Today the D.C. Circuit heard oral argument in a challenge to Salim Ahmed Hamdan’s conviction for material support for terrorism–a crime that Congress recognized after Hamdan committed the acts for which he was convicted but which the Government says is derived from “the domestic common law of war.” Reports by Lyle Denniston at SCOTUSblog and by Wells Bennett at Lawfare suggest that Judge Kavanaugh is sympathetic to Hamdan’s challenge to the ex post facto application of the criminal law. Chief Judge Sentelle was reportedly “somewhat less skeptical than Judge Kavanaugh.” The panel ordered further briefing on whether Hamdan’s case is moot because he has served his time and been released–an issue of particular interest to Judge Ginsburg, who authored a recent decision holding a challenge to Guantanamo detention (but not a criminal conviction) to be mooted by the detainee’s release.
Denniston suggests that Hamdan will lose even if there is only one vote to uphold his conviction, assuming Judge Ginsburg believes, as his questions suggest, that the case is moot.
[I]f it should turn out that Hamdan wins only Kavanaugh’s vote against the material support charge, and loses Judge Sentelle on that point and loses Judge Ginsburg on mootness, the legal result would be that Hamdan would lose, because such a divided result would simply mean that the Circuit Court did not have the votes to overturn a ruling against Hamdan, on all points, by the U.S. Court of Military Commission Review in a decision last June.
This isn’t necessarily so. Although affirmance is the likely outcome if the panel divides as Denniston predicts, it is not the only possibility.
Courts usually decide cases by the “outcome voting” method that Denniston assumes will carry the day here. Under the outcome voting method, each judge casts only one vote–a vote for the outcome he believes is correct, taking into account only his own answers to the logically antecedent questions that must be resolved to reach that result. Under this method, two judges on a three judge panel may create a majority even though their rationales are mutually inconsistent. Thus, because Judge Ginsburg (in Dennison’s hypothetical) thinks the case is moot, he would vote to dismiss the appeal regardless of his views on the merits. And because Chief Judge Sentelle (hypothetically) would vote to affirm the conviction on the merits, Judge Kavanaugh’s (hypothetical) vote for reversal would be in the minority.
But that is not the only way courts decide cases involving multiple outcome-determinative decision points. Under the “issue voting” method, the panel would first address the logically antecedent question of mootness and–after holding, two to one, that the case presents a live controversy–turn to the merits. Judge Ginsburg would then have to decide whether the conviction is lawful even though he would not reach that question if he were the only judge on the bench. If Judge Ginsburg were to agree with Judge Kavanaugh that the conviction is illegal, his vote would create a majority for reversal.
This panel’s voting method could mean the difference between affirming and reversing Hamdan’s criminal conviction.
Update (5/3/12): Adam J. White points out that Chief Judge Sentelle employed something like issue voting in Massachussets v. EPA to resolve a similar three-way split in that case. Although he would have held that the petitioners lacked standing, Chief Judge Sentelle concluded that the contrary view of his two fellow panelists was the “law of the case” and joined Judge Randolph in denying the petitions rather than dismissing them for lack of standing:
My conclusion [that petitioners lack standing] leaves a slight problem. . . . The problem vexes only as to petitions for review . . . which Judge Randolph would deny and Judge Tatel would grant. I would dismiss those as well, as I would hold that we have no jurisdiction to either deny or grant them. How then are we to reach a judgment? [¶] The Supreme Court has suggested a way, or at least Justices of the Supreme Court have. Most recently, in Hamdi v. Rumsfeld, 542 U.S. 507 (2004), Justice Souter, joined by Justice Ginsburg, differed from the plurality in a fragmented opinion adjudicating the due process rights of alleged enemy combatants held at Guantanamo Bay by the United States military. Justices Souter and Ginsburg would have vacated the judgment of the Court of Appeals and remanded for proceedings consistent with their view that the government had failed to justify holding the petitioner. However, because that view did not command a majority of the court, and because of “the need to give practical effect to the conclusion of [a majority] of the court rejecting the government’s position,” Justice Souter (joined by Justice Ginsburg) joined with the plurality “in ordering a remand on terms closest to those I would impose.” I will take a similar course here.
