Many Tajikistanis share a belief, perhaps more emotional than rational, that their country enjoyed a state of 'national consolidation' prior to independence. They assume that irresponsible politicians damaged this consolidation and thrust the country into the flames of civil war. From this assumption follows the conviction that because the war is over, national unity and consolidation have been restored and will now be sustainable and permanent. Yet it is unrealistic to assume that national consolidation is something that is achieved once and forever. It is, rather, the outgrowth of a continuous process of developing timely and appropriate solutions to problems and contradictions. These solutions may not satisfy everyone all the time, but taken as a whole they must address the specific interests of the different social strata, political groupings, and regional concerns. The process of reaching these solutions requires compromise between different interests. National consolidation can therefore be understood as a permanent process of finding and maintaining compromises on as many as possible of the specific problems and disagreements that arise in the economic, social and political development of a country. If politicians from both ruling and opposition parties can achieve these compromises, then national accord and political stability can be secured. If the spirit and practice of compromise is not maintained, tensions will escalate and politics will radicalise, thus generating a crisis.

These dynamics must be taken into account when considering the peacemaking and peacebuilding process in Tajikistan. Political elites in the late 1980s and early 1990s failed to find suitable strategies to address a complex set of problems related to inter-regional contradictions. These included disparities that had arisen from the increasing economic role of southern regions, the demographic structure of the population, ideological diversification, and unequal participation in political decision-making in a country dominated by a northern political elite. Thus the war did not actually cause national disintegration. Rather it followed from the unresolved confrontation among various political forces in 1990-92 and was a manifestation and aggravation of divisions that already existed within Tajikistani society. Yet even in the depths of war, the combatants never fundamentally questioned the legitimacy of Tajikistan's post-Soviet statehood and all its institutions. Theirs was a fight for control of the state and its institutions.

By 1994, it was clear that there was no military 'solution' to the conflict. All the factions began to realise that continued fighting posed a profound threat to the existence of the Tajik state and the people's survival as a nation. This article will explore the structures and opportunities the peace process has provided for Tajikistani leaders – in various roles and representing various interests – to practice compromise. Through the practice of political dialogue and decision-making, they have re-established a unified government for the country. This challenge has continued into the transitional phase and beyond.

The CNR was the mechanism created to oversee implementation of the General Agreement and to design a set of reforms to the government structure. The Commission included representatives of both the government and the United Tajik Opposition (UTO). Because most opposition leaders were in exile when the Agreement was signed, the CNR's first meeting was held in Moscow shortly after the signing ceremony. It began work in Dushanbe in September 1997, after the UTO members returned to Tajikistan. Both the beginning of CNR meetings and the fact that prominent exiles had come home marked important symbolic turning points for post-conflict peacebuilding. The initial working meetings focused on organisational problems. UTO leader Said Abdullo Nuri was elected Chair of the CNR and Abdulmajid Dostiev, then First Deputy Chair of the Parliament, was elected as his deputy. Under their leadership, the Commission negotiated its work plan and schedule. The CNR had four subcommissions on refugee-related, military, political and legal issues. Mirroring the power-sharing principles of the Agreement, these subcommissions were chaired respectively by Shukurjon Zuhurov (government), Habib Sanginov (UTO), Ibrohim Usmonov (government), and Otakhon Latifi (UTO). With its organisational structure settled, the CNR began work on full implementation of the General Agreement.

The most rapid progress in the initial period was with implementation of the Protocol on Refugee Issues. Within three to four months – with the intensive participation of the UN Mission of Observers in Tajikistan (UNMOT), the UN High Commissioner on Refugees (UNHCR), the Russian Border Forces, the Commonwealth of Independent States Collective Peacekeeping Force (CIS/PKF) and other international agencies – tens of thousands of refugees returned home from Afghanistan and CIS countries. They were permanently resettled and provided with shelter and other necessities. This required large-scale restoration of houses damaged or destroyed by war and the return of illegally occupied properties to their owners, as well as social rehabilitation and reintegration of the erstwhile refugees.

The comparative success of the implementation of the Protocol on Refugees was due to the 'push and pull' factors inherent in the changing political conditions brought about by the peace process and the altered circumstances in Afghanistan, combined with a well-resourced return programme. After the General Agreement was signed, none of the parties had an interest in keeping large numbers of Tajikistanis in exile, particularly in Afghanistan where most refugees had fled. The government had long maintained that it prioritised the return of refugees – in part, perhaps, to prevent their recruitment as UTO fighters. It therefore welcomed the international attention (and resources) given to their return. Before the agreement was finalised, the UTO – aware that the refugee communities provided a power-base and also concerned for their safety and security in Tajikistan – had been reluctant to encourage their return.

