Violence in Iraq
by Aymenn Jawad Al-Tamimi
DATA GATHERING ON VIOLENCE
When it comes to statistics on violence in Iraq, there are a number of sources one can use for information, including the "Iraq Body Count" (IBC), the Iraqi government, and the United Nations. The IBC's methodology is to rely solely on reports put out in media outlets and in press releases from various organizations. For example, IBC uses the American news channel ABC News, the British newspaper The Independent, and the Iraqi newspaper al-Mada.
Of course, the limitation of this method is that there will inevitably be some attacks and civilian casualties that are missed by media reports, and so the totals collected by the IBC almost certainly constitute an undercount. Indeed, those who run the IBC stated the following back in 2007: "Recent public debate has rather focused on the number of deaths we don't record, and how much of an undercount that might be. Our own view is that the current death toll could be around twice the numbers recorded by IBC and the various official sources in Iraq."
In any event, the IBC's figures–based on corroborating evidence from multiple outlets with no record of outright fabrication–can certainly be taken as a much more reliable guide to death tolls in Iraq than the statistics put out by the Iraqi government. For instance, while the IBC records 320 and 309 violent civilian deaths for March and April 2012 respectively, government figures–derived from a source in the Health Ministry–gave numbers of 78 and 88 civilians killed for those same months.
These numbers are too low to be put down to lack of resources in data gathering. On the contrary, they reflect clear political motivations on the part of the government, currently led by Prime Minister Nouri al-Maliki, who not only holds the positions of Interior and Defense Minister but also leads a coalition under the mantle of "State of Law," which has promoted itself among Iraqi voters with a platform of ensuring security for the population at large. After all, as prime minister, Maliki has presided over the dramatic drop in violence since the days of the all-out sectarian civil war in 2006.
However, the problem of underestimating the number of civilian casualties did not always afflict the Iraqi government. In this context, one must turn to the question of the total number of violent civilian deaths since 2003. As of August 21, 2012, the IBC gives a tally in the range of 108,225-118,264 civilian deaths and notes that an analysis of the Iraq War Logs leaked by Wikileaks may add an additional 13,000 fatalities. In the period January 2003 until June 2006, the IBC gives a total figure of 51,407 violent deaths. Meanwhile, the Iraq Family Health Survey (IFHS), "a cross-sectional, nationally representative survey of 9345 households, [which] was conducted by relevant federal and regional ministries in Iraq in collaboration with the World Health Organization," estimated 151,000 violent deaths in the period March 2003 until June 2006, with a 95 percent uncertainty range spanning a minimum of 104,000 to a maximum of 223,000.
In contrast, a report published in the British medical journal The Lancet estimated that in the same period, "654,965 persons have died as a consequence of the conflict. Of these, 601,027 have died from violence." This survey, which gave an uncertainty range of 392,979-942,636, followed on from a study that was published in The Lancet in October 2004, which asserted the following:
Unsurprisingly, The Lancet surveys attracted widespread public attention, as commentators on both the left and right of the political spectrum debated the figures in relation to questions like the morality of the decision to invade and coalition or insurgent responsibility for civilian deaths. Many pundits who accepted the figures given by The Lancet would later go on to trumpet figures loosely based on the findings of a survey by the UK-based polling firm "Opinion Research Business" (ORB), which estimated 1,220,580 violent deaths with the range 733,158-1,446,063.
For example, the left-wing polemicist Johann Hari has repeatedly made reference to the "Iraq war, which has led to the death of a million people." Previously, Hari had invoked the second Lancet survey's numbers, as when he wrote in one piece: "The US troops cannot be an agent of anything positive in Iraq, after using chemical weapons in civilian cities, after using torture routinely, after overseeing the death of 650,000 Iraqis."
From the opposite side, neoconservative Irish blogger Mark Humphrys vigorously attacks The Lancet studies, highlighting in particular the fact that the editor of The Lancet, Richard Horton, appeared at a demonstration organized by the Stop the War Coalition in Manchester in September 2006. At this rally, Dr. Horton took the stand to address the protestors, denouncing the "axis of Anglo-American imperialism." Humphrys then offers a counter "NoBody Count," which claims that 475,000 lives have been saved by the Iraq War, a calculation said to be "based on Saddam's average rate of killing when he was in power."
