The UN has been involved in the conflict in Abkhazia since Georgian forces stormed the Abkhaz parliament in Sukhumi in August 1992, triggering a war that remains unresolved today. In 1993, the UN and the CSCE agreed that the international lead on the conflict in Abkhazia should be taken by the UN, while that in South Ossetia should go to the CSCE.

In the same year the UN, faced with urgent requests from the government of Georgia to deploy a peacekeeping force to Abkhazia, decided to establish an observer mission for Georgia (UNOMIG) to monitor implementation of the July ceasefire agreement between the two sides which had been mediated and guaranteed by the Russian Federation. The decision to send an observer force rather than a fully fledged peacekeeping force reflected the desire of the Russian Federation to take the lead in the management of conflict in the ‘former Soviet space’, and the unwillingness of the other permanent members of the Security Council to challenge Russian prerogatives. There was also a general concern that the peacekeeping apparatus of the UN was overloaded, and disagreement among the parties as to what the mandate of a more substantial force would be. The UN Secretary-General also designated Swiss diplomat Eduard Brunner as Special Envoy for the conflict. He served until 1997 when Liviu Bota, a Romanian diplomat, was appointed Special Representative (SRSG) for the Abkhaz conflict. Both were responsible for the mediation of a process of negotiation leading to a political settlement of the conflict. Bota has had a more or less permanent presence in the conflict zone, whereas Brunner was only delegated to visit intermittently. Russia’s special status in this process was recognised in its designation as ‘facilitator’ of the talks. In the early years of negotiation matters were not helped by the passive attitude taken by the Special Envoy to mediation of the conflict. The UN’s failure to take a more engaged approach was one factor among several contributing to the obvious lack of movement towards a political settlement in 1994–96. The fact that the more proactive approach adopted by Liviu Bota has also not produced a settlement would suggest, however, that the extent of UN activism is not the determining factor in conflict resolution. While the first personnel of UNOMIG were being deployed, the ceasefire collapsed and hostilities resumed. The UN Security Council condemned the renewal of conflict and associated displacement of population and demanded that the parties cease fighting. They also decided to extend the mandate of UNOMIG pending clarification of the situation.

The rapid exodus of displaced persons from Abkhazia, many of whom fled via mountain passes into Svanetia in winter conditions and with no shelter, created a humanitarian emergency, occasioning a second UN response, this time by specialised agencies including UNHCR, WFP and UNICEF. These agencies and their partner NGOs moved quickly to stabilise the situation of the internally displaced. They also assisted in addressing the humanitarian consequences of Georgia’s economic collapse, the product of the country’s multiple wars and the collapse of the Soviet command economy. Over the period 1994–97, UNDHA mounted three consolidated interagency appeals for the Caucasus, with approximately $87 million going to UN agency and NGO activities in Georgia. In the first years of UN involvement in the humanitarian response to the emergency in Georgia, the statist nature of the organisation revealed itself in the exclusion of areas under Abkhaz control from needs assessment and delivery of services by the UN. This omission may have impeded the negotiation of a settlement by enhancing the Abkhaz sense of isolation and creating an appearance of UN bias in favour of Georgia’s central government. In consequence, the ground was left to NGOs such as the ICRC and Médecins Sans Frontières.

UNHCR took a prominent role in early efforts to secure a return of the displaced to Abkhazia. This role was most obvious in the negotiations in 1994 which led to agreement on the deployment of a CIS peacekeeping force to be interposed in a security zone between the two parties and the associated agreement on return of displaced persons. The CISPKF was deployed in mid-1994 and secured the line of contact, but the agreement on return failed miserably with only 311 families out of the estimated 240,000 affected people actually being approved for return. This was largely due to the provision within the agreement that the Abkhaz side had the right for security reasons to vet those returning. The 1994 agreement on return also failed to address the security needs of Georgians returning to Abkhaz-controlled areas. In this respect it was probably a good thing that so few did return. More broadly, UNHCR’s involvement in the process of negotiation raised important questions about whether its status as an advocate for the welfare and rights of IDPs and refugees had been jeopardised by its diplomatic role.

