From Struggle to Permanent Failure
Why the Karabakh Attempt at Secession Failed
After 44 days of fighting, the Second Karabakh War came to an end on 10 November 2020 as a result of a Russian‑brokered ceasefire agreement. The most important questions here appear to be: what led to this dangerous military escalation, and what does it mean for the conflict, given that it seems to have now entered into a (new) political phase, again?
In the declining Soviet Union, what was originally a status dispute over the autonomous Nagorno‑Karabakh region escalated into an international violent conflict between Armenia and Azerbaijan in the early 1990s. Following the end of a bloody war in 1994 (the First Karabakh War, 1992‑ 1994), a fragile situation around the conflict region took root: the “frozen conflict,” as it came to be known, lasted for nearly three decades and led to conditions of neither war nor peace. And during this period, it was feared that the longer the sides had to wait for a peace agreement to be reached, the more likely the conflict would re‑escalate and eventually erupt again into a hot war. As it turned out, this is exactly what happened: an all‑out six‑week war erupted again unexpectedly between the conflict parties in late September 2020, and, as a result, the Armenian side more or less capitulated.
But first things first: in the First Karabakh War, Azerbaijan suffered a major defeat, ceding to Armenian forces not only the secessionist region itself but also seven surrounding territories. These other lands were, as a whole, twice the size of Nagorno‑ Karabakh itself and contained five times the old oblast’s population, the entirety of which was expelled by the time an armistice was signed in 1994. And that is why during this war the UN Security Council responded by passing four resolutions demanding the withdrawal of Armenian forces from the occupied areas of Azerbaijan. However, the UN resolutions failed to have any effect.
Since that time, no international protagonists felt a strong, compelling need to try to resolve the Nagorno‑ Karabakh conflict. In addition, all international actors dismissed the idea of “power mediation.” Moreover, although Russia as a key international actor is directly involved in all the conflicts on the territory of the former Soviet Union, its involvement in the Nagorno‑Karabakh dispute has been rather indirect: in the Karabakh case, Moscow has been both a critical and a questionable actor. On the one hand, the Kremlin has taken a central position in mediating a peaceful settlement to the conflict while, on the other hand, it has been delivering weapons to both sides. This last represents perhaps the most striking situation regarding the international dimension of the conflict. Russia is militarily allied with Armenia and has a military presence in the country. It has provided security guarantees to Yerevan, primarily through their shared membership in the Collective Security Treaty Organization (CSTO), which neutralized to a certain extent the potential effects of Russian arms being sold to financially strong Azerbaijan on a purely commercial basis.
From the First to the Second Karabakh War
Following the end of the First Karabakh War, Armenia and Azerbaijan could not reach a political solution to the conflict: countless attempts and numerous rounds of negotiation failed; an attitude of resignation creeped in. Particularly, as nearly three decades went by, Azerbaijan got justifiably frustrated with a lengthy peace process that produced no tangible progress. The OSCE Minsk Group co‑chairs (France, Russia, and the United States), as the key peace brokers to the conflict, were reproached for not placing enough political or diplomatic pressure on the Armenians to withdraw from the occupied Azerbaijani territories, which especially precluded any settlement via negotiations.
Although the conflict was sparked by the status of Nagorno‑ Karabakh, the issue of the occupied surrounding territories complicated the nature of the conflict as a whole.
Although the conflict was sparked by the status of Nagorno‑ Karabakh, the issue of the occupied surrounding territories complicated the nature of the conflict as a whole. In this regard, the Nagorno‑ Karabakh conflict brought with it the risk of an additional shift in former state boundaries, in contrast to other conflicts in the region. Overall, after the First Karabakh War the conflict situation featured a structural asymmetry: Armenia wanted to use the power of facts (i.e., military control) to maintain the territory’s de facto status whilst changing its de jure status; Azerbaijan wanted to use the force of law (i.e., international law) to preserve the de jure status and change (back) the de facto status.
Having lost the First Karabakh War, Baku was particularly dissatisfied with the seemingly permanent occupation of its territories and the plight of IDPs; at the same time, it interpreted Armenia’s negotiating practices as representing a kind of salami‑slice tactic: Yerevan was trying to make only rhetorical— or at most, minimal—concessions in order to prolong negotiations because it was not at all interested in changing the status quo established by the ceasefire that ended the First Karabakh War. Armenia counted on the negotiations either coming to an end with it having to offer minimal concessions or being broken off with absolutely no results. The positions thus remained entrenched. The peace process was leading nowhere, which was why, from time to time, the Azerbaijani side made a point of asking what the point of the negotiation process was exactly, and threatened to use its ultimate form of pressure—its military—in order to prevent the Nagorno‑Karabakh conflict from remaining “frozen.”
