In a lightning strike on Sept. 19, Azerbaijan finally extinguished more than 30 years of de facto self-governance by ethnic Armenians in the embattled enclave of Nagorno-Karabakh. Framing its military assault as a “counterterror operation,” the Azerbaijani army overwhelmed Karabakh Armenian forces within 24 hours. The terms of the subsequent cease-fire included the disbanding of all local Armenian armed forces and the dissolution of the de facto institutions of the Nagorno-Karabakh Republic, the unrecognized entity that had declared independence from Azerbaijan on Jan. 6, 1992.
Three days later, Azerbaijan reopened the Lachin Corridor, the sole road connecting the enclave with Armenia, which it had sealed off to civilian traffic nine months before. Over the following six days, more than 100,000 Karabakh Armenians, comprising the entire Armenian population of Karabakh, poured through the corridor to become refugees in Armenia. Only a handful of the elderly and infirm remained as the region was reincorporated into the Azerbaijani state.
The exodus of ethnic Armenians brings their millennial presence in the eastern reaches of the Lesser Caucasus mountains to an end. Karabakh, or Artsakh as many Armenians know it, is fabled in Armenian culture as a bastion of survival during long centuries when no Armenian state existed. With its landscapes dotted with iconic Armenian churches and monasteries, Karabakh had come to symbolize a much greater array of Armenian ideals than just the claim to self-determination of its population. Its loss is perceived as a catastrophe on a level unseen since the era of the Armenian Genocide during World War I and another excruciating Armenian reckoning with the fickle calculations of great powers.
For Azerbaijan, the dissolution of the Nagorno-Karabakh Republic restores the country’s territorial integrity after three decades of fragmentation. Control over all of Karabakh completed what had been only a partial military victory in the 2020 war, which ended with Baku recovering most but not all of the territories it had lost to Armenian forces in 1992-1994. The cease-fire that ended the fighting in 2020 also saw a reassertion of Russia’s presence in the region that threatened the congealing of a new “frozen conflict” under Moscow’s control. For many Azerbaijanis, the outcome of September’s fighting represented the end of a homeland war and the dawn of a new sense of sovereignty, now complete.
The enabling context for Azerbaijan’s offensive was the accumulated erosion of Russian control over the new status quo that Moscow had introduced when it brokered the end to the last major conflagration of Armenian-Azerbaijani violence in 2020.
Arrived at even as Azerbaijani forces assumed a commanding military position in Karabakh, the trilateral Cease-Fire Statement of Nov. 10, 2020—signed by Armenia, Azerbaijan and Russia—denied Azerbaijan a complete victory. Instead, the cease-fire installed the traditional architecture of a “frozen conflict” in Eurasia: a small and dependent territory under Russia’s protection, a protracted and unproductive peace process in which Russia had a deciding stake, Russian peacekeeping boots on the ground and securitized relations between the conflict parties necessitating Russian “policing.”
Turkey—whose military involvement had enabled Azerbaijan to mount its overwhelming Blitzkrieg campaign and whose diplomatic cover allowed Baku to reject international calls for deescalation—was relegated to a largely symbolic involvement in the form of a presence at a cease-fire monitoring center near the Azerbaijani town of Agdam.
Russia’s power play in November 2020 stunned many observers, yet it brought with it several tensions. If there had been one point of consensus before 2020 between Azerbaijan and Armenia, which had previously acted as Karabakh’s patron state and Azerbaijan’s interlocutor in discussions to resolve the dispute, it was that neither wanted a Russian monopoly on the mediation of the conflict between them. Yet this was precisely the outcome institutionalized by the 2020 cease-fire. This had important implications later on, as it ensured that Yerevan and Baku would welcome a diversification of the mediation landscape.
Russian mediation also rested on the paradoxical assumption that Moscow could deliver stability and even rapprochement between Armenia and Azerbaijan while preserving its own desired outcome, namely irresolution of the conflict. Since the mid-1990s, in stark contrast to its response to Eurasia’s other secessionist conflicts, Russia’s policy in Nagorno-Karabakh had been predicated on not choosing between Armenia and Azerbaijan. Irresolution, however, pitted a Russian-sponsored status quo against Azerbaijani impatience to obtain a final outcome while it held the advantage.
