So, if party leaders can get away with saying almost anything and have their loyal followers believe it -- even if it's totally made up and completely discredited -- why shouldn't they just do it all the time? Why shouldn't they all go Palin one better?
There are a couple of good reasons, and they relate to the question of whether Republicans "are bouncing back from their thorough smashings in 2006 and 2008," as Rod Dreher asks. Does the GOP have to get their act together in order to win? Or is it smarter to just embrace the crazy, bring the president's approval ratings down by convincing everyone that he wants to kill their grandmother, and faster than you can say Betsy McCaughey you're running things again and everything is hunky dory? Is there a disincentive to embracing the crazy?
Well, actually, yes, this are disincentives. At least two. I'll start with elections.
It is true that elections are mostly about the party in power; Obama is popular, Republicans will have no chance to defeat him in 2012, and will likely do poorly in 2010 as well. As for the out party, well, what they do doesn't really matter a lot. Much of what we see in campaigns is sideshow, and most of the rest matters around the margins. The bottom line -- what you can learn from election predictions that are blind to candidates and campaigns -- is that if Obama hadn't been nominated, then Clinton or Dodd or Richardson or Edwards-minus-scandal would have won in 2008, and the results would have been within a couple of percentage points of what Obama did. That can matter, if it's a 2000 or 1960 type election, but mostly it doesn't really change anything.
But Dennis Kucinich...that's another story. If the Republicans had nominated one of their fringe candidates, a Keyes or a Bauer, they would have done a lot worse than they did with McCain. We don't have a lot of data on this, because it happens rarely; the two cases that everyone always cites are Goldwater and McGovern, and really those are the only two cases (at least since we have public opinion data) to look at. But to the extent we can say anything, it sure seems like a bad idea to nominate someone who is near the ideological extremes.
The electoral problem with embracing the crazy is that it can lead to nominating extreme candidates. Suppose Senator Johnny Isakson wanted to run for president...would he be attacked as the author of the Death Panels? What do you think Iowa and New Hampshire voters would think if he tried to explain it? What about Georgia primary voters when Isakson is up for reelection? Could he lose a primary over it?
Indeed, the more that Republican leaders -- candidates, party officials, talk show hosts, and the rest -- convince their followers that the crazy is real, the harder it is for the party to nominate good, solid, non-crazy conservatives. And that may cause them to miss an electoral opportunity. It's most likely already cost them seats in the House and Senate.
Last bit: if the crazy was a good way to take down Obama, then it might be a good trade-off. Fortunately for those who despair of democracy, that doesn't seem to be the case -- telling lies about health care might well scare some Democrats in Congress into voting the other way, and it certainly might yield marginal changes in a bill that gets attacked, but basic fundamentals such as the economy are far more likely to affect Obama's approval ratings. Most of the effect of the crazy is letting followers know the why they should oppose something, not convincing them to do so. Legitimate points (normal, slimy, spin) would do the job just as well.