The Kosovo Verification Mission
By Alex J. Bellamy

School of Political Science and International Studies
The University of Queensland, Australia

From Kosovo and International Society (Palgrave Macmillan, 2002), chapter 4, pages 114-120

The Racak massacre

On 15 January 1999 there was a major exchange of fire between MUP forces and the UCK around the village of Racak. The precise reason for the MUP's decision to attack the village on this particular day remains obscure because the Kosovar Albanian rebels had held it since October. At around 8 a.m. MUP forces entered Racak accompanied by a camera crew from Associated Press. This in itself was unusual. Given that even attacks directed against the UCK contravened the Holbrooke-Milosevic agreement and Yugoslavia's subsequent agreements with the OSCE and NATO it appears strange that the Serbs would have wanted international media coverage. Moreover, never before had the Serbs actively encouraged a foreign media presence to witness an attack on a village. Associated Press had been invited to film the fighting between the MUP and UCK and thus to direct attention away from the main business of the day in Racak: A reprisal against Kosovar Albanian civilians for the killing of three Serb policemen in Dulje on 8 January. The KVM was also invited to observe the exchange and verifiers stationed themselves on a hill overlooking the fighting, but with only a partial view of Racak.

Witnesses reported that the Serb operation began at 6.30 a.m. At this point it is widely believed that the population of Racak numbered around 3-400. T-55 tanks belonging to the VJ and MUP armoured vehicles were deployed around the perimeter of the village. Seeing these deployments, the first families began to flee to Petrovo and UCK members also withdrew from the village to positions in the hills around it. A female eyewitness told Human Rights Watch that around 40 MUP officers opened fire on the village from the hillside. The fire was aimed at the people fleeing below. Bullets hit the witness's husband and son. A 74-year-old man in the same group said he saw his grandson, granddaughter, and daughter injured at this time. Many villagers sought refuge from the fire in Sadik Osmani's house. At around 11 a.m. MUP officers took 30 men from the house. They were forced to lie on the ground in the courtyard. Two hours later, 23 of these men were led away by the officers. The people left in the house suspected that they were being taken into police custody in Stimlje. However, at 3 p.m. villagers heard shots coming from the direction where the bodies were later found.

Although KVM verifiers were able to see some of the events from their hillside location their view was only partial. Contradicting their apparent openness with regards to Associated Press the Serbs explicitly told the verifiers not to enter Racak. The verifiers first entered the village in the late afternoon and immediately heard unconfirmed reports that men and women had been separated and the men taken away by blue-uniformed MUP officers. They withdrew from the village as darkness fell in accordance with the KVM's standard operating procedures. At about 4 a.m. villagers discovered the bodies of 45 people scattered on a hillside, 25 together in a small gorge. They also discovered the bodies of nine uniformed UCK fighters. The verifiers returned early on the morning of 16 January accompanied by foreign journalists. The government had reported that 15 'terrorists' had been killed in the fighting but informal briefings given by verifiers to the press in Pristina on the night of the 15 January aroused a great deal of interest about what had happened in Racak. The KVM's spokesman, Berend Borchardt was amongst this first inspection team. He told the press that he believed the dead to be 'ordinary farming people' most of them wearing Wellington boots and none of them wearing the sort of footwear appropriate for military activity. Two British journalists who were also present at the scene were given access to the bodies and confirmed the verifiers reports that most of the victims had been shot at close range and that many had also suffered serious knife wounds.

William Walker arrived at Racak at around midday. He was first briefed by the verification team on the ground and then visited the site where the bodies were. He later recalled:

The dead were mostly elderly men, all in peasant work clothes, and had died where they were discovered. The evidence of deliberate execution was overwhelming. I so described the incident in a press conference that afternoon, blamed the security forces and asked the government to permit the entry of investigators from the ICTY. (75)

Walker was immediately declared a persona non grata by the Serbian authorities and ordered to leave Yugoslavia.