Overall, following the end of the First Karabakh War until the onset of the Second Karabakh War, a fragile situation around the conflict region took root. However, an equilibrium favoring the status quo appeared to be established around this “frozen conflict” in basically three ways. First, militarily: an offence‑defense balance between Armenia and Azerbaijan (favoring defense); second, internationally: a regional balance of power with Russia as the key stabilizing actor; and third, socio‑psychologically: a political inertia (habituation effect) in the conflict countries.
For many years after the First Karabakh War, the offence‑defense balance appeared to overwhelmingly favor Armenia, which had clear defensive advantages favored by military and geographical factors. It is no surprise that Nagorno‑ Karabakh has been among the most militarized regions in the world: heavy defensive fortifications—including many kilometers of tunnels interlinking with each other along the ceasefire line and dense minefield—offered the Armenian side a false sense of invincibility for a long time.
In the years leading up to the Second Karabakh War, however, the offence‑defense balance changed gradually, ultimately shifting in favor of Azerbaijan.
In the years leading up to the Second Karabakh War, however, the offence‑defense balance changed gradually, ultimately shifting in favor of Azerbaijan. Its extensive military buildup, which took place over the last several years, became the first important indicator of this shift. One visible element of this is the fact that, several years ago, the Azerbaijani government established a Ministry of Defense Industry to build up the country’s military capabilities. In addition, Baku imported high‑tech modern weaponry in large quantities, including drones and loitering weapons (i.e., kamikaze munitions) from countries like Israel and Turkey, thus creating considerable offensive advantages. It came as no surprise that these weapons proved to be very effective in the recent war: within a few weeks, Azerbaijani troops were able to break through the Armenian defense line at several places and retake significant swaths of occupied territory. That is why Azerbaijan’s President Ilham Aliyev proudly stated during the war that “in this case, unmanned aircraft, both Turkish and Israeli drones, of course, helped us a lot.”
Turning next to the geopolitical context of the conflict in the last decades, Russia’s role as an external veto power has also been central in at least two respects. On the one hand, Moscow has been the only external actor that was believed to be able to contain and actually stop a new war between the conflict parties, as was evident during the April 2016 clashes (what some call the Four‑Day War), when the Kremlin forced them into a ceasefire. On the other hand, any amicable resolution to the conflict that goes against Moscow’s will is unimaginable. As such, Russia appeared to create a state of geostrategic stability or balance around the military and political status quo on the ground.
In recent years, Turkey’s rapid rise in power and Ankara’s introduction of a more assertive foreign policy in its neighborhood, resulted in a gradual shift in the region’s balance‑of‑power system that came to favor Azerbaijan. Specifically, Turkey and Azerbaijan built a very effective alliance— encapsulated in Heydar Aliyev’s “one nation, two states” phrase— which in turn weakened the “stabilizing” impact of the Armenia‑Russia alliance that had been effectively designed to perpetuate the status quo. But Baku also tried to maintain close relations with Russia as part of its “balanced” and “multivectoral” foreign policy, which had a constraining effect on the scope of Russian commitments towards Armenia.
In addition to military and geopolitical factors, starting in the second half of the 1990s, political stability set in also domestically in both Armenia and Azerbaijan. And a decades‑old conflict situation, coupled with unsuccessful negotiations, created a lasting condition of “No War, No Peace,” which the adversaries appeared to accept implicitly and gradually. Most importantly, over time it led to the effect that they appeared to avoid new costs or “extreme” measures in terms of both military escalation and substantive compromises. In other words, the willingness to take high risks declined continuously on both sides. Being full of uncertainties and insecurities, “No War, No Peace” implied a potential source of instability—but what amounted to a “stable” one.
Paradoxically as it may sound, “stable instability” worked in practice for decades: the conflict parties got used to this in‑between situation.
Paradoxically as it may sound, “stable instability” worked in practice for decades: the conflict parties got used to this in‑between situation. Thus, “No War, No Peace” became a new normal of sorts and established its own particular form of equilibrium. And this inertia became more sustainable the longer it lasted.
