Federalization, Land-for-Peace or a War Once More: Secession Conflicts in Post-Soviet Countries

Abstract

This essay conceptualizes five recipes to solve secession conflicts that have taken place in postcommunist territories—federalization, land-for-peace, protectorate policy, reconquest, and the destruction of the parent state by the patron state—and investigates their merits and demerits. This essay provides case studies of the South Ossetian War in 2008 and its aftermath (as an example of the protectorate policy), the Second Karabakh War in 2020 (reconquest), and the Russo-Ukrainian War in 2022 (destruction of the parent state by the patron state). We observe the tendency that the ineffectiveness of federalization and land-for-peace induces parties of conflict to move on to unilateral or even coercive recipes.

Russia invaded Ukraine on February 24, 2022, justifying this action by the need to save the Donbas people from “genocide.” Irrespective of how one judges this justification, the very fact of invasion demonstrates the danger causable by the international community’s indifference to secession conflicts. The lack of international attention may exacerbate the conflict into a repeated war with many more casualties than the first one. Among the five secession conflicts in former Soviet countries (the Karabakh, Abkhazian, South Ossetian, Transnistrian, and Donbas conflicts), only the Transnistrian conflict has not experienced a second war and the status quo established in the early 1990s is still upheld there. Yet this exception was possible due to the nonbelligerent attitude of both parties of conflict—namely the Moldovan and Transnistrian governments, a condition that their Caucasian and Ukrainian counterparts can only envy.

This essay overviews secession conflicts in postsocialist territories and classifies five recipes—federalization, land-for-peace, protectorate policy, reconquest, and destruction of the parent state—and identifies advantages and disadvantages of these recipes. This typological analysis highlights the chronological tendency that conflicting parties’ discontent with ineffective diplomatic recipes (federalization and land-for-peace) induces them to shift to unilateral (protectorate policy) or even coercive recipes (reconquest and destruction). In other words, momentum for escalation of post-Soviet secession conflicts is embedded not only in the conflicts themselves but also in the recipes to resolve them. In my view, the substance of this momentum is the desire shared by conflicting parties, mediator countries, and international organizations to solve the conflicts within the sovereign-state principle. Therefore, we must return to Robert Jackson’s idea, proposed at the beginning of this century (Jackson 2000). Jackson found it difficult to solve secession conflicts unless we relativize the single-layered image of international society universally covered by sovereign states and proposed to adopt formulas such as mandated territories and (possibly joint) protectorates based on agreement between conflicting parties. The reignitions of late-Soviet secession conflicts after 2008 confirm the validity of Jackson’s criticism of the existing sovereign-state system.

In order to verify Jackson’s opinion, we must perhaps include in our analysis examples from outside postcommunist territories. Therefore, this essay, focused on postsocialist secession conflicts, seeks to encourage argument rather than to provide a sophisticated answer. I hope to provoke the readers’ doubt that we might have started from the wrong point. What we are witnessing might be an old phenomenon requiring new thinking, rather than a “war unprecedented in post–World War II Europe” solvable by familiar rhetoric.

In 2003, Bruno Coppieters criticized the normative and deductive characteristics of the then studies on secessionist conflicts (Coppieters 2003, 2). I wonder how far we have moved forward during the last two decades. In their recent monograph on land-for-peace, Eiki Berg and Shpend Kursani argue that, by dogmatizing the territorial integrity of states, the presently dominant approaches to secession conflicts do not pay attention to the number of lives that could have been saved had a “geography of peace” (meaning territorial adjustments) been activated before and during the warfare (Berg and Kursani 2022, 4–5).

Continuity from Late Soviet Secession Conflicts

All the major armed conflicts in the post-Soviet space after 2008 were reignitions of the secession disputes taking place in the last months of the Soviet Union. ¹

• — the Second Ossetian War and a small-scale war in Abkhazia in 2008

• — the Crimean crisis and the beginning of the Donbas War in 2014

• — the Second Karabakh War in 2020

• — the Russo-Ukraine War in 2022 under the pretext of protecting the Donbas people

How may one understand this continuity? Soviet federalism gave privileges to titular ethnicities of union republics but obliged them to use the privileges to manage their republics in a consociational manner but not to pursue their own ethnic greed. In other words, Soviet federalism requested the titular ethnics to be the benevolent eldest brother in their republic, a status that titular ethnicities began to find disadvantageous for themselves in the late Soviet period. Tired of being the benevolent eldest brother, the titular ethnics began to shout slogans such as “Georgia for Georgians,” exactly as the Russians had become irritated by the request to be the eldest brother of the whole USSR. Self-assertions made by titular ethnics of union republics shattered the Soviet Union, and the hastiness of this dissolution left no room for considering requests posed by ethnic autonomies and discontented regions, such as Karabakh, Crimea, Donbas, and Transnistria.

In the post-Soviet era, self-assertion of ex-union republics’ titular ethnicities was reinforced by international organizations’ almost unconditional support for the new independent states in their disputes with secessionist entities, which had become de facto states. This support demotivated the new independent states from taking secessionist entities’ requests seriously.

Ethnic autonomies and some discontented regions in the late Soviet period desired to stay in the Soviet Union when their parent union republics tried to secede from it. These autonomies and regions did not resist their parent union republics’ desire to secede from the Soviet Union, but requested the same right for themselves to stay in the Soviet Union. The Union Law adopted on April 3, 1990, allowed autonomies and other sub-republican localities to separate from their parent union republic to stay in the Soviet Union if a majority of its population wished to do so (Zakon SSSR 1990). If this law were implemented in the course of the Soviet Union’s split, the process might possibly have been more peaceful and balanced than it actually was. Post-Soviet state borders would have become somehow different from the former administrative borders between the union republics. Except for Abkhazia, where the titular nationality (the Abkhazians) had become a minority, in most secessionist entities, the position of wishing to stay in the Soviet Union would have won the referendum. As a result, new independent states would have become smaller territorially than their precedent union republics, yet instead they would have been more homogeneous ethnically and have enjoyed more favorite conditions for their nation building in the 1990s.

The Union Law on April 3, 1990, had much in common with land-for-peace arrangements and had historical parallels. Facing an independent movement within one’s own country, it seems natural for the suzerain to request the secessionist entity to leave behind some part of its territory. In 1648, the Spanish Crown agreed with the Netherlands’ independence but retained the Catholic-dominant southern provinces within its sovereignty. In 1920, Ireland became independent, but the United Kingdom retained its northeastern counties within its sovereignty. Facing the Poles’ independence movement in the nineteenth century, Russian Slavophiles, such as Yurii Samarin, requested that the Poles be satisfied with an ethnographical Poland while abandoning their traditional ambition to restore a historical Poland (Rzeczpospolita) together with Ukraine, Belarus, and Lithuania (Samarin 2013, 454–455, cited by Yamamoto 2016, 37).

