David's op-ed this morning was intelligent as he almost uniformly is. He's absolutely right that Rove's obsession in reshaping Republicanism to cater to a post-Clinton landscape was pure tactics and no strategy. It was a game-plan, above all, devoid of philosophical coherence and, in many cases, as Josh Green points out as well, antithetical to conservatism as broadly understood for the last four decades. Bush, however, was integral to this abandonment of conservatism as well. A man who had never understood conservatism except as an esoteric form of class resentment had no reason to push back against a mix of short-term highs and long-term lows. But I must differ from David on this:

Wedge politics unites a large constituency on one side, while splitting the coalition on the other side. In the 1970s, crime was a wedge issue: pushing white urban Democrats away from their black and liberal New Deal allies. In this strict sense, the only wedge issue Mr. Rove deployed was immigration, and he deployed it against his own side, dividing business donors from the conservative voting base.

Obviously, the gay marriage issue was a wedge issue. And it was central to Rove's re-election strategy; and in helping win Ohio in 2004, it may have made the difference in the election.

Rove had pledged not to use gays as a wedge to divide elite Democrats from more socially conservative (and often religious) ones, but he did when it became clear it could make a difference in the election. The disingenuousness of the president's argument - there was never any threat that civil marriage would be mandated nationally by a court decision in one state - was proof of the cynicism behind it. It relied on demonizing a minority of people, one of whom was Rove's own stepfather. It was a sign that the Bush administration took politics much more seriously than policy, or even decency. It was a clear sign that they would divide if it meant they could rule. David cannot whitewash that from history, or his own role in promoting and defending it.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.