Medak investigation

United Nations - Security Council

S/1994/674/Add.2 (Vol. I)
28 December 1994

Final report of the United Nations Commission of Experts
established pursuant to
security council resolution 780 (1992)

Annex VII
Medak investigation

Under the Direction of:
William J. Fenrick
Member and Rapporteur on On-Site Investigations,
Commission of Experts Established Pursuant to
Security Council Resolution 780 (1992)

Principal Legal Analyst:
Major J. C. Holland, Canadian Armed Forces; Member of
Canada's Contributed Personnel to the Commission of Experts

Annex VII
Medak investigation

  1. Introduction
  2. Overview
  3. Description
  4. Analysis
  5. Recommendations

I. Introduction

       The Medak Pocket Operation comprises the military operations of the Republic of Croatia (Croat), United Nations Protection Force (UN or UNPROFOR) and, to a lesser extent, the «Republic of Serbian Krajina» (Serb) forces near Medak, Croatia, in September 1993.

       Sadly, it is only a typical example of how war is, and apparently always has been, waged in the Balkans.

       It is especially amenable to legal study for a variety of reasons. The operation was confined in both time and geography. The units involved were limited in number. UNPROFOR reported on the operation in a particularly detailed and helpful manner. UN forces anticipated law of war violations and gathered much relevant information during the operation. Lastly, it was a recent event so evidence and witnesses were still available.

II. Overview

       The operation took place just north of the town of Medak and just outside the United Nations Protected Area designated as Sector South. Medak is about 150 kilometres south-west of Zagreb.

       Before 9 September 1993, the Medak Pocket was a collection of small rural villages and hamlets forming a finger of Serb- controlled land jutting into Croat territory.

       On 9 September, at about 6:00 a.m., Croat forces attacked the Pocket. An artillery, mortar and/or tank fire barrage preceded an infantry and tank advance. Croats attacked from the north-east and quickly killed or routed the few Serb defenders. Overrunning the Serb defences, the Croat forces soon captured Divoselo (Strunici), Citluk (Licki), Donje Selo, and the surrounding villages. By 10 September, the Croatian army was in charge of the area.

       The rationale for the Croat attack is impossible to determine with certainty. Speculation includes: a rehearsal by the Croats for a larger operation, a test of their forces by the Croats, retaliation by the Croats for Serb shelling of Gospic, a desire by the Croats to straighten their front, or simply a Croat desire to seize territory.

III. Description

       Evidence of events during the Medak Pocket Operation emerges from various witnesses' experiences. The following is an extremely summarized version of their experiences.

       Captain "1" was in command of the Serb forces in the Divo Selo area. In the face of the Croat attack, he ordered his soldiers to withdraw. This they did along with many local civilians. Over the next days, this mixed group made its way by foot to Serb territory. During the escape, an unidentified Serb soldier escaping with the group told the Captain that Croat soldiers had spared him. Indeed, the Croats told the unidentified soldier to escape when they could easily have killed or captured him.

       "2" was a soldier in the Serb army on 9 September. Upon the Captain's orders, he withdrew in the face of the Croat attack. During his escape to Serb territory, he saw the Croats burn houses and steal livestock. He came across a body. Someone had either badly mutilated it after death or had tortured the victim before death. During his escape, he entered Citluk to try to find food. There he came within 10 to 15 metres of a group of Croat soldiers. He heard instructions translated into German for some soldiers. The apparent commander said in Croatian: «Those houses are Serbian houses and you can do anything you wish.»

       "3" was a Serb soldier on the front line at the beginning of the Croat attack. He retreated with his comrades but became separated from them. He came across a female body with an eye, an ear, and all right-hand fingers cut off. "3" saw Croat soldiers setting fire to houses and stealing sheep.

       Three other Serb soldiers all fled the Croat attack. They all report that the only Croat activity they saw was legitimate «soldier against soldier» combat.

       "4", a resident of "A", was in the local Serb militia. He fled the Croat attack and immediately returned to his home. There he saw his elderly sister-in-law dead. A search of the scene after UNPROFOR took control of the area revealed only some clothing. She was wearing this clothing when "4" saw her dead body. During "4's" escape to Serb territory, he saw Croat soldiers killing sheep belonging to local civilians and stealing five or six tractors owned by the local populace.

