19 - IRAQ’S YEAR OF VOTING DANGEROUSLY (Ka)

IRAQ’S YEAR OF VOTING

DANGEROUSLY

Adeed Dawisha and Larry Diamond

Adeed Dawisha is professor of political science at Miami University in Ohio and author of Arab Nationalism in the Twentieth Century: From Triumph to Despair (2003). He has written extensively on competitive politics in Iraq during the first half of the twentieth century and in the recent postwar period. Larry Diamond is senior fellow at the Hoover Institution, coeditor of the Journal of Democracy, and author of Squan-dered Victory: The American Occupation and the Bungled Effort to Bring Democracy to Iraq (2005). This essay originally appeared in the April 2006 issue of the Journal of Democracy.

The year 2005 may prove to have been one of the most politically consequential in the modern history of Iraq. In the space of less than eleven months, the country held three elections. Two of these, at the beginning and the end of the year, elected parliaments under a transitional and then a permanent constitution. Sandwiched between these, in October, was a referendum to approve the draft constitution. The elections took place successfully, with Iraqis voting in large numbers despite widespread logistical challenges, terrorist intimidation, and insurgent violence. But in their entrenchment of ethnic and sectarian fissures as the main organizing principle of politics, the three votes highlighted the role and limits of electoral-system design in the quest to manage and contain potentially polarizing divisions.

As the postwar reconstruction of Iraq began to move past the chal-lenge of immediate stabilization into a phase of political reconstruction, one of the key challenges that had to be faced was the selection of a system to elect first the Transitional National Assembly, and then ulti-mately a permanent National Assembly.1 Some months after its inception in May 2003, the Coalition Provisional Authority (CPA) commissioned experts from IFES (established as the International Foundation for Elec-tion Systems) and elsewhere to advise on how elections might be administered in Iraq. Meanwhile, Iraqi politicians and their advisors

Adeed Dawisha and Larry Diamond

225

were also beginning to consider the question. In theory, all options were on the table: one or another form of proportional representation (PR); the single-member first-past-the-post system (FPTP); some kind of mixed system combining these two methods; the alternative vote (AV); the single transferable vote (STV); and the single nontransferable vote (SNTV).

Most of these options were eliminated early on because they were either impractical or unsuited to Iraq’s political circumstances. FPTP was appealing in that it promised to limit the power of parties with autocratic tendencies and to encourage the election of independent and locally rooted representatives, diminishing the prospect of the election morphing into a polarizing referendum based on religious and ethnic identity. As a distinctly majoritarian system, however, FPTP in its pure form was clearly inappropriate for as deeply divided a country as Iraq, where it was important to ensure some degree of fairness in the represen-tation of all major groups as well as smaller minorities.2 AV was in theory more appealing, for the reason that Donald Horowitz describes: By requiring the voters in each single-member district to rank the can-didates on the ballot, it forces candidates to compete for the lower-order preferences of voters and thus induces moderation.3 But it quickly be-came apparent that neither of these systems would be practical, for the simple reason that a new Iraqi electoral administration—no matter how well assisted and advised it might be by the international community— was not going to have the time, the resources, the reliable census data, or the political credibility to draw the boundaries of some 275 indi-vidual districts (the size of parliament that the interim constitution established in March 2004).

One also had to consider the Iraqi voter. Although there were plans for an extensive program of civic education, it was still a priority to keep the ballot and the voting process as simple as possible. Requiring people who had not voted in a democratic election in fifty years to rank several choices on a ballot seemed to be asking too much. For the same reason, STV—which requires the voter to consider and rank even more candidates in multimember districts—was considered even more im-practical. All of these systems thus fell by the wayside fairly early in the planning for elections in Iraq. So did any kind of mixed system that would rely in part on single-member districts. So did SNTV—in which the voter selects a single candidate in a multimember district, with no possibility for excess votes to be transferred to other candidates of the same party. Although SNTV is for the voter a very simple system, all the experts knew of its perverse practical implications, which require a con-siderable degree of planning and coordination among parties and between parties and their supporters to ensure reasonable degrees of proportionality and rationality in outcomes.

Both substantive and practical considerations therefore strongly in-

226 Iraq’s Year of Voting Dangerously

clined the outside experts as well as the Iraqi political leaders and their advisors toward some form of PR. In fact, PR was a compelling choice for Iraq for several reasons. First, it fit with the power-sharing or “conso-ciational” logic of institutional design that Iraq was moving to embrace. Proportionality had become a basic principle of Iraqi political life with the July 2003 appointment of the Iraqi Governing Council—whose 25 members represented a delicate balance among Shi’ites, Sunnis, and Kurds in numbers approximating their shares of the population (it also included one member each from the Turkoman and Assyrian Christian minorities).4 PR sustained the logic that each group should expect to have a share of power roughly proportional to its weight in society. This expectation—which quickly became an entitlement—was deeply wor-risome to some Iraqis and foreign advisors, but once the logic was established it became inescapable.

