- - º A’ATHIFICATION PROGRAM IRAO’S FUTURE AND ARAB REVOLUTIONS ſeW Terrill U.S. ARMY WAR COLLEGE ºur li TRATEGIC studies INSTITUTE S S S STRATEGIC STUDIES INSTITUTE The Strategic Studies Institute (SSI) is part of the U.S. Army War College and is the strategic-level study agent for issues related to national security and military strategy with emphasis on geostrate- gic analysis. The mission of SSI is to use independent analysis to conduct strategic studies that develop policy recommendations on: • Strategy, planning, and policy for joint and combined employment of military forces; • Regional strategic appraisals; • The nature of land warfare; • Matters affecting the Army's future; • The concepts, philosophy, and theory of strategy; and • Other issues of importance to the leadership of the Army. Studies produced by civilian and military analysts concern topics having strategic implications for the Army, the Department of De- fense, and the larger national security community. In addition to its studies, SSI publishes special reports on topics of special or immediate interest. These include edited proceedings of conferences and topically-oriented roundtables, expanded trip re- ports, and quick-reaction responses to senior Army leaders. The Institute provides a valuable analytical capability within the Army to address strategic and other issues in support of Army par- ticipation in national security policy formulation. – Strategic Studies Institute Monograph LESSONS OF THE IRAQI DE-BA’ATHIFICATION PROGRAM FOR IRAQ'S FUTURE AND THE ARAB REVOLUTIONS W. Andrew Terrill May 2012 The views expressed in this report are those of the authors and do not necessarily reflect the official policy or position of the De- partment of the Army, the Department of Defense, or the U.S. Government. Authors of Strategic Studies Institute (SSI) publica- tions enjoy full academic freedom, provided they do not disclose classified information, jeopardize operations security, or mis- represent official U.S. policy. Such academic freedom empow- ers them to offer new and sometimes controversial perspectives in the interest of furthering debate on key issues. This report is cleared for public release; distribution is unlimited. xxx-k-k This publication is subject to Title 17, United States Code, Sec- tions 101 and 105. It is in the public domain and may not be copy- righted. \?\) 'kºkkkºk Comments pertaining to this report are invited and should be forwarded to: Director, Strategic Studies Institute, U.S. Army War College, 45 Ashburn Drive, Bldg. 47, Carlisle, PA 17013- 5046. -k-k-k-k-k I would like to thank Dr. Steven Metz and Mary J. Pelusi for useful and insightful comments on earlier drafts of this mono- graph. All mistakes in this work of fact, omission, interpretation, and speculation are, nevertheless, entirely my own. -k-k-k-k-k All Strategic Studies Institute (SSI) publications may be downloaded free of charge from the SSI website. Hard copies of this report may also be obtained free of charge while supplies last by placing an order on the SSI website. SSI publications may be quoted or reprinted in part or in full with permission and ap- propriate credit given to the U.S. Army Strategic Studies Insti- tute, U.S. Army War College, Carlisle Barracks, PA. Contact SSI by visiting our website at the following address: www.Strategic Studies.Institute.army.mil. *k-k-k-k-k The Strategic Studies Institute publishes a monthly e-mail newsletter to update the national security community on the re- search of our analysts, recent and forthcoming publications, and upcoming conferences sponsored by the Institute. Each newslet- ter also provides a strategic commentary by one of our research analysts. If you are interested in receiving this newsletter, please subscribe on the SSI website at www.StrategicStudies.Institute. army.mil/newsletter/ ISBN 1-58487–527–5 lºgº zº, 3. |olſ. ii - - - - notes that the initial principles of Ba'athism were suf- ficiently broad as to allow their acquisition by a tyrant seeking ideological justification for a merciless re- gime. His comprehensive analysis of Iraqi Ba'athism ensures that he does not overgeneralize when draw- ing potential parallels to events in the Arab Spring countries. Dr. Terrill considers the nature of Iraqi de-Ba'athification in considerable depth and carefully evaluates the rationales and results of actions taken by both Americans and Iraqis involved in the process. While there are many differences between the for- mation of Iraq's post-Saddam Hussein government and the current efforts of some Arab Spring govern- ing bodies to restructure their political institutions, it is possible to identify parallels between Iraq and Arab Spring countries. Some insights for emerging governments may, correspondingly, be guided by a comprehensive understanding of these parallels. The Arab Spring revolutions that have overthrown the governments of Tunisia, Egypt, Libya, and Yemen at the time of this writing are a regional process of stun- ning importance. While these revolutions began with a tremendous degree of hope, great difficulties loom in the future. New governments will have to appor- tion power, build or reform key institutions, establish political legitimacy for those institutions, and accom- modate the enhanced expectations of their publics in a post-revolutionary environment. A great deal can go wrong in these circumstances, and it is important to consider ways in which these new governing struc- tures can be supported, so long as they remain inclu- sive and democratic. Any lessons that can be gleaned from earlier conflicts will be of considerable value to the nations facing these problems as well as to their re- gional and extra-regional allies seeking to help them. iv ABOUT THE AUTHOR W. ANDREW TERRILL joined the Strategic Stud- ies Institute (SSI) in October 2001, and is SSI's Middle East specialist. Prior to his appointment, he served as a Middle East nonproliferation analyst for the Inter- national Assessments Division of the Lawrence Liver- more National Laboratory (LLNL). In 1998-99, Dr. Terrill also served as a Visiting Professor at the U.S. Air War College on assignment from LLNL. He is a former faculty member at Old Dominion University in Norfolk, Virginia, and has taught adjunct at a variety of other colleges and universities. He is a retired U.S. Army Reserve Lieutenant Colonel and Foreign Area Officer (Middle East). Dr. Terrill has published in nu- merous academic journals on topics including nuclear proliferation; the Iran-Iraq War; Operation DESERT STORM; Middle Eastern chemical weapons and bal- listic missile proliferation, terrorism, and commando operations. He is also the author of Global Security Watch-Jordan (Praeger, 2010). Since 1994, at U.S. State Department invitation, Dr. Terrill has participated in the Middle Eastern Track 2 talks, which are part of the Middle East Peace Process. He has also served as a member of the military and security working group of the Baker/Hamilton Iraq Study Group. Dr. Terrill holds a B.A. from California State Polytechnic Uni- versity and an M.A. from the University of Califor- nia, Riverside, both in political science, and a Ph.D. in international relations from Claremont Graduate University, Claremont, California. vii it was deemed especially important to eliminate the last vestiges of Saddam Hussein's regime to prevent a similar type of government from reestablishing itself. In making this choice, advocates of deep de- Ba'athification pointed to the history of Ba'athist con- spirators rising to power through infiltrating govern- ment institutions and seizing power in undemocratic ways. This comprehensive approach nevertheless made it extremely difficult for Iraq's Sunni Arab lead- ers to accept the post-war political system. Many U.S. leaders became concerned about this problem over time, but they had increasing difficulties moderating Iraqi administration of de-Ba'athification efforts. Despite the time that has elapsed since the initial decisions on de-Ba'athification, these issues remain vital for the future of Iraq. The Sunni Arab insurgency that developed after the U.S.-led invasion reinforced the popularity of de-Ba'athification among many of Iraq's Shi'ite Arabs, thereby keeping the policy alive. Many Shi'ites also agreed with U.S. concerns about the potential emergence of a new Sunni-dominated regime that would once again seize and retain power. A quasi-legal de-Ba'athification Commission (now known as the Justice and Accountability Commis- sion) continues to exist in Iraq and recently played a dramatic role in disqualifying some leading Sunni candidates in the 2010 parliamentary elections. This commission could not have remained relevant with- out the support of a variety of important Iraqi politi- cians, including the current prime minister. Likewise, Iraqi Prime Minister Maliki arrested large numbers of so-called “Ba'athists" in 2011, shortly before the final withdrawal of U.S. troops. Under these circumstances, the legacy of de-Ba'athification and the future of this concept within the Iraqi political system may yet have serious consequences for Iraq's ability to build a uni- fied and successful state. Many Americans and Iraqis of diverse political ori- entations have argued that de-Ba'athification and the nature of sectarianism in Iraq involved a large number of lessons that other countries may wish to consider in the context of future political transitions. This argu- ment has found considerable resonance among some citizens in the “Arab Spring" states where popular uprisings have ousted some long-serving dictators. Many of the new revolutionaries consider Iraq's prob- lems as a cautionary tale that must be understood as they move forward in establishing new political sys- tems. In particular, it is now understood that loyalty commissions led by politicians and set up to identify internal enemies can take on a life of their own and become part of a nation's power structure. Once this occurs, such organizations are exceedingly difficult to disestablish. Likewise, the basic unfairness of collec- tive punishment has again been underscored as an en- gine of anger, resentment, and backlash. Conversely, the importance of honest and objective judicial institu- tions has also been underscored, as has the importance of maintaining a distinction between revenge and jus- tice. Moreover, officers and senior non-commissioned officers (NCOs) of the U.S. Army must realize that they may often have unique opportunities and unique credibility to offer advice on the lessons of Iraq to their counterparts in some of the Arab Spring nations. The U.S. Army has a long history of cooperating with some of the Arab Spring militaries and has a particu- larly strong relationship with the Egyptian military. These bonds of trust, cooperation, and teamwork can be used to convey a variety of messages beyond exclu- sively military issues. xi LESSONS OF THE IRAQI DE-BA’ATHIFICATION PROGRAM FOR IRAQ'S FUTURE AND THE ARAB REVOLUTIONS There was a tendency among promoters of the [2003- 2011 Iraq] war to believe that democracy was a default condition to which societies would revert once liber- ated from dictators. Francis Fukuyama' I pleaded with Bremer not to dissolve the [Iraqi) army, and warned him that it would blow up in our faces. I told him that I understood the rationale behind the process of de-Baathification, but that it needed to apply only to those at the top with blood on their hands....I said I hoped he understood that if he was going to de- Baathify across the board, he would be setting him- self up for major resistance and would create a power vacuum that someone would have to fill. King Abdullah II of Jordan” You cannot build a country if you don't have recon- ciliation and forgiveness. Aref Ali Nayed Libyan National Transitional Council” INTRODUCTION The presence of U.S. combat troops in Iraq has now come to an end, and the lessons of that conflict, includ- ing those involving de-Ba'athification, will be debated for some time to come. De-Ba'athification for Iraq was º initiated by U.S. policymakers in 2003 as the process of eliminating the ideology of the Iraqi Ba'ath Party from public life and removing its more influential ad- herents from the Iraqi political and administrative sys- tem. This policy constituted a central part of the effort to eliminate all significant aspects of the Saddamist state and remake Iraq into a democratic nation. It has also emerged as one of the most controversial aspects of U.S. post-war activities in Iraq. While supporters claim that the approach was unavoidable if Iraq was to be reformed, critics maintain that the approach, as practiced, amplified sectarian divisions in Iraq and also served as an important enabler of enhanced sec- tarianism and the post-invasion Iraqi insurgency. U.S. Government decisionmaking about the nature and depth of the de-Ba'athification effort centered on the conflict between pragmatists who were attempt- ing to prevent U.S. and Iraqi post-war authorities from losing their capacity to manage the emerging crisis in Iraq and various hardliners—often called neoconser- vatives—calling for a fundamental restructuring of Iraqi society. The dominant fear of the first group was that Iraq would degenerate into chaos without some effort to rehabilitate and retain those Ba'athist bureau- crats and officials not directly implicated in the Sad- dam Hussein regime's crimes. For the second group, the primary concern appeared to be ensuring that a favorable outcome for regime change was permanent. Their greatest fear was often that a system of "Sad- damism without Saddam" would dominate the post- war environment unless large-scale Societal restruc- turing took place within Iraq.” In both groups, there was a wide range of opinion, and some individuals (perhaps most prominently National Security Council [NSC] Advisor and later Secretary of State Condoleez- za Rice) were open to the arguments of both sides and sought to synthesize them into coherent policy. While de-Ba'athification still retains some defend- ers in the United States, most Middle Eastern politi- cians and observers consider it to have been deeply misguided, and many Arabs view it as a warning of the ways in which a transition from dictatorial rule can go wrong and lurch dangerously close to civil war. A strong exception to this belief can sometimes be found among Iraqi and other Arab Shi'ites, who ba- sically approve of a policy that punishes Iraq's Sunni Arab community from which Saddam drew most of his supporters and that suffered less than other Iraqi communities under the dictatorship. The future of Iraq as a cohesive and modernizing country remains uncertain, and it is unclear if that society can over- come simmering sectarian differences, which current approaches to de-Ba'athification continue to inflame. The ways in which Iraq deals with the legacy of de- Ba'athification, as well as ongoing policies for national reconciliation, will have a great deal to do with decid- ing the Iraq future. While Iraqis often dream of build- ing a society as prosperous as the Arab Gulf states, the danger remains of an Iraqi society that looks more like Lebanon during its 14-year sectarian civil war. The onset of the Arab Spring has revived a number of questions about the problems with de- Ba'athification. At the time of this writing, Tunisia, Egypt, and Libya have experienced Arab Spring popular uprisings in which long-standing dictators have been ousted. Syria is also experiencing a serious mass uprising led by brave and extremely committed revolutionaries struggling against an entrenched and ruthlessly tenacious dictatorship. None of these states has ever experienced a government as authoritarian as Saddam Hussein's Iraq, although the Syrian dic- tatorship clearly comes the closest to the Saddamist model. All of these states face considerable difficul- ties in establishing legitimate and moderate post-rev- olutionary governments, and some face the danger of prolonged civil conflict. Lessons that can be gleaned from the Iraqi experience may therefore be especially important for their future. THE BA’ATH PARTY AS AN INSTRUMENT OF SADDAM HUSSEIN'S DOMINATION OF IRAQ In order to understand problems surrounding the effort to remove Ba'athism from Iraq, it is necessary to give some consideration to the central tenets of Ba'athism as a political ideology and then to examine the ways in which this ideology was applied and prac- ticed within Iraq under Saddam Hussein. In undertak- ing this analysis, it is worthwhile to consider that a number of dictatorial regimes have used official ide- ologies to justify the power of a particular elite rather than to guide their actions. Some individuals within the ruling elite of such systems may view themselves as seeking to adjust their approaches to emerging problems by emphasizing those aspects of the ideolo- gy that seem most useful for addressing a given prob- lem, while de-emphasizing those that are less useful. Such people remain ideologues despite their willing- ness to show a limited degree of flexibility. Others do not take the national ideology particularly seriously but value its supporting party infrastructure to justify and generate support for the decisions of the political leadership, regardless of how ideologically inconsis- tent those decisions may be. These people are political opportunists in ideological garb. The Ba'ath movement was founded in the 1940s by two Syrian teachers, Michel Aflaq and Salah al-Din Bi- tar, and stressed Arab unity, socialism, and efforts to modernize the Arab World. The party, which emerged in its modern form in 1947, sought to unite all Arab states and to provide them with a set of moderniz- ing principles to help them overcome problems with poverty and backwardness. The word Ba'ath means renaissance or rebirth in Arabic. The movement also sought to address the problems of the entire Arab World and was not to be confined to any individual country. Ba'athists throughout the Arab World were often viewed as committed Arab nationalists who were particularly devoted to the concept of a strong, unified Arab nation. Their slogan is, “One Arab nation with an eternal mission." Aflaq and Bitar met at the Sorbonne in Paris, France, in 1929 where both of them became especially interested in Western literature and philosophy with an emphasis on Marxism and socialism. This form of study was a fairly conventional approach for Arab students in France, since only the French communists and socialists showed much sympathy for Syrian inde- pendence within the political spectrum of Paris in the 1930s. Moreover, Marxism's emphasis on moderniza- tion and scientific socialism appealed to the two men as they struggled for a solution to widespread Arab impoverishment and underdevelopment. The Ba'ath modernize the Arab World in ways that were rooted in leftist, European political and social thought. Is- lam was not seen as a major part of this modernizing outlook. In this regard, Aflaq was not