       "5", a resident of "A", was also a member of the Serb militia. On 9 September, he was on the front line. Upon the attack he fled, returned to his home, and warned his family to flee. While he was in hiding, before he reached safety, he saw his tractor being stolen by the Croats. Croat soldiers wounded him during his escape.

       Two persons, "6" and "7", witnessed the murder of an 83 year- old blind woman.

       "6" was a resident of "B". He left his home at the beginning of the Croat attack. Upon leaving, he saw the victim and a younger unidentified woman outside her home, from about 200 metres away. About 20 unidentified Croat soldiers came up to the victim's house and ordered the younger woman away. Then, the soldiers gunned down the victim. After this, "6" fled to the forest with relatives, where he eventually joined a mixed group of Serb military and civilians. They walked to safety.

       "7", a resident of "A", was hiding in the woods also near the victim's house. From the woods, she saw 10 unidentified Croat soldiers approach the victim, who was standing alone outside her home, and simply kill her.

       "8" was a resident of "B". During the early morning of 9 September, he tried to evacuate two wounded Serb soldiers in his private vehicle. Croat forces ambushed the vehicle, their gunfire hitting all three vehicle occupants. "8" believed the gunfire killed both of his passengers. "8" escaped and hid in bushes approximately 20 metres from the ambush site. Ten to 15 unidentified soldiers approached the vehicle, dragged the two dead Serb soldiers out, placed the bodies near a building and set the building on fire. In "8's"; original statement, he said one of his passengers was alive when taken from the vehicle by the Croats. He also said that they placed both the wounded soldier and the body of the other dead soldier in the building before setting it afire. The correction of the original statement was not placed in all versions of the various reports compiled by UN organizations.

       "9", a resident of "A", fled the initial attack with her family. When shortly thereafter she returned to her house to get shoes, she was shot and wounded by unidentified Croat soldiers. She heard the soldier's conversation from 300 to 500 metres away. The conversation showed that the Croats deliberately targeted her as a civilian. One soldier objected to shooting at her with, «No, it's a woman»; the other replied, «It does not matter».

       "10" resided in "A". During the 9 September attack, she was in her home. "10" son fled the house at the start of the attack. He took his rifle with him. No one has seen him since, and he is presumed to be dead. An unidentified Croat soldier saw her through a window then threw a grenade into her house. The subsequent explosion wounded her. The same soldier entered the house and fired into the room in which she lay. This gunfire did not hit her. Wounded but still in her house, "10" observed two Croat vehicles pull up disguised as UN vehicles (i.e. white with «UN» lettering). These vehicles carried Croat military forces. She heard instructions, translated between German and Croatian, to slaughter everything and leave nothing. She saw Croats killing her sheep and pigs. During her escape, she also saw Croats killing domestic animals, burning houses and stealing roof tiles. She eventually made her way to safety.

       "11" lived between "B" and "A". He and his family escaped to the woods at the beginning of the attack. He fled with his rifle. From hiding, he returned to his house on 10 September. There he found all his possessions destroyed, his animals mostly dead or injured and his house burning. During the several days it took "11" to reach safety, he was shot at several times and eventually wounded. A neighbour accompanied "11" during most of the time.

       "12", a resident of "B", also fled during the attack carrying his rifle. On 10 September, while still in the area, Croat soldiers discovered him and a friend. The Croats arrested both and ordered them to turn over their weapons. A Croat soldier then used his rifle to hit "12". "12" fled with the Croats trying unsuccessfully to shoot him. His friend apparently did not escape, not having been seen since. He spent the next several days hiding in various houses and in the forest in the area. During this time, he discovered Croats had taken the furniture and animals from his house. "12" observed the Croats stealing sheep, cows, and horses belonging to others. He was shot at by Croats twice and wounded in the leg on the second occasion.

       "13", a resident of "A", fled the 9 September Croat attack. During his escape, he saw civilian Croats stealing cattle, including his own cows and calves.

       "14" a resident of "A", escaped from her house on 9 September. During the several days she spent in the woods before she made her way to safety in Medak, she saw Croat soldiers burning houses and throwing grenades into houses. Other Croats fired upon her while she was walking to Medak.

       "15" lived in "C". She entered the Pocket on 26 September to look for her sheep. While searching, she came across the body of a dead female, whose fingers were cut off.