The personal sentiments of the authors of The Lancet studies–or of the editor of the journal–regarding the invasion do not necessarily throw the findings into doubt. However, it is apparent that a fundamental problem with the research for both studies has been a lack of transparency on the part of the authors, such that the results can be considered discredited. The flaws in The Lancet studies highlighted by numerous academics and researchers over the years–and frequently overlooked in polemical debates between commentators–are aptly summarized by analyst Joel Wing of Musings on Iraq.
For example, the authors of The Lancet studies have refused to make their data public (i.e., entirely available to everyone, rather than only in limited form to certain individuals on request), which violates the scientific spirit of enquiry that entails being open about one's data and willingness to share them so that they may be verified and corroborated. For example:
In addition, the authors were guilty of multiple inconsistencies concerning explanations for how they carried out the research. For instance, when it came to the first study, Roberts said that a cluster (collection of randomly selected households) in a rural area could take as long as six hours to complete. Over time, the claims as regards time taken diminished, such that for the 2006 paper, the authors of the studies were claiming that for a given cluster of 40 households, the task of conducting interviews could be completed in two hours. This would mean requiring just three minutes to interview one household. Wing summarizes the implications of this point:
That all these procedural steps could be accomplished in just three minutes is highly implausible. Anticipating such criticism, Gilbert Burnham (one author of The Lancet studies) claimed that for the 2006 study, when the researchers arrived in a neighborhood, they would divide themselves into two groups. However, a contributing Iraqi researcher denied that this was the case, and it should be noted that Burnham's assertion was not mentioned in the published article.
It can therefore be concluded that either Burnham is lying or that members of the research team(s) assigned to collect data misled him as regards their methods. In fact, it is also possible that the researchers fabricated data here, but as Wing notes, "This [controversy] could have been resolved if the authors released the start and end times for the survey work. This proved impossible, because according to them that information was destroyed to protect the interviewees' safety."
In any event, the authors of The Lancet studies were caught out for lying about the inclusion of spaces in the survey forms to fill in the names of individuals. It is in light of this lack of transparency–and the serious reservations that follow regarding integrity and credibility–that one can now invoke the issue of political leanings. The first study was intended for publication in the run-up to the 2004 U.S. presidential elections, while the latter was released at the height of the sectarian civil war, a development that naturally provoked considerable debate as to the best course of action for the United States and other intervening powers to undertake.
By pushing figures generally deemed to be vast overestimates, the authors of The Lancet–as well as the editor of the journal–were evidently trying to advance a partisan anti-war agenda. To gain an idea of the scale of the overestimates, it would mean that data gathering and corroboration from multiple media outlets in the period June 2005-June 2006 was missing out on some 90 percent of violent deaths. Could so many fatalities have really gone unreported?
The ORB's survey was even more problematic and has been thoroughly debunked in an article by Michael Spagat (a professor at Royal Holloway, University of London) and Josh Dougherty of the Iraq Body Count. Both authors begin by noting, as with The Lancet studies, a lack of transparency in disclosing details of the methodology behind the ORB survey. For example, "ORB declines to disclose the field versions of its questionnaire in Arabic and Kurdish, on the ground that such disclosures would lead to what ORB believes would be unproductive discussions of the quality of its translations;" nor does it disclose details of its sampling frame.
Lacking sufficient information on sampling methodology, one cannot give a thorough overview of sampling error, but problems with error calculation are highly significant. This is because ORB treats "error margins as if it had drawn a simple random sample," while completely ignoring "clustering effects" in error calculation. In addition, although the survey includes weighting by governorate and rural-urban divides, neither of these divisions is accounted for in calculating error.
More significantly, however, Spagat and Dougherty uncovered problems in the data percentages provided by three polls conducted by ORB in the period from 2007-2008. In the first poll, 26 percent of those interviewed said a "member of my family/relative" had been murdered in the last three years. In the second poll, 20 percent of respondents said that a household member (i.e., one actually living in the same house) had perished as a result of the violent conflict, while in the final poll, 24 percent of those interviewed said that that "a member of my family/relative" had been murdered since the invasion.
