Kaufman, C.D. (1998) ‘When all else fails: ethnic population transfers and partitions in the twentieth century’

Kaufman, C.D. (1998) ‘When all else fails: ethnic population transfers and partitions in the twentieth century’, International Security 23 (2), 120-156.

Introduction

1. Separating warrying populations is best solution to many of the worst ethnic conflicts.
2. Key questions: do partitions and population transfers reduce or increase loss of life?
3. Move away from integrated multi-ethnic societies in post-conflict situations, as this may promote escalation of violence: separation of warrying groups may dampen conflict.
4. This remains controversial because of high human costs – only justified if it saves lives of people who would otherwise be killed in ethnic violence.
5. Key case-studies: Ireland, India, Palestine, Cyprus.
6. In all four cases, separation of warrying groups reduced subsequent violence, which resulted not from partition / separation but from incompleteness thereof.

State of Debate

1. Case for Separation:
   1. Security dilemma: no group can provide for own security without depressing security of others
   2. Power sharing techniques cannot work because they do not resolve such security dilemmas created by mixed demography.
   3. International community should endorse separation for some communal conflicts to avoid much higher human costs.
2. Case against Separation: Partitions and population transfers have 3 main flaws:
   1. They cause violence;
   2. They generate new conflicts;
   3. They create undemocratic rump states that perpetuate communal hatred.
   4. Therefore, reintegrating ethnic groups in conflict is both more moral and more practical than partition.
3. This conclusion is wrong because the security dilemma generated by intermixed populations “cannot be stopped except by permanent military occupation or genocide, or by not having the war in the first place” (10).

Solving the Debate

1. High-violence partition case studies are best to evaluate effects of international intervention
2. Four cases: Ireland, India, Palestine, Cyprus.
3. Ireland:
   1. Violence caused not by partition, but because partition did not fully separate antagonistic communities, especially in the North: fairly intense security dilemmas, in both North and South.
   2. Solution was a better partition line separating the two groups “as fully as possible”, resulting in “a smaller but safer Northern Ireland”.
4. India and Pakistan:
   1. Security dilemmas created by withdrawal of British imperial power: Muslim / Hindu SD nationally (incl. Kashmir), Muslim / Sikh in Punjab (more severe);
   2. Muslims remaining in India were two few and too dispersed to resist the Indian government (!);
   3. “The problem with Indian independence was not partition, but that partition did not go far enough” (76).
5. Palestine and Israel
   1. Security dilemmas generated by Israeli independence, not partition.
   2. In 1948 Israel, pattern of ethnic cleansing followed security dilemma logic and varied based on strategic needs of each place and time.
   3. Violence after 1949 was not caused by partition, but by existence of Jewish state.
   4. Settlements in West Bank generated new security dilemma; must be removed.
6. Cyprus
   1. Stable situation since 1974.
   2. Result of an intense security dilemma.
   3. Turkish invasion “did save thousands who would have been murdered” (97).
7. Politics of Successor States
   1. Critics of partitions and populations transfers overestimated the risks these “remedies” pose to political development
   2. Democratization
   3. Treatment of ethnic minorities
8. Challenges for Separation and Partition – Three lessons:
   1. Identify threshold of intergroup violence beyond which we must resort to separation and partition;
   2. Partition should only be undertaken if national communities are or will be separate at the same time: defensible borders are essential.
   3. Refugees from ethnic conflicts should be moved away from threats of massacres and resettled permanently – no return, because this would re-create the security dilemma that triggered the conflict in the first place.