Paradoxes in Political Science

Kevin Drum wonders at the perplexing results of the exit polls from New Hampshire that show that Democrats who want quick withdrawal from in Iraq supported Hillary by substantial margins, while Democrats who want a more gradual withdrawal or to keep troops in Iraq voted for Obama by substantial margins. The same trend occurred on the Republican side--voters who disapproved for the war supported McCain, and voters who approved supported Romney. He asks, "huh?"

I think this might be explained by a phenomenon in political science called "Nixon-in-China" syndrome: the public trusts the leader to take action that he is least ideologically predisposed to. It took Nixon to go to to China because the right could trust him not to betray American interests; similarly, it takes left-leaning governments to undertake reforms in areas like labor and welfare because it is easier for them to sell such programs to liberal interest groups (since left-leaning governments ostensibly have those groups' interests as a priority). That's why it also took uber-hawk Ariel Sharon to withdraw from Gaza, and why it will probably take someone who opposed the war to make the decision to tough it out if that's what the situation calls for; and vice versa.