The results of the second poll, conducted in August 2007, were used to extrapolate to the ORB's estimate of more than a million violent deaths. Yet the ORB's estimate is difficult to reconcile with the findings of the first and third polls, which deal with violent deaths of extended family members rather than those in the immediate household. The dilemma is elaborated on as follows:
The problem is ultimately traced to the significant likelihood of inflated data in the central regions of the country for the second poll in comparison with the third and first polls. To illustrate the contrast, in what ORB defined as central regions, 36 percent of respondents reported the death of a household member in the second ORB poll, compared with 26 percent reporting the violent death of a family member or relative over the past three years (i.e., 2004-2007) in the first ORB poll. It is the first aforementioned result that is very odd. In southern regions, only 7 percent said they had lost a household member in the second poll, compared with 35 percent who noted the death of a family member or relative in the first poll.
Such a pattern is hardly unreasonable, and one can infer two reasons the data for central regions seem to be so high. First, there may have been an accidental mistake on the part of the researchers who were conducting the second poll in central regions; that is, they may have posed the question from the first poll to interviewees for the second poll. Alternatively, there may have been deliberate tampering with the data in an effort to attain a large estimate for the number of violent deaths in the Iraq War. In any case, it is clear that the OCB survey is deeply flawed and, like The Lancet studies, cannot be taken as a reliable guide to the true number of excess deaths in Iraq since the 2003 invasion.
Taking into account the actual documentation from the IBC, and the research undertaken by Joshua Dougherty in particular to refute exaggerations of death tolls, it seems most reasonable to take IBC as the best guide to violence in Iraq. Whatever the personal sentiments of those behind the IBC, it is clear they have not allowed politics to interfere with their work. In this author's view, taking into account the statistics compiled by the IBC and the Iraq Family Health Survey, the total number of violent excess deaths since 2003 is likely to be in the range of 150,000-300,000.
As can be inferred from Figure A in the appendix, there were thousands of violent deaths as a direct result of coalition military operations in the invasion that deposed Saddam Hussein's regime in early 2003. The anomaly in 2005 aside, violence peaked in the period mid-2006 to mid-2007, thereafter declining dramatically in spite of the brief surge in violent deaths in early 2008. From late 2008 onwards, it is apparent that violence reaches what many analysts have termed an "irreducible minimum." That is, the violence has generally stabilized at levels much lower than those observed during the sectarian civil war. Indeed, the weekly and monthly fluctuations by this time give no evidence of an overall trend either way. Instead, an examination of fluctuations in deaths on a monthly basis demonstrates the existence of cyclical patterns. For a better understanding of violence in Iraq now, January 2009 is worth taking as the starting point.
Using the numbers in Figure B, it can be calculated that the summer seasons of 2009-2011 averaged 434.78 violent deaths per month, while the autumn seasons in 2009-2011 have a mean of 317.78 violent deaths per month. In the same period, spring seasons had an average of 369.33 violent deaths per month. It can therefore be inferred that as the climate warms, insurgents generally step up their operations. However, the number of attacks may not necessarily bear a perfectly direct correlation with casualties, because one or two successful mass casualty attacks can easily lead to an upsurge in statistics for violent deaths in a given month.
In any event, it is also worth noting that insurgents step up operations around the time of religious festivals, in which Shi'i pilgrims travelling on foot or on bus are easy targets for bombs. Thus, in January 2011 and 2012, there was the festival of Arba'in, which helps explain the sudden upsurge in violent deaths on both occasions. In a similar vein, June 2012 saw more than twice as many fatalities as May 2012, partly because there was a wave of bomb attacks targeting pilgrims who had gathered to commemorate the death of Musa al-Kadhim, who was the great-grandson of Muhammad. December 2011 should also be taken into consideration, as this month saw a spike in violent deaths, with the twenty-second day of the month being particularly noteworthy for a series of coordinated bomb attacks in Baghdad that killed at least 69 people, only days after the last U.S. troops had withdrawn from Iraq. Many attempted to tie the attacks to the political crisis that had emerged following the issuing of an arrest warrant against the Sunni vice president Tariq al-Hashimi.
