The other day I said I would make a few more comments about Joshua Green’s article on Rove in The Atlantic. I missed my own deadline, but here goes.

The lesson Green takes from what he takes to be Rove’s failure is that–as Green put it in the New York Times the week it was announced that Rove is leaving the White House–”government cannot effectively refashion American society.” In Green’s view, which I share, Rove was trying to use policy to change American society to make it more amenable to Republican dominance. I don’t think it’s quite right to say that this type of ambition is categorically doomed to fail because of some conservative insight about the intractability of societies or the perils of reform. The New Deal is a classic case of how public policies made this country more statist in its political orientation, not just by being popular but by creating and organizing new constituencies for federal activism. (Nor was this effect a coincidence: Recall “tax and tax, spend and spend, elect and elect.”) Reagan had a more modest achievement in the reverse direction, making Americans more self-reliant economically and thus better disposed to free markets.

The paradigmatic Bush-Rove policy designed to change American politics was of course Social Security reform. The theory was that if it went through, voters would become less dependent on government and more dependent on private capital markets. Therefore, they would either start looking less favorably on the Democrats as the party of government, or force the Democrats to become a less statist party.

Whether or not that theory was correct–and I wrote more than just about anyone to support it–you can at least see how it would work. But many of Green’s other examples don’t fit the model. Social Security reform, if it worked to realign America, would work by altering voters’ material interests. The No Child Left Behind Act, which Green holds up as a model of bipartisanship that Rove and Bush should have followed, wouldn’t have done that even if it had been extremely successful.

When it comes to NCLB, then, Green (or Green’s Rove) must be using a different model of how public policy can help a party. It’s the more conventional model: A party enacts a policy that proves popular and wins it credit. But how could NCLB even have worked in this more modest fashion? Democrats voted for it more heavily than Republicans. Yet the public was supposed to be grateful to the GOP for it–for a generation? Comprehensive immigration reform, another item Green mentions, would have done little for a Republican realignment, even on the most optimistic scenario, for the same reason. Assume it had passed. Why would Hispanics have been grateful to Republicans for it?

More generally, how would the bipartisan strategy Green promotes have furthered Rove’s ambition? It’s one thing to say that a bipartisan strategy would have resulted in a more successful presidency and thus left the GOP in a better position than it is now. That’s a reasonable view, although I’m not sure that the Democrats, coming out of the Florida recount, would have made it possible. But there’s no way it could have led to a realignment.

Maybe the quest for a realignment was quixotic and led to Rove’s making mistakes that left the GOP worse off. But I read Green as saying something a little different: that Rove’s methods subverted his goal of a realignment, that is, that he might have gotten where he wanted to go if he had done things the way Green implicitly suggests he should have. And I’m not sure that’s true.