The Next Unrests: Predicting World Uprisings

The Next Unrests: Predicting World Uprisings
1 Star2 Stars3 Stars4 Stars5 Stars (No Ratings Yet)

Despite the number of seasoned observers stationed at the world’s main hot spots to explore tensions within a society, nobody seems to have seen coming the 2011 uprisings in the Arab countries. With the model designed by Jean-Charles Grey for his research degree in criminology, it would have been possible to see that the conditions were met for such uprisings.

While studies on the subject are usually limited to considering only a single element predisposing to a type of trigger, the researcher addressed the issue taking into account all economic factors identified by various authors and applied to all types of violent disorder groups. According to the researcher, “The subject had never been treated in such quantitatively macroscopic and comprehensive manner”.

For the purposes of this exploratory research, the violent events selected are those that fall within the definition of a riot, which is popular demonstrations that turn violent until the mass uprising could lead to the downfall of a government . Only excludes conflicts involving the presence of armed groups, whether the legitimate armed or insurrection and riots from sporting event.

A total of 410 violent events occurred in 80 countries between 2000 and 2010 were selected for analysis. “These events were important enough to be reported by the media,” says the researcher.

Distribution of riots rates between 2000 and 2010 – Jean-Charles Gris.

Multiplicity of factors

Depending on the context and trigger, the uprising riot can be qualified as political, socioeconomic or ethnic-religious. For the period studied, the Kyrgyz Republic, which topped the list for ethnic and religious riots at a rate of more than 13 violent incidents per 10 million inhabitants. The first and second ranking of the most unsettling countries goes respectively to Somalia in terms of political riots (rate of 17.5 events) and Guinea-Bissau to the riots socioeconomic rate (42).

All categories, Canada has a rate less than 1 lift for 10 million reported. “If Canada is on the map, it is mainly because of the riots during protests against police brutality and riots in North Montreal in 2008,” says the researcher.

Analysis of Jean-Charles Grey showed that if each category of riots is primarily correlated with political, economic or ethnic triggers, all other factors may also play a role in the outbreak of a riot.

We can see, for example, that the democracy index is inversely correlated not only with political riots, but also with economic riots. Unemployment, for its part, is not a predisposing factor to economic disturbances, it also promotes ethnic and religious riots. Health expenditures are significantly related to economic and political riots, but not religious riots. The level of corruption is in turn linked to both political riots and economic unrest.

The main factor associated with ethno-religious riots remains the division of a country, either by ethnic, linguistic or religious. Followed by military expenditure, precarious democracy  and low urbanization. “In rural areas, there is less pluralism and those who are different from themselves appear more threatening,” says the researcher to explain this last factor.

The intersection of all these data leads Jean-Charles Grey to note that collective violence does not occur mainly in dictatorships or in democracies but in hybrid systems, which are a transition between dictatorship and democracy. “Once a dictatorship begins to open to the outside, people see that they can be better-off than what they have and begin to demand more rights, resources. China, for example, can indefinitely hold billions of people demanding more democracy. ”

Correlation between GDP per Capita (Horizontal Axis) and the rate of Riots (Vertical Axis). Jean-Charles Gris

Arab Spring

When applied separately to each category of riots, the statistical models developed by the researcher explained 33 to 40% of the variance of such events in a given country. Applied to the three classes at a time, the single model explains up to 50% of the variance. “It is statistically robust, he says. In the social sciences, being able to explain 50% of the variance of behavior is certainly not a negligible success and such a tool can be very useful. ”

Jean-Charles Gray also tested the predictive value of the model by applying it to a sample of countries that have experienced the “Arab Spring.” For Tunisia, Egypt, Morocco, Yemen and Iran, the model predicts riots higher rate than the media reported during the period 2000 to 2010. “This disparity between the forecast and what has been reported shows that the conditions for an increase in the riots were present in these countries,” he said.

Only Algeria, which shows a higher rate of riots prediction, responds less well to the model, while Libya was not part of the core sample.

This research was co-led by masters Samuel Tanner and Marc Ouimet, professors in the School of Criminology at the University of Montreal.

Source: Translated from Revolts could be better predicted. Université de Montréal. Daniel BarilThe Full Study (in French) by Jean-Charles Gris can be found here.
Photo Credits: Untitled By Cosmopolita. / FlickR