But then a revolution took place in Armenia: a new opposition leader, Nikol Pashinyan, came to power after a popular uprising in 2018, also raising expectations—hopes, even—in Azerbaijan for progress in negotiations. Initially, it looked as though “he was an open interlocutor ready to discuss thorny issues,” as Robert Cutler put it in an October 2020 Foreign Policy essay. Yet, gradually, quite the opposite happened. Tensions escalated, as the democratically elected Armenian government started making increasingly populist statements with respect to the Nagorno‑Karabakh conflict. Most prominently, Prime Minister Pashinyan said in his address at the opening ceremony of the Pan‑Armenian games held in Nagorno‑Karabakh in August 2019: “Artsakh is Armenia. Period.” He also repeatedly led the crowd in chants of “miatsum” (the Armenian word for “unification”)—a pan‑nationalist slogan that gained popularity during the original escalation of the conflict in the late 1980s. In this way, to refer to another Cutler formulation, Pashinyan apparently yielded to an “irredentist nationalism seemingly required to survive in Armenian domestic politics.”
In general, the increasingly provocative statements and actions by the new Armenian leadership were probably motivated by reasons of domestic power consolidation: they sought to increase their legitimacy by attempting to appear more nationalistic than the forces they had deposed. But by doing so—whatever the reason—Yerevan came to be seen as taking a harder and thus dangerously populist line on the Karabakh issue. And, most importantly, these moves were perceived in Azerbaijan as insulting and hurtful to the country’s national pride, thus amounting to, as the saying goes, adding insult to injury in the public’s perception. It can be argued that such actions by the Armenian authorities upset both the people and government of Azerbaijan, which in turn upset the political inertia that had characterized each country’s posture towards the other beginning in the years that followed the end of the First Karabakh War. As Aliyev made clear during the Second Karabakh War, “insulting the Azerbaijani people” proved to be “too costly” for the Armenian government.
Overall, the Nagorno‑ Karabakh conflict had been a typical dispute in the interwar period (1994‑2020), having reached an advanced stage of attempted secession that had been brought about by military force used by a neighboring patron state. Despite these military‑political advantages, however, Nagorno‑ Karabakh could not become independent.
after Armenia’s military defeat in the Second Karabakh War, it can be argued that Nagorno‑Karabakh’s struggle for secession has now been transformed from a unilateral attempt to a permanent failure.
And after Armenia’s military defeat in the Second Karabakh War, it can be argued that Nagorno‑Karabakh’s struggle for secession has now been transformed from a unilateral attempt to a permanent failure.
There may be many reasons— whether they be actor‑ or process‑centered—for why, against all odds, the Azerbaijani side never accepted the attempted secession of the breakaway region, despite its complete defeat in the First Karabakh War. To develop a deeper understanding of Baku’s invariable stance, we must first (and foremost) consider structural factors, such as geographic and historical preconditions, the ethnic composition of the state, and the country’s dominant legal system. From today’s perspective, these structural factors appear to be relevant to providing an explanation for the ultimate failure of Nagorno‑ Karabakh’s attempt at secession.
Geography
Covering an area of just 4,400 square kilometers, the Nagorno‑Karabakh region is relatively small. As such, it comprises only 5 percent of Azerbaijan’s state territory. Along with this great asymmetry between Azerbaijan proper and Nagorno‑Karabakh, the political and physical geography of the breakaway region differs from that of the other conflict cases in the post‑Soviet space and beyond. The fact of being an enclave should have hindered the region’s secessionist aspirations: it certainly strengthened the Baku central government’s resistance all along. At the same time, Nagorno‑Karabakh’s specific geographic position helped to expand the conflict beyond its boundaries: the Armenian side’s military strategically occupied the adjacent Azerbaijani regions, thereby creating an extensive “security belt” around Nagorno‑ Karabakh to offset the enclave’s precarious isolation and facilitate Armenian control by shortening the length of the front line. Armenia also sought to create an overland connection to Nagorno‑Karabakh, thus expanding the original conflict, which was at bottom about the region’s political status, into a territorial conflict that involved the desire to shift state borders. On the other hand, the issue of the surrounding territories complicated the nature of the conflict as a whole, in contrast to other conflicts in the region. In particular, the permanent occupation of these districts by Armenian troops precluded any peaceful settlement in the last decades.
There is a further geographical factor playing an important role in the conflict’s dynamics. In ethno‑territorial conflicts, a peripheral location (a border region or an island) is generally said to have strong centrifugal effects; whereas the contrary (an enclave in a heartland) is expected to foster centripetal tendencies and cause secessionist efforts to be strongly resisted. Nagorno‑Karabakh is an ethno‑territorial enclave within the Azerbaijani heartland that is separated from Armenia by the high mountains of the Lesser Caucasus, which make access from Armenia even more difficult.
