Almost from the start of the Nagorno-Karabakh conflict, outside or third parties have sought to mediate the conflict, but the conflict remains unresolved. And that, in turn, has sparked a debate, sometimes intense and sometimes less so, as to whether such media contributes to conflict resolution or in fact keeps the conflict alive.
From the outset, the conflict was perceived as important both by regional powers and international organizations. From 1992 to 1994, when the former Soviet republics joined the CSCE and a ceasefire was signed, the CSCE intervened as a third party mediator. In March 1992, the CSCE Council of Ministers held an emergency meeting in Helsinki where they designated the organization to be the main instrument in settling the conflict. A committee  was formed to find a peaceful solution to the conflict. However, its members could not even hold a conference to assess the issues arising from Minsk. From that time on, the OSCE negotiation group took the name “the Minsk group.”
The Minsk group had several obvious shortcomings: “the CSCE’s lack of experience with these type of conflicts and reduced solidarity among its members, combined with Russia’s regional ambitions and Turkey’s advocacy role serve to weaken the intervention…” (Mooradian & Druckman 1999, p. 710). The CSCE’s role was relatively weak compared to Russia’s and it, in fact, made it difficult for them even to create an agenda for a potential meeting. At that time, there was fairly high level Western interest in the Nagorno-Karabakh conflict, and it was “this high level American intervention” that led to the creation of the Minsk group (Maresca 1996, p. 260). At the beginning of 1993, in an attempt to restart the stalled CSCE negotiations, Russia, Turkey and the United States held a series of private discussions resulting in what became known as the “3+1 initiative.” 
According to their proposal, Armenian forces would withdraw from Kalbajar in two months; this term would be accepted as ceasefire and during this time new peace talks would start. Azerbaijan and Armenia accepted the plan. However, Armenia declared some of its concerns about the Karabakh Armenians’ point of view. In May 1993, Karabakh officials rejected proposals due to their lack of “guarantees for the Karabakh Armenians and the proposal’s exclusion of the elimination of the embargo put on Karabakh by Azerbaijan” (Lutem 2001, p. 14). Following and despite the latter, in October 1992, the US Congress passed the Freedom Supports Act 907a, which denied Azerbaijan all forms of governmental US aid, unless it respects international human rights standards, abandons its blockade of Armenia and Nagorno-Karabakh, and searches a peaceful solution to the conflict (Cornell 1997, p. 10). While Armenia was on the top of the list of the countries that received US aid in the region, Azerbaijan was left out of US aid distribution. US foreign policy towards the conflict favored Armenia until Caspian oil came on the scene.
In July 1993, the UN Security Council passed Resolution 853 regarding the Armenian-Azerbaijani fighting. The Resolution condemned seizure of Aghdam and called on all parties to cease supplying weapons to the belligerent sides, especially to the Armenians, because military assistance by third parties, especially by Armenia and Russia, intensified the fighting and resulted in the continued occupation of Azerbaijani territories. The resolution also called on Armenia to use its influence with the Nagorno-Karabakh authorities to force them into compliance with the UN Resolutions and Minsk Group initiatives. The resolution also called for the lifting of all economic and energy blockades in the region (Gurbuz 2003, p. 55). However, Karabakh authorities rejected the resolution.
From 1993 to December 1994, Russia was influential in brokering peace in the region. Russian diplomats and Russia’s special envoy to the Minsk Group employed shuttle diplomacy between the conflicting parties. However, in contrast to other contributing peace efforts in the region, Russia undermined the CSCE’s role and caused confusion among the parties and other mediators.
The Commonwealth of Independent States (CIS), as a peacemaking structure entered the mediating process when Azerbaijan accepted the invitation to enter the CIS Tashkent Agreement on security in May 1992. Armenia was already a member and the CIS became a legitimate peace building force when Russia declared that OSCE could not even secure a cease-fire in the region (Barseghyan & Karaev 2004, p. 10). The CSCE also accepted the CIS participation in the peace process. Finally in 1994, the CIS, Russia and the CSCE managed to convince the parties to sign a cease-fire agreement and terminate violence, although most of the disputes between the parties stayed unsolved. After the cessation of hostilities, Armenia and Azerbaijan entered a state of frozen conflict, in which mass violence had ended, but the political dispute was unresolved (Waal 2003, p. 251).