even a Muslim, having been raised as an Orthodox Christian.” Batar was a Sunni Muslim and, like Aflaq, had no interest in religion as a basis for the state. Within the context of Ba'athist ideology, Islam was primarily viewed as part of the Arab heritage rather than a way to organize contemporary political life. Their outlook was corre- spondingly deeply secular. Like Marxist-Leninist organizations, the Ba'ath Party sought to enter power through the actions of a revolutionary elite operating in a variety of states, including Iraq. In the 1950s and 1960s, these tactics caused the Ba'ath to compete with a number of other conspiratorial movements to infiltrate the military and other centers of state power. Subversion and coups seemed the only way in which to achieve pow- er since contested elections were almost never held in any Arab country except perhaps Lebanon. Major emerging political trends throughout the Arab World included communist and Nasserite movements as well as the Ba'ath. Thus, to achieve power within the various Arab countries Ba'athists had to operate clan- destinely as one of many secretive opposition move- ments dealing with government counterintelligence units and their own splinter groups. Despite these dif- ficulties, Ba'athists seized power in Syria and Iraq in 1963.” The Iraqi Ba'ath Party remained in power for less than a year but once again seized power in 1968 partially as a result of the maneuverings of a young revolutionary named Saddam Hussein. Additionally, the previous Iraqi government had been unable to provide significant help to the other Arab countries at war with Israel in June 1967. Iraqi Ba'athist leaders portrayed this failure as a form of treason, and made anti-Israeli invective a centerpiece of their rhetoric fol- lowing their seizure of power.” The Iraqi Ba'ath Party began its existence with a commitment that all party members should have a broad set of rights to elect officials and present their views in party forums. Unfortunately, this approach changed rapidly over time, and by 1964 Aflaq was complaining about the stratification of the party and the consolidation of power by a limited number of "active members" with influence that dramatically ex- ceeded that of the rank and file. He stated that such an approach “was wholly out of keeping with the spirit of our party's rules.” Nevertheless, the requirement for the Ba'ath Party to carry out its activities in secret until it seized power for the second time in 1968 re- mained a central part of Ba'ath organizational culture throughout the organization's existence. During its underground years, the Ba'ath became increasingly hierarchical, secretive, and accustomed to violence as a political tool. These mindsets carried over to the years in power when such an approach was viewed as equally necessary to cope with real and imagined in- ternal and foreign enemies. The Ba'ath leaders contin- ued to see conspiracies against their government from a variety of sources including the Western powers and Israel. The failure of Iraq's first Ba'ath government to remain in power more than a year underscored the looming danger of a countercoup. Ba'athism appeared to have some problems estab- lishing a popular base in the first years after the 1968 coup. Some Iraqi citizens appreciated Ba'ath ideology for its emphasis on modernity and its rejection of eth- nic/sectarian divisions, tribalism, and religion as the basis for a modern state. Unfortunately, in both Iraq and Syria, these principles had a more insidious func- tion as well, helping to serve as a smokescreen for the domination of one social group over the others in each country. Secular principles in Syria were used to mask the almost complete domination of Syrian society by the Alawite minority, which is usually identified as an offshoot of Shi'ite Islam. In Iraq, the Ba'athist age as a way to achieve high rank within the party. Saddam's ongoing relationship with Aflaq was useful to him throughout his life. Unlike the Syrian Ba'ath Party, which ousted Aflaq and Bitar from power in February 1966, Saddam remained aware of the value of maintaining Aflaq as an honored but powerless member of the Iraqi leadership. Aflaq, for his part, had hoped to be a positive and moderating influence on Saddam once the dictator achieved power, but most of his suggestions on important issues were ig- nored. Saddam did flatter the older man by agreeing to Some of his minor concerns. Such cosmetic conces- sions were an acceptable trade-off for the public sup- port of one of Ba'athism's co-founders. By consorting with the dictator, Aflaq allowed Saddam to exploit him and Ba'athism as window dressing for one of the world's most oppressive regimes. Bitar, by contrast, spent the remainder of his life in Europe. Aflaq died in 1989 in Paris, and Saddam let it be known that he used his personal funds to build a suitable tomb for the co-founder of Ba'athism. Saddam was not a military man, and as a youth was rejected for entry into the Iraqi military academy due to poor performance on his entrance examina- tions.” Throughout his rise to power, Saddam was cor- respondingly wary of the danger of a military coup and used the Ba'ath Party to help him secure full control over the Army. This concern is easily under- standable since coups were the traditional means of ousting an Iraqi leader once his enemies were able to organize against him. In establishing an iron grip over the military, Saddam made heavy use of Ba'athist po- litical officers and frequently promoted Cronies within the military over more qualified officers. Officers with particularly heroic reputations in the Iran-Iraq war, as well as brilliant planners, were quietly sidelined, since 9 there was room for only one "military genius" in Sad- dam's Iraq. Saddam understood the value of efficient officers during times of war, but tended to place these officers in less important positions when he no longer had an immediate need for them. The Ba'ath Party was also useful to Saddam in other ways than simply controlling the military and providing an ideological veneer for the regime. The creation of the Saddam personality cult had nothing to do with original Ba'ath ideology, but it was admin- istered and energized by Ba'ath Party activists. As Saddam Hussein consolidated his rule over Iraq, he consistently viewed the Ba'ath Party as an instrument of dictatorial power and social mobilization. He did not take its ideology and values seriously as principles for leadership, and individuals at the highest levels were noted for their public and ostentatiously blind loyalty to the President rather than their knowledge of Ba'athist principles and political thought. While many members of the top leadership were Sunni, this was not an absolute requirement. Proven Saddam loyalists included Shi'ites, Kurds, and various sects of Christians." If Saddam believed a subordinate was a proven and committed loyalist, he did not particu- larly care what that person's sect or ethnicity was. On the other hand, Saddam often viewed his own family and Sunni Arabs from the Tikrit area as having a head start on loyalty." Saddam and his cronies also seemed to view Sunni Arabs as being more likely to remain loyal, because they were usually more hostile to the traditional enemy of Iran and were likely to fear a new Shi'ite government in which they could be viewed as accomplices in Saddam's crimes. Consequently, the Sunni Arabs were disproportionately represented in the Ba'ath's senior ranks and the regime's security units. 10 Once in power, the Ba'ath Party did follow through on some of its modernization rhetoric. Saddam was committed to building a modern state, although he basically sought this goal primarily to improve the ef- ficiency of the dictatorship rather than to benefit the Iraqi people. Consequently, serious and intense Ba'ath Party literacy drives did more than teach Iraqi citizens how to read.” They also opened an intellectual path- way that allowed them to be more thoroughly bom- barded with regime propaganda. Efforts to reduce the power of the tribes and to limit the role of religion in public life were similarly presented as modernization efforts, although their primary purpose was to further centralize power in Baghdad. Moreover, such policies could be reversed when they were no longer conve- nient to the regime, as occurred in the late 1980s and throughout the 1990s when Saddam's regime sought to encourage some increased religious devotion, so long as such sentiments were properly channeled into activities that the regime viewed as useful.” Addition- ally, Saddam was also willing to work through tribal elements when it suited his purposes. On the eve of the 2003 U.S.-led invasion, Saddam's Iraq was a one of the most rigid totalitarian states in the world, with a privileged elite composed of mili- tary leaders and Ba'ath Party members, virtually all of whom were terrified of the leader." The Ba'ath Party had at least two million members at that time, with Some estimates reaching 2.5 million. Nevertheless, membership in the senior ranks of the Ba'ath Party did not protect individuals from Saddam's terror, which was applied to them to ensure that rival centers of power did not develop within the party.” Saddam was particular wary of ambitious “overachievers" who might be interested in political advancement in of how individual Ba'athists outside of Iraq's top circles would be treated. In the immediate aftermath of Saddam's ouster, both the U.S. military and the newly established Organization for Reconstruction and Humanitarian Assistance (ORHA) seemed to be showing some clear flexibility. ORHA was willing to allow former Ba'athist administrators and profession- als such as doctors and professors to keep their jobs so long as they were not implicated in regime crimes and were willing to renounce their previous Ba'athist affiliations.” This approach was viewed as necessary to keep the economy from further declining or even collapsing. The U.S. Army also showed considerable pragmatism by sponsoring renunciation ceremonies in which thousands of people burned their Ba'ath membership cards, renounced violence, and pledged to help build the new Iraq.” This approach was par- ticularly successful in the area around Mosul, where then Major General David Petraeus presided over such ceremonies. Mosul, at this time, remained quiet, despite its tradition of supplying large numbers of Sunni Arab officers to the Iraqi military. Later, af- ter more comprehensive de-Ba'athification policies were instituted over the objection of the U.S. military leadership there, everything changed, Mosul became much more difficult to manage, and a strong al-Qaeda presence was established in the region. --~~ As noted above, the more tolerant approach of ORHA was not to last. An order to de-Ba'athify Iraqi Society was the first major official act of Ambassador L. Paul Bremer upon his arrival in that country to assume control of the newly created Coalition Provisional Au- thority (CPA), which replaced ORHA. Bremer issued this order on May 16, 2003, after being provided with the directive in draft form by Undersecretary of De- 13 fense for Policy Douglas Feith. According to Bremer, Feith told him that such an order was absolutely es- sential to Iraq's rehabilitation.” The order disestab- lished the Ba'ath Party and removed members of the four highest ranks of the party from government posi- tions. It also banned them from future employment in the public sector. Additionally, the order required that anyone holding positions in the top three management layers in government institutions be interviewed to determine their level of involvement with the Ba'ath Party as well as their possible involvement in criminal activities. Those determined to be senior members of the party were to be removed from their positions and banned from any future public employment. The or- der also called for the creation of a rewards program to pay individuals providing information leading to the capture of senior Ba'ath Party members. The supporters of the de-Ba'athification program frequently maintained that this approach was inspired by the de-Nazification efforts that followed World War II in Germany. Iraqi exiles were fond of the term, which they may have viewed as loaded in a way that made it a useful public relations tool to advocate war and to help clear a way for prominent roles for them- selves in the new Iraq. Additionally, some U.S. senior officials had, by this time, begun viewing Iraq through the lens of Nazi Germany with Saddam as Hitler and the Ba'ath Party as the Nazis.” Such analogies cor- rectly point out the moral repugnancy of the Saddam Hussein regime, but they also allow one to glance over the particulars of Iraqi society and argue about Iraq's future on the basis of analogies rather than conditions within Iraq itself. In just one important difference, it may be significant that the Nazis rose to power as a large and powerful mass movement, whereas the 14 Ba'athists rose to power in Iraq through the actions of a group of conspirators. Individuals joining the Ba'ath movement after it seized power may have done so with motives other than loyalty to Saddam Hussein. The de-Ba'athification order and the subsequent CPA Order #2 (issued shortly afterward on May 23 to disband Iraq's military and intelligence forces) re- flect the priorities of both Under Secretary Feith and Deputy Secretary of Defense Paul Wolfowitz. These priorities centered on the destruction of all forces previously involved in supporting the old regime and particularly those forces that they believed had a chance of reconstituting that regime. The Ba'ath had a long history of underground activity as well as a past pattern of infiltrating key institutions and then at- tempting to seize power by illegal means. The revival of Ba'athism through conspiracy and intrigue there- fore seemed a realistic danger. Unfortunately, such a revival was not the only serious danger facing Iraq at this time, and it was not clearly so dangerous as to trump all other security concerns. It is also not clear if the U.S. leadership fully understood the numbers of enemies that they were making by undertaking such policies or the backlash such actions could produce. The possibility that such a backlash could lead to a serious Sunni military challenge to the new Iraq was apparently dismissed on the grounds that such "dead- enders" were a marginalized force and would not be able to establish a popular rather than a conspiratorial movement within Iraqi society. Ahmad Hashim, in his insightful study of the Iraqi insurgency, quotes an anonymous U.S. policymaker as stating, "We underes- timated their [the Iraqis] capacity to put up resistance. We underestimated the role of nationalism. And we overestimated the appeal of liberation [as trumping all other considerations for Iraqi political behavior].” Another even more biting critic stated that the civil- ians within the George Bush administration had made the fundamental mistake of confusing strategy with ideology.” Some authors also claim that the CPA's policies were deliberately anti-Sunni and pro-Shi'ite because of a belief within the Bush administration that Sunnis were more dangerous to U.S. interests, while Shi'ites were more likely to be grateful to the United States for ousting Saddam, since they had suffered more under his regime.” This charge about administration policy- making is more popular in the Arab World than in the United States and is difficult to confirm. Some Bush policymakers did speak forcefully against Sunni con- trol in Iraq, but they justified their concerns around the theme of democracy rather than the inherent un- trustworthiness of the Sunni Arabs.” In some regional media, as well as in Iraq, the de-Ba'athification policy was sometimes referred to as "de-Arabization.” The central tenets of the Ba'ath Party are Arab nationalism, anti-imperialism, and Arab socialism. Such ideals are not usually viewed as offensive by themselves, and many Arabs consider them to be noble and praisewor- thy. Treating Ba'athism, instead of Saddam's version of Ba'athism, as corrupt was therefore a problem for many Arabs and the pan-Arab media including the satellite television stations where Iraqis often sought to get the news. In an effort that further complicated the situation, some leading Iraqi Shi'ites attempted to play upon U.S. fears by suggesting that Sunnis were “Arab na- tionalists." This is a label that is seldom viewed as a slur in the Arab World, but in this instance was ap- parently used to suggest an anti-American and anti- 16 Israeli worldview. Throughout the years following the invasion, some Shi'ite leaders consistently sought to convey the view that Sunnis were irredeemably wedded to radicalism, and needed to be marginalized to protect both Iraqi and Shi'ite interests. In one par- ticularly revealing incident, Shi'ite leader Abdulaziz Hakim made it clear that he supported democracy so long as his organization and sect benefited from that democracy. In conversations reported by journalist Bob Woodward and others, Hakim told members of the Baker/Hamilton Iraq Study Group that the gov- ernment of Iraq represented 80 percent of the popula- tion of that country (Shi'ites and Kurds) so democracy was served, and nothing had to be done about the re- maining Sunnis.” When Bremer informed the senior staff of the CPA (and especially the ORHA holdovers) of the new de- Ba'athification approach, he met immediate resistance over the scope of the order that he had brought from Washington. Retired Lieutenant General Jay Garner, the outgoing Director of ORHA, was reported to have been disturbed by the order, which he characterized as “too deep.” Charlie Sidell, the Baghdad Cen- tral Intelligence Agency (CIA) Chief of Station who worked with Garner during this period, stated, "Well if you do this, you're going to drive 30,000 to 50,000 Ba'athists underground by nightfall, and the number is closer to 50,000 than it is to 30,000.” Garner and Sidell went to Bremer to attempt to dissuade him from issuing the order until it had been moderated to reflect the realities that they were facing. They recommended eliminating the top two levels of Ba'athist leadership, which was about 6,000 people.” According to Garner, Bremer stated, “Look, I have my orders. This is what I am doing.” Since Bremer held the rank of Presiden- 17 tial Envoy in direct communication with the President, it is not immediately clear who issued such orders. Undersecretary Feith could not have done so on his own authority. President Bush had previously given Defense Secretary Donald Rumsfeld managerial con- trol of the occupation, so it is possible that Feith spoke for Rumsfeld who spoke for Bush and Vice President Dick Cheney.” A complicating factor in this situa- tion is that throughout his time in office, Bremer was willing to ignore the advice of the Defense and State Departments on other issues later in his tenure. If he did not do so in this