The battle of Dubrovnik

United Nations - Security Council

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

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

Annex XI.A
The battle of Dubrovnik and
the law of armed conflict

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)

Prepared by:
Lieutenant-Colonel Dominic McAlea, Canadian Armed Forces;
Member of Canada's Contributed Personnel to the Commission of Experts
Colin Kaiser, Consultant to the Commission of Experts
Major Terje Lund, Norwegian Armed Forces; Member
of Norway's Contributed Personnel to the Commission of Experts
Major Oyvind Hoel, Norwegian Armed Forces; Member
of Norway's Contributed Personnel to the Commission of Experts

Annex IX.A
The battle of Dubrovnik

  1. Introduction
    1. Mandate
    2. Methodology
    3. Sources and analysis of information
  2. Cultural-historical perspective of Dubrovnik
    1. Background
    2. The Old Town of Dubrovnik
  3. The political situation
  4. The military situation
    1. Sources
    2. The Croatian Army order of battle
    3. The «Yugoslavian» order of battle
  5. The occupation of the district of Dubrovnik
    1. Sources
    2. Konavle - the south
    3. Primorje - the north
  6. The «siege» of Dubrovnik: october-december 1991
    1. Sources
    2. Situation of the city area
    3. Bombardments during October and November - military positions and weapons
    4. A month of negotiations
    5. The St. Nicholas day bombardment - 6 December 1991
    6. The inquiry of the JNA
  7. Civilian casualties
    1. Sources of information
    2. Analysis
    3. The dead and wounded
  8. Damage to civilian property
    1. Sources
    2. Categorization of damage
    3. Financial evaluation
  9. Damage to cultural property
    1. Sources - Old Town and New Town
    2. Sources - the district of Dubrovnik
    3. Damage to the Old Town of Dubrovnik during the bombardments of October-December 1991
  10. Applicable law
    1. The conflict
    2. The International Criminal Tribunal for the Former Yugoslavia
    3. Offences
    4. Personal and command responsibility
    5. Protection of victims and witnesses
    6. Military objectives and other customary international law principles
  11. Responsibility for war crimes
  12. Recommendations for further investigation
    1. Criteria
    2. Crimes against persons
      1. Wilful killing
      2. Unlawful confinement or abuse of civilians
      3. Arrest of protected persons
    3. Crimes against civilian property
    4. Crimes against religious, historical and cultural property
  13. Lessons learned
    1. Interpreters
    2. Pre-deployment preparation
    3. On-site secretarial support
    4. Co-ordination with UNPROFOR and ECMM
    5. Team members in uniform
    6. Team composition

I. Introduction

A. Mandate

       In the fall of 1991, the region surrounding the town of Dubrovnik on the southern Dalmatian coast was subject to siege and occupation by what was then known as the Yugoslavian National Army (JNA). Within a matter of weeks, the City of Dubrovnik itself was surrounded on both land and sea by besieging forces of the Former Republic of Yugoslavia (FRY). This occupation continued well into the fall of 1992, when the JNA withdrew from the hinterland or District of Dubrovnik in what had come to be recognized as the Republic of Croatia. This military action received wide international attention because of the historical and cultural significance of an area in the City of Dubrovnik known as the Old Town which sustained considerable damage during these attacks.

       The importance in terms of cultural heritage of the Old Town of Dubrovnik has been recognized by UNESCO through its inclusion on the World Heritage List in 1979. For this reason, the Battle of Dubrovnik is significant for law of armed conflict since the military actions of the JNA, especially in the early stages of the battle, appear to have caused inordinately extensive damages not only to civilian but also to the cultural, historical and religious property in proportion to what would reasonably have been expected given the number, location and nature of valid military objectives within the District of Dubrovnik.

       It was with these circumstances in mind that the Rapporteur for On-site Investigations, Mr. William Fenrick, visited Dubrovnik in May 1993. As a result of his recommendation, the Commission sent a team to Dubrovnik to study the Battle of Dubrovnik. Accordingly, the Commission approved the study and a plan of action was provided by the Commission. The plan of action called for the deployment, on behalf of the Commission, of a small team of law of armed conflict experts (Lieutenant Colonel Dominic McAlea, Major Terje Lundt and Major Oyvend Hoel) and an art historian (Mr. Colin Kaiser) to the area of Dubrovnik from 10 October to 4 November 1993 in order to:

  1. review and collect copies of relevant documents and reports in situ;
  2. meet with United Nations, European Community Monitoring Mission, Croatian and Serbian officials;
  3. visit and analyse relevant sites in and around Dubrovnik;
  4. videotape and photograph appropriate sites, and
  5. interview witnesses.

       The ultimate objective of the team was to prepare this analytical survey focusing upon injury to civilians and to civilian objects (particularly cultural objects) with a view to:

  1. determine whether and when indiscriminate attacks or deliberate attacks on civilians or civilian objects have occurred;
  2. quantify the loss of civilian life, injury to civilian persons, and damage to civilian property, including cultural property, and
  3. attribute responsibility for apparent violations of the law of armed conflict.

B. Methodology

       The methodology contemplated in the original plan of action was deviated from, to a certain extent, as some circumstances differed from what had been anticipated. Therefore, the survey became less ambitious in some aspects. For example, the documentary evidence available on some matters was so voluminous that rather than seeking to collect substantive evidence with respect to all potential violations of the law of armed conflict during the Battle of Dubrovnik, the Commission assessed the main sources available in terms of type, quality, category, comprehensiveness and probative value with a view to providing substantial assistance to a future prosecution team. This does not mean that the Commission failed to gather substantive evidence with respect to some particular, potential violations of the law of armed conflict. Certain promising cases were examined as well as the subject of specific recommendations in this report (see chapter XII) for further prosecutorial investigations.

       In addition to the problem caused by the volume of the available documentary evidence, it became apparent soon after the commencement of its survey that the Commission would be unable, given the time constraints, to conduct comprehensive, or even effectively representative interviews of witnesses or visits to sites. The number of potential witnesses of violations of the law of armed conflict was so great, and they were so widely dispersed, that a team twice as large as the present would be required to carry out this task, as well as several months on- site to complete the investigation thoroughly. Many suspected violations took place in the occupied areas of the District of Dubrovnik in addition to what was suspected to have occurred in the City of Dubrovnik. The part of the District of Dubrovnik, which was the subject of the Commission's survey, included hundreds of towns and villages spread along a seventy-five kilometre band in which approximately three to five thousand residents remained throughout its occupation by the JNA (which was for more than a year in some places).

       Although the Commission was provided with a plethora of information from different sources, other crucial sources of information proved to be inaccessible. Information held by the Croatian Ministry of Defence with respect to the constitution, location, and movement of military forces in and around the District of Dubrovnik from September 1991 until October 1992 was the most notable gap. This is only one example of how access to important sources of information was effectively denied, even though the Commission had previously visited all the key local government officials and secured their promise of positive support for the Commission's investigative efforts. In a related vein, the need to acquire some information from other than Croatian sources also proved to be problematic because of the short time frame available to the Commission to complete the report, and because of the Commission's limited ability to cross political boundaries.

C. Sources and analysis of information

       The evidence assessed by the Commission came in many forms:

  1. oral narrative from first hand witnesses;
  2. written witness statements or memoranda prepared at the time of, or shortly after, the event in question;
  3. hearsay;
  4. photographic and videotape evidence;
  5. physical evidence such as unexploded ordnance;
  6. documentary evidence including the reports of previous investigations by national bodies, such as the civilian police, or international bodies, such as UNESCO, and
  7. a local district criminal court judgement.

       The actual sources of information were also quite varied:

  1. former JNA Captain Zeljko Soldo and many other first hand witnesses;
  2. the Dubrovnik civilian police war crimes investigation files;
  3. the report and files of the Dubrovnik Civil Defence agency;
  4. the ECMM HQ files in Zagreb relevant to the period;
  5. the UNESCO report on the damage to the Old Town of Dubrovnik;
  6. the Council of Europe report on the damage to the small towns and villages of the District of Dubrovnik;
  7. the reports of the Institute for the Protection and Conservation of Historic Monuments on the damage to the Old Town of Dubrovnik;
  8. damage repair estimates from the Restoration Institute of Dubrovnik;
  9. the original judgement of the District Court of Dubrovnik in the case of former JNA Captain F. Soldo as well as F. Soldo's appeal;
  10. guided, as well as self-directed, site visits to various locations within the Old and New Towns in the City of Dubrovnik and District of Dubrovnik;
  11. the report of the Croatian Academy of Arts and Sciences on the damage to the old Arboretum of Trsteno;
  12. casualty reports from the Dubrovnik Red Cross;
  13. casualty reports from the databank Dr. Ivan Kostovic (Vice-President of the Republic of Croatia) in Zagreb;
  14. videotapes made by private citizens of various military activities of the JNA during the relevant period, and by team members, of the current state of damage to the Old Town, and
  15. the files of the office of the Dubrovnik District Prosecutor's Office.

       The Commission always assessed this evidence, with a view to identifying what would be relevant and useful for the purposes of future war crimes prosecutions before the International Criminal Tribunal for the former Yugoslavia (ICTFY). Therefore, evidence with respect to essential elements of potential war crimes was sought, and when found, it was analysed in terms of type, quality, category, comprehensiveness and probative value. In collecting evidence or identifying and analysing sources of evidence, the Commission's team first sought evidence which was independent and directly probative. Evidence of lesser quality was sought in default of this or as corroboration. It also attempted to build redundancy into the bases for potential future prosecutions by always seeking out secondary, non-connected, sources of evidence as supplementary proof of the essential elements of suspected violations of law of armed conflict.

