WEBLOG
Previous Month | RSS/XML | Current | Next Month
Monday, May 27, 2002 ( 9:10 PM ) (Permalink)
How Not to Connect the Dots
Recent revelations of some F.B.I. memos and other reports prior to the terrorist attacks of 9/11 have provoked both questions and accusations. Should the government have been able to "connect the dots" and prevent the attacks? If so, why didn't it?
While some type of investigation into the failure to foresee or prevent the attacks seems warranted, there is a danger in after-the-fact second-guessing: hindsight bias. This is a cognitive bias that is summed up in the familiar saying that hindsight is 20/20.
Hindsight bias consists in the tendency to overestimate the predictability of past events based on current knowledge of the outcome. When we know how things turned out, it is easy to fool ourselves into thinking that we could have foreseen those consequences, or that others should have.
Of course, it's still possible that an investigation will reveal that sufficient information was available prior to 9/11 to have made it possible to prevent the hijackings. However, given the human propensity for hindsight bias, the investigators need to be cautious about connecting those dots, knowing as they do how the finished picture will look. As the psychologist Piattelli-Palmarini says:
"[W]e should not overlook the disastrous results of this cognitive illusion. One doesn't have to take up the case of the cynical and merciless way in which politics must always find a scapegoat: It can be scientifically demonstrated that this predictability in hindsight strikes at even the best balanced and well-intentioned people. An unfortunate victim can be fired, his or her career blocked, or worse under the pernicious effect of this illusion. Even sadder is the fact that it won't be particularly difficult to convince the victim that in fact he or she should have been able to predict what happened on the basis of the data he or she had available." (P. 125)
Sources:
- Baruch Fischhoff, "For those condemned to study the past: Heuristics and biases in hindsight", in Daniel Kahneman, Paul Slovic, and Amos Tversky, editors, Judgment Under Uncertainty: Heuristics and Biases (Cambridge University Press, 1982), pp. 335-351.
- Massimo Piattelli-Palmarini, Inevitable Illusions: How Mistakes of Reason Rule Our Minds (John Wiley & Sons, 1994), pp. 123-125.
Friday, May 24, 2002 ( 1:23 AM ) (Permalink)
Fallacious Analogies
Victor Davis Hanson has a recent article complaining about questionable analogies used in pro-Palestinian rhetoric, and citing evidence for their weakness.
Update (10/13/2012): This article is apparently no longer available at its previous address.
Saturday, May 18, 2002 ( 2:50 AM ) (Permalink)
Fallacy Watch
In a recent column, law professor Alan Dershowitz appears to be arguing against a petition which calls for universities to divest themselves from holdings in companies which do business with Israel. While Dershowitz may be correct that the petition is part of a "foolish and immoral campaign for divestiture", his column provides almost no evidence for this conclusion. He claims that "there is no intellectually or morally defensible case" for divestiture, but neither does he present such a case against it. Instead, he uses most of the article for a series of fallacious attacks on those who support the petition.
Two-thirds of the article is a personal attack on Noam Chomsky, whom Dershowitz calls "the inspiration" behind the petition campaign. While everything that he says about Chomsky may well be true, it is a classic Ad Hominem fallacy to attempt to discredit a position by attacking its advocate. There is an old legal cliché that Dershowitz should know: when you don't have a case, attack the other side's lawyer.
In the next to last paragraph, Dershowitz widens his attack to everyone who supports the petition, thus switching into the similar fallacy of Guilt by Association. Who would want to join "a motley assortment of knee-jerk anti-Zionists, rabid Anti-Americans, radical leftists…, people with little knowledge of the history of the Arab-Israeli dispute…."?
In the last paragraph, Dershowitz leaves off attacking the petition's supporters, and finally gives some reason for opposing the petition: "Universities invest in a wide array of companies that have operations in countries that systematically violate the human rights of millions of people. … Yet this petition focused only on the Jewish State, to the exclusion of all others, including those which, by any reasonable standard, are among the worst violators of human rights." Perhaps this is Dershowitz's legal training leading him astray, for the appeal to precedent is often a valid legal argument, but a fallacious moral one. (Unfortunately, this fallacy does not yet have an entry in the F-Files, but it is similar to the Bandwagon fallacy.) However, what judge would accept the argument that one defendent should be let off just because other, even worse ones, roam free? If this were a good argument, it would constitute a "Get Out Of Jail Free Card".
There must be better arguments to be made against the petition than this, as they surely couldn't get much worse.
Thursday, May 16, 2002 ( 1:28 AM ) (Permalink)
What's New
The F-Files Bookshelf now has a third volume, namely, the below-mentioned Historians' Fallacies.
Tuesday, May 14, 2002 ( 4:03 AM ) (Permalink)
Q & A
Ken Spell writes in to ask:
On your website, you seemed to have organized each fallacy according to type. I've found a couple that don't have types, such as the one called "Complex Question". May I ask, under what category this should be typed?
You sure may, Ken, this is a good question! "Complex Question" is a difficult fallacy to categorize because it involves asking a question, rather than making an argument. For this reason, I have placed it in a category of its own, with the hope of adding further question-asking fallacies in the future. David Hackett Fischer, in his book Historians' Fallacies, devotes an entire chapter to "Fallacies of Question-Framing", of which there are ten, including "Many Questions".
Another approach to categorizing "Complex Question" is taken by T. Edward Damer who, in his textbook Attacking Faulty Reasoning, considers this fallacy a type of Begging the Question. If I had to put Complex Question into one of the non-question-asking categories, I would follow Damer. However, it seems to me that question-asking fallacies have special characteristics and problems, and should be treated as falling into a category of their own.
Sunday, May 12, 2002 ( 2:27 AM ) (Permalink)
What's New?
The David Manning case study has a new link to an old USA Today article on "blurbmeisters" and "quote whores" which came out last year shortly after the Manning affair became public. Coincidentally, one of the critics mentioned in the article, Earl Dittman, had a quote in an ad for the movie High Crimes which appeared in Friday's New York Times: "FANTASTIC!"
Update (10/13/2012): The case study cited above is no longer online.
Friday, May 10, 2002 ( 2:43 AM ) (Permalink)
Fallacy Watch
Bryan Keefer has an article on Spinsanity criticizing the loaded term "Enron conservatism" which also includes a mention of some "false dichotomies" in recent political rhetoric:
Framing current policy debates as "a choice between progressive reform and Enron conservatism," [Robert] Borosage presented a number of false dichotomies designed to frame large social issues as "Enron conservatives" vs. liberal policy prescriptions. Asking his listeners to "put the case to the American people," Borosage suggested a number of slogans: "Invest in education or let Enron conservatives starve even the reforms they celebrated last year," "Launch a drive for energy independence…or let Enron conservatives push their Big Oil energy plan," "Make worker rights and environmental protection central to our trade accords or let Enron conservatives foster a global race to the bottom" and "Political reform to get big money out of politics…or let Enron conservatives continue a politics where private interests dominate our public life."
Tuesday, May 07, 2002 ( 1:21 AM ) (Permalink)
What's New?
The F-Files Bookshelf now has two books; the new one is a reference on logical fallacies.
Monday, May 06, 2002 ( 1:48 AM ) (Permalink)
What's New?
The Fallacy Files Bookshelf is a new feature of the F-Files, providing reviews and recommendations of books on logical fallacies. So far, the shelf has only one book—an introduction to fallacies—but I will be adding two more in the near future: a reference book and a book applying fallacies to a particular subject.