The failure of the 1994 agreement on return did much to poison the atmosphere in talks on a resolution of the conflict. From the Georgian perspective, the Abkhaz behaviour reflected bad faith in the effort to resolve a humanitarian issue that was creating a substantial burden on government resources and contributing to the propensity for instability in Samegrelo. To judge from subsequent events, however, there is some validity in Abkhaz concerns about the security consequences of indiscriminate return.

The deployment of CISPKF resulted in a change in the circumstances of UNOMIG. The observer mission was expanded from 40 to 136 (in early 1999 it stood at 102 from 20 countries), given the task of observing the activities of CISPKF in the security zone and monitoring compliance with provisions for a weapons exclusion zone on both sides of the security zone. The mandates of both CISPKF and UNOMIG included provision for the promotion of conditions conducive to the return of the displaced population. The mandates of both forces have been renewed at six-month intervals since the beginning of their cooperative deployment in 1994. The six-month renewal process has the advantage of bringing the Georgian issue back to the Security Council on a regular basis. On the other hand, reopening the issue invites regular posturing and mutual recrimination by the parties. This does little to further the peace process. Moreover, the CIS has repeatedly failed to renew the mandate of its force in a timely manner, raising doubts about the legal status of the force during periods when the mandate has lapsed and increasing uncertainty and tension on the ground.

It is worth stressing that the interaction between the UN and the CIS in peacekeeping in Georgia is an important example of the sharing of security tasks between the UN and regional organisations. The collaboration has not been easy, although it has improved with time. Russian soldiers deployed to the security zone in the early days were ill suited to peacekeeping, were perceived by both Georgians and UN personnel as lacking impartiality and frequently engaged in harassment of the local population. UN and Russian norms regarding rules of engagement differed markedly. UN personnel were troubled by the corruption evident in some Russian units. Initially, UN observers had difficulty in securing full access to, and freedom of movement in, their areas of operation.

These problems have not disappeared. Accusations by Georgians that CISPKF has been complicit in Abkhaz sweeps through the Gali region, or by Abkhaz that it has failed to prevent the penetration of guerrillas, continue. However, those familiar with the operation generally accept that the Russian performance has improved, in part because the presence of UN personnel and the reasonably close contact between the forces has served as a transmission belt for international peacekeeping norms. On the other hand, the neutrality of the force continues to be questioned by many Georgians, despite the presence of UNOMIG.

The organised return of the displaced failed in 1994, resulting in a gradual process of spontaneous return in 1995–97. This occasioned substantial and repeated violations of the human rights of the civilian population in the Gali region Initially, neither CISPKF nor UNOMIG made any serious effort to prevent these incidents, arguing that the protection of human rights lay outside their mandates. This damaged the credibility and impartiality of both.

After serious human rights violations in 1995, CISPKF and UNOMIG took a more proactive approach to the protection of the returning population, despite the fact that their mandates did not entirely provide for this. Other attempts to address this issue have included the establishment of a joint UN/OSCE Human Rights Office in Sukhumi in 1996 (although it only really became active in 1999) and increased discussion since Spring 1999 about the establishment of a joint investigation unit to explore violations of agreements as well as some cases of criminal activity in the security zone. This has yet to be agreed upon.

The number of returnees to Gali grew gradually through 1996 and into 1997. These returnees were accompanied by guerrilla groups who attacked both CISPKF and Abkhaz personnel. By the spring of 1998 the security situation deteriorated to the point that CISPKF ceased patrolling in the security zone while UNOMIG closed its team bases and concentrated its personnel in Gali and Zugdidi. The Abkhaz de facto authorities then renewed their attacks of the returning Georgian population in May 1998, leading to further mass displacement. This renewed violence might have been prevented or moderated had CISPKF and the UN taken a more proactive stance at this time. The impunity with which the operation was conducted reflects the erosion of the credibility of both the CISKPF and UNOMIG.