Another more ruinous tension, which could not be foreseen in 2020, was introduced by Russia’s invasion of Ukraine in February 2022 and the subsequent trajectory of its war effort there. In both material and reputational terms, Russia’s role as a security patron in the South Caucasus precipitously declined.
At the same time, the extent of Western support to Ukraine, the imposition of sanctions and the realization that the war would be long forced Russia to reevaluate its interests and commitments in the Karabakh conflict. Specifically, a new calculus emerged regarding the relative value to Moscow of Armenia and Azerbaijan that challenged Russia’s prior preference of avoiding a choice between them.
Two dynamics that preceded Russia’s invasion of Ukraine accelerated in its wake. The first was Azerbaijani challenges to the 2020 cease-fire. These had already been evident since May 2021, when a series of escalations, skirmishes and incursions into Armenian territory along the international border between the two states began. Subsequent Azerbaijani military operations in March and August 2022 strengthened local Azerbaijani positions along lines of contact in Karabakh.
The second trend was the mobilization of a mediation effort by the European Union, which for years had been criticized for playing no role in a major interstate conflict in its neighborhood. In December 2021, at a summit in Brussels with Azerbaijani President Ilham Aliyev and Armenian Prime Minister Nikol Pashinyan, European Council President Charles Michel asserted his readiness to work with Baku and Yerevan on a peace agreement. A series of meetings followed in April, May and August 2022 that at the time appeared to define a structured agenda for Armenian-Azerbaijani dialogue.
However, the EU’s mediation effort crystallized differing interpretations of that agenda among the various participants. Azerbaijan framed EU mediation as encompassing only the issues relevant at the interstate level with Armenia, rejecting EU mediation of its relations with the Armenian population in Karabakh. The EU, on the other hand, stressed its commitment to a comprehensive peace including mechanisms that would address the rights and security of the Karabakh Armenians. A potential quid pro quo emerged whereby the Armenian leadership expressed its willingness to recognize Azerbaijan’s territorial integrity, if mechanisms guaranteeing the rights and security of the Karabakh Armenians were agreed.
This approach unequivocally brought the EU’s strategy into line with its positioning on other Eurasian conflicts, finally quelling Azerbaijan’s grievance over Brussels’ hypocritical approach to its territorial integrity compared to that of Ukraine or Georgia. Yet in those settings, EU support had aligned with aspiring democratic regimes willing to discuss variable approaches to governance in contested areas.
In contrast, Azerbaijan not only expressly ruled out discussions of autonomy or distinct governance arrangements for Karabakh Armenians, it was engaging in a campaign of intimidation against them. Since February 2022, reports of Azerbaijani vehicles encircling Karabakh Armenian villages with loudspeakers urging the population to leave, as well as periodic interruptions of gas and other supplies, had become common.
This impasse highlighted the vulnerability of the EU’s approach, implemented in tandem with the United States. Having committed to resolve, rather than refreeze, the conflict, Euro-Atlantic negotiators sought credible commitments on guarantees for Karabakh Armenians that ran counter to realistic appraisals of Azerbaijan’s capacities to offer such guarantees given its internal regime politics.
In September 2022, Azerbaijan sought to break the impasse by leveraging Armenia’s own territorial integrity, striking targets deep inside Armenia itself and occupying new pockets of territory in a two-day offensive. This triggered the increased involvement of two key EU member-states, France and Germany, leading to a decision in October 2022 to mobilize an EU monitoring mission to Armenia’s border with Azerbaijan. It was perhaps the first time in eight years that military escalation had resulted in outcomes not welcomed by or advantageous to Azerbaijan, which rejected the EU monitors’ access to its side of the border.
This international mobilization to prevent interstate war resulted in a shift in strategy on the part of Baku. In December 2022, the Azerbaijani government blocked the Lachin Corridor to civilian movement, under the guise of an “eco-activist” protest against the exploitation of natural resources in Nagorno-Karabakh. The blockade was also justified by persistent Azerbaijani claims that landmines and other materiel were being smuggled through the corridor. Although not independently verified, these claims were taken to substantiate the Azerbaijani charge that Armenia was not abiding by the terms of the 2020 cease-fire.