Yet the Union Law of April 3, 1990, did not play a relevant role in the subsequent dissolution process of the USSR. Why? The five most convinced secessionist union publics (the three Baltic, Georgian, and Azerbaijan republics) pretended that they were restoring their pre-Soviet statehoods, but not seceding from the Soviet Union, and that they were therefore not obliged to follow the procedure prescribed in the Union Law of April 3, 1990 (let us call this discourse restorationism). Among independence-oriented union republics, only the Armenian SSR declared independence via a referendum prescribed by the April 3, 1990 Law (Shiokawa 2021, 2069).

An important corollary of restorationism was that these union republics denied the legal continuity from the SSR to the emerging new state. They argued that the Soviet legal system was a result of occupation and therefore retrospectively invalid. This rhetoric could deliver a suicidal blow to territorial integrity of the Lithuanian, Georgian, and Azerbaijan SSRs because their territories were larger than their pre-Soviet precedents. Sajudis leaders, who came to power as a result of the republican Supreme Soviet (Seimas) elections in February 1990, legitimately feared that restorationism would open the way for Poland to retrieve Vilnius and that the ethnic Poles’ secessionism in the Pole-dominant counties surrounding Vilnius would intensify. On March 11, 1990, the Lithuanian Seimas restored Lithuania’s pre-Soviet statehood (the 1938 Constitution) but on the same day practically reintroduced the Brezhnev Constitution (1978), within the limit that it did not contradict Lithuania’s independence (Sovetskaia Litva, March 13, 1990).

In contrast to Lithuania’s tactful maneuvering, the Georgian and Azerbaijan Supreme Soviets did not take measures to soften the legal consequences of restorationism. Identifying the emerging Georgian and Azerbaijan statehoods as the restored Georgian and Azerbaijan Democratic Republics, which existed for a few years from 1918 to 1920/21, these SSRs denied succession of statehood from the Georgian and Azerbaijan SSRs to independent Georgia and Azerbaijan. Objectively, these SSRs negated the adoptability of uti possidetis juris, a principle of international law requesting automatic transfer of administrative borders of the precedent state into state borders.

Because the UN ignored the fact that Georgia and Azerbaijan, the most significant beneficiaries of uti possidetis juris, rejected adoption of this legal norm, their hyperbole caused no diplomatic disadvantage. Yet this cannot be said for their subordinate autonomies. The South Ossetian Oblast Soviet set forth the counterargument that South Ossetia became a constituent of the Georgian Soviet Republic after what Georgia called “Soviet occupation.” If the Georgian Republic denied the effectiveness of Soviet law retrospectively, South Ossetia’s belonging to the Georgian Republic would lose judicial basis as well. The South Ossetian Soviet found no other way but to become directly subordinated to the USSR authorities, bypassing the Georgian Republic (Moskovskii 2008, 185–186). This logic was shared by other secessionist autonomies as well—by Karabakh, Abkhazia, and Transnistria (Grosul, Babilunga, and Bomeshko 2001, 48–49; “Deklaratsiia” 2008, 703; Moskovskii 2008, 112–115).

After the attempted August 1991 coup, autonomies were excluded from the nationwide debate on an “innovated union” and became orphans (de facto states) after the quick demise of the Soviet Union. The opportunities for peaceful territorial adjustments between the post-Soviet states were lost and confrontations between the former union republics and autonomies became extremely violent.

Quickly Born International Society Indifferent to Empirical Statehood

After the dissolution of the Soviet Union, the international community quickly accepted the fait accompli. The UN granted membership to all ex-union republics of the USSR, though four of them were experiencing serious secession conflicts and two of them (Azerbaijan and Georgia) were in overt civil wars. It is true that Georgia’s UN accession was delayed because the Russian Federation, the successor to the Soviet Union as a permanent member of the UN Security Council, opposed it for the reason of hostilities in South Ossetia. Yet, after the Dagomys Ceasefire Agreement (June 24, 1992) put an end to the First South Ossetian War, Russia changed its attitude. On July 31, Georgia became a UN member, whose first great task was to go to war with Abkhazia two weeks later (on August 14, 2022).

On March 2, 1992, the UN General Assembly decided to accept Azerbaijan as a new UN member, based on the recommendation of the Security Council meeting on February 14, 1992. At this meeting, the chairman made the following statement: “Azerbaijan’s solemn commitment to uphold the Purposes and Principles of the Charter of the United Nations, which include the principles relating to the peaceful settlement of disputes and the non-use of force, is noted with great satisfaction by members of the Council.” ² It is difficult to believe that Security Council members were too naïve to be aware that this kind of “solemn commitment” was meaningless without obliging Azerbaijan to realize a ceasefire in the Karabakh region as a precondition for its UN accession.

The UN’s extremely tolerant attitude toward ex-union republics of the demised Soviet Union had a historical and judicial background. According to Robert Jackson, international norms expanded from Europe to other parts of the globe, from the Paris Conference in 1856 to the establishment of the League of Nations in 1919. At that time, having their own sovereign state and gaining a seat at the League of Nations meant enfranchisement to a club of elite nations. New states were requested to fulfill empirical criteria of statehood (effective control of a territory, citizens’ well-being, minority rights protection, prohibition of slavery, prevention of mass killings, etc.) to be recognized as a state and to become a member of the League of Nations. This situation changed after World War II. The establishment of the United Nations consolidated an image of international society composed of universally existing, equal sovereign states. This image, combined with the very existence of the UN, became a driving force decolonizing the world. UN General Assembly Resolution 1514 (“Declaration on the Granting of Independence to Colonial Countries and Peoples”) in 1960 made empirical attributes of statehood irrelevant. Recognition of new states has become de jure (formalistic), but not de facto (substantive) (Jackson 2000, 296; Berg & Kursani 2022, 14, 16). The criteria for UN accession became much lower than those for League of Nations accession. The League of Nations did not accept the Georgian and Azerbaijan Democratic Republics as members because of territorial conflicts within them, whereas, as mentioned above, post-Soviet Azerbaijan and Georgia could become UN members easily, despite the civil wars ongoing there.

The new international society has produced legal norms, such as the UN Charter and CSCE Helsinki Final Act of 1975, yet the quick emergence of globe-wide international society indifferent to empirical statehood made selective adoption of these norms inevitable. Article 4 of the UN Charter reads, “Membership in the United Nations is open to all other peace-loving states which accept the obligations contained in the present Charter and, in the judgment of the Organization, are able and willing to carry out these obligations.” In fact, “will” can be proclaimed emptily and “ability” is not examined when a country applies for UN membership. Once a state has become a UN member, it will be protected by Article 2 prescribing sovereign equality, territorial integrity, and nonintervention in domestic matters. It is this very discrepancy in compliance enhancement between Articles 2 and 4 that generates fertile soil for failed states—a structural problem having emerged repeatedly in the decolonization of Asia and Africa and the dissolution of the Soviet Union and Yugoslavia. Most failed states are successor states of defunct colonial empires or communist federations (Jackson 2000, 296).

Likewise, the 10 articles of the Helsinki Final Act together represent a systematic worldview. Its Articles 3 (inviolability of frontiers) and 4 (territorial integrity of states) should be interpreted in harmony with Articles 7 (respect for human and minority rights) and 8 (self-determination of people). If a state violating Article 7 may defend itself by referring to Articles 3 and 4, it will cause injustice. In the worse cases, Articles 3 and 4 can be used to justify massive violence, as was the case with many post-Soviet secession conflicts.