Second, the leaders of the principal Shi’ite and Kurdish political par-ties were attracted to party-list PR because it promised to reinforce their weight in the political system and give them tighter control over who would run on their party label. PR also made it easier for the various Kurdish and Shi’ite parties to coalesce into common lists. Finally, advo-cates of a guaranteed quota for women’s representation in parliament—such as the Iraqi Higher Women’s Council—were drawn to PR because of its greater technical suitability to ensuring women’s representation. Interna-tional experts estimated, for example, that if parties were required to place women at no worse than every third interval on their ranked lists, it was quite likely that a 25 percent minimum quota of female representatives in parliament—the target written into the interim constitution and retained, at least for a transitional period, in the permanent one—could be achieved, even if PR were conducted in a series of multimember districts. Devising a mechanism to ensure a minimum percentage of women in parliament would be much more difficult and cumbersome in a system of single-member districts, or even in a mixed system.

What Kind of PR?

Before too long, then, discussions over electoral-system design turned to the question of what kind of PR would be most suitable for Iraq. In addition to selecting a formula for allocating seats to parties, there were two other issues: whether there should be a minimum threshold for entry into parliament, and the choice between a district-based or national-list system (or some combination of the two). Some CPA officials, includ-ing Administrator L. Paul Bremer, wanted to discourage the extreme fragmentation of parliament into an unwieldy body of many small par-ties while also ensuring representation for ethnic minorities and liberals whose support might be thinly spread across parts of the country. This suggested a modest electoral threshold of no more than 2 or 3 percent of

Adeed Dawisha and Larry Diamond

227

the vote. But as the alarming implications of arbitrarily excluding small religious minorities and secular liberal forces became clear, both the logic of inclusion and the desire to have in parliament as many moder-ate and independent voices as possible pressed in the direction of no threshold at all.5

In early 2004, several Iraqi and international officials began to de-velop proposals for a two-tiered PR system, based mainly in electoral districts, but with enough seats filled from national lists to ensure over-all proportionality between votes and seats. At roughly the same time, in March 2004, an Iraqi advisor to the Governing Council and a U.S. advi-sor to the CPA independently developed virtually identical proposals for a system in which 220 of the 275 seats would be apportioned among Iraq’s 18 provinces—each serving as an electoral district—based on their shares of the population.6 These district seats would be filled first, using the PR rule within each province, and then the remaining 55 seats (20 percent of the total) would be allocated to national party lists to ensure that each party’s or coalition’s overall seat share would match as closely as possible its national vote share. (IFES had estimated that if 20 percent of the seats were held back to be filled from national lists, the system could ensure a high degree of overall proportionality.)

There were compelling reasons for basing PR primarily in moder-ately sized districts coinciding with Iraq’s provincial boundaries. First, the provinces were already established. Second, even if the population balance among provinces was in dispute, the second-tier distribution of seats to national lists would ensure overall proportionality. Third, a district-based system would allow some ties to emerge between elected representatives and geographical constituencies. And fourth, it would be easier for voters to judge the individual candidates on each list, making for greater local accountability.

In the end, the decision on an electoral system for Iraq was made prin-cipally by Carina Perelli, the chief of the United Nations Electoral Assistance Division (and the head of its mission to Iraq), then adopted by the Iraqi Governing Council and signed into law by Bremer on 15 June 2004 as one of his final acts as CPA administrator. Somewhat unexpect-edly, Perelli did not opt for two-tiered PR, but rather for PR in a single nationwide district. She did so for several reasons. It was by far the easiest to administer. It avoided having to determine how to apportion seats to districts, which was a thorny issue given the absence of broadly accepted census data. It neutralized concerns over where people would vote—a problem for natives of one province who might be living in another (and a particular problem for Kirkuk, where the Kurdish parties were energeti-cally trying to resettle Kurds to reverse a previous campaign of ethnic cleansing under the Ba’ath Party). And, as Perelli explained in a June 4 press conference, the single nationwide district would make it easier “for [ethnic and religious] communities that have been broken up and dis-

228 Iraq’s Year of Voting Dangerously

persed around Iraq . . . to be able to accumulate their votes and to vote with like-minded people.”7 Yet a district system with a national-list com-ponent would also have enabled dispersed communities to aggregate their scattered votes to elect one or more members of parliament even if they lacked the electoral weight to do so from any one province. Without question, however, the nationwide system was easiest.

