In the period of 16–18 March 1988 Halabja and its surroundings were attacked with chemical weapons (CW). According to Iranian figures, there were 12,500 casualties, including more than 5,500 fatalities. A Kurdish researcher later concluded that at least 3200 residents are known to have died. It is impossible to reconstruct exactly the events in and around Halabja. Nevertheless, it is widely accepted that the Iraqi military forces were solely responsible for the attack against a civilian target. Yet, shortly after the events sources in the United States began to hint that Iranian troops might have actually killed the majority of the civilians. The latter version assumes that the Iranians had not occupied Halabja and that the Kurdish inhabitants were killed by a chemical warfare agent that was never in the Iraqi arsenal. Most significantly, it suggests that Iran had achieved an advanced stage of assimilation of CW into its military doctrine.
According to the generally accepted account of events, Halabja was attacked with CW, which included mustard agent and sarin, beginning on 16 March 1988. Three days earlier Iran had launched a new offensive in the area and begun to infiltrate the town together with Kurdish Peshmerga. By the night of 15 March, they had all but captured it. However, as Iraqi public employees had been ordered to evacuate Halabja that day, there was widespread anticipation of reprisals by the Iraqi regime. The next morning Iraqi forces shelled the town with conventional munitions and launched sustained air strikes, dropping incendiary ordinance (possibly napalm or phosphorus). In the mid-afternoon chemical warfare agents were used in the northern part and the chemical attacks may have continued into the evening. Iranian soldiers wearing protective clothing and gas masks were seen in the streets of Halabja. Local people took refuge in cellars and other shelters against the bombing but did not appear to anticipate the CW attacks. A joint Netherlands–Belgian team of Médecins sans Frontières visited Halabja on 24–27 March 1988 as the first foreign medical organisation. It noted in its report ‘that people seemed to be killed by surprise during their daily activities (car driving, eating, water collecting, …)’. Narratives recorded by US doctors treating civilian victims from the Halabja area in New York also described how the chemical agents were delivered by planes while the families were still at home.
US claims about responsibility for the attack
On 23 March 1988, US State Department spokesman Charles Redman said that Iraq appeared to have used CW, adding that ‘there are indications that Iran may also have used chemical artillery shells in this fighting’. He did not elaborate. Later reports stated that Iraqi planes initially bombed the town with mustard agent. When the Kurdish civilians began to flee the Iranians, thinking that they were Iraqi troops, fired munitions filled with hydrogen cyanide (HCN). The US claims were said to be based on intercepts of battlefield communications and other highly classified, but unspecified intelligence sources. However, the original allegation of Iran’s use of a cyanide compound in Halabja may have been based on conjecture. On the same day of Redman’s statement, the Special Security Offices of the US Defense Intelligence Agency circulated an update of the Iran–Iraq war, which stated in paragraph 6:
‘Most of the casualties in Halabjah were reportedly caused by cyanogen chloride. This agent has never been used by Iraq, but Iran has shown interest in it. Mustard gas casualties in the town were probably caused by Iraqi weapons because Iran has never been noted using that agent.’
Cyanogen chloride is a chemical warfare agent of World War 1 vintage and may have indeed been explored by Iran in the early stages of its CW programme. Investigation is still a long step from weaponisation and deployment.
Some of the unspecified intelligence sources may have been Iraqi, which cannot be ruled out in the light of the increasing US tilt towards Iraq. The Iraqi ambassador in London, Dr Mohammed Al-Mashat, was quick to blame Iran for the attack, but his statement was not considered credible. On 29 March 1988 a Jordanian newspaper quoted an Iraqi spokesperson who claimed that Iran had used CW against Halabja and many other villages inside Iraq. Consequently, Iraq reserved the right to retaliate with CW against Iranian cities as a deterrent. On 4 April the Iraqi News Agency reported that Iranian artillery and aircraft had attacked Iraqi troops in the Halabja area with CW during 30–31 March.