Noting the differing compliance enhancement between Articles 2 and 4 of the UN Charter, Jackson makes an interesting counterfactual assumption. Opposite to the real world, were Article 2 not respected and Article 4 not disregarded, failed states would probably not exist. Instead, colonies, protectorates, trust territories, and other kinds of dependent states would still be in existence. Jackson thinks that the examples of conflict solutions in Bosnia and Kosovo show that the world is returning to an era of international trusteeship (Jackson 2000, 297).

Federalization

Table 1 highlights definition, examples, merits, and demerits of the five recipes to settle (mainly post-Soviet) secession conflicts.

Table 1. Five Recipes to Solve Secession Conflicts

The desire of parent states and the international community to consolidate the Soviet administrative border into a border of newly independent states facilitated the federalist (power-sharing) method. ³ According to this method, the breakaway polity disarms and disbands itself, whereas the parent state is federalized to retrieve the breakaway polity. The federalization arrangement fits the liberal assumption of the Helsinki Final Act, which believes it possible to solve secession conflicts without changing state borders by making states multicultural and decentralized. In fact, contemporary nationalisms have proved themselves to be more communal and intolerant than the international organizations expected (Broers 2019, 295–307).

Federalization policy often causes commitment problems. After a secession polity disarms and dissolves itself, there is no guarantee that its parent state will keep its promise of federalization. To prevent commitment problems from occurring, federalization policy requires a strong arbiter standing above parties of conflict and armed enough to enforce the agreement in full scale. For example, during the Civil War in Russia of 1918–1921, Karabakh, South Ossetia, and Azerbaijan were placed under the effective control of the Azerbaijan and Georgian Democratic Republics. In 1920–1921, the Bolsheviks overthrew these regimes and established Soviet powers in South Caucasus. The Bolsheviks at the Central and South Caucasian levels confirmed the effective control of the three disputed regions that the Azerbaijan and Georgian SSRs inherited from the precedent states but, instead, obliged the latter to grant the disputed regions autonomous status. The Azerbaijan Soviet leaders were happy with the first part of the decision to subordinate Karabakh to them (not to Armenia) but, understandably, neglected the second part requesting the creation of Karabakh Autonomous Oblast. In 1923, the South Caucasian Regional Party Committee ordered the Azerbaijan SSR to create it, and this order was implemented in 1924 (Saparov 2015, 111, 117).

Another example of the need for coercion to implement a federalization policy is the 1995 Dayton Agreement, which put an end to the Bosnian War. Whereas UN Peacekeeping Forces were not effective enough to halt the Serbian forces’ offensive, NATO’s active involvement in the Bosnian conflict since 1994 had a game-changing effect. Operation Storm by the Croatian government, supported by the US, liquidated the Srpska Krajina Republic and left the Serbians in Bosnia no alternative but to accept the Dayton Agreement (Jackson 2000, 270–274; Toal & Dahlman 2011, chap. 5).

Even when we can reckon on the presence of an arbiter armed enough to prevent commitment problems, the federalization policy often contradicts the conflict parties’ real intention. On one hand, the parent state, in its heart, feels relieved by the breakaway of the trouble-making entity and does not think it necessary to federalize itself to retrieve the prodigal son. Especially when the breakaway region has a significant demographic and economic weight, its reintegration may drastically change the political balance of power in the parent state’s domestic politics, as is the case with Ukraine/Donbas and Moldova/Transnistria. This is why the Moldovan government does not hurry for “solution” of the Transnistrian conflict but waits for “natural death” (socioeconomic and demographic catastrophe) of Transnitria, aware that time goes in Moldova’s favor (Serebrian 2022). ⁴

The Minsk Accords were a product coauthored by German Kanzler Angela Merkel and French President François Hollande, who feared to change state borders, and Putin, who then tried to stop Ukraine’s NATO accession from within by pushing Donbas back to Ukraine. Founding fathers of the Donetsk People’s Republic (DPR), skeptical of the Minsk Accords, were mercilessly purged from the DPR leadership under the patron state Russia’s pressure (Matsuzato 2022, 52–56). For Ukraine, the Accords were no more than a compromise forced by its military fiascos in Ilovaisk in August 2014 and Devaltsevo in February 2015. The return of the People’s Republics with more than three million voters would have changed the electoral balance in Ukrainian politics and made it impossible to continue the post-Euromaidan course (D’Anieri 2018).

On the other hand, a breakaway community often prefers a de facto independence to any legalized dependent status. They will agree to be downgraded to a federal constituent only when the arbiter promises a future, almost confederation with constituents’ strong veto power, as was the case with the Dayton Agreement and Kozak’s memorandum to reunite the left and right banks of Moldova (Hill 2012, chaps. 8–11). Unsurprisingly, confederation is unacceptable to the parent state. ⁵

Parties of conflicts often receive generous financial aid from international organizations and mediator states. So they do not confess their real intention to these aid providers even if they regard the federalization policy as hopeless. The international organizations and mediator states, in turn, may not notice or make the appearance of not noticing the approaching crisis behind stagnant negotiation, as was exactly the case with the (non-)implementation of the Minsk Accords on the Donbas conflict during 2015–2021.

Last, federalization does not work when the parent state does not recognize the breakaway polity’s subjectivity. For Azerbaijan and Georgia (since 2008), Karabakh, Abkhazia, and South Ossetia are mere subjects of foreign (Russia or Armenia’s) occupation. In their view, there is no room for a federal solution as long as this occupation continues.

Land-for-Peace and Land to be Sacrificed

The poor performance of the federalization policy to solve secession conflicts has pushed forward another recipe—land-for-peace, whereby the breakaway polity sacrifices part of the land it controls to the parent state to receive the latter’s recognition. In contrast to federalization, land-for-peace arrangements take changes in state borders for granted. As is shown in Table 1, land-for-peace is becoming all the more popular among policy makers. The aforementioned book coauthored by Berg and Kursani (2022) investigates various attempts to adopt land-for-peace. Land-for-peace is more realistic than federalization in the sense that neither the breakaway polity nor the parent state needs to change their existing political regimes. Indeed, this recipe requests neither the parent state’s artificial federalization nor the bringing back of the trouble-making breakaway region, which in turn does not need to abandon its de facto independence. Moreover, the land-for-peace arrangement synchronizes the border change and recognition of the breakaway polity by the parent state, so prevents commitment problems from taking place. Yet land-for-peace also has demerits listed in Table 1.

A serious limitation of land-for-peace is that this policy can be realized only when the secession polity controls certain land that can be sacrificed to the parent state in reward for recognition. Let me examine this condition concerning five post-Soviet secession conflicts.

