The consequences of the Karabakh conflict in Azerbaijan and Armenian societies include large numbers of refugees and displaced persons, huge post-war migration out-flows and other significant demographic changes in both countries. Former combatants have come to be an important constituency in both societies and, particularly in the case of Azerbaijan’s government, viewed with great suspicion. But in reviewing social changes in both societies, Baghdasarian and Yunusov distinguish between these consequences and the ‘no war, no peace syndromes’ related to the current impasse. These include stalled democratic development, taboos against questioning militarism and military spending, and the internalisation of ‘victim’ and ‘victor’ identities. Political elites ‘manage antagonism’ to sustain hostility towards the enemy without allowing it to reach a crisis point.
War, social change and 'no war, no peace' syndromes in Azerbaijani and Armenian societies
Like any war, the Nagorny Karabakh conflict has wrought numerous significant social changes, including waves of refugees and humanitarian and social crises. However, when addressing change in Armenian and Azerbaijani societies it is useful to distinguish between ‘post-war’ consequences of the conflict and what could be termed ‘no war, no peace’ syndromes relating to the current impasse. The latter include militarisation and the integration of combatants into the ‘peace process’, the stalling of democratic development, the internalisation of identities of victor (Armenia) and victim (Azerbaijan) and contradictory approaches to mediation. The prevalence of these syndromes and their role in maintaining animosity towards the ‘enemy’ warns against labelling them as ‘post-war’. On the contrary, they can be seen as syndromes potentially leading to a second round of armed hostilities. This ambiguity is a defining feature of the situation today: while certain radical forces within government and opposition in both states seek to maintain a certain level of public antagonism towards the ‘other’, there is also a need to prevent this condition from reaching crisis point. ‘Managed antagonism’ affords key players certain political dividends, encouraging the deployment of the Karabakh factor in internal political struggles. An important consequence is the perception that it is societies, and not political elites, who are not ready for resolution of the conflict and that hostility and hatred define Azerbaijani-Armenian relations.
Consequences of the war and social change
Before looking more closely at the ‘no war, no peace’ syndromes, it is useful to review some of the changes in Azerbaijan and Armenia societies as a consequence of the war. Above all, large-scale population movements during and after the war have reshaped the Armenian and Azerbaijani demographic and political landscapes. At the outset of the conflict the refugee issue was a rallying cry and key argument between the conflicting parties. Over the years, however, refugees and displaced persons have assumed different functions in the discourse of Armenian and Azerbaijani negotiators. The continued existence of ‘tent camps’ (or ‘tent cities’) holding displaced Azerbaijani populations has served as an unequivocal reminder of the unresolved status of the conflict for Azerbaijani society as a whole, and as a graphic demonstration of Azerbaijan’s suffering as a result of the conflict for international actors visiting these camps. By contrast, the issue of Armenian refugees has not been so politically charged. The following sections show how the two countries have been affected differently by population shifts.
Refugees and displaced persons
After the beginning of the conflict in 1988 both sides were overwhelmed by mass population movements. By 1990 the Azerbaijani government was registering large numbers of refugees from Armenia (see figure 1) as well as 48,000 Meskhetian Turks from Uzbekistan.
Calculating the number of Azeris internally displaced from Nagorny Karabakh was hugely complicated, but in 1993-94 on the basis of State Statistical Department figures (see figure 2) the Azerbaijani government announced the presence of more than one million refugees and displaced persons in the republic, accounting for 12 per cent of its population. Despite subsequent political stabilisation in Azerbaijan, the government continues to cite similar figures. However, according to the data of independent experts, the United Nations (UN) and the International organisation for Migration there may now be around 750,000 refugees and displaced persons in Azerbaijan, accounting for slightly more than 9 per cent of the population.
Figures regarding the number of refugees and displaced persons in Armenia are again difficult to break down. At the end of 1993, according to the official figures, the number of refugees and displaced persons from the Karabakh and Georgia-Abkhazia conflicts amounted to nearly 11 per cent of the population of Armenia (see figure 3). These figures are probably exaggerated and have worked their way into international sources: the UN cited a figure of nearly 500,000 refugees in Armenia on the basis of such official information. Over time, part of the refugee population returned to Karabakh or otherwise left Armenia, so by December 2000 refugees accounted for more than 8 per cent of the republic’s population. Some 30,000 refugees from Azerbaijan took Armenian citizenship at the beginning of 2002.