Central to the US interpretation of events is the identification of the chemical warfare agent. Early descriptions of the consequences of the attack almost invariably pointed to the use of a cyanide-based compound. The clinical findings (blue cheeks; blue finger and toe nails, blisters) led the team of Médecins sans Frontières to conclude with a very high degree of probability that ‘cyanide’ and mustard agent had been used. Analysis of blood and urine samples indicated that the victims from Halabja being treated in the United States had been exposed to mustard agent, a nerve agent and a cyanide compound. During a tour of Halabja the Iranians also showed foreign journalists the remnants of what they claimed to be a 100-litre cyanide container dropped by an Iraqi plane.
At the time few people doubted that a cyanide-based compound had been used in Halabja, but several US sources noted that Iraq never employed HCN during the war and claimed that the compound was part of Iran’s array of chemical warfare agents. Iraq relied more on persistent agents because it was on the defensive, whereas Iran had developed rapidly dissipating agents of the chlorine and cyanide types in order not to hamper its advances. Iran was also said to have adapted plastic plants built under the Shah to help produce HCN. The inspections by the UN Special Commission on Iraq, which was created after the 1990–91 Gulf War, did not contradict this specific claim. In 1991 Iraqi officials categorically denied to UN Special Commission (UNSCOM) inspectors that Iraq had ever used or produced with the intent of using hydrogen cyanide or cyanogen chloride as a chemical weapon. A summary report of the UNSCOM findings submitted to the UN Security Council in January 1999 does not mention HCN or cyanogen chloride. However, while Iraq’s weaponisation of HCN cannot be clearly demonstrated, it did have the starting materials for HCN production as cyanide is used in the manufacture of the nerve agent tabun. It also possessed equipment for filling CW at low temperatures, which would have been required for munitions containing HCN.
The Halabja attack was not the first time that the possible use of HCN or another cyanide-based compound had been reported. During their second visit to Iran in February and March 1986, the UN experts recorded several Iranian claims regarding the possible use of HCN by Iraq. However, some Iranian victims treated for HCN intoxication at the front had died quickly. Other casualties no longer showed any symptoms upon arrival in an infirmary as these normally disappear fast. The clinical records of the patients in question revealed that they were administered medication in the belief that they had been exposed to HCN. In their May 1987 report the UN specialists again recorded the claim by an Iranian physician that he had successfully treated persons exposed to HCN with the inhalation of amyl nitrite and the injection of thiosulphate. They were unable to confirm the diagnosis, but noted that these patients had suffered from exposure to an acetylcholinesterase-inhibiting substance, presumably tabun. Shortly after the Halabja attack Iranian officials were quoted to believe that cyanide-based agents were responsible for many of the deaths. In 1989 an Iranian medical doctor estimated that approximately 500 Iranians had died from exposure to cyanide. The figure represents about 10 per cent of all Iranian fatalities from chemical warfare agents. In 1984 Israeli Science and Development Minister Yuval Ne’eman made an isolated accusation that Tehran was using Zyklon B to kill Iranian Jews, Druse and Baha’is and threatened to bomb Iran in retaliation.
Plausible alternative explanations
While there appear to have been cases of cyanide poisoning—especially among Iranian soldiers—during the Iran–Iraq war, the available literature does not allow for a strong linkage between the reported poisonings and the use of HCN. There are at least two plausible alternative explanations for the symptoms observed in Halabja and after earlier chemical attacks.
First, exposure to a high concentration of a nerve agent like sarin will produce outward signs similar to the clinical findings by the team of Médecins sans Frontières. Another mission to Halabja on 10–14 April 1988 could not trace any cyanides and analogues in six samples brought back to Belgium. This was attributed to the time delay between the attack and the visit. However, using the Morand–Laborit Method, acetylcholinesterase activity was found in three of the six samples, suggesting the presence of an organic phosphorus agent, such as tabun or sarin.
Second, hydrogen cyanide may be produced during the decomposition of the nerve agent tabun through hydrolysis or combustion. Although the UN experts investigating Iranian allegations of chemical warfare left open the possibility HCN bombs in their 1986 report, they noted that the explosion of a bomb containing tabun (which the Iraqis were known to have used since March 1984) may produce hydrogen cyanide. Exposure to the decomposition products of tabun may thus account for the several diagnoses of HCN poisoning made by Iranian doctors. The Iraqi nerve agents were very unstable and decomposed quickly.
