Saturday, October 15, 2011

Just the Facts Redux

Prateek was skeptical about this post. (But Prateek, aren't you in India? Why weren't you sceptical about it? Is American spellling spreading there?)

Here are Prateek's remarks:

"Are you saying that facts admit of some cause-effect relation by themselves?"

"Cause" and "effect" ideally shouldn't be employed in history at all, because they create confusion. Historical "causes" are understood situations, and historical "effects" are intelligible responses to those situations.

 "Between events A and B, there could be a million reasons for A leading to B. One might cite a plausible reason for A leading to B. But plausibility has little to do with what actually happened."

There could be, Prateek. It is the job of the historian to examine the evidence until she knows what did lead from A to B! If she can't determine that, she doesn't yet know the facts, because, without seeing what led from A to B, she doesn't really know that A and B really happened at all. "The facts" are only facts once they all make sense in a coherent narrative... and that narrative explains the facts, and how one of those facts led to the next.

"Plus, there is the human error, that in trying to understand or rationalize certain events, we may use confirmation bias and narrative fallacy to force our preferred explanation into them."

Of course historians are human beings and may be mistaken. No one would claim otherwise. Critical history is the practice of minimizing the possibility of such errors, but of course they never disappear. But by citing "narrative fallacy" it looks like you are cribbing from Taleb. There is a man who has no clue what history is!


  1. "But by citing "narrative fallacy" it looks like you are cribbing from Taleb."

    Argh! I have been outed.

    By the way, after reading Taleb's Black Swan, I went a little crazy for a few weeks. Every time somebody suggested, "We should implement X policy to get Y result", I responded, "But how do you know that will happen? How do you know something else won't happen? How do you know something worse won't happen? If you don't know what will happen, why should you do anything?"

    And always the same response to me: "This is pure sophistry! You might as well worry about crossing the street." I realized it was, and that I should not have taken the idea to the absurd conclusion.

    Now, into this mix, consider that I was reading a bit of Rothbard as well, and when you combine all their ideas about a priori, confirmation bias, and narrative fallacy, the result is intellectual chaos. "You can never use history to determine the future", I kept saying to myself, until I realized how futile and pointless it was.

    The real world is about casual observation and generalization, not worrying about minimizing the margin of error to zero. Still Taleb's ideas are still well impressed in my mind.

  2. Prateek, I like a lot of what Talen says, just not on history.


The NP Turkey

We have a real problem this Thanksgiving: "With a big turkey, you start running into some big problems. It takes longer to thaw ...