II. Cultural-historical perspective of Dubrovnik

A. Background

       In 1991, the City of Dubrovnik was capital of a district or community that covered some 979 square kilometres and stretched about 200 kilometres from the peninsula of Prevlaka on the Boka Kortorska in the south-east to the end of the Peljjesac peninsula in the north-west. The mainland area itself is narrow, ranging from 500 metres to 15 kilometres in width.

       According to the census of 1991, the population of the City of Dubrovnik and the surrounding district was 71,419 of whom 49,728 lived in the city itself. Of this population, 82.4 per cent were Croatian, 6.8 per cent were Serbs, and 4 per cent were Muslims. The rest were comprised of 20 other ethnic groups, not including a small group of «others».

       The city and the rest of the district contain symbols of cultural heritage of outstanding value. The Old Town of Dubrovnik, for example, was «consecrated» by the World Heritage Committee by inscription on the World Heritage List in 1979. The region contains: small towns whose urban structure is of high historic interest (e.g. Mali Ston); numerous monastic establishments, especially Franciscan; and the summer residences of the old Dubrovnik aristocracy. The villages in the south- eastern region, Konavle, and particularly in the poorer north- western region, Primorje, present an extremely high degree of architectural authenticity, the latter forming a remarkable landscape of limestone houses in hilly and frequently arid settings, amidst the remnants of a traditional Mediterranean forest of oaks.

       The settlement of Dubrovnik was first mentioned in written records by an anonymous cosmographer of Ravenna in 667. Known in Latin as Ragusium, it was long known by its Italian name, Ragusa, before its Croatian name Dubrovnik (from Dubravka, «forest of oaks») acquired general acceptance.

       In the political vacuum left by the collapse of the Western Roman Empire, Dubrovnik found itself faced by numerous dangers from the Slavic peoples on the mainland and from invasion by sea. Geography forced it to turn towards the sea and to seek protection, or at least the benevolent indifference, of greater powers. This exigency would greatly influence its history up to the end of its independence which occurred de jure in 1808 with its attachment to the Illyrian provinces governed by France.

       Sequentially, Dubrovnik experienced the suzerainty of Byzantium, the Normans, and Venice, which was to remain its great rival for the Adriatic and eastern Mediterranean trade. Venice ruled Dubrovnik directly from 1000 to 1018, from 1171 to 1172, and intermittently after 1232 until finally losing control in 1358. From this year until the disastrous battle of Mohac in 1526, Dubrovnik recognized as its sovereign the Croatian- Hungarian kings. Until the fall of the republic of Dubrovnik in 1808, Dubrovnik's sovereign was the Sultan of the Ottoman Empire. During this period, Dubrovnik paid tribute, but at the same time enjoyed considerable trading privileges throughout the Empire. In 1699, Dubrovnik recognized the good offices of Vienna in settling its territorial disputes with Venice, and for a century added the protection of the Habsburgs to that of Constantinople. It is important to realize, however, that these dominations were nominal only and that Dubrovnik has effectively been an independent republic throughout most of its history.

       Militarily weak, Dubrovnik deployed treasures of diplomacy and amassed treasures from the sea and the land routes that wound through Bosnia. It concluded numerous treaties with Italian city states and negotiated advantageous arrangements with its powerful protectors. Toward the end of its golden age, in the last decades of the 16th century, it had a merchant fleet of 200 vessels. Devastated by the earthquake of 1667, which decapitated the town of its merchant aristocrats, it arose from its ruins thanks to its fleet and enjoyed a notable economic recovery in the 18th century.

       In 1814, Dubrovnik became a part of the Austro-Hungarian Empire and part of the Kingdom of Yugoslavia in 1918. Throughout the 19th century it continued to prosper, but its economic significance dwindled. The building of the Imperial Hotel in 1898, was a harbinger of the economic function that would predominate after the Second World War and expand enormously in the 1970s - tourism.

       The wealth of its trading aristocracy, combined with the loyalty to its own identity, played a vital role in the establishment of the Old Town that was honoured by the international community in 1979.

B. The Old Town of Dubrovnik

       The Old Town is contained in a small area. The distance from Fort Minceta to St. Peter's on the sea is 450 metres, while the distance from Fort Bokar in the west to Fort St. John in the east, parallel to the axis of the Stradun, is 550 metres. This dense urban space is confined by fortifications whose present aspect is derived mainly from the building campaigns of the 15th and 16th centuries. These fortifications are 6 metres thick on the land side, and 1.5 metres on the sea side, but individual bastions (Fort St. John and Fort Minceta) are thicker. The imperatives of organizing life within this defensive structure explain to a large degree the astringency of subsequent urban legislation.

       The settlement on the promontory that overlooks the sea, supposedly the home of the refugees who fled Epidaurus (present day Cavtat), was joined to the burg on the slopes of Mount Srdj by filling in an arm of the sea during the tenth and eleventh centuries. The municipal regulation of 1272 dictated building standards and forms that were long to be respected. Such regulation also softened the impact of changing architectural styles in the Mediterranean world and prevented the rebuilding that followed the catastrophic earthquake of 1667 from radically changing the scale and design of the new structures. Dubrovnik engaged Italian architects, whose creations (even those of the Baroque period such as St. Blaise, the Cathedral and, to a lesser extent, the Jesuit church) would not clash excessively with the surrounding architecture. All buildings were of hard limestone with the same pale yellow and light red roofing tiles. Aristocratic residences and most public buildings were sober in decoration, and important buildings attained monumental status through perspective and not size. The slow, regulated accretions of the ages created a remarkable architectural ensemble of great coherence. The architectural safety valve of expansion into neighbouring districts outside of the Old Town walls (Ploce, Pile) and the economic decline of the Old Town conserved its authenticity.

       The Old Town is not only an architectural and urbanistic ensemble of high quality; it also contains museums and libraries of interest (e.g. the collection of Ragusan masters in the Dominican Monastery, the Museum of the History of Dubrovnik, the Icon Museum, the libraries of the Franciscan and Dominican Monasteries). Above all, it houses the archives of Ragusa. Kept continuously since the last quarter of the 13th century, they are «the most important source for Mediterranean history» according to the French historian, Fernand Braudel.

III. The political situation

       Until its dissolution in 1991, the Peoples' Federal Republic of Yugoslavia was a federal state composed of six republics (Bosnia-Herzegovina, Croatia, Macedonia, Montenegro, Serbia and Slovenia), with two autonomous provinces attached to Serbia that were abolished in 1989. The federal state which was created at the end of the World War II was an attempt to escape the centralizing tendencies of the Serbian monarchy of the first state of Yugoslavia (1918-1941). Under the leadership of Marshal Tito nationalist tensions were kept under control, but after his death in 1980, these aspirations became stronger and were exacerbated by the economic crisis within the country.

       Tensions increased markedly after 1985 - with the nationalist revival in Serbia, riots in Kosovo (one of the autonomous provinces) in 1989, and the end of the dominant role of the single party - the Communist League - in Slovenia the same year. In 1990, there were free elections throughout the Federal Republic which brought to power nationalist parties, such as the HDZ of Franjo Tudjman in Croatia. Slovenia declared that it was a «sovereign state» within the Federal Republic in July, and Croatia modified its own constitution in July and December. At the same time, a Serbian autonomous movement emerged in the Krajina region of Croatia, which had a large Serbian population, giving rise to a declaration of separation from Croatia in February 1991.

       The federal presidency proved incapable of bringing about either effective economic reforms or satisfactory constitutional changes, and the dissolution of the Federal Republic accelerated. The first violent incidents occurred in the Krajina at Plitvice (31 March 1991), and they were followed by others (killing of Croatian policemen in Borovo Selo on 6 May 1991). In Croatia, a referendum on independence was carried out in May 1991, and it was followed by simultaneous declarations of independence by Croatia and Slovenia on 26 June 1991. The JNA (Yugoslavian National Army) intervened in Slovenia the following day, but mediation by the European Community led to agreement to withdraw JNA forces and a three-month moratorium on the independence of Croatia and Slovenia.

       In July and August of 1991, the situation in Croatia worsened due to numerous attacks staged by Serbian para-military forces who were supported by the JNA. A coalition government was formed in Croatia under the presidency of Tudjman on 1 August 1991, and two weeks later, the JNA intervened massively in Eastern Slavonia with the attack on the town of Vukovar beginning on 26 August. JNA garrisons were blockaded by Croatian forces, and further operations were carried out by the JNA in the Krajina and the Dalmatian coast, where Croatian towns were blockaded from the sea. On 25 September the United Nations introduced an arms embargo on all the republics of the former Yugoslavia. On 1 October 1991, the JNA invaded the District of Dubrovnik on multiple axes; the District was blockaded by sea and was subjected to bombardments by land, sea and air.

IV. The military situation

A. Sources

       As mentioned in Chapter I, the Commission's ability to obtain primary source information with respect to the Order of Battle and activities of the JNA and Croatian Army was effectively foreclosed for reasons which remain speculative. Other sources, less comprehensive or authoritative, provided a useful perspective. Those sources are:

  1. interview with former JNA Captain Zeljko Soldo on 26 October 1993;
  2. Civilian Police files;
  3. ECMM reports concerning the period in question;
  4. first hand witnesses, such as the members of the Dubrovnik negotiating team (crisis committee) which met regularly with representatives of the JNA forces, and
  5. UN reports.