Insights from the assimilation model
The US version of events in Halabja suggests that by March 1988 Iran had assimilated CW into its military doctrine. This implies that the Iranian military had such munitions in large quantities and that they stockpiled them with or near frontline units in order to be able to respond immediately to the exigencies on the battlefield. This in turn suggests that the political and military leadership had delegated the authority to use CW to battlefield commanders. Such military preparedness also requires a sophisticated logistical train to supply the artillery units with different kinds of conventional and chemical munitions (which includes elaborate safety precautions and training for handling the toxic shells). There is hardly any evidence in the public record to suggest that the Iranian forces had been able to assimilate CW into their military doctrine in any significant way. Even the Iraqi allegations and presentation of low casualty figures from chemical warfare point to sporadic Iranian use of CW.
Assimilation of CW into military doctrine also presumes knowledge of the relative utility of the warfare agents under various conditions of combat. HCN is a first-generation chemical warfare agent, and the establishment or conversion of production lines to make HCN was probably not beyond the capabilities of a relatively advanced country like Iran. Yet, while cyanide-based warfare agents have been part of the arsenals of several powers, they have in general not been very successful. The inferior combat properties of HCN are widely reported in the traditional literature on chemical warfare agents. Moreover, the climatic conditions along most of the frontline did not favour the standardisation of a highly volatile agent. If CW were indeed highly assimilated in the Iranian military doctrine, these limitations would have been observed. Even assuming favourable meteorological conditions in the Halabja area, batteries of multiple-rocket launchers rather than standard artillery would have been required to achieve critical doses of HCN in order to kill thousands of people while fleeing.
Finally, the ability to wage chemical warfare effectively on a militarily significant scale is the outcome of a multifaceted process, which includes research and development, production, testing and training, organisational adaptation, and so on. The Libyan and Iraqi CW programmes—two recent cases of CW proliferation that have been described in considerable detail—took several years to establish and Iraq only managed to successfully integrate CW in offensive operations early in 1988, more than four years after the first large-scale chemical attacks. War may accelerate the CW armament dynamic, but as several Allied powers experienced in World War 1 the process remains complicated, time-consuming and fraught with danger. Until the Armistice in 1918 they invariably lagged the Germans as regards new warfare agents and operational doctrine governing their use. Moreover, they learned many lessons directly on the battlefield, leading to changes in equipment and operational procedures. In 1984, Iran was not believed to have a chemical warfare capability. In other words, if Iran is responsible for Halabja it must be presumed to have set up a sophisticated CW capability (weapons, doctrine, policies, etc.) within a matter of four years. The available data on Iran’s CW acquisition efforts do not support such a conclusion. A review of Iraq’s few allegations of Iranian chemical warfare and low number of officially reported CW casualties shows that Iran definitely did not acquire its offensive chemical warfare expertise on the battlefield.
Contradicted by Iraqi intelligence documents
The explanation that holds Iran responsible for the chemical attack on Halabja suffers from several internal inconsistencies and is difficult to support in the absence of concrete public evidence. In 1995 Human Rights Watch/Middle East published the contents of some documents the Kurds had retrieved from Iraqi government agency buildings during their uprising after the 1990–91 Gulf War. A telex dated 11 April 1988 from the Iraqi Military Intelligence Directorate in Suleymaniyeh referred to a video on sale showing ‘the Iraqi chemical attack on Halabja’. A letter from the same agency dated 27 March 1988 describes the fighting around Halabja and refers euphemistically to the chemical bombing and shelling that produced the high number of casualties.
- Preface (Background to the original research note)
- Iran’s CW defensive preparations
- Iran’s CW offensive preparations
- CW use against Halabja
- Conclusion: Was Iran responsible for the CW atrocity in Halabja?
CD/PV.450, 22 March 1988.
Human Rights Watch, Iraq’s Crime of Genocide: The Anfal Campaign Against the Kurds (Yale University Press: New Haven, 1995), p. 72. The figure of 3200 fatalities was reached by the Kurdish researcher Shorsh Resool, who had assembled a list of names of people who died in the attack.
There exist also some accounts in which the Iranian troops passed through Halabja and nonetheless attacked the town with CW reportedly with the intention of blaming Iraq. See, for example, Henderson, S., Instant Empire: Saddam Hussein’s Ambition for Iraq (Mercury House: San Francisco, 1991), p. 114. The author stated that this version is compatible with that of the United States but based his claim on an unidentified diplomat who investigated the incident soon after it occurred.