In reality, however, the perpetrators had probably been planning the coordinated operation well before the announcement of an arrest warrant against Hashimi on December 19. That Washington had made it clear that all troops would be brought home before Christmas time was publicly known some two months in advance following the breakdown in negotiations over a possible troop extension, a development that owed itself to the U.S. refusal to compromise on the issue of immunity for its troops from prosecution in Iraqi courts. Al-Qa'ida in Iraq (AQI), which claimed responsibility for the bombings, had evidently been planning the attacks well in advance–irrespective of the political crisis–in a bid to show that the group had still not been wiped out by U.S. troops or the Iraqi security forces, and that the foreign occupation forces had essentially been "defeated" by the resistance. Such a demonstration of strength would undoubtedly serve to win financial support from supporters both in Iraq itself and abroad.
With U.S. troops gone, it is worth assessing who the main perpetrators of violence in Iraq are, and what, if any, are the exacerbating factors behind the violence. Both questions are relevant to consider in predicting how trends in violence might pan out in the future.
On account of the U.S. withdrawal, the Shi'i militant groups–in particular, the Iranian-backed Special Groups–that had been fighting U.S. troops and to a lesser extent the Iraqi security forces have since disbanded and decided to join the political process. Such an approach is especially advantageous from Iran's point of view, because these groups tend to be at odds with the already well-established Shi'i political factions. For example, the League of the Righteous, led by Qais al-Khazali, originated as an offshoot from Muqtada al-Sadr's Mahdi Army. Following the pullout of the last U.S. troops, Khazali announced on December 26, 2011 that it would lay down its arms, prompting a sharp rebuke from Sadr, who accused the League of the Righteous of being responsible for killing Iraqi soldiers and police, and of serving as agents for Iran. The point is that should the Special Groups gain any substantial influence in Iraqi politics in the years to come, divisions among the Shi'i factions will increase, and so Iran can take on the role of mediator and kingmaker.
At the same time, it should not be thought that these militants have necessarily handed over their weapons. Notably, there have been reliable reports that many of these very same Shi'i militants are now fighting inside Syria against the rebels. In addition, members of the Special Groups have been putting up posters of Ayatollah Khamene'i in Baghdad, essentially threatening violent intimidation against those who might dare to protest openly (and indeed, there has been much discontent among the public, Shi'a included).
With the Shi'i insurgency virtually defunct, the remaining insurgent organizations are entirely Sunni Arab. Of these groups, the two most prominent are AQI (also known as the Islamic State of Iraq) and the Naqshibandia. While the former plays the most prominent role in attacks on government, the security forces, and civilians, its name should not be used as a cover term for all insurgent attacks.
The Naqshibandia (fully: Jaysh Rijal Tariqah al-Naqshibandia) is the main surviving Ba'thi-Islamist insurgent group, and as Michael Knights of The Washington Institute for Near East Policy points out, it is probably the only insurgent group to have gained strength even as other militants were suffering heavy losses in the face of the Sons of Iraq movement, whose rise was facilitated in part by the American troop surge. In general, while AQI is responsible for suicide bomb attacks and the majority of car bomb attacks, many smaller-scale operations is the work of the Naqshibandia.
The Naqshibandia is under the command of Izzat Ibrahim al-Douri, who was deputy head of the Revolutionary Command Council during the Saddam era. While it is certainly true that the Naqshibandia "has successfully created a hybrid of Islamist themes and nationalist military expertise" (having drawn its mid-level operatives from "former military and intelligence officers who had attained ranks between lieutenant colonel and brigadier general under the Ba'athist regime"), it is far from apparent that Douri is "politically active within Iraq." On the contrary, given that there have been no credible reported sightings of Douri in Iraq since 2003, it seems more likely that he is operating in hiding outside Iraq, and the key to ascertaining his probable whereabouts may lie in a video of Douri that appeared in April to mark the 65th anniversary of the Ba'ath Party.
As Reidar Visser notes, Douri's political outlook on the wider region reflects an alignment with the Sunni Arab states against the regime of Bashar al-Asad, together with an opposition to any perceived Israeli, American, and Iranian interference in regional affairs. Thus, he is critical of the intervention against the Qaddafi regime, which entailed NATO airstrikes against loyalist forces, but at the same time he praises the Saudi king for trying to resolve the political crisis in Yemen. His criticism of the Syrian government means that it is unlikely that he is in Syria. Instead, a location in the Gulf area or the southern Arabian Peninsula seems to be most plausible, with Saudi Arabia or Yemen as possible hiding places.