       Many witnesses report joining up with groups of fleeing Serb soldiers or mixed groups of fleeing Serb civilians and Serb soldiers.

       Many male civilian residents of the area carried or fled with their rifles.

       There are many witnesses available who, while not seeing any illegal activity by Croatian forces, can establish the general non-damaged nature of the area prior to the attack and the non- military use of most of the civilian houses.

       Within several days of their attack, Croatian authorities showed a willingness to withdraw to their 8 September positions. Serb artillery attacks on Karlovac and a Serb missile attack on Zagreb may have prompted this willingness.

       Negotiations took place, and the parties eventually agreed that the Croats would withdraw to their 8 September positions, and UN forces would occupy the territory vacated by the Croats. UN forces consisted of Canbat I, the Canadian battalion of the Princess Patricia's Canadian Light Infantry. It was augmented by two companies of infantry from two separate French battalions. The Croat withdrawal and insertion of UN forces started on 15 September and were originally to be completed by 6:00 p.m., 16 September.

       During the evening of 15 September, UN and Croat authorities held a meeting to iron out the implementing details of the original agreement.

       As the UN forces began to deploy into the Pocket on 16 September, they could hear tens of explosions and see new smoke rising from Croat-controlled territory. There were no Serb forces in those areas nor had there been for many days. Such explosions and smoke had not been seen before 15 September. They also heard small arms fire from the same area. There are many witnesses to this including nearly all Canbat I personnel, UNMOs, UNCIVPOLs, UN civilian personnel, UN and Canadian Forces public affairs personnel and news reporters. All suspected that the Croats were engaged in ethnic cleansing of the Pocket before turning it over to the UN.

       During the morning of 16 September, several more meetings took place between Croat and UN authorities. As a result, the time by which UN forces were to complete the takeover of territory evacuated by the Croats was delayed 24 hours until 17 September.

       At noon, 16 September, Croat forces prevented Canbat I soldiers from crossing into the Pocket. This was a violation of the agreement. UN public affairs video and radio, plus Reuters news agency, videotaped this delay. Additionally, there were numerous witnesses to this delay. UN personnel felt the delay was a deliberate tactic used by the Croats to give them more time to complete their ethnic cleansing of the Pocket. UNPROFOR pressed the Croats, and after a delay of about two hours, they allowed UN forces to enter the Pocket.

       The explosions and smoke from fresh fires continued to be evident from Croat-controlled parts of the Pocket throughout the Croat withdrawal. Again, there are many witnesses to this, including all UN personnel involved in the operation.

       As UN forces entered the Pocket, they found every building burning or demolished. There were hundreds of such buildings in the several villages and hamlets, none of which were habitable. Special sweep teams assessed and recorded damage, searched for survivors and collected bodies. The teams included UNPROFOR medical officers, UNCIVPOLs, and soldiers.

       On 16 September, in the Medak Pocket, "16" saw the crest of the Croat Ninth Mechanized Brigade (Ninth) on several Croat army trucks. He also saw the crest of the Croat 111th Home Defence (HD) Brigade (111th) on a Croat army truck.

       "17" noted Croat soldiers with the shoulder flashes of the Ninth Brigade. On 16 September, he saw Croat police of the Special Police in the Medak area. He witnessed the explosions and fires that preceded the Croat withdrawal. He also saw the complete destruction throughout the Medak Pocket.

       "18" heard the explosions and saw the fires within Croat- controlled territory on 16 September. He questioned a Croat liaison officer about these. The Croat officer replied that maybe the Croats were destroying their own homes in the area. "18" also observed that the Croat troops within the territory did not react to the explosions as they would to incoming artillery. He witnessed the total destruction and devastation throughout the area. He saw no house undemolished and took photographs of the damage.

       Also on 16 September, "19" saw about 15 Croat Special Police at Drjlei in the Pocket. These Special Police appeared fresh, probably having recently arrived in the area.

       During a 16 September meeting held at 10:00 a.m. an UNPROFOR officer specifically asked a Croatian officer to stop the explosions and fires in Croat-controlled territory. The Croatian officer clearly ordered to ensure his troops stopped those actions.

       Two more representations to the same effect were made to the Croats that day. The Croats said that Croat soldiers were firing into the houses, but that Croat forces were causing the explosions by detonating mines to make the area safe.