In the end, there were two chief consequences of Perelli’s choice of electoral system for Iraq. First, in the absence of districts, the January 2005 elections became almost purely a national-identity referendum, untempered by any local component or flavor. And second, the lack of any minimum floor of representation for geographical areas, along with a host of other grievances, led most Sunni political parties to boycott the elections.8 Each of these consequences would have major implica-tions for Iraqi politics.

The First Two Votes

The January 30 elections were to choose a Transitional National Assembly, whose charge was to write a permanent Iraqi constitution, hold a referendum to approve it, and prepare the country for general elections to be held no later than December of the same year. In the midst of endemic violence, January saw a heated election campaign featuring thousands of candidates belonging to more than a hundred parties and coalitions. Noticeable by their absence were the Iraqi Sunnis. A demographic minority constituting some 20 percent of Iraq’s popula-tion, they had been the country’s political and power elite for centuries, a position of dominance that had been cemented under the blatantly sectarian rule of Saddam Hussein. Fearful of losing political power, they tried to thwart and delegitimize the elections, publicly boycotting and belittling them as a U.S. imperialist endeavor, while giving a silent nod to the threats and intimidation employed by the Sunni insurgents.

Ranged against the Sunni rejectionists were the Shi’ites and Kurds, who constitute roughly 60 percent and 20 percent of the country’s popu-lation, respectively. Both communities had suffered considerably under Saddam’s rule and were now determined to seize the moment that would elevate them to political ascendancy. Senior Shi’ite clerics, especially the most venerated Grand Ayatollah Sayyid Ali al-Sistani, publicly urged the Shi’ite majority to go to the polls. The Kurds, on the other hand, needed no encouragement to vote in heavy numbers; they had a major stake in designing a constitution that would safeguard their cultural and political autonomy. In the end, even with an almost total Sunni boycott, a healthy 58 percent of registered voters nationwide went to the polls.

From a vast field of party lists, three blocs emerged to dominate the 275-seat Transitional National Assembly. The largest group, garnering 140 seats, was the United Iraqi Alliance (UIA), a collection of mostly

Adeed Dawisha and Larry Diamond

229

Shi’ite parties and individuals led by the Supreme Council for Islamic Revolution in Iraq (SCIRI) under the leadership of the cleric Abd al-Aziz al-Hakim and the al-Da’wah party led by Ibrahim al-Jaafari. The Kurdistan Alliance, a coalition of the two major Kurdish parties under the leadership of Masoud Barzani and Jalal Talabani, won 75 seats; and the Iraqi List, a secular group of politicians under the leadership of interim prime minister Iyad Allawi, won 40 seats. Nine small parties accounted for the remaining 20 seats. Given the breakdown of the vote, it was abundantly clear that Iraqis overwhelmingly voted in accord with their ethnic and sectarian identities. Almost all Kurds (probably more than 95 percent) voted for the Kurdistan Alliance, some 75 percent of Shi’ites voted for the UIA, and at least 75 percent of Sunnis opted for their sectarian choice—boycott. Consequently, only 17 of those elected were Sunnis, and most belonged to secular and nationalist lists.

The election outcome was a harsh wakeup call for the Sunnis. Having bet on its failure, they instead emerged as the real losers. Controlling barely 5 percent of the seats in parliament, the Sunnis were left out of the political bargaining through which the new permanent constitution was going to be designed. Very soon after the April 2005 formation of the transitional government, the Sunnis changed their tune and demanded to join the constitution-drafting committee. Realizing that Sunni political inclusion was essential to diminishing the insurgency, the Americans pressed hard for a deal to include the Sunnis, and an agreement was fi-nally reached with Shi’ite and Kurdish political leaders to add 15 voting and 10 nonvoting Sunni delegates to the 55-member drafting committee.

All the horsetrading meant that the constitutional committee did not begin serious deliberations until early July. Faced with an August 15 deadline and a host of contentious yet far-reaching and consequential issues, the committee produced a compromise document that left no fewer than 53 articles to be resolved at some point in the future.9 Even so, the Sunni committee members refused to validate the document. Their main fear was that Iraq would collapse into a loose federal struc-ture with a Kurdish region in the north and one massive Shi’ite region in the south—together controlling all of Iraq’s oil and gas wealth—thus leaving the Sunnis with a central region bereft of resources and power.

As the October 15 referendum approached, intense bargaining among the various groups continued, this time with heavy U.S. involvement. The goal was to reach some political compromise that would address Sunni concerns, particularly those relating to Iraq’s envisioned federal structure. Three days prior to the referendum, the Shi’ite and Kurdish leaders consented to the demand of the largest Sunni party, the Iraqi Islamic Party (IIP), for an additional article that called for a new consti-tutional committee to be formed after the general elections, with the task of implementing within four months of its creation “necessary amendments to the constitution.”10 Its wish having been granted, the IIP

230 Iraq’s Year of Voting Dangerously

urged its followers not only to participate in the referendum, but also to vote in favor of the constitutional document. The following day, insur-gents bombed IIP headquarters in the radical Sunni city of Fallujah.