Karabakh

In the latter half of the First Karabakh War, during 1993–1994, the Karabakh armed forces marched beyond the border of the former Nagorny Karabakh Autonomous Oblast’ and occupied Azerbaijan’s internal territory, which amounted to almost twice the size of NKAO’s own territory. ⁶ This was an exceptional event for post-Soviet secession conflicts, which usually end when the de facto state restores the former Soviet autonomy’s territory. As was the case with the First Abkhazian War and the Second South Ossetian War, the de facto state does not dare to march into internal regions of its parent state (former union republic) even when it enjoys favorable military conditions. The occupation of Azerbaijan’s internal territory was reckless behavior in which Karabakh paramilitaries engaged against the Armenian leaders’ objection (Babayan 2022). This adventurism put Armenia and Karabakh in a diplomatic predicament after the war but gave Karabakh huge territory usable for deals with Azerbaijan. As a result, the Karabakh peace talks had become a forerunner of land-for-peace arrangements since the 1990s.

South Ossetia

The 1992 Dagomys Agreement clearly divided the territory of the former South Ossetian AO, subordinating ethnic Ossetian and Georgian settlements to the South Ossetian and Georgian governments, respectively. Therefore, the South Ossetians did not have any land to be sacrificed to Georgia.

Abkhazia

As mentioned above, Abkhazia restored the former autonomy’s territory by the end of the Abkhazian War, which included Gali County with a Mingrelian population, regarded as a branch of the Georgian nationality. Therefore, there emerged the possibility of gaining Georgia’s recognition of Abkhazia’s independence by sacrificing Gali County to Georgia.

Transnistria

For Transnistria, perhaps, Bendery City could be an object of land-for-peace because this city is located on the right bank and belonged to Romania during the interwar period. In 2004, however, the Russians and Ukrainians constituted more than 60% of the city population. The city suffered the largest number of victims among Transnistrian cities during the military conflict of 1992, so concern about the Pridniestr Moldovan Republic’s legitimacy as a state will not allow its leaders to sacrifice this city to Moldova. Last, according to the Moldovan authorities, depopulation of Transnistria raised Bendery’s demographic portion in Transnistria to one-third (Serebrian, 2022).

Donbas

Though Minsk-II was shelved for seven years, no party proposed any alternative, let alone alternatives based on land-for-peace. Counterfactually, a possible land-for-peace arrangement would have been to recognize the publics’ secession from Ukraine with the territories that they effectively controlled then. In other words, on the condition that the DPR and Luhansk People’s Republic (LPR) changed their constitutions to abandon their pretensions to Mariupol, Kramatorsk, Severodonetsk, and other Ukraine-controlled territories of the former Donetsk and Luhansk Oblasts, their separation could have arisen for discussion. In hindsight, as I believe, this was the only way to evade a repeated war, but such a peace plan, even had it been made, would have pleased none. Perhaps, German and French representatives feared border changes much more than a possible future war. At least until 2019, Donbas within Ukraine interested Putin more than an independent Donbas. The status quo, or an unresolved situation, was comfortable for both Ukraine and the DPR/LPR. So, all parties continued casting Minsk-II like a spell, though none of them believed it viable.

Overall, only Karabakh and Abkhazia possessed the controlled land usable for land-for-peace settlement with their parent state. Yet, even in these regions, secession conflicts could not be solved peacefully. Further, we trace the process through which unsuccessful land-for-peace arrangements for these breakaway regions resulted in Russia’s protectorate policy on Abkhazia and South Ossetia in 2008 (the third recipe) and in Azerbaijan’s reconquering of Karabakh in 2020 (the fourth recipe).

Unsuccessful Land-for-Peace to a Protectorate Arrangement—Abkhazia and South Ossetia

During 2007–2008, the Saakashvili administration’s quest for Georgia’s NATO accession intensified the Russo-Georgian talks on redivision of Abkhazia and South Ossetia. The Russian Foreign Ministry under Sergei Lavrov illusorily thought it possible to prevent Saakashvili’s approach to NATO by helping Georgia to solve its territorial problems.

The Abkhazian and South Ossetian conflicts had some advantages for their solution: (1) Georgia, in contrast to Azerbaijan, did not categorically oppose its own territorial adjustments. (2) Both of the patron and parent states—namely Russia and Georgia—tended to put Abkhazia and South Ossetia on the same plate and then carve them. Therefore, both Russia and Georgia could compensate a possible excessive concession in one of the two disputed territories by gains in the other. (3) Russian politicians and experts, especially the South Ossetians’ “elder brothers” (North Ossetians) hardly believed in the viability of South Ossetia’s statehood (Kochiev 2009). If the South Ossetian statehood needed to be sacrificed anyway, it should be exchanged for maximum gains in Abkhazia.

The Russian leaders’ condescending attitude toward South Ossetia or preference for Abkhazia became even more manifest after the Russian Olympic Committee earnestly pushed Sochi as a candidate venue for the 2014 Winter Olympic Games and Russian leaders began to count on Ochamchire on the Abkhazian coast as a candidate for the Russian Black Sea Fleet’s naval base after 2017, when Russia was obliged to return Sevastopol Port to Ukraine. Russian foreign minister Igor Ivanov (Sergei Lavrov’s predecessor) previously intended to “return” Abkhazia to Georgia while keeping South Ossetia within Russia’s sphere of influence. Because of Sochi, Ivanov reversed his priority (Baburin 2010). ⁷

Ideas of dividing Abkhazia into Abkhazia- and Georgia-dominated parts have their beginnings in the late Soviet period (Markedonov 2008). Various partition plans were proposed during the Abkhazian War during 1992–1993. ⁸ After the war, the UN, the main conflict-regulating institution in Abkhazia, dogmatically pursued federalization arrangements, so the rare advantage of the Abkhazian conflict that the parent state, Georgia, was flexible toward its own territorial adjustments could not be exploited. In 1997, after repeated failures of a federalization arrangement, the Republican Party, the then opposition to the Shevardnadze administration, published a peace plan to divide Abkhazia almost equally. According to this proposal, the southeastern half of Abkhazia would be returned to Georgia, leaving its northwest to an “Abkhazian Republic.” After the Rose Revolution of 2003, the Republicans developed the contents of the 1997 plan. The Saakashvili administration criticized this opposition proposal but, in 2008, plagiarized and proposed it to Russia and Abkhazia (Haindrava n.d., 12).

In 2007, Russian specialist on the Caucasus A. E. Semirechnyi argued that the Abkhazian/South Ossetian conflict should be managed in the context of Georgia’s attempts for NATO accession. According to him, the federalization policy pursued by international organizations was not acceptable for Abkhazia and South Ossetia at all because it would establish vertical intergovernmental relations between Georgia and Abkhazia/South Ossetia. Instead, Semirechnyi proposed to recognize a horizontal coexistence of the three governments of Georgia, Abkhazia, and South Ossetia, joined by a treaty or association, in the internationally recognized border of Georgia. Special status would be granted to the territories with a Georgian majority in Abkhazia and South Ossetia, such as the Kodor Valley, Gali District, “Georgian enclaves” surrounding Tskhinval, and Leningor District. If Georgia did not agree with this solution or if Georgia became a NATO member in the future, Russia should unilaterally recognize the independence of Abkhazia and South Ossetia (Semirechnyi 2007). Thus, Semirechnyi explicitly rejected the federalist solution and made a proposal similar to land-for-peace vis-à-vis regions with a Georgian majority.