On the day of the initial attack on 15 January the Serbian Deputy Minister, Nikola Sainovic spoke on the telephone with the Interior Minister General Sretan Lukic. Unbeknown to them, the conversation was being tapped by the CIA and the subsequent tapes found their way to the Washington Post newspaper. Speaking from Belgrade Sainovic opened the conversation by enquiring about on-going operations around Racak. The Deputy Prime Minister was clearly concerned

About international news reports of a massacre there and the two engaged in lengthy and detailed discussions about what to do. Sainovic recommended two courses of action. First, the border with Macedonia should be sealed to prevent Louise Arbour, the chief prosecutor for the International Criminal Tribunal for Former Yugoslavia (ICTY), who was on her way to the region from entering Yugoslavia. Secondly, Lukic was instructed to order his forces to retake and hold Racak, which had fallen back into UCK hands on the night of the 15 January after the MUP had withdrawn. (76)

On 17 January the Serb security forces launched a renewed attack on Racak. Around 100 people were in the village at the time (including verifiers and journalists) and were forced to flee in the face of the attack. This meant that the bodies, now laid out in the village mosque, were left unattended. On 19 January, Louise Arbour was prevented from entering Kosovo.

Having retaken Racak, MUP officers took the victims bodies from the mosque to the Serbian Forensic Medicine Institute in Pristina. Two days later a combined Serb and Belarussian team of pathologists began an autopsy on many of the bodies. Two days after that, on 21 January, a Finnish team of pathologists headed by Helena Ranta arrived having been instructed by the EU to investigate and report on the causes of the deaths. According to the Serb and Belarussian pathologists the bodies showed no signs of execution and all had gunpowder traces on their hands indicating that they were indeed UCK guerrillas as Belgrade had claimed. Helena Ranta disputed these findings. Two months later in a detailed report Ranta pointed out that the Serbs reached these conclusions by using a paraffin test method. She argued that this method had been recognised as unreliable as long ago as 1968 and that using modern methods revealed there to be no evidence that any of victims had fired weapons prior to death. (77)

Ranta released her final report on 17 March. She concluded that there was no evidence that the victims were anything other than unarmed civilians and that many of them had been shot a close range, execution-style. Ranta went on to conclude that what happened at Racak on 15 January was in her mind without doubt 'a crime against humanity'. Seven key points emerged from this and other reports.

1. The dead were unarmed civilians who had not fired weapons.

2. The bodies had been moved before inspection but there were many reasons why this would have happened and nothing suggests that they had been moved by the UCK in order to create the appearance of a massacre.

3. KVM verifiers were correct to identity that many of the dead had been shot at very close range.

4. The Serbian pathologists concluded that there had been fabrication of evidence before they had conducted autopsies.

5. The fact that the bodies were left unsupervised between 17 and 21 January increases the possibility that they had been moved and tampered with by the Serbs.

6. The Serb and Belarussian pathologists used outdated methods and reached unsupported conclusions. (78)

7. The Serbs had actively obstructed the ICTY investigation, (79)

By piecing together reports by different organisations, many of which were frequently critical of NATO, it is possible to build a detailed picture of events in Racak. Nevertheless many commentators in the West continue to argue that the massacre was 'staged' by the UCK in an attempt to persuade NATO to intervene against the Serbs. Their key arguments are that, first, the Serb attackers were accompanied by the cameras of Associated Press and verifiers of the KVM, neither of which reported a massacre. Secondly, the television pictures show the village to be deserted, so any Kosovar Albanian casualties must have been UCK fighters. Thirdly, the verifiers were not shown the bodies until the next morning, giving the UCK time to gather the bodies of its fighters and arrange them to give the appearance of a massacre. Fourthly, the Finnish pathologists generally agreed with their Serb counterparts and Ranta steadfastly refused to describe what happened at Racak as a massacre. Finally, William Walker deliberately overreacted and refused to investigate the events properly because his remit was to find grounds to legitimise an armed NATO intervention. (80)

These arguments are wildly inaccurate. The Associated Press cameraman's pictures actually show very little. Indeed, witness reports suggest that the killings probably took place after he had left Racak. The fact that there were no villagers caught on camera was because those who remained were hiding in Sadik Osmani's house at the time. The claim that the Serb forces were accompanied by KVM verifiers is simply erroneous. Verifiers were expressly told not to enter the village and were only given a partial view by their vantage point. If most of the killings were committed in the general area where the bodies were discovered the KVM verifiers could have neither seen them or observe people being led out of the village in that direction. The verifiers did not see the bodies until next morning because they had not been discovered until the early hours of that morning. Most villagers believed that the victims had been taken to a police station but some did tell verifiers on the night of the 15 January that they heard gunshots from the direction of where the bodies were subsequently found. Much was made of the fact that Ranta did not describe what happened in Racak as a massacre. The Finnish report is a highly detailed and technical piece of work. The word 'massacre' is not a precise legal term, so Ranta instead referred to what happened as a 'crime against humanity', which is a crime that falls under the jurisdiction of the ICTY. The Finnish report departed substantially from the Serb and Belarussian pathology report. The final point concerning NATO's desire for a war ignores the fact that international consensus on using force had waned considerably since October.