All this changed when the military confrontation ended. The peace agreement, backed by international guarantees in combination with improved conditions in Tajikistan, encouraged refugees to return home. The UTO also had an interest in their return because they were likely to strengthen its domestic support base. At a more localised level, returning refugees also brought much-needed labour power, which could be deployed in depleted agricultural and industrial enterprises. A significant 'push' factor for those who had found refuge in Afghanistan was the escalation of war in the north of that country with the Taliban's military offensive in the second half of 1997. This encouraged many to seek safety in Tajikistan. The combined effect of these political, security, economic and humanitarian considerations was a consensus on the need to implement the refugee protocol rapidly. Return programmes were therefore implemented with minimal controversy between the leaders of the parties to the peace process.

Just as significant in character, content and scale were the results of the implementation of the Military Protocol. The agreement aimed both to integrate Tajikistan's many armed forces into a unified military and to promote decommissioning and demobilisation. Military cooperation between the parties began before the CNR started operations. They first reached agreement on a ceasefire in the Tehran Agreement of September 1994, which provided for a Joint Commission (JC) of government and UTO representatives to monitor implementation. The experience of working together helped members to progress from barely tolerating their political and military opponents to being able to work together to develop effective solutions to practical problems. Members of the JC, although representing different factions, formed unified teams to negotiate with commanders from the different sides. They often worked in potentially life-threatening circumstances to investigate and prevent violations of the Tehran Agreement. In so doing, many were not only expressing the interests of their own parties but also promoting the interests of the country as a whole. This experience of joint problem-solving between 1994 and 1997created both the mechanisms and habits of working cooperatively and served as a precedent for the work of the CNR.

Implementation of the Military Protocol centred on the deployment of UTO fighters into ten officially established assembly areas where they could be registered for further integration into the Tajik military and other security services. One CNR priority was to draft laws for a general amnesty, which created the legal framework for releasing opposition members from prison and granting amnesty to more than 5,000 UTO fighters. Between July and November 1998, all UTO fighters in Afghanistan were relocated to Tajikistan through the combined efforts of the government, the UTO, the Russian Border Forces and the CIS peacekeepers. At the end of 1998, the UTO leadership announced the closure of all its military training camps abroad.

The next stage was the gradual integration of UTO troops into the national armed forces, partly through the establishment of provisional military units composed of UTO fighters. From early September 1997, one unit of UTO fighters was deployed in Dushanbe. It later became the 25th Battalion. By March 2000 – when CNR activity was completed – 4,498 UTO fighters had been integrated into the armed forces, mostly within their own units. The strategy of retaining separate units was used in the early stages of unification to reduce the potential for tensions between people who had been fighting each other. The greatest levels of integration were achieved in the Committee for the Protection of State Borders because protection of Tajikistan's borders from armed intrusion by foreign forces was a common objective of both the government and the UTO.

While there have been successes in implementing the military protocol, the process was often difficult. It tested the capacity of the political leadership on both sides to control their own military commanders. During the war, the power base of the commanding officers from both sides was not derived from politicians but rather from local people who considered the military leaders as their protectors. The command and control structures were often weak and political leaders had few points of leverage over military commanders. Thus the ending of military confrontation was made possible also by the direct negotiations amongst the commanders themselves. For example, representatives of the government's forces made contact with UTO commanders in the Qarateghin valley in September 1996 and signed the Gharm Protocol. This agreement gave a form of legal recognition to UTO field commanders that signalled the government's acknowledgement of them as partners in the peace process. In turn, these commanders sent a cable to the negotiators at the December 1996 meeting in Moscow and politely but firmly demanded that the negotiators continue their efforts to reaching an agreement. Their action had a positive effect on the UTO delegation's position.

Perhaps the most important condition for the successful implementation of the Military Protocol was the cooperation of the field commanders. The latter were never merely the followers of political leaders; they were a political power in themselves. Most had strong and charismatic personalities and were able to form armed units from people they recruited. They typically enjoyed the full loyalty of those who served with them, as well as the support of the population in the territories they controlled. When these commanders agreed to the peace treaty it was not difficult to convince their fighters to follow, particularly because many were ready for the war to end. Many engaged in direct contacts with their counterparts, which enabled them to gradually develop sufficient trust to address outstanding differences through negotiation rather than armed confrontation. Trust was facilitated by the tendency of military personnel to accord respect to other commanders, in contrast to a shared suspicion of politicians. Cooperation was furthered by the recognition that they could gain more political, social and economic benefits in peacetime than in war. The improving relationship between former military opponents gave further impetus for politicians to continue implementing the General Agreement in order to preserve their political control over the peace process.