B. The Croatian Army order of battle

       The Croatian military forces, present in the District of Dubrovnik in September 1991, were virtually non-existent. A platoon-minus sized body with light weapons manned the Imperial Fort on Mount Srdj (built during the Napoleonic period). It was stationed about one kilometre up Mount Srdj above the Old Town of Dubrovnik. No other regular army units were present in the district, except for a national territorial defence force or «homeguard», which could be considered operational from a military perspective. A Yugoslavian Naval installation, controlled by Belgrade, was present in the Prevlaka area at the extreme southerly tip of the District of Dubrovnik.

       The immediate reason for the absence of any significant Croatian military presence could not be attributed to the fact that the Croatian authorities' assessment of the threat to this district was insignificant. Certainly, the proportion of the Serbian population in this area was relatively low (6.8 per cent). Further, the military authorities in Belgrade had affirmed that it was not their intention to attack Dubrovnik. In addition, the District of Dubrovnik (apart from Prevlaka) had long before been largely demilitarized. The last regular army unit of any significance located in the District of Dubrovnik was moved in 1968, and the territorial defence armament was transferred in 1972 to Duzi and Grab, also in Bosnia-Herzegovina.

       Another important factor to this equation is the General Order issued in May-June 1989, in Belgrade, at the JNA Headquarters, by which it would appear that the territorial defence forces of Croatia and Slovenia were disarmed. This element is essential for understanding the Croatian military situation, as its knowledge can be imputed to the senior JNA authorities. This may fundamentally effect the validity of their direction to subordinate JNA authorities with respect to the presence and nature of military objectives in the District of Dubrovnik.

       The presence of the Croatian Army in the District of Dubrovnik has increased significantly since January 1992, nevertheless, the information available to the Commission was scant and limited at this stage.

C. The «Yugoslavian» order of battle

       Even though the JNA and Yugoslav Navy and Air Force were involved in the bombardment and occupation of the District of Dubrovnik, only limited information regarding their conclusion and activities were made available to the Commission.

       The JNA unit which occupied the hills closest to, and immediately above, the City of Dubrovnik was the III Motor Vehicle Battalion of the Trebinje Brigade detached for the purposes of that operation to the Military Naval Sector with headquarters in Herceg Novi, Montenegro. It was comprised of three sharpshooter (rifle) companies, one anti-tank company, one mixed-mortar company and one tank detachment under the command of a named officer not disclosed for confidentiality and prosecutorial reasons. It had a strength of approximately 800 to 1000 personnel; 50 per cent professional JNA soldiers, and 50 per cent were reservists from Montenegro and Bosnia-Herzegovina.

       As previously stated, elements of this unit were positioned all around the hills, directly overlooking the City of Dubrovnik (Zarkovia and Bosanka in particular), and they participated extensively in the bombardment of both the Old and New Towns of Dubrovnik on 6 December 1991. They appear to have possessed a wide assortment of former Warsaw-Pact type weapons, as well as Yugoslavian variants of the same. These included: 60 milimeter, 82 milimeter and 120 milimeter mortars; 82 milimeter rockets; 20 milimeter anti-aircraft guns; 85-105 milimeter tank guns; 62 milimeter recoilless rifles, and 7.62 milimeter assault rifles and general purpose machine guns. With the approval of the sector commander, this unit also had access to the employment of 130 milimeter artillery.

       The sector commander was a JNA officer of Vice-Admiral rank. A number of other officers involved in this operation both on sea and on land are known but not disclosed for confidentiality and prosecutorial reasons.

       In regard to the strength of the JNA forces which occupied the District of Dubrovnik, estimates range from 7,500 to 20,000 men. A more precise estimation of the number of JNA personnel involved in this operation and a determination of the place of origin of most of them remains to be discovered. Little else is known about the constitution of the JNA forces apart from the names of several individual officers who came into contact with local Dubrovnik authorities.

V. The occupation of the district of Dubrovnik

A. Sources

       As mentioned in Chapter IV, the lack of information from Croatian and Serbian military sources precludes, at present, any detailed or authoritative description of the military actions which resulted in the occupation of the District of Dubrovnik by JNA forces in the fall of 1991, or on the events which occurred during that occupation. This gap in information prevents the Commission from determining many facts which could be crucial for the attribution of criminal responsibility. For example, it was often possible for the Commission to determine when and where a civilian was killed or wounded. The attribution of criminal responsibility, however, requires clear answers to the following questions:

  1. Did the person die as a result of military action or accident?
  2. Prior to his/her death, was this «civilian» acting consistently with his/her status as a non-combatant?
  3. If the person died as a result of military action, was he/she specifically targeted or was he/she a collateral victim of an attack upon a valid military objective or was he/she the collateral victim of an attack upon an unlawful target?
  4. Who committed the act which resulted in the civilian's death?
  5. Was the person, who caused the death of the civilian, acting on his/her own initiative or pursuant to orders? If such orders were given, who gave them or passed them on?
  6. If the person who committed the act was acting on his/her own initiative, what guidance had he/she been given by his/her superiors regarding appropriate conduct? Who are his/her superiors?
  7. Did the superiors of the person who committed the act monitor his/her conduct and the conduct of all their subordinates? Who are their superiors?

       These particular questions focus upon the death of a civilian or non-combatant, but they represent the type of questions which must be answered with respect to all types of violations of the law of armed conflict, whether they refer to lawful combatants, civilian property, religious or historical property.

       Without information on the marginal activities to the ad hoc military units formed in Croatia to resist the JNA advance, the determination of what constituted a valid military objective for JNA attention is very difficult. Similarly, without information about the names and locations of commanders and command structure of JNA units and formations, finding the answers to the questions posed in paragraph 37 becomes a matter of an enormous expenditure of investigative energy and luck. Therefore, this basic information should be obtained before any future prosecutorial teams begin assembling cases with respect to specific law of armed conflict violations.

       Although the Commission's team was able to obtain a limited amount of information which exists on the subject, the Commission has managed to construct a general outline of how most of the District of Dubrovnik came to be occupied by JNA forces and some of the events which occurred during this period which suggest a violation of the law of armed conflict. This has been done with information obtained from first-hand witnesses and reports from the Dubrovnik Civilian Police, the Dubrovnik Civil Defence authorities, the Dubrovnik Red Cross, the United Nations Military Observers Detachment Dubrovnik and the former ECMM Detachment Dubrovnik.

       Independent information on the constitution and actions of the respective military forces is expected to be provided in the near future through the Commission of Experts. This additional information should contribute to a better understanding of the events in question.

B. Konavle - the south

       The political tensions described in Chapter III manifested themselves militarily at the end of September 1991 in Konavle, located in the south-east of Dubrovnik. At that time the Yugoslavian Navy established a blockade along the southern coast of Dubrovnik, and the JNA deployed its forces along the border with Croatia. On 23 September 1991, the southern village of Vitaljina was bombarded. Many villages throughout Konavle were subjected to artillery bombardments, including the village Brgat, located five kilometres from the city of Dubrovnik.

       On 1 October 1991, the JNA entered Konavle by several routes and began to move steadily north-west through Konavle toward the city of Dubrovnik (this action was co-ordinated with a similar, multiple-axis advance toward the city of Dubrovnik through Primorje and the area known as the Rijeka Dubrovacka areas to the north-west of the City of Dubrovnik). As far the Commission's team has been able to determine, the resistance to the JNA's advance took the form of small, poorly armed, and hastily formed, bodies of «homeguard», composed of men from the Croatian police and former members of Croatian descent who had deserted the JNA. Information regarding their numbers, weaponry and activity is limited. It can be said, however, that the nature of this resistance was such that there was no «front line» distinction between the JNA and Croatian military forces - only the edge of the area up to which the JNA had advanced.

       By the middle of October 1991, the JNA forces had advanced to an area approximately fifteen kilometres south-east of the City of Dubrovnik, near the village of Mocici. This advance caused thousands of residents of Konavle to seek refuge in the City of Dubrovnik, thereby severely straining the city's already diminished resources (see Chapter VI). Thousands of people remained in their residences throughout the occupation, and they represent an enormous potential source of information on alleged violations of the law of armed conflict. At that time, the ECMM arranged for representatives of the JNA forces and the City of Dubrovnik to meet in order to negotiate on matters such as cease- fires, public utilities and prisoner exchanges. These negotiations and cease-fire arrangements had no real effect upon the advance of the JNA toward the City of Dubrovnik. During the next two weeks, the JNA forces advanced north-westward through Cavtat, Plat, and Zupa Dubrovacka until they occupied, on 26 October 1991, the promontory of Zarkovica, which commands a direct view of practically all of the City of Dubrovnik.

       The occupation of the Zarkovica promontory represented the end of the JNA's north-westward advance and the beginning of what was to be a year long occupation of the entire area of Konavle by the JNA forces. By this time, many civilians in Konavle had been killed or wounded as a result of the military action incidental to occupation of Konavle by JNA forces. As the entire details of the circumstances surrounding these deaths and wounding were not available to the Commission's team, this report cannot present, at this stage, a complete and balanced account of the events. Nevertheless, due to the nature of the resistance mounted against the JNA during its advance through Konavle and the district paucity of any valid military objectives, it is suggested that many of these deaths and wounding were not militarily necessary. Subject to the concerns set out in paragraphs 31 and 32, the extensive destruction of civilian property and cultural property by JNA forces during their advance or during the occupation, such as the damage to the Franciscan Monastery in Pridvorje or the Ethnographic Museum in Cilipi, would appear to fall into the same category. For these reasons, the conduct of the JNA forces during the advance north-westward will be the subject of a recommendation in Chapter XII.