Post-war migration outflows
After the ceasefire in 1994 migration outflows from both republics continued, but were now associated with political instability and economic hardship, reflected in the fact that these outflows were composed mainly of Azerbaijanis and Armenians rather than minority groups. Out-migration remains uncontrolled and it is impossible to determine reliable figures, but official sources in Azerbaijan acknowledge that the figure of 800,000 outward migrants since 1994 is an underestimate. Between 1991 and 2000 more than 1.5 million left for Russia alone, where according to unofficial sources up to 2 million Azerbaijani citizens (equivalent to 25 per cent of the population of Azerbaijan) live and work today.
As in Azerbaijan, the Karabakh ceasefire put an end to the flow of refugees from the conflict zone to Armenia, but as the socio-economic situation in Armenia deteriorated the number of migrants leaving the country increased noticeably. According to the data of independent experts up to one million people left Armenia in the period 1990-2001 (see figure 4), suggesting at least 26 per cent of Armenia’s population left in the first decade after independence.
The demographic profiles of both republics show further changes wrought by the war and resulting socio-economic and political developments. During the conflict not less than 600,000 Azerbaijani citizens belonging to national minorities left the country; as a result, more than 90 per cent of the population are now ethnic Azeris. The composition of the non-Azeri population has also sharply changed: if before Armenians and Russians were the dominant non-titulars (see the political glossary), now Lezgins, Talysh and Kurds have taken their place.
The biggest wave of migrants from Armenia was composed of Azeris and Muslim Kurds. The almost complete removal of Armenia’s Muslim population occurred during the most difficult years of the conflict. Migration of representatives of other groups from Armenia, already demographically marginal in the Soviet period, did not have such a significant impact on Armenia’s population profile.
All of the above data bear witness to the humanitarian disaster caused by the conflict. Behind these statistical facts lie real people with serious psychological traumas, even if some have established new lives in new surroundings. The displaced Azerbaijani population living in ‘tent camps’ finds itself in the worst situation: confronted by disease, poverty, declining humanitarian aid and a government that would sooner exploit it for propaganda purposes than address its problems, this population is increasingly vulnerable to radicalisation. Efforts to voice its problems have resulted in public disorder, blocked highways and clashes with security forces. Opposition forces also attempt to exploit their frustration, further aggravating their predicament and complicating approaches to alleviating it.
During the war combatants were seen in their own societies as heroic defenders of the homeland. Their political role, however, has been perceived more ambivalently. Armenian and Azerbaijani militias and armed bands appeared spontaneously in the period 1988-91 and were used by many political forces, including political parties, as vehicles for banditry and the removal of political opponents. Armenia perceived the danger posed by armed bands first and in 1990-91 nearly all militias were incorporated into the body of regular armed forces, while many militias simply moved to the conflict zone in Nagorny Karabakh. At least half of the arms in the republic (some tens of thousands of units) remained in the possession of the population, a factor reflected subsequently in levels of violent crime.
In Azerbaijan the process of forming regular military units took place later, and as late as 1993 self-defence volunteers and militias affiliated with political organisations were engaged on the Karabakh front. During the fiercest fighting in Karabakh in 1992 Azerbaijani forces were made up of 21,000 regular army soldiers, 7,000 volunteers from battalions of the Popular Front and other political parties and up to 4,000 members of special police units. The scale of volunteer engagement was of no small concern to the government in Baku: in 1993 President Heydar Aliyev disbanded 33 volunteer battalions consisting mainly of opposition followers. Aliyev’s disbandment policy to a great extent accounted for the subsequent crisis on the front and contributed to the fall of seven regions around Karabakh to Armenian forces. In 1994-95 Aliyev dealt his internal security forces a similar blow, arresting 710 officers and disbanding its militias.
Since the ceasefire the role of former combatants in each society has been different. organisations composed of former combatants have assisted veterans, the war-wounded and bereaved families, as well as seeking to educate younger generations in a military-patriotic tradition. These have been viewed popularly as entirely legitimate agendas aimed at providing justice for deserving elements of society short-changed by the state.
