This report is a narrative account of a campaign of extermination against the Kurds of northern Iraq. It is the product of over a year and a half of research, during which a team of Middle East Watch researchers has analyzed several tons of captured Iraqi government documents and carried out field interviews with more than 350 witnesses, most of them survivors of the 1988 campaign known as Anfal. It concludes that in that year the Iraqi regime committed the crime of genocide.
Anfal--"the Spoils"--is the name of the eighth sura of the Koran. It is also the name given by the Iraqis to a series of military actions which lasted from February 23 until September 6, 1988. While it is impossible to understand the Anfal campaign without reference to the final phase of the 1980-1988 Iran-Iraq War, Anfal was not merely a function of that war. Rather, the winding-up of the conflict on Iraq's terms was the immediate historical circumstance that gave Baghdad the opportunity to bring to a climax its longstanding efforts to bring the Kurds to heel. For the Iraqi regime's anti-Kurdish drive dated back some fifteen years or more, well before the outbreak of hostilities between Iran and Iraq.
Anfal was also the most vivid expression of the "special powers" granted to Ali Hassan al-Majid, a cousin of President Saddam Hussein and secretary general of the Northern Bureau of Iraq's Ba'ath Arab Socialist Party. From March 29, 1987 until April 23, 1989, al-Majid was granted power that was equivalent, in Northern Iraq, to that of the President himself, with authority over all agencies of the state. Al-Majid, who is known to this day to Kurds as "Ali Anfal" or "Ali Chemical," was the overlord of the Kurdish genocide. Under his command, the central actors in Anfal were the First and Fifth Corps of the regular Iraqi Army, the General Security Directorate (Mudiriyat al-Amn al-Ameh) and Military Intelligence (Istikhbarat). The pro-government Kurdish militia known as the National Defense Battalions, or jahsh, assisted in important auxiliary tasks.1 But the integrated resources of the entire military, security andcivilian apparatus of the Iraqi state were deployed, in al-Majid's words, "to solve the Kurdish problem and slaughter the saboteurs."2
The campaigns of 1987-1989 were characterized by the following gross violations of human rights:
· mass summary executions and mass disappearance of many tens of thousands of non-combatants, including large numbers of women and children, and sometimes the entire population of villages;
· the widespread use of chemical weapons, including mustard gas and the nerve agent GB, or Sarin, against the town of Halabja as well as dozens of Kurdish villages, killing many thousands of people, mainly women and children;
· the wholesale destruction of some 2,000 villages, which are described in government documents as having been "burned," "destroyed," "demolished" and "purified," as well as at least a dozen larger towns and administrative centers (nahyas and qadhas);
· the wholesale destruction of civilian objects by Army engineers, including all schools, mosques, wells and other non-residential structures in the targeted villages, and a number of electricity substations;
· looting of civilian property and farm animals on a vast scale by army troops and pro-government militia;
· arbitrary arrest of all villagers captured in designated "prohibited areas" (manateq al-mahdoureh), despite the fact that these were their own homes and lands;
· arbitrary jailing and warehousing for months, in conditions of extreme deprivation, of tens of thousands of women, children and elderly people, without judicial order or any cause other than their presumed sympathies for the Kurdish opposition. Many hundreds of them were allowed to die of malnutrition and disease;
· forced displacement of hundreds of thousands of villagers upon the demolition of their homes, their release from jail or return from exile; these civilians were trucked into areas of Kurdistan far from their homes and dumped there by the army with only minimal governmental compensation or none at all for their destroyed property, or any provision for relief, housing, clothing or food, and forbidden to return to their villages of origin on pain of death. In these conditions, many died within a year of their forced displacement;
· destruction of the rural Kurdish economy and infrastructure.
Like Nazi Germany, the Iraqi regime concealed its actions in euphemisms. Where Nazi officials spoke of "executive measures," "special actions" and "resettlement in the east," Ba'athist bureaucrats spoke of "collective measures," "return to the national ranks" and "resettlement in the south." But beneath the euphemisms, Iraq's crimes against the Kurds amount to genocide, the "intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such."3
In 1970, the Ba'ath Party, anxious to secure its precarious hold on power, did offer the Kurds a considerable measure of self-rule, far greater than that allowed in neighboring Syria, Iran or Turkey. But the regime defined the Kurdistan Autonomous Region in such a way as deliberately to exclude the vast oil wealth that lies beneath the fringes of the Kurdish lands. The Autonomous Region, rejected by the Kurds and imposed unilaterally by Baghdad in 1974, comprised the three northern governorates of Erbil, Suleimaniyeh and Dohuk. Covering some 14,000square miles -- roughly the combined area of Massachusetts, Connecticut and Rhode Island -- this was only half the territory that the Kurds considered rightfully theirs. Even so, the Autonomous Region had real economic significance, since it accounted for fully half the agricultural output of a largely desert country that is sorely deficient in domestic food production.
