Why You Might Not Want a Judge Who Already Ruled Against You to Recuse Himself:

The Supreme Court just agreed to hear Hamdan v. Rumsfeld, which involves a challenge to the military commissions set up to try enemy combatants; the issues are whether the commissions are duly authorized by statute or possibly by the President's inherent powers, whether the Geneva Convention bars certain procedures used by those commissions, and whether the Geneva Convention is even judicially enforceable by courts (on the theory that it's self-executing). The D.C. Circuit upheld the commissions (correctly, in my view), in an opinion that then-Judge John Roberts joined. Chief Justice Roberts has therefore recused himself from considering the case.

But here's an interesting tactical twist: Today's AP report contains the following line — "Hamdan's attorneys may ask Roberts to participate in the case to avoid a 4-4 tie." Why would Hamdan's lawyers want to do such a thing?

If Chief Justice Roberts recuses himself, and the Court ties 4-4, that will affirm the lower court decision. So if you're Hamdan's lawyer, you'd reason: If Roberts would have voted against us, and this produced a 5-4 result for the government, Roberts' recusal wouldn't help us, since we'd still lose 4-4. But if Roberts would have voted for us, at least in part, and this produced a 5-4 result for us, Roberts' recusal would hurt us. The chances of Roberts voting for us, given his vote below, are small, but not zero — the lower court decision rested on certain Supreme Court precedents that bound Judge Roberts, but perhaps Chief Justice Roberts might consider reversing or modifying some of those precedents. Or maybe we can talk him around with one more argument; or perhaps his new colleagues can talk him around.

Yet here's the downside: If Roberts isn't recused, he can talk around some of his colleagues. The Justices' attitudes aren't independent factors that simply need to be aggregated; each Justice can try to persuade the others, through questions at oral arguments, discussion at their end-of-the-week conference during which they vote (a very brief discussion, but one that might still affect people), and through the exchange of opinion drafts and related memos. If Roberts is recused, then he's supposed to not participate in these discussions, and not just not vote (and I'm pretty sure that this is in fact the practice, not just the theory). But if Roberts isn't recused, then he will participate, and maybe turn a 5-3 vote in favor of Hamdan into a 5-4 vote in favor of the government. And Roberts' status as Chief may be especially influential here, not just because he'd be able to assign the majority opinion, but also because Justices are probably a little more likely to accommodate the wishes of the Chief.

There's also one other downside, albeit hypothetical: If Justice O'Connor's replacement is confirmed after the argument in Hamdan but before the decision is handed down, then a Roberts vote for the government wouldn't mean a 5-4 pro-government decision as opposed to a 4-4 pro-government decision — it might mean a 4-4 pro-government decision as opposed to a 4-3 pro-Hamdan decision. This won't happen, of course, if Alito sails through in January, as seems likely. But who knows what the Senate will do? If I were Hamdan's lawyer, I'd be figuring out the latest time I can file a motion asking Roberts not to recuse himself (there's an answer to this, but I just don't know it), since I'd want to — if possible — wait until I see whether Alito (or someone else) is confirmed before the Hamdan oral argument.

Finally, one other twist: From the perspective of people who oppose the military commissions, a 5-4 Supreme Court decision to uphold the commissions is worse than a 4-4 vote that has the result of upholding the commissions. The former sets a binding precedent, and puts the persuasive weight of the Court's majority behind the commissions; the latter leaves the issue up in the air, at least for now.

But Hamdan's lawyers, even if they personally care mostly about the legal question, must focus on the welfare of Hamdan and Hamdan alone (at least unless they've somehow expressly agreed with Hamdan otherwise, though I'm not even sure to what extent they can do that). So they need to take the course that maximizes Hamdan's chances of winning, even if increasing his personal chances of winning by 1% hurts the broader anti-commission cause by considerably more.

Finally, a warning: These are just a few quick ruminations of a Court-watcher who isn't a Court litigator; I think they're right, but I'd love to be corrected if I'm mistaken.

