Let's see if I got this right.
5 votes to dismiss as improvidently granted. (Rehnquist, Scalia, Thomas, Kennedy, Ginsburg).
1 vote would have also preferred staying the case: (Ginsburg).
4 votes to remand to the Fifth Circuit to resolve all of the issues raised by the parties at the Court as well as the new ones created by the President's intervention. (O'Connor, Souter, Stevens, and Breyer).
The Court's ultimate disposition is not surprising, as Orin Kerr
has pointed out, but the Court doesn't usually spend 40 pages of the U.S. Reports explaining why it is DIGging a case.
It would be dangerous to read too much into the breakdown here. I still think that it's fair to guess Ginsburg would ultimately have been happy to find in favor of Medellin, and that O'Connor is more skeptical. In theory, this whole exchange is really about whether the Court should deal with big cases while they have them, or hold back into all the issues are cleanly presented. Ginsburg appears to think that the Court should hold back, and O'Connor does not.
Paul Stephan of Virginia has put his finger on the most curious aspect of Justice O'Connor's reasoning. Her view is NOT that the Court should resolve the case, but rather that it should have granted, vacated the lower court opinion, and remanded to the Fifth Circuit. But why should the Fifth Circuit be better positioned to resolve the important issues that the Supreme Court has identified than either the Texas courts or the Supreme Court itself?He writes:I admit to being deeply puzzled by Justice O'Connor's dissent. On the first page, she identifies three issues involving Avena and the VCCR that "deserve further consideration." She also says that "It seems to me unsound to avoid questions of national importance when they are bound to recur." Yet the three issues she identifies are fairly encompassed by the grant of certiorari. Why she concludes that the court ought to remand the case to the Fifth Circuit, rather than decide those three issues, is not explained in the opinion. She identifies as reversible error the Fifth Circuit's denial of a COA. Yet she does not clearly indicate what else the Fifth Circuit should do differently, given that it did address the VCCR issues on their merits and the independent COA issue arose only in the Supreme Court proceedings.
Prof. Stephan also notes: "that the four dissenters do not reject the 2253(c)(2) issue on its merits (although there is plenty of dicta indicating their leanings), but rather regard the issue as forfeited in this case."
The 2253(c)(2) issue was probably Texas' strongest argument: that the federal statute limiting federal appellate jurisdiction barred treaty-based claims. So it is possible that Texas could have still won on the merits (but not likely).
In any case, I think Medellin is in an even better position now than I first thought. Four members of the Court are on record as to wanting this case back, and all justices appear to agree that the Supreme Court will have jurisdiction from any final Texas state court judgment. But this time, when Medellin comes back (which seems all but assured), he avoids all the nasty federal habeas obstacles and gets a clean argument on the treaty's protections and, most importantly, the President's order.
Now we might have a different Court by then because that is at least a year away. But even so, Medellin faces tough obstacles. One dilemma that future case will raise is whether the Court will defer to an Executive determination on foreign policy, as it did in American Insurance Association v. Garamendi. In that case, Justice Souter wrote the majority, and Justice Ginsburg's dissent focused on the dangers of executive authority over the states. So it is hard to say how even that case will break down.
Delightfully, however, it should provide even more fodder for discussions on this blog and law review articles. Which, of course, is the really important thing...