The geopolitics of armed conflict settlement : How did Russia affect the settlement of the 1991-1994 Nagorno-Karabakh war?

The Caucasus region is the cradle of several protracted ethnic conflicts. One example of this Caucasusturmoil is the Nagorno-Karabakh ethno-secessionist conflict which opposes Armenians and Azerbaijanis. Nagorno-Karabakh (NK) is a region within the state of Azerbaijan which is today inhabited by around 95% of ethnic Armenians1 who seek to secede from Azerbaijan with the backing of the neighbouring Republic of Armenia2. It can thus be characterised both as an internal and an international conflict (between NK and Azerbaijan as well as between Armenia and Azerbaijan). It is possible to date back the conflict to 1922 when Stalin, under a divide-and-rule logic, arbitrarily assigned the region of NK to the Soviet Republic of Azerbaijan. With the weakening of the USSR, the ethnically-driven aspirations of NK re-emerged in 1988 when the Armenian delegates of the NK parliament petitioned Moscow in order to be reunited with Armenia. The conflict escalated to a full-scale war in 1991, the same year NK unilaterally declared itself an independent state. Russian mediation attempts thus started in 1991 and led to the signature of many unsuccessful ceasefires (Yerasimos 1994, 69). It was only in May 1994 that parties to the conflict signed the Bishkek protocol which led to the still-in-vigour 1994 ceasefire. Although the roots of the discord still remain unresolved and parties are still negotiating to find an issue to this protracted conflict, the 1994 ceasefire successfully put an end to the war which had cost the lives of around 25,000 people (Mooradian; Druckman 1999, 712). As explained by the current Armenian Ambassador to the United Kingdom, although the conflict is still frozen today, at least parties don’t shoot each other anymore3. The Secretary of Security Council of NK, Karen Baburian, also stated in 2001 that ‘the Bishkek Protocol was the only real achievement in the Karabakh settlement’4. This essay will thus explore the reasons of the success of the 1994 ceasefire and will analyse the particular role of the Russian mediation. Through a chronological analysis of the war, this essay will show that although the ceasefire was brokered by Russia, Russian mediation during most of the war was actuallyinhibiting the likeliness of a conflict settlement. The acceptance of the latter is seen as a result of the mutually hurtful stalemate in which parties found each other at the time.

It is possible to understand Russia’s involvement in the conflict using the archives of the New York Times. When the conflict re-emerged in 1988 with the weakening of the USSR, the three primary actors in the conflict were still subordinated to Moscow’s rule. Moscow refused to transfer the region of NK to the Republic of Armenia in 1988 and despite its promises of economic and social aid, NK’s aspirations didn’t vanish. With the escalation of the conflict, Moscow sent its troops to patrol in the region and toreduce the scale of theviolence. Because the situation was getting out of hands, Moscow decided to put the region of NK under its direct control in January 1989. After 11 months, NK’s control was transferred back to Azerbaijan which led to the discontent of Armenians in both Armenia proper and NK. Azerbaijan was also clearly unsatisfied because the Observation Commission which remained in NK meant that Moscow was still keeping some sort of control over the region. During the period in which the USSR was still in place, Russia tried to stop the conflict both through the use of its military and inviting presidents of Armenia and Azerbaijan to negotiate. However, all attempts of the Soviet central government to reduce the scale of the conflict were being totally unsuccessful as violence only escalated. The reason is that each party in the conflict had hopes of winning over the other through fighting and thus had hopes ofbeing better off through violence.

From 1991 onwards, several important changes occurred. The conflict escalated to a full-scale war and with the expectation of a soon coming dismantlement of the Soviet Union, the first mediation attempts of Russian President Yeltsin with Kazakh President Nazarbayev took place in September 1991 (Mooradian; Druckman 1999, 710). After the Soviet Union collapsed, former Soviet troops in the region refused to intervene in the conflict and finally decided to leave the region in 19925. The collapse of the USSR thus shows how Russia had to change its strategy to put an end to the conflict. With Armenia and Azerbaijan becoming independent states, Russia had to shift from military intervention to diplomatic mediation. For the first mediation attempt of Russia, President Boris Yeltsin travelled to Stepanakert (the capital of NK) on a peace-keeping mission. After the negotiations held in Russia between Armenia and Azerbaijan and with the presence of a NK delegation, a joint communiqué was issued6. However, this first mediation attempt was unsuccessful as violence retrieved right after. Several reasons can explain why the strategies of the Russian mediation could not lead to a successful settlement. First of all, although NK delegates were present during the talks, they were not contracting parties to the ceasefire in 1991 (Yerasimos 1994, 69-70). This was due to the pressure of Azerbaijan which didn’t want to recognise NK as a party to the conflict. It feared that recognition might grant legitimacy to NK’s independence movement (Svensson 2009, 11). This non-recognition of NK as a contracting party made that NK forces had no incentive to respect the ceasefire as it was not binding to them.Russia thus did not put enough pressure on Azerbaijan to negotiate directly with NK and, considering Russia’s internal affairs, it is possible to provide an explanation. Russia was clearly in favour of territorial integrity as the contrary would have had negative effects domestically (Baser 2008, 107). Russia’s own internal secessionist conflicts impeded it from backing the recognition of NK as this would have granted legitimacy to Chechnya, for example, which had unilaterally declared its independence from Russia in 1991 as well7. From the perspective of NK at the time, it knew that if it respected the ceasefire, Azerbaijan with the backing of Russia would concede nothing to NK which would not be recognised as an independent state. Consequently, it believed that it would be better off if it continued fighting in order to win over Azerbaijan and impose its secession through violence. Also, as the personal agenda of Yeltsin was actually more important than bringing peace into the region, there was no real commitment from him to de-escalate the conflict and his attention was turned elsewhere when more important national issues appeared on his agenda (Mooradian; Druckman 1999, 710).