UPDATE: Reader Dan Schmutter points out: "One other pro-recusal thought. As a member of the majority below, Roberts may be more persuasive with his colleagues having more familiarity with the record below than another member of the court simply reading the briefs for the first time."

Dean Kimball (mail):
Euguene - you stated you would uphold the commissions. What is your reasoning? Do you have any prior posts that address this?
11.7.2005 1:57pm
Eugene Volokh (www):
Too swamped to get into this right now, but I basically think the D.C. Circuit's opinion was quite persuasive.
11.7.2005 2:03pm
This is all very interesting, but I find it highly unlikely that CJ Roberts would un-recuse himself based on a request from Hamdan's lawyers.

Roberts' recusal says, in effect, "my impartiality might reasonably be questioned in this case."

Hamdan's lawyers' response says, in effect, "maybe so, but our cost-benefit analysis tells us that putting you back in the case is more likely to help us than hurt us."

Who in his right mind would grant such a request?

Unless the court has something weird happen, like 3 Justices die, it seems to me that the only good reason to un-recuse oneself would be if the reason for recusal is conclusively proven to be wrong. In a recent patent case involving "Merck KgaA v. Integra," Justices Breyer and O'Connor did not participate in the decision to grant cert, but rejoined the case for the argument and the disposition, apparently because they had originally thought that Merck &Co, in which they owned stock, was connected to the petitioner. The petitioners' counsel explained in a letter "to address a possible misunderstanding" that this was not the case, and that while the companies shared a common ancestor, they had been entirely separate since 1917. And O'Connor and Breyer rejoined the case.

Other than something analogous to that fact pattern (e.g. "this isn't the same Salim Hamdan on whose case you ruled earlier"), I don't think there are many good reasons to un-recuse, and I think there's zero chance that Roberts will do so here.
11.7.2005 2:30pm
In one case I argued before my state supreme court, a visiting judge sat to fill a vacancy. The issue in the case considered an issue that had been bouncing around the lower courts with different results.

In the other court of appeals cases, the visiting judge had ruled against my client's position dozens of times. I almost asked for her recusal. She ended up on my side of the case. The margin of victory was her one vote.

I think she changed her mind because the lower court opinions mostly parsed the opinions of the state supreme court, not the underlying statutes.

In the Hamdan case, the DC Circuit spent a lot of time parsing U.S. Supreme Court decisions. As Chief Justice, Roberts will have the authority to give less weight to those decisions and more weight to the underlying legal theories.

Will he change his vote? I agree it's unlikely. But given the effect of a 4-4 split, it's probably worth the risk for Hamdan's lawyers to try to keep Roberts on the case.
11.7.2005 2:45pm
Anderson (mail) (www):
Reading the tea leaves on Kennedy, including his concurrence in Rasul, I don't think he's a lock for a 4-4 split by any means, which is another reason to keep Roberts off the case--especially given EV's sage comment that Roberts arguing in chambers might be persuasive. I daresay Kennedy would be better able to fend off Scalia and Thomas.
11.7.2005 3:17pm
cfw (mail):
Kennedy looks likely to support Hamdan a good bit, judging from his recent interview for Academy of Achievement.

I agree Roberts probably will not agree to unrecuse. I see no point in having him around - too much chance he could sway folks in favor of the US.

I suggest EV needs to think more about what persuades him in the DC Circuit case. A derelict on the waters of the law, in my view.
11.7.2005 3:40pm
Greedy Clerk (mail):
Wow. This is a real tough tactical call. And it's fascinating. Another pro for asking for Roberts not to recuse: It may garner some goodwill from Roberts -- I am not saying enough to swing his vote or anything but every little bit counts.

