Peter Beinart argues in his op-ed “A Democratic Call to Arms” in the Sunday Washington Post that Democrats must make big progress among military voters to counter the Republicans’ big success among black and Hispanic voters. As Beinart puts it:

Republicans have been conquering their demographic challenge, while Democrats have not. Between 2000 and 2004, George W. Bush increased his share of African American votes from 9 percent to 11 percent….Among Hispanics, Bush’s total rose from 35 percent to as much as 44 percent. But despite widespread talk about military disaffection over Iraq, John Kerry won only 41 percent of Americans with military experience.

I think it’s a fine idea for Democrats to increase their share of the military vote–but not because they need to match big Republican gains among blacks and Hispanics. Indeed, the idea that, based on the last election, Republicans have somehow conquered “their demographic challenge” is absurd.
In terms of the black vote, Kerry’s 88-11 margin is the highest obtained by a Democratic candidate since the exit polls started in 1976, except for 2000 and Mondale’s 1984 campaign. To say the 2004 result represents a breakthrough for the Republicans is ridiculous.
In terms of the Hispanic vote, if Bush had really gotten 44 percent of the Hispanic vote, that would represent some kind of a breakthrough. But he almost certainly did not. The 44 percent figure Beinart alludes to is the NEP national exit poll figure which has more or less been repudiated by the exit pollsters themselves, due to sampling problems in the 2004 poll. The best exit poll figure for the Hispanic vote at this point is 40 percent, based on aggregating all the state exit polls. And there are good reasons for thinking that the true figure may actually be closer to 39 percent.
So, by all means, go after the military voters. But Democrats should go after them not out of desperation (Help! We’re losing blacks and Hispanics!), but because it’s a good idea in its own right.