It is not clear, however, that the Chief Judge’s method was true issue voting. Instead, like Justice Souter in Hamdi, Chief Judge Sentelle seems to have selected the available outcome (denial) that was most similar to his preferred outcome (dismissal), rather than truly setting aside his justiciability concerns and deciding the merits on a blank slate. This seems to be more like a modified version of the usual outcome voting paradigm. For Chief Judge Sentelle in Massachussets v. EPA, this may have been a distinction without a difference–he probably agreed with Judge Randolph on the merits–but it could matter very much in Hamdan if Judge Ginsburg believes the conviction violates the Ex Post Facto Clause. True issue voting would require reversal, but joining a decision “on terms closest to those [he] would impose,” would counsel affirming the conviction.
From the oral argument transcript:
JUDGE KAVANAUGH: To the extent it codifies a new war crime, can it be applied retroactively, consistently with the ex post facto clause?
MR. PUE [for the United States]: If it codified a new war crime, no, it could not. But –
JUDGE KAVANAUGH: Okay. So the question turns on, then, whether this was a war crime under the law of war –
MR. PUE: Exactly.
. . .
JUDGE KAVANAUGH: . . . You agree material support for terrorism is not a war crime under international law, correct?
MR. PUE: That’s correct. So, we have to look to another source other than customary international law to codify that.
. . .
JUDGE KAVANAUGH: — going backward to what we are talking about in this case, . . . if you are not relying on a statute or a treatise, Justice Stevens’ plurality opinion [in Hamdan v. Rumsfeld] says, you have to have a precedent, a precedent that’s plain and unambiguous. And you just acknowledged, I think, that we don’t have that.
MR. PUE: Well, I don’t think, I don’t think that that, as I said earlier, applies to an act of Congress. I think that you need to defer. You need to afford some deference to Congress –
JUDGE KAVANAUGH: For the acts before 2006, we don’t have an act of Congress. That’s what I’m trying to say.
MR. PUE: Yes, you didn’t have an act of Congress. You had Congress saying in 2006 and 2009 what it was doing here was codifying offenses that had been traditionally triable under the law of war or otherwise by military commission. And recognition, in my view, that it understood that there were two sources of offenses triable by military commission.
JUDGE KAVANAUGH: But we can’t just defer to Congress’ view on what’s ex post facto because that would eliminate the ex post facto clause.
- Lewis A. Kornhauser & Lawrence G. Sager, The One and the Many: Adjudication in Collegial Courts, 81 Cal. L. Rev. 1, 41 (1993) [Lexis] (“In paradoxical appeals from criminal convictions, one of two propositions is always true: a majority of the judges on the appellate court agree that the conviction of the defendant was illegal, or in the alternative, shifting majorities agree on each of the subordinate propositions necessary to support the proposition that the conviction of the defendant was illegal. Either of these circumstances ought to give an appellate court pause. Serious criminal convictions alter the status quo in a drastic way. They stamp an individual guilty and subject her to imprisonment. Even in an era in which we are frightened and dismayed by the propagation of lawless violence around us – perhaps especially in such an era – we would do well to insist that we not cheat on our own standards of fair process or shortchange criminal defendants of their substantive legal rights. A court sensitive to these concerns might well take the view that any conviction compromised to the point of paradox ought to be overturned.”)
- David Post & Steven C. Sal0p, Rowing against the Tidewater: A Theory of Voting by Multi-Judge Panels, 80 Georgetown L.J. 743 (1992).
- David Post, McDonald and the Voting Paradox, Volokh Conspiracy (July 7, 2010)
- Gul v. Obama, No. 10-5117+, 652 F.3d 12 (July 22, 2011) (Ginsburg, J., joined by Tatel & Brown, JJ.) [Guantanamo, mootness] (affirming order of detainability)
- Massachussets v. EPA, No. 03-1361+, 415 F.3d 50, 60-61 (2005) (Randolph, J., with Sentelle, C.J. dissenting in part and concurring in the judgment, and Tatel, J., dissenting)
- Steve Vladeck, My Three Takeaways from the Hamdan Argument, Lawfare (May 3, 2012) (“[T]he superficial impression one takes away from the argument is that there’s one vote to dismiss on mootness grounds (Ginsburg), one vote to uphold on the merits (Sentelle), and one vote to reverse on the merits (Kavanaugh). Needless to say, that would be a remarkably messy (and confusing) result—virtually necessitating en banc review. If Judge Kavanaugh is as skeptical of the government’s position as he appeared to be at oral argument, however, perhaps he might find a way to convince one or both of his colleagues to join a narrow opinion relying on the Ex Post Facto Clause and invalidating Hamdan’s conviction.”)