Bringing the IIP on board did not overcome Sunni opposition to the constitution, however. While Shi’ite and Kurdish areas voted over-whelmingly for the adoption of the constitution, two Sunni provinces voted heavily against it, and a third registered a 55 percent negative vote. In fact, the Sunnis narrowly missed defeating the constitution altogether, since Iraq’s interim constitution stipulated that a two-thirds majority against the constitutional document in any three of Iraq’s 18 provinces would result in its rejection. While the Sunnis failed again to achieve their goal, the feeling among their political leaders was that they made a much greater statement through participation in the refer-endum than they had by boycotting the January elections.

The December General Elections

As a result, there were hardly any calls among Sunnis for boycotting the December 2005 election of a four-year National Assembly that would put in place a Presidency Council (of a president and two deputies), a prime minister, and a cabinet. Another important factor in the Sunni decision to participate in the December election was a significant change in the electoral system. While retaining the general system of PR with-out any threshold for representation, the Transitional National Assembly opted this time to employ a two-tiered system very close in design to the one recommended in March 2004 by both Iraqi and international advisors. Slightly more than 80 percent of the seats (230 of 275) were allocated to the provinces as multimember districts, while the remain-ing 45 seats were to be filled as “compensatory seats” from national lists to achieve overall proportionality. Each province would have a share of the 230 seats proportional to its share of registered voters in the country. Thus, regardless of how many people actually went to the polls in their provinces, the Sunnis were effectively guaranteed a certain bloc of seats in the new Assembly.11

Under this system, the seats in parliament were filled in two steps. First, the number of votes cast in each province (formally called governorates) was divided by the number of allocated seats to produce the “governorate quota.” Any party that achieved the quota was entitled to one seat, and additional seats were allocated in multiples of the quota. This process was repeated in every province to fill the 230 district-based Assembly seats. The remaining 45 seats were then distributed first among the parties and entities that did not win seats on the provincial level but were able to accumulate votes nationally equal to or higher than the “national elec-toral quota.” This quota was arrived at by dividing the total number of votes in the country by the number of Assembly seats (275). Only one

Adeed Dawisha and Larry Diamond

231

party was able to earn a seat through this compensatory process. The re-maining 44 seats were then distributed among all participating parties in accord with their respective national vote shares. Expatriate voting, which took place in 15 countries, was designed to claim a portion of these seats.

The Independent Electoral Commission of Iraq (IECI) announced on October 29 that 228 political parties, coalitions, and other entities had registered to compete in the elections. It was obvious from the start, however, that five coalitions would claim the lion’s share of Assembly seats. These were the UIA (having acquired stronger Islamist credentials with the withdrawal of some Shi’ite secularists such as Ahmad Chalabi’s Congress party and the incorporation of a group loyal to firebrand Shi’ite cleric Muqtada al-Sadr), the Kurdistan Alliance, Allawi’s secular Iraqi National List, and two Sunni lists—the Iraqi Accord Front and the Iraqi Front for National Dialogue.

The most noticeable aspect of the campaign was its seeming local-ization. The confluence of geographic and ethnosectarian lines in most areas of Iraq precluded serious competition between the various parties and coalitions within their respective provincial strongholds. Thus nei-ther the UIA nor the Sunni fronts bothered to electioneer in the Kurdish areas, and there was hardly any Kurdish presence in Shi’ite and Sunni provinces. Even in the large multiethnic cities—such as Baghdad, Mosul, and Kirkuk—there was little crossing of ethnosectarian lines. Rather, each alliance focused its energy on cementing support among its own base, while doggedly obstructing intrusions by other alliances into its home area. The logic of electoral politics as an identity referendum became further entrenched.

Under such arrangements, the groups that stood to lose most were the secular nonethnic parties and coalitions, whose performance in the elec-tions depended on conducting an effective national campaign. For example, Allawi’s Iraqi National List, the most sophisticated and best financed of the secular parties, encountered staunch resistance in areas considered by the UIA to be its own exclusive domain. Allawi’s party headquarters were attacked by gunmen in Basrah and Nasiriyah, bombed in Najaf, and burned to the ground in Karbala.12 Such activities were not confined to the UIA, but involved also the Sunni coalitions and the Kurdistan Alliance. It seems that what one group considered legitimate campaigning, the other saw as unacceptable electoral trespassing.

