How Civil Wars Start: And How to Stop Them

by Barbara F. Walter

When Saddam Hussein was captured, researchers who study democratization didn’t celebrate. We knew that democratization, especially rapid democratization in a deeply divided country, could be highly destabilizing. In fact, the more radical and rapid the change, the more destabilizing it was likely to be.

The United States and the United Kingdom thought they were delivering freedom to a welcoming population. Instead, they were about to deliver the perfect conditions for civil war.

Former Baathist party leaders, intelligence officials, and Iraqi army officers, along with Sunni tribal chiefs, soon realized that if they wanted to retain any power in the new democracy, they had to act fast. Nascent insurgent organizations began to form as early as the summer of 2003.

OVER THE PAST one hundred years, the world has experienced the greatest expansion of freedom and political rights in the history of mankind.

President Bush, after invading Iraq, felt confident that the United States would establish “a free Iraq at the heart of the Middle East” and inspire a “global democratic revolution.”

Full democracies are less likely to go to war against their fellow citizens and against citizens in other democracies.

But the road to democracy is a dangerous one. When scholars around the world first began collecting data on civil wars, in the early nineties, they noticed an interesting correlation: Since 1946, right after World War II ended, the number of democracies in the world had surged—but so had the number of civil wars. They seemed to be rising in tandem.

Civil wars rose alongside democracies. In 1870, almost no countries were experiencing civil war, but by 1992, there were over fifty.

It turns out that one of the best predictors of whether a country will experience a civil war is whether it is moving toward or away from democracy.

It is in this middle zone that most civil wars occur.

Experts call countries in this middle zone “anocracies”—they are neither full autocracies nor democracies but something in between.

It’s why we were so critical of President Bush’s decision to try to catapult Iraq from autocracy to democracy in 2003. We understood that a major political transition in Iraq was likely to trigger civil war instead. Experts have seen this pattern repeated around the world over the last century.

most conflict researchers tend to rely on the one that has been compiled by the Polity Project at the Center for Systemic Peace—a

One of the most influential measures in the dataset is called the Polity Score, which

If a country receives a score of +10, for example, its national elections have been certified as “free and fair,” no important social groups are systematically left out of the political process, and major political parties are stable and based on mass national constituencies. Norway, New Zealand, Denmark, Canada—and, until recently, the United States—all have a +10 rating.

The Political Instability Task Force (the one I later joined) came up with dozens of social, economic, and political variables—thirty-eight, to be precise, including poverty, ethnic diversity, population size, inequality, and corruption—and put them into a predictive model. To everyone’s surprise, they found that the best predictor of instability was not, as they might have guessed, income inequality or poverty. It was a nation’s polity index score, with the anocracy zone being the place of greatest danger. Anocracies, particularly those with more democratic than autocratic features—what the task force called “partial democracies”—were twice as likely as autocracies to experience political instability or civil war, and three times as likely as democracies.

In the chaos of transition, these leaders often fail.

These weaknesses set the stage for civil war because impatient citizens, disgruntled military officers, or anyone with political ambitions can find both a reason and an opportunity to organize a rebellion against the new government.

A primary reason for revolt is that democratic transitions create new winners and losers: In the shift away from autocracy, formerly disenfranchised citizens come into new power, while those who once held privileges find themselves losing influence. Because the new government in an anocracy is often fragile, and the rule of law is still developing, the losers—former elites, opposition leaders, citizens who once enjoyed advantages—are not sure the administration will be fair, or that they will be protected. This can create genuine anxieties about the future: The losers may not be convinced of a leader’s commitment to democracy; they may feel their own needs and rights are at stake.

From their perspective, they were better off fighting while they were still relatively strong, rather than wait for their rivals to consolidate power.

A painful reality of democratization is that the faster and bolder the reform efforts, the greater the chance of civil war. Rapid regime change—a six-point or more fluctuation in a country’s polity index score—almost always precedes instability, and civil wars are more likely to break out in the first two years after reform is attempted.

But democratization is possible. Though the path to democracy is treacherous, the risk of civil war fades when a country takes its time, evolving its political system gradually. Mexico weathered democratization relatively peacefully. Its transition lasted nearly twenty years, from 1982 until 2000, when the National Action Party (PAN) became the first opposition party to win a presidential election since 1929.

We’ve seen this in Poland, where the Law and Justice Party won elections in 2015; the president, prime minister, and deputy prime minister have since systematically taken control of the courts, restricted free speech, targeted political opponents, and weakened the electoral commission.

Democratic countries that veer into anocracy do so not because their leaders are untested and weak, like those who are scrambling to organize in the wake of a dictator, but rather because elected leaders—many of whom are quite popular—start to ignore the guardrails that protect their democracies. These include constraints on a president, checks and balances among government branches, a free press that demands accountability, and fair and open political competition.

gaining support by exploiting citizen fears—over jobs, over immigration, over security.

