That's actually a word, but it's not particularly related to what I will say here.
Last week George Will wrote a column sufficiently full of outright falsehoods–non facts as it were–such that people all over were demanding retractions, corrections, etc. Rightly so, as the falsehoods were fairly egregious (it wasn't just forgetting who argued that philosophy is preparation for dying). He completely misrepresented research on global warming in order to claim that it wasn't happening. Strangely enough, the Washington Post has issued nothing by way of a retraction, and they have claimed that the column made it through their multi-layered editing and fact-checking process. So much the worse for them. Click here for a discussion of the factual errors, and here for our original post.
There is something slightly odd about the whole matter, however. He got some facts wrong. So did Bill Kristol (formerly of the New York Times now–get this–of the Washington Post). People screamed about that as the height of sloppiness. Which it is. But op-ed columnists do not really deal primarily in facts anyway. They deal primarily in arguments–the argumentative ones at least (which is mostly the conservative ones by the way–don't know why that is, it just is).
Arguments are one part fact, one part inference. As a matter of fact, they're mostly inference. Getting facts wrong is bad, but I think it is a comparably less egregious problem than repeatedly authoring crappy, to use a technical term, arguments. For the sake of clarity, a crappy argument is one whose conclusion likely wouldn't follow even if they premises were true. Seems to me that if one repeatedly makes these sorts of arguments, then one's getting the occasional fact wrong is a comparatively minor problem–easily correctable by a (competent) fact checker.
Correcting crappy arguments is rather more difficult, as many lay people don't seem to have a good idea what a crappy argument is. Many lay people think that pointing out an argument's crappiness, in fact, is a kind of crappy argument. I think someone–maybe me, maybe our Australian friends–suggested a name for this–the fallacy fallacy fallacy (the fallacy of thinking pointing out fallacies is a fallacy of some sort). It takes some amount of training to point out sophistries. But it's at least as important if not more than the pointing out errors of fact. So I'd like to propose that the Post, in addition to hiring a new fact-checker (not one paid by George Will–seriously, he's got like two of them), hire a kind of sophistry checker.
That way we wouldn't be subjected to things like this (from Will's column yesterday):
Although liberals give lip service to "diversity," they often treat federalism as an annoying impediment to their drive for uniformity. Feingold, who is proud that Wisconsin is one of only four states that clearly require special elections of replacement senators in all circumstances, wants to impose Wisconsin's preference on the other 46. Yes, he acknowledges, they could each choose to pass laws like Wisconsin's, but doing this "state by state would be a long and difficult process." Pluralism is so tediously time-consuming.
The sophistry detector might say the following: Mr.Will, you're going to have be a lot more specific here. In the first place, who are "liberals"? Second, you're equivocating on "diversity." "Liberals"–whoever the hell they are–can consistently be for one kind of diversity but not another. Aside from this, "impose" seems to have a different meaning for you. Feingold seems to be advocating a change in electoral procedure by democratic means–amending the Constitution. Such an activity requires signifcant electoral participation and agreement. One can hardly call that "imposing."
These, I think, are serious and egregious problems with just this one paragraph, yet none of them are factual problems in any direct sense. They do, however, make the argument here rather sucky.