       Numerous allegations have been made with respect to the abuse of civilians who remained in Konavle during its occupation by JNA forces. Also, many allegations have been made with respect to the destruction and pillage of civilian and religious or historical property by JNA forces during their occupation of Konavle. The Commission's team has identified some victims of, and witnesses to, these alleged violations, as well as reports prepared by local authorities. These allegations and this information will also be the subject of a recommendation in Chapter XII.

C. Primorje - the north

       The co-ordinated, multiple-axis advance of JNA forces through Primorje, in the north-west, toward the City of Dubrovnik experienced the same negligible resistance as the JNA forces which advanced from the south-east. However, they did not follow a similar timetable. The JNA forces from the north-west entered Croatia at the beginning of October 1991, as did those from the south-east, but they did not occupy their final position surrounding the City of Dubrovnik until approximately 24 November 1991, almost one month after the JNA forces from the south-east halted their advance on the City of Dubrovnik. There may have been several reasons for this.

       First, the control by the JNA forces of the strategically located Pigeon Rock on the border of Croatia and Bosnia- Herzegovina from the beginning of October 1991 effectively hemmed in the City of Dubrovnik to the north-west by land. By sea, the City of Dubrovnik was blockaded by the Yugoslavian Navy, and there was no infiltration without its permission. In combination, these factors would not have required the JNA forces to «hurry» down the coast from the north-west.

       The JNA forces may also have approached this area more slowly because it was more internationally significant at that time. It may be recalled that the Kouchner Plan of 19 November 1991, called for the JNA forces advancing from the north-west to stop at Mokosica and for this halt to be monitored by ECMM monitors. This plan was followed closely by the Geneva Agreement of 23 November 1991 which also called for no further troop advancement. On the other hand, both of these agreements were violated as soon as they had become effective. However, with the information currently available, it is not possible to determine why the JNA forces advanced toward their final static position the way that they did.

       The evidence available does clearly suggest that, (even more so than in Konavle), the nature of the resistance to the south-eastward advance of the JNA forces and the paucity of any valid military objectives in Primorje were such that many of the deaths and wounding of civilians, (significantly higher in Primorje than in Konavle), and much of the destruction of civilian property during this advance was not militarily necessary.

       The concerns expressed in paragraph 46 above regarding the destruction and pillage of civilian property and religious or historical property during the occupation by JNA forces, (which lasted until the end of May 1992), are equally applicable to Primorje. Subject to the concerns set out in paragraphs 37 and 39 above, it is hard to imagine how the looting and desecration of the Franciscan Monastery of St. Jerome in Slano or the destruction of parts of the old Arboretum of Trsteno can be anything but two glaring examples of destructive acts which were not militarily necessary. As in Konavle, the occupation of Primorje caused many people to seek refuge in the City of Dubrovnik. Of course, many residents of Primorje remained in their homes during the occupation by JNA forces, and the information which they can provide with respect to the violations of the law of armed conflict during the occupation of Primorje should not be overlooked. Therefore, the conduct of JNA forces in the advancement through, and occupation of, Primorje will also be the subject of a recommendation in Chapter XII.

VI. The «siege» of Dubrovnik: october-december 1991

A. Sources

       In addition to the sources noted in Chapters IV and V, information regarding the siege of Dubrovnik was derived from the reports of the UNESCO observers and the Institute for the Protection of Cultural Monuments and the Natural Environment Institute. UNICEF was also present in Dubrovnik throughout November and December 1991, and possesses potentially useful information. Extensive videotape records made by ITN, local amateurs, the Dubrovnik Civilian Police and the Dubrovnik Civil Defence agencies during their investigations, and from Montenegran television are also available.

B. Situation of the city area

       The invasion of the District of Dubrovnik by the JNA led to masses of its residents seeking refuge in the City of Dubrovnik. These refugees numbered approximately 15,000, and most of them took shelter in the numerous hotels in the New Town (west of the Old Town); some stayed with relatives or friends. Approximately 7,000 others were evacuated by sea in October 1991. These evacuations became more difficult as the siege progressed, but negotiations with the JNA carried out by international organizations, such as the ICRC and UNICEF, with the assistance of Croatian humanitarian organizations and the local crisis committee, ensured that a further 9,000 persons, (mainly pregnant mothers, mothers with small children, the elderly and the sick), were able to leave by sea.

       On 1 October 1991, the electricity and water to the city were lost as a result of aerial bombardment by JNA forces of the district's electrical grid. Except for those few people who had generators, the people of the City of Dubrovnik lived without electricity until the end of December 1991, and they were dependent on ships for their supply of fresh water.

C. Bombardments during October and November - military positions and weapons

       It is difficult at this time for the Commission's team to present a precise chronology of the bombardments of, or firing on, the city area. With the exception of the discussion of the bombardments on the Old Town, the following comments will require revision as more information becomes available.

       The Imperial Fort on Mount Srdj above the city and the telecommunications tower next to it, the hill of Zarkovica and the Ploce district (south-east area of city) were bombarded on 1 October 1991 and again the following day. On 3 October, the Hotel Belvedere, (in which there was a Croatian military post), was shelled, and JNA aircraft dropped bombs in the sea near the Hotel Argentina. There was further shelling of Ploce on 5 October 1991 and an airstrike on the Imperial Fort on 6 October 1991.

       The city was relatively quiet thereafter until 18 October 1991, when airstrikes fell on Ploce again. Two days later, there was another airstrike on the Imperial Fort and further shelling of Ploce. On 22 October 1991, the hotels inhabited by displaced persons in Lapad (north-west area of city) were subjected to naval bombardment.

       On 23 October 1991, the Old Town was bombarded for the first time. In particular, 120 milimeter mortar fire hit the Old Granary (Rupe), and 82 milimeter rockets struck a house on the Ulica Boskoviceva. It was also reported by the local radio station that the Sponza Palace had been hit. This bombardment probably originated from he JNA positions consolidating on the hills above the city.

       On 24 October 1991, there was a naval bombardment of the Hotel Belvedere and uninhabited Lokrum Island. The next day, the JNA representatives presented an «ultimatum» to the Dubrovnik crisis committee and the ECMM representatives. This «ultimatum» demanded the disarmament of the city and the immediate departure of all Croatian military forces and «foreign mercenaries», as well as the removal of the public officials who had been elected during the previous free elections. On 26 October 1991, the Yugoslavian flag was hoisted at Zarkovica, a point on a promontory 2.3 kilometres south-east and up from the Old Town. Zarkovica looks directly down onto virtually all of the City of Dubrovnik, and it is a textbook example of a point from which military actions can be directed against a location like the City of Dubrovnik.

       According to an ECMM report, the City of Dubrovnik was shelled again on 30 October 1991, and 6 civilians were wounded. This shelling was also responsible for the deaths of a mother and her children on the following day. This same source details the shelling on the Gruj and Lapad (western city zone) on 1 November 1991. A complex of hotels in Babin Kuk, (also in the western city zone), was bombed on 2 November, and several refugees were wounded as a result. According to the ECMM report, shrapnel from JNA mortar and artillery fire struck the Hotel Argentina the next day.

       Automatic weapons were fired on the Old Town from the lines of the III Motor Vehicle Battalion on 3 November 1991, wounding one person on the Stradun, (the main east-west thoroughfare in the Old Town). On 4 November 1991, similar fire was directed toward the Hotel Argentina, and on the following day, the Imperial Fort was bombarded again.

       On 7 November 1991, an ultimatum was announced on the local radio, demanding that all Croatian forces surrender by noon. On 8 November 1991, Vice-Admiral Jokic announced that discussions with the Croats had not succeeded and that he would spare only the Old Town. Bombardments continued from 9 to 12 November, some of which were recorded on videotape by ITN. On 9 November, the Imperial Fort, Gruj and Lapad were bombarded. The following day, the Imperial Fort was hit again, as well as the Hotel Belvedere, the Excelsior Hotel, the Hotel Argentina and the Old Town. Naval vessels participated in this bombardment. The Imperial Hotel, just west of the Old Town, the Hotel Argentina, the Tirena Hotel on Babin Kuk, the new (St. Blaise) Hospital, and the Old Town were hit the following day. On 12 November, the Ploce district, the Imperial Hotel and the Old Town were bombarded again. On 13 November, only the Imperial Fort was attacked.

       Apart from the targets mentioned, there is little precise information available on targeting except with respect to the Old Town. Bombardment damage to the Old Town was recorded by the Institute and for the most part verified by UNESCO observers. Up until this point, mainly 82 milimeter rockets had been fired at the Old Town. The front of the Small Port in the south-east of the Old Town and the houses along the Stradun had been hit by these projectiles, as were the St. John Fort and St. Peter's Bastion, which were being used as shelters. Among notable monuments hit were the Sponza Palace (roof and south-east facade), the Jesuit Church (roof and corniche damaged by a 120 milimeter mortar), the Franciscan Bell Tower, and the Town Clock Tower. The home for the elderly south of the St. Claire Monastery and the city wall behind it were also hit. The location of these impacts were such that the UNESCO observers concluded that the bombardment had come almost entirely from the south-east. Pleasure craft in the Small Port were also hit with wire-guided rockets.

D. A month of negotiations

       Throughout the rest of November 1991, there was a lull in JNA activity during which intensive negotiations between the JNA representatives and the Dubrovnik crisis committee took place. Such meetings had been brokered throughout October by the ECMM, but they were withdrawn in mid-November after an incident in which the ECMM claimed to have been deliberately targeted by the JNA. At this time, Bernard Kouchner, the French Minister of Humanitarian Affairs, and Stephan Di Mistura, Chief of the UNICEF Mission, took an active role in attempt to broker a cease-fire.

