Reagan took office in January of 1981, he was determined to take a stronger stance against terrorism than the preceding administration. The failed rescue of the hostages in Tehran, and their subsequent release upon Reagan s inauguration, steeled the administration to take a tougher stand against terrorists. 58 In March of 1981, Secretary of State Alexander Haig, Jr., publicly claimed that Libya was supporting terrorist training camps and closed the Libyan diplomatic mission in Washington in May. 59 In August of 1981, the Reagan administration reversed previous policy and permitted the navy to enter the Gulf of Sidra. During a confrontation with SU-22 fighters from the Libyan air force, two US F-14 Tomcats were fired upon. They managed to evade the missiles and downed both of the Libyan aircraft. 60 Although they claimed that the confrontation was undertaken to safeguard navigation in international waters, the Reagan administration also hoped that the display of military might would deter future terrorist attacks or generate internal dissent sufficient to mount a coup against the Libyan leader. 61 Although Qaddafi did not issue any further overt challenges to the U.S. military in this arena, in October 1981 the United States imposed export controls on various types of aircraft and aircraft parts, in an effort to limit Libya s ability to support military 57 Peter Huchthausen, America's Splendid Little Wars: A Short History of U.S. Military Engagement: 1975-2000 (New York: Viking, 2003), 89. 58 David B. Cohen, "Revisiting El Dorado Canyon: Terrorism, the Reagan Administration, and the 1986 Bombing of Libya," White House Studies 5, no. 2 (2005): 162. 59 Hufbauer et al, Case 78-8, US v. Libya, case history on supplemental CD-ROM for Economic Sanctions Reconsidered, 1. 60 Huchthausen, America s Splendid Little Wars, 90-91. 61 Litwak, Regime Change, 173. D. Pfundstein 21 ISA 2011

interventions in other African states. 62 Yet, Qaddafi did not remain dormant: some observers claimed Libyan involvement in the December 6 murder of a U.S. defense attaché in Paris, and on the following day Reagan claimed that the U.S. government had uncovered plots by Qaddafi to assassinate U.S. officials. On 10 March 1982, drawing on section 232 of the Trade Expansion Act of 1962, the Reagan administration instituted an embargo on crude oil imports from Libya. Later in the year, the CIA also assisted Libyan exiles in mounting opposition to the Libyan occupation of Chad, with the hope that such activities might eventually lead to Qaddafi s fall from power. This was part of a limited pattern of attempted covert operations directed against Qaddafi during the early 1980s, a pattern to which I will return below. 63 In an extension of the 1982 embargo, the United States banned the import of refined petroleum products from Libya on 15 November 1985. 64 Nevertheless, this unilateral embargo on Libyan oil could have only limited coercive impact as long as other states were willing to purchase Libyan oil a willingness that would survive until Libya s support for terrorism garnered the attention of the United States main European allies. Libya s support for international terrorism shifted into high gear in 1985. In June, TWA flight 847 was hijacked and an American sailor on the plane was murdered. 65 In October, terrorists seized a cruise ship and killed an American passenger. Later that month, a shopping mall at a U.S. Army base in Frankfurt was bombed, wounding 23 62 Hufbauer et al, Case 78-8, US v. Libya, 1. 63 Litwak, Regime Change, 174. Litwak argues that the limited scope of covert operations posed little immediate threat to Qaddafi s regime, but were viewed as a means of augmenting the pressure of unilateral economic sanctions. 64 Hufbauer et al, Case 78-8, US v. Libya, 2. 65 The experience of managing this hostage standoff seems to have played a large role in shaping Secretary of State George Shultz s later approach to handling terrorism. Throughout the crisis, Shultz insisted that the United States could not make an open deal to trade the hostages in exchange for the release of Israeli prisoners. See George P. Shultz, Turmoil and Triumph: My Years as Secretary of State (New York: Charles Scribner's Sons, 1993), 653-668. D. Pfundstein 22 ISA 2011

Americans. In November, hijackers forced a plane to land in Egypt and killed an American on board; the Egyptian raid to recover the aircraft killed fifty-eight people and all but one hijacker. 66 In December, coordinated attacks in the Rome and Vienna airports by a Palestinian faction allegedly backed by Qaddafi killed 19 and wounded 110. On 7 January 1986, Reagan invoked the International Emergency Economic Powers Act to bar the export and import of most goods to and from Libya, technology transfers, and many financial transactions, and he ordered Americans to leave the country. The next day, Reagan froze the Libyan government s assets in the United States. In February, the Reagan administration revised these sanctions to permit U.S. oil companies to continue their operations in Libya, in order to prevent the Libyan government from profiting from the oil companies absence. 67 Reagan cited Libya s support for terrorist activity as the main reason for declaring the state of emergency and imposing the harsh sanctions that he had declined to initiate in 1983. Cohen notes that, Officially, the administration had the view that economic sanctions would deter Libya from any further support of terrorism. Unofficially, sanctions appeared to be a formality opening the door for a military response. 68 This suggests that the choice of economic sanctions was viewed as a half-measure between military action and inaction, and also suggests that the administration was hoping to build support for later military action by attempting to coerce Libya through non-military means. Thus, in addition to their coercive properties, economic sanctions played a useful political role: The imposition of sanctions was a necessary step before the use of force 66 Huchthausen, America s Splendid Little Wars, 91. 67 Hufbauer et al, Case 78-8, US v. Libya, 2-3. 68 Cohen, Revisiting El Dorado Canyon, 167. D. Pfundstein 23 ISA 2011

should the Qaddafi regime s sponsorship of terrorism not cease. 69 Reagan also cited concern for U.S. citizens on the ground in Libya: he noted that the sanctions allowed time for Americans to leave the country before military action, thereby denying Qaddafi the ability to seize them as hostages. 70 1986: Military Clashes and the U.S. Bombing Raid The conflict between the United States and Libya erupted into military confrontation once more in March 1986, when the U.S. Sixth Fleet again challenged Qaddafi s claim to the Gulf of Sidra. On March 24, after two days of inaction, Libya fired two surface-to-air missiles at two US F-14 fighters and missed by wide margins. The rules of engagement permitted U.S. forces to return fire: two navy aircraft attacked and destroyed two Libyan air-defense radar sites, while other forces damaged a Libyan patrol ship. The Sixth Fleet then withdrew on March 27. 71 Zimmermann argues that granting authorization for the U.S. military to cross into the gulf was a deliberate attempt to provoke Qaddafi into combat, a way of demonstrating U.S. resolve both to Libya and to its allies, and a means of testing and wearing down Libya s antiaircraft capabilities in anticipation of a future attack. 72 He also argues that the Reagan administration hoped that any military defeat might have driven the Libyan armed forces to overthrow Qaddafi. 73 Litwak argues that the military engagement was a further attempt by the Reagan administration to deter Libya from sponsoring terrorist acts, and a signal to Qaddafi that the United States would not fail to 69 Litwak, Regime Change, 174. 70 Ronald Reagan, An American Life (New York: Simon & Schuster, 1990), 511. 71 Huchthausen, America s Splendid Little Wars, 93. 72 Tim Zimmermann, "Coercive Diplomacy and Libya," in The Limits of Coercive Diplomacy, ed. Alexander L. George and William E. Simons, Second Edition (Boulder, CO: Westview Press, 1994), 209. 73 Ibid., 212. D. Pfundstein 24 ISA 2011