In the wake of the autonomy decree, the Ba'ath Party embarked on the "Arabization" of the oil-producing areas of Kirkuk and Khanaqin and other parts of the north, evicting Kurdish farmers and replacing them with poor Arab tribesmen from the south. Northern Iraq did not remain at peace for long. In 1974, the long-simmering Kurdish revolt flared up once more under the leadership of the legendary fighter Mullah Mustafa Barzani, who was supported this time by the governments of Iran, Israel, and the United States. But the revolt collapsed precipitately in 1975, when Iraq and Iran concluded a border agreement and the Shah withdrew his support from Barzani's Kurdistan Democratic Party (KDP). After the KDP fled into Iran, tens of thousands of villagers from the Barzani tribe were forcibly removed from their homes and relocated to barren sites in the desert south of Iraq. Here, without any form of assistance, they had to rebuild their lives from scratch.
In the mid and late 1970s, the regime again moved against the Kurds, forcibly evacuating at least a quarter of a million people from Iraq's borders with Iran and Turkey, destroying their villages to create a cordon sanitaire along these sensitive frontiers. Most of the displaced Kurds were relocated into mujamma'at, crude new settlements located on the main highways in army-controlled areas of Iraqi Kurdistan. The word literally means "amalgamations" or "collectivities." In their propaganda, the Iraqis commonly refer to them as "modern villages"; in this report, they are generally described as "complexes." Until 1987, villagers relocated to the complexes were generally paid some nominal cash compensation, but were forbidden to move back to their homes.
After 1980, and the beginning of the eight-year Iran-Iraq War, many Iraqi garrisons in Kurdistan were abandoned or reduced in size, and their troops transferred to the front. In the vacuum that was left, the Kurdish peshmerga--"those who face death"--once more began to thrive. The KDP, now led by one of Barzani's sons, Mas'oud, had revived its alliance with Teheran, and in 1983 KDP units aided Iranian troops in their capture of the border town of Haj Omran. Retribution was swift: in a lightning operation against the complexes that housed the relocatedBarzanis, Iraqi troops abducted between five and eight thousand males aged twelve or over. None of them have ever been seen again, and it is believed that after being held prisoner for several months, they were all killed. In many respects, the 1983 Barzani operation foreshadowed the techniques that would be used on a much larger scale during the Anfal campaign. And the absence of any international outcry over this act of mass murder, despite Kurdish efforts to press the matter with the United Nations and Western governments, must have emboldened Baghdad to believe that it could get away with an even larger operation without any adverse reaction. In these calculations, the Ba'ath Party was correct.
Even more worrisome to Baghdad was the growing closeness between the Iranians and the KDP's major Kurdish rival, Jalal Talabani's Patriotic Union of Kurdistan (PUK). The Ba'ath regime had conducted more than a year of negotiations with the PUK between 1983-1985, but in the end these talks failed to bear fruit, and full-scale fighting resumed. In late 1986 Talabani's party concluded a formal political and military agreement with Teheran.
By this time the Iraqi regime's authority over the North had dwindled to control of the cities, towns, complexes and main highways. Elsewhere, the peshmerga forces could rely on a deep-rooted base of local support. Seeking refuge from the army, thousands of Kurdish draft-dodgers and deserters found new homes in the countryside. Villagers learned to live with a harsh economic blockade and stringent food rationing, punctuated by artillery shelling, aerial bombardment and punitive forays by the Army and the paramilitary jahsh. In response, the rural Kurds built air-raid shelters in front of their homes and spent much of their time in hiding in the caves and ravines that honeycomb the northern Iraqi countryside. For all the grimness of this existence, by 1987 the mountainous interior of Iraqi Kurdistan was effectively liberated territory. This the Ba'ath Party regarded as an intolerable situation.
