In Isaac Asimov's science fiction series Foundation, the mathematician Hari Seldon invents a method, called psychohistory, that predicts social behavior as accurately as statistical mechanics predicts the behavior of gases. The catch is that the predictions may be thwarted if influential people learn about them and consequently change their behavior. Seldon's model predicts that civil war will destroy his galactic civilization. He never publicizes his prophecy, so it comes true.

A conference I attended last week at Ohio State University (OSU) got me thinking about psychohistory and the paradox of prediction. Titled "Hybrid Warfare: The Struggle of Military Forces to Adapt to Complex Opponents from the Ancient World to the Present," the conference examined past wars for lessons that could help us prevail in Iraq and Afghanistan and similar wars in the future. The speakers were military historians.

The implicit assumption of the entire conference was that there will always be wars. The meeting's chief organizer was Peter Mansoor, who teaches military history at OSU and has written well-received books about World War II and the war in Iraq. With his close-cropped hair, square jaw and no-nonsense demeanor, Mansoor looks and speaks like a soldier, and in fact he was one. After graduating from West Point in 1982, he served in the Army for 26 years, rising to the rank of colonel. He commanded the 1st Brigade, 1st Armored Division in Iraq from 2003 to 2005 and later served as executive officer to General David Petraeus, commander of U.S. forces in Iraq. Mansoor knows war, as a warrior and scholar.

"No," Mansoor replied immediately when I asked him if he thought international war would ever end, as some scholars have recently proposed. He acknowledged that since World War II there have been relatively few international wars and no wars between major powers (although of course the U.S. and the Soviet Union fought through proxies). But he likened our era to the century of relative calm following the end of the Napoleonic Wars in 1815. "You had a long period of stability, punctuated by some regional conflicts—the Crimean War, the Franco-Prussian War—but nothing that drew the entire continent into a massive conflict."

By the early 20th century, some prominent intellectuals proposed that the growing economic interdependence of major European powers made war between them unlikely. "World War I and World War II show that obviously wasn't the case," Mansoor said. "And I don't think there's anything that could convince me that major war or even another world war couldn't happen in the future. Again, we could be in this long period of stability in the wake of the end of the cold war, where there are regional conflicts, little brushfire wars, as we're seeing in Iraq and Afghanistan. It may be decades or even a century before there is another major war or even a world war. There might not be. But I wouldn't rule it out."

The U.S. must remain prepared to wage war not only against hostile states, insurgencies and terrorist groups but also against major powers, including those possessing nuclear weapons, Mansoor said. Rather than banning all nuclear weapons, as Barack Obama has proposed, Mansoor believes the U.S. should retain a nuclear arsenal, "because you could never know if your opponent has somehow re-created the capability in secret." Nuclear weapons may deter but ultimately not prevent war between major powers from breaking out. "I just think it means that if it does break out, it would be a lot more deadly than what we've seen in the past."

I see a couple of differences between our era and the early 20th century that make me more optimistic than Mansoor. There are many more democracies in the world now, and democracies rarely fight against each other (although they obviously fight against non-democracies, as Mansoor pointed out). Moreover, modern media rub our noses in war's ugliness as never before.

What bothers me most about Mansoor's vision of the future is its potential to be self-fulfilling. To the extent that our political and military leaders see the world as a dangerous place, filled with actual and potential enemies, nations are more likely to remain fearful, distrustful and heavily armed. States will also be more likely to carry out preemptive attacks, as Japan did against the U.S. in 1941, and as the U.S. did against Iraq in 2003.

Consider these fictional counterfactuals: If Hari Seldon had publicized his forecast of civil war, the fighting might just have started sooner, as one faction sought to gain an early, decisive victory over the other. Conversely, if he had predicted enduring peace, maybe that vision would have become self-fulfilling by engendering hope and trust instead of fear.

Just because predictions of peace have failed in the past does not mean that they will forever. Does it?