President George W. Bush hoped to establish a flourishing multiethnic democracy in Iraq by overthrowing Saddam Hussein; instead, the colonial contraption, formerly held together by monarchy and dictatorship, has disintegrated along ethnic lines. Iraqi Kurds have enjoyed de facto independence since the Gulf War of 1991. At times, much of the Sunni-majority area of Iraq has been part of the Islamic State, which has incorporated much of Syria’s Sunni territory, another European colonial creation.

ISIS, which combines revolutionary terror with apocalyptic theology, cannot be incorporated into a global state system whose very legitimacy it rejects. But most revolutions either end in counterrevolution or in mellowing. Whether ISIS is overthrown or burns out, a future decision to recognize a nonradical Sunni majority state in its region, should one emerge, might make more sense than trying to reassemble the shattered European-created states of Iraq and Syria. Similar considerations are relevant in the case of the state formerly known as Libya.

As of this writing, the U.S. government is officially encouraging the restoration of both Iraq and Syria to their preconflict borders, as multiethnic states with liberal democracy and multiparty political systems. Public criticism that this project is almost certainly doomed and utopian is discouraged. In 2006, then Sen. Joe Biden and Leslie H. Gelb, president emeritus of the Council on Foreign Relations, were widely criticized for proposing ethnic federalism in Iraq—a step well short of partition.

Then in August 2015, in his final news conference, Gen. Ray Odierno, the former top U.S. commander in Iraq, acknowledged that in the future “Iraq might not look like it did in the past.” Asked about partition, he said,

I think that is for the region and politicians to figure out, diplomats to figure out how we want to work this, but that is something that could happen. It might be the only solution but I’m not ready to say that yet.

One headline read “U.S. Army Gen. Odierno Retires amid Controversy over Iraq Remarks.”

Writing over a year ago in Vox, journalist Max Fisher denounced the idea of formally partitioning Iraq. Sounding like an arrogant European colonial administrator of a century ago, Fisher argues that Iraq should not be partitioned along sectarian lines even if most of its inhabitants want it to be. “The fact that Iraqis want this does not make it any wiser or sounder.”

What is the alternative? According to Fisher,

The only real way to solve sectarianism is by solving sectarianism, to overcome it by getting people to abandon the idea that they exist in a zero-sum contest for security with other sectarian groups that can only be regarded as innately hostile. It means building a new social contract in which security and rights are guaranteed irrespective of ethnicity or religion, signing everyone on to that new contract, and then proving it can actually work.

Fisher concludes, “That is a tremendous political and military challenge that will take years or decades.” But to judge by the congressional election of 2006 and the presidential election of 2008, which were among other things referendums on Bush’s Iraq policy, the American people do not think that a decades-long commitment to nation building in the Middle East is worth the blood and treasure it has already cost.

In any event, like the failed seventy-year project in Moscow to create a “New Soviet Man,” replacing Russians, Ukrainians, Georgians, Kazakhs, Azeris and others, the project of creating a postethnic, nonsectarian “New Iraqi Individual” is a fantasy.

More than four thousand U.S. military personnel have died in Iraq, while more than two thousand American soldiers have died in Afghanistan since 2003. How many more Americans should die for the Sykes-Picot borders or the Durand Line?

If Middle Eastern borders created by long-dead European colonial administrators are allowed to unravel, where will the process end? The result may resemble the map of the Middle East imagined by the U.S. military expert Lt. Col. (ret.) Ralph Peters in an essay entitled “Blood Borders” in the Armed Forces Journal in 2006. Among other things, Peters envisioned the partition of Iraq into a Shia state, a Sunni state and a “Free Kurdistan,” as well as the creation of a “Free Baluchistan” from parts of present-day Afghanistan and Iran. Around the world, conspiracy theorists interpreted the map as a secret plan by the U.S. government to redraw the borders of the region.

