The Washington Post today ran the least convincing op-ed by William Kristol that I’ve ever read. I’ll deal with his general claim about Republican overconfidence in a piece for the site tomorrow. Two points now:
1) Kristol says that “[t]he voters. . . have consistently preferred divided government.” Previous periods of unified government have been immediately ended by voters–in 1980, with the election of Reagan, and in 1994, with the election of a Republican Congress. That’s a very limited sample, and it could mean that voters have consistently voted to end unified Democratic governments. It is clearly untrue that most voters have wanted divided government. Most voters have wanted either unified Democratic or unified Republican government, and in recent years the percentage of voters who have chosen unity has increased. Nor can it be assumed that everyone who split their ticket did so on the basis of a conscious preference for divided government. More of them probably voted for an incumbent congressman, or voted a split ticket for some other reason.
2) Kristol suggests that voters may not back Bush over Dean on the war on terror because too many of Bush’s subordinates are insufficiently “committed to victory in that war.” He writes that “his secretary of state seems committed to diplomatic compromise, and his secretary of defense to an odd kind of muscle-flexing-disengagement.” (Didn’t Kristol encourage Powell to run for president in 1996?) I’d like to have a Department of State more committed to the president’s policies, too, and I’m certainly open to the argument that Rumsfeld’s version of military transformation short-changes the need for boots on the ground. But I’ve always respected Kristol as a political analyst because he usually resists the temptation to assume that adherence to his views is the precondition for political success. That assumption seems especially foolish in this case.