In February 2008, I interviewed Aleksander Rondeli, director of the Georgian Foundation for Strategic and International Studies and President Mikheil Saakashvili’s advisor. On the verge of the Bucharest NATO Summit in April, which resolved to approve Georgia’s near-future candidacy for NATO membership, he told me,

[r]epresentatives of the Georgian and Russian Foreign Ministries meet in this building of the GFSIS when it is inappropriate to negotiate in the Georgian Foreign Ministry building… . If John McCain wins in the coming US presidential elections, Georgia will join NATO within three years. Even if Hilary Clinton wins, the same will happen within four years. Gaining NATO’s support, Georgia will internationalize the Abkhazian issue and force the Russians to negotiate with us as equals. We may be flexible as to the contents of solution. It will be acceptable that Georgia takes Gali and Ochamchire Counties and gives Russia the north of them. (Rondeli 2008)

Though Rondeli reserved that he was talking on his personal opinion, it seemed to me that he was exposing the content of the then ongoing informal Russo-Georgian talks, which produced a tentative peace plan by May 2008. The Russian leadership invited Abkhazian president Sergei Bagapsh to Moscow, showed him the plan, and asked his opinion.

The newspaper Kommersant uncovered its contents as follows, on which Sergei Markedonov commented: (1) Officially, the whole of Abkhazia will continue to be Georgia’s domain; (2) In practice, Abkhazia will be divided by the Kodor River and the southeastern counties (Gali, Ochamchire, and Tkvarcheli) will be passed to Georgia. In the north of them, a pro-Russian, de facto independent state, Abkhazia, will be created; and (3) The CIS Peacekeeping Forces now stationed by the Ingur River will be pulled back to the west bank of the Kodor River. Instead, in the three southeastern counties Georgia and Abkhazia will create a joint police force (Kommersant 2008, June 27; Markedonov 2008; See also Fuller 2008). The three southeastern counties to be passed to Georgia only amounted to 33% of the whole Abkhazian territory, so it seems that Georgia conceded too much. Yet, if we add up the territories of the three counties with South Ossetia, it turns out that Georgia and Russia agreed to divide Abkhazia and South Ossetia approximately equally. ⁹

At that time, Kommersant and Markedonov supposed that, if this peace plan were implemented, Saakashvili would be able to advertise that he had restored Georgia’s territorial integrity and saved the Georgians’ coethnics, the Gali Mingrelians left beyond the border of Georgia’s effective control. Kommersant believed that Russia’s assistance in restoring Georgia’s territorial integrity would weaken Georgia’s desire for NATO accession, whereas Markedonov more realistically thought that Georgia would become a NATO member in any case, so the existence of an independent Abkhazia as a buffer state between Russia and Georgia would be beneficial for Russia’s security. It is obvious that Kommersant shared the then Russian Foreign Ministry’s optimist illusion, whereas Markedonov’s view was closer to reality. Yet the very fact that the Russian Foreign Ministry had an illusory perspective provided a rare, favorable environment for the land-for-peace arrangement.

Markedonov (2008) and Berg and Kursani (2022, 164) argue that the peace plan of May 2008 would have released Abkhazia from the Mingrelian question, a source of headache for Abkhazian leaders, and instead make it possible for Abkhazia to gain international recognition. Yet Abkhazian President Bagapsh, summoned to Moscow, reacted to the peace plan negatively. The Abkhazian government had been arguing for 10 years until then that Gali Mingrelians were one of the native peoples of Abkhazia, which was by nature a multinational state. Bagapsh was one of the first advocates of this multicultural discourse and indeed thanks to the Gali Mingrelians he won the 2004 presidential election. It seemed improbable that he would sacrifice southeastern Abkhazia to solve the Mingrelian question according to Georgia’s nationality discourse (Matsuzato 2009, 253; Matsuzato, 2011, 815–16).

As a whole, the international environment in the first half of 2008 was advantageous for Georgia. Since 2008, researchers continue to discuss why Saakashvili launched a reckless war when it seemed possible to restore Georgia’s territorial integrity via peace talks. ¹⁰ A possible explanation is that 2008 was a year of American presidential election. Outdone by Hilary Clinton first and Barack Obama later, the Republican Party and George Bush Jr.’s administration needed to pursue a hard line vis-à-vis Russia. The same can be said for Saakashvili, suffering from a decline in his approval rate since he coercively suppressed protests in Tbilisi in November 2007. I would restate here that the Russian and Georgian leaders treated the Abkhazian and South Ossetian problems as one secession conflict. The rumor that Russia was ready to “return” South Ossetia to Georgia was widespread even among the South Ossetians at that time and this was a reason, along with Georgia’s successful military and economic pressure, for the South Ossetians’ massive outflow from the region on the eve of the 2008 war. This situation might give Saakashvili an erroneous message that Russia would not intervene even if he liquidated South Ossetia by force.

Russia’s Quick Shift to the Protectorate Arrangement on Abkhazia and South Ossetia

After the Second South Ossetian War, there was a serious disagreement among the Russian leaders on the possibility of recognizing South Ossetia and Abkhazia’s statehood. Some were apt to do so to prevent further military attacks from Georgia, whereas others, the Russian Foreign Ministry above all, legitimately anticipated that few countries in the world would follow Russia’s recognition of Abkhazia and South Ossetia and, therefore, the problem of these territories will not be solved. However, because Russian TV broadcast atrocities in South Ossetia during the hostilities, anger and compassion swept Russian society and eventually overwhelmed the realists within the government.

After the hostilities ended in the former SOAO territory, Russian troops marched into Georgia proper and occupied part of the Shida Kartli region until August 22. During this short period of occupation, Russian president Medvedev asked Abkhazian president Bagapsh to make Abkhazia return to Georgia because Russia, this time decisively, would request Georgia to transform itself into a confederation (Gurguliia 2017). Perhaps Medvedev made the same proposal to South Ossetian President Eduard Kokoity. Possibly, Russia occupied Shida Kartli as a hostage to enforce Saakashvili’s agreement with the confederalization of Georgia. It seems unrealistic if Medvedev thought it possible to solve the Abkhazian and South Ossetian questions by pushing them back to the imagined Georgian confederation after the severe hostilities, but perhaps Medvedev was desperately trying to evade unilateral recognition of these polities as a bad move. Because Bagapsh (and perhaps Kokoity as well) rejected Medvedev’s proposal immediately, the Russian leadership did not find any alternative but to have its army retreat from Georgia proper on August 22 and recognize Abkhazia and South Ossetia on August 25.

In hindsight, after witnessing the outbreak of the Russo-Ukrainian War in 2022, we may legitimately speculate that the Russian leaders had another option to be pursued, apparently contradictory to the creation of a Georgian confederation. Russian troops could have marched to Tbilisi and forced Saakashvili to recognize the secession of South Ossetia and Abkhazia, as the Russian delegation requested Ukraine to recognize the secession of Crimea and Donbas from Ukraine in the Istanbul peace talks on March 29, 2022 (the fifth recipe described below). Yet, in 2008, Russia lacked sufficient military and economic resources to do so.