instance, he probably believed in the policy that was being put forward or considered it to have come directly from the President. It is also likely that he did not fully understand the importance of the advice he was receiving from Garner and the CIA, since he later stated that he did not recall the con- versation.” Garner left Iraq shortly afterward, sharing his concerns over de-Ba'athification with U.S. Central Command (CENTCOM) Deputy Commander then Lieutenant General John Abizaid, who also feared that the deep de-Ba'athification effort would feed the de- veloping resistance.” General Abizaid would become CENTCOM commander after General Franks' retire- ment. In a related event, President Bush later appeared to blame Bremer for disbanding the Iraqi Army (al- though not for deep de-Ba'athification), suggesting that presidential guidance on one of the most impor- tant issues of the occupation was not reflected in CPA decisionmaking. Rather, Bush told journalist Robert Draper, “The policy had been to keep the [Iraqi army intact. Didn't happen.” Bremer responded angrily to the President's statement, saying that he had been ordered to disband the Army by Rumsfeld, and the 18 White House had approved the move. He also made the unusual claim that disbanding the Iraqi Army had been the correct choice, but he was not the one respon- sible for this decision.” Clearly, these are very differ- ent versions of the truth, and no one wants to take responsibility for disbanding Iraqi security forces in spite of Bremer's professed belief that it had been the correct approach. Despite this inconsistency, Bremer's arguments have a certain level of resonance, since it is difficult to believe that he would have implemented such dramatic policy changes without at least a gen- eral understanding of President Bush's priorities on de-Ba'athification and the future of the Iraqi military. At this point, Bremer was imposing Washington's priorities and appeared primarily concerned about preventing the possible reconstitution of the Ba'ath regime. These fears may have been enhanced by Sad- dam's status as a fugitive at that time. Moreover, Bremer also entered Iraq with the determination to establish himself quickly as a decisive leader willing to make decisions that were unpopular with his staff, the military, and others in the U.S. Government. In his book, Bremer relates an incident in which his son gave him a pair of desert combat boots as a going away gift with the note that they were to help him "kick some butt.” He was apparently in total agreement with that sentiment.” Bremer clearly felt that asserting his will over subordinates was exceptionally important if he was to maintain effective control of the CPA and Iraqi policy.” He made this effort in the face of consid- erable local unhappiness about CPA policy, and de- Ba'athification was especially unpopular in the U.S. military because U.S. officers lost their hardest work- ing and most competent counterparts.” In response to the order, some commanders, and most notably Gen- 19 eral Petraeus, sought wide authority to grant waiv- ers from the de-Ba'athificaiton requirements for local individuals to limit the disruptions caused by this policy." Bremer claims in his book that he expected the de-Ba'athification order to be applied to only about 20,000 people, or what he identified as 1 percent of all party members. The program would therefore include the ranks officially designated as “Senior Party Mem- bers." Bremer also claims to have been sensitive to the needs of lower-ranking Ba'ath Party members to join the organization to make a living. He later maintained that his order was applied in ways that he never in- tended, and that many more people were purged than he had envisioned under the original program. This included people of much lower rank than the levels of Ba'ath membership outlined in the order as well as individuals whose links to the Ba'ath Party leadership were tenuous at best. He was also apparently unre- sponsive to Ambassador Barbara Bodine's argument made earlier to General Garner that some senior mem- bers of the party were not criminals, while various ju- nior members had engaged in serious crimes, making a blanket approach based on rank alone unfair and ineffective.” Another problem for the CPA was that the justice of the de-Ba'athification order was not clear to many Iraqis. Joining the Ba'ath Party in Saddam's Iraq was a rational decision for anyone seeking to feed their family and live in conditions other than squalor and poverty. The best and most numerous jobs in Iraq are found in the government and in state-controlled en- terprises such as the oil industry. In Iraq, as in most Middle Eastern countries, there is not a strong private sector with a wide variety of good jobs. Socialism and 20 W- llS he jut all de the Oin led láſ] his Wels las state control of the economy were official parts of the Ba'ath ideology, further weakening the nongovern- mental sector, while years of United Nations (UN) sanctions (1990-2003) undermined foreign investment in the Iraqi economy and also retarded private sector development. Yet, it is also within the government that one was most vulnerable to pressure to show enthusiasm for Saddam's rule. In this environment, the greatest and most direct system of rewards and punishment had been put into place for rewarding loyalty to the government and the party. In Iraq, a non-Ba'athist primary school teacher would usually be paid the equivalent of U.S. $4 per month, while a Ba'athist in the same position, doing the same work, would be paid around $200 per month.” Unfortunately, Bremer's estimate of 20,000 people being purged as a result of his order did not hold up. While exact numbers are impossible to obtain, most estimates place the number as at least 30,000 and possibly up to 50,000 individuals.” A few estimates place it even higher and note that the party members' families, as well as ousted Ba'athists, were harmed by the mass firings.” Blanket de-Ba'athification pun- ished Iraq's managerial class merely for being part of that class, and not because of individual miscon- duct, abuse of power, or other crimes. Moreover, other choices were available to address the problem, although they clearly would have been more cumber- some. According to one observer, the best alternative would have been to place the Ba'athists on trial and then punish those found guilty of human rights viola- tions, corruption, incompetence, and other crimes. A truth and reconciliation commission could then have been established along South African lines. Such an option would have avoided the approach of treating 21 all Ba'athists in responsible positions as criminals.” Additionally, there was also the possibility supported by Garner and others to dismiss only the top two lev- els of the Ba'ath Party leadership and thereby try to avoid plunging Iraq into an administrative vacuum by eliminating managers and technocrats, many of whom were only “nominal Ba'athists.” As will be examined later, Bremer maintains that his de-Ba'athification order was issued with a full un- derstanding of the complexities of Iraqi society, but it was overzealously applied. Yet, if Bremer's authority and the approach of his order were abused, he still cannot be fully absolved for the difficulties that fol- lowed. In addition to problems with the decision it- self, it is unclear that the CPA leadership paid enough attention to how his order was being implemented throughout the process rather than simply issuing a fiat and expecting it to be carried forward without dif- ficulty, first under the authority of the CPA and then by the Iraqi government. Lieutenant General Ricardo Sanchez, a former commander in Iraq, excoriated the CPA on these grounds noting, "[T]he CPA treated [de- Ba'athification] like they were issuing an academic, theoretical paper. They simply released the order and declared success. But there was no vision, no concept, and in my opinion, no desire to ensure that the policy was properly implemented. On the other hand, it did look good on paper.” While Bremer was to become more pragmatic over time, his first few days in Iraq resulted in what have arguably emerged as some of the worst mistakes asso- ciated with the war, and these mistakes were impos- sible to reverse by the time he started to understand their negative implications.” It is nevertheless also useful to understand the context of Bremer's actions 22 by looking at the reaction to these policies in Washing- ton. In his memoir, Douglas Feith minimizes the chaos created by de-Ba'athification, and takes issue with Bremer's later second thoughts about the policy.” Un- like Bremer, he was unprepared to admit that the de- Ba'athification policy may have been producing bad results. Rather than adjust his focus to the real and emerging problems as Bremer eventually did, Feith, at least publicly, continued to support policies that were proving disastrous. ADMINISTRATION OF THE DE-BA’ATHIFICATION PROGRAM Nine days after the issuance of CPA Order Num- ber 1, Bremer established a de-Ba'athification Coun- cil, which he was to supervise and which would report "directly and solely" to him.” Later, on No- l vember 3, 2003, the responsibility for implementing de-Ba'athification was passed from the CPA to the U.S.-created Iraqi Governing Council (IGC).” The IGC made de-Ba'athification the responsibility of Govern- ing Council member Ahmad Chalabi, who was placed in charge of the newly-created “Supreme National Commission for De-Ba'athification." Chalabi was sup- ported in his efforts at deep de-Ba'athification by the Shi'ite religious parties such as Dawa and the Supreme Council for the Islamic Revolution in Iraq (SCIRI, later the Islamic Supreme Council of Iraq) and by vari- ous Kurdish groups. Former post-Saddam Defense Minister Ali Allawi (not to be confused with Ayad Allawi) describes Iraqi Kurds as favoring broad de- Ba'athification, but with so many exceptions that their actual priorities were difficult to sort out.” Most Sunni Iraqi Arabs did not favor deep de-Ba'athification, al- 23 Inder&# e Sumik "sºft; itépì. malſº ySmi. tes: iddî: iArabſ. ideaſ: 3 Othº jº ing managed by either Chalabi or by local committees that had set themselves up using Bremer's order as the rationale for their activities. Chalabi, who had strong allies in the U.S. civilian leadership of the Pentagon, may have been particularly difficult for Bremer to moderate. Much later, in retrospect, Bremer indicated that de-Ba'athification should have been conducted by a judicial body rather than a commission led by Iraqi politicians.” The collapse of large segments of the Iraqi educa- tional system harmed not only teachers but students and Iraqi families by rendering Schools and universi- ties increasingly dysfunctional. It also created pools of high school and college age males who could some- times be approached about the possibility of participat- ing in the insurgency. Other state-controlled bureau- cracies were decapitated as well, but these leadership gaps did not always last for long. In the south and the Shi'ite sections of Baghdad, Shi'ite clergy and their supporters quickly established their leadership over a variety of local government institutions.” Many of these people were affiliated with Muqtada Sadr's Sadr II movement (so named to indicate continuity with his murdered father's charitable activities). Holdover of- ficials within the establishments seized by the Sadrists or other groups were quickly made to feel unwelcome or even in danger unless they pledged loyalty to the new leadership. These new political leaders often had no concept of the technical or administrative issues associated with the enterprises that they seized. Nev- ertheless, the rise of Shi'ite clerics to fill the political vacuum in their own community is not surprising. The Shi'ite political establishment was one of the only organized forces outside of the Ba'ath Party in Iraq at the time of the invasion. Moreover, it had a strong 25 – and loyal following, a system of self-financing, and a record of long-standing persecution by the regime. Later, the Sadrists lost some of their initial power fol- lowing Muqtada Sadr's political and military confron- tations with the Iraq government led by rival Shi'ite politician Nuri al-Maliki. Many Ba'athists who held ranks below the highest four levels of the Ba'ath Party were also purged un- der the 2003 de-Ba'athification order, because it was often difficult to discern an individual's rank within the Ba'ath Party. Often such standing was not clear to those around the person, and a large number of re- cords were destroyed in the immediate aftermath of the invasion and the looting of Iraqi government of- fices that occurred following the fall of the Saddam regime. Individuals who held important administra- tive positions were therefore often simply assumed to be high-ranking Ba'athists and removed from office. Ironically, some individuals who were not important in the Ba'ath Party were strong pro-Saddam sympa- thizers, while some important Ba'athists sought to rise within the Iraqi government and bureaucracy through whatever means available. Allowing junior officials to assume the jobs of their former superiors did not necessarily lead to a bureaucracy that was inherently more anti-Saddam or pro-democracy. The decision to place Chalabi in charge of the de- Ba'athification process was also unfortunate. At least some U.S. leaders were aware of exactly what they were getting with a Chalabi-led de-Ba'athification Commis- sion, and they should have understood that he was not likely to show restraint on this issue.” Chalabi had been an advocate for wide-ranging de-Ba'athification well before the war against Saddam had begun in 2003. He had previously published his concerns that t the United States would invade Iraq but would not at- tempt to eliminate all aspects of the Ba'ath Party with the comprehensiveness that he favored. In a February 19, 2002, Wall Street Journal editorial, Chalabi attacked what he called the plans for the future occupation of Iraq, which he apparently believed he understood on the basis of testimony before Congress by U.S. mili- tary and Bush administration officials. According to Chalabi, "[T]he proposed U.S. occupation and mili- tary administration of Iraq is unworkable. Unwork- able because it is predicated on keeping Saddam's ex- isting structures of government in place—albeit under American officers.” He went on to claim that, “Iraq needs a comprehensive program of de-Ba'athification even more extensive than the de-Nazification effort in Germany after World War II." Chalabi has often been identified as the least popu- lar member of the Governing Council among the Iraqi population at the time of his appointment by the IGC to head the de-Ba'athification Commission. His sta- tus as an exile caused at least some to view him as an outsider who had no experiences of the challenge of living under Saddam.” The strong and public ties Chalabi held to both Israelis and pro-Israeli figures in the U.S. Government were well-known and not universally appreciated throughout Iraq.” Later, the December 2005 elections underscored his unpopu- larity when his political party failed to win a single seat in the 475-person Parliament, despite a massive political campaign under the slogan, “We Liberated Iraq.” The decision to move forward with Chalabi at the head of the Commission rather than seeking a more reconciliation-oriented figure indicated a con- tinuing determination to impose a harsh peace on the Sunnis and anyone associated with the old regime. 27 This approach was consistent with the priorities of the senior Pentagon civilians who remained concerned that a regime similar to the one led by Saddam could reemerge. This danger was also worrisome to many of Iraq's Shi'ite and Kurdish leaders who were aware that the Ba'ath had previously come to power twice through coups. As noted, the Shi'ite religious parties and other community leaders were among the groups most inter- ested in comprehensive de-Ba'athification priorities. U.S. policymakers seeking to justify a more sweeping de-Ba'athification policy were quick to point out that failure to do this would potentially harm U.S. relations with these parties.” Nor is it difficult to understand the intense hatred Shi'ites and Kurds held for Saddam and the Ba'ath. Shi'ite religious parties, as well as the Shi'ite-dominated Iraqi Communist Party, had suf- fered intensively under Saddam, and most prominent members of these organizations had lost a number of friends and family members to torture and execution by the regime. The rise of a Shi'ite Islamic republic in Iran through revolution was particularly frightening to Saddam, who unleashed an especially high level of brutality against Iraqi Shi'ites who seemed even the slightest bit comfortable with the Iranian concept of Islamic government. An overly political definition of Shi'ite identity during the 1980–88 Iran-Iraq war was especially dangerous. Nevertheless, revenge (or jus- tice) was not the only motivation for the Shi'ite par- ties in supporting de-Ba'athification. Many of these groups also wanted as much power as possible for themselves. Destroying the political viability of the Sunni leadership in Iraq helped to move them toward that goal. Some Shi'ite leaders may have also hoped to reverse the situations of Sunnis and Shi'ites perma- 28 nently. In contrast to Iraq's first 8 decades of existence, Shi'ites would hold the important positions, and Sun- nis would be politically marginalized. Under these circumstances, some Sunni Arabs believed that they were being offered second-class citizenship at best. The CPA de-Ba'athification order was sometimes taken as at least a partial green light for some Iraqis to exact revenge on former Ba'athists who had per- secuted them or were their personal enemies. In- deed, a Shi'ite assassination campaign against former Ba'athists did take place, although it is doubtful that a more reconciliationist approach by the CPA would have prevented these outbreaks of violence, once the dictatorship had been removed." Many of these assas- Sinations were carried forward in a highly profession- al manner, rather than as frenzy or sloppy revenge at- tacks. It is correspondingly possible, if not likely, that Iranian intelligence units coordinated with friendly Shi'ite groups to ensure that Ba'athist enemies of Teh- ran were never in a position for them to cause trouble for Iran again.” According to the London-based news- magazine, The Middle East, Iranian Supreme leader Ali Khamenei put the commander of the al-Quds Force in charge of setting up a network of covert operatives in Iraq as early as September 2002, with the mission of expanding Iranian influence in that country in the aftermath of the invasion.” If Chalabi hoped to use the de-Ba'athification Com- mission as an avenue for his own rise to power, he was deeply disappointed by the outcome of the 2005 election. While he may have helped to