Any sane and decent person wants this war to be short and the Iranian people to be free from the brutal theocracy that has ruled them for half a century.

But hope is not a strategy.

At this point—Day Five of Donald Trump’s war of choice against Iran—both of those outcomes look unlikely. We Americans have learned the hard way over the last 60 years that it is easy to start wars, and hard to end them. And wars rarely end the way their architects imagine.

That familiar line of military strategists is worth calling to mind again: No battle plan survives first contact with the enemy.

Or, as Mike Tyson put it, “Everyone has a plan until they get punched in the mouth.”

The US military is the best in the world and it’s not close. Our forces have demonstrated that fact in spectacular ways during this presidency; Trump has ordered them into action more than any president in the modern era. But Iran is different.

It is a proud, mountainous, heavily defended nation of 90 million people with an armed force of over a million active-duty and reserve personnel and a vast, indigenous arsenal of ballistic/cruise missiles and drones. Iran’s forces are no match in a direct confrontation with the combined power of the US and Israeli attack—but the leadership has long recognized that fact, and prepared extensively for an asymmetric fight.

A short war that topples the regime and triggers an uprising that liberates the Iranian people is possible, but unlikely.

The more realistic scenario is a conflict that ends unsatisfactorily without becoming catastrophic. It may well be the best that can be hoped for at this point.

Trump could declare victory and walk away after taking out Iran’s missile programs and nuclear capacities (even though he boasted after the Israeli-US strikes last year that the Iranian nuclear program had been “obliterated,” it wasn’t), leaving behind an ungodly mess—and the ayatollahs in power—in one of the world’s crucial regions.

Iran’s odious leadership—what’s left of it—would pick a new Supreme Leader (the son of the old one looks to be the frontrunner) sue for peace, and get back to repressing their people, rebuilding their missile/drone arsenals, and preparing for the next conflict.

Oil would flow. The region’s abiding conflicts would return to normal—tense, volatile, but manageable. Trump would have a parade.

It could be worse. Far worse. Let’s look at how.

The chaos of war inside Iran could ignite ethnic uprisings inside and outside its borders.

In western Iran, the Kurds have been fighting Tehran on and off for decades. Now, according to several reports, the US and Israel are in active talks with Kurdish forces in Iran to arm them and foment an uprising aimed at overthrowing the regime.

But Kurdish revolts rarely stay confined to one country.

There are an estimated 25-30 million Kurds; they are the largest ethnic group in the world without a state, spread across parts of Turkey, Iraq, Iran, Syria, and Armenia. In each of those places they already possess armed movements, political organizations, and—in northern Iraq and northeastern Syria—semi-autonomous territory.

If Iranian Kurdish regions erupt in rebellion, the idea of a united Kurdish state, something Kurds have dreamed about for more than a century, could suddenly feel possible.

That would alarm every government in the region, Turkey in particular.

Suddenly, a war that began as an American-Israeli confrontation with Iran could trigger a regional ethnic conflict stretching across four countries.

History has seen stranger things.

Another dark possibility involves not a Kurdish state but the fragmentation of Iran itself.

Iran is not a homogeneous country. Roughly half its population belongs to ethnic minorities—Kurds, but also Azeris, Arabs, Baluchis, Turkmen and others—many living in border regions with strong cross-national ties. Western imperialists drew borders for generations in the Middle East as they saw fit, but those lines are still resented by many in the region.

There are those in Washington and Tel Aviv who quietly speculate that a weakened Iran might break apart along these lines. Some Israeli analysts and former intelligence officials have floated the idea of exploiting Iran’s ethnic divisions—and even partitioning the country—but it has never been an official Israeli policy. And the neoconservative Foundation for Defense of Democracies, based in Washington has also argued for partition.

The argument: A smaller, divided Iran would pose less of a military threat to Israel and to the Gulf states.

There are even maps floating around of what a “fractured Iran” might look like:

But history offers a warning. When Yugoslavia collapsed in the 1990s, the result was not peaceful partition but a decade of civil war, ethnic cleansing, and outside intervention.

I was in Bosnia in the early 1990s and Kosovo in the late 1990s. It was vicious. And heartbreaking. The Balkans are still reckoning with the aftermath of those wars.

The third scenario is less dramatic but more likely. The Islamic Republic survives.

Iran’s leaders have spent decades preparing for precisely this moment. They anticipated air strikes, regime-change rhetoric, and attempts to decapitate their command structure.

If the regime survives the opening phase of the war—as looks likely now—it will shift into a long campaign of attrition. That war has in fact already begun.

Iran’s missiles and drones are a real problem for the the US, Israel and the Gulf states. Our missile/drone defense systems inventories are not infinite.

And Iran can still activate proxies in the region for attacks against American bases and businesses. The shipping lanes of the Persian Gulf and the Bab-el-Mandeb Strait remain vulnerable.

The US—in our eternal naïveté about the world as it is—dreams of an uprising of the Iranian people to throw off their chains. Let us hope that happens. But external attacks often strengthen authoritarian regimes rather than weakening them. Nationalist sentiment can rally populations around governments they otherwise despise.

If Iran absorbs the initial blows and fights back effectively—even modestly—the regime could emerge battered but politically strengthened.

And the United States could find itself trapped in a war it did not intend to fight for very long.

Before any nation chooses war, its leaders have a duty to ask hard questions.

What if the enemy doesn’t collapse?

What if the region destabilizes?

What if events spiral beyond control?

Nightmare scenarios are useful. They aren’t necessarily arguments against taking military action. They are arguments for thinking about it first.

And based on almost everything we have seen from this president and from senior officials in his administration—thinking was in short supply. They can’t even get their stories straight about why Trump took our country to war against Iran. Getting out will be much harder because of that abdication of responsibility, that hubris.

—Terry