Table 1 shows the advantages and disadvantages of Russia’s protectorate policy for South Ossetia and Abkhazia. It is true that Russia’s protection allowed these de facto states to concentrate on tourism and other civilian businesses. However, as the Foreign Ministry of Russia feared, relations between the patron state (Russia) and the parent state (Georgia) worsened irreversibly. Had Russia not behaved so, Russo-Georgian relations would have improved under the Georgian Dream government since 2012.

Because only a handful of countries in the world followed Russia’s recognition of Abkhazia and South Ossetia, this move did not normalize the military, political, and economic situation in the South Caucasus. For example, indefinite closure of the Transcaucasian Highway since 2008 has been causing tremendous loss to the post-Soviet, Turkish, and Iranian economies. The high cost of transportation is passed to the consumers of these countries. Even in the 1990s, when both South Ossetia and Russia were extremely poor, South Ossetia could survive thanks to income from the Transcaucasian Highway. Having lost this source of income, South Ossetia became even more dependent on subsidies from Russia. Russia may be capable of feeding the about-forty-thousand population of South Ossetia but would definitely be unable to do so vis-à-vis the DPR/LPR population of between three and four million on the eve of the 2022 war.

In sum, making de facto states its protectorates by unilaterally recognizing them is neither an advantageous nor permanently affordable policy for the parent state. Wherefore another, ultimate “solution” comes to the fore—that is, destroying the parent state militarily and making it recognize the breakaway polities (the fifth recipe).

Unsuccessful Land-for-Peace to Military Reconquest: The Second Karabakh War

It is often the case that tangible disparity in military and economic resources between a breakaway polity and its parent state culminates with the parent state’s reconquest of the breakaway polity. This scenario has many examples, such as the Biafra War in 1967–1970, destruction of the Krajina Srpska Republic in 1995, surrender of the Liberation Tigers of Tamil Eelam in Sri Lanka in 2009, and the Second Karabakh War in 2020.

A significant merit of reconquest is that this is the most clear-cut solution, leaving no seed for future troubles. According to the widespread interpretation of international law, the breakaway polity is illegal by nature, so the international community will recognize the parent state’s military victory. This is demonstrated by the fact that, during and after the Second Karabakh War, no country in the world, including Russia and France, cochairs countries of the OSCE Minsk Group, renowned for their pro-Armenian inclination, remarked on Azerbaijan’s violation of the Bishkek Ceasefire Protocol of 1994, which Azerbaijan itself signed. Because the enemy population is wiped out of the territory, future security threat to the parent state will also be uprooted.

A demerit of reconquest, besides huge human sacrifices accompanying this policy, is its infeasibility. Though a parent state is usually richer and better armed than its breakaway polity, secession conflicts often confuse politics and damage the economy of the parent state and make it difficult for it to recover their strength enough to conquer the breakaway polity. In the post-Soviet context in particular, the very existence of secession conflict induces the West to heroize and spoil the parent state, which thus loses a self-critical attitude and geopolitical sensitivity and is eventually driven into deep debt. Azerbaijan was an exception in the sense that, while enjoying abundant military aid from Turkey and oil income, the lack of Western spoiling kept the Azeris sober. Meanwhile, Armenia after the populist revolution in April 2018 destroyed its own defense capacity. Neither Ukraine, nor Moldova, nor Georgia enjoys these favorable conditions.

Despite the presence of a huge occupied territory, usable for deals with Azerbaijan, the Karabakh conflict could not be solved by land-for-peace arrangement. Texts of peace plans became all the more sophisticated, from the 1996 OSCE Lisbon Summit, to the 2007 Madrid Summit and revised Madrid Principles, but a fundamental contradiction that “Azerbaijan will never accept Karabakh’s independence, but Karabakh will never abandon its desire for independence” could not be solved. Rather, peace plan texts became long and sophisticated to hide this fundamental contradiction.

In Azerbaijan’s perception, Karabakh is neither an unrecognized state nor a secession polity but just an occupation of Azerbaijan’s territory by Armenia, so recognition of Karabakh as a state cannot be an option. During the First Karabakh War, Azerbaijan recognized Karabakh’s subjectivity and Karabakh was one of the signers of the Bishkek Ceasefire Protocol of 1994, but, when the former Karabakh president Robert Kocharyan became Armenia’s president in 1998, he declared himself to represent Karabakh’s position, too, and practically deprived Karabakh of its status of party of conflict (de Waal 2013, 274). After the April 2018 Revolution in Armenia, new Prime Minister Nikol Pashinyan tried to invite Karabakh back to the negotiation table, but Azerbaijan rejected any recognition of Karabakh’s subjectivity.

According to the revised Madrid Principles, Karabakh was obliged to return five of the seven occupied counties, located on the internal side of Azerbaijan, and then a referendum would be held in Karabakh to determine its status. Subsequently, Karabakh would return two occupied counties located between Karabakh and Armenia (Lachin and Kalbajar) to Azerbaijan with the exception of a necessary corridor joining Armenia and Karabakh (Broers 2019, 293). Yet neither peace plan documents nor the Minsk Group cochairs (US, France, and Russia) answered a simple question—can the option of Karabakh’s independence be printed on the ballot of the referendum? Aware that Azerbaijan would never agree to include Karabakh’s independence as an option, Karabakh returned to Azerbaijan none of the occupied territories. If Karabakh’s fate could only be determined by a soon-to-be resumed war, and neither by peace talks nor a referendum, the occupied counties meant for Karabakh a lifeline for defense. Even an opposition activist from Karabakh says, “Eventually, the occupied territories saved Karabakh. Had the Second Karabakh War begun from the NKAO (a Soviet-era autonomy) borderline, Karabakh would not have existed today” (Grigorian 2022). As time passes, the Minsk Group cochair countries, including Russia, became so irritated that they began to regard Karabakh and Armenia as beneficiaries of the status quo.

Because of the Obama administration’s unsuccessful management of the Syrian crisis and the subsequent Trump administration’s isolationism, the Syrian crisis began to be managed by the Astana Process (negotiations between Russia, Turkey, and Iran) since 2017. The fact that radical Islamists, once within a whisker of seizing Damascus under Turkey’s tacit protection, had become contained in Idlib means that Turkish President Recep Erdogan conceded to Putin more than Putin did to Erdogan. Erdogan began to request compensations from Putin in the Karabakh conflict. To my mind, however, a saliently pro-Turkish attitude is peculiar to Russia’s presidential administration or personally Putin, whereas Russia’s Defense and Foreign Ministries, security organ, and army continue to hold the Turkey threat theory, a historical instinct of Russian bureaucrats and militaries since the tsarist period. So there was much room for Armenian diplomacy to regain Russia’s support, but the Pashinyan government lacked sufficient knowledge of international affairs and the Karabakh question. ¹¹ This government looked on the Armenian army, construed as close to the old regime, as the enemy and devastated Armenia’s defense capacity within only two years.