create a power vacuum by purging a number of potential rivals, he did not have the ability to fill it through the electoral process. Rather, the most important players in Iraq at this stage were quickly proven to be the Shi'ite reli- 29 gious parties who were also enthusiastic supporters of de-Ba'athification. After the election, Chalabi moved in and out of a variety of governmental jobs, which he held for various lengths of time. Throughout his political maneuvering, he was unable to obtain real power within the top leadership of the government. As noted above, many Iraqi Sunnis viewed the effort to remove large numbers of Sunni leaders and bureaucrats from power through the vehicle of de- Ba'athification as part of a new political system in which Shi'ites would dominate Sunnis. The politiciza- tion of sectarian differences also led Iraqi political par- ties to adapt an approach whereby they viewed failing to fill a political post with one of your supporters or allies as tantamount to allowing that post to be filled by enemies.” In addition to Sunni Muslims, some “establishment Shi'ites" had also risen to high ranks within the Ba'ath Party and were also caught up in de- Ba'athification. A key problem here is that Saddam ac- tively reached out to secular Shi'ites to serve as “dem- ocratic ornaments," while attempting to marginalize the Shi'ite clergy, which he felt was at least potentially loyal to Iran.” Some secular Shi'ite leaders, including those with advanced degrees from Western universi- ties, took the bait for a variety of reasons including the hope that they could gain some reasonable level of patronage for their own communities. Some of these people were also well-educated and talented enough to be of real use to the regime in performing adminis- trative tasks. These links with the regime allowed such individuals to become targets for de-Ba'athification in ways that the more persecuted opposition clerics did not once the regime had been removed.” The most prominent example of the problems faced by “establishment Shi'ites" was the case of Saa- SOf Wed lich real ed h doun Hammadi, the former Iraqi premier who died of leukemia in Germany in March 2007.” Saadoun Hammadi had previously served as Iraq's Foreign Minister, Deputy Prime Minister, Prime Minister, and most recently, Saddam's last Speaker of the Assem- bly, thus becoming the highest ranking Shi'ite within the regime. Hammadi held a Ph.D. in economics from the University of Wisconsin and has been described as having a “thoughtful and scholarly demeanor.” He also is the author of a number of academic articles on Arab affairs and political philosophy.” Hammadi favored economic and political liberalism in the past, and was presented to the world as a reform prime minister after the 1991 Gulf War. He apparently took his reform charter a little too seriously for Saddam and was removed for overzealousness after 7 months in power.” As an articulate, respected Shi'ite intellectual who held high-profile/high-prestige government posi- tions, Hammadi helped give Saddam's government the appearance of broad-based Iraqi support across re- ligious sects. Saddam thus presented Hammadi with the option of being co-opted and in return gaining a few crumbs of power for himself and some economic assistance for his Shi'ite supporters. This Faustian bargain was occasionally made available to Western- educated secular intellectuals, but it was almost never an option for important members of the Shi'ite clergy. Although Saddam sometimes sought to appear reli- gious, formal clerical participation in the Ba'athist government was largely unacceptable to him. Certain- ly, no ayatollah would hold any of the governmental positions Hammadi held. Hammadi was arrested and placed in prison shortly after the U.S.-led invasion, while his son and members of his al-Karakshah tribe Stringently protested his arrest on grounds that he did 31 m; had been fired by Ba'athists from various bureaucra- cies under the Ba'athist regime became strong candi- dates to replace them following the change of regime. The problem here is that such individuals sometimes (perhaps often) were fired for nonpolitical reasons, including incompetence and corruption. Upon being returned to their former jobs or those of their former Supervisors, they returned to old patterns of behavior, showing little responsibility, effectiveness, or commit- ment to even a limited work ethic. To be fair, it might be noted that these people had no monopoly on the shortcomings noted here. Most Iraqis had never had any preparation to work in an efficient, modernizing bureaucracy, and corruption permeated the society during the Saddam years as it still does.” At various times, the Iraqi government announced that it was relaxing the de-Ba'athificiation policy, of- ten as a response to U.S. pressure. Chalabi would usu- ally announce the policy “changes" and then provide grandiose projections of how many people would be rehabilitated under new more lenient rules. In early 2007, for instance, he publicly agreed to soften the de-Ba'athification policy, announcing that his office had begun removing hiring restrictions from former Ba'athists who had not committed crimes during the Saddam years. Elaborating on this change, he stated that more than 2,300 former high-ranking Ba'ath Party members were either being reinstated in their former jobs or granted pensions.” On the same day, Chalabi stated that over 700 former Ba'athists had returned to their old government jobs, suggesting that the balance of the 2,300 people he cited were given pensions if his figures are correct.” Chalabi’s commitment to reform nevertheless remained tactical, and there is no inde- pendent evidence for the figures he cited. Additional- 33 MILITARY IMPLICATIONS OF DE-BA’ATHIFICATION The decision to dissolve the Iraqi Army and the Ba'ath Party within the first few days of establish- ing the CPA administered an overwhelming blow to organized Iraqi life. This radical shock therapy was deemed by some members of the Bush administration as vital to the establishment of a stable democracy in Iraq. Of all of the CPA actions in this time frame, the abolition of the Iraqi Army was the most controversial and disconcerting to many Iraqis, who often viewed the military as something more than a pillar of the Sad- dam Hussein regime. Supporters of the decision often claim that the Iraqi Army dissolved itself, and that the reality of the post-war situation was simply being rec- ognized. This argument implies that the United States only had only two choices, reconstituting the 600,000- man Iraqi Army in its Saddamist form or bringing the Iraqi Army down to zero. The choice, however, was never that binary, and the CPA order was issued at a point when U.S. Army General David D. McKiernan and various CIA officials were already working on a third option, that of reconstituting certain units of the Iraqi military on a voluntary basis under vetted offi- cers.” These efforts had to be discontinued following the CPA announcement. Armed resistance to U.S. forces at Some level fol- lowing the invasion was probably inevitable, no mat- ter how well the post-war reconstruction effort was handled. The question was, would this resistance comprise small groups of terrorists or would it en- compass much larger forces drawn from alienated Social groups that were able to organize into a strong 35 network of resistance organizations. At this stage in the conflict, the Bush administration was loath to ad- mit that segments of the Iraqi population were waging war against U.S. forces rather than welcoming them. At a June 2003 press conference, Secretary of Defense Rumsfeld stated, "I guess the reason I don't use the phrase ‘guerrilla war' is because there isn't one.” In general, the administration seemed to believe that the Iraqis would be sufficiently grateful for liberation that they would be granted sweeping ability to do anything they wanted in Iraq without much of a backlash.” This view was emphatically reinforced by some of the most pro-war Iraqi exiles who maintained that Iraqis were so oppressed that they did not care about much else other than their deliverance from Saddam Hussein.” The de-Ba'athification order, as unpopular as it was with Sunni Arab Iraqis, was not as unpopular as the disbanding of the Iraqi Army. Yet, if the United States was determined to implement a de-Ba'athification order, the rationale for dissolving the Army becomes much less clear. Senior Ba'athist officers could have been retired under the de-Ba'athification order, and low-ranking Ba'athists and non-Ba'athists could have been offered the option to remain in the military pro- vided that they were not complicit in regime crimes. Ba'ath political officers, who were often resented by regular army officers, could easily have been removed from service, and elite units with special loyalty to Saddam could have been dissolved.” The Iraqi Army under new leadership could then have been used to help provide order rather than be left disgraced with many of its members facing destitution. The special relationship of the Iraqi Army to Iraqi society went far beyond Saddam. Even a number of anti-Saddam Iraqi exiles urged that it not be abolished.” 36 - amined by special technical experts. The special tech- nical experts confirmed in fact that the records were authentic and not tampered with. One of the key find- ings of those records which was shared with [CPA Di- rector for National Security and Defense) Mr. [Walter] Slocombe, was that in fact you did not have large-scale Ba'ath issues in the army until you got to the major general rank, and at the major general rank, 50 percent of the major generals were Ba'athists and 50 percent weren't.” An important caveat is in order here, since the Iraqi Army was extremely top heavy and had more than 10,000 generals.” Nevertheless, the database that Colonel Agoglia mentions could have been an invalu- able tool in reconstituting the Iraqi Army and then using it to help provide security for the new govern- ment. This effort would have to include extensive use of other intelligence means to confirm all aspects of the database to the greatest extent possible. In the aftermath of CPA Orders 1 and 2, Ba'ath offi- cials became natural allies to the angry and financially troubled ex-soldiers of the Iraqi Army after the Army was disbanded, with no effort made to recall those former soldiers who may have remained interested in serving.(The ability of senior Ba'ath leaders to obtain 9 and provide funding to the insurgency was particular- ly important in helping to organize it into an effective force able to include unemployed and desperate Iraqis willing to strike at U.S. forces for money. Ba'ath fund- ing for such efforts appears to have been drawn from a variety of sources. Some Ba'ath leaders had signifi- cant reserves of cash within Iraq when the invasion oc- curred. This group included many mid-level Ba'athist officials as well as more senior leaders.” Others had access to funds in foreign banks, particularly in Syria. 38 the |Oſt that all- hen The Syrians, at this time, seemed willing to turn a blind eye to many provocative Iraqi Ba'athist activities out of some ideological kinship and, more concretely, from a fear that Damascus might also be targeted for a U.S.-led regime change, unless the United States was bogged down elsewhere. Saddam loyalists who were trusted enough to have access to millions of dollars from the old regime were naturally few in number, but their ability to provide funding to unemployed ex- Soldiers at the early stages of the occupation served as the lifeblood of the emerging insurgency. Later, the in- surgency was to become dominated by al-Qaeda radi- cals, with their own funding sources from outside the country (as well as various “taxes" and asset seizures within Iraq). The senior Ba'athists able to distribute money would probably have remained committed en- emies of the new Iraq under any conceivable scenario, but it would have been much more difficult for them to establish the initial insurgent networks without the large and discontented groups created by CPA Orders 1 and 2. According to Stanford professor and former CPA Senior administrator Larry Diamond, there were im- portant warning signs that the former Iraqi officers would create severe problems if they were not given other options than simply walking away from their military careers with virtually nothing. Bremer did not seem prepared to listen or initially adjust his policies on military pensions and possible return to service by vetted individuals in the face of changing events. Ac- cording to Diamond, “Bremer has set out on a decisive course—establishing the American political occupa- tion of Iraq, dissolving the Iraqi Army and institut- ing a sweeping process of de-Ba'athification—and he did not want to be steered off course.” The refusal 39 up to fail on these grounds. The U.S. delay in recog- nizing the dangers inherent in Iraq's lack of adequate military forces correspondingly gave the Iraqi insur- gents huge advantages in establishing control over ar- eas that would subsequently be outside the control of the central government. The U.S. leadership therefore made a deliberate decision to deny Iraq the type of force that could allow the Iraqi government to survive in the absence of U.S. forces in order to ensure that a Ba'athist coup from the military could not take place, although U.S. leaders did not seem to see this as the trade-off at the time. The Ba'ath Party's “Political and Strategic Pro- gram," issued after Saddam's ouster, stated that its immediate priority was to “expel the occupation forces from Iraq and preserve the country as a unified homeland for all Iraqis.” Alienated Ba'athists did not, however, always join Ba'ath resistance organiza- tions to fight against the Coalition and the Iraqi gov- ernment. Many who wanted to fight reached out to violent Islamist groups after repenting their "sins" of supporting Saddam's secular regime.” These people then fought against the United States and the Iraqi government as supposed Islamic warriors. Some of this solidarity may have resulted from a decision by various Sunnis to resist the Shi'ite-dominated govern- ment by whatever means available. More pragmati- cally, there is also the possibility that over time the Islamist groups would be more effective than the Ba'athists in finding foreign sources of funding. Some Sunni Iraqi leaders were also forced, or at least strong- ly pressured, to support the fighting by al-Qaeda once it had established itself in that country. Many insur- gents would again change sides when their al-Qaeda allies became too authoritarian to tolerate and when 41 the United States offered to fund the anti-al-Qaeda, "Awakening Councils." The ability of Iranian intelligence and paramilitary organizations to function in Iraq was also aided by the portion of CPA Order 2 dissolving Iraqi intelligence organizations. This order and the de-Ba'athification order made it difficult, if not impossible, to return key personnel to intelligence duties focused on anti- Iranian counterintelligence and the containment of Iranian power. This situation is especially tricky since the intelligence organizations were important pillars of the regime, and were correspondingly riddled with potential war criminals and human rights violators. Saddam first came to power by consolidating his con- trol over these organizations, and it appears that his youngest legitimate son, Uday, was being groomed to assume future powers as president by serving an apprenticeship within the intelligence organizations. Under a less sweeping purge, trade-offs would have had to be considered in assessing the criminality of various officials as juxtaposed against their useful- ness in opposing Iran. The example of the CIA's use of former German Nazi-era general Reinhard Gehlen and his intelligence gathering organization against the Soviet Union after World War II may have served as a starting point for consideration of how this could be done.” EFFORTS TO REVERSE THE EXCESSES OF DE-BA’ATHIFICATION The idea that the United States could enter Iraq and then rapidly depart after removing much of the leadership from all major administrative structures, as well as the military as a whole, reflected an optimism 42 that was difficult to justify. This belief was influenced by the experience of some Eastern European countries where various communist parties were removed from power without the types of problems that were to oc- cur in Iraq.” Feith suggests that Iraq could have been turned over to exiles, but it was certainly not clear how these people could have established civil order or im- plemented a set a policies that had already been seen to create a major Sunni insurgency. In the years since the invasion, various neoconservatives have stressed that the United States should have trained a large exile army prior to invading Iraq and that by choosing not to do so helped to foreclose the option of installing exiles.” This approach was supposed to be based or at least inspired on the model of the French resistance in World War II and in some extreme versions may even have considered Chalabi to be a latter-day Gen- eral Charles De Gaulle. Even if one accepts the logic of such an argument, the fact remains that the United States did not have such a force in place in 2002-03, and this in no way altered the ideologically based be- lief that the Iraq war would not require a significant occupation force to keep order after the Iraqi regime was defeated and the Iraqi Army was disbanded. The U.S. difficulties in Iraq may also have increased due to the hostility of neighboring Arab governments and populations to policies that they viewed as anti- Sunni, such as de-Ba'athification. After the ouster of Saddam's regime, there were few credible news out- lets for average Iraqis to use in an effort to understand the situation within their country. Newly emerging Iraqi newspapers were often hopelessly biased and based on a poor understanding of journalistic stan- dards, which were unknown in Saddam's Iraq where the media's chief function was propaganda dissemi- 43 nation. The fledgling Iraqi television network was almost totally ignored in preference to the pan-Arab news stations such as al-Jazeera and, to a much lesser extent, al Arabibiya as well as Iran's al-Alam news broadcasts in Arabic. Some of the pan-Arab media outlets, especially al-Jazeera satellite television, were particularly hostile to de-Ba'athification Commission leader Ahmad Chalabi for a variety of reasons, not all of which involved his activities on the Commission. In response, Chalabi was interviewed on al-Arabiya television where he stated that al-Jazeera was com- pletely infiltrated by Iraqi intelligence. This statement is widely viewed as untrue, reckless, and a little des- perate. This feud between Chalabi and al-Jazeera con- tinued for some time." Moreover, as the difficulty of stabilizing Iraq be- came increasingly