They understand that if they can persuade citizens that “strong leadership” and “law and order” are necessary, citizens will voluntarily vote them into office.

Then, once in power, these leaders plunge their countries into anocracy by exploiting weaknesses in the constitution, electoral system, and judiciary.

they are able to consolidate power in ways that other politicians are unable, or unwilling, to stop. This increasing autocratization puts countries at higher risk of civil war.

For a decaying democracy, the risk of civil war increases almost the moment it becomes less democratic.

As a democracy drops down the polity index scale—a result of fewer executive restraints, weaker rule of law, diminished voting rights—its risk for armed conflict steadily increases.

To many Russian-speaking Ukrainians, his autocratic tendencies were the lesser of two evils; they would rather have an authoritarian who was on their side than a democrat who was not.

When Yanukovych announced his intention to strengthen economic ties with Russia, rather than the European Union, citizens—many of them young people from European-leaning western Ukraine—decided they’d had enough. Demonstrations, known as the Euromaidan protests (“Euro” for the desire for ties with Europe, “maidan” for the main public square in Kyiv) began in the capital city and spread across the country. Protesters toppled the statue of Lenin in Kyiv, clashed with police, and demanded new elections, freedom of speech, and closer ties to the EU.

Within weeks of Yanukovych’s ouster, separatist militias declared their own autonomous states—the Luhansk and Donetsk people’s republics—and quickly captured stocks of weapons to defend their territory.

THE LOVE AFFAIR with democratization that marked the twentieth century and the very beginning of the twenty-first century is over. It ended in 2006, when the number of democracies around the world reached its peak.

Others have evaded civil war by using a gradual, sneakier hand, like Putin in Russia or Orbán in Hungary. These leaders maintain the guise of democracy—elections and limited individual freedoms—and cement their popularity with effective propaganda, control of the media, and sometimes xenophobia. Citizens, rather than revolt, have acquiesced to their rule.

Why do some countries safely navigate the road through the anocracy zone, while others become engulfed in cycles of chaos and violence? The story of Iraq again offers a clue. When I asked Noor to describe what changed before civil war erupted in her homeland, she looked at me for a moment. Soft-spoken and reserved, she radiated the quiet confidence of someone who doesn’t break easily. Her face, however, was heavy with sadness. “People began asking whether you were Shia or Sunni,” she said.

There are three widely used datasets that measure a country’s system of governance: Polity V, Freedom House, and V-Dem.

Polity V, for example, is particularly interested in the durability of different types of governments and their political institutions, focusing heavily on the democratic and autocratic features of a country.

V-Dem (which was introduced in 2014) is keen to uncover the many varieties of democracy around the world and includes five detailed dimensions of democracy (electoral, participatory, egalitarian, deliberative, and liberal dimensions).

Freedom House focuses heavily on individual freedoms and includes detailed measures on citizens’ politic...

scholars have found a high level of agreement in terms of how countries are coded in each dataset, and high intercorrelations between th...

This man had single-handedly—albeit ruthlessly—held together one of the most ethnically diverse countries on the planet.

Instead, that June, on the six hundredth anniversary of a historic battle between the Serbs and the Ottomans—a key event in Serbian history—he delivered a speech in Kosovo to a crowd of roughly one million Serbs from all over Yugoslavia. It was the obligation of Serbs, he asserted, to “remove disunity, so that they may protect themselves from defeats, failures, and stagnation in the future.” He vowed that Serbs would no longer compromise or suffer defeat.

In the first five years after World War II, 53 percent of civil wars were fought between ethnic factions, according to a dataset compiled by James Fearon and David Laitin, two civil war experts at Stanford University. Since the end of the Cold War, as many as 75 percent of civil wars have been fought by these types of factions.

One particular feature of countries turned out to be strongly related to political instability and violence. It was an acute form of political polarization that they called “factionalism.” Countries that factionalize have political parties based on ethnic, religious, or racial identity rather than ideology, and these parties then seek to rule at the exclusion and expense of others.

They found that the biggest warning sign of civil war, once a country is in the anocracy zone, is the appearance of a faction.

Two variables—anocracy and factionalism—predicted better than anything else where civil wars were likely to break out.

Countries that are considered “factionalized” have identity-based political parties that are often intransigent and inflexible. Boundaries between them are rigid, leading to intense competition and even combat.

The groups that are competing are often about the same size. In fact, it’s this balance of power between the two groups that creates such fierce rivalry; the stakes of winning or losing are high. These parties can also be personalistic in nature, revolving around a dominant figure who often appeals to ethnic or religious na...

Experts assess the level of factionalism in a country based on a five-point scale that goes from a fully competitive political system (5) to a fully repressed system...