Civil warsHow to stop the fighting, sometimes

Bringing an end to conflicts within states is vexatious. But history provides a guide to the ways that work best

WHEN Hussein el-Husseini moved into a modest flat with a sea view in Beirut in 1983, the surrounding streets were littered with the detritus of an eight-year-old civil war. When Mr Husseini became Speaker of the Lebanese parliament the following year, the war still had six years to run. By the time it ended it had claimed 150,000 lives.

Yet the solution, says Mr Husseini, was clear more or less from the beginning. The country’s various religious groups, each with its own militias, had to share power. Lebanon could not be conquered by one side, nor divided among all. Its people are too mixed; Mr Husseini’s prominent Shia Muslim family includes Christians and Sunnis, and that is par for the course. “But the militias were against it,” he says.

Latest updates

See all updates

Attempts by Mr Husseini and others, notably the tycoon Rafik Hariri, to reach the obvious but fugitive solution took him to the outside powers sponsoring the militias: America, France, Iran, Israel, Syria and Saudi Arabia. He was repeatedly rebuffed until, in 1989, finally despairing of the war, the outsiders agreed to stop paying their proxies. Mr Husseini quickly convened representatives from the various communities and militias in Taif, a resort in Saudi Arabia. After a lot of haggling, they signed an accord that led to peace a year later.

Ending civil wars is hard. Hatreds within countries often run far deeper than between them. The fighting rarely sticks to battlefields, as it can do between states. Civilians are rarely spared. And there are no borders to fall back behind. A war between two states can end much where it began without the adversaries feeling in mortal danger. With nowhere safe to go home to, both sides in a civil war often feel they must carry on fighting if they are to escape slaughter. As those fighting in Syria know, defeat often looks like death, rather than retreat (see article).

New mutiny

Yet civil wars do end. Of 150 large intrastate wars since 1945 fewer than ten are ongoing. Angola, Chad, Sri Lanka and other places long known for bloodletting are now at peace, though hardly democratic.

And recently civil wars have been ending sooner. The rate at which they start is the same today as it has been for 60 years; they kick off every year in 1-2% of countries. But the number of medium-to-large civil wars under way—there are six in which more than 1,000 people died last year—is low by the standards of the period. This is because they are coming to an end a little sooner. The average length of civil wars dropped from 4.6 to 3.7 years after 1991, according to Kristian Skrede Gleditsch, a professor at the University of Essex.

Mr Gleditsch is one of a growing number of political scientists studying civil wars. The field, long overshadowed by studies of superpower conflict, is coming into its own. Its participants do not claim that all civil wars are the same—the range of causes and types of conflict is obvious. But the sheer number of civil wars allows scholars to attempt, at least, a quantitative approach to the factors that affect the wars’ outcomes. And governments are keen to learn from their insights. When Roy Licklider, a professor at Rutgers University in New Jersey, was invited to the State Department this summer to conduct a seminar for officials dealing with Syria, he found that officials were “frantically trying to read the underlying dynamics.”

Click here for chart 1

So far, nothing has done more to end the world’s hot little wars than winding up its big cold one. From 1945 to 1989 the number of civil wars rose by leaps and bounds, as America and the Soviet Union fuelled internecine fighting in weak young states, either to gain advantage or to stop the other doing so. By the end of the period, civil war afflicted 18% of the world’s nations, according to the tally kept by the Centre for the Study of Civil War, established at the Peace Research Institute Oslo, a decade ago. When the cold war ended, the two enemies stopped most of their sponsorship of foreign proxies, and without it, the combatants folded. More conflicts ended in the 15 years after the fall of the Berlin Wall than in the preceding half-century (see chart 1). The proportion of countries fighting civil wars had declined to about 12% by 1995.

The outcomes of civil wars changed, too, according to Scott Gates, the director of the centre. Until 1989, victory for one side was common (58%). Nowadays victories are much rarer (13%), though not unknown; the Sri Lankan government defeated Tamil rebels in 2009. At the same time negotiated endings have jumped from 10% to almost 40%. The rest of the conflicts peter out, subsiding to a level of violence below the threshold of war—though where that threshold should lie is a matter of some debate (see article).

