Arnold Kling summarizes quite succinctly just how much the stars aligned to put Obama in the lead:

Since my father's death, I have been trying to think about how to articulate his views on politics. Insider politics, as exemplified by Congressional earmarks defended here, struck him as normal and rational. Outsider politics, coming from libertarians or other radicals, struck him as irrational.

For the public at large, he took seriously the results of studies of voting behavior. Based on those, he predicted that Obama would not win the nomination, much less the Presidency. Historically, one's vote can be predicted quite well by one's parents' party affiliation, by one's ethnic group, and by one's economic class, in that order. I don't think my father took into account the Democratic Party's rules, which worked out this year to the detriment of Clinton by putting caucuses in states that she might have won as primaries, by negating a state with a large elderly population (Florida), and by negating a state with a large blue-collar population (Michigan).



Arnold thinks that as soon as the dust has settled, the party will unite behind the nominee.

I would guess that this will be less true of a Hillary Clinton nomination than of a Barack Obama nomination. The identity politics just doesn't resonate the same way for her base. For reasons that I can't quite articulate, I think that even the sixty year old women who strongly personally identify with her will be less angry and disappointed with a Clinton loss than blacks will at an Obama loss. If Kennedy had had the nomination snatched from him at the last minute because the party elders thought a Catholic couldn't win--or worse, because some Democratic voters were uncomfortable with a Catholic president--you'd have had a great deal of trouble motivating Irish-American turnout come November.

There are also the swingy Democrats who liked McCain in earlier Republican primaries. Those people are in the Obama camp right now. The war has changed the picture somewhat, of course, but Hillary will have a much harder time keeping Obama's supporters from defecting to the other side than he will hers. Obama also appeals to some of McCain's support among independents.

That said, I'm overall unconvinced by the large numbers of people who say that they'll vote for McCain if their candidate doesn't win. Most of them will fall back in line, and of the ones who don't, most of those will stay home. What problem there is comes down to turnout. If Barack Obama is the nominee, I expect that blacks will react the way the Irish-Americans, and to a lesser extent the Catholic community, did about Kennedy--i.e. if they had to stand in line on a bed of hot coals to vote for him, they'd happily do it. You'd barely need an urban turnout machine. Hillary motivates some women this way, I think, but not as many, and too dispersed to do the party much good.

Meanwhile, if Clinton is the nominee, the Republican turnout problem is largely taken care of--even people who are sick of Bush and don't much care for McCain will hustle to vote against her. No obstacle will be to great for those people to overcome; the polling place could be destroyed by a flash flood, and they'd just swim to the next town.

But it's not clear to me how big a problem this is for either party. The election still seems mostly like a referendum on Bush. Which spells Democratic victory in November.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.