How the War in Ukraine Might End

Hein Goemans grew up in Amsterdam in the nineteen-sixties and seventies, surrounded by stories and memories of the Second World War. His father was Jewish and had hidden “under the floorboards,” as he put it, during the Nazi occupation. When Goemans came to the United States to study international relations, he recalled being asked in one class about his most formative personal experience of international relations. He said that it was the Second World War. The other students objected that this wasn’t personal enough. But it was very personal for Goemans. He recalled attending a commemoration of the fortieth anniversary of the liberation of Amsterdam by Canadian forces, in May of 1985. Many of the Canadian soldiers who took part in the liberation were still alive, and they re-created the arrival of Canadian troops to liberate the city. Goemans remembers thinking that the people of Amsterdam would be too blasé to attend the commemoration, and being moved that he was wrong. “The entire city was packed with people by the roadside,” he told me recently. “I was really surprised at how deeply it was felt.”

Goemans, who now teaches political science at the University of Rochester, wrote his dissertation on war-termination theory—that is, the study of how wars end. A great deal of work, Goemans learned, had been done on how wars start, but very little on how they might conclude. There were, perhaps, historical reasons for this oversight: the nuclear armament of the United States and the Soviet Union meant that a war between them could end human civilization; not just some dying, but the death of everything. The study of war during the Cold War thus gave rise to a rich vocabulary about deterrence: direct deterrence, extended deterrence, deterrence by punishment, deterrence by denial. But the Cold War ended, and wars kept happening. Goemans saw an opportunity for an intellectual intervention.

In his dissertation, and in his subsequent book, “War and Punishment,” Goemans laid out a theory of how and why some wars ended quickly and others dragged brutally on. The war in the title was the First World War; the “punishment” was what leaders in Germany, in particular, feared awaited them if they brought home anything less than a victory. When Goemans’s book came out, in 2000, it was the first modern full-length study devoted entirely to the problem of war termination, and it helped launch the field.

Traditionally, Goemans writes, wars were thought to end because one side surrendered. “Until the vanquished quits, the war goes on,” as one author put it, in 1944. But the empirical record showed this to be at best an incomplete account. It usually took two sides to start a war, even if they had different culpability, and it usually took two sides to end it; the vanquished may accept the terms that were proposed last week, but what was to keep the winner from inventing new terms? The classic example from the First World War was the Bolsheviks’ refusal, in the wake of their seizure of power in Russia, to continue the fight against Germany; proclaiming “neither war nor peace,” they simply left the negotiations at Brest-Litovsk. “Literally the vanquished quit,” Goemans writes. But the Germans, instead of accepting this, kept advancing into Russia. Only after the Bolsheviks agreed to even harsher terms than had been proposed just three weeks earlier did the Germans agree to their exit from the war.

More recent theoretical literature had acknowledged the two-sidedness of war, Goemans writes, but here, too, important aspects had been missed. War theory imported from economics the concept of “bargaining,” and wars were thought to begin when the bargaining process—over a piece of territory, usually—broke down. The most common cause of the breakdown, according to war theorists (and again borrowing from economics), was some form of informational asymmetry. Simply put, one or both sides overestimated their own strength relative to their opponent’s. There were many reasons for this sort of informational asymmetry, not least of which was that the war-fighting capacity of individual nations was almost always a closely guarded secret. In any case, the best way to find out who was stronger was to actually start fighting. Then things became clear quite quickly. Many wars ended in just this way, with the sides reëvaluating their relative strengths and opting to make a deal.

But there were other kinds of wars, in which factors besides information predominated. These factors, in part because they did not play prominent roles in economics, were less well understood. One was the fact that contracts in the international system—in this case, peace deals—had little or no enforcement mechanism. If a country really wanted to break a deal, there was no court of arbitration to which the other party could appeal. (In theory, the United Nations could be this court; in practice, it is not.) This gave rise to the problem known as “credible commitment”: one reason wars might not end quickly is that one or both sides simply could not trust the other to honor any peace deal they reached. In his 2009 book “How Wars End,” Goemans’s colleague Dan Reiter used the example of Great Britain in the late spring of 1940, after the fall of France. Britain was losing the war and had no certainty that the United States would enter in time to save it. But the British fought on, because they knew that no deal with Nazi Germany could be trusted. As Winston Churchill put it to his Cabinet, in his inimitable way: “If this long island story of ours is to end at last, let it end only when each one of us lies choking in his own blood upon the ground.”

