Sambanis, N. (2000) ‘Partition as a Solution to Ethnic War: An Empirical Critique of the Theoretical Literature’, World Politics 52 (4), 437-483

1. Partition theorists have not produced operational criteria for applying their theories consistently across cases.
2. Empirical testing could help elucidate these theoretical debates.
3. Focus on claims that once ethnic violence starts, civil politics can only be restored if ethnic groups are demographically separated into ‘defensible enclaves’ (438) and solutions aiming to restore multi-ethnic civil politics will not work because they do not resolve existing security dilemmas (‘SD’).
4. SDs are at heart of partition theory: when communities distrust each other, one community’s actions to increase its own security is seen as threatening to security of others.
5. Therefore, the claim is that partition in homogenous separate regions becomes inevitable to end conflict.
6. No proof for this has been provided.
7. Will test three core hypotheses:
   1. Partitions facilitate post-war democratization;
   2. Partitions prevent war recurrence;
   3. Partitions significantly reduce low-level ethnic violence.
8. Result: Partitions do not prevent recurrence of ethnic war and may not even be necessary to stop low-level ethnic violence.
9. Criticism of partition:
   1. too limiting a solution: ethnic cooperation may be possible even after a civil war.
   2. too severe a solution: forced population movements cause tremendous human suffering.
   3. Will create undemocratic successor states: repression of residual minorities.
   4. Endorsing some partition will encourage others, leading to wars: not true; this is rare.
10. SD ignores the fact that conflict is often due not to ethnic groups’ security needs, but to ‘predatory’ goals of their leaders; therefore, partition will not solve the SD of partitioned ethnic groups if it exacerbates ‘predatory’ incentives of predecessor states.
11. Civil wars tend not to end in negotiated settlements, unless supported by external security guarantees that prevent predatory predecessor states from restarting wars against successor states (442).
12. Alternative to both Partition (‘P’) and creating a new balance of power is to create a regional hegemon responsible for regional peace.
13. Ethnic diffusion might mitigate the SD (Byman) because it reduces the possibility that a single ethnic group might become dominant (443): ‘ethnic balancing’ against threatening groups is both possible and stabilizing.
14. “The probability of civil war drops significantly at very high levels of ethnic diversity and it is greatest in ethnically polarised societies” (443).
15. Four key questions related to partition must be investigated:
    1. What are the main determinants of P?
    2. Does P create democratic or undemocratic states?
    3. Does P prevent war recurrence?
    4. Does P end low-level ethnic violence?
16. Complied data set with 125 civil wars and 21 partitions.
17. Testable hypotheses.
18. Findings:
    1. Type of war is a significant determinant of partition: ethnicity matters for the onset of partition.
    2. As ethnic heterogeneity increases, probability of partition decreases significantly; as size of ethnic groups increase, so does likelihood of partition.
    3. Partitions are positively and significantly correlated with levels of violence; but violence may well be caused by partition itself.
    4. “Partition is more likely after identity than ideological war, after truce or rebel victory following war, in a country with large ethnic groups and little heterogeneity, and higher levels of economic development.
19. Three critical hypotheses of P theory:
    1. Ps create successor states that are at least as democratic as predecessors: evidence unclear; more research needed.
    2. Ps reduce risk of war recurrence: evidence does not support this. Therefore, separating ethnic groups does not resolve the problem of violent ethnic antagonism.
    3. Ps reduce low-level ethnic violence after war ends: very weak evidence in favor.
20. Ps are coerced, painful, costly, may saw seeds for future conflicts. International policy towards P must rely on rigorous, empirical testing and arguments.
21. New Hypothesis: “The strategy of supporting ethnic diffusion by combining rather than partitioning large ethnic groups may be worth pursuing” after a civil war: “If borders can be credibly and securely redrawn, then combining several large ethnic groups in a larger multiethnic state may reduce the probability of new wars” (479).
22. “Partition, as we have seen, does not help reduce the risk of war recurrence. Partitions are in fact positively (though not significantly) associated with recurrence of ethnic war. The probability of a new war rises in tandem with the human toll of the previous war and with non-decisive outcomes to the war” (480).
23. “Negotiated settlements, a strong government army, and a lengthy previous war all reduce the probability of war recurrence” (481).
24. “To reduce residual violence, it is important to prevent war recurrence, as patterns of large-scale violence over time seem to encourage lower-level violence” (481).
25. “Strategies to support the government’s prewar institutions and its military may also achieve peace, but they may do so at the expense of justice” (481).
26. Empirically-derived strategy for resolving ethnic war: “This strategy demands action by the international community, which must promote democracy as its number one conflict-prevention strategy. If violence does erupt, its priority should be to facilitate a negotiated settlement, as well as to integrate and downsize the government’s military… If border redefinition is a viable option – and it should be an option only if it does not assist one party at the expense of another – then ethnic integration rather than ethnic partition may be a winning strategy. In addition to having the potential for greater success than partition, this strategy is also not loaded with subjective and arbitrary assumptions about the necessity for ethnically pure states and about the futility of interethnic cooperation” (481).
27. “On average, partition may be an impossible solution to ethnic civil war” (482).

