Over at the American Prospect, Bob Kuttner has an interview with Doris Kearns Goodwin that's well worth a read. About seven different times Kuttner asks a variation on the same question, which is essentially: "Don't you think Democrats can be as far to the left as I want them to be and still win elections? Give me some historical examples." He asks:
What I admire about Reagan and even about Bush, if I can use the word admire in the same sentence with Bush, is they were willing to say, "Polls be damned, I'm going to spend some political capital on what I believe, and I'm going to move public opinion in my direction and people will admire me for standing for what I believe"--the opposite of what many political scientists teach about elected officials seeking the median voter. Do you think the next Democratic president is capable of saying: "Polls be damned, I'm going to be a leader?"
(Kuttner, apparently, has just as little regard for mainstream political science as he has for mainstream economics.) Goodwin, good-natured and knowledgeable as always, responds with examples of instances in which, in fact, politicians have been able to take to the bully pulpit, shun the center, and succeed. The problem, though, is that they're all from what can best be described as extraordinary periods in American history--Lincoln during the Civil War, LBJ during the Sixties. It's not very convincing to simply cite two historical examples from the past 150 years and conclude that the median voter theorem is invalid, while ignoring all those who have fallen victim to it, from Hoover to Goldwater to McGovern (or used it to their advantage, like Bill Clinton).
Indeed, it's a bit strange that Kuttner expresses admiration for Bush's brand of politics, which has his approval rating parked solidly at 30 percent, while at the same time urging Democrats to move to the left. There's a tendency on the part of some liberals to maintain, on the one hand, that Bush is unpopular because his right-wing ideology turns off most Americans, while insisting at the same time that Democrats can eschew centrism without suffering the same fate. It seems like you have to choose one or the other: either the median voter theorem has more to recommend it than Kuttner wants to believe, or Bush's conservatism isn't to blame for his unpopularity.