ethnic enclaves are bad. if you want to keep people alive and prevent ethnic conflict, that is.
a couple of real world examples…
first, guatemala. in “Spatial and Temporal Patterns in Civil Violence: Guatemala 1977-1986,” timothy gulden found, amongst other things, that [pg. 29]:
“The amount of killing in a municipality was correlated with its ethnic mix….
“The distribution of ethnic population among municipalities was highly polarized and the distribution U-shaped, as suggested by the gray bars in the histogram presented as Figure 3.
“Thus, at both ends of this histogram, where one group or the other [ladinos vs. mayans] represented more than 90 percent of a local population, and also in the middle, where neither group was more than 75-percent dominant, killings of Mayans, the principal victims, were fewer.”
here’s the histogram:
more from gulden [pg. 34]:
“The ethnic mix histogram showed the greatest number of killings in municipalities with predominantly Mayan populations, but fewer killings in the most heavily Mayan areas. This pattern might have been produced by an attempt to clear Mayans from mixed Mayan-Ladino areas but a disinclination to do the same, to attempt ‘ethnic cleansing,’ in the Mayan heartland. The ethnic mix histogram also showed a small rise in killings where Mayans were between 10 and 25 percent of the population. Minority populations of such proportions might somehow have threatened or annoyed a local Ladino majority without effectively restraining it. Members of the smallest Mayan populations were safer; they might typically have been individual servants or laborers, with or without families, well known to Ladino employers.”
re. switzerland and the former yugoslavia — from an article i posted about the other day, “Good Fences: The Importance of Setting Boundaries for Peaceful Coexistence”, the authors found that, in both switzerland and the former yugoslavia, there was the least amount of ethnic conflict in those areas where topographical and/or political boundaries clearly separated different linguistic, religious and/or ethnic groups, but where there were porous borders, there was more conflict. the former situation is more characteristic of switzerland on the whole, while the latter characterizes the former yugolavia, ergo more ethnic conflict in the former yugoslavia [pgs. 10-12]:
“Our work clarifies the ambiguities of mixed language and religion Swiss cantons by showing that in most cases the natural geography of the populations conspires to lead to a low level of violence, so that additional boundaries were not necessary; where they were needed, as in Graubunden, they were established. The highest calculated propensity to violence is between linguistic groups in the northern part of the canton of Bern, where historically unresolved real world tensions actually exist. Our analysis indicates that both administrative and natural barriers can play a significant role in mitigating conflict between religious and linguistic groups. Historical evidence suggests that for religious groups the boundaries in Switzerland were created to provide autonomy to a group with a shared identity and avoid conflict among multiple groups. Ongoing efforts to reduce tensions in Bern include introducing new political boundaries. The many political, social and economic factors that play roles in reducing violence build on a strong foundation of geographical borders.“
diversity is strength … in killing games. =/
(note: comments do not require an email. swiss diversity!)