Do ethnic and non ethnic wars have the same causes; summary

[EvanKaufman]

This summary is based on an article by Nicholas Sambanis. Sambanis, N. 2001. “Do Ethnic and Nonethnic Civil Wars Have the Same Causes?” Journal of
Conflict Resolution 45(4):259-82.
I have tried to upload the article with this summary but it says that the pdf is not allowed. Hopefully we can get it uploaded soon or you can find the article on the internet.

Sambanis is interested in civil wars and the role ethnicity plays in the start and continuation of such conflicts. He notes that in the past, much research has been done on civil wars, but it has been done in an aggregate way, grouping all civil wars together. According to Sambanis, it may be more appropriate to categorize civil wars into ethnic and non ethnic wars, thereby drawing different distinctions that can help policy makers predict, prevent, or terminate violent civil wars.

In this study, Sambanis attempts to integrate IR theories, specifically neorealism and neoliberalism, into his large N quantitative study of civil war. He intends to find whether each of these theories for international affairs would relate to phenomena at the civil war level.

He defines ethnicity in his study according to Horowitz’s (1985, 17-18) study which explains that “ethnic groups are defined by ascriptive differences, whether the indicum is color, appearance, language, religion, some other indicator of common origin, or some combination therof…” This is a very inclusive concept of identity which facilitates comparison.

Five factors are used to code civil wars which include: (1) there were more than 1,000 war-related deaths during the entire war and in at least a single year of the war, (2) the war challenged the sovereignty of an internationally recognized state, (3) the war occurred within the territory of that state, (4) the state was one of the principal combatants, and (5) the rebels were able to mount an organized military opposition to the state.

In his literature review, he cites some interesting economic theories of civil war by Collier and Hoeffler (2000) and Fearon and Laitin (2000). These theories view war as the outcome of an expected utility calculation: potential rebels evaluate their expected gains from war, given their grievances, and compare these expected gains with the expected losses, which include the opportunity costs of forgoing productive economic activity. Whether rebellions will be observed is determined by the financial viability of a rebel organization. War will produce private and public gains and losses that are unevenly distributed among the parties, and private gains explain why war may be rational for some groups when it is collectively irrational (Sambanis 2001).

There are also transaction and coordination costs to rebellion. Collier and Hoeffler use the degree of ethnolinguistic fractionalization (ELF) as a proxy for determining this factor. They argue that ethnic kin groups facilitate within-group coordination and hinder cross-group coordination. The hypothesis is that coordination for rebellions is easier at low levels of ELF and becomes harder as ELF increases. Essentially, they argue that the more diverse a society, the less likely it is to experience a civil war because of the difficulty of cross group coordination(Sambanis 2001)

In addition, Fearon and Laitin argue that insurgencies are determined by mainly economic and not political factors. War will occur when its economic opportunity costs are low and finds that a lack of democracy and ethnic fragmentation are non-significant correlates of war.

In light of this past research, Sambanis notes that one assumption made by this research is that there are no differences between civil wars. He intends to test the differences between war types and question the applicability of the economic theories of civil war to the subset of ethnic civil wars.

His six hypothesis’ for this study include:
H1: The probability of ethnic war should be an increasing function of the degree of ethnic heterogeneity
H2: Greater levels of economic development should reduce the risk of ethnic civil war by raising the economic opportunity costs of violence
H3: Lack of democracy should significantly and substantially raise the probability of ethnic civil war because it threatens the core of ethnic identity.
H4: The greater the level of democracy in neighboring countries, the lower the probability of ethnic civil war.
H5: The presence of war in neighboring country should increase the risk of more war in the neighborhood.
H6: The cold war should be negatively correlated with the onset of ethnic civil war.

Sambanis counters the economic theories of civil war by noting that diversity of a nation would have no affect on the ethnic rebellion of one group. Ideology based revolutions that are prone to collective action problems are not relevant to war mounted by a single group. In addition, in ethnic war, people are less concerned about economic opportunity and more concerned with preserved ethnic identity through political and civil rights. He sides with neoliberalism in that democratic institutions offer solutions to help defuse ethnic conflict and prevent its escalation. The core of ethnic conflicts may include preserving regional autonomy, linguistic education rights, ethnic representation, or religious freedom. The survival of ethnic identity is said to have utility in itself and can explain why members of an ethnic group would offer free labor to the rebellion as economic opportunity costs are outweighed by the higher expected costs of suppression of ethnic identity. Political variables should be more robust determinants than economic variables.