More unsettling still was a wave of assassinations and armed attacks in Baghdad and other Iraqi cities. A number of these were aimed at Sunni politicians who advocated electoral participation and rejection of violence. Two days after declaring that “nonparticipation in the com-ing elections is tantamount to treason,”13 a prominent leader of the Accord was gunned down in broad daylight just outside Baghdad. In-deed, the Front claimed that no fewer than ten of its members were killed, including the head of its organization in the Sunni city of Ramadi.

232 Iraq’s Year of Voting Dangerously

Workers for other parties and groups were also killed either in indi-vidual attacks or in more coordinated assaults on party headquarters.14 While such levels of electoral violence and intimidation are hardly unknown in emerging democracies, they did compound the difficulties already presented by the general lack of security in Iraq, and the result-ing climate severely limited the party leaders’ freedom of movement. They consequently seemed to settle on campaigning through the mass media, especially the electronic media. Satellite and television stations carried incessant commercials for the various parties; in this domain, the Iraqi National List was by far the slickest and had the deepest pock-ets. But this was balanced by a plethora of feature stories, interviews,

and debates involving prominent members of most parties.

Concern over the lack of security, basic services, and economic op-portunity dominated much of the campaign rhetoric. The most blistering verbal attacks came from Iyad Allawi and his fellow candidates, who accused the UIA-led transitional government of indecisiveness, incom-petence, and corruption while relentlessly decrying the country’s many ills—the decrepit sewage system, the paralyzing shortages of electric-ity, the dearth of jobs, inadequate health care, and above all the violence and chaos of everyday life. In a widely-reported speech shortly before the elections, Allawi declared that the government “has sunk into fi-nancial and administrative corruption . . . [and] has left Iraq a weak and divided country on the verge of civil war.”15 He also strongly implied that a UIA government would continue doing Iran’s bidding in Iraq, and therefore was enshrining sectarianism in a manner that would lead to the disintegration of the country.

The two Sunni coalitions focused their campaigns on what could be construed as a reversal of the security issue: They attacked the transi-tional government and the Ministry of the Interior for allegedly unleashing, or at least encouraging, irregular Shi’ite militiamen to as-sassinate innocent Sunni men in retaliation for atrocities committed by Sunni insurgents. Echoing Allawi’s accusation, Sunni politicians also invoked the ghost of bloody civil war, which they placed squarely on the shoulders of the transitional government and its allegedly sectarian and partisan posture.

The second major theme of the campaign was the debate over feder-alism. From the very beginning of the new political order, the Kurds had made it clear that the price for their willingness to be part of Iraq was an acceptance by the Shi’ites and Sunnis of a loose federal structure in which Kurdistan would be accorded substantial autonomy within Iraq. While debate initially raged over the issue, it was not long before Kurdistan’s autonomy had become a fait accompli that even the Sunnis grudgingly accepted. But the federalism debate reappeared with renewed intensity when SCIRI chief and UIA leader Abd al-Aziz Hakim pro-posed the creation of a Shi’ite super-region uniting all nine southern

Adeed Dawisha and Larry Diamond

233

provinces. Sunnis saw the move as a bid to assert control over the 80 percent of Iraq’s oil and gas wealth that lay deep in the Shi’ite south, and even as a precursor to the dismantling of Iraq. Bitter denunciations of the plan became a staple of the Sunni campaign.

Another issue on which the parties clashed was the role of religion in politics. The secular Iraqi National List railed against any move to insert religion into politics, implicitly and at times directly criticizing the Shi’ite Islamist UIA. Interestingly, the Sunni electoral alliances—whose most powerful party, the IIP, was an offshoot of the Muslim Brother-hood—also advocated the separation of religion from politics. Obviously, an Islamic state of the Shi’ite variety was hardly palatable to the Sunni fundamentalists. Meanwhile, the UIA, visibly constrained in its defense of the inseparability of religion and politics, held that it sought to re-main true to its Islamist essence while reassuring voters that religion would constitute only one of several factors shaping public policy.16

Hovering over the religion-in-politics debate was the figure of Grand Ayatollah Sistani, the most senior and revered Shi’ite cleric. While the reclusive ayatollah himself did not make public comments or state-ments during the campaign, his assistants and disciples spoke on his behalf, emphasizing his neutrality. This did not, however, stop the UIA and other Shi’ite groups from appropriating Sistani to their cause by invoking his name, declaring their fidelity to him, and putting his im-age on their posters—a practice bitterly denounced by other groups, particularly members of the Iraqi List.

Finally, the campaign saw many references to the “occupation” of Iraq by U.S. and other coalition forces. But this topic never lived up to expectations that it would be the one burning issue to dominate the campaign. No party supported open-ended occupation, and they all pledged to negotiate with the United States some kind of timetable for withdrawal, taking into consideration existing constraints. Naturally, the Sunni parties were the most insistent, but apart from the radical Council of Islamic Scholars—which had rejected participation in an election under occupation—no one advocated immediate withdrawal.

