So Kurtz looked at the places where gay marriage has been instituted the longest. And the results are that marriage is the weakest of all there.
Now Kurtz does show a direct causal relationship there, at least not empirically. But the evidence is enough to suggest that Eskridge's and Sullivan's position is utterly untenable.
Gay marriage may or may not in fact damage the institution itself. In fact, it is apparent (especiall in Scandinavia) that most of the damage was done before same-sex marriage and by heterosexuals to boot. But the idea that gay marriage will strengthen marriage itself is simply not tenable based on the evidence today. In fact, it is flatly absurd.
I'm not sure if it is or isn't untenable. I think what we're dealing with is a lot of smoke and mirrors on both sides. Rather than keeping things clear and comparing apples to apples, each is clouding the picture by putting their own interpretation on two different sets of statistics (number of divorces and number of marriages) and concluding that the amalgamation of the numbers support their position.