“A few months ago,” Mr. Brooks concluded, “Mr. Obama was riding his talents. â€¦ Now, Democrats are deeply worried their nominee will lose in November.”
Eh, not really. That logic fixates on all of the ammunition that Republicans have at their disposal against Mr. Obama. But it ignores the more basic question of whether voters, upon being exposed to the caricature, will actually buy into it. ~Steve Kornacki 
Hope really does spring eternal, doesn’t it? Consider John Judis’ latest article , in which he seems to confirm Brooks’ description of growing Democratic anxiety:
Meanwhile, Obama’s weaknesses as a general election candidate grow more apparent with each successive primary.
The main difference is that Judis wrote this after the Pennsylvania results, and Brooks wrote his column before them. They are otherwise making closely related arguments. Judis went on to say:
Indeed, if you look at Obama’s vote in Pennsylvania, you begin to see the outlines of the old George McGovern coalition that haunted the Democrats during the ’70s and ’80s, led by college students and minorities. In Pennsylvania, Obama did best in college towns (60 to 40 percent in Penn State’s Centre County) and in heavily black areas like Philadelphia.
Its ideology is very liberal. Whereas in the first primaries and caucuses, Obama benefited from being seen as middle-of-the-road or even conservative, he is now receiving his strongest support from voters who see themselves as “very liberal.” In Pennsylvania, he defeated Clinton among “very liberal” voters by 55 to 45 percent, but lost “somewhat conservative” voters by 53 to 47 percent and moderates by 60 to 40 percent. In Wisconsin and Virginia, by contrast, he had done best against Clinton among voters who saw themselves as moderate or somewhat conservative.
Obama even seems to be acquiring the religious profile of the old McGovern coalition. In the early primaries and caucuses, Obama did very well among the observant. In Maryland, he defeated Clinton among those who attended religious services weekly by 61 to 31 percent. By contrast, in Pennsylvania, he lost to Clinton among these voters by 58 to 42 percent and did best among voters who never attend religious services, winning them by 56 to 44 percent. There is nothing wrong with winning over voters who are very liberal and who never attend religious services; but if they begin to become Obama’s most fervent base of support, he will have trouble (to say the least) in November.
Those who are sympathetic to the idea that Obama represents a break with the past may be both heartened and terrified by this. On the one hand, this evidence could be used to rebut my claim that Obama seems to be an old-fashioned liberal with respect to much of domestic policy. One could argue that he has to appear to be more to the left than he “really” is, because that is where his primary coalition is, and he will eventually, to use Kaus’ word, pivot to the center as all nominees supposedly have to do in the summer, and more importantly would not govern in the manner of a progressive. Besides, you might add, his health care plan puts him to the right of Clinton on that issue. You could push this more and say that he is doing exactly what I said he would have to do, shoring up his support on the left, because this had been weak earlier in the primaries when he was talking up unity, bipartisanship and reconciliation, and that he will later turn out to be the coalition-expanding, independent-attracting dynamo that many hoped/feared he would be. Then again, the demographic profile of his coalition for most of the primaries has actually remained pretty constant, and to the extent that it has changed it has only grown more predictably liberal over time, which suggests that even if I am wrong about the content of Obama’s liberalism his coalition of supporters may end up identifying him in the public mind with their politics and thus make him less electable.
Whether you buy the “Greater New England” thesis that explains Obama’s success in such places as Wisconsin and Iowa, he seems to prevail in primaries in those almost entirely white states where there has been a particularly strong progresstive tradition in the politics of those states, so the electoral limitations of a candidate identified with this coalition seem clear. Technically, the composition of his coalition and the content of Obama’s liberalism are distinct matters, but if his primary coalition is liberal or very liberal this will tend to draw attention to those areas where Obama is more liberal than Clinton and will cause people to neglect those instances when he is moving to her right (for instance, on Social Security to the endless aggravation of Paul Krugman). Of course, perception counts for a great deal, and if Obama becomes identified with the profile of his voters, as McGovern (who was personally more conservative than Obama) was, as a practical matter it won’t make any difference whether he is actually neoliberal or “centrist” on a number of things, because he will be perceived according to the views of those who support him rather than his own stated positions one way or another.
This is perhaps the same reason why so many on the left became convinced that George W. Bush was some great right-winger, since his primary coalition in the fight with McCain was made up of the most conservative members of the GOP and may also help explain why conservatives embraced him as “one of their own,” despite plentiful evidence to the contrary, out of resistance to McCain’s campaign. The images forged during primary battles do not disappear quickly (hence McCain remains in the fantasies of the media the reasonable, moderate Republican!), and the opposing party has every incentive to exaggerate the “extremism” of the nominee. I still think my earlier assessment makes sense based on what Obama has said about policy, but perhaps we on the right are mistaking episodes of necessary primary pandering to a voting coalition for Obama’s convictions and that Obama would govern as a liberal every bit as much as Bush has governed as a conservative, which is to say not much.
Whether or not Obama really is as far to the left as I think him to be, the profile of Obama primary voters will reinforce that image throughout the year. That is really why Pennsylvania will be so damaging to him. It isn’t just that certain demographic groups in Pennsylvania voted against him by large margins, though that isn’t a good sign for general election performance, but it is the impression the rest of the country gets from the results is that he appeals to a fairly small part of the country and to people whose ideological persuasion is not shared by all that many Americans. He runs the risk of being pigeonholed as Huckabee was, except that this is happening much later in the process. What happened in Pennsylvania will end up, in a somewhat circular fashion, ensuring that the pattern of voting in Pennsylvania will occur again in November, because the Pennsylvania results seem to confirm the limits of Obama’s appeal and can then be used and re-used frequently to characterise Obama as past Democratic nominees have been characterised. In Iowa Obama once boasted that every place he visits becomes Obama country, but after spending the better part of the last six weeks in Pennsylvania this obviously didn’t happen. The results came back almost as if they had been generated by a computer program that plugged in the demographics of the state and ran an equation based on Ohio’s voting. Everyone knew that Obama wasn’t going to win in Pennsylvania, but the sheer intractability of the demographic groups who didn’t vote for him is what has to worry those who want to see the Democrats win the White House in the general election.