In 1992, Russia’s Foreign Minister Andrei Kozyrev invited the Foreign Ministers of Armenia and Azerbaijan to Moscow where he moderated the talks. After eight hours of talks, the Armenian and Azeri Foreign Ministers newly called for a truce without the agreement of NK and without its presence during the talks8. The talks thus didn’t help de-escalate the conflict although President Yeltsin also held separate individual talks with President of Armenia Levon Ter-Petrossian in Moscow. In March 1992, the OSCE also started intervening in the conflict as a mediator and soon became known as the Minsk Group. Its mission since then has been to provide a political solution and a peaceful settlement to the conflict without much success9.Furthermore, from 1992 to 1993, Armenian forces started occupying Azeri territories. They occupied the territories which separated Armenia proper from Nagorno-Karabakh in order to sidestep Azerbaijan’s blockade against NK. Territories such as the Lachin corridor as well as the Kelbajar region unofficially moved under NK administration (Yerasimos 1994, 79-80). One interesting fact that can explain why the mediation didn’t manage to improve the situation in the region is that the parties in the conflict rejected to a certain extent the mediation held by Moscow. From the perspective of NK, any solution which supported the principle of territorial integrity was firmly rejected and no compromise was sought or accepted (Baser 2008, 91). From the perspective of Azerbaijan, although Russia was in favour of territorial integrity, it saw Russia as a biased mediator towards Armenia. Since Armenia was Russia’s main strategic ally in the Caucasus, Azerbaijan felt that Russia didn’t put enough pressure on Armenia and NK to stop the conflict and prevent the occupation ofAzeri territories by Armenian forcesand it interpreted the great bilateral economic relations between Russia and Armenia as a threat (Baser 2008, 107). Azerbaijan also accused Russia of having provided a logistical help to Armenian forces in order to occupy Kelbajar. It suspected that this was a strategy from Russia to put pressure on Azerbaijan to join the Commonwealth of Independent States (Yerasimos 1994, 79-80). Consequently, this made Azerbaijan reluctant to accept Russia’s mediation on the grounds that it was not being impartial.

Later at the beginning of 1993, Russia together with Turkey and the USA started private discussions and made different proposals to Armenia and Azerbaijan in order to find anissue to the war. They insisted that Armenian forces leavethe region of Kelbajar in exchange of the recognition of the autonomy of NK and they wanted parties to conclude a ceasefire in order to start peace talks. Although Armenia and Azerbaijan accepted the peace plan, it was again ineffective (Yerasimos 1994, 80). The problem was the fact that NK rejected the proposal and this highlighted the obstacle of not recognising NK as a contracting party since its positions started diverging from those of the Republic of Armenia. Indeed, although Armenia accepted the peace plan, NK rejected it on the grounds that the plan didn’t present any guarantee to NK and that it didn’t urge Azerbaijan to stop its blockade against the enclave (Baser 2008, 92). This showed that at this stage of the conflict, not considering NK as an independent party to the conflict was still a very significant obstacle to its settlement. Also, with the intervention of the OSCE in the conflict, another problem emerged. Due to the regional ambitions of Russia, there was a lack of coordination between Russian mediation and the OSCE. Consequently, mediators didn’t have enough leverage on the parties of the conflict who weren’t put under enough pressure to accept compromises. In that particular case, parties to the conflict had the luxury of choosing the mediating options that made them better off. In that sense, they weren’t urged enough to settle the conflict through compromises because they had the possibility to choose the solutions which better satisfied their national interests. Since Russia was willing to maintain its influence over the Caucasus which it considered as its backyard, it started creating a sort of counter-mediation against the OSCE instead of cooperating with the organisation in order to join forces to settle the conflict. As explained by Svensson, ‘[w]hereas the OSCE foresaw an international peacekeeping mission that could pave the way for durable peace, the Russians were anxious that they were about to lose control over their own backyard’ (2009: 12). Russia was thus trying to marginalise the OSCE in order to maintain its position in the region and was thus inhibiting the chances for a conflict settlement in order to best satisfy its own ambitions. As explained by an Armenian diplomat: ‘[i]t [was] easier to bring the positions of Baku and Yerevan closer to each other than to reach an agreement between the mediators- Russia and the Minsk Group’ (Baser 2008, 86).