The resulting blockade was initially manageable through the continued access of the International Committee of the Red Cross, or ICRC, and Russian peacekeepers to Nagorno-Karabakh. But in June, Azerbaijan tightened its grip to exclude even ICRC and Russian access, causing severe shortages of food, fuel and medicine in the territory.
Amid growing reports of malnutrition, anemia and crippling fuel shortages in Nagorno-Karabakh and a dispute over which road should be used to provide humanitarian relief, intense behind-the-scenes diplomacy finally succeeded in enabling the arrival of a lone Russian Red Cross truck carrying food, sanitary items and blankets on Sept. 12, the first such delivery in three months. That was followed by a second on Sept. 18. Azerbaijan launched its offensive the next day.
A vital question that will be discussed for decades is whether Azerbaijan’s “one-day war” was really necessary. There is much that we still do not know about the chronology and content of secret contacts between the Karabakh Armenian leadership and Azerbaijani officials in the days and weeks prior to Sept. 19. At a minimum, it is clear that negotiations were pointing toward Azerbaijan’s desired diplomatic outcomes.
Pashinyan, Armenia’s prime minister, had repeatedly asserted Armenia’s willingness to recognize Azerbaijan’s territorial integrity. This had been reinforced by EU messaging, to the point where an EU statement in May 2023 had gone as far as to enumerate the territorial areas in square kilometers of both Armenia and Azerbaijan—29,800 and 86,600 respectively, the latter figure inclusive of Karabakh—in order to underline Azerbaijan’s undisputed sovereignty over the region.
Despite this, however, Azerbaijan maintained a nine-month blockade and staged a major military offensive that in hindsight make sense as a phased strategy to physically and psychologically weaken a civil population in advance of a major military assault, incentivizing their mass displacement through intimidation and violence; the fallacy of “voluntary departure” has consistently been used to explain away the coercive reordering of demography in the South Caucasus since the late 1980s. The totality of the exodus that followed and what some journalists on the ground reported as the resignation of the refugees to the finality of their departure indicate that it worked.
Azerbaijan’s choice to use force against a weak and isolated opponent may be puzzling seen through the prism of the ongoing peace process, since Baku held all the cards already and diplomacy, albeit falteringly, was delivering long-sought-after commitments to Azerbaijan’s territorial integrity. Azerbaijan’s calculus makes more sense when we see the conflict over Nagorno-Karabakh—and conflicts more generally—not just as a set of outcomes but as a strategy for shaping political community, agency and legitimacy.
Azerbaijan’s approach to the Karabakh conflict over the past three years will become a textbook case of authoritarian conflict management, or ACM, an approach to conflict that uses a variety of coercive methods to suppress grievances, impose stability and uphold power verticals within the state deploying it. Its spread reflects the wider decline of the liberal international order and the latter’s emphasis on negotiated settlements and peacebuilding.
ACM functions through the dominance of a single hegemonic discourse that foregrounds state actors at the expense of all others. The period since the 2020 war has been notable in Azerbaijan for the further tightening of political controls over various forms of autonomous political association, from political parties to media to religious organizations. This extended in 2023 to the scattered and atomized network of Azerbaijani peace activists critiquing Baku’s militarism, which was increasingly targeted, leading many of them to go into exile.
ACM in Azerbaijan is tied to a powerful emotional culture of resentment that is used to justify the humiliation of vanquished opponents, with a stark individualization of Azerbaijan’s military success in the person of its president, Ilham Aliyev. His recently filmed tour of Karabakh’s regional capital highlights both features, depicting Aliyev walking alone through abandoned cityscapes, clad in military fatigues and at one point stepping on the former de facto republic’s flag underfoot. Such acts of ritualized humiliation are hardly accidental. To the contrary, they lay the foundation for Aliyev’s personalized legitimacy as the icon of Azerbaijani victory.