In 2013, in the New York Times, Robin Wright engaged in a similar exercise, this time envisioning a “Sunnistan” stretching from western Syria to eastern Iraq, along with three successor states to Libya: Tripolitania, Fezzan and Cyrenaica. Wright’s map also featured the division of the Arabian peninsula into “Wahhabistan,” North Arabia, South Arabia, Eastern Arabia, Western Arabia, North Yemen and South Yemen.

Like Mazzini’s vision of a postimperial Europe of nation-states, exercises like these will inevitably get details wrong. But the larger point is that Mazzini was broadly right about what Europe would look like. Will the map of the Middle East in 2050 or 2100 look more like those of Peters and Wright than the map of today?

Outside of the Middle East, Africa is the area with the greatest mismatch between nationalities and the borders drawn by European colonial administrators generations ago. A subtle form of racism is evident in the use of the term “tribes” for linguistic and cultural communities outside of Europe, some of which have far more people than many European nations.

The economists Stelios Michalopoulos and Elias Papaioannou recently published a study on the correlation between ethnic violence in Africa and the legacy of borders drawn by European powers beginning with the nineteenth-century “Scramble for Africa.” They discovered that conflict likelihood is 8 percent higher and conflict intensity 40 percent higher in places where borders divide members of the same ethnic group than in countries that are ethnic homelands. Moreover, the study continues, “The within-country analysis shows that partitioned ethnicities are significantly more likely (11 percent–14 percent increased likelihood) to engage in civil wars that have an explicit ethnic dimension.”

Writing in 2010 in Foreign Policy, in an essay entitled “Africa Needs a New Map,” G. Pascal Zachary effectively rebutted those who argue that ethnic conflicts that now take place within arbitrary multinational states would simply continue as wars among partitioned nation-states.

Of course, splicing up Africa’s countries is no panacea for the continent’s woes. You might argue, for example, that conflicts would not be stopped at all; they would just go from being civil wars to interstate conflict between two divorced neighbors. That may well happen, and of course no conflict is good news. But the international community has much stronger deterrents for such country-to-country spats than internal civil war. And new states would likely be reluctant to incur the repercussions of diplomatic and economic isolation.

AS A to foreign policy, the “Pottery Barn Rule” was enunciated by former secretary of state Colin Powell, as a caution to bear in mind the consequences of wars of regime change. “You break it, you own it.” To the Pottery Barn Rule might be added the Humpty Dumpty Rule: if all the king’s horses and all the king’s men cannot put Humpty Dumpty back together again, then maybe the king shouldn’t bother trying. There are human and economic costs even to bloodless partitions, like the Velvet Divorce of the Czech Republic and Slovakia.

Nor need national self-determination always take the form of national independence. In some cases, ethnic federalism, or ethnic power-sharing—“consociational democracy”—can reconcile overarching unity with a high degree of autonomy for ethnocultural communities. Switzerland, a successful multiethnic state, employs both kinds of arrangements. As we have seen, both Wilson and Lloyd George suggested Balkan power-sharing arrangements as an alternative to the “Balkanization” of the Habsburg Empire which eventually took place.

But when a multinational state like Iraq or Cyprus has broken down into war, when de facto borders have been established and when, in some cases, population transfers have taken place voluntarily or by ethnic cleansing, it is worth asking how much value the United States and the international community should put on restoring lines drawn on a map long ago by European imperial officials.

While the process has sometimes been tragically violent, the formation of new nation-states—either from the partition of former multinational states, or the unification of “multi-state nations” like West and East Germany—has been the necessary outcome of democratic self-determination. As John Stuart Mill wrote in his 1861 Considerations on Representative Government, “One hardly knows what any division of the human race should be free to do if not to determine with which of the various collective bodies of human beings they choose to associate themselves.” Washington has no business encouraging the breakup of functioning multinational states into new nation-states. But neither should America, born of secession from a multinational empire, reflexively reject that option for other nations that seek to assume a “separate and equal station” among “the powers of the earth.”