Despite the violence and disruption that infused the campaign, Ira-qis went to the polls with a broad awareness of where the main coalitions stood on the dominant issues. The election debate had been generally absorbing and at times fairly sophisticated, suggesting that it would have an impact on voter behavior. Yet when the voters finally went to the polls, they ended up turning to their primordial loyalties very much as they had in the January 30 elections.

More than 12 million Iraqis—representing over 77 percent of regis-tered voters—cast ballots on December 15.17 The two Sunni provinces of Salah al-Din and Anbar, the three Kurdish provinces of Erbil, Duhok, and Sulaymaniya, and Kirkuk with its Kurdish majority had the highest turn-outs (84 percent and above). The average turnout in the nine Shi’ite

234		Iraq’s Year of Voting Dangerously
TABLE 1—RESULTS OF THE DECEMBER 2005 IRAQI ELECTIONS
	PERCENT OF	PERCENT OF	GOVERNORATE	NATIONAL	TOTAL
PARTY OR ALLIANCE	NATIONAL	NATIONAL
	VOTE	SEATS	SEATS	SEATS
United Iraqi Alliance (Shiite)	41.2	46.5	109	19	128
Kurdistan Alliance	21.7	19.3	43	10	53
Accord Front (Sunni)	15.1	16.0	37	7	44
Iraqi List (secular)	8.0	9.1	21	4	25
National Dialogue Front (Sunni)	4.1	4.0	9	2	11
Kurdish Islamic Union	1.3	1.8	4	1	5
Liberation & Conciliation (Sunni)	1.1	1.1	3	0	3
Risaliyoon (Shi’ite Sadrists)	1.2	0.7	1	1	2
Turkomen Front	0.7	0.4	1	0	1
Iraqi Nation (secular)	0.3	0.4	1	0	1
Yezidi Front	0.2	0.4	1	0	1
Rafidayn List (Christian)	0.4	0.4	0	1†	1
Total	95.3*	100	230	45	275

* The remaining votes were won by more than 200 other lists that failed to qualify for a single seat.

† This was technically a compensatory seat. This was the lone list to qualify for a national compensatory seat after failing to win any governorate seats.

provinces fell below the national figure at 71 percent, while Baghdad had one of the lowest turnouts, yet still managed a respectable 70 percent.

To ensure the integrity of the elections, the IECI deployed 126,125 observers in all 18 provinces. In addition to 949 international monitors, the political parties and coalitions themselves spread another 272,295 agents across the 31,348 polling stations around the country. As the chief electoral officer remarked, “This election has been one of the most observed in the whole world.”18

Still, when partial results were announced a few days after the elec-tion, showing a victory for the UIA, a deluge of complaints alleging widespread fraud erupted in Baghdad and Sunni areas. Thousands of demonstrators took to the streets denouncing the IECI and accusing it of doing the UIA’s bidding. The pressure became so intense that the IECI refrained from publishing the full and final results until an international commission, which arrived in Baghdad in late December, looked into the 1,985 complaints received by the IECI. The international commission decided that while infractions had indeed occurred, they had been mostly minor and would not affect the final distribution of Assembly seats. Hence, the final and complete results were announced on 17 January 2006 and certified on February 10, confirming once again both the victory of the UIA and the heavily sectarian character of the voting (Table 1).

The results show a clear preponderance of ethnosectarian loyalties. Any hopes that the electorate would vote to separate religion from poli-tics or to transcend ethnic fissures were completely frustrated. The wish, certainly harbored by the Bush administration, for a muscular showing by the secular Iraqi List that would cement Allawi’s bargaining posi-

Adeed Dawisha and Larry Diamond								235
TABLE 2—ETHNOSECTARIAN PATTERNS OF DECEMBER 15 VOTING
		# OF		UIA &	IRAQI	SUNNI	KURDISH
GOVERNORATE				OTHER
	ETHNICITY	SEATS			LIST	LISTS	LISTS	OTHERS
				SHI’ITES
Basrah	Shi’ite	16		13	2	1	–	–
DhiQar	Shi’ite	12		11	1	–	–	–
Babel	Shi’ite	11		9	1	1	–	–
Najaf	Shi’ite	8		7	1	–	–	–
Qaddisya	Shi’ite	8		7	1	–	–	–
Misan	Shi’ite	7		6	1	–	–	–
Wasit	Shi’ite	8		7	1	–	–	–
Karbala	Shi’ite	6		5	1	–	–	–
Muthana	Shi’ite	5		5	–	–	–	–
Anbar	Sunni	9		–	–	9	–	–
Salah al-Din	Sunni	8		1	1	6	–	–
Sulaymaniya	Kurdish	15		–	–	–	15	–
Irbil	Kurdish	13		–	–	–	13	–
Duhok	Kurdish	7		–	–	–	7	–
Baghdad	Mixed1	59		35	8	14	1	1
Nineveh	Mixed2	19		2	2	10	4	1
Diyala	Mixed2	10		2	1	5	2	–
Kirkuk	Mixed3	9		–	–	3	5	1
1 Shi’ite majority; 2 Sunni majority; 3			Kurdish majority