For a detailed and documented account, see Human Rights Watch, 1995, pp. 68–72.
Artsen zonder Grenzen, MSF Holland, ‘Assessment of the by war affected population’, Report mission Kurdistan, 24–27 Mar. 1988, p. 4.
Statement of Dr Deborah Lief-Dienstag, 9 February 1989 in Global Spread of Chemical and Biological Weapons, Hearings before the Committee on Governmental Affairs and its Permanent Subcommittee on Investigations, US Senate (US Government Printing Office: Washington, DC, 1990), pp. 46–47 and 249.
Reuters, ‘Iran chemical war threat’, Guardian, 24 March 1988, p. 24. This statement was also published in the Washington Times of 23 March, which is the earliest public source quoted in later US analyses. Some further details about the US allegations were provided in Ottaway, D. B., ‘Chemicals and missiles alter Middle East warfare’, International Herald Tribune, 6 April 1988, p. 1.
A detailed overview of events was presented in a paper prepared by Anthony H. Cordesman entitled ‘The Iran–Iraq War: gas warfare and the prospects for the use of nuclear weapons’ inserted in the statement by Senator John McCain (Republican, Arizona), Congressional Record–Senate, 30 Sep. 1988, pp. 13804–07. It erroneously stated that the Halabja attack took place on 26 February 1988. This sequence of events was also included in Cordesman, A. H. and Wagner, A. R., The Lessons of Modern War, Volume II: The Iran–Iraq War (Westview Press: Boulder, Colorado, 1990), p. 517. The details (including the wrong date) were incorporated in the unpublished study by White, D., Characterization and Historical Review of Chemical/Biological Weapons: Mid-Term Review (Science Applications International Corporation: San Diego, 1992). Some later analysts treated this document as an original source. Another often quoted publication is Pelletiere, S. C., Johnson II, D. V. and Rosenberger, L. R., Iraqi Power and U.S. Security in the Middle East (Strategic Studies Institute, U.S. Army War College: Carlisle Barracks, Pennsylvania, 1990), p. 52. The authors did not elaborate on their claim, but it was reportedly based on a operational history of the final stages of the Iran–Iraq war by the Pentagon.
United States Defense Intelligence Agency, Special Security Offices, ‘Iran-Iraq: war update’, 23 March 1988, Envelope PTTSZYUW RUEKJCS2867 0850428-SSS–RUEALGX. (Text as in original.)
Muir, J., ‘Iraqi gas attacks revive horrors of the Great War’, Sunday Times, 27 March 1988, p. B3.
Al-Ra’y (Amman), 29 March 1988, p. 1, as translated from Arabic in Foreign Broadcast Information Service, Daily Report–Near East and South Asia (FBIS-NES), FBIS-NES-88-060, 29 March 1988, p. 34 (as summarized in the Harvard–Sussex CBW Events Database).
INA (Baghdad), 4 April 1988, as translated from Arabic in Foreign Broadcast Information Service, Daily Report–Near East and South Asia (FBIS-NES), FBIS-NES-88-065, 5 April 1988 (as summarized in Chemical Weapons Convention Bulletin, no. 1 (summer 1988), p. 11). The allegation formed the basis of the Iraqi request to the UN Secretary-General to have an investigation. See ‘Report of the Mission …’, 25 April 1988, p. 2.
Artsen zonder Grenzen, MSF Holland, ‘Assessment of the by war affected population’, Report mission Kurdistan, 24–27 Mar. 1988. The report did not include the laboratory results of the toxicological tests on samples. It should be noted, however, that following exposure to HCN, the victims would have been expected to have a reddish rather than a blue colouring.
Heyndrickx, A., ‘Clinical toxicological reports and conclusions of the biological samples of men, sent to the Department of Toxicology at the State University of Ghent, for toxicological investigation’, Report no. 88/NY/PJ881, Ghent (Belgium), 6 April 1988, 6p., in Global Spread of Chemical and Biological Weapons, Hearings before the Committee on Governmental Affairs and its Permanent Subcommittee on Investigations, US Senate (US Government Printing Office: Washington, DC, 1990), pp. 254–59.