Moreover, Roberts may turn down the request, feeling that he is obligated to recuse regardless of what the lawyers think (which I would think highly likely). By asking for him not to recuse, the lawyers gain goodwill from Roberts for future cases -- this being especially important if there is in fact a 4-4 tie. (I recognize of course that there is a huge chance that the same lawyers appearing again on similar issues may not happen.) Thus, asking for recusal can be no-lose situation. In fact, I would think it very likely that Roberts would decline to recuse himself.
11.7.2005 3:56pm
Would the lawyers really get any goodwill from asking to unrecuse Roberts? It would be clear that they were doing so for only tactical reasons.
11.7.2005 4:11pm
I'm curious - has it ever happened before that a judge ruled one way at the Court of Appeals level and then was elevated to Supreme Court and voted the other way on the same exact case at the Supreme Court level?

Can anyone think of an example?
11.7.2005 5:15pm
Using the only supreme court that I know well (my state's), I think it would be a bad idea to file the motion. Gaming a court for votes is bad form. It is low-ball lawyering. Judges know it, and they generally don't like it.

Usually, gaming is done to get someone off (like in Tom Delay's case), it's no different when you're trying to get someone on a case who shouldn't be there.

Again, I'm using my state supreme court as my framework. I'd reconsider if frequent SCOTUS litigators said that SCOTUS worked differently.
11.7.2005 5:46pm
CharleyCarp (mail):
I'm not sure the reporter understands what's going on.

On another note, rejected intervenor al Oteibi has 45 days from October 11 to seek en banc review in the DC Cir from denial of his motion to intervene. If granted, and if Judge Roberts is retroactively recused from the panel opinion, the result on the Common Article 3 issue could become 1-1. This would leave Judge Robertson's opinion in place on this issue should the Supreme Court split 4-4.
11.7.2005 6:15pm
CharlieCarp, why would Roberts be "retroactively recused from the panel opinion"? That seems ridiculous to me.

On another note, as Public Defender notes, JR is highly unlikely to unrecuse himself based on a "gaming"-type motion by Hamdan's lawyers.
11.7.2005 6:20pm
CharleyCarp (mail):
WB: because he should have recused from the panel when the Attorney General interviewed him for a coming opening.
11.7.2005 6:57pm
CharleyCarp (mail):
Or put the meeting on the record, and given the parties an opportunity to waive the obligation to recuse.
11.7.2005 6:57pm
bolivar (mail):
I think this may well be a move on the part of the defendant's lawyers to signal confidence: They probably think that Roberts will recuse himself regardless but by asking they come across as assured of their position.
11.7.2005 7:17pm
Ted Frank (www):
I'm curious - has it ever happened before that a judge ruled one way at the Court of Appeals level and then was elevated to Supreme Court and voted the other way on the same exact case at the Supreme Court level?

It's also possible from the days when Supreme Court justices regularly rode circuit; it was not that unusual for them to be deciding their own appeals.
11.7.2005 9:13pm
Ira Brad Matetsky (mail):
I don't know that recusal would be waivable in the Hamdan situation. See 28 U.S.C. sec. 47: "Disqualification of Trial Judge To Hear Appeal. No judge shall hear or determine an appeal from the decision of a case or issue tried by him." Unlike other recusal statutes (cf. 28 U.S.C. sec. 455), sec. 47 has no provision for waiver by the parties.

To be sure, section 47 is in the Code chapter about Courts of Appeals, not the Supreme Court, but the principle clearly applies and if I recall correctly the statute was enacted specifically to block Justices on circuit from sitting in appeal of their own decisions below (which indeed had been the standard practice for the first 100 years or so of the federal court system).
11.7.2005 9:29pm
I don't think asking Roberts to un-recuse himself would show that the lawyers were "assured of their position." Just the opposite. It would show that they were insecure about their chances of getting five votes from the remaining eight justices.

The lawyers should spend their time researching the issues, not figuring out ways to game the Bench. To paraphrase Rumsfeld, you go to court with the panel you have, not the panel you wish you had.
11.8.2005 2:07pm