A destruction process has an inherent pattern. There is only one way in which a scattered group can effectively be destroyed. Three steps are organic in the operation:
Concentration (or seizure)
This is the invariant structure of the basic process, for no group can be killed without a concentration or seizure of the victims, and no victims can be segregated before the perpetrator knows who belongs to the group.4
The Kurdish genocide of 1987-1989, with the Anfal campaign as its centerpiece, fits Hilberg's paradigm to perfection.
In June 1987, al-Majid issued two successive sets of standing orders that were to govern the conduct of the security forces through the Anfal campaign and beyond. These orders were based on the simple axiom on which the regime now operated: in the "prohibited" rural areas,all resident Kurds were coterminous with the peshmerga insurgents, and they would be dealt with accordingly.
The first of al-Majid's directives bans all human existence in the prohibited areas, to be applied through a shoot-to-kill policy. The second, numbered SF/4008, dated June 20, 1987, modifies and expands upon these orders. It constitutes a bald incitement to mass murder, spelled out in the most chilling detail. In clause 4, army commanders are ordered "to carry out random bombardments, using artillery, helicopters and aircraft, at all times of the day or night, in order to kill the largest number of persons present in these prohibited zones." In clause 5, al-Majid orders that, "All persons captured in those villages shall be detained and interrogated by the security services and those between the ages of 15 and 70 shall be executed after any useful information has been obtained from them, of which we should be duly notified."
Even as this legal and bureaucratic structure was being set in place, the Iraqi regime became the first in history to attack its own civilian population with chemical weapons. On April 15, 1987, Iraqi aircraft dropped poison gas on the KDP headquarters at Zewa Shkan, close to the Turkish border in Dohuk governorate, and the PUK headquarters in the twin villages of Sergalou and Bergalou, in the governorate of Suleimaniyeh. The following afternoon, they dropped chemicals on the undefended civilian villages of Sheikh Wasan and Balisan, killing well over a hundred people, most of them women and children. Scores of other victims of the attack were abducted from their hospital beds in the city of Erbil, where they had been taken for treatment of their burns and blindness. They have never been seen again. These incidents were the first of at least forty documented chemical attacks on Kurdish targets over the succeeding eighteen months. They were also the first sign of the regime's new readiness to kill large numbers of Kurdish women and children indiscriminately.
Within a week of the mid-April chemical weapons attacks, al-Majid's forces were ready to embark upon what he described as a three-stage program of village clearances or collectivization. The first ran from April 21 to May 20; the second from May 21 to June 20. More than 700 villages were burned and bulldozed, most of them along the main highways in government-controlled areas. The third phase of the operation, however, was suspended; with Iraqi forces still committed to the war front, the resources required for such a huge operation were notavailable. But the goals of the third stage would eventually be accomplished by Anfal.
In terms of defining the target group for destruction, no single administrative step was more important to the Iraqi regime than the national census of October 17, 1987. Now that the springtime village clearances had created a virtual buffer strip between the government and the peshmerga-controlled zones, the Ba'ath Party offered the inhabitants of the prohibited areas an ultimatum: either they could "return to the national ranks"--in other words, abandon their homes and livelihoods and accept compulsory relocation in a squalid camp under the eye of the security forces; or they could lose their Iraqi citizenship and be regarded as military deserters. The second option was tantamount to a death sentence, since the census legislation made those who refused to be counted subject to an August 1987 decree of the ruling Revolutionary Command Council, imposing the death penalty on deserters.
In the period leading up to the census, al-Majid refined the target group further. He ordered his intelligence officials to prepare detailed case-by-case dossiers of "saboteurs'" families who were still living in the government-controlled areas. When these dossiers were complete, countless women, children and elderly people were forcibly transferred to the rural areas to share the fate of their peshmerga relatives. This case-by-case, family-by-family sifting of the population was to become a characteristic feature of the decisions made during the Anfal period about who should live and who should die.
Last, but not without significance, the census gave those who registered only two alternatives when it came to declaring their nationality. One could either be Arab or Kurdish--a stipulation that was to have the direst consequences for other minority groups, such as the Yezidis, Assyrians and Chaldean Christians who continued to live in the Kurdish areas.5
The Iraqi authorities did nothing to hide the campaign from public view. On the contrary, as each phase of the operation triumphed, its successes were trumpeted with the same propaganda fanfare that attended the victorious battles in the Iran-Iraq War. Even today, Anfal is celebrated in the official Iraqi media. The fifth anniversary in 1993 of the fall of Sergalou and Bergalou on March 19, 1988 was the subject of banner headlines.