Although most UTO commanders supported the peace process, not all were fully content with the practical outcomes. Causes of dissatisfaction included delays in implementing decisions on government appointments, restoring pre-war jobs, closing criminal cases against commanders and fighters to fulfil the amnesty and unreliable distribution of supplies to fighters in assembly areas. In some cases, dissatisfied commanders balked at the integration arrangements and tried to preserve their autonomy. Most such problems were settled by the efforts of senior working groups of CNR members and representatives of the national Security Council and the armed forces. In a few cases, negotiation strategies were combined with the use or threat of armed force to subdue recalcitrant commanders. For example, a decisive test for the newly integrated armed forces occurred in November 1998, when former government and UTO fighters combined to defeat an uprising by the formerly pro-government Colonel Mahmud Khudoiberdyev and his militia in the Leninabad region.

Compared with the protocols on refugee and military issues, there were many difficulties with implementing the Political Protocol, which was based on the principle that the former combatant parties would share government power. Under the agreement, 30 per cent of government positions were reserved for UTO representatives, with specific arrangements for ministerial positions. The first set of difficulties coalesced around the integration of some prominent UTO figures. There had been an initial agreement that the UTO's First Deputy Chair Khoji Akbar Turajonzoda would become Tajikistan's first vice premier and that UTO Commander-in-chief Mirzo Ziyoev would head one of the 'power ministries'. Implementation of these agreements was delayed by a combination of factors, including opposition within the government. Considerable international pressure was applied to the parties to reach a settlement on these sensitive issues. Rakhmonov and Nuri exercised their considerable political skill to reach an acceptable compromise. Turajonzoda was allowed to return to the country to assume the post of first vice-premier and Ziyoev was promoted to the rank of major-general and appointed head of the newly established Ministry of Emergency Situations. Other political appointments followed and, by the time the CNR's work was concluded, thirteen UTO representatives had been appointed as the heads of ministries and important committees. It was, however, more difficult to achieve the full 30 per cent allocation of UTO representatives in the second-tier positions of government.

Many anxieties surrounded the creation of the CNR and the return of opposition leaders from exile. Some government supporters feared that the return of UTO politicians would strengthen the opposition – and particularly the Islamicists – at Rakhmonov's expense. Some feared that the UTO leadership would attempt to impose a lifestyle similar to that in post-revolutionary Iran. Others feared that they would have to leave their positions in the public and private sectors to make room for the newcomers and returnees. Over time, however, many of these anxieties began to fade as they were proved groundless.

The CNR also had considerable powers to draft constitutional amendments and new legislation. It drafted reforms that enabled former opposition parties – and particularly those of a religious character – to hold their congresses openly and participate in multi-party elections. After the UTO leadership announced the dissolution of its military formations in August 1999, the Supreme Court lifted the ban on UTO member parties and organisations that had been imposed in June 1993. With the presidential elections in late 1999 and parliamentary elections in 2000, the transition period was officially regarded as successfully concluded.

Perhaps one of the most significant achievements of the transition period was that it demonstrated that the fears many had harboured about opposing parties had little rational basis. It also demonstrated that power-sharing – if it means sharing responsibilities for developing solutions to real political, economic, and social problems – can transform even the most radical people into pragmatists with greatly moderated views, thoughts and practices. Many CNR members had participated in the second-track Inter-Tajik Dialogue. Through this experience, they had improved their negotiating skills with an emphasis on trying to understand the positions of others and to seek integrative solutions to problems. It can be argued that the outcomes of implementing the General Agreement were related only minimally to its quality as a document. The determining factor in practice was the very political process of developing consensus between parties on a range of concrete issues and problems. Yet the CNR was only one of the necessary elements in the process.