Destruction of Parent State by Patron State: The Russo-Ukrainian War of 2022

A historical precedent of Russia’s attempt to destroy Ukraine “to save Donbas” in 2022 is the NATO bombing of Yugoslavia “to save Kosovo” in 1999. Both military actions were motivated by the irritation of leaders (Bill Clinton and Putin) that the half-way measure of protectorate policy (the Dayton Agreement for the United States and recognition of South Ossetia and Abkhazia for Russia) did not eliminate the root of evil and could not establish a desirable regional order. In other words, an unsuccessful protectorate policy (the third recipe) induced the leaders to proceed to a more radical, “final solution.” ¹²

A demerit of the solution to destroy the parent state is that, in contrast to ordinary secession conflicts, which affect the peripheral territories of the parent state, hostilities between the patron and parent states take place in the parent state’s core areas, which causes tremendous human sacrifices. Moreover, with the exception of Kosovo, even if the patron state and the breakaway polity win the war, the international community will not recognize the result and the conflict eternalizes.

Russia’s pro-Putin mass media and the Ukrainian opposition criticized Zelenskyy for breaking his promise of Donbas peace given in the 2019 presidential election. In fact, however, Zelenskyy was skeptical of the Minsk Accord even before the election and criticized Poroshenko for failing to propose an alternative despite the obvious ineffectiveness of the accord (Shramovich 2019). Having won the presidential election, Zelenskyy proposed to “reset” the Minsk Accord (Baranovskaia 2019). The contents of this reset were publicized in the autumn of the same year. Zelenskyy proposed to reverse Steinmeier’s formula and restore control of Ukraine’s eastern border first and then hold local elections in Donbas. He did not find it necessary to amend the Constitution to grant Donbas special status because the decentralization reform ongoing in Ukraine proper was sufficient to retrieve Donbas. The president would not pardon Donbas leaders and activists. Thus, Zelenskyy completely denied the contents of the Minsk Accord (TASS 2019).

During the summit in Paris, based on the Normandy Format, in December 2019, Zelenskyy rejected Russia’s interpretation of Minsk-II and made Putin furious. An article published in The Wall Street Journal in April 2022 supposes that this summit was the turning point eventually leading Putin to warring against Ukraine (Gordon 2022). It is difficult to support this assumption unconditionally because immediately after the Paris Summit Putin replaced his advisor Vladislav Surkov in charge of the Ukrainian, Donbas, Abkhazian, and South Ossetian affairs since autumn 2013 with Dmitry Kozak. Reportedly, Surkov was blamed for excessively supporting the independence of the People’s Republics and thus estranging Ukraine while Kozak was Putin’s trusted troubleshooter who nearly reunified Moldova and Transnistria by his memorandum in 2003 (Hill 2012) and helped to make a grand coalition of the Socialists and Liberals to terminate Vlad Plahotniuc’s government in Moldova in 2019 (Solov’ev 2020). If Putin had decided to go to war with Ukraine in December 2019, he would not have changed the hardline negotiator (Surkov) to a moderate (Kozak).

Indeed, in the first half of 2020, despite the pandemic, Ukraine, the People’s Republics, and Russia cooperated to implement technical agreements reached at the Paris Summit: POW exchanges, mine-clearing, increasing checkpoints on the military border, etc. (Oleshchuk 2020). Since the autumn of 2020, however, Ukraine’s policy vis-à-vis the DPR/LPR became tough again. The Ukrainian leaders pretended that they had worked out a plan to “de-occupy Donbas” (Interfaks-Ukraina 2019) and began to refer to “Plan B” in a suggestive manner. Why did this stiffening take place after more than half a year of thaw? One reason was the revitalizing Southeast opposition (former Party of Regions), who could not even proceed to the final round in the 2019 presidential election and had no voice in the process up to the Paris Summit.

Another, more substantial reason would seem to be Azerbaijan’s overwhelming victory in the Second Karabakh War during September–November 2020. Armenian Prime Minister Pashinyan attributed this defeat, caused by his own self-destructive policy, solely to Armenia’s unpreparedness for unmanned aircraft and drones. This excuse caused a worldwide Bayraktar syndrome, which influenced the Zelenskyy administration as well. Zelenskyy and his men became convinced that it was possible to solve another secession conflict in Donbas by force and that unmanned air vehicles and drones could defeat tanks and howitzers, with only the former being correct. In 2021, the Zelenskyy administration increased the import of Bayraktars, which had started under Poroshenko, and even built a joint factory with Turkey to start domestic production of Bayraktars (Izvestiia 2021). Zelenskyy desired to be blessed by lucky charms.

If the ineffective federalization policy induced Ukraine to imitate Azerbaijan’s successful reconquest policy, for the same reason, the Russian political and military leaders split into two competing groups: one proposing to make Donbas Russia’s protectorate (the third recipe) and the other requesting the destruction of Ukraine “to save Donbas” (the fifth recipe). On March 25—namely after a month since the beginning of the war—the Russian General Staff held a press conference and declared its intention to concentrate Russia’s military operation on the Donbas Front. At this press conference, the deputy chief of staff for military operation, Sergei Rudskoi, overtly recognized the existence of disagreement among staff officers regarding the main target of operation on the eve of the war. Rudskoi justified the General Staff’s bet on the fifth recipe, arguing that Russia’s geographically stretched military operation prevented the Ukrainian army from sending its main troops to Donbas (MR.RU 2022).

In his TV address on February 21, 2022, Putin recognized the DPR and LPR’s statehood. This was the third, protectorate policy, familiar to observers by the examples of South Ossetia and Abkhazia in 2008. Even for some Ukrainian leaders, this option was not a surprise. ¹³ In the early morning of February 24, however, Putin suddenly declared war on Ukraine and thus shifted to the fifth recipe. General Rudskoi’s report on March 25, Defense Minister Sergei Shoigu’s near disappearance from the media between February 24 and March 25, and the lack of logical necessity of dividing the declaration of war into two parts delivered on February 21 and 24 make me speculate that Putin could not secure the consensus of Russian political and military leaders on executing the fifth recipe and needed to impose a fait accompli on his opponents by forestalling them.

At first, apparently, the level of destruction that Putin’s group of Russian political and military leaders intended was no more than a regime change. The Russian army rushed from Belarus to Kyiv via Chernihiv Oblast to replace the Zelenskyy administration with Russia’s puppet government or make Zelenskyy surrender. Simultaneously, the Russian army rushed from Crimea to Mykolaiv. Its final targets were the South Ukrainian Nuclear Power Plant and the city of Odesa, which had become a peculiar item of Putin’s war purposes. In his February 21 address, Putin promised to organize a (perhaps show) trial of those who set fire to the Odesa Trade Union House on May 2, 2014, which caused more than 40 deaths (Kremlin 2022). At the ceremony of Victory Day on May 9, 2022, Putin included the victims—Odesites—in the objects of the one-minute silence tribute together with victims of World War II and the ongoing war with Ukraine. This looked very strange.