The other factor that had been ignored in the literature, according to Goemans, was domestic politics. States were considered unitary actors with set interests, but this left out the internal pressures placed on the government of a modern nation-state. Goemans created a data set of every leader of every war-fighting country between 1816 and 1995, and coded each according to a tripartite system. Some leaders were democrats; some were dictators; and some were in between. According to Goemans, democrats tended to respond to the information delivered by the war and act accordingly; at the very worst, if they lost the war but their country still existed, they would get turned out of office and go on a book tour. Dictators, because they had total control of their domestic audience, could also end wars when they needed to. After the first Gulf War, Saddam Hussein was such a leader; he simply killed anyone who criticized him. The trouble, Goemans found, lay with the leaders who were neither democrats nor dictators: because they were repressive, they often met with bad ends, but because they were not repressive enough, they had to think about public opinion and whether it was turning on them. These leaders, Goemans found, would be tempted to “gamble for resurrection,” to continue prosecuting the war, often at greater and greater intensity, because anything short of victory could mean their own exile or death. He reminded me that on November 17, 1914—four months after the First World War began—Kaiser Wilhelm II met with his war cabinet and concluded that the war was unwinnable. “Still, they fought on for another four years,” Goemans said. “And the reason was that they knew that if they lost they would be overthrown, there would be a revolution.” And they were right. Leaders like these were very dangerous. According to Goemans, they were the reason that the First World War, and many others, had dragged on much longer than they should have.

I recently spoke to a number of war-termination theorists, including Goemans, to see what the theoretical perspective could tell us about the war in Ukraine. The theorists turned out to be an engaged and lively group, most of them glued to Twitter and Telegram, in multiple languages, as they tried to follow the war in real time. They believed the wars that they had studied could shed light on the current conflict. Apparently, they were not the only ones who thought so. The war theorist Branislav Slantchev, one of Goemans’s former students and a professor at the University of California, San Diego, told me that in August he’d been asked to participate in a Zoom symposium on war termination convened by a U.S. intelligence agency.

Reiter, the author of “How Wars End,” was intrigued by the fact that the conflict in Ukraine was such an old-fashioned war. There was very little cyber warfare, and Russia had used just a few hypersonic missiles. He said that, on the Russian side, “it’s artillery, armor, infantry, brutality against civilians. That’s the twentieth century.” And on the Ukrainian side it was the same: “They have reasonably sophisticated weapons, coupled with enough training, coupled with a lot of bravery. Things have not changed as much as we had thought.”

Tanisha Fazal, a scholar at the University of Minnesota who is writing a book on battlefield medicine, was struck by the low ratio of Russian wounded to killed. The historical ratio in the last hundred and fifty years has been around three or four to one. In recent wars, such as in Afghanistan, the U.S. had managed to get the wounded-to-killed ratio as high as ten to one, meaning that fewer soldiers were dying after being wounded. The Russians were estimated to be back down to four to one. The reason, Fazal said, was that the Russians have not managed to establish air superiority; they cannot get their wounded soldiers out quickly enough, and therefore many of them die.

More broadly, the war had exhibited many traits that were familiar to theorists of war. Vladimir Putin’s initial miscalculation that he could overrun Ukraine in a matter of days was a classic case of informational asymmetry; it was also a classic instance of a repressive regime being fed poor information by its own people. Everyone agreed that we were faced with a “classic” credible-commitment problem. Russia claimed that it could not trust Ukraine to not become, in essence, a NATO state; Ukraine, for its part, had no reason to trust a Russian regime that had repeatedly broken promises and invaded it in February with no provocation. But the resolution of the credible-commitment problem was complicated. In the Second World War, it was resolved by the destruction of the Nazi regime, the rewriting of Germany’s constitution, and the partition of Germany. But not many wars end with such absolute outcomes.

To add to the complications, this war, like others, is dynamic. A great deal has happened since Russia invaded Ukraine on the morning of February 24th. The revelations of Russian weakness and Ukrainian strength have buoyed the Ukrainian public; the discovery of the massacres of civilians at Bucha and now Izyum have enraged it. If once there was space in Ukrainian public opinion for concessions to Russia, that space has now closed. “Sometimes war generates its own causes of war,” Goemans said.