Dog Whistle Politics and the McWhorter Fallacy

Pithlord watches ex-Berkeley professor John McWhorter commit the shameful act of laying down covering fire for Republican Senator from Virginia George "Let's give a welcome to macaque, here--welcome to America and the real world of Virginia" Allen. He focuses on the fact that McWhorter's argument is self-denying--that it is valid only if it is invalid; and that if it is valid, then it is invalid.

Pithlord writes:

Pith and Substance: The McWhorter Fallacy: Those who pay attention to the scandals of the American political blogosphere doubtless know that George Allen -- a more-than-expectedly unappealing potential Republican nominee for President in 2008 -- called one of his opponents' staffers a "maccaca" and bid him "Welcome to America." The staffer in question is of South Asian descent and the hivemind quickly discovered that "maccaca" was a slur used by French colonists against North Africans. Allen's mother was a French colonist in Algeria, and it seems plausible (particulalry in light of the absence of competing explanations) that he consciously or sub-consciously had this meaning in his head.

The Allen campaign can't admit any of this, of course, because it would suggest that Senator Allen is part French.

All of that is by way of background for those of readers and spambots who do not obssessively follow American politics.

The point I wanted to make came from John McWhorter's defence of Allen, related by publius here:

Imagine for a moment that Allen actually knew that a "macaque" is a kind of monkey, or that in French the term is sometimes used as an insult for North Africans (Allen denied having known about either). Who, then, believes that Allen would use the slur against an opposition campaigner aiming a camera straight at him?

The logic, as I understand it is as follows:

• If P said X, then P would be criticized for violating some public norm.
• P therefore could not have said X since it would result in public criticism, which would reduce P's electability.
• Therefore, public criticism [of P] is unjustified.

The second premise depends on the idea that politicians, even in moments of stress, always act rationally. It also assumes that everything blameworthy will, in fact, be blamed. Since these are unsuound assumptions, the argument is a stupid one. This could actually be used as a warning about rational choice models, or just about the fact that in the public discourse today, someone will defend anything.

But more interesting is that the argument undermines itself to the extent that it is believed.

McWhorter's Fallacy logically compels the conclusion that no politicians will ever say antyhing blameworthy. If people accept McWhorter's Fallacy, then, of course, they won't criticize politicians for what they say. But if that is true, then rational politicians, fearing no criticism and seeking the psychic satsifaction of insulting people will start saying blameworthy things again.

In other words, racial slurs by politicians become frequency-dependent strategies. If they are uncommon enough, they will not be criticized (on the assumption that their use is too irrational to occur). But then they will become a cheap way of letting loose. But if too many politicians start employing this method of stress relief, the McWhorter reasoning loses its hold and politicians start being criticized for being boors and bigots again.

It would then follow that there is a slur E[volutionary ]S[table ]S[trategy].

Anyone thinking of modelling this should give me credit. I realize it might be a bit embarrassing to suggest your idea came from an anonymous fellow on the Internet named "Pithlord", so I suggest P. Lord of the Institute for Pith and Substance. You're welcome.

John McWhorter is Senior Fellow at the Manhattan Institute and until 2003 was Associate Professor of Linguistics at UC Berkeley. He is the author of Winning the Race: Beyond the Crisis in Black America and The Power of Babel: A Natural History of Language.