Iraqi troops tore through rural Kurdistan with the motion of a gigantic windshield wiper, sweeping first clockwise, then counterclockwise, through one after another of the "prohibited areas." The First Anfal, centered on the siege of the PUK headquarters, took more than three weeks. Subsequent phases of the campaign were generally shorter, with a brief pause between each as army units moved on to the next target. The Second Anfal, in the Qara Dagh region, lasted from March 22 to April 1, 1988; the Third, covering the hilly plain known as Germian, took from April 7 to April 20; the Fourth, in the valley of the Lesser Zab river, was the shortest of all, lasting only from May 3 to May 8.
Only in the Fifth Anfal, which began on May 15 in the mountainous region northeast of Erbil, did the troops have any real difficulty in meeting their objectives. Encountering fierce resistance in difficult terrain from the last of the PUK peshmerga, the regime called a temporary halt to the offensive on June 7. On orders from the Office of the Presidency (indicating the personal supervisory role that Saddam Hussein himself played in Anfal), the operation was renewed twice in July and August, with these actions denominated Anfal VI and Anfal VII. Eventually, on August 26, the last PUK-controlled area was declared "cleansed of saboteurs."
By this time, Iran had accepted Iraq's terms for a ceasefire to end the war, freeing up large numbers of Iraqi troops to carry the Anfal operation into the Badinan area of northern Iraqi Kurdistan. The Final Anfal began at first light on August 25, and was over in a matter of days. On September 6, 1988, the Iraqi regime made its de facto declaration of victory by announcing a general amnesty for all Kurds. (Ali Hassan al-Majid later told aides that he had opposed the amnesty, but had gone along with it as a loyal party man.)
Each stage of Anfal followed roughly the same pattern. It characteristically began with chemical attacks from the air on both civilian and peshmerga targets, accompanied by a military blitz against PUK or KDP military bases and fortified positions. The deadly cocktail of mustard and nerve gases was much more lethal against civilians than against the peshmerga, some of whom had acquired gas masks and other rudimentary defenses. In the village of Sayw Senan (Second Anfal), more than eighty civilians died; in Goktapa (Fourth Anfal), the death toll was more than 150; in Wara (Fifth Anfal) it was thirty-seven. In the largest chemical attack of all, the March 16 bombing of the Kurdish town of Halabja, between 3,200 and 5,000 residents died. As a city, Halabja was not technically part of Anfal--the raid was carried out in reprisal for its capture by peshmerga supported by Iranian Revolutionary Guards--but it was very much part of the Kurdish genocide.
After the initial assault, ground troops and jahsh enveloped the target area from all sides, destroying all human habitation in their path, looting household possessions and farm animals and setting fire to homes, before calling in demolition crews to finish the job. As the destruction proceeded, so did Hilberg's phase of the "concentration" or "seizure" of the target group. Convoys of army trucks stood by to transport the villagers to nearby holding centers and transit camps, while the jahsh combed the hillsides to track down anyone who had escaped. (Some members of the militia, an asset of dubious reliability to the regime, also saved thousands of lives by spiriting people away to safety or helping them across army lines.) Secret police combed the towns, cities and complexes to hunt down Anfal fugitives, and in several cases lured them out of hiding with false offers of amnesty and a "return to the national ranks"--a promise that now concealed a more sinister meaning.
In addition to this argument of principle, many features of Anfal far transcend the realm of counterinsurgency. These include, first of all, the simple facts of what happened after the military goals of the operation had been accomplished:
· the mass murder and disappearance of many tens of thousands of non-combatants--50,000 by the most conservative estimate, and possibly twice that number;
· the use of chemical weapons against non-combatants in dozens of locations, killing thousands and terrifying many more into abandoning their homes;
· the near-total destruction of family and community assets and infrastructure, including the entire agricultural mainstay of the rural Kurdish economy;
· the literal abandonment, in punishing conditions, of thousands of women, children and elderly people, resulting in the deaths of many hundreds. Those who survived did so largely due to the clandestine help of nearby Kurdish townspeople.