The main reason for jaw-jaw outpacing war-war is a change in the nature of outside involvement. In the Cold War neither of the superpowers was keen to back down; both would frequently fund their faction for as long as it took. Today outside backers are less likely to have the resources for such commitment. And in many cases, outsiders are taking an active interest in stopping civil wars.

Civil hands unclean

The motives vary. Some act out of humanitarian concern. Others seek influence, or a higher international profile. But above all, outsiders have learned that small wars can wreak preventable havoc. Fractious Afghanistan bred al-Qaeda; the genocide in tiny Rwanda spread murder across a swathe of neighbours. In coastal west Africa, violence is passed back and forth between Guinea, Liberia, Sierra Leone and Ivory Coast like a winter cold round an office. “The best predictor of a civil war is having one next door,” says Mr Licklider.

Outsiders can weigh in on one side, backing their desire for peace with cold steel. In Mali a brawl involving a mutinous army, ethnic rebels and Islamic extremists ended after less than a year thanks to French soldiers, who intervened in January and forced a partial reconciliation.

Ever fewer powers, though, have the stomach for an overt armed intervention. At the same time, there is rarely great enthusiasm for following the advice of Edward Luttwak, author of a famous essay called “Give war a chance”. It is true that military victories tend to provide more stable outcomes than negotiated settlements, which—especially in the absence of external peacekeepers—often break down when the underlying problems that led to the conflict in the first place resurface.

There may be some conflicts better fought to their conclusion than left unresolved. But the charms of victory can be overstated. “The violence needed for a military victory also tends to destroy the state institutions required to stabilise a country in the long term,” warns Andrew Exum, a former Pentagon adviser. And the factors that draw in foreign countries—a commitment to humanitarian values, the exhibition of diplomatic prowess, and the forestalling of contagion—also encourage outsiders to seek negotiation.

Orchestrating talks towards an end like the one brokered in Lebanon requires strong nerves and stomachs. Civil wars tend to end as messily as they are fought. Negotiations often take place in parallel with combat. For years Nepalese guerrillas negotiated with the government while also pummelling it, finally signing a peace deal in 2006. The prospect of an ending can quite often intensify the fighting.

Sometimes the dispute is so intractable that no agreed solution short of the break-up of the state seems possible. Wars of identity—those in which populations are mobilised by grievances that have ripened over decades or centuries—are the most likely to belong to this category.

The drawbacks to partitions in such cases, especially where they require large-scale population movements, are well rehearsed. Sects and tribes are rarely neatly divided, waiting for a line to be drawn between them. Separating them, if need be by force, will make some safer, but it will cause others great misery and may well spark new conflicts. When Pakistan split from India, it was saddled with a coup-prone state and a war in Kashmir. And many nations with fissiparous tensions at home recoil from the idea of any partition anywhere, lest it be seen as a precedent.

Ancient grudge

Still, some break-ups do make sense. South Sudan’s government is lousy, and fighting continues along the border set up with the rest of Sudan two years ago. But most independent observers agree that the south made the right choice in negotiating to split off. The Arab elite in the north was never going to change its murderous attitude toward black southerners that brought about decades of miserable war and the death of 2m people. And there is little worry that South Sudan will look so attractive as to encourage secession elsewhere. Few minorities would accept such pain to win a seat at the UN.

In talks aimed at a one-state solution, history suggests that several things can better the odds of success. The prospect of UN blue helmets is one. Combatants often require security guarantees. In Bosnia the outgunned Muslims could only imagine resting what rusty arms they had when assured of protection by trusted outsiders. In conflicts where parties agree not just to pause but also to disarm—thus further reducing the chances of more war—this is essential. Guerrillas worry that, without weapons, they will face oppression once again and stash some away. Since its founding, the UN has completed 53 peacekeeping missions (see chart 2). The 15 ongoing ones employ almost 100,000 in uniform. Civilian mediators can be useful too, sometimes opening up negotiating tracks states cannot, and being trusted to operate without their own political agenda.

Another essential in peace negotiations is combatants’ acceptance, at least privately, that the hope of winning has died away. Anyone still contemplating victory will find negotiated compromises unbearable. Were fighters to listen to the experts with the databases, they would come to the table earlier; a majority of victories come in the first year of a civil war. But most cling to their original dreams long after all possibility of attaining them has faded.

