It’s true that a lot of the arguments that emerge from Atlantic Ideas occupy that Slateian niche of onanistic — and completely impracticable — cleverness, but this Conor Clarke idea seems especially misguided:
Polls are as integral to the American political tradition as sex scandals or earmarks. Yet it’s not clear that they serve any beneficial purpose.
Really? It seems to me one of the only reasons legislators are still pursuing a public option in the face of intense industry pressure is because polling shows it to be popular. But to engage a bit more with the argument as constructed, I’ll outline the points Clarke makes.
- Polling “uncomfortably expands the domain of democracy.” In essence, Clarke argues polls can act as a malign force to advance “tyranny of the majority.”
- “Many polls,” owing to the misinformed state of much of the American public or a difference between stated and revealed preference, don’t accurately reflect actual sentiment.
- Polls can create self-fulfilling prophecies.
With respect to uncomfortable expansions of democracy (the political equivalent, I suppose, of post-Thanksgiving Dinner stupor), Clarke’s argument is essentially that some institutions — like the Senate — could use more democracy, and on the other hand, some institutions — like the referendum process in California — could use less democracy. You’ll get no argument from me on either score. However, it’s not clear why this makes polling a bad thing. Indeed, it would seem that polling occupies the perfect middle ground. It’s results do not compel action or force the will of the majority upon the minority. In cases where polling shows wide margins of support — as with a public option in health care — politicians know that they have cover to support a particular initiative. By contrast, if an initiative has only small margins of support, politicians understand the controversial nature and will likely adapt legislative priorities accordingly.
Second, it’s true that polling on some topics is practically worthless. It’s also true that some ideas posed in public discourse are daft. Neither means we should abandon the practice of polling or coming up with smart ideas. Indeed, the poll Clarke cites showing most Americans don’t even know what cap-and-trade is (thus, how can their opinions on the matter be valid) evinces the need for polling: legislators understand that if we wish to avert catastrophic climate change, the public will need to be educated. As for the difference between “stated” and “reveled” preferences, this doesn’t seem to be a major problem outside the margins (that is, it happens in small numbers on totally uncontroversial subjects, and happens in larger margins on much more controversial subjects). More importantly, this argument shifts into shaky epistemic questions that if you follow only a bit further, suggest we should do away with asking people questions because we can’t be sure of the veracity of their response.
Finally, it’s true that polls can create momentum unto themselves. So can journalists who set “expectations.” Just think about the Vice Presidential debate between Sarah Palin and Joe Biden. Sarah Palin, at that point, would have surpassed expectations had she managed only not to drool on herself. Because of this, she “held her own,” and thus somehow “won” despite the fact that by almost any measure, public sentiment strongly favored Biden winning the debate. I realize this isn’t an argument in defense of polling per se, but it demonstrates that instruments should not be abolished because they have flaws.
This went on a bit longer than I expected, but I really find these exercises in cleverness annoying.