Kumar, R. (1997) “The Troubled History of Partition’, Foreign Affairs 76 (1) (Jan./Feb. 1997), 22-34.

1. 1995 Dayton Peace Accord: a partition agreement with an exit clause for outside powers.
2. Claim that partitions as solutions to ethnic conflicts save lives, safeguard rights of contending ethnic groups through intervention of impartial outside power, and create homogenous territories through population transfers.
3. Instead of doing so, partitions “fomented further violence and forced mass migration” (24).
4. Partition assumes irreconcilable ethnic identities and capacity to separate ethnic groups.
5. Usually, partitions are “driven by considerations extraneous to the needs and desires of the people displaced” and “end up stimulating further and even new conflict” (26).
6. Case studies of Cyprus, Palestine, Northern Ireland, India, Bosnia show that partitions are seen as temporary solutions to crises; however, once implemented “ethnic partitions have never been reversed; their implementation has inexorably driven communities further apart”. (33).
7. Ethnic partition can hamper development of postwar economies.
8. “Divide and Quit” approach rarely works: it turns into “Divide and Be forced to stay” (34).
9. “Investment in reintegration may be discovered as the easier route to withdrawal.” (34)

Horowitz, D.L. (2003) ‘The Cracked Foundations of the Right to Secede’, Journal of Democracy 14 (2), 5-17.

1. Two international legal developments in regulation of conflict and warfare between ethnic groups:
   1. Tribunals to punish genocide and crimes against humanity;
   2. Elaboration of various doctrines of human rights, including a possible right of ethnic groups to secede.
2. Such a right is ill-considered and dangerous.
3. Certain theorists see secession as answer to problems of ethnic conflict and violence; that is wrong: it is likely to make such problems worse.
4. Secession does not create homogenous successor states or reduce conflict, violence, or minority oppression in successor states (5-6).
5. Minorities’ condition can best be improved by devising institutions to increase their satisfaction within existing states.
6. “Partition can be accomplished reluctantly, as a matter of prudence, without recognizing a right to secede” 96).
7. Atlantic Charter, 1941: self-determination limited to peoples living under foreign domination, resulting in decolonization.
8. From the end of WWII to end of Cold War territorial boundaries were remarkably stable.
9. After the end of the Cold War boundaries became less stable. Badinter Commission on Yugoslavia legitimised breakup of this federation.
10. Various theories of self-determination leading to secession:
    1. An integral right of peoples to be free of authoritarian oppression: right to live under a democratic regime;
    2. Right of people in general of with common group characteristics to choose with whom they wish to associate politically – although collective identity fluctuates;
    3. Remedial right: last-ditch response to discrimination or oppression by central government;
    4. All assume secession can result in homogenous successor states or at least ones that will guarantee minority rights.
11. Treatment of minorities in new states will not improve if minorities were not respected in undivided state.
12. “Secession merely proliferates the arenas in which the problem of intergroup political accommodation must be faced” (9).
13. Secession encourages the former minority, now a majority, to cleanse the secessionist state of its own minorities and induces the rump state to do the same.
14. There are no truly natural boundaries that secession can institutionalise.
15. Secessions and partitions convert domestic ethnic disputes into more dangerous international ones and trigger irredentist claims that often are followed by ethnic cleansing.
16. A right to secede will undermine efforts to achieve interethnic accommodation within states (10).
17. Devolution efforts are most effectively undermined by a right, recognised under international law, to secession.
18. “A right to secession effectively advantages militant members of ethnic groups at the expense of conciliators.” (11)
19. A right to secession grounded in extreme oppression of minorities derives from an alleged commitment of international law to democracy. Such a commitment is tenuous at best.
20. Self-determination for peoples or groups within a state is to be achieved by participation in its constitutional system, on the basis of respect for its territorial integrity.
21. Secession is an anti-state movement and undermines the very foundation of current international legal order – state sovereignty.
22. Solution is to foster interethnic accommodation within states, through institutions that can mitigate conflict:
    1. Consociational democracy: neglect of democratic opposition and propensity for excessively limited government and immobilism; it is attractive to minorities, not majorities.
    2. Use of political incentives to encourage ethnic moderation: electoral systems capable of inducing moderate behaviour by politicians.
23. Political engineering can work in specific circumstances, but is no panacea.
24. “Efforts at conciliation will not be helped by providing either a liberal or constrained right to secede” (15).

