Sane Democrat touches upon a major expectation issue for us
here re the idea that voters won't dis a sitting president to the pollsters absent a scandal of Nixonian proportions (which this is not - yet, and perhaps never will be). That doesn't necessarily translate into votes for Bush, btw, as the WaPo pollsters have taken pains to say.
Josh Marshall has an excellent analysis of the WaPo editorial page thinking out loud about this dilemma for Americans. On the one hand, it's increasingly obvious Rice will have to testify under oath, and the WaPo calls for that. OTOH, they insist she'll breeze though is as she obviously has nothing to hide, plus she'll give a strong performance. And they know that in advance.
The way they can function with such cognitive dissonance is easy for them... like most Americans, they supported the war in Iraq and will have to be kicked dragging and screaming to the idea that maybe they made a mistake. But like the multitude (upwards of 65%) who claim Clarke didn't change their mind regarding assessing Bush's performance on and before 9/11, his terrorism scores nonetheless fell 18 points in the Newsweek poll (all above polls at Polling Report). When Bush loses the WaPo editorial page, it'll be like Johnson losing Cronkite.
John Dean was initially disbelieved, and 44% of the American public in 1997 told pollsters Watergate was primarily about politics. Americans take things in slowly, and won't change their vote to Kerry until they get to know him better. Whether it's from the convention or the debates remains to be seen. Polls show trends but do not predict outcome. Not this far out, anyway.