       As a result of these efforts, a General Protocol Agreement was signed by both sides on 19 November 1991, which provided for mutual withdrawal of armed forces behind a neutral zone and, «under neutral control», the delivery of weapons and withdrawal of military elements for Dubrovnik. Another agreement, signed in Geneva on 23 November 1991, among other things called for a cease- fire and withdrawal of JNA forces from the District of Dubrovnik. Yet another agreement, specific to Dubrovnik, was negotiated on 5 December 1991, punctuated by the 6 December 1991 bombardment and then signed on 7 December 1991. None of these agreements had any real effect upon the JNA activity in the district.

       On 2 December 1991, automatic weapons fire was directed on the Old Town. Witnessed by one of the UNESCO observers, it wounded one civilian. The following day, automatic weapons fire wounded several civilians at Ploce. On 4 December, the Imperial Fort was subjected to mortar fire.

E. The St. Nicholas day bombardment - 6 December 1991

       At 5:48 hours, on 6 December 1991, the single heaviest bombardment that the Old Town was to experience began. In its opening stages, 82 milimeter rockets struck the Old Town, (48 impacts according to the Institute). The bombardment continued with the utilization of 82 milimeter mortars (232 impacts) and 120 milimeter mortars (364 impacts). The intensive shelling lasted until about 11:30 hours. Wire-guided rockets (22) were used on yachts in the Small Port, and others were fired on the Old Town itself. For example, videotape records show one of these hitting the drum of the Cathedral.

       It is possible that heavier calibre weapons were used. This could be the case since two impacts on the interior parapets of the Old Town caused more damage than would normally be expected from 120 milimeter mortars. Local records indicate that, in addition to those already mentioned, about 300 other projectiles landed in the Old Town, but did not explode. This information has yet to be verified. Photographic evidence and rough crater analysis suggest that, as with previous bombardments, most of the firing originated from the south-east. Combining these findings with the known range of projectiles involved, the source can reasonably be established as coming from a particular unit whose name and emplacement are not mentioned for confidentiality and prosecutorial reasons.

       On 6 December 1991, the targeting was different than earlier bombardments. The south-eastern front of the Old Town, except for the Dominican Convent, was relatively unscathed. This means that the monumental zone at the end of the Old Town, including the Sponza Palace, the Rector's Palace, St. Blaise Church and the Cathedral, maintained its integrity. Similarly, the southern part of the Old Town on the promontory over the sea did not receive many impacts either.

       While there were many impacts north-east of the Stradun and on the Stradun itself, the preponderance of the heavy mortars fell upon the north-western part of the Old Town. The Franciscan Convent, the buildings near the Stradun and the buildings in the zone south-west of the Stradun were most seriously hit. This is also the area in which most of the burning damage occurred. According to the Institute report, a large proportion of the 120 milimeter mortars fell there. Some of those mortar rounds probably contained white phosphorus. This conclusion is suggested because of the distinctive red stains which were evident on the surfaces where these mortars exploded.

       It might be more than coincidental that the zone which received the worst bombardment damage was also the area which had been most seriously damaged by the earthquake of 1979. However, this concordance might just as easily have been a coincidence. In a similarly cognitive vein, some evidence suggests the mortars fired upon the Old Town on that day were systematically firing upon pre-established grid co-ordinates, while other evidence suggests that the targets were selected opportunistically as those witnessing the effects of the bombardment from the observation post at Zarkovica monitored the activity of the JNA forces.

       This bombardment not only struck the Old Town, but the Inter-University Centre west of the Old Town burned down as well, along with its library of 20,000 volumes. In addition, the worst civilian casualties (see Chapter VII) were incurred on this date, at the Hotel Libertas which is in the New Town and clearly visible from Zarkovica. Unfortunately, the information on the bombardment outside of the Old Town has not yet been organized in a way which would permit its proper analysis. It is buried in the reports of the Institute, the Civilian Police agencies, the Dubrovnik Red Cross, and probably the files of the Ministry of Defence.

F. The inquiry of the JNA

       The international media, the Director-General of UNESCO (Federic Mayor), and Cyrus Vance, (the UN negotiator for the Yugoslavian crisis), protested strongly about the bombardment of 6 December 1991. The ECMM in Zagreb also lodged a strong protest the same day. In response, the JNA promised «a full investigation of the incident» and asked Croatia to do the same. Later that day, a statement of regret and apparent intention to discipline those responsible was received from the JNA. The team was unable to determine if this promised investigation was actually carried out. We know that on 7 December 1991, a delegation from the JNA visited the Old Town to investigate the damage, but we do not know if any other follow up occurred. A request has been made by the representatives of the Commission of Experts stationed in Zagreb to the Belgrade authorities, on the status of the promised investigation and disciplinary action taken on this matter, but at the time of writing, no response had been received.

       After 6 December 1991, the JNA forces surrounding the City of Dubrovnik were relatively inactive until April-May 1992. In the meantime, the Croatian Army's capability in the city was increased dramatically. This resulted in a proliferation of valid military objectives in the City of Dubrovnik (but not in the Old Town). This increase in number and distribution of valid military objectives, combined with the Commission's lack of information on the constitution, location, and movements of these Croatian military forces, makes any realistic assessment of violations of the law of armed conflict after this build up impossible. Therefore, at present, this survey will be largely limited to the period up to 7 December 1991.

       From the perspective of the law of armed conflict, it is clear that there were valid military objectives in and around the City of Dubrovnik. Above the city was the Imperial Fort on Mount Srdj which was besieged continuously, but never overtaken by the JNA Forces. In addition to this, the team is aware that there were Croatian military forces located in Sustjepan and the Container Port on the Rijeka Dubrovacka; these locations were the only spots in the area which remotely resembled a «front line». Finally, evidence suggests that there may have been some Croatian military forces in Babin Kuk and Petka, however, this is uncertain at this point in time.

       One point that remains clear is that the closest that any of these valid military objectives were to the Old Town was within one kilometre of the Imperial Fort, and most of them were between three to four kilometres away from the Old Town. They also appear to have been insignificant in terms of the numbers of military personnel and the calibre of their weaponry. In addition, knowledge of the special historical status of the Old Town which was recognized by UNESCO in 1979 upon the sponsorship of Yugoslavia can be imputed to the JNA. In addition to these aforementioned points, there is evidence that the Dubrovnik authorities, (aided by UNESCO observers), appear to have been scrupulous about keeping weapons out of the Old Town, that the besieging forces could see virtually everything that was going on in the Old Town, and that the Old Town was clearly subject to indiscriminate, and possibly even deliberate, targeting. Therefore, this conclusion will also be the subject of a recommendation for further investigation with a view to prosecution in Chapter XII.

VII. Civilian casualties

A. Sources of information

       Sources of information with respect to those killed and wounded during the Battle of Dubrovnik are numerous and varied, and the Commission's team studied as many cases as were made available to them. In particular, the Commission's team studied the data available from the Dubrovnik Red Cross, the databank from Dr. Ivica Kostovic (Vice President of the Republic of Croatia), the Civilian Police and the data available along with the report of the Dubrovnik Civil Defence agency.

       While using so many different sources may seem to represent unnecessary redundancy, such an approach was not wasteful since each source reflected a different approach to the data and contained unique information. For example, the Dubrovnik Red Cross data dealt with everyone from the District of Dubrovnik who had been killed as a result of hostilities. In addition, it made a distinction between civilians, the military, and police, and it provided the age of the victim as well as the date and place of his or her death. This data was presented by listing the victims alphabetically by name.

       Dr. Kostovic's data listed all dead and wounded from the district. However, it made no distinction between civilians, the military, or police. Further, it provided the date of birth of the victims and the date of wounding with respect to those who had been killed, as well as those who had been wounded. This data was provided by listing the victims name in alphabetical order. The approach of the Dubrovnik Civil Defence agency was different again. It only listed, in chronological order, civilians from the district, as well as their age, the place they were injured, the date they were wounded and their degree of gravity, (from lightly to mortally wounded). Contrary to other lists, information continued to be provided beyond, and it extended until March 1993.

       The data provided by the Civilian Police arrived too late and it could not be analysed by the Commission.

B. Analysis

       Although evidence with respect to the dead and wounded was plentiful and well presented, real problems subside. The data differs, not just in perspective, but also in substance. For example, the names of some people who have been killed appear on all three lists, in other cases their names appear only on two lists and, in other cases, only on one. In addition, the lists often contradict each other with respect to crucial information such as the date or place of injury.

       These shortcomings do not indicate that the data is un- useful. For prosecutorial purposes, it will have to be independently verified and properly cross-referenced in terms of names, dates, places and nature of injury.

C. The dead and wounded

       With respect to civilians who died as a result of hostilities, at least 82, (and possibly as many as 88), from the District of Dubrovnik were killed as a result of the JNA advance and occupation of the area from September 1991 until the end of December 1992.

       It is also clear that most (about 50) of these civilians died as a result of JNA action prior to the St. Nicholas Day bombardment on 6 December 1991. Thirteen civilians died as a result of the bombardment on that date, and this represented the greatest number of civilian deaths that had, or would, be sustained on a single day during the Battle of Dubrovnik. About 20 more civilians were killed over the next twelve month period as a result of the hostilities.

       Fewer conclusions can be made with respect to the available data on civilians wounded during this period. However, once the work contemplated in paragraph 82 above is completed, a more accurate account will be available regarding wounded civilians. In the meantime, it is clear that hundreds of civilians were killed or wounded, and while these figures may not seem high in absolute terms, when put into perspective, they become shocking.