In Budapest in 1994,  the OSCE held a summit and appointed co-chairs to lead the Minsk Group. Until 1995, Russia and Sweden served as the co-chairs. The Nagorno-Karabakh intervention marked the first time in history that the OSCE committed itself to resolving a conflict as a mediator in a peace conference (Mooradian 1998, p. 6). The OSCE failed to bring parties to any preliminary statement of general principles, to guide the process of conflict settlement (Barseghyan & Karaev 2004, p. 10). At the Lisbon Summit in 1996, the main peacemaking proposal in NK was accepted by all the participating countries except Armenia.  Since the OSCE resolution supported territorial integrity of Azerbaijan, the Armenian side found it unacceptable.
In 1995, Sweden vacated its co-chair position to Finland and when Finland ceased to be the co-chair with Russia in 1997, a French representative was appointed to succeed the Finish position. This initiated opposition from the Azerbaijani side, which requested an American representative instead. The dispute was resolved by appointing an American as third co-chair and since then this troika has not changed.
Unfortunately, Russian-OSCE coherence problems did not end after the ceasefire. Until the time when Swedish co-chairs left their position to Finland, Jan Eliasson’s efforts of shuttle diplomacy were challenged by Russian diplomacy. Since that time, the Minsk group has come up with several proposals. Those proposals have been rejected once by Azerbaijan and twice by Armenia, a development that led OSCE to change its negotiation strategy in 1999 and to give greater priority to face-to-face meetings between Azerbaijani and Armenian officials. 
In March 2002, the Minsk group co-chairs offered another peace plan to the representatives of Azerbaijan and Armenia. The talks were held in privacy and the plan was not disclosed to the public. Following a decade of fruitless talks, a new format of the meetings, the Prague Process, involving direct bilateral contact between the foreign ministers of Armenia and Azerbaijan was initiated in 2004. Two years later, the Minsk Group declared that 2006 could be the “ripe moment” for a settlement. The mediators tried to convince Azerbaijan to accept a referendum just within NK, but Baku rejected that. It called for a nationwide Azerbaijani referendum on the final status of NK, but Yerevan in turn rejected that.
Because of this experience, many analysts agree that mediation efforts in NK have turned out to be a complete fiasco and that these efforts themselves have prolonged the conflict (e.g. Betts 1999; Mooradian 1998). Moreover, from the beginning, the OSCE has been criticized by both Baku and Yerevan on many grounds. The co-chairs, in response, have laid the blame on the parties, but there clearly are several aspects of their activity and structure that hindered the peace process. This concerns, in particular, the specific views of the co-chair countries and their biases, real and perceived, for one or another side; something that has limited their ability to influence both parties.
It can be said that the mediators were successful, when they acted as communicators, still they could not change the perceptions of the parties from zero sum game to win-win solutions. The mediators, however, lacked the capacity of maneuvering and formulation. At the same time, the mediators also could not afford to withdraw or terminate the mediation efforts, since their national interests were at stake and they did not want another mediator to undertake initiation. One way or another, no peace plan was fully acceptable to both of the warring parties, and the lack of leverage gave them room to avoid pressure. The OSCE Minsk Group’s impartiality, in turn, has been a matter of dispute for a long time.
Several conclusions follow from the above. Above all, there needs to be a change in mentality; that change is more than necessary both among the parties and the mediators. The parties need to understand compromise is inevitable for a peaceful settlement, and the mediators need to change their strategies, which seem to have failed up until now. They must pay more attention to the roots of the conflict and try to better understand why the parties are not willing to surrender certain principles. Now, more than 20 years on, Armenia’s and Azerbaijan’s demands are mutually exclusive and the international community’s approach is usually divided. As a result, peace remains a long way off.
If the two parties are not ready, there will not be a settlement. The third party mediators may come up with more effective peace proposals and more innovative resolution packages, but they cannot dictate a settlement to the parties. Expecting a miracle from the third parties is not rational, but without their involvement the immediate parties are unlikely to be able to find a resolution on their own.
Bahar Başer, Dr.
Postdoctoral Research Fellow
Politics & International Studies
University of Warwick