O’Leary, B. (2016) ‘Debating Partition’, K. Cordell and S. Wolff, eds., The Routledge Handbook of Ethnic Conflict, 2.ed., London, Routledge, 138-154.

1. Partition: “a fresh political border cut through at least one community’s national homeland with the goal of resolving conflict” (138).
2. Key political and moral arguments to resolve antagonisms through partition:
   1. Historicist: partition is inevitable once ethnic conflicts pass a certain threshold. It is seen as both informed and realistic, but record shows that there can be peace without separation.
   2. Last resort: if alternative strategies fail, partition should be chosen to avoid genocide or large-scale ethnic expulsions.
   3. Net benefit: “partition should be chosen when, on balance, it offers a better prospect of conflict reduction than maintaining the existing borders”. It is desirable in its own right, not just as last resort. (139).
   4. Better tomorrow: without partition there will be more conflict and conflict recurrence; therefore, after partition parties will conduct themselves better.
   5. Realist rigour: partition must lead to radical demographic restructuring, to reduce military and political significance of new minorities.
3. They all suggest that “it is foolish to insist on maintaining unviable multinational polities” (141).
4. Modalities:
   1. Proceduralists: consultations with affected parties to achieve reciprocal consent on new border: deploy fairness and feasibility requirements. A. Lijphart set requirements for a fair partition: negotiated not imposed; fair division of land and resources; results in substantially less plural independent states.
   2. Paternalists: local parties cannot reasonably agree, therefore a sufficiently powerful outsider must determine partition that will be durable and reduce conflict fast.
5. Anti-partitionist arguments share views of partitions as perverse, of jeopardising existing relationships, and of impossibility of achieving fair partitions:
   1. Rejection of rupturing of national unity: majority of original state opposes secession as violation of its right to self-determination, seen as only in interest of privileged elites.
   2. Constructive possibility of bi- and multi-nationalism: pluri-national arrangements must be properly exhausted before partition is considered genuinely as a last resort. Often, minority leaders refuse or block all other options.
   3. Practical impossibility of just partitions:
      1. a number of key difficulties that a boundary commission would need to decide:
         • Which should be the units around which new boundaries should be drawn?
         • Should there be subunit optouts?
         • How should units’ preferences be determined?
         • Should local popular preferences be considered just one criterion to be balanced among others?
         • Should non-preferential factors be considered in designing new borders, and should local popular preferences be subordinated thereto – and who should decide this?
         • Should there be constitutional amendments to ratify proposals and referendums, and should there be provisions for their subsequent revision?
      2. “partitions are perverse: they achieve exactly the opposite of what they nominally intend” (147).
      3. Kaufmann (1998) is wrong when he argues that partition reduced violence in his 4 case-studies. He shows “it is easy to slip from a defense of partition as a last resort to tacit support for ethnic expulsions” (148).
      4. Partitions often lead to post-partition wars., creating inter-state ‘security dilemmas’.
   4. Elusive mirage of homogenization without expulsion:
      1. Partition alone is unlikely to create desired levels of homogenization.
      2. Assimilation, expulsion, even genocide will follow.
      3. “Partitions are never enough for rigorous homogenizers.” (149)
      4. “Partitions without comprehensive expulsions generate two kinds of orphaned minorities: former prospective majorities, and formerly dominant minorities” (149). They both may become part of irredentist movements or campaign for further partitions.
   5. Damage to successor states:
      1. Partitions generate new inter-state security crises and cause significant economic disruption by disturbing established monetary and exchange networks, increasing transaction costs, protectionism and border-related criminal activity.
      2. Post-partition states have functional and infrastructural interests that leads them to consider cross-border cooperation or confederal arrangements that put into question the need for partition itself.
      3. Usually, one of the two post-partition states is significantly disadvantaged and significantly underperforms.
   6. Failure to make a clean or elegant cut:
      1. Post-partitionists’ maps bleed and “do not look good” (149).
      2. New borders are usually less compact and create adverse security and transport connections.
6. Anti-partitionists’ arguments are more compelling judged by realistic, political and moral criteria and are endorsed by international law.
7. “Give power-sharing a chance” (151): complex power-sharing settlements are possible even after protracted ethno-national wars.
8. Partitions deserve their poor press, as they do not generate better security environments and are biased towards privileged or dominant minorities.
9. Post-partition arrangements are worse than predicted for at least one successor state.
10. “Prudence therefore mandates opposing partition as a tool of international public policy-making, and placing the burden of proof on its advocates” 152.
11. Hard to find a good 20^th-century partition.
12. Implementing a new border destabilizes inter-group relations in ways that may take generations to repair. Secessions harden existing administrative borders and may be easier to accomplish.

