Optimism in War

An article in Foreign Policy, “Why Hawks Win,” by Daniel Kahneman, a Nobel laureate in economics at Princeton, and Jonathan Renshon, a graduate student in Harvard’s department of government, argues that people proposing the use of military force tend to win arguments over government policy because of fundamental psychological predispositions. As the two put it, “These psychological impulses—only a few of which we discuss here—incline national leaders to exaggerate the evil intentions of adversaries, to misjudge how adversaries perceive them, to be overly sanguine when hostilities start, and overly reluctant to make necessary concessions in negotiations. In short, these biases have the effect of making wars more likely to begin and more difficult to end.” Political leaders are on this account “receptive to advisors who offer highly favorable estimates of the outcomes of war.” The evidence Kahneman and Renshon cite is necessarily historical evidence, since while our propensity to overestimate our abilities as drivers of automobiles is the sort of thing that can be tested in a lab, that quirk is not what interests them here. They are interested in war and foreign policy, and propositions about those fields can only be tested historically, if at all.

Does history bear them out? It is hard to say, since the long data series used to test such propositions tend to be assembled by political scientists, whose professional deformation makes them strikingly insensitive to the difficulties of comparing historical data for purposes of this kind. But we certainly can look at the specific historical evidence cited in this article by Kahneman and Renshon, and see if it holds up. For example, the two write that “the optimistic bias and the illusion of control are particularly rampant in the run-up to conflict. A hawk’s preference for military action over diplomatic measures is often built upon the assumption that victory will come easily and swiftly. Predictions that the Iraq war would be a ‘cakewalk,’ offered up by some supporters of that conflict, are just the latest in a long string of bad hawkish predictions. After all, Washington elites treated the first major battle of the Civil War as a social outing, so sure were they that federal troops would rout rebel forces.”

Actually, the first phase of the Iraq War—the part that ended with the fall of Baghdad and destroyed Sunni Arab domination of Iraq, probably forever—was easier than many predicted; Saddam’s regime was destroyed at a trivial cost to the Allies , around a third as many Allied losses as in the first Gulf War, itself an astonishingly one-sided victory. In the first Gulf War, staggering Allied losses had also been predicted, but in the event hundreds of thousands of Iraqi troops were routed at a cost to the Allies of a few hundred men. In the second Gulf War, what was underestimated was not the cost of destroying the Iraqi army that underpinned Saddam’s tyranny—that was overestimated—but the cost, perhaps the very possibility, of shaping a decent Iraq after we had destroyed Baathist rule. We underestimated our power to destroy and overestimated our power to create, which is not quite what Kahneman and Renshon are suggesting.

How about the American Civil War? It is true that Washington elites overestimated our chances on the eve of Bull Run. But shortly thereafter, McClellan again and again underestimated his chances and overestimated the strength of his enemies, for the next long phase of the war. This arguably protracted the Civil War and increased suffering. And consider the implicit bias here, which assumes that an accurate estimate of your chances lowers your will to fight, which is further assumed to be a good thing. Is it in all cases? What if Northern elites had in 1861 accurately estimated the cost of preserving the Union, and subsequently of abolishing slavery? Had they then decided that the game wasn’t worth the candle, that would have probably have been a catastrophe; an illiberal, even more racist nineteenth century would have been a plausible outcome of such a decision. So overestimating your chances in war can be a blessing for all concerned. Similarly, if British elites had acted on a rational calculation of their chances in June of 1940, they would have made peace, which would probably have meant a comparable disaster.

Back to Kahneman and Renshon on military history. Writing of the First World War, they assert, “In fact, almost every decision maker involved in what would become the most destructive war in history up to that point predicted not only victory for his side, but a relatively quick and easy victory. These delusions and exaggerations cannot be explained away as a product of incomplete or incorrect information. Optimistic generals will be found, usually on both sides, before the beginning of every military conflict.”

Some obvious exceptions spring to mind. British and French generals fatally underestimated their chances of fighting Hitler between 1936 and 1939, and statesmen were eager, very eager, to listen to them. Kahneman and Renshon acknowledge that the hawks were right on that occasion, and the doves wrong, but they do not, for my taste, acknowledge it extensively enough, since the surplus mortality that can be laid to the door of the doves that time is on the order of 50 million people. And is that the only significant exception to this alleged rule? I was last week reading about Ludwig Ritter von Benedek, the Austro-Hungarian general charged with stopping the Prussians in 1866, who probably underestimated his chances and is normally faulted for having been beaten in his own mind before he began. That underestimate may have been a significant misfortune, for German unification achieved in the historical fashion is not normally thought to have been a smashing success.

On another note, Kahneman and Renshon assert that “during the run-up to World War I, the leaders of every one of the nations that would soon be at war perceived themselves as significantly less hostile than their adversaries.” They did indeed, but some of them were right about that, and others wrong. Kahneman and Renshon imply that a more accurate reading of potential adversaries’ desires would have forestalled war. The reverse is at least as likely: We now know that Germany was determined to go to war with Russia, and Austria-Hungary to go to war with Serbia. That first determination was the overwhelming cause of the First World War. It is possible that a clearer understanding of Allied hostility to the prospect of German triumph might have put off the war, but that seems the reverse of what Kahneman and Renshon are arguing.

Kahneman and Renshon also make a few remarks about political choices in the Middle East, with the implication, at least to my eye, that mistaken assumptions about the malevolence of adversaries have been a serious problem. Maybe so. It is at least possible, however, that from the early 1990s through the year 2000 the Israelis were more prone to overestimate than to underestimate the good faith of their Palestinian interlocutors. In 1979-1980, and at various points since, the American government was at least as likely to underestimate as to overestimate the malevolence of the people running the Islamic Republic of Iran. Given the century just past, when powerful states were led by Hitler, Stalin, and Mao, and also by Wilhelmine and Japanese militarists, and other adversaries included people like Pol Pot, it seems strange to assert that overestimating the malevolence of adversaries was the worst sin of Western statesmen.