Diplomatic Historians and the Return to Theory

18 Historically Speaking November/December 2006 Politics and psychology often conspire to make people believe what they need to believe in order for them to have faith in their policies. This is necessary to give them the confidence to move forward and to persuade others to join them. My theory suggests that the main reason why the Bush administration relied on the expectation diat Ahmad Chalabi could quickly stabilize Iraq and permit U.S. forces to withdraw was that this belief allowed them to proceed widi the war with relatively little psychological conflict or political opposition. To take another example, John Coogan argues quite convincingly that at the start of World War I Woodrow Wilson was able to— indeed, had to—delude himself into believing that he was upholding American neutrality against the British blockade. To have seen clearly would have forced him to make an unpalatable choice between threatening Great Britain, whose defeat he believed would have been very much against the American interest, and openly rejecting the defense of neutral rights that was deeply embedded in international law and American history. Better, both politically and psychologically, to convince himself that he was not making such a choice.2 There is a high cost to such behavior. Many choices are made implicitly rather than explicitly. Courses of action that might appear attractive upon hard and prolonged thought are never contemplated. It is often very difficult to understand which beliefs in fact drive behavior. Mediocre scholars often imagine they are smarter than decision makers, but good ones have to grapple with the fact that it is hard to reconstruct the pressures that leaders feel and must adapt to. Trachtenberg is of course correct that we are not without weapons in this struggle for understanding . We can often see which beliefs form first and which seem to be dragged along in their wake, and we can look for consistency between beliefs and actions. Silences in the documents also are diagnostic : it is telling when questions that are not only obvious in retrospect but that normally would be addressed go unexamined. In the endeavor to understand beliefs and how they fit with behavior, political scientists are more inclined than historians to make explicit use of the comparative and hypothetico-deductive methods. There is no magic in either of them, and they are not unfamiliar to historians, although the latter often forego the labels. In its essence, the comparative method simply means seeking to establish causation by comparing several cases to see if the effect changes when the possible cause has changed. Trachtenberg often does this, and so does Ernest May in his study of American expansionism and the Spanish -American War.· The hypothetico-deductive method is less familiar, especially in its explicit form. The basic point is the importance of taking one's theory or causal argument seriously and probing it by asking what one would expect to find if die argument is correct. In other words, in addition to moving from facts to explanations, the scholar asks herself what evidence she would expect to see (aside from that which produced the hypothesis) if the hypothesis is in fact correct. This is difficult to do because it not only requires drawing out the implications of one's argument, but, even more important , putting aside what one knows happened in order to figure out what should have happened if the hypothesis is correct. Even though the method rarely works out neady or without dispute, it can be powerful for pointing scholars toward important evidence and, perhaps even more, for leading them to understand the implications of the arguments they and others are making. It may seem unhistorical because it moves away from direcdy examining what happened, but this method is central to how we can understand the past. A final issue combines both method and substance . This is how much consistency we expect to find in the behavior of leaders and states. In his book more than in the piece under discussion,' Trachtenberg 's analysis rests on the idea that there is a high degree of consistency in state behavior both over time and across related issues. I do not mean to exaggerate...