Working the refs pays off
. Just ask Mike Krzyzewski...or the GOP...
One of conservatism's most effective ploys of the last fifty years has been the charge of liberal media bias--a charge that even some conservatives have admitted is really just a matter of working the refs.
It has seemed to me for some time that Politifact, facing the generally greater mendacity of the right, has bent over backwards trying to avoid the charge of liberal bias. This, of course, produces the opposite bias--in order to appear even-handed to the less-honest of two competing groups, the allegedly neutral judge holds the more honest group to higher standards.
That's what's happened with the Reid-Romney dust-up about taxes. Politifact has asserted that Reid is lying about the claim that Romney did not pay taxes for ten years--it gives his claim a "pants on fire" rating. And WaPo FactChecker gives Reid four Pinocchios. Both are supposed to indicate that the claim in question is a lie. However, it is in no way clear that either rating is warranted, given that we do not know whether Reid's claim is true or false
. A lie, on the strictest and most common definition, is something like a falsehood knowingly misrepresented as being a truth
. Since neither Politifact nor FactChecker know whether Ried's claim is true, neither is in a position to call his claim a lie, so neither is in a position to give the rating they have given.
Now, they might claim that Reid's claim is unjustified and/or irresponsible, and that's not crazy. However, Reid is actually in a better epistemic position than they are--he at least knows who the source of the rumor is, and Politifact and FactChecker do not. As Nyhan basically notes, the rating scale of both organizations is set up to represent something more like lies vs. non-lies rather than justified vs. unjustified / responsible vs. irresponsible assertions. (A wildly irresponsible claim can be the moral equivalent of a lie--just as bad as a lie, that is. Most of the lies the Bush administration told in the spin-up to the Iraq war, for example, were the moral equivalent of lies, but (in a textbook display of consciousness of guilt) they normally were very careful not to lie technically
and on the narrowest definition.) At any rate, and more importantly, neither Politifact nor FactChecker know exactly how good Reid's evidence is
. Thus, even if their rating scales were well-suited to rating the justification/responsibility of assertions, they're not in a position to make a clear judgment in this case.
Details matter in such cases, incidentally, and it does matter--in some way that's hard for me to articulate off the top of my head--that Romney could easily disprove Reid's assertion, but chooses not to. If, for example, the claim could only be disproved with difficulty, we might have a different case on our hands.
All this does put Politifact and FactChecker in a rather tight spot epistemically. If they claim that Reid's assertion is a lie because he doesn't know whether the proposition in question (Romney did not pay taxes for ten years
) is true, then they seem to have to rate their own assertions as lies, since they don't know whether or not their own claims are true
. After all, if no one knows whether the proposition in question is true, then no one knows whether or not Reid is telling the truth when he says that it's true. And if Reid is lying for saying that the proposition is true without knowing it, then Politifact and FactChecker are lying for saying that this other proposition (Reid is lying) is true without knowing that it is. But clearly they are not lying, they're just making unjustified claims...as Reid may (or may not) be doing...
OTOH, one could assert that Reid has the burden of proof, and so the positions are asymmetrical. However this seems to be not quite right. Politifact and FactChecker can plausibly (though not unimpeachably) claim that Reid's claim is irresponsible. But that's not what they have, in effect, said. What they've said is that he is lying
. And that they are not in a position to do.
Reid has told us, in general terms, what his evidence is for the first proposition. By committing himself to that proposition, he is committing himself to the reliability of his source. At least he is being up front about the general character of his evidence. (He's also, again, making a claim that Romney could easily refute, and refute by doing something he ought to do anyway...)
None of this is terribly complicated, and it all reinforces the worked-ref hypothesis: Politifact and FactChecker have both reached fairly far beyond the known facts to tag Reid with a liar
rating. Their rating is, I think, irresponsible, and my guess is that it hints at an unfair standard.
There's a rather easy way to test this hypothesis, however--they hypothesis that one or both organizations are holding one side to a higher standard than the other. We could ask, for example, whether either Politifact or FactChecker gave a pants-on-fire/4 Pinocchio rating to birther claims about Obama's birth before Obama had produced evidence about his birth certificate. The cases seem similar at a glance. If they did
give such ratings to birthers before
evidence was produced, then I'll happily apologize to both organizations. I haven't check this out, so I don't know the answer to my question.