Second, there is the matter of how Anfal was organized as a bureaucratic enterprise. Viewed as a counterinsurgency, each episode of Anfal had a distinct beginning and an end, and its conduct was in the hands of the regular army and the jahsh militia. But these agencies were quickly phased out of the picture, and the captured civilians were transferred to an entirely separate bureaucracy for processing and final disposal. Separate institutions were involved--such as Amn, Istikhbarat, the Popular Army (a type of home guard) and the Ba'ath Party itself. And the infrastructure of prison camps and death convoys was physicallyremote from the combat theater, lying well outside the Kurdistan Autonomous Region. Tellingly, the killings were not in any sense concurrent with the counterinsurgency: the detainees were murdered several days or even weeks after the armed forces had secured their goals. Finally, there is the question of intent, which goes to the heart of the notion of genocide. Documentary materials captured from the Iraqi intelligence agencies demonstrate with great clarity that the mass killings, disappearances and forced relocations associated with Anfal and the other anti-Kurdish campaigns of 1987-1989 were planned in coherent fashion. While power over these campaigns was highly centralized, their success depended on the orchestration of the efforts of a large number of agencies and institutions at the local, regional and national level, from the Office of the Presidency of the Republic on down to the lowliest jahsh unit.
The official at the center of this great bureaucratic web, of course, was Ali Hassan al-Majid, and in him the question of intent is apparent on a second, extremely important level. A number of audiotapes were made of meetings between al-Majid and his aides from 1987 to 1989. These tapes were examined by four independent experts, to establish their authenticity and to confirm that the principal speaker was al-Majid. Al-Majid was known to have a distinctive, high-pitched voice and the regional accent of his Tikrit district origins; both these features were recognized without hesitation by those Iraqis consulted by Middle East Watch. As a public figure who frequently appears on radio and television in Iraq6, his voice is well known to many Iraqis. One Iraqi consulted on this subject pointed out that the principal speaker on the many hours of recordings in Middle East Watch's possession spoke with authority and used obscene language. In contrast, he said: "Others in those meetings were courteous and respectful with fearful tones, especially when they addressed al-Majid himself." Al-Majid, two experts noted, was often referred to by his familiar nickname, "Abu Hassan."
The tapes contain evidence of a bitter racial animus against the Kurds on the part of the man who, above any other, plotted their destruction. "Why should I let them live there like donkeys who don't know anything?" al-Majid asks in one meeting. "What did we ever getfrom them?" On another occasion, speaking in the same vein: "I said probably we will find some good ones among [the Kurds]...but we didn't, never." And elsewhere, "I will smash their heads. These kind of dogs, we will crush their heads." And again, "Take good care of them? No, I will bury them with bulldozers."
Loyalty to the regime offered no protection from al-Majid's campaigns. Nor did membership in the pro-government jahsh. Al-Majid even boasted of threatening militia leaders with chemical weapons if they refused to evacuate their villages. Ethnicity and physical location were all that mattered, and these factors became coterminous when the mass killings took place in 1988.
The 1987 village clearances were wholly directed at government-controlled areas, and thus had nothing whatever to do with counterinsurgency. If the former residents of these areas refused to accept government-assigned housing in a mujamma'a, and took refuge instead in a peshmerga-controlled area -- as many did -- they too were liable to be killed during Anfal. The same applied to other smaller minorities. In the October 1987 census, many Assyrian and Chaldean Christians -- an Aramaic-speaking people of ancient origin -- refused the government's demands that they designate themselves either as Arabs or Kurds. Those who declined to be Arabs were automatically treated as Kurds. And, during the Final Anfal in Dohuk governorate, where most Christians were concentrated, they were in fact dealt with by the regime even more severely than their Kurdish neighbors. Those few Turkomans, a Turkic-speaking minority, who fought with the Kurdish peshmerga were not spared, because they too were deemed to have become Kurds.
Almost continuously for the previous two decades, the Ba'ath-led government had engaged in a campaign of Arabization of Kurdish regions. The armed resistance this inspired was Kurdish in character and composition. In 1988, the rebels and all those deemed to be sympathizers were therefore treated as Kurds who had to be wiped out, once and for all. Whether they were combatants or not was immaterial; as far as the government was concerned they were all "bad Kurds", who had not come over to the side of the government.