Horowitz, D.L. (2015) ‘Irredentas and secessions: Adjacent phenomena, neglected connections’, in K. Cordell and S. Wolff, eds. The Routledge Handbook of Ethnic Conflict, 2.ed., London, Routledge, 155-164.

1. Secessions and irredentas have traditionally not been treated together, but they are deeply interconnected: the strength of one movement is related to the fact that the other may well rise – they are plausible alternatives to each other.
2. Secession: attempt by an ethnic group claiming a homeland to withdraw with its territory from the authority of a larger state of which it is part (155).
3. Irredentism: a movement by members of an ethnic group in one state to retrieve ethnically kindred people and their territory across borders.
4. Both secession and irredentism contain various levels of intensity and various strategies (irredentism: incorporation into another state; creation of a new state from various irredentist groups).
5. Secession: subtracting from an existing state; Irredentism: subtracting from one state and adding to another.
6. There have been very few irredentas in postcolonial states, but many secessionist movements.
7. Some international border disputes have no ethnic component; others contain compact ethnic groups that nevertheless do not dominate their region.
8. When faced with a choice between the two, groups find secession the more satisfying choice. “Indeed, the potential for irredentism may increase the frequency and strength of secession, but not vice-versa” (157).
9. Three issues that connect S and I:
   1. Convertibility of S and I:
      1. violence is convertible from one to the other;
      2. so is mutability of ethnic group claims and of trans-border affinities.
      3. State policies towards both also change over time: they tend to be inconstant towards I, which drives groups towards S.
      4. Ethnic affinity in one state may not extend to the irredentist group, also favoring S.
   2. Relative frequency of S and I:
      1. Irredentist action by the potential retrieving state is uncommon.
      2. Aid to S movements can be strategically terminated; aid to I movements cannot because they are underpinned by “an ideology of common fate” that does not lend itself to abrupt termination (160).
      3. In turn, irredentist groups find retrieval by another state undesirable because of their own personal interests as political leaders. S given them a better access to power, whilst I allows them to be challenged by leaders of retrieving state.
      4. Retrieving state itself may be heterogenous.
      5. Multiple secessions to build a new state (eg. Kurdistan) is almost impossible.
      6. In practice, even groups that have theoretically an I option do not really have it in practice and end up opting for S.
      7. “In short, all else being equal, the fewer the irredentas, the larger the number of secessionist movements” (161).
   3. Relative strength of S and I:
      1. Relative strength of a movement I affected by whether they chose S or I and whether the other is also available.
      2. States that are reluctant to engage in I claims may assist groups to achieve S (India / Bangladesh).
10. When will an ethnoterritorial separatism movement take S or I courses? It’s a strategic choice based on calculations of rational interest.
11. However, emotional factors cannot be discounted (eg. Ethnic affinity): secession movements continue to arise although secession almost always fails.