clear, Bremer became more open- ly critical of Chalabi's methods of conducting de- Ba'athification, which went beyond the scope of the original order. In April 2004, Bremer moved to narrow de-Ba'athification in response to the abuse of the sys- tem and to establish a more reasonable set of policies to reverse Iraq's escalating violence. He stated that the policy had been applied "unevenly and unfairly." By this time, the crisis between the Sunnis and Shi'ites was exceptionally serious, and the outlines of a po- tential civil war were becoming increasingly clear. In response, Bremer hoped to empower a more recon- ciliationist Iraqi leadership that would be able to help stem the tide of insurgency. He believed that he found the right individual to lead this effort in the person of Ayad Allawi, a secular Shi'ite leader who believed in achieving national unity by reaching out to Sunni Arabs and Kurds. ----" 44 Interim Iraqi Prime Minister Ayad Allawi was in- stalled in office by the United States on June 1, 2004. On June 28, 2004, the CPA handed over formal po- litical power to Allawi and the Iraqi Interim Govern- ment. Allawi was a longtime opponent of deep de- Ba'athification, and sought to limit the scope of the de-Ba'athification effort throughout his time in office. In a strategic vision that anticipated the Awakening Councils, he also strongly favored efforts to reconcile with Iraq insurgents and thus draw them away from hardcore Saddamists, Iraqi al-Qaeda members, and foreign terrorists. He also hoped to negotiate with Iraqi Sunnis and peel them away from their emerging alliance with foreign fighters such as al-Qaeda in Iraq leader Abu Musab al-Zarqawi. Allawi pushed for- ward with his efforts at reconciliation by seeking to al- low Ba'athists without blood on their hands to return to state jobs. Chalabi’s de-Ba'athification Commission seems to have significantly reduced its purges in re- sponse to Allawi's pressure. In justifying this policy, Allawi stated, “This country needs every single citi- Zen" and “we will not repeat the policies of Saddam Hussein, who favored some while excluding most of the population.” He is also the only post-2003 Iraqi Prime Minister not to visit Iran, although he was in- vited to do so.1% Bremer presented the decision to appoint Ayad Al- lawi as Interim Prime Minister of Iraq as a decision made by the Iraqi Governing Council more or less on its own, which he was asked to ratify. This interpre- tation of events is almost universally discounted as a useful fiction designed to help Allawi by indicating that he was selected by other Iraqis and not by the U.S. leadership. Most Iraqis believe that the U.S. Govern- ment chose Allawi, and this interpretation permeates 45 most published accounts of the event. Whatever the origins of Allawi's appointment, he nevertheless ap- peared to have maintained a well-reasoned and for- ward-looking agenda for the Iraqi future, although he also had his shortcomings on such issues as con- trolling corruption. Allawi met Iraqi opposition to his relaxation of de-Ba'athification from predictable sources. The leadership of SCIRI stated that “improp- er persons" were being given positions in the security field and this was a violation of the principles of the new Iraqi government.” Feith states in his memoir that the CIA and State Department favored the Allawi appointment, but he felt that Allawi and his party were insufficiently committed to democracy. He noted that Allawi's Iraq National Accord (INA) Party had leaders who were, "supported by Sunni-controlled Arab governments [and] wanted the country's Sunni-controlled military to continue to play a key role in Iraq." It is inter- esting that Feith was concerned not simply about the dangers presented by a Ba'athist leadership for the Iraqi Army, but he also seems to have been worried about a Sunni leadership, in general, becoming pow- erful within Iraq. Additionally, it is not clear why an exile group receiving support from other Arab coun- tries allied with the United States was considered as dangerous as Shi'ite parties accepting support from Iran. Feith also notes his own serious concern about the Iranian connection to other Iraqi political parties, but does not appear to view them as any more or less troubling than Allawi's links to Arab nations such as Saudi Arabia.” Prime Minister Allawi's strongly reconciliation- ist approach might have made progress in defeating the insurgency over time, but the decisive defeat of 46 al: h | his party in the 2005 elections ended the chances for his approach to go forward. Sunni Arabs boycotted the election, and most Shi'ite and Kurdish voters sup- ported parties with a clear and direct sectarian or eth- nic agenda. Allawi was also undermined by reports of Staggeringly high levels of corruption in his govern- ment (although this scandal did not involve him per- sonally).” Allawi's term as prime minister ended on April 7, 2005, when he turned power over to Ibrahim Jaffari of the Shi'ite Islamist party, al-Dawa. Jaffari remained in power until May 2007 and was then fol- lowed by Nouri al-Maliki of the same party. Both in- dividuals were selected through a process of internal bargaining in a Parliament dominated by Shi'ites and, to a lesser extent, Kurds. Distrust among the Shi'ite groups led to the choice of two consecutive Dawa Party prime ministers, since Dawa unlike many of the other parties has no militia, and its leaders were there- fore considered safe compromise candidates. The principles of the 2003 de-Ba'athification de- cree were also enshrined in the 2005 Iraqi Constitution that was largely put together by Shi'ites and Kurds (Sunni Arabs unwisely boycotted the election for the Constitutional Assembly). According to Article 131 of the Constitution, “The High Commission for De- Ba'athification shall continue its functions as an inde- pendent commission, and in coordination with the Ju- dicial Authority and the Executive institutions within the framework of the laws regulating its functions. The Commission shall be attached to the Council of Representatives." The inclusion of this statement in the Constitution was not taken well by many Sunnis, although there were a few limited efforts to reassure them. The de-Ba'athification Commission became much more active after Allawi left office and the new 47 Constitution was ratified. One of their chief functions seems to have been removing people who managed to regain their jobs while Allawi was prime minister. By the summer of 2006, there was another softening of de-Ba'athification with significant numbers of people reportedly returned to the Ministries of Information, Interior, and Defense.” By this time, various political parties had established control over at least some im- portant ministries and the domestic situation in Iraq appeared at its nadir. Many within the Iraqi govern- ment may have felt pressure to announce some prog- ress on reconciliation-related issues simply to prevent the United States from giving up on Iraq. THE "ACCOUNTABILITY AND JUSTICE ACT OF 2008” The passing of the Accountability and Justice Act in Parliament came in early 2008 and was meant to reform the process of de-Ba'athification, as well as re- verse some of its earlier excesses. The law was passed after long and tortured debate within the Iraqi Parlia- ment in which many Shi'ite leaders made it clear that they were content with a political system that margin- alized the Sunni Arab community, which they noted was only 20 percent of the Iraqi population. An earlier effort along the lines of the 2008 act has been derailed by Shi'ite opposition, including statements made by Grand Ayatollah Sistani and the three other most se- nior Grand Ayatollahs in Iraq.” The 2008 law was enacted to respond to one of 18 benchmarks set by the U.S. Government to mea- sure political reconciliation in Iraq. Moreover, U.S. observers often viewed it as a particularly important benchmark since it dealt with an effort to heal the 48 Shi'ite-Sunni divide that was poisoning Iraqi politics and undermining national reconciliation. The Iraq Parliament was correspondingly under tremendous pressure to produce some sort of a restructured and reformed de-Ba'athification law. It was also supposed to be designed to convey the message that there was now a place for the Sunni Arab elite in helping to gov- ern Iraq. Many Ba'athists who were purged from their former positions were told that they could apply for pensions and even reinstatement in their jobs as a re- sult of this law. There were nevertheless problems with the new law. Many Sunnis did not view the law as liberal re- form. Rather, they charged that the obtuse and am- biguous language of the law could be used to con- duct further purges of ex-Ba'athists and fire soldiers and state bureaucrats who were then employed by the government. More dramatically, some former Ba'athists claimed that the law was a ploy to lure them into situations in which they could be killed.” Those Ba'athists in exile outside of the country have been particularly suspicious. Others assume that there is no future for them in ministries dominated by Shi'ite politicians and activists, even if their lives are not ac- tually threatened by working there. The law therefore did not bridge differences between the Sunni and Shi'ite communities to the extent that U.S. observers had hoped it would. Under the 2008 law, the Parliament was to ap- point a new board and a new staff for the restructured Commission. The individuals appointed to the board were supposed to act under amended rules and take a more reconciliationist approach to the issue of de- Ba'athification. Unfortunately, Parliament failed to meet its obligations to make these changes at the time 49 —r of this writing. This failure occurred through both pro- crastination and possibly because some Members of Parliament were unprepared to approve a list of gov- ernment appointments to the commission for fear that a successful effort to address this issue could be politi- cally beneficial to Prime Minister Maliki. Under these circumstances, de-Ba'athification board chairman Ali Faisal al-Lami (then released from detention) stated in January 2010 that his board should continue to func- tion as the new Accountability and Justice Commis- sion (AJC) until a new board was in place. This ar- gument was not accepted by a number of critics, and the organization continued to exist under an uncertain legal status. Although Chalabi was no longer playing a day-to-day role, it was widely believed throughout Iraq that he continued to pull most of the strings. THE CONTINUING LEGACY OF DE-BA’ATHIFICATION IN IRAQ A new crisis for Iraqi national unity arose on Janu- ary 6, 2010, when the AJC under Chalabi and al-Lami announced the disqualification of 511 candidates in the March 7 parliamentary elections for supposed Ba'athist connections. This was done under the old quasi-legal commission's board in a sloppy and hur- ried manner that introduced an earthquake into the Iraqi political system. Fifty-nine of those identified for disqualification were cases of mistaken identity where individuals had names similar to those found on the commission's database. Ayad Allawi's Iraqiya Alli- ance suffered most under these initial rulings, with 72 of its candidates identified as disqualified. Most of the individuals identified for disqualification were from Sunni religious or secular parties, which were often 50 0. supportive of Sunni Muslim political rights. Many t of the most powerful Shi'ite politicians, including W. Muqtada al-Sadr, supported the ban. To his credit, | Iraqi President Jalal Talabani strongly denounced the i. ban. Iraqi critics began referring to the election process Sº as the “Iranian form of election." In Iran, a Council of | Guardians evaluates the credentials of each candidate for parliament and makes a decision about their fitness lſ. to hold office as a way to constrain democracy and # limit voter choice. Numerous international observers àſ. question the legitimacy of these Iranian procedures. mi American presidents and other political leaders have im often been among those most critical of such a vetting ſ: process. \| The commission's decision to disqualify such a large number of candidates in a highly opaque process also threatened to undermine the legitimacy of the election with Sunni voters. In response to complaints about the commission's actions, an appeals court ini- tially ruled that these candidates could run for election and clear up the issue of Ba'athist affiliation later. Un- der reportedly heavy pressure from Maliki, the court then reversed itself on February 12, 2010, and disquali- fied 145 candidates.” If these reports are true, the AJC was enjoying political cover from the Prime Minister's office. Nevertheless, these candidates and some who were disqualified earlier were quickly replaced on their party slates so that their coalitions could contin- ue to contest the elections with minimal disruptions. The Iraqiya political coalition was particularly hard- hit by these changes because of the disqualification of two of its leading Sunni candidates, Salah al-Mutlaq and Dhafir al-Ani. Mutlaq, who headed the National Dialogue Front of the Iraqiya coalition, was replaced by his brother on the national slate. Later, in Decem- ber 2010, he became one of Iraq's two vice presidents. 51 As this process unfolded, the legal and ethical lim- bo of this entire approach to candidate disqualifica- tion was severely tainted by the fact that both Chalabi and al-Lami were running for seats in Parliament, and thereby may have hoped to benefit from the disquali- fication of rivals through the commission, which they dominated. The senior U.S. military leadership in Iraq was deeply disturbed by this process, and accused both Chalabi and al-Lami of being under the sway of Iran, whose leadership had a potential interest in a weak, divided Iraq that could not rise to become a regional power. In a public presentation at a Washington, DC, event, General Raymond Odierno stated that the two Iraqi politicians “are clearly influenced by Iran.” He added that, “We have direct intelligence that tells us that." General Odierno's comments seem to raise the concern that Iran was using Chalabi as a tool to undermine Iraqi national unity. Many Iraqi leaders were more direct and strongly raised the possibility that both Chalabi and al-Lami were implementing Iraqi election procedures in coordination with Tehran for their mutual benefit. Ironically, Chalabi and al-Lami's quasi-legal ac- tivities seem to have boosted Sunni Muslim turnout and may have caused the two leaders to lose support among Shi'ite voters. Al-Lami was not elected in the 2010 election, although Chalabi won a seat because of his party's inclusion within the Iraq National Alliance coalition list. More generally, the 2010 election did not lead to a clear outcome. Rather, strong competing po- litical blocs emerged and were unable to put together a coalition government for 9 months. Iraqiya) won a plurality of 91 seats in Parliament, but was unable to put together a coalition government. Maliki's State of Law alliance, which won 89 seats, was eventually able 52 lim. lić. hâlî lºſt jū. h t; iniº idbº Yilm | Wö. egº: InDſ. he W "AH. tells; to ſix toºlſ lead: to do so, but only by including around 40 followers of radical anti-American cleric Muqtada al-Sadr. A few months prior to the U.S. military withdraw- al from Iraq, the Maliki government seemed intent on repeating previous mistakes by alienating Sunni Mus- lims. This time Maliki's actions occurred in a political environment where the United States had a decreasing level of influence. In October 2011, Iraqi authorities ar- rested at least 240 individuals whom they identified as former leading members of the Ba'ath Party or former important members of Saddam Hussein's army, The charges seem to imply that the arrests involved a plot by these individuals to seize power after U.S. forces depart the country.” Some Iraqis also appear to be- lieve that the arrests may not have involved a specific plot as much as a general feeling that these individuals were a threat and that arresting them was a useful pre- caution/In defending his actions, Maliki stated that he continued to be concerned about Ba'athist “coups and conspiracies.” In a Washington Post opinion article addressed to the American public, he also asserted that, "I refute characterizations that the detentions were a sectarian action based on political motives.” Nevertheless, Iyad Allawi's Iraqiya political bloc has been harshly critical of the arrests and demanded the release of “all detainees held on false charges.” Serious Sunni-Shi'ite differences began to escalate further, 1 day after the completion of the U.S. military withdrawal from Iraq in December 2011. At this time, Maliki moved against one of his most important Sunni rivals by issuing an arrest warrant for Vice President Tareq al-Hashemi for supporting terrorism by run- ning a death squad. The Maliki government also ar- rested three of Hashemi's bodyguards and charged them with terrorism. More Sunni arrests on similar 53 The increasing Sunni-Shi'ite differences within Iraq also paralleled the reemergence of al-Qaeda in many parts of the country. According to an August 2011 statement by Iraqi Lieutenant General Hussein Ali Kamal, Iraq's Deputy Interior Minister for Intelli- gence, “There was a thought that al-Qaeda had ended in Iraq. No, they regrouped and now the third gen- eration of al-Qaeda is working actively to reorganize itself with weapons and training.” These words are difficult, if not impossible, to dispute. Terrorist bomb- ings continue throughout the country, some of them with spectacular coordination and large numbers of casualties. Pilgrims visiting Shi'ite religious sites are a favored target. The government has responded to these outrages with considerable brutality against Suspected Sunni terrorists to avoid charges of incom- petence. Iraq is now in a position where it will have to man- age its sectarian differences without U.S. forces in that country. While the United States initiated the de- Ba'athification campaign, the Iraqi leadership twisted these policies into instruments of sectarian revenge, which its Shi'ite-dominated government is never quite willing to give up. This legacy will have to be borne by Iraqis in the absence of U.S. troops who attempted to contain sectarian differences in that country for almost 9 years. Both Iraqi Sunnis and Shi'ites (although not Kurds) celebrated that final departure of U.S. combat troops from their country, but such celebrations were held separately. Iraq will either manage its sectarian problems, or it will fail to do so. This is the choice that awaits Iraqis and will have to be faced at some point except in the event of a permanent U.S. military occu- pation, which is unacceptable to both countries. Any actions against the Sunnis that appear to be war crimes 55 have infinite potential to make the situation in Iraq