Hegre et al. 1999 found that countries in the middle of the autocracy-democracy spectrum are most at risk of civil war because they are neither autocratic enough to suppress a rebellion nor democratic enough to suppress political grievance.
Sambanis also looks at the factor of being in a bad neighborhood. A bad neighborhood is a region that may be more prone to war and have low levels of democratic institutions. Whereas in a good neighborhood, neighboring countries may help in mediation of conflict and try to stop their ethnic group from getting involved in violent conflict.

Sambanis describes his proxies and statistics for testing said hypothesis which can be found in his article.

FINDINGS:

Ethnic heterogeneity is among the most robustly significant variables and is positively correlated with the onset of ethnic war. As a country becomes more heterogeneous, the probability of occurrence of an ethnic war increases.
Another major finding is the strength of the neighborhood effects. Countries that have land borders with countries at war are significantly more likely to experience an ethnic war of their own.
He also suggests that economic variables may be more important determinants of non-ethnic war onset than ethnic war onset.
Identity conflicts are largely contributable to political factors and are unlikely to occur in politically free (i.e. democratic) societies.
Whereas economic theories posit that ethnic heterogeneity may reduce the risk of war or is insignificant, Sambanis finds the exact opposite to be true for ethnic wars.

When looking at different theories, he shows that some theories are better predictors of ethnic civil war then others. “…it appears that modernization theory, with its emphasis on economic variables, is less applicable than theories of political development or neoliberalism, whose emphasis on the role of political institutions seems highly relevant as an explanation of the onset of ethnic war. Neorealism so far seems not as relevant as some scholars have thought because we found no evidence of a security dilemma, although more precise tests are needed before we reach a final conclusion on this matter (Sambanis 2001; p.280).”
Furthermore, Sambanis finds that it is important to look at ethnic civil wars differently from other types. Ethnic civil wars may be caused and defused by different factors compared to other types of rebellions. The survival of an identity may be more important in terms of political, regional, and religious freedoms than just purely economic opportunity and development factors.

QUESTIONS TO PONDER:
Is poverty and underdevelopment or political repression more ethnically concentrated?

How important is it to study more distinct characteristics of civil war rather than as an aggregate?

Do you think the Israeli-Arab conflict could be considered an ethnic war and if so, does it seem that political and democratic factors outweigh economic ones?
Does the preservation of identity seem more important than economic opportunity?

How much importance should be placed on identity and ethnicity when trying to resolve conflicts? What are some of the problems that you can foresee that would stem from this?

What specific factors of the Israeli-Arab or other conflicts would uphold Sambanis’s theory vs. economic theory or vice versa?

What are specific measures that states can take in order to prevent ethnic conflicts from arising?

[DanielShvartsman]

Hi Evan and everybody,

I'm going to respond to the question about studying civil wars (or any social phenomenons) distinctly or through the aggregate method. Our most recent (10th May) Civil War lecture had a guest lecturer talk about his research on distinct civil wars in El Salvador and Peru. We also talked about the different benefits of case studies and larger studies. The discussion is one we have had in several other classes, and some of them are obvious: case studies provide more details, a clearer understanding of the cases in question, and a more obvious connection between causes and effects, but they are harder to generalize with, while larger studies give signs of common, frequent relationships that are generalizable. One thing I liked about the 10th May discussion was the way the teacher pointed out that the large studies can uncover relationships between two factors (such as ethnicity and conflict, for example), but case studies give a clearer view on how those relationships work.

I studied Economics and History as an undergrad, so I have an appreciation for the way numbers can reveal relationships and suggest more scientifically solid theories, but I feel it is necessary to ground those theories in real examples and understanding, which studying history provides. This blog, to me, seems to be an example of using two different cases of conflict, reconciliation, and (hopefully) peace building. By sharing insights on Rwanda's issues and Israel's issues, we see things we can learn from one another and use, as well as differences that inform one another's conflicts and the broader problems of conflict and so on.

So as an answer to the question of aggregate versus distinct, I think we definitely need to study both and go back and forth from one to another. Large, aggregate studies can give broader frameworks, and then more focused studies can explain those frameworks and fill in the details of those frameworks (or change those frameworks).

We should have the discussion on Evan's other questions, but follow up questions for me would be: What have you learned from studying other cases that applies to the problems in your country? Do the specific differences in each country make it difficult or impossible to compare between different conflicts?

Have a nice weekend everyone,
Daniel