Scott Sumner makes a point I alluded to, when he discusses some influential “texts”:
Perhaps you saw the movie Metropolitan. There is a scene where a young man is debating the merits of Jane Austen with a young woman at a New York cocktail party. Finally in exasperation she asks the guy “Which Jane Austen books have you actually read?” He replied “I don’t actually read novels, I read literary criticism.” I’m kind of like that asshole. I haven’t read a lot of the intellectual classics, but can spend 30 minutes telling you what is wrong with each of them. Yes, I’m quite aware of how unfair this is; I know that when you boil an argument down to its essentials the work can lose much of its persuasive power. But I did read Pride and Prejudice.
I haven’t read all of the books that I consider influential, but I have read extensively about them.
His point about history ties in with my own comment on the lesson of Taleb’s Fooled by Randomness:
I once read all the New York Times from 1928–38. History seems really different when it is actually happening. The people back then seemed just as smart as we are. Of course we have a bit more history to learn from, so we did a bit better with monetary policy this time around. But we still made many of the same mistakes, just to a lesser degree.The class distinctions back then seemed bigger — which surprised me. I knew that was the case for African-Americans, but I didn’t realize that class divisions among whites were also much greater, and that the upper class was so uninterested in the suffering of average farmers and workers. Or how much wealth was concentrated in New York City at that time.
I also developed a much greater respect for the stock, bond, and commodity markets’ ability to forecast the economy. They reacted to lots of things that seemed very important at the time, and that I think actually were very important, which are totally ignored by historians. A good example is the gold panic of early 1937 and the dollar panic of late 1937.