The damage wrought by civil wars in Syria, Yemen, Libya, and Iraq since 2011 is staggering. Perhaps around 600,000 people have died—likely including more than half a million Syrians—and 17 million people have been displaced from their homes. From Jordan to Tunisia and beyond, refugee flows have strained budgets and inflamed domestic politics.

Driven by the emergence of the Islamic State and the resurgence of Al-Qaeda, the number of global suicide terrorist attacks averaged 478 in the last five years, compared to 374 in the five years before 2011. Growing authoritarianism and deepening regional rivalries in the wake of these wars have seemingly prevented coordinated responses to urgent regional challenges, such as climate change, food and water scarcity, and burgeoning employment and development needs.

In light of the urgency of addressing this string of civil wars, I reviewed conflict datasets and the academic literature for insight into the changing trends of conflict and why these particular wars have become so intractable. (I followed the convention of defining wars as passing the threshold of 1,000 annual battle deaths and the broader category of conflicts as passing the threshold of 25 annual battle deaths.) The data have suggested an evolving physical and geopolitical conflict geography since the end of World War II. The transformations in the nature of civil conflict and in the global order might be divided into three rough periods: the Cold War (from around 1946 to 1991), the post-Cold War (from 1991 to around 2011), and the post post-Cold War (around 2011 to the present).

Notwithstanding entrenched civil wars in contexts such as Colombia, Angola, and Indochina, Cold War era civil conflicts—many of which were national liberation or ideological proxy conflicts, or both—were, on aggregate, shorter and therefore generally of lesser lethality. One study found that the average duration of civil wars increased from five years in 1950 to fifteen years by 2000. Meanwhile, the number of armed conflicts increased from fewer than 20 before 1950 to a post-World War II peak of 52 in 1991, the vast majority of them civil conflicts. These new conflicts brought new patterns of violence. Formerly conflict-prone regions such as Central America and Southeast Asia had stabilized by the time of the collapse of the Soviet Union, even as internal conflicts engulfed large regions of Africa and Eurasia in the early 1990s.

But after the genocidal horrors of Rwanda and Bosnia, the post-Cold War era saw a sharp reduction in the number of conflicts to below 35 by the early 2000s. The number of civil wars declined to as low as four and for several years there was not a single interstate war. This was a period of unusually intense international mediation efforts. According to one study, internal conflicts in the late Cold War period were seven times more likely to end in military victory than with peace settlements. However, by the 2000s the ratio had been reversed: internal conflicts were five times more likely to end in peace settlements than with military victories. Thus, some 47 United Nations peacekeeping missions were initiated between 1991 and 2011, nearly three times as many as during the previous four decades.

But that happy post-Cold War period—during which one prominent social scientist speculated, “we may be living in the most peaceable era in our species’ existence”—is now over. Since 2012, the global incidence of conflict has again spiked, matching the modern peak of 52 in 2015. In 2016, there were twelve civil wars, the most since the end of World War II, all but one of which—the war in Nigeria—occurred in Muslim-majority countries from sub-Saharan Africa to South Asia. Not only is conflict spreading, its lethality is surging, with the Middle East as its epicenter. Since 2011, the number of global combat deaths has nearly quadrupled to the highest level since 1994 during the Rwandan genocide, roughly 70 percent of which have occurred in conflicts within Arab League states.

This new era of Middle East conflict also coincides with a breakdown of the international order. In today’s post post-Cold War period, both the capacity and willingness of the international community to meaningfully respond to conflicts has evaporated. Of the four UN peacekeeping missions created since 2012, not one is currently active in any of the twelve civil wars.

Amid these structural evolutions, three conceptual problems highlighted by the academic literature shed light on the persistence of Middle East conflict. First is the stakes (or indivisibility) problem. Whereas modern interstate wars almost always end with the belligerent states intact, it is difficult to identify intermediate solutions between victory and defeat in internal conflicts. As Fred Ikle noted in his 1971 classic Every War Must End: “If partition is not a feasible outcome because the belligerents are not geographically separable, one side has to get all, or nearly so, since there cannot be two governments ruling over one country.”

Several quantitative studies [for example here, here, and here] have found that the stakes problem is acute in conflicts defined by ethnic and sectarian divisions, as in Iraq, Syria, and Yemen. By implication, so long as regional and international opposition to the formal partitioning of Middle Eastern states remains strong, zero-sum outcomes will be difficult to avoid. Thus, negotiated settlements that preserve state sovereignty may only be possible through the devolution of political power through increased local autonomy.

Second is the commitment problem. As Barbara Walter has argued, it is difficult to give belligerents confidence that self-enforcing settlements—those lacking international enforcement mechanisms—will be honored. Conflict settlements are necessarily signed at periods of government weakness. But as these governments gain strength, they have little incentive to honor agreements, particularly if rebel groups have simultaneously been demobilized. This dynamic helps to explain why recidivism rates are significantly higher for civil conflicts ending in political settlements than outright victory.

The implication is that even if political settlements of Middle Eastern wars can be found, the likelihood of a recurrence of fighting is high without international enforcement mechanisms, such as peacekeepers. Given the lack of great power consensus on the Middle East’s wars and the current dysfunction in the UN Security Council, there is a need to explore possible mechanisms to support Middle East settlements, including consideration of the difficulty of peacekeeping amid high risks of terrorism.

Third is the veto-player problem. As David Cunningham and others have noted, conflicts with multiple independent actors last longer because the range of acceptable agreements to all parties is greatly reduced, alliances are prone to shift, and there are incentives for fighting groups to hold out as long as possible. Several quantitative analyses [for example here, here, and here] suggest that external military interventions render civil wars particularly difficult to resolve, especially when they result in counter-interventions. One recent study found that whereas only 4 percent of civil wars were internationalized in 1991, 40 percent had become so in 2015. This is certainly true of the Middle Eastern civil wars, all of which began as local conflicts, but have become internationalized, greatly complicating efforts to deescalate and contain them.

An implication of the veto-player problem is that big-tent multilateral processes, such as have commonly been employed to deal with Syria or Libya, are unlikely to be successful. When everyone has a seat at the table, consensus may be impossible. A better approach might be to carefully sequence multiparty negotiations around only the essential set of veto players, perhaps beginning with the key external actors and then proceeding to principal factions on the ground.

After the mid-1990s, the surge of African and Eurasian conflicts began to recede, in part because of a changing geostrategic environment. Overcoming the deep structural factors exacerbating the new Middle Eastern civil wars is perhaps an even more daunting task that will require prudent decisions in the years ahead.