much worse. Iran already has a major role inside Iraq, and the Sunni Arab nations are unlikely to allow this influence to go uncontested. No Sunni nation wanted to supply arms to Iraqi insurgents and then find that American soldiers had been killed with weapons that could be traced to them. Now, in the absence of U.S. forces (except contractors), the rules have changed. EVALUATION OF THE IRAQI DE-BA’ATHIFICATION PROGRAM The de-Ba'athification program as it was practiced in Iraq is widely understood to have been deeply flawed in both its original conception and in the way that it was carried out. It seriously complicated all of the major challenges that the United States and its allies within Iraq faced after the ouster of Sad- dam Hussein's regime. These problems included the alienation of Sunni Arabs, the politicization of sectar- ian differences, and the rise of the Sunni insurgency. De-Ba'athification also helped to cause a number of Sunni Arab populations in neighboring countries to become more virulently opposed to the U.S. presence in Iraq, helping to undermine the U.S. presence in the Middle East and create difficulties for pro-American governments throughout the Arab World. A less dra- matic approach to restructuring the Iraqi government would have substantially reduced these problems, but it would also have been seen as rendering it less likely that the United States could completely change the character of the Iraqi nation to be a close and reli- able ally. Many Bush administration leaders believed that such a transformation was possible and that a new Iraq would favor permanent U.S. bases, seek U.S. 56 investment, support cheap oil sales, and work with U.S. allies. The decisions that were made with the idea of maximizing these goals nevertheless did very little to advance them, while instead aggravating massive problems with security and stability. A key failure here may be a lack of understanding about the limits of what can be accomplished in the aftermath of extra- regional coercive regime change in countries that are disinclined to accept foreign tutors for very long. U.S. decisionmakers did not want to allow Ba'athist values to continue in Iraq, nor did U.S. leaders wish to allow any kind of a reformed Ba'ath to be used as a vehicle to oppose the U.S. presence and agenda in Iraq through participation in the government and admin- istration of the country. This approach may have been aggravated by the embarrassing failure to find an Iraqi nuclear weapons program or even residual chemical and biological weapons programs. The administra- tion, by every indication, believed such weapons ex- isted prior to the war and, in their absence, needed to justify the intervention in other ways. The emergence of a disarmed Iraq in which some Ba'athists were still present in the government was probably not some- thing they would view as acceptable within these pa- rameters. Nor was a new Iraq with a traditional Arab foreign policy of opposing the West and Israel seen as a particularly compelling result for U.S. casualties and the financial cost of the war. The new Iraq was expected to look a lot more like America, or at least Turkey, than the Arab World.” The worst U.S. decision made in Iraq during this time frame was not, however, a single choice. Rather, it was a combination of two decisions that reflected the desire of the United States to destroy real and imagined vestiges of the old regime at any cost. The 57 decision by the CPA to engage in comprehensive de- Ba'athification and dissolve the Army simultaneously created the building blocks of the insurgency and pro- vided it with the organizational capacity to create that insurgency. As noted, the Ba'ath Party did not control the military primarily through propagandizing the senior officers. Instead, it used a system of political officers and counterintelligence officers throughout the military as a way of bending the military to the regime's will. While still on balance an unwise policy, deep de-Ba'athification would seem to have preclud- ed the need for the even worse decision of disestab- lishing the military. This one major advantage of deep de-Ba'athification was squandered when the second CPA order was issued. As the increasingly harsh U.S. approach unfolded, people treated like enemies cor- respondingly became enemies) Attempting to understand Saddam Hussein's rule by assuming that he was a committed Ba'athist seek- ing to live up to Aflaq's ideals was also a serious mis- take. Modern totalitarian regimes never rule in the name of naked power. They have an ideological cloak that is meant to mask the centrality of a terror state, Iraq under Saddam had a parliament, but parliamen- tary power was known to be a fraud. No one could have seriously held Iraq's Parliament responsible for Saddam's crimes. In the same way, the principles of Ba'athism are not necessarily offensive to all anti-Sad- dam Iraqis or Arabs more generally. Saddam did not build a criminal regime to serve Ba'athism. He built a Ba'athist infrastructure to support a criminal regime. The United States was correct to outlaw the Iraqi Ba'ath Party due to its hijacking by a criminal regime, but there also should have been a much greater will- ingness to tolerate individual noncriminal Ba'athists 58 even if they did believe in Arab nationalism, secular- ism, and Arab modernization. - It might also be noted that Ba'athism is not wide- ly viewed as the wave of the future in Middle East politics. Angry young men are seldom inspired by it, and are much more likely to join radical Islamist or- ganizations if they seek to confront the West. In their struggle with the Israelis, Palestinian radicals are now much more likely to turn to the Islamist organization Hamas than they are to work with Fatah, the mostly secular former torchbearer of the Palestinian cause. A few Palestinians even consider Hamas and Islamic Jihad to be too tame and are seeking to affiliate with al-Qaeda-like organizations.” Some vigilance against the virulent alternatives to Ba'athism might have been considered. The simplistic belief that the Ba'athists were all Nazis and most other Iraqis were waiting for a Chalabi-like figure to lead them into a secular, pro-Western government created a situation in which many of the best options and opportunities for the United States were squandered. Retrospectively, it might be noted that the prob- lems identified with de-Ba'athificaiton in Iraq do not clearly suggest a need for changes in the structures of the military organizations involved in the war or in the ways in which they interface with civilian leaders. Many military officers were working well with for- mer Ba'athists and saw clearly the disadvantages of a deep de-Ba'athification program. General Petraeus was doing especially well in the Mosul area in the im- mediate post-war time frame by implementing gen- erous peace terms, which he eventually had to revise and make more punitive in response to orders from higher authorities. ORHA Director Garner, General Abizaid, and the senior CIA representative in Gar- ner's staff were horrified by the implications of deep de-Ba'athification but were unable to argue against it effectively. The practical concerns of officers and in- telligence professionals on the ground seem to have been overruled by civilian leaders who felt that the invasion of Iraq offered a historic opportunity to re- make the Middle East in ways that benefited both the United States and the region. Once civilian authorities, including the President, had embraced this vision, the military did its best to achieve favorable results within the parameters set by civilian leaders. All leaders in- volved in the effort desperately wanted Iraq to be suc- cessful, but ultimately the de-Ba'athification policies proved counterproductive to achieving that goal. PARALLELS AND DIFFERENCES BETWEEN IRAQ AND THE ARAB REVOLUTIONARY NATIONS As noted throughout this work, some U.S. poli- ticians feared a Saddamist restoration in Iraq to the point that they supported a policy of deep de- Ba'athification, which became a major and ongoing problem in Iraqi nation-building. Such logic was based on the Ba'ath Party's conspiratorial history and its past ability to infiltrate government and military institutions and then use them to infiltrate the gov- ernment. The threat was especially serious to Bush administration leaders, since the Ba'athists in Iraq were ousted by a foreign military campaign and not a popular uprising. Consequently, the United States was placed in the position of attempting to manage what constitutional scholar Andrew Arato called, “an externally imposed revolution.” These U.S. policy- makers believed that they had to build a radical new 60 kind of political system in Iraq while remaining uncer- tain about the level of Iraqi public support for such an enterprise, which could appear to Iraqis as a foreign project designed to serve U.S. and perhaps even Israe- li interests. This situation inclined some civilian U.S. leaders at the Defense Department and CPA to push for the deep de-Ba'athification that was so damaging to Iraq's national unity, causing many Sunni Muslims to believe that they were facing an externally imposed revolution that was directed primarily against them. In the Arab Spring countries, few seem to believe that regimes based on the old values and elites are likely to reestablish themselves, although equally dismal re- gimes could still emerge in these societies and regime "remnants" can still create difficulties. The political edge that the Arab Spring countries have is that they do not appear to have the same level of fear about the old regimes reemerging as the Bush administration leaders did about Iraq. This difference may reflect the contrast between an imposed revolution and an indig- €In OUIS Oſle. Many Arabs involved in the 2011-12 revolutionary wave consider post-invasion Iraq to be an example of what can go wrong following the fall of a long-standing dictator. Some Arab revolutionary activists are corre- spondingly seeking to learn from the Iraqi experience despite the very different circumstances under which Saddam and the other Arab dictators were ousted. While many within the Arab World tended to blame the United States for all of the problems associated with de-Ba'athification, this monograph has shown that many of the worst excesses resulted from Iraqi government policies undertaken long after respon- sible U.S. leaders had become concerned about such activities and attempted to moderate them. The horror 61 of Iraq in 2005-06 reminds the world of the capacity of some human beings to slaughter people who lived quietly for decades as their neighbors once the re- straints of a crushing dictatorship have been removed. Revolutions by their nature are illegal, and questions of law and authority are often thrown open in eras of revolutionary transformation. The new governments established in 2011 clearly wish to avoid such a phase in their own revolutions, although many face equally daunting challenges, including the dangers of new kinds of dictatorship and civil war. Like Iraq, Some of these states, including Syria and Libya, have no democratic tradition. Others, such as Tunisia and Egypt, have had more open societies and limited power-sharing among diverse elites at earlier points in their history, although authoritarian tradi- tions have tended to be much more prevalent, espe- cially in recent years. Moreover, at least currently, there are no leading opposition figures with the dig- nity, moral authority, and heroic status of a Nelson Mandela or Vaclav Havel. Such individuals can use their standing as towering historical figures to lead an effort to build a tolerant and progressive society. Weaker politicians face more serious obstacles, no matter how well intentioned they might be. The two countries where the lessons of de- Ba'athification may have the most relevance are prob- ably Libya and Syria, should the regime in Syria be overthrown by the revolutionaries currently strug- gling against it. Both of these countries have long his- tories of especially intense repression and no tradition of democracy. This approach contrasts with many of the Arab monarchies that are organized along the lines of a patriarchal model. In these instances, the monarch often feels the need to show that he cares 62 about the population and is making strong efforts to ensure their welfare. While such regimes are inher- ently undemocratic, they are often much less brutal than the governments of nonmonarchical dictators. Social mobilization regimes, by contrast, are much more inclined to demand the society grant its full sup- port to the self-anointed guiding leader. Monarchs have obligations to their citizens that they usually freely acknowledge. The recently deposed dictators in Iraq and Libya needed only to implement their vision for society as they defined it, while maintaining solid control over the instruments of repression. Ba'athism in Iraq was what Saddam said it was, while Libya was Supposed to be guided by Colonel Muammar Qad- hafi's incoherent, “Third Universal Theory.” Syria is organized along similar lines as Iraq with its form of Ba'athism used to legitimize rather than guide actions undertaken by the regime. Citizens in such systems have obligations to the government, but they have no right to question the leadership or the leader's vision in any public way. The Qadhafi and Assad political Systems made only cosmetic concessions to more lib- eral or tolerant societies. * Nevertheless, it remains appropriate to begin the discussion of the lessons of de-Ba'athification and the Arab Spring with the two countries where the Arab Spring began, Tunisia and Egypt. These were the first and easiest of the 2011 Arab revolutions. Tunisia and Egypt. The first two Arab Spring dictators overthrown in 2011 were President Zine al-Abidine Ben Ali of Tu- nisia and then President Hosni Mubarak of Egypt. While parallels between Saddam Hussein's Iraq and 63 the dictator. Tunisia's ruling party, the Constitutional Democratic Rally (RCD) Party, was forced from power in February 2011, not long after Ben Ali fell. The RCD had served as Tunisia's ruling party since that country achieved independence in 1956, and was often viewed as a path of political advancement. Party leaders at- tempted to save the organization and their own power base after Ben Ali fled to Saudi Arabia by formally ex- pelling him and his closest associates from the RCD.” This act had no practical effect and was largely viewed as a desperate public relations stunt. Some senior poli- ticians also quit the party during this time frame in an effort to remain in office, although these machina- tions usually did not save their positions. The RCD was formally dissolved in March 2011 by the post-Ben Ali leadership, and its funds were impounded. Con- currently, a number of senior party officials were ar- rested but always on specific charges, usually related to corruption.” Other government and party leaders fled into exile to avoid trial. This orderly and respect- able approach to the old regime is not surprising. In January 2011, Dr. Moncef Marzouki, one of Tunisia's leading dissidents, who was soon to become the first post Ben-Ali president, commented on the RCD by stating, "We don't want any revenge, but we are fast with our principles that this horrible party does not return." A variety of other Tunisian revolutionaries echoed these statements, while explicitly maintaining that they did not want to repeat the mistakes of Iraqi de-Ba'athification.” Their focus remained arresting those at the top of the regime, although Saudi Ara- bia ignored repeated requests to extradite the former Tunisian dictator. Around 40 other senior leaders are currently being prosecuted for the killing and wound- ing of anti-Ben Ali protestors.” 66 On October 23, 2011, competitive elections were held in Tunisia for a 217-member National Assembly to draft a new constitution. The newly legalized En- nada Party won 41 percent of the total votes, trans- lating into 91 seats within the National Assembly.” This party is usually viewed as moderately Islamist, although some Tunisian secularists and others fear it could become more dogmatic if it is able to obtain and consolidate power.” Much of the support for Ennada appears to have been a result of the organization's un- flinching opposition to the Ben Ali dictatorship, which outlawed and persecuted it for a number of years. Even some secular voters cast their ballots for this or- ganization in the belief that an empowered Ennada is the most likely way to ensure that former Ben Ali sup- porters never regain power. Additionally, during the election campaign, Ennada strongly emphasized its sterling record of confronting the previous regime.” The challenge in Tunisian society now seems to be between secular, European-oriented citizens and those interested in a government more closely guided by Islamic religious principles. There are parallels with Iraq's first competitive elections. Islamists did quite well in those instances, although extremists such as Muqtada al-Sadr are not as important in the Tunisian system.” As in Iraq, it appears that the defeat of a cor- rupt but secular dictatorship helped to open a serious path for Islamist political victories in at least the short term. Such victories are not inherently alarming un- less the Islamists attempt to consolidate their power and remain in office through undemocratic means. Ennada's current success is a result of a “backlash election" in which its popularity directly mirrored the unpopularity of the RCD. As that party involves it- Self in the more mundane aspects of governance, such †: # | : m ū. 1% continue and could intensify over time. Even this ap- proach would nevertheless be significantly milder than de-Ba'athification, which fired large numbers of people from state bureaucracies as well as leading politicians. As in Tunisia, Islamists did exceptionally well in Egypt's first set of openly contested elections. The Muslim Brotherhood's newly established Freedom and Justice Party won the most seats of any political party, while the more hardline Islamist Nour (light) Party came in second in the elections for the lower house of Parliament. The Muslim Brotherhood’s im- pressive performance is not surprising. Even while illegal, this organization had been deeply involved in charitable activities, including, in particular, provid- ing support for the poor. The Mubarak government tolerated the Brotherhood's charitable role because it had no interest in diverting its own resources to address the problems of the poor. This background strongly parallels that of the Sadr Movement in Iraq both before and after the U.S.