Dan Markel has an interesting post at PrawfsBlawg about how retributive theory fits into the Kennedy case:
This article of mine from a couple years ago (update: sorry, I had the wrong link earlier) tries to systematically undermine that conclusion. In fact it gives strong reasons to think retributive theory, at least in the version I find most compelling, is actually hostile to the death penalty. If that argument succeeds, it succeeds against the death penalty's use not only against murderers but also child-rapists. And since the Court has moved away in large measure from predicating its 8th Amendment analysis on non-retributive goals, it entails the conclusion that the Court should make the small move of extending Coker v. Georgia, which held that the Eighth Amendment prohibits the execution of a rapist of an adult, to this scenario.
I haven't read Markel's article (I've now added it to my to-do list), but I think he raises an interesting point. Maybe I'm just a pessimist, though, since I think actual retributive theory doesn't really factor into USSC opinions. Too often, it is just taken for granted that all punishment serves a retributive and deterrent function. Certainly, if Baze is any indication, the Court can't make sense out of empirical deterrence studies and I believe the same will be true of any discussion of retribution. I'd love to be proven wrong.
Related Posts: Kennedy v. Louisiana Resource Page