       On 17 September, after personally viewing the destruction in the Pocket, an UNPROFOR officer expressed his disappointment at this devastation to Croat officers. The Croats offered no explanation, but said they would again warn their troops.

       Also on 17 September, during a dispute between UN forces and the Croatian Army on the exact proper location of their respective forces, an UNPROFOR officer met a Croatian officer in the Pocket. The Croatian officer had a map and was plainly in charge of the Croat forces. The map had been obviously and crudely altered, and it no longer represented the earlier agreed upon placement of various forces.

       On 19 September, Croatian and UNPROFOR officers met and discussed the exact boundaries of the UN zone of responsibility in the Pocket.

       On that same date, an UNPROFOR officer met Croatian officers. During this meeting, a Croatian officer threatened to have his Croat forces fire on UN forces, if he did not receive the cooperation he wanted.

       On 22 September, a meeting was held to finalize the agreed upon positions of Croat and UN forces in the Medak Pocket. A Croatian officer signed a map depicting the agreed relative positions.

       The Medak Pocket area falls within the area of responsibility of the Croat OZ Gospic. Croat units involved were largely from OZ Gospic. Some Special Police Forces from OZ Split were seen during the operation, but their role seemed to be secondary.

       The nominal OZ Gospic Commander was a particular Croatian officer. However, during the relevant prelude to the attack and the attack itself, he was not in command due to his absence on leave and to illness. The actual OZ Commander was another officer.

       The major Croat unit involved in the attack was the Ninth Mechanized Brigade (Ninth), which is unofficially called the «Wolves». It had been designated the 6th Mechanized Brigade until about July 1993. The Ninth was directly responsible to OZ Gospic. Graffiti left on the walls of buildings in the Pocket by Croat forces included the «Ninth» and «Wolves».

       The 111th also participated in the attack. Miscellaneous other Croat army units formed a minor part of the attacking forces.

       The UN Military Information (MI) Branch (at UNPROFOR headquarters and elsewhere) compiles orders of battle (orbats) for all the protagonists in the former Yugoslavia.

       UNPROFOR units sweep teams recovered 18 bodies in the Pocket in the immediate aftermath of the operation. Croat authorities turned over another 64 bodies they said they recovered in the Medak Pocket. These bodies were all given to the Serb authorities. Of all the bodies recovered, 59 (71 per cent) were probably those of soldiers and 23 (29 per cent) those of civilians. Among other factors complicating the determination of military status is that many civilians wear items of military clothing and many local military wear items of civilian clothing.

       Serb authorities have identified most of the bodies. There are no reported witnesses to the deaths of any of the bodies identified.

       Medical officers examined many of the 18 bodies recovered by UNPROFOR. The preliminary field examinations and the circumstances in which the bodies were found revealed:

  1. Some suspicious circumstances, e.g., two badly burned bodies were found in a concrete chicken coop that could have been used as a jail, spent casings found near bodies, one body tied up, etc.;
  2. Some bodies had injuries that might have occurred before death, e.g., broken legs, a broken neck, a smashed face;
  3. Some evidence of either pre-death torture or post-death mutilation, e.g., missing ears, eyes or fingers; and
  4. A perhaps higher portion of head and close range wounds than might be expected.

       These examinations led a medical officer to place the times of death from 24 to over 96 hours before discovery of the bodies, with six having died after 14 September.

       Serbian authorities had a doctor conduct an examination of the bodies turned over to them by the Croats and UNPROFOR. Only one full autopsy was done. UNCIVPOL describes the rest of the examinations as «cursory». After this, Serb authorities quickly turned over the bodies to relatives for burial.

       The Serb authorities prepared a postmortem report and gave it to Major Holland. He passed it on to Dr. Robert Kirschner, an experienced pathologist with the independent group, Physicians for Human Rights. Dr. Kirschner's report states that regarding the bodies recovered by UNPROFOR, «there is insufficient evidence to document an execution style slaying». Of those bodies turned over to the Serbs by the Croats, «I could find no evidence to suggest a pattern of extra-judicial executions». His opinions regarding both groups cannot exclude some murders and admit some suspicious circumstances, but in summary the evidence is ambiguous.

       Dr. Kirschner's further verbal opinion was that the earlier preliminary field examinations and those conducted by a Serbian doctor are not reliable. This unreliability results from the necessarily rudimentary conditions of the field examinations, e.g., the bodies could not be washed, no x-ray equipment was available, etc. Additionally, the medical personnel involved understandably lacked forensic experience.