The detection of rather high levels of cyanide compounds in these victims is surprising. The analysis of cyanide in biological fluids is difficult and almost never available during the treatment phase (in part because death will occur in 5–8 minutes at high dosages on the battlefield and in 15–30 minutes at lower dosages). Detoxification also occurs within the body.
Aubin Heyndrickx led another mission to Halabja on 10–14 April 1988, but he could not trace any cyanides and analogues in six samples brought back to Belgium. He attributed this to the time delay between the attack and the visit, but nonetheless concluded ‘with no doubt’ that cyanogen or derivatives had been used. Heyndrickx, A., Clinical and toxicological reports and conclusion of the biological samples of men and of the environmental samples, brought to the Department of Toxicology at the State University of Ghent, for toxicological investigation, Report no. 88/KU2/PJ881, Ghent (Belgium), 27 April 1988, p. 5.
Hirst, D., ‘Iran puts dead on show after gas raid’, Guardian, 22 March 1988.
Tyler, P. E., ‘Iran, too, faulted in gas attack on Kurds’, International Herald Tribune, 4 May 1990, pp. 1 and 4. See also the debate about the claim in Pelletiere, et al., 1990 , as reported in ‘8 November’, Chemical Weapons Convention Bulletin, no. 11 (March 1991), p. 5. Nevertheless, according to a statement in the Congressional Record of the US Senate of September 1988 Iraq produced hydrogen cyanide or cyanogen chloride. ‘The Iran–Iraq War …’, 30 September 1988, p. 13805. Press reports in 1985 noted that Western intelligence sources then believed that Iraq was producing ‘cyanide gases’ north of Baghdad. See Beecher, W., ‘US aides say Iran ready to use toxic gas’, Boston Globe, 19 April 1985, p. 1.
Ottaway, D. B., ‘Chemicals and missiles alter Middle East warfare’, International Herald Tribune, 6 April 1988, p. 2.
‘The Iran–Iraq War: gas warfare and the prospects for the use of nuclear weapons’, note , p. 13806.
Gee, J., ‘Iraqi declarations on chemical weapons: How much did they really have, and what was it?’, paper prepared for an informal meeting on Iraq’s chemical arsenal, Stockholm, 8 June 1992.
Letter dated 27 January 1999 from the Permanent Representatives of the Netherlands and Slovenia to the United Nations addressed to the President of the Security Council, UN document S/1999/94, 29 January 1999, appendix 2, ‘Status of the verification of Iraq’s chemical weapons programme’.
Private communication by a former Swedish member of UNSCOM with the author, 17 June 2000.
There are many press reports referring to anonymous US and Western intelligence officials prior to the Halabja attack that mention Iraqi possession or use of HCN. One article quoted a ‘high-ranking Arab officer familiar with Iraq’s chemical warfare program’ who said that ‘Iraq also appeared to be using hydrogen cyanide, either by its own manufacture or supplied by the Soviets’. The article further identified an Iraqi HCN production plant in Ramadi, about 60 miles west of Baghdad. Segal, D., ‘Poison gas in the Persian Gulf’, Combat Weapons, vol. 2, no. 1 (Winter 1986), pp. 16 and 84..
‘Report of the Mission Dispatched by the Secretary-General to Investigate Allegations of the Use of Chemical Weapons in the Conflict Between the Islamic Republic of Iran and Iraq’, UN Security Council document S/17911, 12 March 1986, p. 12.
‘Report of the Mission Dispatched by the Secretary-General to Investigate Allegations of the Use of Chemical Weapons in the Conflict Between the Islamic Republic of Iran and Iraq, Addendum, Appendix III: Report on Patients Examined by Dr Manuel Dominguez with Relevant Clinical Data’, UN Security Council document S/17911/Add.1, 14 March 1986, pp. 11 and 13. The patients were administered sodium thiosulphate intravenously and amyl nitrate together with intravenous sodium thiosulphate respectively.
‘Report of the Mission …’, 8 May 1987, p. 10.
Muir, J., ‘Iraqi gas attacks revive horrors of the Great War’, Sunday Times, 27 March 1988, p. B3.
