The reason it doesn't say is that Clinton's path to the nomination is pretty repulsive. She isn't going to win at the polls. Barack Obama has a lead of 144 pledged delegates. That may not sound like a lot in a 4,000-delegate race, but it is. Clinton's Ohio win reduced that total by only nine. She would need 15 more Ohios to pull even with Obama. She isn't going to do much to dent, let alone eliminate, his lead.

That means, as we all have grown tired of hearing, that she would need to win with superdelegates. But, with most superdelegates already committed, Clinton would need to capture the remaining ones by a margin of better than two to one. And superdelegates are going to be extremely reluctant to overturn an elected delegate lead the size of Obama's. The only way to lessen that reluctance would be to destroy Obama's general election viability, so that superdelegates had no choice but to hand the nomination to her. Hence her flurry of attacks, her oddly qualified response as to whether Obama is a Muslim ("not as far as I know"), her repeated suggestions that John McCain is more qualified.

Clinton's justification for this strategy is that she needs to toughen up Obama for the general election-if he can't handle her attacks, he'll never stand up to the vast right-wing conspiracy. Without her hazing, warns the Clinton memo, "Democrats may have a nominee who will be a lightening rod of controversy." So Clinton's offensive against the likely nominee is really an act of selflessness. And here I was thinking she was maniacally pursuing her slim thread of a chance, not caring--or possibly even hoping, with an eye toward 2012-that she would destroy Obama's chances of defeating McCain in the process. I feel ashamed for having suspected her motives.

Still, there are a few flaws in Clinton's trial-by-smear method. The first is that her attacks on Obama are not a fair proxy for what he'd endure in the general election, because attacks are harder to refute when they come from within one's own party. Indeed, Clinton is saying almost exactly the same things about Obama that McCain is: He's inexperienced, lacking in substance, unequipped to handle foreign policy. As The Washington Monthly's Christina Larson has pointed out, in recent weeks the nightly newscasts have consisted of Clinton attacking Obama, McCain attacking Obama, and then Obama trying to defend himself and still get out his own message. If Obama's the nominee, he won't have a high-profile Democrat validating McCain's message every day.

Second, Obama can't "test" Clinton the way she can test him. While she likes to claim that she beat the Republican attack machine, it's more accurate to say that she survived with heavy damage. Clinton is a wildly polarizing figure, with disapproval ratings at or near 50 percent. But, because she earned the intense loyalty of core Democratic partisans, Obama has to tread gingerly around her vulnerabilities. There is a big bundle of ethical issues from the 1990s that Obama has not raised because he can't associate himself with what partisan Democrats (but not Republicans or swing voters) regard as a pure GOP witch hunt.

What's more, Clinton has benefited from a favorable gender dynamic that won't exist in the fall. (In the Democratic primary, female voters have outnumbered males by nearly three to two.) Clinton's claim to being a tough, tested potential commander-in-chief has gone almost unchallenged. Obama could reply that being First Lady doesn't qualify you to serve as commander-in-chief, but he won't quite say that, because feminists are an important chunk of the Democratic electorate. John McCain wouldn't be so reluctant.

Third, negative campaigning is a negative-sum activity. Both the attacker and the attackee tend to see their popularity drop. Usually, the victim's popularity drops farther than the perpetrator's, which is why negative campaigning works. But it doesn't work so well in primaries, where the winner has to go on to another election.

Clinton's path to the nomination, then, involves the following steps: kneecap an eloquent, inspiring, reform-minded young leader who happens to be the first serious African American presidential candidate (meanwhile cementing her own reputation for Nixonian ruthlessness) and then win a contested convention by persuading party elites to override the results at the polls. The plan may also involve trying to seat the Michigan and Florida delegations, after having explicitly agreed that the results would not count toward delegate totals. Oh, and her campaign has periodically hinted that some of Obama's elected delegates might break off and support her. I don't think she'd be in a position to defeat Hitler's dog in November, let alone a popular war hero.

Some Clinton supporters, like my friend (and historian) David Greenberg, have been assuring us that lengthy primary fights go on all the time and that the winner doesn't necessarily suffer a mortal wound in the process. But Clinton's kamikaze mission is likely to be unusually damaging. Not only is the opportunity cost--to wrap up the nomination, and spend John McCain into the ground for four months--uniquely high, but the venue could not be less convenient. Pennsylvania is a swing state that Democrats will almost certainly need to win in November, and Clinton will spend seven weeks and millions of dollars there making the case that Obama is unfit to set foot in the White House. You couldn't create a more damaging scenario if you tried.

Imagine in 2000, or 2004, that George W. Bush faced a primary fight that came down to Florida (his November must-win state). Imagine his opponent decided to spend seven weeks pounding home the theme that Bush had a dangerous plan to privatize Social Security. Would this have improved Bush's chances of defeating the Democrats? Would his party have stood for it?

Jonathan Chait is a senior editor at The New Republic.