McGarry, J. (1998) ‘Orphans of Secession: National Pluralism in Secessionist Regions and Post-Secessionist States’, in M. Moore, ed., National Self-Determination and Secession, Oxford, OUP, 216-28.

1. National pluralism problem: many states, even while respecting individual rights, do not treat all national groups with equal respect.
2. ‘New liberalism’ school attempts to remedy this, but do not extend their analysis to debate on self-determination and secession: still offer a restrictive view of right to secession (eg. A. Buchanan: only if group is victim of injustice). They do not consider the nationalist basis of most secession movements (216)
3. Neither do they think adequately about national pluralism in post-secession states, even when they recognize a broader right to self-determination (eg. D. Philpott: right to self-determination, including secession, grounded in expression of autonomy).
4. Not helpful to elaborate theoretical principles that make abstraction from national diversity, diversity of seceding area, majority – minority relations, type of existent group identities, which are all critical to each specific secession case.
5. Problems involved in governing in post-secession states rooted in fact that these regions were seriously divided on the secession project. The new states are probably just as heterogenous as predecessors and just as likely to abuse their own minorities and engage in conflict.
6. Solution: move away from nation-state model and towards that of a ‘multinational state’ in which all groups are treated equally. “[T]he principle of equal treatment involves moving away from idea of independent states and embracing transborder or supra-state political institutions” (217).
7. Secessionism is usually highly contested not just by remainder state but also within secessionist region, by local ethnic minorities, against the largest regional ethnic group that achieved secession.
8. Secessions usually occur along administrative rather than national lines; therefore, national heterogeneity continues. Administrative boundaries did not create homogenous subunits on purpose, to better “control minority passions” (219).
9. Sometimes ethnic minorities do support secession (eg. Baltic Countries and Ukraine in former USSR). In other cases opposition can come even from large sections of local ethnic majority against the elite driving the project.
10. Conclusion: “demands of these groups for autonomy can be satisfied short of secession, and that if secession is to occur, it is unlikely that there will be a consensus behind it”. This is because “[o]utside of polarized conflict zones, individuals frequently have nestled identities, and feel part of several communities simultaneously” (220).
11. In many cases, “secession does not solve the problem of national diversity: it merely places it in a different state context.”
12. New states are often ‘nationalizing states’ seeking to promote interests of national majorities (language, culture, symbols) at the expense of their own minorities (Brubaker).
13. Most new minorities do not revolt; many migrate back to the rump state where they are in the majority. Those who remain mobilize to secure their individual and group rights.
14. Members with dual identities accept the new state but “regret the passing of the old” (222).
15. Two key issues arise out of attempt to reconcile nationally-defined self-determination with liberal values of equal respect and individual autonomy:
    1. Do these dynamics, present especially in Eastern Europe, apply to Western Europe and North America as well? There is no clear East-West dichotomy: cases are rather on a continuum, depending on their specific circumstances (eg. Quebec has its own minorities opposed to secession that are highly mobilized).
    2. Is the focus on the pre-and post-secession periods an unfair context for theorizing, rather than taking a longer-term approach to allow identities to adjust to the new state? Conditions which produce conflict during state formation may be durable (eg. Northern Ireland, Israel, Romania). Triadic Nexus will continue to evolve in a vicious circle of conflict (new states worry about minority revols and frontiers security; minorities continue to withhold their loyalty; neighboring states continue irredentist claims).
16. Similar problems may continue in rump states as well, where remaining minorities may be subjected to ‘nationalizing projects’ of remaining majority (eg. Catholics in Northern Ireland).
17. Need to address causes of minority discontent before support for secession reaches a tipping point because the original state continues behaving like a nation-state with only one nation. Timely and genuine decentralization often effectively addresses the causes of such grievances. Extensive decentralization is consistent with state unity.
18. Arrangements should be also made for minorities to be represented in central governments, through partnership strategies resulting in power-sharing regimes.
19. Nationalizing states and even liberal nation-states are not suitable frameworks for this: need to develop institutions based on accommodating all national groups.
20. Breakaway states should seem to win support from their own minorities by adopting similar decentralist and consociationalist strategies.
21. “The appropriate way to address these problems of parallel and overlapping national loyalties is to move beyond the notion of traditional ‘Westphalian-style’ independent states and construct supra-state partnerships and institutions” (227).
22. “Such accommodation of minority nations is, however, very much the ideal, and rarely the practice” (228).