Instability also spilled over into Georgian-controlled areas. The UNOMIG sector headquarters in Zugdidi were invaded in February 1998 and several members of the force were kidnapped. This was linked to efforts earlier in the year to assassinate President Shevardnadze and coincided with a number of serious terrorist attacks on government targets in Zugdidi, signalling a general decline in the situation. In July, a UN employee who had previously worked in Sukhumi was murdered in Tbilisi. The problem was not limited to Georgian-held territory. In the summer NGO personnel involved in the demining programme were attacked in Abkhazia. In the autumn, UNOMIG headquarters in Sukhumi was targeted in a series of grenade incidents and three members of the force were wounded during an assault on a UNOMIG vehicle. This series of events led one important contributor to UNOMIG (the United States) to withdraw its personnel and to ban travel by its citizens to Abkhazia.

Ironically, perhaps, the same period was marked by a quickening of the long-stalled peace process, the result of a more proactive role taken by the newly arrived SRSG, Liviu Bota, who initiated the Geneva Process, an intensive series of meetings, among the parties and other interested states and organisations. UN activities were paralleled by a more active Russian diplomacy towards the conflict. The formation of the Group of Friends of the Secretary-General on Georgia and their formal association with the peace talks diluted the dominance of Russia in the process of mediation, as did the participation of OSCE representatives. In August 1997 President Shevardnadze met with the Abkhaz leader Vladislav Ardzinba in Tbilisi for bilateral talks under the patronage of Russian Foreign Minister Evgenii Primakov. By the end of 1997 the Geneva Process had produced agreement to a programme of action on the peace process and the establishment of working groups to address three clusters of issues: non-resumption of hostilities, the return of refugees and IDPs, and economic and social issues. At this time, it was also agreed to establish a Co-ordinating Council for the peace process that would institutionalise the role of the group of friends. Bilateral contacts extended into 1998 with UN facilitation and UNOMIG logistical support, and despite the events in Gali in May. Subsequently, the SRSG organised two further meetings between the parties (in Athens and Istanbul) to push the process forward.

By the autumn of 1998 the two sides had prepared a draft agreement on repatriation of refugees after reiterating their commitment to a non-resumption of hostilities, but Shevardnadze and Ardzinba failed to meet and sign it. The issue of partial return remains unresolved with the two sides still quarrelling over such issues as whether women and children should be allowed to return first, with men of military age being subject to Abkhaz screening, and whether returnees would be obliged to take Abkhaz citizenship. Although UN agencies have assisted returnees since 1995, they have always been ambivalent about spontaneous return without a political settlement, not least as a result of well-founded fears for the security of returnees. The events of May 1998 have deepened this scepticism.

The acceleration on the diplomatic side was accompanied by serious exploration of the possibility of using economic assistance as a means of facilitating a diplomatic resolution of the conflict. In this context UNDP, drawing upon its success with similar activities in South Ossetia, mounted a needs assessment mission to Abkhazia in February 1998. A number of donors, including the USA and the EU, committed several million dollars to reconstruction and other assistance programming in anticipation of agreement between the parties on an approach to the reconstruction of Abkhazia.

Like many others this initiative failed, largely because the developmental objectives of the two parties remained far apart and because they could not agree on implementation. The Georgian side perceived such assistance to be a means of tying Abkhazia in practical terms back into an integrated Georgian economy. The Abkhaz side perceived reconstruction assistance as a way of rendering Abkhazia itself more viable.

The collapse of UNDP’s effort to use economic assistance to push the peace process forward is only one manifestation of the broader fact that there has been little progress towards a political settlement of the Abkhaz conflict. This dismal conclusion is the result of several factors.

