Noah Millman wrote:

In retrospect, what suffered the most lasting damage from the terrorist attacks of ten years ago was my belief in my own rationality. I believed that I was thinking things through seriously, and coming to difficult but true conclusions about what had happened, what would happen, what must happen. Here is part of what I wrote, to friends and family, several days later:

Our President has made it clear: we are at war. I do not anticipate that this will be a short or an easy war. Our enemy has operations in dozens of countries, including this one. He is supported, out of enthusiasm or fear, by many governments among our purported friends as well as among our enemies. He has shown his cunning, his ruthlessness, and most of all his patience, in his successful plot to kill thousands of innocents and bring down the symbols of our civilization. And in striking at him, as we must, we will bring down others who will in turn seek their own vengeance upon us.

There is not a single factual assertion in that paragraph that I had any reason to believe I could substantiate. I did not know anything about the enemy. I had no idea whether or not there were “operations” in dozens of countries – I don’t even know what I meant by “operations.” I know what I was referring to with the business about being “supported” by friends and enemies, but “support” is a deliberately fuzzy word; I wouldn’t have used it if I was trying to make a concrete assertion with clear implications. The purpose of that assertion, like everything else, was to build up my first assertion. We were at war. And it wouldn’t be short or easy. Because that conclusion, though grim, was one that imparted meaning to the murder of 3,000 people. I thought I was being serious – examining the facts, calculating the likely negative consequences of necessary action, preparing myself for the unfortunate necessities of life. But I wasn’t doing anything of the kind. I was engaged in a search for meaning in which reason was purely instrumental.

Link (which includes additional good retrospectives) thanks to Ampersand.

This article may have more political content than is suitable for LW-- if you'd rather discuss it elsewhere, I've linked it at my blog. I've posted about it here because it's an excellent example of updating and of recognizing motivated cognition even if well after the fact.