According to conventional thinking, the end of the Cold War helped unleash a spate of civil conflicts among ethnic groups whose simmering animosities had been stifled by superpower hegemony. It was commonly accepted that ethnic and religious diversity made countries more prone to civil war, and it was believed that conflicts could be predicted to break out in areas with the strongest ethnic or political grievances.
But research on the causes of civil wars by political scientists David Laitin and James Fearon refutes these popular theories.
"We're finding that none of this holds up when we look at the data," Fearon said in a recent interview. Instead, the professors argue, internal wars are more likely to happen in mountainous, impoverished, politically unstable regions that favor rural guerrilla warfare or insurgency. "It's not that grievances are irrelevant, rather they are ubiquitous," said Laitin. "When there is a rebellion, there is no assurance that solving its stated grievance will cause it to stop, because it might be a proxy for something else."
Finding the root causes of civil wars is critical, the researchers stress, because they are so devastating. "This is an immense public health problem," Laitin said.
Since the end of World War II, 16.5 million people have died in internal conflicts, compared with 3.3 million in interstate wars. About 122 civil wars have raged since 1945, compared with 25 conventional wars. Internal conflicts last six years on average and bring about widespread refugee dislocations and economic devastation -- as seen in Afghanistan, Somalia and Lebanon. Despite this heavy toll, the professors say, civil wars have been studied far less than conventional conflicts and are not properly understood.
Fearon and Laitin have detailed their findings in a paper titled "Ethnicity, Insurgency and Civil War" that will be published early next year in the American Political Science Review. They argue that the prevalence of internal war is mainly the result of the accumulation of protracted postcolonial conflicts since the 1950s and 1960s, rather than a sudden change associated with the post-Cold War international system. The number of civil conflicts has increased over time because they break out faster than they end -- on average, 2.3 wars have begun annually but only 1.7 conflicts have been resolved each year.
"We shouldn't assume, as some policy makers in the Pentagon have sometimes assumed, that this is a post-Cold War phase involving civil wars that will go away and we'll be back to great power politics," said Fearon. "Rather, if our analysis is right, what we're really seeing is a longer-term trend based on decolonization that created a system dominated numerically by weak states."
Based on data on civil wars from 1945 to 1999, Fearon and Laitin argue that, after controlling for per capita income, ethnically or religiously diverse countries have been no more likely to experience significant civil violence than more homogenous states. The criterion for including an insurgency in the study is a minimum of 100 deaths annually and 1,000 during the course of the conflict. (Consequently, the Basque conflict in Spain and France is not included because 836 people have died during its three-decade-long campaign for an independent homeland.)
Grievances such as economic inequality, lack of democracy or civil liberties, or state discrimination against minority religions or languages are less effective predictors of civil conflict than weak states marked by poverty, large size and instability, Fearon and Laitin explained. Instead, the paper noted, "The factors that explain which countries have been at risk for civil war are not their ethnic or religious characteristics but rather the conditions that favor insurgency."
According to the political scientists, the "technology of insurgency" makes it difficult to end civil conflicts. In a guerrilla war, a small number of rebels can inflict considerable damage. The standard rebel tactic, they explained, especially at the start of a war, is to provoke indiscriminate retaliation by the government in order to help rebels recruit more sympathizers to their movement.
Weak states without effective counter-insurgency capabilities, such as Indonesia and Sierra Leone, take the bait and burn entire villages instead of trying to determine the exact identity of the rebels, Laitin said. By driving people off the land, this tactic tends to give them few alternatives except to join the rebels. "It's just a really difficult thing to defeat once it gets started," Fearon said. Furthermore, insurgency is a very flexible military strategy that can be used by various political agendas, ranging from communism in Southeast Asia and Latin America to Islamic fundamentalism in Afghanistan, Algeria and Kashmir.
According to a military rule of thumb, a government needs a force of 10-to-1 to defeat a guerrilla insurgency, Fearon said. In conventional conflicts, the ratio is 3-to-1. "If you have 5,000 rebels, you need 50,000 soldiers -- less than a quarter of the countries in the world have an army that large," he said. "That's why this problem is unlikely to go away."
The spread of democracy and tolerance for ethnic religious minorities should be major foreign policy goals because they are desirable for their own sake, Laitin and Fearon said -- not because they are "magic bullets" for preventing or ending civil war. Establishing ethnic partitions, on the other hand, is a dangerous course because it increases the chances of launching an insurgency.
Implementing policies to redress grievances could be an important tool for resolving an ongoing conflict, although the researchers found little evidence that civil wars occur where large cultural divisions or grievances exist. "But it seems quite clear that broad and powerful grievances are produced by civil war -- indeed, this is often a central objective of rebel strategy. This could well pose obstacles to settlement," the political scientists said.
Economic growth should be supported because it tends to be
associated with more competent governments. More important, the
professors argue, international and nongovernmental organizations
should develop programs that improve legal accountability within
developing world militaries and police. They should also make aid
to governments fighting civil wars conditional on their employing
effective counterinsurgency strategies that do not, through their
brutality or killing innocent people, create additional insurgents.
When a state fails, however, or when regional stability is
threatened, Laitin and Fearon have proposed a form of
"neo-trusteeship," through which the United Nations would delegate
responsibility to a lead nation and a group of international
organizations to rebuild the country.
Stanford Report, September 25, 2002