It reflects issues and processes over which the UN has little control. Ultimately, the conflict is not yet ripe for resolution: the two sides remain unwilling to accept compromises on the key issue of status. Although the Abkhaz have retreated from the objective of full independence, the two sides remain divided on whether status should be confederal or meaningfully federal, on whether the relationship between Sukhumi and Tbilisi should be horizontal or vertical. Lack of progress on the matter of status prevents movement on other issues such as the return of IDPs and refugees and economic and social questions.

An additional problem is engagement in the conflict by external powers and notably Russia. The Russian Federation played a substantial role in the active phases of the conflict, apparently seeing it as a means of bringing Georgia back into the fold. A complete resolution of the conflict, particularly if this occurred in a negotiating process controlled not by Russia but by the UN would result in a further decline in Russian influence over Georgian politics and policy. There is little doubt that the existence of parallel UN and Russian tracks in the negotiations has impeded the effort to reach a compromise. The existence of a parallel channel has made it easier for the parties to resist concessions. However, given the other obstacles to successful negotiations it is unlikely that this has been a significant determining factor.

The failure of the UN in Abkhazia also reflects problems internal to the organisation, most notably generating consensus at the level of the Security Council. The issue here is that the fundamental interests of the permanent members differ substantially one from another. The Western powers seek conflict resolution, the consolidation of the sovereignty and territorial integrity of the states of the region, and the integration of their economies into a global economy dominated by the West. UN involvement is seen as an instrument in the pursuit of these objectives. Russia, on the other hand, has for much of the post-Soviet period sought to maintain or to re-establish its influence over the Caucasus region and the dependence of the smaller states on Russia. It has claimed special rights and responsibilities in the region on the basis of its preponderance of power and its historical role there. Russia’s control over the northern Caucasus is vulnerable to instability in the Caucasus itself, giving the Russians a more direct security interest in the affairs of Georgia and Azerbaijan. The completion of the Baku–Supsa oil pipeline and the consequent end to the Russian monopoly on oil export from the Caspian basin gives Russia yet another incentive to sustain its influence in Abkhazia. For Russia, a robust UN role might well be a threat to its regional agenda.

This said, it is not clear that UN effectiveness would be dramatically enhanced were Russia to change its policy. The other permanent members of the Security Council have remained unenthusiastic about a more direct UN peacekeeping role, despite Georgia’s apparent desire to replace CISPKF with a genuinely multilateral force. This reflects the general crisis in UN peacekeeping in the post-Somalia, post-Bosnia context. Experiences in these cases and elsewhere suggest that the UN has neither the resources nor the will for robust and effective peace operations in civil wars. In the meantime, the disaster for US forces in the Somali conflict has removed any enthusiasm the Americans have for substantial participation in potentially dangerous UN operations. Nor have any viable regional peacekeeping alternatives appeared on the horizon. Notably, although NATO’s operations in Kosovo have encouraged speculation about the possibility of similar operations in the Caucasus, there is very little likelihood that NATO would oblige.

Although the UN has not delivered peace (and it is not clear whether it could have done so given the attitudes of the two parties and the limited capacities of the UN itself) it has made a positive contribution to the management of the conflict and ceasefire. The presence of UNOMIG personnel in the field enhances transparency and limits the capacity of CISPKF to pursue a unilateral agenda in the conflict zone. It has probably had some effect in improving the security of civilians in zones patrolled by the mission, but has not curtailed the persistent, if low, level of violence and criminality in the region. UNOMIG has been of great use in facilitating humanitarian assistance in the Gali region and elsewhere in Abkhazia by providing a modicum of security of movement in often quite dangerous circumstances. More generally, the UN and other international agencies were instrumental in preventing what otherwise might have been a complete meltdown of Georgia and total collapse of order within its borders.

The presence of the UN in the early days also reduced the sense of isolation and desperation on the Georgian side. Although the lack of involvement by UN humanitarian agencies on the Abkhaz side in the first years of the conflict may have had the opposite effect in Abkhazia, since 1996 the specialised agencies have made a conscious effort to pursue proportionality in its delivery of assistance to both sides.