       To create this perspective, one must consider that civilians are non-combatants, (assuming that they do not do anything to jeopardize that status), and they may not legally be the object of attack by opposing forces. The law of armed conflict does contemplate the death or wounding of civilians as being legally possible, but this may only occur incidentally or collaterally to attacks upon valid military objectives. When one considers: the overwhelming military superiority of the JNA forces throughout the period during which most of the civilians were killed or wounded (1 October 1991 to 6 December 1991); the negligible size, offensive impotence and isolated location of the pockets of resistance to the JNA occupation and siege of the District, and City, of Dubrovnik; the weapons used by the JNA (thousands of rounds or mortar and artillery fire of various calibers), and the ability to identify and neutralize specific Croatian military objectives which JNA forces had because of the weapons systems made available to them. Due to the ideal observation capacity that the JNA enjoyed through its command of the high ground, the air, and the sea, it seems clear that (at best) the JNA was indifferent to the civilian casualties it caused or (at worst) it deliberately and systematically targeted civilians and civilian objects throughout this period.

       As stated above, discrepancies and incompleteness of data with respect to civilian casualties and lack of information with respect to their nature are more comprehensive or precise than the location of activity as the respective military forces preclude. Therefore, the state and cohesiveness of the data with respect to the civilian casualties caused by the Battle of Dubrovnik will also be the subject of a recommendation for further investigation in Chapter XII.

VIII. Damage to civilian property

A. Sources

       The Institute for the Restoration of Dubrovnik is responsible for evaluating the war damage to private and public property in the District of Dubrovnik, with the exception of the cultural heritage, which is the responsibility of the Institute for the Protection of Cultural Monuments and the Natural Environment.

       So far, the former institute has completed its survey only for the housing of the district of Dubrovnik. Categorization of damage and the financial estimates for public buildings such as hotels, factories, shops, forests, etc. should be ready in the spring of 1994.

B. Categorization of damage

       The statistics made available by the Institute cover the housing of the district, which includes thousands of buildings in the villages that are vernacular cultural heritage. The categorization used comprises six levels of damage:

  1. 1-3 represent minor damage to surfaces to slight damage to the building structure;
  2. 4 is heavy damage to the building structure;
  3. 5 is partial destruction of the building, and
  4. 6 is total destruction of the building.

       Unfortunately, this system of categorization gives no indication of the actual means of destruction. Therefore, much of it is of insignificant use for purposes of attributing criminal responsibility. In the District of Dubrovnik, for example, burning was the cause of most damage except for the damage caused by shelling which is apparent in the «fighting» zones near to towns. In the New Town of Dubrovnik, shelling is the main cause of damage.

       The following table represents the categories of damage for three zones, including the Old Town, but it must be kept in mind that the figures for the World Heritage Site do not include damage to sacral structures and many public buildings and monuments:

              1       2       3       4         5       6
 New Town    823     368     235      48        12       3
 Old Town     43      70      68       5        13       1
 District   2401    2218    1754     813      1650     882
 TOTALS:    3267    2656    2075     866      1675     886
Total Number of damaged buildings: 11,425

       These statistics reveal that the occupied areas of the district suffered the worst damage to housing, 93.9 per cent for category 4, 98.5 per cent for category 5, and 99.5 per cent for category 6. This explains why Chapter XII will include a recommendation that further investigations into war crimes include intensive investigation of damages to small towns and villages of the District of Dubrovnik.

C. Financial evaluation

       Specifically concerning only the question of housing, the Institute has calculated that the cost of reconstructing the buildings comes to DM 69,000,000, and the cost of complete reinstallation of the families (reconstruction, appliances, furniture, etc.) is DM 480,000,000 (prices on 31 December 1990). The methodology used was elaborated by the Building Institute of Zagreb, and modified for local conditions.

IX. Damage to cultural property

A. Sources - Old Town and New Town

       There have been numerous descriptions and evaluations of damage to the cultural heritage of the Old Town and, to a lesser extent, to the New Town of Dubrovnik. These have been prepared by the local Croatian cultural authorities, the Institute for the Protection of Cultural Monuments and the Natural Environment of Dubrovnik as well as by UNESCO. There has been a great deal of co-operation between the Institute and UNESCO in the preparation of this information since the arrival of the UNESCO observer mission of November-December 1991, in culmination with the preparation of the Action Plan for the Old Town of Dubrovnik, published in February 1993. With the exception of the first report of the Institute, which was prepared in English, all the other local reports exist only in Croatian.

       The first report of the Institute, dated 24 November 1991, contained maps and photographs and was prepared without outside assistance. It was entitled, «The Report on the War Destruction in Dubrovnik», and it covers damage in the Old Town which occurred during the bombardments of October and November 1991. The UNESCO observer mission checked most of the entries contained in this report and provided an evaluation of the damages for the Director-General of UNESCO.

       The observer mission elaborated on a methodology which was adopted by the Institute. This methodology was simplified and did not have the same exactitude as the methodology elaborated on by the National Institute for Monuments in Zagreb, which was only prepared in the latter part of 1992. Using the UNESCO observers' methodology, the Institute drew up a preliminary Report, contained in 7 volumes, on the Destruction in the Old Town Of Dubrovnik for the months of October, November, December 1991, which was completed on 6 January 1992. This document contains maps and photographs in addition to individual fiches on monuments and houses. It also covers the sections on the contact zones of the Old Town, and on Gruz and Lapad as well as second hand information about the occupied zones.

       While in Dubrovnik, from 27 November to 20 December 1991, the UNESCO observers prepared reports that were transmitted to the Director-General. The first was on 3 December 1991, and two others followed the bombardment of 6, 7 and 8 December 1991. The second two reports included maps of impacts. Their final report («Rapport sur l'etat du patrimoine culturel dans la ville de Dubrovnik � la suite des bombardments d'octobre, de novembre et de decembre 1991»), submitted in January 1992, included first- hand information as well as summarized data already included in the January report of the Institute. The observers also took approximately 1,000 colour slides of the damage caused by the JNA bombardments of October-December 1991.

       The Institute extended its work to include the New Town, using the same methodology, but working only on monuments and parks. This «Preliminary Report on War Destruction - October, November, December 1991», consists of 8 volumes and was completed in February 1992. This document contains many photographs and maps in addition to individual fiches.

       The present mission did not closely examine the bombardments of May-July 1992, because of a lack of crucial information on the contested nature of the District of Dubrovnik during that period. Mention should be made of the «Preliminary Report on War Destruction in the Old Town in June-July 1992», which was completed on 27 August 1992, using the same methodology.

       It should be pointed out that reports on damage to specific architectural elements were also prepared, but these grouped all the damage done in the bombardments of 1991 and 1992 («Reports on Destruction of Architectural Elements and Stone on Monuments of the Old Town of Dubrovnik, August 1992, Project Programme for Renewal of Roofs destroyed in the 1991-92 War, August 1992).

       The UNESCO Action Plan contains cost estimates for restoration of the most important buildings, roofs, street surfaces, fixtures and the walls. However, it does not extend to other areas regarding damage.

B. Sources - the district of Dubrovnik

       Analyses of damage to cultural heritage in the district are few and far between. The Island of Lokrum has been the subject of one report of the Institute (December 1991). Mali Ston was also the subject of a report (summer of 1992). The Croatian Academy of Arts and Sciences in Zagreb had prepared a report on the damage to the Old Arboretum of Trsteno. Also, the Institute is slowly collecting information on moveable heritage, the monuments, and the vernacular architecture of the villages, but, at the present stage, it cannot present any overall appreciation of damage in the district.

       The information report on «The Destruction by War of the Cultural Heritage in Croatia and Bosnia-Herzegovina», dated 2 February 1992 and submitted to the Committee on Culture and Education of the Parliamentary Assembly of the Council of Europe, contains a long section on damage in the district. This report was based on statistics presented by the Institute for the Restoration of Dubrovnik and on first hand observations carried out by its consultant experts. No exhaustive approach was used for its preparation. As a result, a number of villages which did not suffer extensive damages were not visited. The utility of this report is close in proximity to the time of withdrawal of the JNA from Konavle in late October 1992. The Commission is in possession of about 300 colour slides and prints taken during the mission, with captions on this subject.

C. Damage to the Old Town of Dubrovnik during the bombardments of October-December 1991

       The damage suffered by the Old Town during the bombardments of October and November 1991 was slight, relative to the damage sustained in December 1991. Caused primarily by 82 milimeter rocket fire, this damage was confined mainly to the stone surfaces of the Small Port, Fort St. John, St. Peter's Bastion, and the pavement of the Stradun. A number of houses were also hit, mainly in the roofs and the upper stories. In several instances, 120 milimeter mortars were shot, which caused extensive damage to roof structures. There were no fires caused by the bombardment during the October-November period, although the internal structures and contents of some buildings sustained some heat damage.

       The damage done by the bombardment of 6 December was far more serious, however, due to the type of projectiles used (notably 120 milimeter mortars); the number fired, and the duration of continuous bombardment prevented the fire department and citizens from extinguishing fires before getting out of control.

       In the report of January 1992, the Institute estimated that 55.9 per cent of the buildings had been damaged:

  1. 1 per cent had burned;
  2. 11.1 per cent had suffered serious structural damage;
  3. 26.4 per cent less severe damage to structural elements; and
  4. 17.4 per cent only minor damage to facades and roofs.