The efforts of leading political and military actors to implement the peace agreement were underpinned by two significant sources of pressure. First, the Tajikistani population as a whole was convinced of its necessity. This was echoed by the realisation among most party supporters and fighters that they could only translate their political, ideological, and economic interests and aims into reality in a sovereign state. Most Tajikistanis believed that the country could retain its independence only through further development of the peace process. A second source of pressure was the array of international actors and foreign governments interested in the successful implementation of the peace agreement. Their support was institutionalised in the role of the UN (and UNMOT in particular) and the Contact Group in monitoring implementation and providing analysis and recommendations to overcome obstacles. With this encouragement, both Rakhmonov and Nuri were able to translate the aspiration for peace into political mechanisms and policies. The core element in making these mechanisms work was their intensive bilateral contacts – including tête-à-tête meetings and regular political consultations – which enabled them to work through some of the most complex problems and crises.

Three significant crises threatened the transition period. The first major crisis was in spring 1998, when armed clashes erupted between pro-government and UTO troops, first in Kofarnihon and then in eastern Dushanbe, resulting in both combatant and civilian deaths. These skirmishes seemed to be a carry-over from the war years, when the main method for settling problems was armed force. Yet the fighting shocked many. Importantly, leaders of the parties in the peace process recognised that the temptation to resort to military-political pressure could escalate to the point where they would lose control as more radical elements gained influence. They also realised that the failure of the peace agreement would jeopardise the support provided by the international community for rehabilitation, reconstruction and development.

The next crisis, from April to June 1999, was more explicitly political. It was sparked by delays in amending the constitution, extending amnesty to UTO supporters and closing criminal cases, incorporating UTO representatives into government and lifting the ban on the activities of UTO member parties. The UTO responded by suspending its activities in the CNR for a period. Both sides issued statements and counter-statements about the problems and their positions. Yet while this unfolded, cooperation in the day-to-day governance of the country continued and the UTO never threatened to suspend its participation or withdraw its representatives. The crisis was ended through a political settlement addressing the contentious issues. While the Tajik leaders did not waiver in their commitment to a negotiated approach to the crisis, their efforts were facilitated by the active mediation efforts of UNMOT, the Contact Group of guarantor states, and the OSCE.

In autumn 1999, crisis loomed once again in the run-up to the presidential elections scheduled for 6 November. UTO member parties claimed that the government had set unequal conditions for both the nomination and registration of candidates for the presidency and for their pre-election campaign. In response, they threatened to boycott the elections. This crisis was also settled with the help of international mediators. The brinkmanship continued until Nuri and Rakhmonov signed an agreement on the night before the election. This agreement provided guarantees for free and unimpeded participation for all registered parties in the parliamentary elections in February 2000, which went relatively smoothly.

Attitudes changed through these experiences of crisis management. Leaders began to recognise that the threat of breakdown was quite normal in the complicated and difficult transition from civil war to peace. Bottlenecks are inevitable. Crises clearly indicate the problems and force the parties to take measures to address them. This awareness underscored the new approach, in which the main form of political struggle became the dissemination of position statements describing problems, rather than the distribution of arms. As a rule, these statements included the important caveat that they were intended primarily to promote development of the peace process rather than undermine it – thus leading to a normalisation of politics as the method for conflict management.

The combined consequences of the civil war and its peace process were effective in resolving the contradictions that had led to the political crisis of 1990-92 and the civil war of 1992-97. Yet new contradictions have emerged in the course of economic, political and social development. Several key ones can be identified. First is the disjuncture between the economic power of the northern region – which has adapted more quickly to the market economy – and the inadequate representation of the region's elites in decision-making. During the transition period, most of the new modernising elites from Leninabad concentrated on consolidating their position within the region. But in the future they will probably begin to demand more effective representation in central government. Another potential contradiction is emerging from the rapid evolution of the presidency as the dominant political institution and the comparatively weak development of civic society. A third dilemma arises from the emergence of new opposition interests and the political forces that claim to represent them, for example the Islamicist Hizbu-t-tahrir (Party of Liberation). Although many in the government tend to perceive such new political movements as representing no more than the ambition of individuals, there are good reasons to believe that they arise from sincerely-held beliefs and objective interests rooted in the development process of the past decade.

If the ruling elite ignores the fundamental causes of these contradictions, there is a risk that they could become the source of protracted conflict. If it ignores the first set of contradictions around regional disparities, the country could return to the dynamics that prevailed on the eve of political cataclysm in the early 1990s. Insufficient attention to the second set weakens democratic development and risks confrontation with international institutions and western governments. Attempts to address the third set of problems through suppression, rather than by addressing the causes of opposition, could stimulate resistance and risk political destabilisation. Future prospects for sustainable peace may depend on whether the lessons learned and experience gained through the peace process by this generation of political leaders will be used to address continuing problems and contradictions.