In 2004, James D. Fearon, a political scientist at Stanford, published a study, “Why Do Some Civil Wars Last So Much Longer Than Others?,” in which he and a colleague analyzed scores of civil wars fought between 1945 and 1999. Some of the findings were intuitive: civil wars end quickly when one side has a decisive military advantage over the other; poor countries with natural resources to export often have long internal wars, because whoever controls the resources also controls the national treasury. Other findings were novel, such as the fact that wars following coups d’état tend to be short. In another study, “Ethnicity, Insurgency, and Civil War,” Fearon and the political scientist David D. Laitin discovered that even though in nations with exceptional ethnic pluralism, like Syria and Iraq, lines of conflict may be defined by ethnic identity, pluralism itself is not a notable predictor of civil war; poverty is a much more significant factor.

Rereading these works in light of the infuriating problem of the Islamic State, two discouraging findings stand out. In 1945, many civil wars were concluded after about two years. By 1999, they lasted, on average, about sixteen years. And conflicts in which a guerrilla group could finance itself—by selling contraband drug crops, or by smuggling oil—might go on for thirty or forty years. The Revolutionary Armed Forces of Colombia, or FARC, has been around since 1964, sustained in no small part by American cocaine consumption.

Steve Coll looks at the history of civil wars to try and shed light on how the war with ISIS might unfold. The last point in the passage is an easy one to forget when it comes to wars. At a very basic level, wars require some stock or resources from which to fund the flow of war. We speak of burn rates in technology, but in the context of war it has a much more serious meaning.

Asymmetry works in both directions here. On one hand, the industrialized nations of the world have asymmetric military power to deploy against entities like ISIS. On the other hand, ISIS and its equivalents have an asymmetric investment in the battle in that region. No industrialized nation has the appetite to stay in that region in the long run, and that only erodes with time.

From the American intervention in Somalia, in 1992, through the French intervention in Mali, in 2013, industrialized countries have been able to deploy ground forces to take guerrilla-held territory in about sixty days or less. The problem is that if they don’t then leave, to be replaced by more locally credible yet militarily able forces, they invite frustration, and risk unsustainable casualties and political if not military defeat. This has been true even when the guerrilla forces were weak: the Taliban possesses neither planes nor significant anti-aircraft missiles, yet it has fought the United States to a stalemate, and the advantage is now shifting in its favor.

If President Obama ordered the Marines into urgent action, they could be waving flags of liberation in Raqqa by New Year’s. But, after taking the region, killing scores of ISIS commanders as well as Syrian civilians, and flushing surviving fighters and international recruits into the broken, ungoverned cities of Syria and Iraq’s Sunni heartland, then what? Without political coöperation from Bashar al-Assad, Russia, Iran, Hezbollah, Iraqi Shiite militias, Turkey, the Al Qaeda ally Al Nusra, Saudi Arabia, the Gulf States, and others, the Marines (and the French or NATO allies that might assist them) would soon become targets for a mind-bogglingly diverse array of opponents.