       The Croats claimed in an 11 October statement to the UN in Geneva that they did not violate the laws of war during the Medak battle. They specifically cited the cases of two elderly but apparently still feisty women killed during the attack. They said one was killed while operating anti-aircraft artillery and another blew herself up with a grenade to avoid capture.

       Most Canbat I personnel entering the Pocket witnessed the total destruction involved. Many buildings were still on fire on 16 September. There may be others with equal experiences. A Canbat I photographer took 1,400 photographs recording the destruction and the 18 recovered bodies. Much of this activity was video recorded by Canbat I personnel.

       Besides the destruction of buildings, all witnesses saw that most livestock was killed and most personal property, including vehicles and farm equipment, was destroyed. They noted that haystacks were set on fire, and wells were polluted. Croatian forces had discarded hundreds of surgical gloves throughout the area.

       Canbat I personnel think the bulk of the destruction in the Pocket was done on 16 September.

       The Canbat I reports state that firewood and other incendiary materials were seen being brought into the area by the Croats. Unfortunately, the report does not identify the specific witnesses to this.

       UN civilian employees and UNCIVPOLs also witnessed the same destruction. A UNCIVPOL team member made sketches of most buildings and detailed damage assessments of over 100 representative buildings. These assessments confirm the total devastation in the Pocket.

       The Canadian War Crimes Investigation Team (WCIT) visited the area from 27 to 31 October and on 10 November 1993. The WCIT consisted of Major Holland and Master Corporal T. McComb, both of the Canadian Forces. The team was accompanied by Dr. Kirschner. The team took video and still photographs, interviewed some witnesses and gathered further materials and reports. The team also obtained the 1,400 photographs taken by the Canbat I photographer and arranged and conducted the videotaped assessment by Major S. Laplante (a combat engineer in the Canadian Forces serving as UNPA Sector South Engineer) and Chief Warrant Officer Bastid (an explosive ordinance disposal expert in the French Army serving with UNPROFOR HQ Zagreb) of the damage to civilian buildings.

       The examinations of the buildings by Major Laplante and Chief Warrant Officer Bastid reveal that the buildings were either set on fire and/or demolished by charges set inside the buildings. Artillery did not cause the damage nor did tank fire, mortar shells, rocket propelled grenades, nor aerial bombardment. Canbat I personnel believed that antitank mines were used to demolish those buildings not burned. However, neither of the above witnesses could be that certain of the type of explosives used.

       Croat authorities say that the widespread destruction in the Medak Pocket was necessitated by the Serbs using the civilian homes for barracks and the storage of ammunition. What evidence there is of military use of the civilian accommodation is ambiguous or point to its military use by Croat forces. The type of garbage (Croat cigarettes, newspapers, etc.) and the direction the builders pointed the defensive positions (towards Serb controlled areas, etc.) support usage by Croat forces.

IV. Analysis

       The preceding reveals several potential broad categories of «serious violations of international humanitarian law» *1 or war crimes within the jurisdiction of the International Criminal Tribunal. These are:

  1. Murder («willful killing»--Article 2(a), «murder»--Article 5(a), unlawful targeting of civilians causing their death being a general unlisted violation of the laws of war--Article 3);
  2. Torture («torture or inhuman treatment»--Article 2(b), willfully causing great suffering or serious injury«--Article 2(c), »torture«--Article 5(f));
  3. Genocide by killing (Article 4(2)(a)) or by causing serious bodily . . . harm (Article 4(2)(b));
  4. Wanton destruction (»extensive destruction and appropriation of property, not justified by military necessity and carried out unlawfully and wantonly«--Article 2(d), »wanton destruction of . . . villages, or devastation not justified by military necessity«--Article 3(b)); and
  5. Plunder (»plunder of . . . private property«--Article 3(e)).

       The war crimes have been analysed in a pragmatic narrow legal fashion. The analysis will seek to decide who specifically, if anyone, can be held criminally responsible for any given crime. Dealing with a war crime, no matter how horrendous or how clear, is in this analysis a fruitless exercise if specific legal responsibility cannot be attributed.