Thus, the breakaway region clearly exemplifies the aforementioned second situation that, all other things being equal, should have inhibited secession because it made it much harder for Azerbaijan to agree to any territorial compromise in the interwar period. Interestingly enough, back in 1921 the Soviet leadership cited Nagorno‑Karabakh’s constant connections with Lower Karabakh and the rest of Azerbaijan as its official reason for retaining the region within the borders of Azerbaijan.
What is more, geographic locations at times also constitute a reference point of one’s national identity. The relevant territory is seen as a site which solidifies the nation’s collective memory into an indispensable component of its “character.” Shusha, a key town in Nagorno‑Karabakh, best illustrates Nagorno‑Karabakh’s national importance for Azerbaijan. Once the regional center for traditional carpet production, Shusha was also home to many Azerbaijani composers and singers who made the town famous as the musical capital of Azerbaijan. During the Soviet era, Shusha was even declared an inspiration for Azerbaijani culture.
It is thus no surprise that Aliyev made the liberation of Shusha a central goal during the Second Karabakh War, because, as he put it, “Shusha has a special place in the hearts of the Azerbaijani people. [...] Without Shusha, our business would be unfinished. Of course, this issue was [also] always on the agenda during the [peace] talks.”
History
Shusha is also a good example of a situation in which geography and history reinforce each other. As the old capital of the Karabakh khanate (1748‑1822), Shusha is also an important component of Azerbaijan’s (political) history. For example, the successful 33‑day‑long defense of the Shusha fortress against the all‑powerful army of the Iranian Aga Mohammed Khan Qajar in 1775 is a lieu de mémoire for a popular national‑historical story of Azerbaijani heroism.
Historically, another factor inhibiting secession is the lack of Armenian statehood in Nagorno‑ Karabakh. Although the Armenian side refers to its bloody fights for sovereignty in the disputed area, Nagorno‑Karabakh cannot invoke an earlier era of political independence under Armenian authority, which is always helpful for legitimizing secession. The region’s lack of any Armenian sovereign tradition contrasts with Abkhazia, for example, another long‑term post‑Soviet conflict in the region: a principality from the fifteenth to the nineteenth century with its own tradition of statehood, Abkhazia was a Soviet Socialist Republic from 1921 to 1931 before it was downgraded by Stalin to being an autonomous republic within Georgia.
What is more, over the decades, Azerbaijan was mostly concerned about losing still more land to its neighbor—in addition to the areas that Moscow had ceded to Armenia in the twentieth century. In Azerbaijani public opinion, Nagorno‑Karabakh’s secession would be thus perceived as Azerbaijan losing part of its territory to Armenia again. Most prominently, a comparison was made with the historical province of Zangezur, which had been transferred to Armenia after the establishment of Soviet rule in the South Caucasus in the early 1920s. That is because Azerbaijan sees in the conflict two complementary processes: first, the violent attempt at secession of a breakaway minority that seeks to expand beyond even its administrative borders; second, the irredentist policy of Armenia, which supports this attempted secession militarily in order to further expand its borders at the expense of Azerbaijani territories.
In this respect, it had been a dominant historical narrative in Azerbaijan in the past years and decades that, as Aliyev said back in 2014, for example, its “historical lands are not limited to Nagorno‑Karabakh and surrounding areas. [...] Today’s Armenia is, in fact, the historical land of Azerbaijan.” That is why the Azerbaijani government repeatedly made its policy plain that it would never allow a “second Armenian state” to be established on Azerbaijani soil. It is no surprise that Aliyev famously announced, already back in 2009, that “Nagorno‑Karabakh will not be an independent state, not today, not in ten years or one hundred years. Azerbaijan’s position is unequivocal. Despite all the pressure, we will defend this position to the end.”
Ethnic Composition
With only 1.5 percent (150,000 people) of Azerbaijan’s total population (10 million) documented as residing in Nagorno‑Karabakh (as of last count), there is a huge asymmetry in the quantitative relationship between the majority and the minority group in the country.
Another relevant factor inhibiting the attempted secession is connected to the ethnic composition and structure of settlements in the secessionist area. Prior to the war, the situation in Nagorno‑ Karabakh proved contradictory. Ethnic Armenians represented more than three‑quarters of the population, but the region also had a substantial number of ethnic Azerbaijanis. However, the Azerbaijani and Armenian settlement areas were not compact, displaying an ethnic heterogeneity in the conflict area: they were spread throughout the region—a situation that generally seems best suited to a system of autonomy with minority protection.