Both the Naqshibandia and AQI rely on extortion from businesses as one of their main sources of financial funding. The former is particularly well entrenched in the central northern region of Iraq (roughly forming a triangle between Mosul, Kirkuk, and Tikrit), while AQI has established a very extensive extortion network in Mosul. This extortion covers sources of financial revenues at all levels, from construction projects in the city to importations of fruit and vegetables, as well as kidnappings of individual citizens and demanding ransom money from the families.
Here, it is important to note that even as AQI was suffering major setbacks elsewhere in the country from mid-2007 onwards, this extortion scheme has remained firmly in place, and efforts by Nouri al-Maliki to dismantle it have failed. As a report reprinted in the Jordanian outlet al-Bawaba noted: "In May 2008, Iraqi Prime Minister Nouri al-Maliki launched a military campaign, called 'Operation Umm al-Rabiain,' to push the ISI out of Mosul. However, the group simply went underground and changed their tactics. 'Instead of targeting homes and businesses with explosive devices, al-Qaeda started kidnapping and assassinating,' a third contractor explained."
The Naqshibandia and AQI also receive support from sources outside Iraq. For the latter, this may include Iraqi and non-Iraqi sympathizers (particularly in the Gulf states) with the group's ideology. The Naqshibandia also gains a share of these funds, besides revenue from former Republican Guard officers residing in Syria, Yemen, and Jordan. In any case, both groups appear to have a working relationship in handling and distribution of financial income.
Today, the foreign component of the insurgency has become virtually irrelevant, because the attention of foreign jihadists has been directed elsewhere. For example, Libyan fighters have largely returned to their native land, where they fought against the Qaddafi regime and have since the end of the Libyan civil war integrated into Abd al-Hakim al-Hasidi's Islamist political faction known as al-Watan. Others–including Syrians, Yemenis, Saudis, and Jordanians (and some Iraqi fighters)–have turned to fighting Asad's regime, and have been playing an important role in combating regime loyalists around Dayr al-Zour.
POLITICS AND VIOLENCE
A big question facing analysts is how far the violence in Iraq today can be linked to the political situation. In this author's view, this link tends to be misinterpreted. There is no evidence that the insurgent groups have gained recruits on account of the current political impasse in Iraq, which has seen a failed effort by factions opposed to Nouri al-Maliki to hold a no-confidence vote against the premier, together with an ongoing arrest warrant against the Sunni Arab vice-president Tariq al-Hashimi. If this were the case, then said insurgents would surely make their specific political grievances and demands clear. Instead, the ideological goals of the insurgents remain clear. Here, for example, are the words of AQI spokesman Abu Muhammad al-Adnani in February 2012, one day after a series of bomb attacks targeting Shi'i districts in Baghdad and security forces in the capital and elsewhere:
For both AQI and the Naqshibandia, nothing short of the overthrow of the government will do. That said, AQI has put out propaganda videos cynically exploiting the political crisis, with an aim not necessarily of winning recruits to take up arms, but rather persuading them not to cooperate with the government or security forces in hunting down the perpetrators of attacks. For instance, Haydar al-Khoei notes one AQI propaganda clip featuring Tariq al-Hashimi, decrying what he saw as American abandonment of the Sunni Arabs in Iraq. Yet in that same video: "Prominent Sunni politicians such as Hashemi, deputy prime minister Saleh al-Mutlak and parliamentary speaker Osama al-Nujaifi are labeled 'traitors' who were deceived by the Shia into joining the national unity government but given no real authority."
This propaganda undoubtedly works to some degree, and the political impasse is one reason why violence is still at the irreducible minimum mentioned above. Frustrated at the lack of political gain for the Sunni Arab community as a whole, despite the cooperation of the Sons of Iraq movement with Baghdad, many Sunni Arab locals in areas where insurgents are based and take refuge harbor resentment against the government and accordingly refuse to cooperate with the security forces, an attitude that is only exacerbated by the heavy-handedness of the army and police. That is, instead of adopting a "soft security" approach, "[army and police] commanders are encouraged to teach the locals a lesson to dissuade them from supporting insurgents. Communities are harassed with blockades, arrests, red tape, local curfews, etc."