-led invasion. Moreover, although tolerated, the Muslim Brotherhood was ille- gal, and its leadership was sometimes subject to arrest and persecution. This Islamist electoral success may pave the way for these organizations to take power, although this is not fully certain. The SCAF is still playing a significant potential role, and presidential elections may still pro- duce a strong secular leader. Additionally, even if the Islamists are able to dominate Egyptian political insti- tutions, they will almost certainly have to behave in a way that does not isolate Cairo from important U.S. and European sources of foreign aid, international investment, and tourism income. Any new Egyptian government that presides over an economic down- 69 ward spiral, let alone a freefall, cannot be expected to last for very long. The Muslim Brotherhood seems to understand this situation and has already asserted that they will not renounce Egypt's peace treaty with Israel.” The leadership of the Muslim Brotherhood has also indicated that it is interested in working with secular parties. The Egyptian Revolution therefore seems to be less comprehensive in reordering society than the process that occurred in Iraq due to the U.S.-led invasion. While the NDP has been outlawed, the SCAF remains dominated by colleagues and decades-long friends of President Mubarak. Moreover, during the Mubarak regime, the military was repeatedly scrutinized to make certain that Islamists did not infiltrate its ranks. The result of this process is a senior leadership that is deeply wary of the Muslim Brotherhood and other Islamists, such as the Salafis associated with the Nour Party. Egyptian politics correspondingly seems to be increasingly dominated by the conflict between the military leadership and the emerging Islamists with a declining economy serving as the domestic context. Secular parties are also a player in this struggle, al- though their influence is limited by failures in com- petitive elections with the Islamists. As this struggle continues, the military seems focused on maintaining its role as a powerful autonomous actor in the Egyp- tian political system, including control over signifi- cant economic resources that are unrelated to military functions. Under such a scenario, the military role will remain similar to its situation under Mubarak. Moreover, state television and print media tend to portray SCAF leader Field Marshal Hussein Tantawi in the same fawning light as they applied to President Mubarak during his years in power, providing some- 70 thing of an echo of the earlier regime.” Conversely, in a key difference, the nonstate media can be extremely critical of the military's actions as was seen during the February 2012 soccer riots. Libya. There are a number of differences between the Iraqi and Libyan situation, but events in Iraq do have lim- ited parallels to Libya, because the legacy of long and brutal dictatorships in both countries. Fortunately, a number of Libyans appear to see Iraq as a cautionary tale for them.” At least some Libyans view the near civil war in Iraq as a result of a cycle of revenge and backlash between Iraq's Sunni and Shi'ite communi- ties, which carries important lessons for them. Al- though Libya is not factionalized along sectarian lines such as those in Iraq, there are important regional, tribal, and some ethnic differences. Responsible Liby- ans are consequently concerned about maintaining national unity and avoiding internal armed conflict. They are also aware of how quickly national unity can break down following the ouster of a brutal dictator. More ominously, building a unified and functioning Society in Libya will be significantly more challeng- ing than rebuilding Tunisia or Egypt, since both of the earlier revolutions were much less violent, and both of these countries retained a functioning political, eco- nomic, and institutional infrastructure that survived the dictatorships. In the aftermath of Qadhafi's defeat and death, it is not even fully clear that Libya will be able to maintain itself as one nation. Libya only be- came a unified nation under its first and only king in the 1950s, after having been liberated from Italian fascism. The major regions are Cyrenaica in the east, 71 -| Tripolitania in the west, and Fezzan in the southern part of the country. Political integration of these re- gions occurred to some extent under the monarchy, but not under the Qadhafi regime, which sought to take advantage of regional disagreements. Libya had no political parties under the Qadhafi regime, and there was no direct equivalent to the Ba'ath Party. There was, however, the Revolutionary Committees Movement set up by the regime to create vehicles for pro-regime activism. Like the Iraqi Ba'ath Party, these organizations are sometimes identified as important components of government, although real power has always resided with the dictator, his family, and his closest associates. Such organizations were nevertheless charged with a role in “defend- ing the revolution," and were often used as a tool of political and ideological surveillance as well as key enablers of corruption. Under these circumstances, the National Transitional Council (NTC) decision to abolish these organizations appears wise and gives up very little administrative talent, unlike the process of de-Ba'athification. While Qadhafi was probably just as ruthless as Saddam Hussein, his regime was nowhere near as well-organized for repression as Iraq.” Many individuals associated with these groups have a lot to answer for, including human rights abuse and torture. The questions that remain are; how to establish a clear dividing line between the worst offenders and the oth- ers, as well as how to ensure that justice is dispensed on an individual basis and does not expand into col- lective punishment. The Iraq example indicates the vital role of fair judicial institutions rather than highly political commissions for such tasks. As noted ear- lier in this work, Ambassador Bremer has openly ac- knowledged his mistake in not engaging the judiciary on these kinds of issues. 72 TST: Oſlº. 10km riſis mgº ingº nºi pºt m tº miſ thed: eikk emtſ; }Th: up : leſſ Oluffſ y ſº pºſt º ſº | Wiltſ: thſ ffſ ey Slſº isſ? % m Wi ſe: | (I& ºft ſº (ſ {{ aſ: the 2011 civil war and experienced a number of deser- tions.1% Most of the prisoners from the pro-Qadhafi units, such as the Khamis Brigade, now claim that they nev- er fired their weapons, they joined only for the mon- ey, and they were not involved in torture. Neverthe- less, pro-Qadhafi troops did commit a variety of war crimes throughout the conflict. Before the imposition of a North Atlantic Treaty Organization (NATO) No- Fly Zone over Libya, Qadhafi ordered the Libyan Air Force to bomb urban areas in the hands of the upris- ing. Most pilots involved in the conflict followed these orders, although a few defected to Malta or Italy. A willingness to refuse unlawful orders was therefore the exception and not the rule. Another particularly ghastly set of crimes against civilians involved the re- gime siege of Misurata, Libya's third largest city. This siege involved regime efforts to retake the city thor- ough a variety of means, including the use of artillery and multiple rocket launchers to fire indiscriminately into civilian areas. These attacks included the use of cluster munitions against the civil population. Orders for such tactics are almost certain to have come from the regime's leadership. Other war crimes, includ- ing the massacre of prisoners and the use of rape as a weapon of war, have also been charged, although it is not clear where the orders for these atrocities actu- ally originated. It therefore seems likely that many of Libya's citizens will seek justice in the aftermath of the conflict. It is, of course, known that some anti-regime forces committed war crimes as well, but these ap- pear to have occurred at the small unit level and were not part of any overall National Transitional Council (NTC) strategy. Senior NTC figures did not order war 75 crimes, and, because of their limited control over the militias, probably had no real ability to halt excesses at the tactical level. Such excesses were serious but appear more limited compared to the actions of the regime, where a furious Qadhafi seemed willing to do just about anything to crush the people he called "rats." Libyan authorities are unlikely to pursue ex- cesses by their own militia troops at this time due to an urgent concern to limit future fissures within the ranks of the revolutionary movement. At this time, it may be more important for them to begin to profes- sionalize the individuals who are expected to remain in the new Libyan Army so that they behave with restraint and military discipline in the face of future challenges and particularly domestic unrest that may continue to plague Libya. Some crimes committed on the rebel side may nevertheless be too serious to over- look indefinitely, and it may be less divisive to address them rather than ignore them. The unsolved murder of NTC General Abdul Fatah Younes on July 28, 2011, may be an example of such a crime. This assassination is widely believed to have been a result of internal dif- ferences in the anti-Qadhafi forces, possible because of Younes' background as a former Qadhafi interior minister.” In addition to former soldiers, there are questions of the future of pro-Qadhafi tribesmen in the new Lib- ya. Libya is a highly tribalized society with around 140 tribes and clans, of which 20 to 30 are considered large and important. Qadhafi's own tribe, the Qadadfa, is sufficiently small and weak, that the dictator had no choice except to reach out to individuals from other tribal units to help establish a base of power. Unsur- prisingly, members of pro-Qadhafi tribes (such as the Magarha) have dominated the Qadhafi regime's secu- 76 l: ºf ić rity forces, and it may be difficult to fully disentangle individual actions from those encouraged and abetted by tribal leaders and authority figures. This situation could invite the prospect of collective punishment from a vindictive successor regime. Some of the most important tribes, such as the Werfella, had combatants on both sides of the struggle to overthrow Qadhafi. This situation helps to mitigate the danger of collec- tive punishment to some extent. Beyond the dangers of unrestrained revenge, many of the new Libyan leaders are also concerned about the potential problems of building a new government with effective and legitimate institutions. Colonel Qa- dhafi had some unusual ideas about government, and his divisive and bizarre approaches to governance left little foundation on which to build. Qadhafi was open- ly contemptuous of political institutions and claimed to have developed a new form of government based on revolutionary democracy expressed at the grass- roots level. As seen with a variety of oil-rich countries, Colonel Qadhafi organized a strong public sector in Libya which he used to increase popular dependency on the government and to support claims of limited unemployment. While the Libyan government has a number of se- rious problems, there are also some positive factors for the country's future. One of the most important of these factors is the country's economic potential. Like Iraq, Libya is an important oil-producing state, but it also has only about a fifth of the Iraqi popula- tion. Moreover, it now appears that Qadhafi diverted huge sums of money to various secret bank accounts and investments, which are now available to the current Libyan leadership. One estimate suggests that there may be as much as $200 billion in such ac- 77 counts." While economic potential did not save Iraq from large-scale sectarian conflict, the possibility of a dramatically higher standard of living for virtually all Libyans would seem to be a positive factor in address- ing problems such as youth unemployment, which can increase the potential for unrest. The experience of both Iraqi de-Ba'athification and the disbanding of the Iraqi Army illustrated the severe dangers of spikes in youth unemployment in post-revolutionary sys- tems. The Iraq experience suggests that the sooner the government is able to use its considerable resources to create reconstruction jobs, the safer Libya will be from a downward spiral into instability. Syria. Unlike Egypt and Tunisia, the ouster of the Syr- ian regime may take years of struggle, if it occurs at all. At the time of this writing, more than 5,500 Syrian civilians had been killed by the military and security forces as part of this conflict, and the rate of civilian deaths may yet increase rather than decline. Syria is ruled by a Ba'athist dictatorship, which has a number of similarities with Saddam Hussein's Iraq. Current President Bashar al-Assad rose to become the leader of Syria in 2000 following the death of his father. Syria is, for all practical purposes, a one-party state with its own branch of the Ba'ath Party providing the ideo- logical façade for a system of government designed to protect the privileged position of those currently in power." Syria, also like Iraq, is a patchwork of dif- ferent sectarian groups, and Syria's ongoing revolu- tionary struggle has profoundly sectarian overtones. Since 1970, Syria has been led by presidents from the Assad family, members of the Alawite sect, which is 78 | generally considered to be a branch of Shi'ite Islam. The Alawites of Syria, at 10 percent of the population, are a much smaller portion of the Syria's total popu- lation than the Arab Sunnis of Iraq who comprise 20 percent of the population of their country. Most of the most important members of the government, military, and Ba'ath Party are from the Alawite sect or other minority communities allied with the Alawites. Sunni Muslims constitute around 70-75 percent of the popu- lation, whereas Christian and other minorities com- prise the remaining portion. Throughout much of Syrian history, Alawites were an ill-treated, impoverished religious minority who were seldom able to obtain anything more than the worst employment that the society had to offer. Their fortunes improved somewhat when opportu- nities to serve in the French colonial army became available after France occupied Syria as a League of Nation's mandate following World War I. The legacy of being near the bottom of Syria's socioeconomic ladder weighs heavily on many current members of the Alawite community as they contemplate their fu- ture under either a Sunni-dominated democracy or a Sunni-led dictatorship. If post-Assad Syria was to be dominated by Islamists, their views of the Alawites could be even further poisoned by claims their sect is heretical. Such beliefs are widespread among Sunni Islamist hardliners. Additionally, many Alawites un- doubtedly fear Sunni revenge for 40 years of brutal- ity and misrule under the Assads. The bloody con- flict that has raged over the last year has also added horrendous new grievances to the already long list of injustices for which the Alawites may be held ac- countable. Saudi writer Essam al-Zamel has acidly Summed up this situation with almost literal gallows 79 “What did I ever do to you" in his last hour of life, ap- parently stunned at the depth of hatred against him. Bashar is under no such illusions. He is aware of his crimes and aware that only a strong dictatorship can prevent him from sharing Qadhafi's fate. The Syrian regime will probably never be over- thrown by nonviolent protesters, whom they are willing to kill in whatever numbers they need to in order to remain in power. Since these tactics are in- creasingly viewed as ineffective, the opposition seems more willing to consider armed opposition. If Assadis to be overthrown, this will have to be done by armed struggle, and the effort to oust the dictator may well continue to evolve into an increasingly bitter civil war. The other side of this approach is that the Syrian re- gime will use armed opposition to justify increasing its own level of repression, which is already at ex- tremely high levels. All of these factors indicate that the Syrian government will use all of the repression of which it is capable rather than liberalize the regime in a way that fundamentally threatens the power of the current elites. Many reports cited by Western news agencies have accused Syrian military units of indiscriminate shelling of civilian areas thought to be opposition strongholds.” The use of such tactics was later confirmed by Arab League monitors, whom the Syrian regime accepted into their country to buy time before its case was sent to the UN Security Council. These monitors have now departed Syria. The Syrian military has held together extremely well throughout much of the uprising, although there are increasing reports of defections.” Some of these defectors have fought against pro-regime units, but they have neither the organizational infrastructure nor the logistics and supply network necessary to maintain themselves as a conventional military force. 82 ić |Stil eft hig tº ſº. º di 㺠|Sºſ sº Various spokesmen for the defectors call these forces the Free Syrian Army (FSA). While some units of the FSA have been allowed to seek sanctuary on Turkish soil, there is no indication that Ankara is involved in any effort to arm, equip, or train these units.” More- over, at the time of this writing, FSA forces in Turkey did not seem to comprise more than a few hundred individuals. Nevertheless, as the carnage continues, it is likely that Sunni conscripts will feel increasing psy- chological pressure to defect. The conscripts make up the majority of forces in the non-elite military units. In light of the above considerations, any program of de-Ba'athification in Syria could be especially brutal because of the ever-growing list of crimes the regime has committed against the Syrian people to remain in power. It has been noted that some Iraqis spoke of de- Ba'athification as de-Sunnization, and the prospect for all Syrian Alawites to be viewed as regime supporters suggests that de-Ba'athification in Syria would take on a sectarian character as well. While most Alawi- tes sympathize with the Assads, not all Alawites have committed crimes to support them. The prospect of a post-revolutionary Alawite bloodbath is therefore Something that all responsible friends of a new Syr- ian government would have to guard against. It is not, however, clear that neighboring Arab governments will push hard in that direction, since all of them, ex- cept Iraq, have Sunni leadership. Many Lebanese fear that sectarian hatred in Syria could spill over into their own country, but Lebanon has very little ability to moderate actions within Syria. Perhaps the best hope is that a post-revolutionary government will need a great deal of aid for reconstruction and development, and any anti-Alawite bloodbath would inevitably cause the United States and Europe to suspend such aid and call upon Arab states to do so as well. 83 Yemen. At this time, only one additional Arab leader has been ousted as part of the Arab Spring upheavals, for- mer President Ali Abdullah Saleh of Yemen. There are probably relatively few lessons that Yemen can draw from Iraq's experience with de-Ba'athification. Ye- men has never maintained a strong centralized gov- ernment such as could be found in Iraq under Sad- dam. President Saleh ruled by political opportunism, manipulation, and bribery as much as by repression. One of Kuwait's post-1991 foreign ministers once characterized him as a dictator like Saddam Hussein, but this is not true. Saleh never had the apparatus of repression that Saddam maintained, and many of his leading officers appear almost as opportunistic as the former president. In situations such as this, Yemeni officers have often had the political space to consider their own interests in deciding if they will continue to support Saleh. Yemen is a multiparty system, and while former President Saleh's political party, the General Peoples Congress, may not be abolished, it will almost certain- ly be weakened. The powerful Islamist party, Islah, is its most likely successor. Such an ascendancy would hardly be good news for the United States, which had designated the Islah party leader as a terrorist supporter. Nevertheless, no new leader in Yemen is likely to establish a strong central government. The powerful Yemeni tribes and the military will probably continue to dominate the political system. General Ali Moshen, a former Saleh lieutenant who defected to the rebels along with the forces under his command, may emerge as an especially important post-Saleh leader. 