       The worst devastation was caused by the burning of the roof and interior structures of seven places from the Baroque period. Among them was the Festival Palace, where the archives from the Summer Festival of Dubrovnik were kept and later destroyed by the fire. In several of these buildings, all floors and internal walls collapsed; in others, they collapsed only partially. Two other houses, of lesser cultural value, also burned down.

       Individual monuments of importance sustained damage to stone and architectural decorative elements:

  1. St. Blaise Church (balustrade, medallion, statues, stained glass);
  2. the Cathedral (drum, statues);
  3. the Franciscan Convent (cloister: balustrade, columns, facade: library);
  4. the Dominican Convent (south door mouldings and the rose window of the church, cloister facades);
  5. the St. Claire Convent (facades, well); and
  6. the Onofrio Fountain (covering, stonework).
All of these monuments suffered damage of varying degrees to their roofs as well. The beautiful veined limestone of the Stradun was also disfigured by 43 mortar impacts.

       Another type of serious damage, one which will permanently mark the town, was the destruction to the roofs of Dubrovnik. The Institute counted 336 direct impacts on roofs by which 254 roof sections were damaged by shell fragments. In the «Plan preliminaire de financement des travaux d'urgence de reconstruction, de reparation et de restauration pour la vieille ville de Dubrovnik a la suite des bombardements de novembre et de decembre 1991» the UNESCO observers and the Institute estimated that 56,747 square metres of roofs had been damaged. Therefore, thousands of traditional tiles were destroyed and probably will be replaced by modern imitations, since the manufacture of the old type of tiles was discontinued in the middle of the 19th century.

       The surveys by the UNESCO observers and by the Institute carried out immediately after the 6 December bombardment under extremely difficult conditions, may suffer from a tendency to exaggerate the gravity of some impacts on facades and the stability of the greater buildings. However, with the exception of a few cases, little analysis was performed on the effects of vibrations from the detonations on buildings that had been shaken by the earthquake of 1979. In some instances, the amount of damage to some roofs was determined to be worse than originally estimated since surface tiles were often lifted by nearby detonations, falling back on the tiles underneath, and, in consequence, breaking them. This damage, as well as the weakening of the roof structures, was not readily visible, and only became visible at a later stage. To summarize, the hidden damage to the town was worse than the immediately assessed, and the obvious and visible damage was somewhat less in significance.

X. Applicable law

A. The conflict

       The geographical area, which once was the state of Yugoslavia, has been the stage for the use of varying degrees of force during the last two years, and these conflicts have involved parties with differing international status. Although the international status of Croatia during 1991 may reasonably be the subject of academic debate, the discussion to follow will be based upon the premise that the Battle of Dubrovnik was an international armed conflict.

B. The International Criminal Tribunal for the Former Yugoslavia

       In proposing the statute for an International Criminal Tribunal for the former Yugoslavia (ICTFY), the Report of the Secretary-General pursuant to paragraph 2 of Security Council Resolution 808 (1993) of 3 May 1993 (S/25704), required that only rules of international humanitarian law which were «beyond any doubt» part of customary international law were to be applied by the ICTFY. The report referred to the principle of nullum crimen sine lege in justifying this position. Among other international instruments, the report the cited the 1949 Geneva Conventions, the 1907 Hague IV Convention Regulations, and the 1945 Charter of the International Military Tribunal as being included in this group.

       The additional Protocol 1 to the Geneva Conventions (AP 1) is conspicuous by its absence from this group, even though the former state of Yugoslavia ratified the AP 1 on 11 June 1979, and its successor states, Croatia, Serbia and Montenegro are also bound at present by this international obligation in spite of the fact that large portions refer to customary international law. This restrictive approach was adopted to avoid the substantive and procedural problems which might be caused by non-universal adherence to AP 1.

       In regard to the Battle of Dubrovnik, the non-inclusion of AP 1 does not represent a problem since the offences and principles of law set out in the statute of the ICTFY adopted 29 May 1993 satisfactorily cover all the violations identified by the Commission.

C. Offences

       Although the statute of the ICTFY lists many possible offences on which the ICTFY could stand its jurisdiction, only those set out in Article 2(a), 2(d), 2(g) and Article 3(b), 3(c), 3(d) and 3(e) have been determined relevant to the Battle of Dubrovnik, such as:

  1. wilful killing;
  2. extensive destruction and appropriation of property, not justified by military necessity and carried out unlawfully and wantonly;
  3. unlawful confinement of a civilian;
  4. wanton destruction of cities, towns, or villages, or devastation not justified by military necessity;
  5. attack or bombardment, by whatever means, of undefended towns, villages, dwellings, or buildings;
  6. seizure of, destruction, or wilful damage done to institutions dedicated to religion, charity, and education, the arts and sciences, historic monuments, and works of art and science, and
  7. plunder of public or private property

D. Personal and command responsibility

       The offences listed in paragraph 5 of the Statute provide perspective by including principles such as personal responsibility and command responsibility. Article 7 of the Statute sets out these principles and briefly states that one can incur criminal liability for the aforementioned offences: committing the offence oneself, ordering, planning or instigating the commission of an offence, or by failing to take reasonable steps to prevent or deter the commission of an offence by subordinates if one knew, or ought to have known, that subordinates had committed or were about to commit an offence.

       This statement may seem convoluted and unworkable, but it represents nothing more than a restatement of principles common to most domestic criminal law systems and those declared by many war crimes tribunals which sat in the years following the Second World War.

E. Protection of victims and witnesses

       In accordance with Article 15 of the Statute, the need to protect victims and witnesses has also been recognized. This first took the form of Guidelines to the Commission's team for the Conduct of Inquiries and Interviews. Later, this took the form of an interview protocol which was developed by the Dubrovnik team as a way of ensuring absolutely that personal interviews do not violate the principle set out in Article 15 of the Statute.

F. Military objectives and other customary international law principles

       The Statute of the ICTFY cannot hope to make all public international law applicable to armed conflict or even to the prosecution of individuals for grave breaches of this law. The Statute must be seen as an essential mechanism which functions as part of the great body of international humanitarian law for its enforcement through the suppression of grave breaches of the same. For this reason, we must remember that in addition to the Statute of the ICTFY and the international instruments specifically named, the great body of customary international law is also applicable to armed conflict.

       A good way to illustrate this point is to consider a well known rule found in a notable international instrument. An example of this is found in the 1907 Hague IV Regulations which states that the right of opposing parties «to adopt means of injuring the enemy is not unlimited». Considered in isolation, this broad prohibition, by way of a double-negative, may seem to be too vague and abstract to be practically useful. However, this broad principle could be useful and could be made applicable to real armed conflicts with the application of concepts such as «military objective» which have become, over the last one-hundred years, part of customary international law.

       As quoted in the Commission's team Report of the Battle of Sarajevo, «military objectives are those objects which by their nature, location, purpose or use:

  1. make an effective contribution to military action, and
  2. whose total or partial destruction, capture or neutralization, in the circumstances ruling at the time, offers a definite military advantage ...»

       A «military objective» is a lawful target, and the knowledge of this criteria allows combatants and non-combatants alike to determine which kind of opposing forces are likely to attempt to destroy or neutralize. This concept also allows the competent authorities to assess the legality of the conduct of opposing forces and react accordingly. Without this concept, the above quoted rule would just be a legal abstraction. Concepts such as «undefended places or things» or «proportionality» or «neutrality» or «collateral damage» are other examples of concepts which make concrete rules of international humanitarian law.

       These kind of principles complement the Statute of the ICTFY; thus they are part of the law applicable to the Battle of Dubrovnik as well as to the prosecutions initiated as a result of it.

XI. Responsibility for war crimes

       At this stage, based upon the collection and assessment of the evidence on the Battle of Dubrovnik, the Commission is not in a good position to suggest indictments for any individual. This is not because a prima facie case cannot be produced, but further evidence is needed. A prima facie case, however, can be made against certain officers with respect to the incident of 6 December 1991. Names are not included for confidentiality and prosecutorial reasons.

       The problem with the prima facie case against a known officer, whose name is not mentioned for confidentiality and prosecutorial reasons, is that according to the Commission team's knowledge, several of the essential elements in the case were only provided by former JNA Captain Soldo - someone who has a lot to gain if he can pass on responsibility from himself a superior. It may very well be that Soldo's evidence is reliable, but when one considers the enormous amount of raw data relevant to this case which has yet to be evaluated by an independent body such as the Commission's team, and the fact that the evidence on all of the essential elements is not as consistent as one would like, it seems reasonable to take the time to consider all the readily available evidence before preparing the indictment.

       It is also possible that these officers could be linked to other offences in the occupied areas such as the unlawful confinement of civilians and plunder of private property. Clearly, more evidence is necessary to even consider issuing indictments for these additional offences. Cases against the commanding officers and other senior officers can be made and are subject to the same comments as those made above in paragraphs 125 and 126.

       On the last day set aside for the drafting of this report, the team started to receive more information on the JNA command structure during the Battle of Dubrovnik. This enabled the Commission's team to have a more complete picture of the military hierarchy, but the Commission did not have enough time to go back to the records available to try to establish a link with any specific offences. This task must be reserved for the future.

       In order to conduct effective prosecutions to punish violators of international humanitarian law and deter others from the same actions, indictments must be prepared dispassionately after all of the available evidence has been reviewed. The prosecutor will need to know:

  1. what offences occurred;
  2. where the offences occurred;
  3. when the offences occurred;
  4. how the offences occurred (by artillery? by mortars? by small arms? by beating? by burning? etc), and
  5. how the victim was conducting himself or how the building or town/village was being used at the relevant time.

