Sambanis, N. and Schulhofer-Wohl, J. (2009) ‘What’s in a Line? IS Partition a Solution to Civil War?’

Sambanis, N. and Schulhofer-Wohl, J. (2009) ‘What’s in a Line? IS Partition a Solution to Civil War?’, International Security 34 (2), 82-118.

1. Partition promises a clean and simple solution to war – but does it work?
2. Arguments for partition:
   1. Ethnic identities are hardened by war;
   2. This makes interethnic cooperation difficult;
   3. It increases risk individuals will be targeted for violence because of their ethnicity;
   4. Separating ethnic groups in conflict reduces risk of escalating violence.
3. Partitions has costs:
   1. Changes political boundaries;
   2. Forcibly relocates populations.
4. Empirical evidence in favor of partition is weak.
5. Chapman and Roeder (2007) (‘C&R2007’) reanalyses data generated by Sambanis (2000) to show partitions have strong pacifying effect after civil wars.
6. This article will demonstrate the fragility of pro-partition empirical results: on basis of available evidence, partition does not have the pacifying effect C&R2007 claim it does.
7.  Necessary to pay attention to data coding issues, historical and political contexts, rigorous theory building.
8. We “find that partition does not work in general and that the set of conditions under which it is likely to work is very limiting” (83).
9. Kaufmann claims partition is a good solution if groups can’t live together in ethnically hetereogeneous states because it resolves the ethnic security dilemma: it reduces threats each group poses to the other.
10. Partition: “a civil war outcome that results in territorial separation of a sovereign state” (84).
11. “Redrawing borders, with or without substantial physical separation of people, is often unsuccessful in reducing the risk of war recurrence” (85).
12. Three historical examples indicate that partition might work under certain conditions: Cyprus, Bangladesh, Croatia, Eritrea, Somaliland.
13. C&R2007 develop an “institutional bargaining” model arguing that de jure partitions resulting in creation of new sovereign states reduce likelihood of escalation in hostilities in short run.
14. Their results are due to methodological mistakes. Their claim that partitions should outperform all other solutions to civil wars over competing nation-state projects because “they simplify the nature of bargaining between elites of the secessionist region and elites of predecessor state, reducing opportunities for violence escalation” (87).
15. Ethnic Security Dilemma (‘ESD’) set out by Posen (1993):
    1. Without impartial state policing, ethnic groups become responsible for own security and risk escalation because of “tactical offensive advantage” (94). Therefore, each group attempts to ethnically cleanse its territory of potentially hostile ethnic groups, leading to rapid escalation of violence in a preemptive war.
    2. ESD puts political geography at the centre and claims that partition removes tactical offensive advantage if near-complete physical separation of antagonist ethnic groups is achieved.
    3. ESD asserts that ethnic power sharing is particularly unstable, that ethnic identity is easily identifiable, and that this makes targeting of individuals for violence after an ethnic war particularly easy.
16. Unstated key condition: presence of powerful coethnics in a neighbouring state is a key component of the escalation logic of the security dilemma; this implies against logic of partition “that neighboring states can both deter and catalize an escalation of violence regardless of demographics” (95).
17.  “Ethnic power sharing need not be inherently unstable; conflict escalation often results from external intervention and not from the country’s ethnic demography. Focus on ethnic demography assumes fixity of ethnic identities and ease of ethnic identification, which underpin position that partition is potentially useful only in ethnic wars (96).
18. “The ethnic security dilemma applies only under conditions of state weakness, and the argument boils down to a credible commitment plan” (96). Partition is only one of several ways through which the credible commitment problem might be addressed” (98).
19. Security dilemma does not apply only to ethnic wars, but also to those arising out of political beliefs and affiliations to any number of social groups (97). It ceases to apply to residual minorities and is exacerbated by certain demographic patterns, all of which must be taken into account when evaluating conditions under which partition might work.
20. Institutionalists argue that partition is supposed to have a pacifying effect and reducing hostility by simplifying postwar bargaining between elites of a secessionist state and elites in the predecessor state (98). They opine that partition eliminates conflict by:
    1. Strengthening the collective identity of the secessionist region’s inhabitants and reducing the claims of the rump state to its territory and inhabitants;
    2. Eliminating causers of conflict by minimizing joint decisions that need to be taken jointly by the central government of the rump state and the leaders of the secessionist region;
    3. Raising the costs of escalating conflict by transforming it into an international one;
    4. Giving both sides more visible and defensive military positions, thus achieving ‘a balance of capabilities’ (99).
21. These arguments rest on ad hoc assumptions that are not proven.
22. Even if partition solves a conflict by separating populations that do not trust each other to cooperate in a unitary post-war state, it can generate new incentives for new identity or distributional conflicts in bot rump and secessionst states (101):
    1. Within new state for government control;
    2. With new group trying to secede from new state and join rump state;
    3. Within rump state over government control;
    4. Within rump state for distributional resources after departure of a resource-rich territory;
    5. Between rump state government and other minorities as result of secondary consequences of partition.
23. Data shows that partitions do not have the anticipated positive and significant effect on post-war peace. (107).
24. Peace transitions are nonlinear processes: one step forward – two steps back.
25. “The best available evidence shows no significant association between partition and postwar stability, defined as a lower risk of a return to war” (116).
26. “Ethnic cleansing, once considered a humane way to manage conflict, has fallen out of favour. In many ways, partition just takes the problem and calls it a solution” (117).
27. The ‘institutional model’ offers weak foundations for arguments in favor of de jure partition as a solution to civil war (118).
28. “Institutional arguments assume that contentious identities will be quickly transformed by partition, whereas security dilemma arguments assume that contentious identities cannot be transformed. Yet neither argument has dealt with the institutional effects of partition on ethnic identity properly; both assume away the problem that new identities and distributional conflicts can be created by partition” (118).