(This is not to suggest that either side has a monopoly on humiliating the other. Few spectacles could have been guaranteed to generate similar feelings among Azerbaijanis than the sight of Pashinyan participating in a folkdance during a May 2019 visit to Nagorno-Karabakh’s previously Azerbaijani-majority city of Shusha, known as Shushi to Armenians.)
ACM in the context of Nagorno-Karabakh has two key implications for the future. One is the tension between mobilization around and social fatigue with conflict. In the years and possibly decades to come, Azerbaijani citizens will be persistently mobilized to celebrate victory, not peace. This will encumber any Azerbaijani leader seeking to transform the relationship with Armenia. At the same time, the Azerbaijani elite will need to navigate social fatigue with the continued mobilization of society around the conflict rather than other values, such as rights and participation, which for many years Azerbaijani officials have declared off-limits until the conflict was resolved.
The second key implication is that ACM does not resolve the underlying issues driving conflict, but rather embeds them in new cycles of injustice. The mass forced displacement of the Karabakh Armenians has set reconciliation back by at least a generation and probably more, and sets the stage for a new cycle of disputed claims in the future. Normative considerations will motivate discussions of rights of return, and even the symbolic return of a small number of Armenians would suit a number of geopolitical agendas. Yet these debates, for many years at least, will remain entirely divorced from realities in which the two societies are mobilized to see returnees as the illegitimate fifth column of a hostile, irredentist power.
What, then, can be expected of the Armenian-Azerbaijani peace process going forward? Azerbaijan’s military operation spelled the end of the two predominant approaches to resolving the conflict associated with outside actors: the “liberal peace” predicated on participation and co-existence advocated by the EU and the U.S., and the “frozen conflict” approach postponing solutions to an indefinite future favored by Russia.
Azerbaijan’s ascendancy instead facilitates a pathway to the realignment of the region away from being seen as a periphery of Europe or a contested European-Russian neighborhood toward becoming a regionalized space bringing local powers Turkey, Russia and, potentially, Iran into alignment around Azerbaijan as the keystone. This constellation will provide for some forms of regional cooperation and connectivity, but through a top-down regionalism that does not seek to resolve underlying fractures and which promotes illiberal norms.
This process of “regionalization” is not new, but Azerbaijan’s military solution in Karabakh resets the terms. A project to eject the South Caucasus out of a “globalized” order regulated by liberal norms into a “regionalized” space managed by local illiberal powers has been significantly strengthened. At the same time, however, Azerbaijan has introduced two critical shifts.
First, Baku has shifted this process away from Russia’s exclusive “tutelage” toward a more diffuse constellation in which Russia is one partner among several. A significant tension for the future is the extent to which Russia can recalibrate its ideological attachment to dominance of the South Caucasus as part of “its” near abroad into a commitment to transactionalism as one stakeholder among several in a regionalized space.
Second, by incorporating Karabakh militarily, Azerbaijan has also shifted the next focus of regionalization from Nagorno-Karabakh to Armenia itself. In 2020, it was the war-ending cease-fire and arrangements in Karabakh itself that drove the regionalization process. Now it is discussion over an interstate agreement between Armenia and Azerbaijan, and the associated arrangements for transit and connectivity involving Armenia, that is doing so.
The outlook for mediation is in many ways paradoxical. On the one hand, the core issue driving the Armenian-Azerbaijani conflict has been “resolved” in the latter’s favor, suggesting that a long-awaited normalization treaty is within reach. On the other, multiple mediation tracks risk prolonging the fragmented circularity of talks. EU mediation continues, although Armenia favors key member-states and Azerbaijan the European Council as the key interlocutors. So, too, does Russian mediation. The good offices of the U.S. also continue to be accepted, while Azerbaijan has also recently welcomed a role for Georgian mediation. Azerbaijani analysts also regularly advocate for direct negotiations with Armenia, without any external “interference.”
These expressions of what might be termed “hyper-forum shopping” are in part a structural corollary of declining multilateralism and a rising multipolar order. The result is an iteration of the “multiple principals problem,” or to put it more bluntly, that which is everybody’s business is nobody’s business. Yet forum-shopping is also a political strategy that prevents any single mediator from bringing all of the possible trade-offs into a composite bargain around one table that could provide the basis for an agreement.