       The prosecutor will then need to correlate that information with reliable information regarding the names of units and unit commanders, as well as details of their activities at the relevant time. Armed with this kind of evidence, the prosecutor will be able to conduct successful prosecutions.

       It should be noted that the local Croatian authorities have been working on their own cases for approximately two years. According to the Dubrovnik Civilian Police, they completed investigations on 127 JNA members and forwarded them to the District Prosecutor's office with recommendations for charges. Out of these 127 cases, 36 are related to offences committed against the Old Town. The Dubrovnik Civilian Police have another 80 JNA members under investigation.

       The local authorities also have a former JNA member, First Lieutenant Ahmet Alasani, in custody pending charges. The local authorities did not specify the charges except that they refer to crimes against Croatia. The Commission team's assessment concludes that local authorities operate under much less stringent rules of proof and evidence than those established by the ICTFY. For example, they include provisions for trials in absentia. Therefore, it is suggested that the statistics regarding these potential prosecutions should not be treated as anything other than interesting facts for assessing the feasibility and prospects of future investigations by the Commission's team.

XII. Recommendations for further investigation

A. Criteria

       As mentioned in previous chapters, there are many avenues along which the initial work of the Commission's team should be followed either by further investigations of events or by reviewing the existing data. Specific incidents should receive priority of attention for any future investigative missions.

       The Commission chose two basic criterion for the selection of specific incidents. The first criterium was determining how close the existing evidence was to establishing a prima facie case of a violation of Articles 2 or 3 of the Statute of the ICTFY against a specific individual or group of individuals. Therefore, cases which are most likely to bear fruit, from a prosecutorial point of view, should be given priority.

       The second criterion for selecting incidents should involve those cases for which, though a strong prosecutorial case did not exist, are considered to be serious in terms of loss of life or health or loss of property. In other words, the most serious breaches of international humanitarian law should be properly investigated even though evidence with respect to them is scant or vague.

B. Crimes against persons

1. Wilful killing

       During the Battle of Dubrovnik, 82 to 88 civilian persons were killed. The circumstances surrounding these deaths appear to be particularly disturbing. The deaths of firemen at the Libertas Hotel in the New Town of Dubrovnik on 6 December 1991 while they were fighting a fire appear to have been the result of a specific target by JNA forces. In that connection, a recording of the JNA radio transmission which gave the order to direct mortar fire on the firemen is in the possession of the Dubrovnik Civil Defence Agency. This case should receive investigative priority.

       Another case of civilian deaths which should receive investigative priority is the apparently deliberate targeting by JNA forces of the hotels in the New Town of Dubrovnik during the fall of 1991. These hotels were filled with thousands of refugees. The JNA forces either had or ought to have had knowledge of their presence.

2. Unlawful confinement or abuse of civilians

       Approximately 5,000 civilians remained in the District of Dubrovnik during its occupation by JNA forces. The Dubrovnik Red Cross and the ICRC have collected information with respect to beatings and other maltreatment of civilians in these occupied areas, and these incidents should be followed up.

       In addition, it appears that several employees of the hydro-electric plant at Plat were unlawfully confined during the occupation of that area.

3. Arrest of protected persons

       Two Red Cross volunteers of a five-member Red Cross team were sent to Cajkovici to collect bodies. They were detained on 26 November 1991 by JNA forces. These volunteers, Muslims from Kosovo, were wearing Red Cross armbands at the time of their arrest. They were arrested by a known JNA officer, whose name is not disclosed for confidentiality and prosecutorial reasons. They have not been seen again, in spite of the efforts made by the Croatian Red Cross. This matter should also be followed up.

C. Crimes against civilian property

       The incidents of wanton destruction and plunder of private property were so numerous and widespread that it is difficult to choose a starting point for further investigation. They must all be dealt with, and the Dubrovnik Civilian Police have already been working for at least one year investigating and documenting these violations by JNA forces. They must all be followed up. No specific recommendation is made, at this stage, with respect to a particular case.

D. Crimes against religious, historical and cultural property

       With respect to the wanton destruction inflicted upon the Old and New Towns of Dubrovnik, immediate attention should be given to the development of cases against certain JNA officers, who are not named for confidentiality and prosecutorial reasons. The cases against all of these JNA officers are close to completion, and have already been the subject of considerable attention by the Croatian authorities. As to essential elements of these cases, such as the identities of the officers, the command structure and the damage to both moveable and immoveable cultural property - both religious and historic - are practically ready for prosecution, and they should be followed up.

       As discussed in Chapters VIII and IX, the wanton destruction of cultural property was not limited to the Old and New Towns of Dubrovnik. The District of Dubrovnik suffered greatly, and was occupied for a considerable period of time. It experienced plunder as well as wanton destruction. The entire District of Dubrovnik should be the subject of further investigation. Starting points in the Primorje region should be the village of Slano and the old Arboretum of Trsteno which suffered both wanton destruction and plunder. It also appears that the village cemetery of Osojnik was subjected to desecration. In the Konavle region, the village of Cilipi and the Franciscan Convent of Pridvorje would be good starting points. These matters should also be the subject of early attention by any future Commission's investigative teams.

       In making these recommendations for further investigative efforts, it is understood that conducting these investigations will involve significant financial and personal resources and that the Commission of Expert's resources are limited. The Commission's team has no solution to this dilemma other than to emphasize how serious the violations of international humanitarian law were during the Battle of Dubrovnik and to state that evidence must be found and promptly secured if the Commission envisages prosecutions. With the passage of time, the physical evidence is most likely to get lost or mislaid. In addition, it is most possible that human memories and the international indignation will fade away.

XIII. Lessons learned

       In all endeavours of this nature, lessons are to be learned. So that future Commission's team investigations may benefit from the experience, and mistakes of the Dubrovnik team, a brief outline recommending a methodology for making future investigations more efficient and effective is provided herewith. It is hoped that these recommendations will become a part of the Commission of Expert's corporate memory.

A. Interpreters

       Independent and professional interpreters are essential to the effective operations of investigations of this kind. Two of the three interpreters used by the Commission's team were provided by the government of Croatia. Although they were charming and well motivated people, the Commission's team often had to discipline them so that they would only interpret and not carry on conversations which effectively excluded the Commission's team members from a conversation. This is a matter of professionalism, and it can normally be dealt with by the team members on the ground.

       This problem was underscored when the team used the interpreter provided by the UNMO Detachment Dubrovnik. This interpreter was conscientious, disciplined and accurate. She also permitted the team to maintain an appearance of impartiality when this was very important (during the interview of former JNA Captain Soldo, for example). Having said this, there were clearly times when having an interpreter who was «connected» to the regime facilitated our investigation. The lesson learned is that independent interpreters are to be sought, but if they are not available, an interpreter who is less than perfectly independent is acceptable, (and advantageous in certain circumstances), as long as the interpreter is professional and disciplined.

B. Pre-deployment preparation

       Much valuable investigatory time was lost once the team was deployed to Dubrovnik because the team had to spend time attempting to secure the co-operation of many local authorities who were prepared to be very bureaucratic in dealing with the team. The amount of time referred to was well beyond that would be dictated by courtesy and protocol, and in the end, the team was still not able to get access to a great deal of crucial, locally held data.

       This could have been avoided if all of the important authorities had been formally contacted in advance so that they could grant the necessary permission and notify their subordinates to expect, and co-operate with, the investigative team. Ideally, this could be accomplished by holding a pre- deployment conference at a convenient location. All of the necessary requests for co-operation could be prepared in draft during such a conference and then forwarded to the Commission of Experts in Geneva or Zagreb for official contact with the relevant authority. This would also be an appropriate occasion for the team members to identify issues and crucial areas and to prepare a very detailed plan of action and to identify and review relevant documentation so that the team would be more efficient and effective once it actually got on the ground.

C. On-site secretarial support

       Although the team members were able to use the personal computer provided by the Commission of Experts and another provided by the Norwegian government in the processing of data and the preparation of this report, the team would clearly have been more efficient if it had some type of dedicated on site secretarial support. It is recommended that this type of support be made available to future investigative teams. If this type of support is not feasible, then each member of the team must have access to a personal computer to avoid the non-productive time which occurred when team members waited for a personal computer to become available.

D. Co-ordination with UNPROFOR and ECMM

       Although the Commission's team received significant support from the UNPROFOR and ECMM personnel individually and from the UNMOs as a group, the team did not receive as much formal support from UNPROFOR and ECMM as it might have. The team would most certainly have been more effective with such support. Therefore, it is recommended that co-ordination be requested at whatever level is necessary to secure the support of these organizations for future investigations.

E. Team members in uniform

       It was readily apparent that being in a military uniform was helpful in some circumstances but a hinderance in others. During this mission three of the four team members would have been entitled to wear a uniform, but only one (the team leader) actually brought one to wear. It is recommended that those future team members who are entitled to wear military uniforms be given the option of doing so according to the circumstances as long as the team leader (if military) is always in uniform.

F. Team composition

       The sheer volume of information available with respect to the Battle of Dubrovnik dictated very early in the deployment that the team would concentrate on an overall assessment of the evidence available with which to prosecute war crimes rather than on the assembling of specific war crimes cases. Therefore, the team was well constituted, with three lawyers and one art historian, to do this. In the future, however, if the mandate of a particular team is to assemble specific war crimes cases and conduct the personal interviews and other associated foot work, the team will have to be more heavily composed of trained investigators such as civilian or military police (at least four). Investigators might be drawn from the civilian or military police of UNPROFOR to conduct these investigations.