Foroutan, A., ‘A review of 5 years of chemical attacks & the role of Iranian medical staff in management of victims’, in International Commission of Health Professionals, International Conference on Combatting the Use of Chemical and Biological Weapons, Conference proceedings, Geneva, 1989, p. 205.
‘Ne’eman Hints Action Possible on Iranian Gas Use’, Jerusalem Post, 9 May 1984, p. 2. Zyklon B was the code name for the hydrogen cyanide used in the Nazi extermination camps during World War 2.
Heyndrickx, A., Clinical and toxicological reports and conclusion of the biological samples of men and of the environmental samples, brought to the Department of Toxicology at the State University of Ghent, for toxicological investigation, Report no. 88/KU2/PJ881, 27 Apr. 1988. While Aubin Heyndrickx is very controversial both in Belgium and internationally, this report gave the first indications based on samples taken in Halabja that a nerve agent may have been used in the attack.
Ellison, D. H., Handbook of Chemical and Biological Warfare Agents (CRC Press: Boca Raton, Florida, 2000), p. 222. Stöhr, R. (ed.), Chemische Kampfstoffe und Schutz vor chemischen Kampfstoffen [Chemical warfare agents and protection against chemical warfare agents] (Militärverlag der Deutschen Demokratischen Republik: Berlin, 1985), p. 121.
‘Report of the Mission Dispatched by the Secretary-General to Investigate Allegations of the Use of Chemical Weapons in the Conflict Between the Islamic Republic of Iran and Iraq’, UN Security Council document S/17911, 12 March 1986, p. 13.
See, for instance, van Zelm, M., ‘A detailed inspection of Muthanna (UNSCOM17/CW5)’, paper prepared for an informal meeting on Iraq’s chemical arsenal, Stockholm, 8 June 1992.
Its physical properties and composition were determined by the Swedish chemist Carl Scheele in 1782. Smart, J. K., ‘History of chemical and biological warfare: an American perspective’, in Sidell, F. R., Takafuji, E. T. and Franz, D. R. (eds.), Medical Aspects of Chemical and Biological Warfare (Office of the Surgeon General, Department of the Army: Washington, DC, 1997), p. 10. In World War 1 the French were the only belligerent to have favoured it for a prolonged period. Cyanogen chloride is another cyanide-based agent the French tried in World War 1. It is heavier and less volatile than HCN, and although its toxicity is similar to that of HCN, it is more effective at low concentrations. In contrast to HCN, its effects are cumulative on its victims and it also has a delayed toxic effect. Cyanogen bromide was introduced by the Austrians in 1916. Baskin, S. I. and Brewer, T. G., ‘Cyanide poisoning’, in Sidell, F. R., Takafuji, E. T. and Franz, D. R. (eds.), Medical Aspects of Chemical and Biological Warfare (Office of the Surgeon General, Department of the Army: Washington, DC, 1997), p. 273.
Jacobs, M. B., War Gases (Interscience Publishers: New York, 1942), pp. 36–37. Baskin, S. I. and Brewer, T. G., ‘Cyanide poisoning’, in Sidell, F. R., Takafuji, E. T. and Franz, D. R. (eds.), Medical Aspects of Chemical and Biological Warfare (Office of the Surgeon General, Department of the Army: Washington, DC, 1997), pp. 272–73.
Meyer, J., Der Gaskampf und die chemischen Kampfstoffe [Chemical warfare and chemical warfare agents] (Verlag von S. Hirzel: Leipzig, 1925), p. 370, noted that the toxicity of HCN decreases rapidly when mixed with air. In World War 1 the French hardly ever obtained lethal or injurious concentrations so that the Germans recorded very few HCN casualties. Prentiss, A. M., Chemicals in War (McGraw-Hill Book Company: New York, 1937), p. 176, stated that systemic poisons such as HCN are the most virulent toxicants known. However, they are practically non-effective until a critical concentration is reached. Therefore, in order to achieve a lethal dose before the concentration falls below its critical effective strength it is necessary to establish the concentration with a high density in a minimum of time and under favourable meteorological conditions. He concluded that systemic poisons are uneconomic and that the results on the battlefield are uncertain.
Human Rights Watch, 1995, pp. 241–42.