Coggins, B.R. (2011) ‘The History of Secession: An Overview’, in A. Pavkovic and P. Radan, The Ashgate Research Companion to Secession, London, Routledge, 24-43.

1. Macro-historical approach: secessionism across time.
2. Rise of nation-state, self-determination, nationalism in 16^th and 17^th centuries required rulers to develop a compelling national myth to legitimise their rule in order to minimise both internal and external challenges.
3. By middle of 19^th century, nationalism took two main forms: civic and ethnic.
4. “Ethnic nationalism fractured multinational empires and the tension between national self-determination and state integrity grew steadily throughout the 20^th century” (25).
5. After WWII, “nationalists demanded that national identity be the presumptive basis for self-determination, sovereignty and membership in the international society of states” (26)
6. Jennings: “let the people decide” seems reasonable but is in fact ridiculous because “the people cannot decide until someone decides ‘who the people are’” (26).
7. The international system went from 25 states in 1816 to 194 in 2008.
8. Secessionism was encouraged by President W. Wilson’s 14 points after WWI setting out a programme of national self-determination requiring a state for each nation.
9. After WWII, anti-colonial secession exploded in the global South, legitimised and managed by the UN.
10. Secessions arising out of anti-colonial movements and emerging from disintegrating host states usually succeeded and were recognised; most others were not.
11. Most secessionist movements are predisposed to violence and have disastrous consequences for host states.
12. National communities asking for self-determination are branded as internal enemies by the host state and are at risk of violence because they begin from a position of weakness: balance of power between the two means that violence favors the host.
13. Leaders of secessionist movements stand to gain wealth, prestige, power and lifestyle upgrade if successful.
14. Few states have ever included legal provisions for secession in their constitutions, and only a few more explicitly outlaw secession: most leave it legally ambiguous. But many secessionist movements are non-violent and end up in independence.
15. Common tactic is formation of independentist parties advocating secession.
16. Enduring characteristics of secession:
    1. Statehood remains a valuable commodity;
    2. Third-party states and international institutions play an influential role in secessionist conflicts, with international recognition as a sovereign state and UN membership being the ultimate prize;
    3. UN maintains that right to self-determination does not imply right to secession in order to avoid violent conflicts and wars; therefore no legal developments regarding secession have taken place over past six decades. Conditions under which secession might be deemed acceptable or justified remain unclear and can only be deduced from precedents.
    4. “Without an international consensus regarding secessionist norms, most conflicts will drag on or reach stalemates only to reignite because foreign capitals will not unanimously ratify battlefield outcomes or compel negotiated compromises between parties” (40).