From the beginning of the conflict until April 1993, Mooradian and Druckman estimated that there were around 2,000 casualties per year, both sides included. However, according to their statistics, the conflict reached a different stage from April 1993 onwards where the conflict dramatically escalated: the fights intensified and became bloodier and bloodier. Never did the parties to the conflict experience that many losses. From 1993 to 1994, there were more casualties than during all the 5 previous years of the conflict together. It is estimated that around 15,000 people were killed that year only (1999: 718)10. This is due to the fact that Karabakh forces carried out a powerful military offensive which led to the situation where they managed to occupy 10,000 square kilometres of Azeri territories in addition to the 4,000 square kilometres of the NK region itself during the fall of 1993 (Yerasimos 1994, 81). Azerbaijan consequently carried out a bloody counter-offensive to try and recuperate what was being taken from them. However, the massive losses suffered by both Armenians and Azerbaijanis during that period led the conflicting parties to perceive that they had reached the stage of a mutually hurtful stalemate where the losses they suffered were greater than the potential gains they could achieve through continued fighting. According to Zartman, hurtful stalemates make conflicts ripe in order to start negotiating for a settlement and mediation usually becomes particularly important at this specific stage because ‘both sides […] realize that they cannot achieve their aims by further violence and that it is costly to go on’ (through Ramsbotham 2011, 178). According to Svensson, this is the main reason why parties to the NK war were particularly keen on settling the conflict through a ceasefire in 1994 (2009: 9). Indeed, the attempts of Russia to make parties sign an armistice were only effective in May 1994 because the conflict had finally reached ripeness. On May 5 1994, Azerbaijan, Armenia and NK delegates signed the Bishkek protocolwhich called for the signature of a ceasefire, on the initiative of the CIS Inter-Parliamentary Assembly, the Parliament of the Kyrgyz Republic, and the Federal Congress and Ministry of Foreign Affairs of the Russian Federation11. The ceasefire agreement, redacted under the auspices of the Minister of Defence and the Minister of Foreign Affairs of the Russian federation, was then finally signed a few days after12 and the conflict never re-escalated to a full-scale war since then because of the balance of power between the parties.

Through the lessons learnt from the Nagorno-Karabakh war, it is possible to draw some conclusions in order to better address the settlement of protracted violent conflicts. First of all, mediation can only be successful if there is a change in the perceptions of the parties. They have to come to believe that they can only be better off if they put guns down and start negotiating. Most of the time, this occurs after the conflict becomes ripe: parties suffer major losses and start losing fate in the purpose of violence. In the Nagorno-Karabakh war, the parties came to this conclusion alone because of the flaws of the mediation. However, it is also possible for mediators to create this sense of urgency. They need to reason with the parties in order to change their perceptions and convince them that their national security and interests can only be guaranteed through negotiations rather than violence. Mediation should thus try and create this perception of ‘ripeness’ arguing about the dangers of continued fighting before the conflict dramatically escalates, in order to avoid major losses, or giving parties incentives. However, for mediators to be successful in changing parties’ perceptions, they need to respect several conditions. Mediators have to be impartial in the sense that they need to prove a real commitment to the settlement of the conflict rather than trying to act in order to satisfy their own ambitions which could otherwise be detrimental to a settlement. Also, they have to be trusted by all the parties to the conflict and this can particularly be achieved if they are not biased towards one side. Although Russia was in a particularly good position to settle the conflict thanks to the historical and cultural links it shared with the parties, failure to respect these conditions led its mediation to be unsuccessful during most of the war. Furthermore, in a case where there are several mediators, actions have to be coordinated in order not to lose leverage on the conflicting parties. Failure to coordinate actions could lead to a situation where conflicting parties actually do not find themselves under enough pressure to make compromises as they are submitted to contradictory flows by the mediators. Finally, and this applies to internal protracted conflicts, there can only be long-term possibilities for peace if a government accepts to negotiate with a sub-state actor by recognising it as an equal party to the conflict. Otherwise this could increase the grievances of the sub-state actor and fail to provide itwith incentives to stop the violence.

Anaïs Chagankerian


1)      Map of the Region


Source : http ://

2)      Map of Nagorno-Karabakh and occupied territories: Autumn 1993.