       Such an analysis will deal with those directly responsible «who planned, instigated, ordered, committed or otherwise aided and abetted» the crime (Article 7(1)). Also dealt with will be indirect (command) responsibility, as referred to in Article 7(3), of a superior «if he knew or had reason to know that the subordinate was about to commit such acts or had done so and the superior failed to take the necessary and reasonable measures to prevent such acts or to punish the perpetrators thereof.»

       The analysis of direct criminal responsibility for any of the disclosed war crimes is very simple. At present, there is no evidence implicating any specific identifiable individual in the direct planning, instigation, ordering, commission, aiding or abetting of any of these crimes. Therefore, this analysis will concentrate on indirect, i.e., command responsibility.

       Proof of command responsibility for war crimes will often be largely circumstantial. Absent capturing or obtaining written orders by the suspects or having the testimony of witnesses to the issuing of these orders, such responsibility can usually only be inferred. Inferences of such orders of a sufficiently reliable nature to satisfy a criminal tribunal can only be derived from clear and convincing patterns. Under the first head of criminal command responsibility, the patterns must be so strong that the only reasonable inference is that the suspect ordered the commission of the crimes. Under the second head of criminal command responsibility, the pattern must be so strong that the only reasonable inference is that the suspect knew of the crimes and failed to control or punish his offending subordinates or that the suspect criminally abandoned his command.

       As mentioned, there is no evidence identifying any specific individuals with any crimes, including any of the murders. The only first-hand evidence of a murder is in the murder of the 83 year-old blind woman. However, even here, neither of the witnesses identified the responsible individuals beyond being «Croat soldiers». In any event, there are serious discrepancies between the two witnesses. "6" describes 20 soldiers, "7" describes 10. Furthermore, "6" describes a second woman with the victim, but "7" omits this.

       The reliability of these witnesses, particularly "6", is questionable. "6" was 200 metres away. "7" was hiding to save her life and almost certainly greatly upset. Their opportunity to reliably observe is moot.

       There is no convincing general pattern in the deaths occurring in the Pocket. It is suspicious that UN troops found virtually no survivors. It may also be suspicious that there were more head wounds and close-range injuries than might be expected. However, the majority (71%) of the located dead were military. Although not strictly comparable, this was a higher percentage of military casualties than the global figures in WWII, the Korean War, or the Vietnam War.

       Despite the importance of the presence of surgical gloves stressed in the Canbat I report and elsewhere, their presence is ambiguous. The sinister explanation is a possible one. However, surgical gloves may have simply been ordinary precautions by the Croats to deal with the legitimate dead and wounded anticipated in any attack. The photographs of Canbat I personnel show them also using surgical gloves.

       A prosecutor cannot use the postmortem examinations of the bodies, as they are unsatisfactory from several points of view. The initial examinations in the field were of necessity cursory, done without equipment and by doctors without forensic pathology qualifications.

       Further, the examinations done by the Serb authorities were unsatisfactory. They were extremely undetailed and were conducted in an unprofessional atmosphere, i.e., the supervising judge's drunkenness and corruptness.

       Thus, conclusions reached in the preceding examinations are unreliable. As earlier mentioned, an independent examination of what material remains (Dr. Kirschner's report) also fails to disclose any convincing pattern.

       Care must be taken in relying upon local witnesses. They tend to be unreliable due to their extremely emotional attachment to their own community's cause and their demonization of the enemy. Witness "9" has given several versions of her experiences. While her versions are not inconsistent, they do not cover the same material, some of which it is surprising not to hear in all versions. Her testimony should be confirmed and carefully considered before presentation.

       The irrationality of the parties to this conflict is shown by the absurd excuse given by the Croat authorities for the deaths of the two elderly women. They say one was killed while operating anti-aircraft artillery and another used a grenade to kill herself rather than face capture.

       Therefore, any reliable prosecution should be based on objective evidence, objective witnesses, or clear patterns. One or even two local witnesses may be insufficiently reliable to obtain a conviction.

       Yet, even using a stringent standard, it is obvious from the many witnesses' experiences that Croatian forces fired on fleeing civilians, killing some. However, this action cannot be proved to be illegal.