At one level, this was a straightforward matter. Under the terms of al-Majid's June 1987 directives, death was the automatic penalty for any male of an age to bear arms who was found in an Anfal area.7 At the same time, no one was supposed to go before an Anfal firing squad without first having his or her case individually examined. There is a great deal of documentary evidence to support this view, beginning with a presidential order of October 15, 1987--two days before the census--that "the names of persons who are to be subjected to a general/blanket judgment must not be listed collectively. Rather, refer to them or treat them in your correspondence on an individual basis." The effects of this order are reflected in the lists that the Army and Amn compiled of Kurds arrested during Anfal, which note each person's name, sex, age, place of residence and place of capture.
The processing of the detainees took place in a network of camps and prisons. The first temporary holding centers were in operation, under the control of military intelligence as early as March 15, 1988; by about the end of that month, the mass disappearances had begun in earnest, peaking in mid-April and early May. Most of the detainees went to a place called Topzawa, a Popular Army camp on the outskirts of Kirkuk--the city where Ali Hassan al-Majid had his headquarters. Some went to the Popular Army barracks in Tikrit. Women and children were trucked on from Topzawa to a separate camp in the town of Dibs; between 6,000 and 8,000 elderly detainees were taken to the abandoned prison of Nugra Salman in the southern desert, where hundreds of them died of neglect, starvation and disease. Badinan prisoners from the Final Anfal went through a separate but parallel system, with most being detained in the huge army fort at Dohuk and the women and children being transferred later to a prison camp in Salamiyeh on the Tigris River close to Mosul.
The majority of the women, children and elderly people were released from the camps after the September 6 amnesty. But none of theAnfal men were released. Middle East Watch's presumption, based on the testimony of a number of survivors from the Third and bloodiest Anfal, is that they went in large groups before firing squads and were interred secretly outside the Kurdish areas. During the Final Anfal in Badinan, in at least two cases groups of men were executed on the spot after capture by military officers carrying out instructions from their commanders.
The locations of at least three mass gravesites have been pinpointed through the testimony of survivors. One is near the north bank of the Euphrates River, close to the town of Ramadi and adjacent to a complex housing Iranian Kurds forcibly displaced in the early stages of the Iran-Iraq War. Another is in the vicinity of the archaeological site of Al-Hadhar (Hatra), south of Mosul. A third is in the desert outside the town of Samawah. At least two other mass graves are believed to exist on Hamrin Mountain, one between Kirkuk and Tikrit and the other west of Tuz Khurmatu.8
While the camp system is evocative of one dimension of the Nazi genocide, the range of execution methods described by Kurdish survivors is uncannily reminiscent of another--the activities of the Einsatzkommandos, or mobile killing units, in the Nazi-occupied lands of Eastern Europe. Each of the standard operating techniques used by the Einsatzkommandos is documented in the Kurdish case. Some groups of prisoners were lined up, shot from the front and dragged into pre-dug mass graves; others were shoved roughly into trenches and machinegunned where they stood; others were made to lie down in pairs, sardine-style, next to mounds of fresh corpses, before being killed; others were tied together, made to stand on the lip of the pit, and shot in the back so that they would fall forward into it--a method that was presumably more efficient from the point of view of the killers. Bulldozers then pushed earth or sand loosely over the heaps of corpses. Some of the gravesites contained dozens of separate pits, and obviously contained the bodies of thousands of victims. Circumstantial evidence suggests that the executioners were uniformed members of the Ba'ath Party, or perhaps of Iraq's General Security Directorate (Amn).
By the most conservative estimates, 50,000 rural Kurds died during Anfal. While males from approximately fourteen to fifty wereroutinely killed en masse, a number of questions surround the selection criteria that were used to order the murder of younger children and entire families.
Many thousands of women and children perished, but subject to extreme regional variations, with most being residents of two distinct "clusters" that were affected by the Third and Fourth Anfals. Abuses by zealous local field commanders may explain why women and children were rounded up, rather than being allowed to slip away. But they cannot adequately explain the later patterns of disappearance, since the detainees were promptly transferred alive out of army custody, segregated from their husbands and fathers in processing centers elsewhere, and then killed in cold blood after a period in detention. The place of surrender, more than place of residence, seems to have been one consideration in deciding who lived and who died. Amn documents indicate that another factor may have been whether the troops encountered armed resistance in a given area--which indeed was the case in most, but not all, of the areas marked by the killing of women and children. A third criterion may have been the perceived "political stance" of detainees, although it is hard to see how this could have been applied to children.