Map 2

Source :

3)      Average Yearly Deaths of the War


Source: Mooradian, Moorad; Druckman, Daniel. ‘Hurting Stalemate or Mediation? The Conflict over Nagorno-Karabakh, 1990-95’. Journal of Peace Research, 1999, Vol 36, n°6, p 718.


1 Country Overview, Office of the Nagorno-Karabakh Republic in Washington DC. 13/04/2012.

2 See Appendix for regional maps.

3 Armenian Ambassador Speech at the University of Kent, 22/03/2012.

4 07/05/2001. 13/04/2012.

5 New York Times, ‘20 Reportedly Killed in Nagorno-Karabakh’, 17/02/1992 Accessed 15/04/2012; New York Times, ‘Ex-Soviet Force Is to Leave Caucasus Area’, 02/03/1992 Accessed 15/04/2012 .

6 ‘The War of 1991-1994’, Ministry of Foreign Affairs of the Nagorno-Karabakh Republic. Accessed 15/04/2012.

7 BBC News, ‘Q&A : The Chechen Conflict’, 10/07/2006. Accessed 15/04/2012.

8 New York Times, ‘Cease-Fire Is Sought in Nagorno-Karabakh’, 21/02/1992,, Accessed 15/04/2012.

9 Background, Minsk Group, OSCE., Accessed 15/04/2012.

10 See Appendix for a graphic.

11 The Bishkek Protocol.NK Republic Ministry of Foreign Affairs. Accessed 17/04/2012.

12 Ceasefire agreement.NK Republic Ministry of Foreign Affairs. Accessed 17/04/2012.

Bibliography:‘Map of Artsakh’ Accessed 18/04/2012.

Baser, Bahar.‘Third Party Mediation In Nagorno-Karabakh: Part of The Cure or Part of The Disease?, 2008, 3:5. pp 86-114.

BBC News,‘Q&A : The Chechen Conflict’, 10/07/2006. Accessed 15/04/2012 Accessed 18/04/2012.

BBC News,‘Azerbaijan Condemns Turkish Pact’ 11/10/2009.http ://  Accessed 18/04/2012.

Ministry of Foreign Affairs of the Nagorno-Karabakh Republic.‘The War of 1991-1994’ Accessed 15/04/2012.

Ministry of Foreign Affairs of the Nagorno-Karabakh Republic.‘The Bishkek Protocol’. Accessed 17/04/2012.

Ministry of Foreign Affairs of the Nagorno-Karabakh Republic .‘Ceasefire agreement’. Accessed 17/04/2012.

New York Times,‘20 Reportedly Killed in Nagorno-Karabakh’, 17/02/1992 Accessed 15/04/2012.

New York Times,‘Ex-Soviet Force Is to Leave Caucasus Area’, 02/03/1992 Accessed 15/04/2012.

New York Times,‘Cease-Fire Is Sought in Nagorno-Karabakh’, 21/02/1992,, Accessed 15/04/2012.

New York Times,‘Officials Seek Way to Stop a Strike in Armenian Enclave’, 03/04/1988. Accessed 18/04/2012.

New York Times, ‘Soviet Troops Said to Patrol Streets of City in Azerbaijan’, 24/05/1988. Accessed 18/04/2012.

New York Times,‘More Soviet Troops Are Sent to Disputed Armenian Area’, 30/08/1989. Accessed 18/04/2012.

New York Times,‘Russian Republic Urges Azerbaijan to End Armenia Blockade’, 03/12/1989. Accessed 18/04/2012.

New York Times,‘Upheaval in the East; Years of Angry Clashes In 2 Southern Republics’20/01/1990. Accessed 18/04/2012.

Office of the Nagorno-Karabakh Republic in Washington DC.‘Country Overview’. Accessed 13/04/2012.

OSCE, Minsk Group.Background’., Accessed 15/04/2012‘Bishkek Protocol Was the Only Real Achievement in Karabakh Settlement, Secretary of NKR Security Council Says’, 07/05/2001. 13/04/2012.

Mooradian, Moorad; Druckman, Daniel.‘Hurting Stalemate or Mediation? The Conflict over Nagorno-Karabakh, 1990-95’. Journal of Peace Research, 1999, Vol 36, n°6, p 718.

Ramsbotham, Oliver; Woodhouse, Tom; Miall, Hugh.Contemporary Conflict Resolution.Chapter 7.‘Ending Violent Conflict: Peacemaking’, pp 171-197. UK: Polity, 2011

Svensson, Isak.‘The Nagorno-Karabakh Conflict: Lessons From The Mediation Efforts’, March 2009, The Initiative for Peacebuilding.

Yerasimos Stéphane.‘Caucase : le retour de la Russie’. Politique étrangère, n°1, 1994, 59th year, pp 61-86.