The most influential combatants’ organisation in Armenia, Yerkrapah (‘Defenders of the Land’) was formed in 1994 on the initiative of the first Armenian Minister of Defence Vazgen Sarkisian. Sarkisian was a key figure in the military-political life of Armenia from the outset of the Karabakh conflict and continued to wield great influence at the Ministry of Defence even after leaving his post. Yerkrapah was effectively financed from the coffers of the Ministry of Defence and its members had the right to wear ceremonial arms. With a sizeable membership (40,000), it played a significant role in the internal politics of the country, especially in the aftermath of the 1996 presidential election. Responding to opposition street protests against the election result, Yerkrapah members enforced the state of emergency declared by the authorities in Yerevan. They did so wearing full camouflage uniforms (which by law can only be worn by regular army soldiers) and bearing machine guns, grenade launchers, sniper rifles and hand grenades. Yerkrapah continued to wield significant influence until Sarkisian’s assassination in 1999.
In Azerbaijan a different situation developed after the ceasefire. While formally declaring a policy of welfare, from the outset the government regarded veterans of the Karabakh war with great suspicion. It spared no effort to prevent the emergence of veterans’ social or political organisations, resorting to repressive measures in some cases. In October 1994 the Nijat (‘Salvation’) organisation, supporting bereaved families, was disbanded and nearly 40 Karabakh veterans sentenced. In the second half of the 1990s the Azerbaijani government initiated a number of proceedings against members of former volunteer battalions, such as the case of the Garangush brigade charged with an attempted coup in the Autonomous Republic of Nakhichevan, and imprisoned dozens of veterans. In January and February 2001 the government mercilessly dealt with protest actions mounted by members of the Society for the Wounded of Karabakh demanding a rise in pensions: 14 disabled protesters were arrested. More than once the government has cracked down on activists of the Karabakh Liberation organisation. Against a backdrop of increasing socio-economic hardship in Azerbaijan and rising disenchantment with the peace process, these actions against veterans (especially the disabled) have provoked indignation in society and predisposed veterans to ever more radical positions. Surveys provide evidence of the more radical stances held by veteran groups, a factor commonly highlighted when public attitudes towards developments in the peace process are probed.
‘No war, no peace’ syndromes
The patriotic mood of societies is reflected in attitudes towards their armies and the strengthening of their combat capacity. Questions of military spending are seen not through the lens of the dangers of militarisation but from the perspective of an evident threat posed by the enemy, and the need to be prepared in case of aggression. Such formulations as, ‘the Azerbaijani army must be prepared to take back territories occupied by the Armenians,’ or, ‘the armies of Armenia and Nagorny Karabakh must be ready and forever prepared to resist Azerbaijani revanchism,’ have been a constant in statements regarding possible changes in the post-war status quo over the past eleven years.
It is no surprise that according to opinion polls conducted by the authors in Armenia and Azerbaijan at the end of 2004 the majority of respondents in both countries considered their own army the probable victor in the event of renewed hostilities. Moreover, the respondents did not connect their belief in the victory of their own side to levels of economic development obtaining in their own or the other country, nor indeed with any other factor. It comes as no surprise that with the prominence of these attitudes and in the context of two possible outcomes of the ‘no war, no peace’ situation, questioning militarism in both Armenia and Azerbaijan is taboo. In each country the logic behind the taboo is different, yet the result is the same. In Armenia the logic is that since the army won the war it has earned the right to be trusted by society; in Azerbaijan, it is claimed that any imposition of review or oversight on the army would obstruct the strengthening of its capacity to re-establish Azerbaijan’s territorial integrity.
The ‘victor’ and ‘victim’ syndromes in Armenian and Azerbaijani societies are undoubtedly among the key consequences of the Karabakh war. We have focused here on some of the less obvious symptoms of these syndromes, such as taboos on public recognition of militarisation. Other aspects of these syndromes are reflected in the results of public opinion surveys carried out by the authors. Victor and victim syndromes affect societies’ perceptions of threat: whereas Azerbaijan sees the non-resolution of the Karabakh conflict as the most serious threat to its security, Armenians perceive greater threat from internal political developments and consider the Karabakh conflict ‘solved’. Similarly public perceptions of foreign countries and international organisations in Azerbaijan are strongly influenced by their stances on the Karabakh conflict; Armenian perceptions are more differentiated. These views derive from skewed information disseminated by the mass media in each country. The mutual isolation of Armenian and Azerbaijani societies will continue until greater efforts to build new bridges are made.