But much of the other conventional wisdom is dubious. Clinton is said to appeal to more conservative, less-educated working-class voters, especially Latinos, and Obama to educated, well-to-do liberals—the Dunkin’ Donuts vs. Starbucks trope. Yet in fact, income hasn’t really correlated strongly and consistently with which Democrat one prefers. Nor does simply being Latino, or liberal or conservative.
In Virginia, for instance, the difference between college graduates and non-college-educated voters was vanishingly small, and in Maryland, the less educated were more pro-Obama. And in the populous northeastern states that Clinton won, such as New York, she won majorities of the college-educated.
The Clinton supporter and CNN pundit Paul Begala said on TV last Tuesday night that Obama had “lost every income group below $50,000 a year” on Super-Duper Tuesday. Right? Wrong: In fact, he won among those people in Georgia, Illinois, Missouri, Utah, and elsewhere.
The other big piece of not-entirely-true conventional wisdom is that Clinton is everywhere beloved and/or Obama rejected by Latinos. In fact, Clinton’s share of the Latino vote has ranged from a high of 73 percent in New York to just 43 percent next door in Connecticut. Go figure. In fact, Obama won the Latino vote in Connecticut, and in Arizona he did better among them than among whites.
Among self-described liberals and conservatives, the results are even more all over the map. Last week, for instance, Obama’s greatest strength in Virginia was among “conservatives,” but among “moderates” in Maryland, where Clinton did best among “liberals.” And in New York, Obama actually beat Clinton at both ends of the ideological spectrum, among the quarter of Democrats who call themselves either “very liberal” or “somewhat conservative.” (As a supporter of Obama who, depending on the issue, wanders from very liberal to somewhat conservative, I don’t find this crazy.)
It’s pretty much the same mixed-up situation with religious observance. In California, Clinton won among people who go to church more than once a week, while in New Jersey Obama won among those most frequent churchgoers.
And speaking of that state just across the Hudson: You are so weird, New Jersey. In Jersey, Obama didn’t win among college graduates or the rich. And if you were to look at the state’s voter numbers in a blind test, you’d swear they were from the Deep South: Only in places where people have southern accents—and New Jersey—did Obama get less than 30 percent of white women’s votes, or Clinton an absolute majority of the votes of white men. I’m hereby adjusting my stereotype.
Last week, however, the political weather changed. In Virginia and Maryland, two of the strongest fixed demographic patterns started to break, like frozen rivers thawing as spring approached. Maybe this, finally, is what momentum looks like. In both states, Obama won convincingly among white men—by 58 to 40 percent in Virginia, the capital of the Confederacy. And, apparently for the first time, Obama won among old voters, too. Only a single group, white women, stuck with Clinton.
Thus, it seems, demographics are destiny, but in this election really only in three particular cases: African-Americans drawn irresistibly to one of their own, white women of a certain age drawn irresistibly to one of their own, and young whites drawn irresistibly to youthfulness and the thrill of the new. Duh.
We have a bit of a double standard about this, of course. On the one hand, who can fail to be heartened by the spectacle of majorities of white southern guys voting for an African-American? We coo approvingly when certain people behave against type. I was even pleased (and slightly shocked) that 43 percent of black women in New York voted for Clinton.
Yet on the other hand, we selectively suspend our disapproval of the historically downtrodden as they behave according to type. We’re sympathetic to all the white women voting for Clinton, and all the black people voting for Obama. But we also hope—don’t we?—that this is another awkward, inevitable moment in our long, long march toward a society in which nonwhite and non-male presidential candidates aren’t thrilling by virtue of their race or gender.
The general election, of course, will be a whole new, fascinating phase of this sociology experiment—though the prospect of McCain vs. Obama is vastly more interesting than McCain vs. Clinton. I think we know pretty much exactly how the latter would set up: The main mystery would be whether the gender gap might break 20 percent. Between McCain and Obama, however, so much would be fresh territory: young versus old, two viable voting options for the (white, male) Bill Clinton Democrats, a possible bifurcation of independents and peeling away of moderate Republicans, and so on. As Hillary Clinton said a couple of months ago, now the fun part starts.