Whatever the precise reasons, it is clear from captured Iraqi documents that the intelligence agencies scrutinized at least some cases individually, and even appealed to the highest authority if they were in doubt about the fate of a particular individual. This suggests that the annihilation process was governed, at least in principle, by rigid bureaucratic norms. But all the evidence suggests that the purpose of these norms was not to rule on a particular person's guilt or innocence of specific charges, but merely to establish whether an individual belonged to the target group that was to be "Anfalized," i.e. Kurds in areas outside government control. At the same time, survivor testimony repeatedly indicates that the rulebook was only adhered to casually in practice. The physical segregation of detainees from Anfal areas by age and sex, as well as the selection of those to be exterminated, was a crude affair, conducted without any meaningful prior process of interrogation or evaluation.
The regime had no intention of allowing the amnestied Kurds to exercise their full civil rights as Iraqi citizens. They were to be deprived of political rights and employment opportunities until Amn certified their loyalty to the regime. They were to sign written pledges that they would remain in the mujamma'at to which they had been assigned--on pain of death. They were to understand that the prohibited areas remained off limits and were often sown with landmines to discourage resettlement; directive SF/4008, and in particular clause 5, with its order to kill all adult males, would remain in force and would be carried out to the letter.
Arrests and executions continued, some of the latter even involving prisoners who were alive, in detention, at the time of the amnesty. Middle East Watch has documented three cases of mass executions in late 1988; in one of them, 180 people were put to death. Documents from one local branch of Amn list another eighty-seven executions in the first eight months of 1989, one of them a man accused of "teaching the Kurdish language in Latin script."
The few hundred Kurdish villages that had come through Anfal unscathed as a result of their pro-government sympathies had no guarantees of lasting survival, and dozens more were burned and bulldozed in late 1988 and 1989. Army engineers even destroyed the large Kurdish town of Qala Dizeh (population 70,000) and declared itsenvirons a "prohibited area," removing the last significant population center close to the Iranian border.
Killing, torture and scorched-earth policies continued, in other words, to be a matter of daily routine in Iraqi Kurdistan, as they always had been under the rule of the Ba'ath Arab Socialist Party. But the Kurdish problem, in al-Majid's words, had been solved; the "saboteurs" had been slaughtered. Since 1975, some 4,000 Kurdish villages had been destroyed; at least 50,000 rural Kurds had died in Anfal alone, and very possibly twice that number; half of Iraq's productive farmland had been laid waste. All told, the total number of Kurds killed over the decade since the Barzani men were taken from their homes is well into six figures.
By April 23, 1989, the Ba'ath Party felt that it had accomplished its goals, for on that date it revoked the special powers that had been granted to Ali Hassan al-Majid two years earlier. At a ceremony to greet his successor, the supreme commander of Anfal made it clear that "the exceptional situation is over."
To use the language of the Genocide Convention, the regime's aim had been to destroy the group (Iraqi Kurds) in part, and it had done so. Intent and act had been combined, resulting in the consummated crime of genocide. And with this, Ali Hassan al-Majid was free to move on to other tasks demanding his special talents--first as governor of occupied Kuwait and, then, in 1993, as Iraq's Minister of Defense.
1 A derisive Kurdish term for the National Defense Battalions, the word jahsh means "donkey foals."
2 "Saboteurs" is the term commonly applied by the Iraqi regime to the Kurdish peshmerga guerrillas and their civilian sympathizers.
3 As defined in the Convention on the Prevention and Punishment of the Crime of Genocide (hereinafter the Genocide Convention), 78 UNTS 277, approved by GA Res. 2670 on December 9, 1948, entered into force January 12, 1951.
4 Raul Hilberg, The Destruction of the European Jews (New York: Holmes and Meier, 1985 student edition), p.267.
5 While the Yezidis, a syncretic religious sect, are ethnic Kurds, the Assyrians and Chaldeans are a distinct ancient people in their own right.
6 Al-Majid has served variously over the past five years as Secretary General of the Ba'ath Party's Northern Bureau, Interior Minister, Governor of Iraqi-occupied Kuwait in 1990 and, as Defense Minister.
7 Rural Kurdish men carry personal weapons as matter of tradition, regardless of their politics.
8 Other mass graves have been found elsewhere in Kurdish-controlled territory, containing the remains of Amn executions before, during and after the Anfal period.