On this analysis, therefore, one need not assume the sensationalist view of a return to sectarian civil war, but it is now possible to understand why the insurgency still poses a serious terrorist threat to the country. Another factor that could be behind attacks on government officials and some figures in the security forces is violence between political factions. The existence of this phenomenon in the period from 2006-2008 is well attested. For example, even as the militias affiliated with the Sadrists and the Supreme Council for Islamic Revolution in Iraq (SCIRI, which later changed its name to the Supreme Islamic Council) were engaged in fighting Sunni insurgents, they were essentially at war with each other.
Observers have often drawn attention to the sophistication of the methods of attacks on certain government officials and officers in the police and army, including the use of silencer pistols (as occurred during the assassination of a shaykh in Diwaniyya in September 2011) and the booby-trapping of officials' vehicles (as was the case in Kirkuk in the same month, targeting the colonel in charge of the emergency police). The problem is that there is no definite evidence at the present time to prove one way or the other, but the notion that political factions in Iraq might be employing hitmen against each other as a means of exerting political pressure is hardly implausible, given their histories of involvement in such violence.
Nevertheless, it is important not to overstate the extent of this phenomenon when it comes to pinning responsibility for attacks on security and government officials. The vast majority of these attacks can still be put down to the insurgency. Indeed, Haydar al-Khoei rightly notes the sophistication of AQI's operations, including "live-fire training exercises in broad daylight," driving in official SWAT-team vehicles, and using night-vision goggles. In fact, so effective are the disguises adopted by AQI operatives, that on one occasion they even mistook each other for members of the Iraqi security forces and subsequently engaged in a friendly-fire attack.
The reason for this increase in sophistication of attacks is likely the fact that following the U.S. withdrawal, the Iraqi government decided to release militant prisoners on the assumption that as there was no longer an occupation force in the country, they would cease from militancy and return to normal life. This assumption was, however, greatly mistaken. These militants had simply been biding their time in general, thinking of new ways to launch attacks should they ever be released from jail, and becoming intricately familiar with the workings of the Iraqi security forces.
CONCLUSION: VIOLENCE AND THE FUTURE
There has been speculation that the overthrow of Asad's regime in Syria will lead to a resurgence of Sunni militancy in Iraq and therefore reignite a sectarian civil war, especially if a Sunni Islamist regime comes to power in Damascus. Certainly, this concern underlies the Iraqi premier's unease about the civil war in Syria, but the fact is that the country has already experienced a civil war that ended decisively in favor of the Shi'i political factions against the Sunni insurgency. For the majority of Sunni Arabs, survival depends on adapting to the fact that the Shi'a lead the political process, the frustration at the impasse notwithstanding. Nor should too much be made of Arab-Kurdish tensions in the disputed areas. On the contrary, in cities like Kirkuk, the status quo has mainly been one of loud noise with protests and counter-protests, rather than violent clashes between rival ethno-religious groups.
A prime example of the problem of sensationalist reporting was the case of the face-off between Peshmerga and the Iraqi army at the disputed Rabia border-crossing town near the Syrian border. No deaths resulted from this incident. Any violence over imminent Kurdish independence remains highly unlikely in the near future, for any realistic notion of an independent Kurdistan is a long way off on account of the heavy financial dependence of the Kurdistan Regional Government on Baghdad.
In sum, the political impasse, the heavy-handedness of the Iraqi security forces, the failure to dislodge AQI from Mosul, lack of understanding of the rule of law on the part of the country's political factions, and the persistence of the Naqshibandia mean that violence is unlikely to decrease substantially for many years to come. The cyclical patterns observed above will almost certainly persist. At the time of writing, the southern regions (in particular, the Basra area) and the Kurdistan territory are the safest parts of Iraq, while Baghdad, Anbar, Diyala, Babil, Salahaddin, and Ninawa remain the most dangerous areas. In short, the current status-quo is unlikely to change for quite some time.
Figure A: Violent deaths on a weekly basis from March 2003 to the end of 2011, compiled by the IBC as of August 25, 2012, and using figures from the IBC database.
Figure B: Violent deaths per month as compiled from the IBC's database (as of November 8, 2012).
 "Sources Used by Iraq Body Count," Iraq Body Count, August 21, 2012.