At one level the answer is obvious: an internal fight for control of a nation. But in the bloody conflicts that split modern states, our policy makers often understand something deeper to be at work. The vengeful slaughter that has ripped apart Bosnia, Rwanda, Syria, and Yemen is most often seen as the armed eruption of ancient and complex hatreds. Afghanistan is embroiled in a nearly impenetrable melee between Pashtuns and smaller ethnic groups, according to this thinking; Iraq is split by a long-suppressed Sunni-Shia feud. The coalitions fighting these wars are seen as motivated by the deepest sort of identity politics, ideologies concerned with group survival and the essence of who we are.

This view has long shaped America’s engagement with countries enmeshed in civil war. It is also wrong, argues Fotini Christia, an up-and-coming political scientist at MIT.

In a new book, “Alliance Formation in Civil Wars,” Christia marshals in-depth studies of the recent wars in Afghanistan, Iraq, and Bosnia, along with empirical data from 53 civil conflicts, to show that in one civil war after another, the factions behave less like enraged siblings and more like clinically rational actors, switching sides and making deals in pursuit of power. They might use compelling stories about religion or ethnicity to justify their decisions, but their real motives aren’t all that different from armies squaring off in any other kind of conflict.

“The idea that today’s enemy can be the next day’s friend was very compelling to me,” Christia said in an interview. “We should not be surprised to see groups switching sides, based on how the balance of power on the ground evolves.”

How we understand civil wars matters. Most civil wars drag on until they’re resolved by a foreign power, which in this era almost always includes the United States. If she’s right, if we’re mistaken about what motivates the groups fighting in these internecine free-for-alls, we’re likely to misjudge our inevitable interventions—waiting too long, or guessing wrong about what to do.

***

Civil wars always have loomed large in the collective consciousness. Americans still debate theirs so vociferously that a blockbuster film about Abraham Lincoln feels topical 150 years after his death. Eastern Europe saw several years of ferocious killing in the round of civil wars that followed World War II.

Such wars have been understood as fights over differences that can’t be resolved any other way: fundamental questions of ideology, identity, creed. A disputed border can be redrawn; not so an ethnic grudge. In the last two decades, identity has become the preferred explanation for persistent conflicts around the world, from Chechnya to Armenia and Azerbaijan to cleavages between Muslims and Christians in Nigeria.

This thinking allows for a simple understanding, and conveniently limits the prospect for a solution. Any identity-based cleavage—Jew vs. Muslim, Bosnian vs. Serb, Catholic vs. Orthodox—is so profoundly personal as to be immutable. The conventional wisdom is best exemplified by a seminal 1996 paper by political scientist Chaim Kaufmann, “Possible and Impossible Solutions to Ethnic Civil Wars,” which argues that bitterly opposed populations will only stop fighting when separated from each other, preferably by a major natural barrier like a river or mountain range.

During the 1990s, this sort of ethnic determinism drove American policy toward Bosnia and Rwanda. It was popularized by Robert Kaplan’s book “Balkan Ghosts,” which was read in the Clinton White House and presented the wars in the former Yugoslavia as just the latest chapter in an insoluble, four-century ethnic feud. Like Kaufmann, Kaplan suggested that the grievances in civil wars could only be managed, never reconciled.