       Many civilians were fleeing in the company of retreating Serb soldiers, armed Serb civilians or were themselves armed. Thus, the actions of many Croat soldiers were possibly lawful and, at least, ambiguous. Much of the firing was at night. They may have believed they were firing on retreating non-surrendering enemy forces, which remain lawful targets. The law of war does not impose a standard of perfection on soldiers. It accepts that unfortunately civilians near military objectives may be incidentally killed without necessarily breaching the law of war. The inherent ambiguities surrounding this aspect of the Medak Pocket Operation make building a prosecution for unlawful targeting of civilians difficult. The evidence is not sufficiently reliable or clear to obtain a conviction.

       There is even a hint of contradictory evidence. The sparing of the Serb soldier, as related to "1" is some, albeit second hand, evidence that either there was no express policy to kill everyone, or at least any such policy was not strictly observed. Additionally, the witnesses naturally would not be aware of the possibly many times they were not fired upon by Croat forces.

       All this being said, some dead were probably murdered. However, no individual can be proved to be directly responsible. Further, there is no strong unambiguous pattern of willful killing emerging from the evidence available to this point. Without such a pattern, it is impossible to affix criminal responsibility upon the Croat commanders.

«Clearly, assignment to command military troops is accompanied by broad authority and heavy responsibility. This has been true of all armies throughout recorded history. It is absurd, however, to consider a commander a murderer or rapist because one of his soldiers commits a murder or a rape.» *2
What is required is something «extensive and widespread». No such pattern emerges from the evidence. Many commentators have criticized the General Yamashita prosecution, which was, at least, an aggressive use of the doctrine of command responsibility. Even at that, the prosecutors of General Yamashita could point to tens of thousands, if not hundreds of thousands, of victims with 286 witnesses and 423 exhibits.

       Unlike the deaths arising from the Medak Pocket Operation, there is a clear, obvious and overwhelming pattern of wanton destruction. Hundreds of homes were destroyed, virtually hundreds of other buildings were destroyed, most animals were killed or taken, virtually all personal property was destroyed or taken, all vehicles and farm equipment were destroyed or taken, haystacks were fired, and many wells were polluted. Devastation was total.

       The timing of the destruction is inconsistent with any legitimate military conduct or of military necessity. The bulk of this destruction occurred on 16 September, according to the many eyewitnesses. The sounds of the explosions, the rising of the smoke from fires, and the fact many buildings were still on fire as UN personnel entered the Pocket establish this.

       The destruction occurred well after all Serb resistance had ended. There was no Serb resistance in the areas from which the explosions were heard and the rising smoke seen. All effective Serb forces had fled the area at the time of the bulk of the destruction. The devastation was wrought in an unopposed withdrawal, not a contested advance or retreat. Even the Croat liaison officer was forced to use an excuse not involving legitimate combat activity by the opposing forces. The scattered nature of the buildings, with the brief and desultory original Serb defence, does not explain this level of destruction.

       The widespread destruction by demolition and fire is also generally inconsistent with legitimate military operations. If buildings are contested, then demolition charges cannot usually be placed inside them. Similarly, absent incendiary weapons, of which there is no evidence, being able to set fire to buildings implies an ability to approach and control them. That is inconsistent with enemy forces being present in them.

       Various contradictory excuses given by the Croats for the destruction suggest the lack of any legitimate excuse for such widespread destruction.

       The reports, if confirmed, of incendiary materials being brought into the Medak Pocket by the Croats is some evidence of planning. However, this evidence is ambiguous. Obviously lumber and even timber can be used to construct defensive positions as well as be used to burn buildings.

       The Canbat I report states that any antitank mines used in demolitions would have been centrally controlled. This indicates that their use to destroy civilian objects would have required high level approval. Such control is usual in a regular well-run military. However, there is evidence that this is not always the case with the forces involved in this conflict. Some minefields in Sector South have overlapped, indicating a lack of such central control. Additionally, all the forces display large measures of ill-discipline, lack of control, and disorganization.

       There are at least two defences to this charge offered or available to the responsible Croat authorities. A Croat press release said that the level of destruction in the Pocket was required by the Serbs' use of civilian buildings as barracks and for ammunition storage. This is clearly an inept excuse. The timing of the destruction, being after Serb resistance ended, argues against it. The extent of the destruction also argues against this excuse. If every destroyed building had contained Serb soldiers and ammunition, the Croats would never have captured the Pocket. Destruction by fire and demolition, as opposed to conventional military assault, also suggests this Croat excuse is invalid.