tion—or even allow him to emerge as the central character in a secular coalition—was shattered when the party ended up obtaining only about 8 percent of the vote, losing almost half the seats that it had received in the January 2005 election. The extent of the ethnosectarian character of the vote is better observed when voting is broken down at the provin-cial level. As Table 2 shows, the UIA dominated the voting in the nine southern Shi’ite provinces, winning 70 of the 81 district seats.19 The two Kurdish lists (secular and Islamic) together won all 35 seats in Kurdistan. And the Sunni lists won 15 of the 17 seats in the predomi-nantly Sunni Anbar and Salah al-Din provinces.

Why did the change in electoral system have no apparent effect in diminishing the sectarian pattern of the voting? There are at least three possible explanations. First, the electoral terrain was strongly defined by the January 30 elections and the old electoral system. Compounding all that had transpired in the nature of governance and politics since the fall of Saddam’s regime, the January elections created such momentum for identity politics that it would have been asking too much of the district-based voting system to effect much change. Second, one might have expected the district-based PR system to aid the election of local and independent candidates. In the climate of widespread violence, however, a great many candidates declined even to reveal their identi-ties, putting the premium on identification with parties and alliances.

236 Iraq’s Year of Voting Dangerously

Third, the only list that transcended the country’s deep ethnic and sec-tarian cleavages was led by a politician, Iyad Allawi, whose tenure as interim prime minister had suffered from most of the flaws of ineffective performance for which he later condemned the UIA-led transitional gov-ernment, and who was easily stigmatized as a former Ba’athist and paid agent of the CIA. This underscores that, when the underlying pressures and constraints are powerful and entrenched, a change in the electoral system may in the near term do little to transcend them.

The New Policy-Making Institutions

Since no party achieved an overall majority, the elected members of the National Assembly will have to cross ethnosectarian lines to get things done. In fact, a two-thirds majority (184 of the 275 members) is required to pass some significant measures. To begin with, no govern-ment can be formed until two-thirds of the National Assembly agrees on a single slate of candidates for the Presidency Council. Moreover, in constructing a viable coalition, factors other than numbers have to be considered. The specter of the insurgency, for instance, makes Sunni demands far weightier than their 55 seats. While it was obvious to all parties that Sunnis had to be included in the government, the dilemma after the election was how to balance the distribution of cabinet portfo-lios among the prospective coalition partners. The debate showed clear disparities in the aspirations and positions of the various parties: While Sunnis talked of a national-unity government, implying a kind of equal-ity among the coalition parties, the UIA emphasized a governmental balance based on the election results.20 For the sake of successful gover-nance and effective policy making, significant compromises on all sides will need to be made and adhered to, not just in the immediate period but throughout the four-year life of the Assembly.

One factor that might enhance the fluidity of alliance-making is that, while the various party lists seem to be fairly cohesive on matters of identity, they differ within themselves on other issues. The UIA is par-ticularly noteworthy in this regard. While Islamist Shi’ism constitutes the raison d’^etre of its four main constituent groups—SCIRI, al-Da’wah, al-Fadhilah, and the Sadrists—personal tensions, historic rivalries, and policy differences permeate their relations with one another. In fact, al-Fadhilah announced its withdrawal from the UIA on November 1, citing unhappiness with some of Hakim’s policies and his handling of Assem-bly-seat distribution. They ended up remaining part of the UIA only because the IECI would not accede to their request.21

Differences surfaced after the election when al-Fadhila talked about the possibility of negotiating with the Allawi group, breaking with a seeming UIA consensus and drawing a sharp rebuke from the Sadrists.22 Neither, for that matter, are relations between SCIRI and the Sadrists

Adeed Dawisha and Larry Diamond

237

consistently fraternal. The Sadrists, who resent Hakim’s seeming de-meanor as the acknowledged head of the UIA, censured him for negotiating with the Kurds after the election, calling it “a personal endeavor that did not enjoy the UIA’s blessings.”23 The Sadrists seem to have a pronounced affinity with Sunni groups, and as such do not sup-port Hakim’s idea of a nine-province Shi’ite region. Moreover, the militias of SCIRI and Sadr have clashed bloodily in the recent past.