Only when the fighters have been disillusioned, can mediators get to work—and then only for a limited period. Civil wars unresolved for more than a decade seem to drag on for ever, with both sides resigned to perpetual fighting, too disgusted or exhausted to face their enemies across the negotiating table. The armed conflict in the dense mountains of Colombia has been going since 1964. In some cases causality may run the other way. Conflicts last because they are unresolvable.

And conflicts recrudesce, too. Peace settlements can break down; indeed some worry that, at the moment, it is particularly easy for rebels to go back to war. Heavy weapons are easier to come by than once they were and insurgency tactics have been refined in Iraq and elsewhere.

Glooming peace

One reason for backsliding is that peace often fails to bring the prosperity that might give it lasting value to all sides. Power-sharing creates weak governments; nobody trusts anyone else enough to grant them real power. Poor administration hobbles business. Ethnic mafias become entrenched. Integration is postponed indefinitely. Lacking genuine political competition, with no possibility of decisive electoral victories, public administration in newly pacified nations is often a mess.

Lebanon is a prime example. When the sects carved up power in 1989 they fixed quotas for all public bodies. Even department heads in the telecoms regulatory authority are appointed according to a religious formula. Loyalty is to sects, not the public. Services are virtually non-existent; reliable electricity supplies are rare. The latest government fell in March and nothing has replaced it. Still, many Lebanese prefer this state of affairs to the bloodletting of the 1980s. Better to condemn one’s children to a poorly run country than to endanger their lives.

The question of how outsiders can push settlements along is among the trickiest in international relations. One idea is to engineer a change in leadership. Warlords who start conflicts are rarely prepared to admit that they cannot win, and their charisma can be central to the cause. The capture of Abdullah Ocalan by Turkish forces in 1999 was such a blow to the Kurdistan Workers’ Party that peace talks have been going on ever since. Peru’s Shining Path withered after the 1992 capture of Abimael Guzmán. Leadership changes are a factor in the termination of between 25% and 40% of civil wars, according to James Fearon, a professor at Stanford University.

Changing leaders is not the only way to intervene. By using military power or curtailing the flow of money, outsiders can engineer what scholars call a “mutually hurting stalemate”. In this neither side can advance and the cost of holding tight is high—making peace the least bad option. The NATO air campaign in 1999 against Serbia to protect Albanians in neighbouring Kosovo is an example; bombs rained down on his capital until Milosevic caved in. In 1980 Britain ended Zimbabwe’s civil war by simultaneously squeezing the government and persuading Mozambique and Zambia to threaten to end the aid that they supplied to the rebels making gains in the field.

Mutually hurting stalemates are hard to bring about. Knowing that the enemy is under the cosh can tempt embattled combatants to hold out. Separate measures are needed for the two sides. Governments often need less pressure, since they find stalemates painful in themselves. Without full control of their territory, legitimacy seeps away. This weakens them and encourages others who have grievances to make a stand, adding to the problems. Rebels, on the other hand, may require extra pressure, since they are less likely to find a stalemate intrinsically painful. Fighting becomes their raison d’être; keeping the ability to fight on is all they need. “The guerrilla wins if he does not lose,” noted Henry Kissinger. “The conventional army loses if it does not win.”

The trickiest part for outsiders is getting both sides into painful positions at the same time. It is easy for the combatants to perceive advantages on their side, or the other, which are not there. It gets a lot easier when one side is willing to accept, tacitly at least, the need for a negotiated peace, and starts to act in ways that stabilise a stalemate, rather than seeking to break out of one. So its forces might conquer territory and then withdraw in order to show strength and willingness to compromise. Or it might counter enemy attacks with enough force to check them, but without following through in a way that escalates the fight. Foreign intelligence can be useful in calibrating such fine-tuned actions.

Mr Husseini, who helped bring peace to Lebanon, says he knew the years of pleading were finally getting results when the militias stopped receiving money from abroad. He never won any prize for his role; the militias eventually pushed him out of politics. Sitting in his home under a picture of Pope John Paul II, he wonders how many more people could have been saved if the guns had fallen silent earlier.

The imposing view of the Mediterranean he once enjoyed from his flat is a distant memory, blocked by new buildings. It doesn’t matter, he says. After all, construction shows that Lebanon has regained a measure of peace and prosperity. It even manages to offer refuge to a million Syrians who have fled their own civil war.

View comments
Reuse this content

Tell us what you think of Economist.com

Need assistance with your subscription?