Then, during the First Karabakh War, ethnic cleansing transformed Nagorno‑Karabakh into a homogeneous, ethnically pure Armenian region. Just as in seven occupied surrounding territories, all ethnic Azerbaijanis either fled Nagorno‑ Karabakh or were expelled. At the onset of the conflict, in Azerbaijan proper only a tiny part of the population living in an equally tiny part of the country was of ethnic Armenian origin.
However, unlike the Israeli‑ Palestinian conflict for example, the ethno‑cultural differences in Nagorno‑Karabakh have not caused it to become an international proxy conflict between two religious groups—despite efforts by Armenia and its diaspora to portray themselves as an endangered Christian outpost in a predominantly Muslim region. Although Christianity is a source of the West’s general effects are limited. For example, the United States was the only Western country to impose sanctions against Azerbaijan in 1992—a sign of one‑sided solidarity helped by the Armenian diaspora’s intensive lobbying.
Dominant Legal Order
Along with the aforementioned non‑political factors, Azerbaijan’s tradition of a centralized state made Nagorno‑ Karabakh’s attempted secession even more difficult to accept. Also, regarding either a federative or a confederative scheme—namely, granting maximal sovereignty to Nagorno‑Karabakh short of a state independence—the following structural constraint immediately strikes the eye: as a unitary state with a presidential system of government, it would be very hard for Azerbaijan to consider even a loose union with Nagorno‑ Karabakh.
While looking at other conflict settlement cases, autonomy arrangements are rather a typical characteristic of centralist unitary states (albeit also found in federations), which was also Baku’s preferred solution in the interwar period. It is no surprise that back in 1998, the international peace broker’s common‑state plan— which foresaw a joint state for Azerbaijan and Nagorno‑Karabakh —failed because Azerbaijan would not accept Nagorno‑ Karabakh as its equal.
In addition to Azerbaijan’s domestic system, it is also the international system that makes Nagorno‑Karabakh’s attempted secession highly problematic. In this respect, Nagorno‑Karabakh’s legal status in the Soviet Union plays a central role. The Soviet leadership first issued a binding decision declaring Nagorno‑ Karabakh an autonomous region (oblast) of Azerbaijan in July 1921. Baku continues to regard this ruling as confirmation of the Azerbaijani nation‑state’s rightful boundaries (uti possidetis jur is—principle of the inviolability of borders). Accordingly, when Azerbaijan became independent— like all other former Soviet republics—it was under international law recognized by the community of states within the boundaries that it had as constituent republic of the Soviet Union. Azerbaijan therefore always saw the conflict first and foremost as an act of aggression by Armenia because it illegally occupied its sovereign territories during the First Karabakh War. That is why the UN Security Council also condemned the Armenian occupation in the early 1990s in four separate resolutions.
Significant Advantages
A sober calculation reveals that an internal settlement within Azerbaijan can present significant advantages for Nagorno‑Karabakh.
A sober calculation reveals that an internal settlement within Azerbaijan can present significant advantages for Nagorno‑Karabakh. One aspect is its geographic link to Azerbaijan: this would facilitate the development of the territory’s economic and transportation connections, which in turn would positively impact upon the surrounding regions. Also, twentieth‑century history reveals another important and positive moment in the collective memory of the two communities: the period of peaceful co‑existence when they lived together and got along with each other day in and day out. Building on these and similar examples could gradually transform the historically antagonistic distortions and enemy images and make it possible to create a new, shared identity.
In addition, Azerbaijan’s economic potential, which is far superior to that of Armenia, along with its financial resources, also presents opportunities for relatively poor Nagorno‑Karabakh. The case of South Tyrol in Italy can serve as an example: a once mostly poor province populated by mountain farmers, South Tyrol is now one of Italy’s wealthiest provinces. South Tyrol benefited not only from Italian government grants, but also from Italy’s membership in the EU, which granted significant regional funds to the autonomous province. In the same vein, if Nagorno‑Karabakh were to become prosperous in comparison to Armenia—like South Tyrol (Italy) did in comparison with North and East Tyrol (Austria), it could develop its own economic interests and self‑confidence. This, however, would require creating incentives, for instance in the form of special offers, such as starting a “Develop Karabakh” initiative and financial transfers. The regional road network, municipal infrastructure, and energy supply urgently need to be upgraded. Creating competitive structures, renovating and modernizing homes, and building new housing are also needed.
But it can be done.