Divisions within the other groups may not be as well publicized, but they exist. It remains to be seen whether the alliance of the two Kurdish leaders, Barzani and Talabani, and their parties will stand the test of time, or whether the two Sunni blocs will consistently vote together and refrain from striking different political bargains and entering into new alliances. There is thus some possibility in the new Assembly of increas-ing fluidity in coalitional politics, as new issues surface and the benefits of power await distribution. With good fortune and artful political man-agement, it is possible that Iraq could gradually edge away from the high degree of ethnic and sectarian polarization produced by the past “year of voting dangerously.” The ability of the parties and politicians to manage and mitigate this polarization, through negotiation and com-promise, will likely determine not only the scope for democratic development but even Iraq’s ability to survive as one country.

NOTES

1. The transition plan announced by U.S. officials on 15 November 2003 provided for direct elections for a constitutional convention but only indirect elections for the Transitional National Assembly. This met with intense Iraqi pro-test and was abandoned after UN mediation in favor of direct elections for the National Assembly—which was to act as both a parliament and a constitutional convention—by 31 January 2005.

2. Although this was the widely shared view of international experts and most Iraqi political leaders, some Americans nevertheless continued strongly to advo-cate FPTP while condemning PR. See, for example, Michael Rubin, “The Wrong Elections for Iraq,” Washington Post, 19 June 2004.

3. Donald L. Horowitz, “Electoral Systems: A Primer for Decision Makers,” Journal of Democracy 14 (October 2003): 115–27.

4. Later, when Iraqi ministers were appointed under the CPA, the ethnic and sectarian balance in the Governing Council was replicated.

5. In the typical situation, an electoral threshold is often favored in part to keep small extremist parties out of parliament. In Iraq, however, it became apparent that radical movements and parties like Muqtada al-Sadr’s probably had enough support to clear any electoral threshold, while some of the more liberal politicians and parties would likely have had to struggle to obtain the needed vote share.

6. The population shares could be estimated based on either the 1997 census or the food-ration data, but in fact they yielded similar percentages for each province. The U.S. advisor also suggested the possibility of breaking up the largest prov-inces, particularly Baghdad (which had almost a fourth of the 220 district seats),

238 Iraq’s Year of Voting Dangerously

into multiple districts.

7. Carina Perelli, Coalition Provisional Authority Briefing, Baghdad, 4 June 2004. Transcript available at www.iraqcoalition.org/transcripts/20040604_Perelli _Prep.html.

8. The Sunni parties feared that without guaranteed district representation, based on the provinces, the much greater degree of violence, intimidation, and resentment in their areas would result in a sharply lower voter turnout, leaving them underrepresented. Yet by boycotting, they left themselves virtually unrepre-sented. Larry Diamond, Squandered Victory: The American Occupation and the Bungled Effort to Bring Democracy to Iraq (New York: Times Books, 2005), 323.

9. There was in fact strong sentiment in the Transitional Assembly to vote for a six-month postponement of the deadline, as the interim constitution allowed on a one-time basis. But the United States was strongly opposed to it, and the Sunnis as well were not enthusiastic about remaining in the inferior status of unelected members.

10. Al-Rafidayn (Baghdad), 13 October 2005.

11. Specifically, all or at least most of the nine seats allocated to Anbar Province and the eight seats allocated to Salah Al-Din, plus a good share of the seats in the mixed provinces such as Nineveh, which has 19 seats.

12. For the above incidents, see Al-Sharq al-Awsat (London), 9 and 11 Decem-ber 2005; Al-Ahali (Baghdad), 7 December 2005; and Al-Rafidayn (Baghdad), 29 November and 17 December 2005.

13. Al-Sharq al-Awsat, 30 November 2005.

14. New York Times, 11 December 2005.

15. Al-Sharq al-Awsat, 11 December 2006.

16. The gist of the UIA position was that political decisions are based on considerations of the national interest, but most Iraqis after all are devout Muslims who will not be comfortable with Western-type secularism.

17. Out of 15,568,702 registered voters, 11,895,756 valid votes, 139,656 in-valid votes, and 62,836 blank votes were cast in all Iraqi governorates.

18. “Observers’ reports tell of good conduct of elections,” Independent Elec-toral Commission press release, 24 December 2005, available at www.ieciraq.org.

19. The disparity between the UIA’s percentage of national votes and parlia-mentary seats is probably explained by the IECI’s decision to use Hare’s “Largest Remainder Method” to distribute seats.

20. Al-Sabah (Baghdad), 14 January 2006; al-Arabiya television station, 14 January 2006.

21. Al-Rafidayn, 2 December 2005.

22. Al-Hayat (London), 6 January 2006.

Al-Ahali, 4 January 2006.