The resulting protracted and performative diplomacy provides cover for the establishment of new facts on the ground. And if there is one lesson from the history of Armenian-Azerbaijani diplomacy since 1992, it is that negotiations have never reversed facts on the ground. With Karabakh militarily subdued, however, the “ground” now in question is Armenia.
There are three sets of issues framing ongoing territorial disputes between Armenia and Azerbaijan. The first is the delimitation of their international borders, which is further complicated by the legacies of skirmishes and incursions from both the 1990s and since May 2021, which mean that lines of actual control vary significantly from presumed de jure boundaries. The second is the fate of a number of small exclaves—three Azerbaijani exclaves in Armenia and one Armenian exclave in Azerbaijan—which are territorial anomalies inherited from the Soviet Union.
Finally, and most consequentially for the wider region, there remains the issue of transit across southern Armenia that would connect mainland Azerbaijan to its larger exclave Nakhchivan and beyond to Turkey. This route is referred to in Azerbaijan and Turkey as the “Zangezur Corridor” and is heavily promoted in Baku and Ankara as facilitating a Middle Corridor route as an alternative to the Northern Route running through Russia.
A transit route across southern Armenia, under Russian supervision, is mandated by Article 9 of the 2020 Cease-fire Statement. Yet with almost all of the other arrangements mandated in that document now obsolete, it is surely a dubious basis for such an ambitious geopolitical project. Russia’s acquiescence to Azerbaijan’s military takeover of Karabakh, a striking departure from the Kremlin’s preference for “frozen conflicts” in Eurasia, is likely tied to a quid pro quo upholding Russia’s role as “guardian” of a trans-Armenian route as the sole relic the Kremlin was able to salvage from the otherwise defunct Cease-fire Statement. This reflects the reality that connectivity has become a real and urgent issue for Russia in the aftermath of the war in Ukraine.
Transit across southern Armenia as foreseen in the Cease-fire Statement consequently faces hurdles with regard to its legal credibility, as well as Armenia’s concerns that its national sovereignty be upheld and Western concerns over Russia’s ongoing role. Turkish and Azerbaijani officials have posited transit through Iran as an alternative. Though tensions between Azerbaijan and Iran have periodically flared since 2020, spiking after an attack on the Azerbaijani embassy in Tehran that killed a security official and injured two others, there has also been a consistent flow of pragmatic agreements on connectivity. Recent accords between Baku and Tehran point to the possibility of a road corridor connecting mainland Azerbaijan and Nakhchivan via Iran.
An Iranian alternative comes with other issues attached, however, namely the exclusion of Western investment in upgrading to costlier rail infrastructure due to Iran’s involvement; the ambiguous role of Russia, since unlike in Armenia there is no needfor a Russian peacekeeping presence in an Iranian-Azerbaijani connectivity arrangement; and the potential need for a wider regional platform providing a legal framework for new transit infrastructure given the number of states involved. A revival of the “3+3” platform—combining Russia, Turkey and Iran with Armenia, Azerbaijan and Georgia, an idea circulating since 2020 without serious uptake—could serve this purpose. Different corridor projects consequently implicate different constellations, and reconfigurations, of regional power.
Since 2020, connectivity has been virtually the sole framework for peace narratives. But connectivity breakthroughs have so far been stymied by the undiminished securitization of Armenian-Azerbaijani relations. Worst-case scenarios foresee the carving out of corridors by force. This, however, would complete the cycle of role-reversal between Armenia and Azerbaijan, establishing a new territorial politics of conquest, occupation and irredentism and foreclosing an alternative future of a reconnected South Caucasus.
Azerbaijani officials reject such scenarios. Yet with its principal goals achieved, Baku may be content to continue hedging among the region’s weakened and distracted hegemons, while consolidating new facts on the ground and protracting a negotiated settlement into an uncertain future.
Laurence Broers is an associate fellow at the Russia & Eurasia Programme at Chatham House and the author of “Armenia and Azerbaijan: Anatomy of a Rivalry” (Edinburgh University Press, 2019).
https://www.worldpoliticsreview.com/armenia-azerbaijan-nagorno-karabakh/?loggedin=1