The U.S. Court of Appeals for the Fifth Circuit is proposing an amendment to Circuit Rule 41.3, governing the effect of a granting a rehearing en banc. The proposed revision provides that if a motion for rehearing en banc is granted and the court subsequently lacks a quorum to decide the case en banc, the case will be returned to the original panel, the mandate reinstated, and the opinion issued as an unpublished, non-precedential opinion. The rule makes sense, even if it addresses a scenario that (one would think) is quite unlikely to occur.
Although there is no explanation accompanying the proposed amendment, it is clearly prompted by the recent case of Comer v. Murphy Oil, in which the precise scenario addressed by the rule occurred — and given that there was no rule in place, the original panel opinion was never reinstated. In Comer, the Fifth Circuit reversed the district court’s dismissal of a class action lawsuit against several energy companies alleging that their activities (including their greenhouse gas emissions) contributed to the intensity of Hurricane Katrina. The Fifth Circuit initially voted to hear the case en banc, vacating the panel opinion in the process, but then one of the judges recused, depriving the en banc court of a quorum. The case was then dismissed on the grounds that the absence of the quorum left the court unable to hear the case, or even to reinstate the panel opinion. (See the order and dissenting opinions here.) The plaintiffs subsequently filed a writ of mandamus with the Supreme Court, seeking to reinstate the panel opinion, but to no avail.
Although I think the initial Fifth Circuit opinion was in error, the result of the Comer proceedings was quite incongruous. If there were not enough judges to hear the case en banc, then en banc review should never have been granted, and the Fifth Circuit opinion should not have been vacated. What appears to have occurred here is that a conflict of interest was discovered, or came into being, after the conflicted judge had already participated in the vote to grant the en banc petition. As a consequence, the initial vote to rehear en banc (and vacate the decision) was procedurally proper and could not be undone as the recusal left the en banc court without jurisdiction to enter any additional orders. Following the rules as written resulted in a result that rubbed most observers the wrong way — hence the need for a revised rule. The proposed amendment will not do anything for the Comer plaintiffs, but it will fix this potential problem for future litigants — should this odd sequence of events ever actually recur.
Steve says:
Forget this specific context for a moment. If a judge issues an order, and then discovers – maybe a day later, or an hour later – that he should have recused himself, should he not at least consider vacating his earlier order sua sponte before recusing himself from the case? I would think so, but maybe I am wrong.
May 24, 2011, 3:46 pmAnderson says:
Glad to see this. Comer was a disgrace. And, given that FRAP 2 expressly allowed the court to suspend its own rules for good cause, it was a deliberate disgrace.
May 24, 2011, 3:47 pmShelbyC says:
Maybe he should consider it, but he probably shouldn’t do it, unless there is a rule requiring him to do so. Wouldn’t taking discretionary action in a case in which he has an interest be unethical?
May 24, 2011, 3:52 pmOrenWithAnE says:
I’m with Steve on this one — once the judge recused himself he not only removed himself from the panel but invalidated every prior action he took with respect to the case.
May 24, 2011, 3:52 pmShelbyC says:
But if that’s true, it should be by rule, not by the discretionary action of the judge. Isn’t that the whole point of recusal?
May 24, 2011, 3:56 pmVSJB says:
A judge doesn’t typically “undo” actions taken prior to recusal. And why should he? Assuming he recuses as soon as he becomes aware of a conflict, the conflict would not have tainted the impartiality of previous decisions.
May 24, 2011, 4:01 pmSoronel Haetir says:
Plus I can see circumstances where the actions were not in fact conflicted. Courts generally seem very loath to have to redo work that no one claims was undertaken improperly.
I could see Steve’s rule making sense if the conflict existed at the time the work was performed but was not realized, but this new rule also covers the case where no conflict existed before but does going forward.
May 24, 2011, 4:05 pmAnderson says:
I’m with Steve on this one — once the judge recused himself he not only removed himself from the panel but invalidated every prior action he took with respect to the case.
Is that really how it works? If a judge recuses after a year on the case, is everything prior wiped out?
May 24, 2011, 4:11 pmSteve says:
Not if the goal is simply to restore the status quo ante so that a new judge can take over the case in an orderly fashion. It depends on the specific situation. Many court orders are administrative matters having little to do with the merits.
By way of imperfect analogy, if I discover that I have to withdraw from a case because I have a conflict, I nevertheless have an obligation to do so in a way that minimizes the harm and inconvenience to the client. I shouldn’t simply drop the case like a hot potato on the theory that any actions I take are tainted by the conflict.
May 24, 2011, 4:24 pmJonathan H. Adler says:
Steve –
Might it not depend on the reason for the recusal? If the judge had not been aware of the conflict, there’s no reason to question the propriety of the prior action. There’s also the scenario where the conflict emerges — say the judge inherits stock in a company, or owns stock in a company that merges with a party to the case during the course of the litigation, etc. In such cases there’s no reason to think the initial vote was improper. In this case, I don’t think we actually know precisely how this occurred (though I would welcome a correction on this point).