Given the weight of evidence it is strange that the conspiracy theory has taken such a hold among Western commentators. The reason for this can be traced back to the discord between Walker and his deputy, Gabriel Keller. It was the French newspaper, Le Monde, which first exposed the so-called conspiracy. A day after the massacre, Keller briefed French journalists that 'there was something wrong' with the apparent massacre. This prompted journalists and academics to piece together the story of a staged massacre, a story that continues to hold credence in the West today. That story hinges on there being significant lapses of time during which the UCK would have been able to move bodies in order to stage a massacre and a belief that some NATO states were intent on unleashing air strikes. Thanks to numerous detailed reports we now know that no such time lapses occurred. Subsequent debates within the Alliance about how to respond to the massacre suggest that there was no consensus on using force.


Although the KVM remained in place until the eve of Operation Allied Force, the Racak massacre effectively sounded the death knell for the period of unarmed intervention. Because it was unarmed, the KVM relied on high levels of consent from both parties and progress towards a political settlement. Ultimately, the Hill plan failed to satisfy either side and there was very little international interest in using sticks and carrots to coerce acceptance.

Herein lay the major problem. Milosevic had only agreed to accept the KVM when threatened with NATO air strikes. Although the Activation Order that was issued in October remained in place it was clear that consensus within the Alliance on using force had diminished. There were two major reasons for this. Firstly, there was a dramatic reduction of violence. Both sides acknowledged that they were constrained by the presence of verifiers. This was greatly aided by the fact that the Serbs believed the UCK to be a vanquished force and because the mission was deployed in winter, the harshness of which precluded significant military operations. There was widespread evidence that both sides were preparing for further clashes in the spring. Nevertheless, the absence of burning villages made it very difficult to sustain the consensus on using force.

The second reason for the loss of consensus was the facade of progress on political dialogue. Maintaining political dialogue based on the Hill proposals remained the preferred policy in the West despite the 2 November deadline having passed almost unnoticed. What Western states shared with the Yugoslav president was a belief that Kosovo should remain part of Yugoslavia and that a political settlement be based on the creation of democratic local political institutions to manage conflict. There was therefore little desire to increase tensions by threatening to use force in a situation where violence initiated by the MUP and VJ had decreased substantially.

With Racak, however, the period of unarmed intervention came to an end. Its exposure in the media galvanised the Contact Group into action. After a two-week period of consultation, the Group adopted a new strategy based on coercive diplomacy and grand summitry.


74. Much of the following is derived from reports by the OSCE KVM, ‘Massacre of Civilians in Racak’ (, Helsinki Human Rights Watch (, International Committee of the Red Cross, Society for Threatened Peoples, Amnesty International ( and the Finnish EU pathologists report written by Helena Ranta, the head of the team. The pathology report can be seen in Weller, The Kosovo Crisis, p. 333. It is important to go into such detail because the incident at Racak has been the source of much debate in the West relating to the merits of Allied Force.

75. Walker, 'OSCE Verification Experiences in Kosovo', note 10, pp. 141-2.

76. Washington Post, 17 January 1999.

77. Report of the Independent Pathologists, in Weller, The Kosovo Crisis, p. 333.

78. It is interesting to note that the professional credibility of the Finnish pathologists has never been questioned by the Serbs.

79. These insights from various sources have been collated by the Society for Threatened Peoples.

80. See Johnstone, 'Humanitarian War', p. 163 and Edward Herman and David Peterson, 'CNN: Selling NATO's War Globally', in Philip Hammond and Edward S. Herman, Degraded Capability: The Media and the Kosovo Crisis (London: Pluto Press, 2000), pp. 117-19.

More Articles on Racak

Deniers of Serbia's War Crimes Index

Balkan Witness Home Page






Contact Balkan Witness

Report broken links