At the Istanbul peace talks on March 29 held after Putin’s “blitzkrieg” failed, the Russian and Ukrainian representatives almost agreed that Ukraine would withdraw application for NATO membership and shelve its pretensions to Crimea and Donbas for 15 years, and Russia promised to have its troops retreat from Kyiv and Chernihiv Oblasts. Yet Ukraine waived this near agreement, and Russia did not simply go back to the third recipe (making Donbas its protectorate). The Russian leadership began to disclose its intention to annex Kherson and Zaporizhzha Oblasts to the Russian Federation to make them a “land corridor to Crimea.” On April 19, a meeting of “representatives” of Rozivka County, the eastern end of Zaporizhzha Oblast, adopted a resolution requesting itself to be incorporated into the DPR (Rossiiskaia gazeta 2022). On April 26, Russia introduced an utter occupation regime in Kherson Oblast where the regional capital’s city hall had functioned until then despite Russia’s military occupation of the region since early March. Thus, the Russian leaders escalated the war purpose from regime change to territorial acquisition rather than retreating from the fifth to the third recipe for solving the Donbas problem.

Meanwhile, the Russian and DPR/LPR armies left Ukraine-controlled, heavily fortified settlements, such as Avdiivka and Mar’inka, surrounding Donetsk City almost intact until July, though from there the Ukrainian army shelled civilian quarters of Donetsk, Gorlovka, and other settlements of the DPR. During the early months of the war, Russia mainly provided assistance from the air, whereas the DPR/LPR armies conducted land battles on the Donbas front. Uneven division of labor between the Russian and DPR/LPR armed forces, modeled after the one between Russia’s Aerospace Forces and Syria’s ground forces in the Syrian War, continued even after Russia’s official declaration to concentrate on Donbas. During the battle in Mariupol, Russian TV often broadcast the Chechen Special Forces’ spectacular performance, but their contribution was often exaggerated and this robbery of feats offended the DPR militaries. ¹⁴ During the first several months of hostilities, the Russian army had preserved its contingents maximally by assigning possible losses of strength to the DPR/LPR armies.

All the above described—namely, Russia’s dispersed military operations in the first month of war, the postponing of mopping-up operations in Avdiivka, and other fortress cities shelling Donetsk and assigning painful land battles to the DPR/LPR army while itself concentrating on aviation assistance—may cause a legitimate suspicion that Putin used the cause of “saving Donbas” for other war purposes. I addressed this question to DPR citizens, whom I interviewed in May and June online (one politician, one military man, one social activist, and two university professors). Surprisingly, all of them answered that, unless the central Ukrainian authorities are made to surrender, Donbas will not be saved from Ukraine’s shelling. Two of them even added that Avdiivka and other fortress cities near Donetsk had been so fortified by Ukraine during the recent eight years that it would have been costly to start a “liberation war” from there. It was rational that Russia started its operation from various “less fortified” parts of Ukraine. It is true that those whom I interviewed compose so-called conscious people aloof from how common people perceive the hostilities. Nevertheless, one may see a social base supporting the view that the recognition of the DPR and LPR (the third recipe) cannot bring peace to Donbas and, for this purpose, it is necessary to make the parent state, Ukraine, surrender (the fifth recipe).

Conclusion

Politicians and intellectuals of union republics dominated the dissolution process of the Soviet Union, leaving little room for dialogue between union republics and subordinated autonomies. Deprived of opportunities for peaceful territorial adjustments with union republics, several autonomies became orphans (unrecognized states) after the sudden demise of the Soviet Union. After decolonization, the UN and other international organizations ceased to examine the empirical parameters of statehood, which made Article 4 of the UN Charter and Articles 7 and 8 of the Helsinki Final Act irrelevant. This tendency in the application of international law also worked in favor of the former union republics (the parent states). It would have been unsurprising if an emergence of secession conflict cast doubt on the parent state’s capacity, yet the situation in the post-Soviet context has been the opposite. The international community often spoiled and even heroized the parent state exactly because of the challenges caused by the de facto state. De facto states accumulated a sense of injustice vis-à-vis international society.

Among the three cases of attempt to overcome secession conflict by federalization (Donbas), land-for-peace (Abkhazia and South Ossetia), or their combination (Karabakh), the Abkhazian and South Ossetian conflict came closest to a solution. Ironically, however, an advantage turned into a stumbling block; Saakashvili interpreted Russia’s readiness to sacrifice South Ossetia and Gali County to secure a pro-Russian independent Abkhazia as a sign of Russia’s hesitance to intervene in hostilities even if Georgia retrieved South Ossetia by force. Azerbaijan’s utter defeat in the First Karabakh War resulted in the vast Karabakh-occupied territories usable in its deals with Azerbaijan. Yet the same reason humiliated Azerbaijan more than any other post-Soviet defeated parent state. A psychological compensation was Azerbaijan’s stubborn denial of Karabakh’s subjectivity, which made both federal and land-for-peace solutions impossible. I see a most serious reason for the escalation of the Donbas conflict in the distance between the mediator countries, which dogmatized the inviolability of borders, and the parties of conflict, which did not believe in federalization requested by the Minsk Accord at all.

Having said this, I do not support at all Putin’s attempt to describe the years from 2015 to 2022, when the Minsk Accord was more or less effective, as a genocidal period. Most of the about fourteen thousand victims of the Donbas War were caused before Minsk-2, in 2014–15. Minsk-2 no doubt mitigated the hostilities and decreased the number of victims. Unfortunately, civilians continued to be killed, though to a limited extent, and a stable peace was not achieved. Yet none of the ceasefire agreements of the four post-Soviet secession conflicts in the 1990s turned into a peace treaty. Nevertheless, these agreements eased the conflicts. Secession conflicts are barely solvable by nature, unless people change the present image of statehood, according to which territorial decrease is perceived as amputation of one’s own limbs. A feasible way for security in post-Soviet contested areas seems to be repeatedly prolonging and modifying the existing fragile ceasefire rather than trying to solve the conflict fundamentally.

Another reservation is that, despite the general momentum for escalation embedded in anticonflict recipes, there was often room for politicians to choose. Saakashvili could continue the negotiation with Russia in the summer of 2008. Putin could choose the third (protectorate), not the fifth recipe (destruction) in 2022. I hope that the matrix presented in this essay, typologizing possible scenarios of secession conflicts, makes politicians aware of where they are, why they are psychologically urged to choose one remedy but not the other, what the possible consequence of the choice is, and what the alternative can be.

We should not forget that the third (protectorate), fourth (reconquest), and even fifth apocalyptic recipes to manage secession conflicts did not derive from a dictator’s evil will alone but enjoyed and continue to enjoy certain popular support. It seems urgent for the international judicial community to work out a theoretical framework to prevent the parties in post-Soviet secession conflicts from proceeding from the ineffective first (federalization) and second (land-for-peace) to the third, fourth, and fifth recipes pretending to cut the Gordian knot. An indispensable element of this judicial rethinking seems to revisit late-Soviet history and return to Jackson’s proposal to relativize the present, excessively universalized sovereign-state system and to restore an empirical approach to statehood.