84 lines, the following recommendations are offered for addressing concerns about the future of Iraq and the Arab Spring with the lessons of de-Ba'athification as a consideration. Some of these recommendations may also be relevant for non-Arab countries, includ- ing Iran, if their populations rise against oppressive regimes in Arab Spring-type revolts. Regional experts and policymakers might wish to consider such paral- lels on a case-by-case basis. 1. The United States must continue to do what- ever it can to support a process of national concilia- tion in Iraq. In order to do this, U.S. leaders must re- main attuned to problems in Iraq and elsewhere that may push the country toward state failure. The Bush administration often seemed to view democracy as a wholly positive transformative system without fully understanding the ways in which it can go wrong in deeply polarized and sectarian societies. As noted, these mistakes were eventually understood by many U.S. leaders over time, but, by then, the dynamics of sectarian conflict were especially difficult to contain and roll back. The United States must nevertheless continue to express its disapproval for politicized and reckless de-Ba'athification procedures in Iraq. Wash- ington must also stress that political institutions used as instruments of sectarian oppression always hold the potential to incite civil war in societies that have not been totally crushed by their governments, one of the worst outcomes imaginable. 2. Officers and senior non-commissioned officers (NCOs) of the U.S. Army must realize that they may often have unique opportunities and unique cred- ibility to offer advice on the lessons of Iraq, includ- ing the problem of de-Ba'athification, to their coun- terparts in some of the Arab Spring nations. The U.S. Army has a long history of cooperating with some 86 : §: h; of the Arab Spring militaries and has a particularly strong relationship with the Egyptian military. These bonds of trust and teamwork can be used to convey a variety of messages beyond exclusively military is- sues. Additionally, many U.S. Army officers gained a great deal of experience in Iraq from which they can draw upon to highlight the ways in which revolution- ary change can go wrong. 3. The United States will have to use the les- sons learned about Iraqi de-Ba'athification to assess ways in which Arab Spring countries may be fail- ing. U.S. leaders must also consider ways to advise or otherwise help struggling governments. One of Iraq’s key problems was the rapid development of the view among some Iraqis that democracy is a winner- take-all form of government. The activities of the de- Ba'athification Commission often helped to reinforce that view among both Shi'ites and Sunnis. All Arab Spring governments will need to take care that new post-authoritarian institutions do not reflect a similar winner-take-all mentality among leaders, sects, tribes, political groups, or regions. The United States can draw upon its experiences in 2003 Iraq to make this message known to new governments, but unlike Iraq, the United States is not in a situation where it must take ownership of the Arab Spring. This means that U.S. leaders will have to have the dexterity to recog- nize when their advice is helpful and welcome and when it may be a source of resentment and conflict. 4. The United States needs to help strengthen and support judiciary institutions in the Arab Spring countries and strongly emphasize the need to re- solve issues of guilt or innocence within a judiciary framework rather than a commission composed of politicians or any similar organization. The United 87 States should laud any efforts in these countries to ad- vance the concepts of rule of law and support the ac- tivities of UN specialized agencies and other nongov- ernmental organizations (NGOs) seeking to provide aid to judicial bodies in the post-revolutionary states. It has already been noted that Iraq's de-Ba'athification Commission was administered by politicians, and that these politicians were running for office against people who they disqualified for participation in the election. In no part of the planet can this be considered fair. Seeking to honestly establish the guilt of those ac- cused of political crimes as well as their punishment is best left to judges if a fair court system can be estab- lished and maintained. Eligibility for particular indi- viduals to run for office should likewise be addressed by unbiased judicial institutions. 5. The United States needs to be careful to avoid the appearance of publicly picking out and then lob- bying for favorite leaders in the new Arab Spring governments, as it did with Ahmad Chalabi in Iraq. If these countries are lucky, they may find a dignified and fair-minded national leader, although it would be a mistake for outside forces to attempt to generate one. The U.S. experience with Chalabi in Iraq might be particularly instructive here. U.S. leaders should not underestimate the distrust that many Arab popu- lations feel for the United States. U.S. support for a leader with strong connections in Washington is not always well received by a populist electorate. No na- tions are so close that their publics want another state to choose their leadership for them. Additionally, problems have already resulted when Qatar chose to support some Libyan politicians and factions over others in their aid programs. Qatar was the strongest Arab backer of the Libyan revolution, and Doha has 88 adopt similar policies to those in Libya. The words of journalist Michael Hirsh, “What happens in Libya, stays in Libya," correctly suggest that a unique set of factors enabled international action against the Qad- hafi regime, and that this set of factors is unlikely to be duplicated elsewhere in the Arab World.” Syria, in contrast, would present a hornets' nest of problems comparable in intensity to those the United States en- countered in Iraq. Even without foreign intervention, it seems unlikely that the Assad regime will be able to crush the opposition by force, which remains its goal at this time. It also seems unlikely that the regime will offer more than simply cosmetic efforts at power- sharing. If the regime is overthrown, the United States must seek to work with post-Assad revolutionaries, but ultimately may be viewed with distrust in the Arab World for any effort to involve itself in combat. Like the Iraqis, the Syrians have enduring decades of propaganda about U.S. conspiracies to control the Middle East. 10. The U.S. leadership must also understand that, to an even greater extent than the Iraqis, Syrian citizens may be susceptible to anti-Western conspir- acy theories and distrustful of U.S. intensions. The close U.S. friendship with Israel is especially likely to create Syrian distrust about any U.S. agenda for their country. Despite some occasional Iraqi participation in the conflict with Israel, including the 1973 Arab- Israeli War and the 1991 Scud missile strikes, Iraq is usually a peripheral player in the Arab-Israeli conflict, while Syria is at the core of the conflict. The Iraqis viewed the U.S. presence in their country with steadi- ly increasing criticism, with Washington receiving very little credit as a liberator, due to a local belief that the United States invaded their country to advance its —ſ own agenda for obtaining cheap oil and establishing military bases that could be used to dominate the re- gion. In Syria, suspicions of an alternative U.S. agenda favorable to Israel could only be stronger. This situa- tion is not an insurmountable problem, but it may also indicate the need for the United States to allow other friendly nations to play a major role in helping to re- solve problems in Syria should Assad be overthrown. 11. The United States must maintain an ongo- ing dialogue with its closest regional allies, in- cluding Israel, Jordan, and Saudi Arabia, about the changes that are occurring in the Arab World, and Washington would do well to consider their views in the formulation of policy. This type of behavior was not always apparent in Iraq on the issue of de- Ba'athification when some allies such as Jordan and Saudi Arabia considered the United States to be dis- missive of their views in favor of its own poorly in- formed assumptions. These assumptions included the belief described by Professor Fukiyama that the de- fault position for a defeated dictatorship is always a liberal democracy. The Jordanians and the Saudis had been living in that region long enough to know better. The Jordanians had even seen a branch of their own ruling family ousted by a military coup in Iraq and then watched a succession of new and more brutal dictatorships emerge in that country culminating in the rule of Saddam Hussein. While the United States may ultimately disagree with these countries, their views have earned serious consideration. 12. The United States must understand that the rise of Islamists in the Arab Spring countries is not an aberration and is likely to continue to occur in additional cases, although it is uncertain that IS- lamists will dominate any of these countries in the 92 governments will not be easy for any of the Arab Spring governments. Libya will be harder than Egypt or Tunisia, and all countries need to understand the importance of avoiding a downward spiral. 14. The U.S. Army will have to be aware that it may be asked to play a limited role in rebuilding the Libyan military. The United States will also have to deal with the Libyan government on a variety of is- sues, such as shoulder-fired anti-aircraft weapons ac- countability, and has a stake in the new government's success. This situation suggests that various forms of military expertise and training may be provided to a responsible Libyan government at some future point. Such efforts will undoubtedly complement efforts by other countries to help Libya, including its major trad- ing partners in Europe and the Arab League. It may also be possible, although not likely, that Libya may have the building blocks of an insurgency. Pro-regime holdouts conducted impressive defense of Sirte and Bani Walid for some time after the rest of the country was liberated. 15. The United States might wish to offer ad- vanced military training and education to Libyan officers that includes Western concepts of civil-mil- itary relations. Such training is already provided to officers of other Arab Spring militaries. Additionally, army officers who defected from the Libyan army and fought for the rebels have a clear expectation that they will have a role in the future of Libyan national Secu- rity. These concerns need to be channeled in ways that allow democratic processes and institution building to move forward. If the Syrian government is over- thrown, the United States may wish to consider ways to reach out to help professionalize the military and keep it out of politics. U.S.-Syrian relations may nev- 94 º | th: til t; d ertheless be complicated by issues involving Israel and the Palestinians in ways that will not occur with Libya. ENDNOTES 1. Francis Fukuyama, America at the Crossroads, Democracy, Power, and the Neoconservative Legacy, New Haven, CT, and Lon- don, UK: Yale University Press, 2006, p. 116. 2. King Abdullah II of Jordan, Our Last Best Chance: The Pur- suit of Peace in a Time of Peril, New York: Viking, 2011, p. 226. 3. As quoted in Tom Heneghan, “Libyans Stress Harmony to Avoid Iraq-Style Chaos,” Reuters, September 4, 2011. 4. On the use of this phrase, see Douglas J. Feith, War and Deci- sion, New York: HarperCollins Publishers, 2008, p. 202. 5. Saddam Hussein's regime later claimed without evidence that Aflaq had made a deathbed conversion to Islam in 1989, at a time when his propaganda machine was attempting to respond to Iranian charges that his regime was "Godless." See Amatzia Baram, Culture, History and Ideology in the Formation of Ba'athist Iraq, 1968–89, New York: St. Martin's Press, 1991. 6. The Syrian and Iraqi branches of the Ba'ath displayed a long-standing inability to work with each other and a continuing struggle for leadership of the Ba'athist movement. Brief periods of partial reconciliation were usually followed by long periods of hostility and conflict. 7. Baram, p. 13. 8. Hanna Batatu, The Old Social Classes and the Revolutionary Movements of Iraq, Princeton, NJ: Princeton University Press, 1978, p. 1010. 9. Efraim Karsh and Inari Rautsi, Saddam Hussein: A Political Biography, New York: The Free Press, 1991, p. 15. 95 | ſ 21. See, for example, Feith, War and Decision, p. 240; and Don Eberly, Liberate and Leave: Fatal Flaws in the Early Strategy for Post- war Iraq, Minneapolis, MN: Zenith Press, 2009, p. 27. 22. Ahmed S. Hashim, Insurgency and Counter-insurgency in Iraq, Ithaca, NY: Cornell University Press, 2006, p. 15. 23. Andrew J. Bacevich, The Limits of Power: The End of Ameri- can Exceptionalism, New York: Henry Holt and Company, 2008, p. 165. 24. Hashim, chap. 5. 25. Feith, p. 202. 26. On the concept of "de-Arabization," see Hashim, p. 385. 27. Bob Woodward, The War Within, New York: Simon and Schuster, 2008, pp. 120-121. 28. Transcript of Interview with Lieutenant General Jay Gar- ner, “The Lost Year in Iraq." PBS.org, posted October 17, 2006. 29. Ibid.; Rudd, p. 313. 30. Colonel John Agoglia, as quoted in Charles H. Ferguson, No End in Sight, New York: Public Affairs Books, 2008, p. 155. 31. Public Broadcasting Corporation Transcript, “Frontline: The Lost Year in Iraq.” October 17, 2006. 32. Vice President Cheney was an exceptionally strong sup- porter of deep de-Ba'athification. See Graham, p. 462. 33. Bremer interview as cited in Public Broadcasting Corpora- tion Transcript, Frontline: The Lost Year in Iraq, October 17, 2006. 34. Bob Woodward, State of Denial: Bush at War Part III, New York: Simon & Schuster, 2006, pp. 196-197. The author has also heard a similar account from an officer who worked with General Abizaid during this time frame. This officer also maintained that no one who knew General Abizaid at this time could fail to un- derstand his opposition to harsh and sweeping de-Ba'athification. 97 35. Draper interview with President Bush as cited in Robert Draper, Dead Certain: The Presidency of George W. Bush, New York: Simon & Schuster, Inc. 2007, p. 211. 36. L. Paul Bremer III, “How I Didn't Dismantle Iraq's Army," New York Times, September 6, 2007. 37. Bremer, p. 21. 38. Chap. 2 of Bremer's book is entitled “Taking Charge." See Bremer, p. 23. 39. Rudd, pp. 307-308. 40. Packer, p. 240; Rudd, pp. 310-311. 41. Rudd, p. 327. 42. Ibid, p. 132. 43. Allawi, p. 149. 44. Ibid., p. 150. 45. For an exploration of these issues, see Christian Caryl, "Iraq's Young Blood,” Newsweek, January 22, 2007, pp. 25-28. 46. Packer, p. 192. 47. On nominal Ba'athists, see Rudd, p. 270. 48. Lieutenant General Ricardo S. Sanchez, Wiser in Battle, A Soldier's Story, New York: HarperCollins Publishers, 2008, p. 186. 49. According to a 2008 article co-authored by Bremer, the CPA sought to reverse deep de-Ba'athification. But by the time it did so, “the CPA's authority was waning, and the dominant (Shi'ite] and Kurdish political factions in Baghdad had little or no interest in allowing even innocent Ba'athists back into public office." See L. Paul Bremer, James Dobbins, and David Gompert, "Early Days in Iraq: Decisions of the CPA,” Survival, August/Sep- tember 2008, p. 31. 98 139. "Tunisia coalition hits trouble on day two," Jordan Times, January 19, 2011. 140. Ibid. p. 3. 141. “Tunisia ministers quit government as protests resume," BBC News, January 18, 2011. 142. Kemel Riahi, "A Night in Tunisia," New York Times, Janu- ary 18, 2011. 143. “Tunisia says Saudi silent on requests to extradite ex- leader," AFP, January 4, 2012. 144. For a complete breakdown of the election results, see Erika Atzori, "Tunisia leads the way,” The Middle East, December 2011, p. 19. 145. The party's name is French for “Renaissance," so, ironi- cally, an Arabic translation would render it as the Ba'ath Party. Nevertheless, the name is used in a completely different context, and Ennahda has nothing to do with Saddam's Ba'ath Party. 146. "Islamists to the fore," Economist, October 29, 2011. 147. Bouazza Ben Bouazza, "Tunisia's Islamist party slams anti-Semitic chants," Associated Press, January 9, 2012. 148. º Mubarak party members allowed to run in Egypt vote," Jordan Times, November 15, 2011; Leila Fadel, "Egyptians Fear Re- turn of Mubarak Allies,” Washington Post, November 17, 2011. 149. Abdelrahman Youssef, “Mubarak-era figures recycle ca- reers in rural Egypt.” Reuters, January 9, 2012. 150 "Brotherhood's laxity on Israel treaty stuns Hoss," The Daily Star, January 14, 2012. * Jeffrey Fleishman and Amro Hassan, "Egypt Military Giving Signs of Not Wanting to Relinquish Power," Los Angeles Times, November 3, 2011. 106 164. Liz Sly, “Report Hint at Cracks in Syria Regime," Wash- ington Post, April 28, 2011; Elizabeth A. Kennedy and Diaa Hadid, "Syrian Army Units Turn on Each other Amid Crackdown," BOS- ton Globe, April 29, 2011. 165. Kathierine Zoepf and Anthony Shadid, “Syrian Leader's Brother Seen as Enforcer of Crackdown," New York Times, June 7, 2011. 166. “Defiant Assad Praises Troops after Deadly Syria Crack- down," Jordan Times, August 2, 2011. 167. Khaled Yacoub Oweis, “Anti-Assad fighters defy odds in Syrian town," Jordan Times, September 29, 2011. 168. Liam Stack, “In Slap at Syria, Turkey Shelters Anti-Assad Fighters," New York Times, October 27, 2011. 169. Sam Dagher, “Tiny Kingdom's Huge Role in Libya Draws Concern," Wall Street Journal, October 17, 2011. 170. Michael Hirsh, "What Happens in Libya Stays in Libya," National Journal, October 20, 2011. 108