Tag Archives: pundit’s fallacy

General discussion

The Pundit’s Fallacy

4 Comments

Internet denizen Matthew Yglesias coined the term “pundit’s fallacy” back in 2010. It goes like this:

The pundit’s fallacy is that belief that what a politician needs to do to improve his or her political standing is do what the pundit wants substantively. So progressive populists think that Barack Obama would have higher approval ratings if he acted more like Ed Schultz while establishmentarian centrists think his ratings would go up if he acted more like David Broder. The truth, of course, is that he really needs to hew more closely to my preferences politics doesn’t work this way.

For this to be a “fallacy” in any meaningful sense of the term, there has to be some reasoning failure. To me it seems the failure consists in a lack of self-awareness about one’s own perspective. Not, mind you, that you have a perspective. Rather, that you assert your perspective as the perspective without, and this is crucial, offering any argument for it.  If you put it this way there is nothing peculiar to pundits. Indeed, it’s just another form of petitio principii–the one where you assert controversial premises without evidence.
Naturally, the phrase, “without evidence” might raise some hackles. Let me specify. It’s to assert controversial things without the right kind or quality of evidence. So, for instance, democrats should have appealed to white working class voters by being less, er, judgmental, moralistic, or whatever. I’ve seen this a lot (one really bad one in today’s Chicago Tribune).
This kind of claim may be true (because any proposition can be true–well, almost any proposition). The problem is the kind of evidence offered for it. If it’s anecdotes about your father-in-law’s dislike of elites, then you’ll have to try again.  Claims about what motivates mass numbers (or important minorities in this case) of people to select one candidate over another require a special kind of evidence; you can’t ask everyone and you cannot easily interpret their selection (or non-selection) as a signal of anything in particular.