Do Educated Ethnic Minorities Choose Nonviolent Resistance?

by Erica Chenoweth on August 4, 2011 · 2 comments

in Education,International Security,Protest

Renat Shaykhutdinov has an interesting piece in the July 2011 issue of the Journal of Peace Education. From the abstract:

ethnic groups that enjoy a higher educational status are less prone to using violent strategies choosing instead peaceful protest. I test this hypothesis using data on 238 ethnic groups in 106 states from 1945 to 2000. The results of the statistical analysis indicate that groups with higher levels of educational attainment are more likely to engage in non-violent protest. Conversely, groups that enjoy lower educational status in their respective societies tend to use violent tactics.

The basic idea here is that ethnic minority groups that have better educational access and privileges than the majority (or “core”) population are more likely to use nonviolent protest to make territorial or group demands. Ethnic minorities that have no significant advantages (or the same educational access and privileges as the majority population) will be a little less likely to use nonviolent resistance, and ethnic minorities with observable disadvantages relative to the majority population should be more likely to adopt violence. You can read the article to see the evidence he brings to bear on this question, his control variables, and the methods he uses.

 

This is a great, under-explored question with important ramifications for the policy and advocacy communities. In general, we should probably think more about how learning shapes world politics. Moreover, I like Shaykhutdinov’s argument, mostly because I can get behind its policy implications (who is pro-educational-inequality-across-ethnic-groups and would say so in public?). Nevertheless, the article brings to mind a couple of issues for me.

  • What is the causal mechanism here? Shaykhutdinov argues that educational attainment (what he codes as “educational advantage” vs. “no significant educational advantage” vs. “educational disadvantage”) should reduce the propensity to use violence because education instills norms, values, and skills. I’d call this the “violence is for dummies” argument. This argument has some appeal, as well as some empirical support when forecasting where nonviolent uprisings will occur. But we also know that a lot of the most dangerous terrorists or insurgents in the world have been educated elites–including many suicide terrorists. Those who use violence aren’t really dummies. Moreover, if education makes people less violent, then why do highly educated people in many societies commit the worst violence? Seemingly this argument would apply to government officials as well as to ethnic groups. But very highly educated people in the world have been some of the 20th Century’s greatest mass murderers (I’m thinking Pol Pot and the Khmer Rouge, of which many at the top were educated in France prior to returning to Cambodia and committing one of the world’s worst genocides). So why doesn’t educational attainment make violence go down in those cases too?
  • To me, one of the most important effects he may be picking up is the fact that ethnic groups that enjoy educational “advantages” may simply be “advantaged” in general. Privilege is privilege. I suspect that educational advantages would be highly correlated with business advantages, for example. The problem is that advantage–or status–may be explaining both the educational status of the groups and their adoption of different protest techniques. What I mean is that ethnic groups that enjoy privileges in society may not wish to threaten that privilege by appealing for more rights through violence. People with less privilege, on the other hand, are already starting from a lower point on the social totem pole. They likely already face considerable barriers to social, economic, and political satisfaction, and educational access is simply more of the same. It is precisely these conditions that may explain both their educational disadvantages and the grievances that they use to justify their violence. This is the classic endogeneity problem (and to be fair, there are no easy ways to overcome this problem statistically).
  • Does the type of education matter? Substance of education might be important. For instance, people who have spent their entire lives in parochial schools may have different feelings about nonviolent and violent resistance than people who have spent their entire lives in public schools. People who receive training in civil resistance methods during their education may be more likely to favor these methods over violence (and vice versa!). A potentially more precise (and theoretically defensible) type of education might be whether the ethnic group has had access to training from other civil resistance or civil society organizations on how to launch an effective nonviolent protest, versus contact with violent insurgents on how to train for a violent uprising.
  • Ironically, I think that oppressive regimes would much rather face a violent insurrection than a nonviolent one. Check out this creepy video released by the Iranian Interior Ministry to see what I mean:

<iframe width=”425″ height=”349″ src=”http://www.youtube.com/embed/lL9MaZQORfI” frameborder=”0″ allowfullscreen></iframe>

Civil resistance campaigns are scary for autocrats. They don’t know how to competently respond to them. Violent insurgencies, on the other hand, are relatively easy for them to dispose of, using a wide range of repressive tools that are readily available to them. If we took Shaykhutdinov’s conclusions to their logical policy implications, therefore, scholarly-inclined autocrats might use this research as a pretext to generate more educational inequality among their ethnic groups. That way, they could continue to suppress these minority groups socially, economically, and politically, while also denying them the fundamental skills and knowledge required to launch effective nonviolent challenges to the regime. Yikes.

But not so fast, autocrats. I think the empirical relationship between educational advantage may be overstated a bit in Shaykhutdinov’s piece. Take a look at the cross-tabulation below (from the article).

What this table tells me is that the preponderance of ethnic groups in the sample are either advantaged or equal to society as a whole. Few ethnic groups in the sample (only 12 out of 238) were really disadvantaged, and among those that were, only 1 adopted a purely violent strategy. Among the most educationally privileged groups, however, over 15% resorted to a purely violent strategy (the highest percentage of all three categories), whereas only 10% of the educationally-equal groups used a purely violent strategy. A roughly equal percentage of them (38-39%) used nonviolent resistance. As such, the “middle” category of a “mixed” nonviolent and violent strategy is doing the most work in the statistical analysis. But the middle category is the one that is the most problematic fro the theory, since the theory relies on the notion that educationally-privileged ethnic groups should avoid violence, not use it occasionally.

To me, the cross-tabulation suggests that Shaykhutdinov’s hypothesis has little support. I am not sure if there is some colinearity in the regression that moves the coefficients into being significant, but my guess is that the substantive effects of educational equality are pretty small.

From my reading, here are the four key takeaways:

  • Shaykhutdinov should be commended for taking on a crucial question that needs further inquiry. We need more research on the relationship that education has on the choice to use nonviolent or violent resistance (or both), using methodological techniques that can help us to account for potential endogeneity.
  • There seems to be a weak positive association between educational advantage (as well as general education of the overall population) and the adoption of nonviolent strategies of protest, though the association needs further testing.
  • If Shaykhutdinov’s hypothesis is robust, then educational inequality may be a “structural” impediment to nonviolent mobilization. This means that people who want to promote the spread of nonviolent resistance (and reduce the spread of violence) should focus on improving the educational status of ethnic minorities in troubled countries.
  • If there ends up being no support for Shaykhutdinov’s hypothesis, we should be encouraged that educational inequality is not a “structural” impediment to nonviolent mobilization. Even the educationally disadvantaged should be able to adopt and practice nonviolent principles. This should scare autocrats, because it means that one of their tools–deprivation of educational rights–doesn’t really make a difference in terms of an ethnic minority’s ability to rise up and make demands of them.

Regardless, we need to know the answers to these vital questions. Kudos to Shaykhutdinov for taking the first cut.

{ 2 comments }

not so fast... August 4, 2011 at 7:41 am

Ever hear of an ecological fallacy? No, probably not…

John Bullock August 5, 2011 at 11:54 am

> groups with higher levels of educational attainment
> are more likely to engage in non-violent protest

A very good working paper just reached a rather different conclusion:

Education as Liberation
Willa Friedman, Michael Kremer, Edward Miguel, Rebecca Thornton
April 2011
http://www.nber.org/papers/w16939

Here is the relevant bit from the abstract:

We assess the social and political impacts of a randomized girls’ merit scholarship incentive program in Kenya that raised test scores and secondary school enrollment [...] the program increased objective political knowledge, and reduced both acceptance of political authorities and satisfaction with politics. However, in our Kenyan context, this rejection of the status quo did not translate into greater perceived political efficacy, community participation or voting intentions. Instead, the program increased the perceived legitimacy of political violence.

Comments on this entry are closed.

Previous post:

Next post: