My First Post
I finally got around to starting a blog today for two unrelated reasons. The first was coming across a speech by Michael Crichton deploring the victory of politics over science in the form of "consensus science." One of his examples ... "According to Sagan and his coworkers, even a limited 5,000 megaton nuclear exchange would cause a global temperature drop of more than 35 degrees Centigrade, and this change would last for three months. The greatest volcanic eruptions that we know of changed world temperatures somewhere between .5 and 2 degrees Centigrade. Ice ages changed global temperatures by 10 degrees. Here we have an estimated change three times greater than any ice age. One might expect it to be the subject of some dispute. But Sagan and his coworkers were prepared, for nuclear winter was from the outset the subject of a well-orchestrated media campaign. The first announcement of nuclear winter appeared in an article by Sagan in the Sunday supplement, Parade. The very next day, a highly-publicized, high-profile conference on the long-term consequences of nuclear war was held in Washington, chaired by Carl Sagan and Paul Ehrlich, the most famous and media-savvy scientists of their generation. Sagan appeared on the Johnny Carson show 40 times. Ehrlich was on 25 times. Following the conference, there were press conferences, meetings with congressmen, and so on. The formal papers in Science came months later. This is not the way science is done, it is the way products are sold." I formed my own opinion on that particular issue many years ago, after reading a scientific article by the authors of one of the articles that fed into the nuclear winter calculations. It conceded that their earlier article contained, as critics had pointed out, a serious error--serious enough so that correcting it reduced the predicted duration of global winter from years to weeks. But they explained that they had now discovered another error in the opposite direction--and correcting it brought the duration back to years. My guess is that they were telling the truth about their analysis. They may even have been correct in their conclusion. But the degree of uncertainty implied by that article was strikingly inconsistent with the confidence with which the nuclear winter conclusion was being trumpeted--largely by people who wanted other people to believe it because they thought that belief would reduce the risk of nuclear war. [All of this was long ago and I am working from memory, so if you want to check my report of the article I am afraid you will have to locate it for yourself.] My second reason to start a blog was a recent experience of the virtues of Chicago style workshops. At most universities, if you are invited to give a paper, you get something like an hour and a half, most of which is spent reading your paper aloud--on the bizarre theory that the professors and graduate students who make up your audience are unable to read it for themselves. Any remaining time can be spent answering questions from people most of whom have just heard your ideas for the first time. Many years ago, someone at the University of Chicago came up with a better model. At a Chicago style workshop, everyone is expected to read the paper in advance. The author gets fifteen minutes to half an hour--the time increases the farther you are from Chicago--to say whatever he wants about his paper. After that it is open season, with members of the audience pointing out errors, raising questions, suggesting ways in which the paper might be expanded or improved. Great fun for all, and the nearest thing I have observed to real time thinking by a group brain. Recently I observed a striking demonstration of the superiority of the Chicago workshop. I was attending one--not at Chicago. Reading the paper, I noticed a mistake in the brief theoretical section that was supposed to motivate the conjecture tested in the statistical analysis that was the core of the paper. The authors had left a term out of an equation--had assumed, in effect, that when A sues B and wins, B pays damages but A does not receive them. Correcting that mistake, and changing nothing else, reversed the conclusion--implied that the conjecture they were testing could not be true. To restore the conclusion it was necessary to make additional changes, dropping one or another of the simplifying assumptions that had gone into the model. The mistake was sufficiently obvious so that any competent graduate student in the field who actually worked his way through the paper should have spotted it. Before the talk started, I mentioned the existence of the mistake to one of the authors. Shortly thereafter, another professor came in--and also described the mistake to him. A few minutes after the talk started a third member of the audience, a very distinguished senior member of our profession, raised his hand and pointed out that the model was clearly wrong. What struck me was not the existence of a careless error in an unpublished draft of a scholarly article--we all make mistakes. What struck me was that this was not the first university the paper had been presented at. Presumably, none of the others followed the Chicago model. If your first contact with a paper is an oral presentation, you are unlikely to check the details of the equations. Apparently nobody at those talks had. [To be fair, I should add that the author told me that the error had been mentioned to him, although not explained, on one earlier occasion--but not by someone whose first exposure to the work was at a university talk.]