       Although not yet raised by the Croats, the second possible excuse is that this destruction was a legitimate use of a «scorched earth» policy. However, pointing out that this excuse was not what the Croats themselves have claimed demolishes it. They chose to rely on the use of these buildings for military purposes by the Serbs.

       Further, this area was not being turned over or being re- occupied by enemy forces. The UN was taking control of it to the exclusion of the Serb military.

       Lastly, as a party to the 1977 Protocols to the Geneva Conventions, Croatia is bound by Part IV, Section I of Protocol I. That treaty prohibits this tactic even if the Croats could otherwise claim they were entitled to lay waste to their own territory as a legitimate defensive tactic.

       It is not practical to prosecute anyone for any torture committed during the Medak Pocket Operation. First, it is not clear anyone was tortured. In each instance of suspected torture, it is possible the body was subjected to post-death mutilation and not pre-death torture. However, mutilation of bodies is itself a conventional war crime. It is not specifically covered in Article 3, but it is perhaps within its general ambit. The determination of this question turns on whether mutilation of corpses is a «serious violation of international humanitarian law» (Article 1).

       Second, as with the suspected murders, no specific individual can be identified as involved in any torture or mutilation. Again, assuming torture or mutilation could be proved, the few cases do not form any convincing pattern that could then be attributed to commanders.

       A particular Croatian officer was clearly in charge of the operation for the Croat forces. The evidence is clear and convincing that he was responsible.

       The thoroughness of the destruction, the open nature of the methods used (the explosions could be heard and the smoke seen by all), the orders overheard by civilian survivors, the inherent control a commander must or should have, and the express concerns raised by all overwhelmingly point to his responsibility. He, by his actions and by attending at various meetings, was clearly the responsible OZ Commander.

       Widespread use of roadblocks by all parties in the conflict, particularly near front lines, means that the responsible commanders would or should know about any large-scale movements of personnel or goods entering or leaving an area of operations. Therefore, the commanders either did know or should have known about demolition or incendiary materials entering the Pocket and plundered goods leaving.

       Can responsibility be found above that officer? Another officer was clearly in charge of the withdrawal of the Croat forces. However, he was only sent from the Zagreb general staff on about 15 September to oversee the withdrawal. The Croats saw the withdrawal as extremely sensitive, requiring a high level of supervision. His responsibility is therefore problematic due to his limited time in the area.

       There is no direct evidence of the commander of the Ninth's role in the wanton destruction. However, based solely on the orbat information, a prima facie case can be made. As the commander of the major unit involved, he is responsible for the destruction inflicted in the Pocket, which must have involved his unit. Graffiti in some of the destroyed buildings implicate the Ninth.

       The OZ Commander was replaced shortly after this battle. This, plus his reported Albanian ethnic background, suggests that the Croatian authorities are perhaps more likely to turn him over to the International Criminal Tribunal than another officer of Croat background and in more political favour.

V. Recommendations

       It is recommended that two Croatian officers be charged with the following war crimes, respectively:

  1. A grave breach of the Geneva Conventions of 12 August 1949, being the extensive destruction and appropriation of property protected under the provisions of the relevant Geneva Convention, such destruction and appropriation not being justified by military necessity and carried out unlawfully and wantonly contrary to Article 2(d);
  2. A violation of the laws or customs of war, being the wanton destruction of villages or devastation not justified by military necessity contrary to Article 3(b); and
  3. A violation of the laws or customs of war, being plunder of private property contrary to Article 3(e).
The charges involve the wanton destruction and plunder which occurred during the Medak Pocket Operation. While the case against one officer is stronger than that against the other, it is believed that there is also a prima facie case against the second officer.

       It is also recommended that no one be charged with any murder (murder, killing, unlawful targeting of civilians or genocide related deaths), torture or mutilation occurring during the operation. No one has been identified as being directly responsible for such crimes. There are some suspicious circumstances and even some probable murders. However, no convincing pattern can be proved such that a commander can be held indirectly criminally responsible using the doctrine of command responsibility.


       *1 From the Statute of the International Criminal Tribunal (Statute) Article 1. All further references to articles are to those in the Statute, unless stated otherwise. It is interesting that virtually all these provisions are echoed in the Yugoslav constitution.

       *2 «Trial of General Yamashita», 4 Law Reports of Trials of War Criminals 35.