Are you ever in a situation, such as when shopping, where it's about time to finally make a decision, like between buying the GyroXdisk 6800 or the MutantBlaster 970, and so you announce, "I choose the ...," but then you are immediately surprised at what came out of your mouth? As the salesman is ringing up your new MutantBlaster 970, you're thinking, "How the heck did that happen? I was sure I was going to say 'GyroXdisk 6800.'"
But pretty soon the relief of having made a decision, any decision, even if you can't tell why you made it, overcomes your dismay, and soon you're home blasting mutants without a care in the world. It's not like you sputtered, "I'll buy that big bag of mulch over in the Gardening Dept. instead of one of these electronic gizmos. And I live in an apartment so mulch makes no sense at all." No, you knew you wanted some kind of gizmo and so you were standing in the gizmo aisle talking to the gizmo salesman, so which precise gizmo you wound up with is moderately random. But it's not at all random that you bought a gizmo instead of a bag of mulch.
This happens to me all the time. I'm probably just the world's worst decision maker and this never happens to you, but maybe it gives a hint of something important in understanding history.
I started thinking about this again because Freeman Dyson, who is one of the last surviving physicists to have done war work during WWII (although John Archibald Wheeler, who worked on the Manhattan Project at Hanford is now 97), says that he's changed his mind and now no longer believes that the two atomic bombs dropped on Japan didn't end the war:
1. Members of the Supreme Council, which customarily met with the Emperor to take important decisions, learned of the nuclear bombing of Hiroshima on the morning of August 6, 1945. Although Foreign Minister Togo asked for a meeting, no meeting was held for three days.
2. A surviving diary records a conversation of Navy Minister Yonai, who was a member of the Supreme Council, with his deputy on August 8. The Hiroshima bombing is mentioned only incidentally. More attention is given to the fact that the rice ration in Tokyo is to be reduced by ten percent.
3. On the morning of August 9, Soviet troops invaded Manchuria. Six hours after hearing this news, the Supreme Council was in session. News of the Nagasaki bombing, which happened the same morning, only reached the Council after the session started.
4. The August 9 session of the Supreme Council resulted in the decision to surrender.
I don't see this timeline as undermining my long-held assumption that the 1-2-3 punch of Hiroshima, the Soviet entry into the war against Japan, and Nagasaki -- three cataclysms in four days -- was what finally broke the ferocious will of the Japanese militarists after endless arguing.
I don't know much about the Japanese decision-making process, but I can imagine what it would be like if I had been emperor on August 9, 1945. The Supreme War Council would turn to me and say they were hopelessly divided, "Your Divine Imperialness, we need you to break the deadlock. Shall we fight until we are all dead ... or surrender?"
They'd all look at me. I'd go, "Uh, uh, uh ... surrender!" And I'd immediately think, "Oh, crap, why did I say that? I meant 'Fight on!' Where'd that 'Surrender' come from? I'd better tell them I didn't mean it. ... But, that wouldn't look very divinely imperial -- now would it? -- to admit I just blurted out the most important decision in Japanese history at random. Maybe it's best just to keep my mouth shut and see what happens? In fact, yes, 'Surrender' is what I intended all along. It's becoming very clear to me now. Indeed, there was never really any doubt."
I'm sure it didn't actually happen that way (just that it would have happened that way if I was the Emperor of Japan).
Why did the Japanese surrender when they did? What weight did the two atomic bombs have in their decision? (Sure, surrender made all the sense in the world, but nothing the Japanese militarists had done since 1941 made much strategic sense.)
In contrast, nobody debates why Zanzibar surrendered 38 minutes into its 1896 war with the British Empire. Of course, they surrendered -- they were Zanzibar. What, they were going to do: beat the Royal Navy? It's just not very interesting because it's an obvious decision: Zanzibar was getting its butt kicked. Randomness didn't play much of a role.
My point is that the most interesting decisions in history, the ones that historians argue over endlessly about why they were made, are the virtually 50-50 tossups that could have gone either way. Those are the most interesting questions, but they are also the least likely to be fully explicated by historians.
The other decisions that are most argued over by historians are the ones that are the most over-determined.
That makes the Japanese decision of 1945 perhaps the ultimate in endless arguability. The Japanese leadership had many good reasons to surrender, each one perfectly adequate, and no reason to fight on other than to save their own necks from war crimes trials, but all the assassinations by the Army of non-insane statesmen before the war had bequeathed an atmosphere of militarist hysteria, so it was also a very close run thing -- overdetermined but also a toss-up.
This is relevant to the distinction between history and social science. I define social science (perhaps idiosyncratically) as fields where statistics are crucial.
Much of history is driven by unique personnel decisions that often didn't necessarily get a lot of reflection at the time -- e.g., the Bush dynasty is due to Ronald Reagan rejecting at the last moment at the 1980 GOP convention the popular idea of Gerald Ford for VP and picking George H.W. Bush. If something else had happened, George W. Bush never would have become President.
Now, Reagan probably put more thought into choosing a VP in 1980 than the average Iowa Caucus voter put into his or her choice, but the results are less random because thousands of semi-random individuals decisions were aggregated. So, the rise of the Bush Dynasty is the reserve of history, while elections are both history and political science.
Much the same is true about forecasting. The forecasts that interest us most are those where the chance of being wrong is highest. If you predict that NAM high school dropout rates will be higher than white and Asian drop outrates 30 years from now, nobody cares, even though it's clearly an important and highly plausible prediction, because it's boring and depressing. But tell me who is going to win the NCAA March Madness basketball tournament and I'm all ears, even though your chance of being wrong is probably at least 90%.