JHA
May 24, 2011, 4:26 pmrequired says:
Nope, not even if a judge is removed is the prior work thrown out. I have to agree with Shelby on this one, once a judge realizes he or she has a conflict then any ruling by judge, even one vacating prior orders, is a ethically suspect. After recusal a different judge without a conflict can review the orders and see if they should be modified, but not the judge with the conflict. Once a judge is out of a case they should stay out, and even undoing prior work is inappropriate, let the new judge without a conflict handle it.
May 24, 2011, 4:34 pmShelbyC says:
But isn’t that the problem? We don’t want the judge saying, well, I ruled in favor of the plaintiff before I found out that I own stock in the defendant and my ruling cost me a buttload of money. Now, let me look at the specific situation and see if I should restore the status quo ante and get my money back or not…
May 24, 2011, 4:43 pmOrenWithAnE says:
Yes, it should be the rule. But in this case, the SCOTUS should have come up with some equitable relief such that a plaintiff can receive the appellate review to which he is entitled.
Quite the contrary, having now perceived the conflict he cannot be sure that anything he did prior was appropriately neutral.
Sorry, I was writing in ‘ought’ not ‘is’ (and didn’t even bother to make that clear, like a jerk.).
At the very minimum, if a judge recuses himself after a year, the new judge should review all the previous findings de novo.
I agree, except that the harm done to the client comes in two flavors: first in having a partial arbiter (or conflicted agent) and second in potentially having to relitigate seemingly settled matters.
I don’t see any way to balance the two harms without conducting at least some inquiry into the proceedings and thus imposing the second harm by reopening those matters for debate.
May 24, 2011, 4:49 pmOrenWithAnE says:
Good point, I hadn’t considered an emerging conflict, only a latent one, in my previous posts.
I can see the merit in this, even though I would prefer a more categorical approach.
May 24, 2011, 4:51 pmSteve says:
For a second time, I’m talking about administrative rulings, not rulings that go to the merits.
You seem to be suggesting that our only safeguard is a hard and fast rule that a judge cannot withdraw any orders once he has decided to recuse himself. But of course there is no such rule. There is no ethical or legal barrier I can think of to a judge entering an order which provides, “I hereby recuse myself, and last week’s scheduling order is withdrawn pending assignment of a new judge.”
May 24, 2011, 5:07 pmShelbyC says:
Well, you didn’t say that. Maybe a judge can consider whether or not he’s doing something administrative or something that goes to the merits, I don’t know. But I’m not sure what that has to do with this thread, doesn’t a grant of re-hearing en banc pretty clearly go to the merits? If that’s what you mean by administrative, then no, conflicted judges shouldn’t be ruling on administrative matters.
May 24, 2011, 5:35 pmShelbyC says:
Steve, shouldn’t the test be whether or not he is required to do it as part of the recusal? In your example, istm that the judge has no choice but to withdraw the scheduling order. But a decision that benefits a party that the judge can exercise judgement to make or not to make is pretty clearly off limits, no?
May 24, 2011, 5:46 pmSteve says:
I don’t know about that. Let’s assume that a judge enters a preliminary injunction that is going to impose significant compliance costs upon one party. The next day, he discovers a previously unknown conflict that requires him to recuse himself. I don’t think it’s obviously illegal or unethical for him to say “you know what, since there’s going to be serious consequences from my order, I’m just going to withdraw it and let the new judge decide whether such an order is appropriate.” I’m not suggesting this is common, but I don’t know where I can find an absolute bar to the concept.
May 24, 2011, 6:36 pmJay says:
But the criticism here isn’t primarily of the specific judge who recused, it’s of the remaining en banc court, for failing to reinstate the panel opinion and/or declare the en banc vote itself void.
May 24, 2011, 7:02 pmShelbyC says:
Although presumably lifting the order would impose costs on the other party. Either way, the judge is picking a winner and a loser. And if the winner is the judge himself, say, that’s probably bad.
May 24, 2011, 7:04 pmShelbyC says:
The criticism I was responding to was this:
May 24, 2011, 7:40 pmSoronel Haetir says:
Except the remaining panel members believed they were bound by rule to not do that. They would in fact seem to be the entire point of the new rule. A gap was found in the old rules so now it is being fixed.
It sucks for the winning litigant below but I would say playing by the rules we have and fixing them going forward is the wisest course. Prospective fixes like this defuse the argument that a change is being made to help any particular party instead of helping everyone who uses the courts.
May 25, 2011, 1:07 amJosh Bornstein says:
I’d be willing to bet that more than one Civ Pro professor used these facts for her or his final exam. What a great fact pattern (and how horrible for the party that got screwed).
May 25, 2011, 4:21 amDJR says:
The bar would be to avoid even the appearance of impropriety. Since judges rarely say why they are recusing, a recusal combined with vacating a PI could certainly raise the appearance of impropriety, particularly if you’re the side who has to convince a second judge that a PI is warranted, and particularly if vacating the PI relieves the other side of significant costs.
May 25, 2011, 12:02 pmSoronel Haetir says:
Steve,
I would also ask how quickly the timer on this idea runs. A minute? An hour? A day, a week, a month a year? And if it’s left up to the judge then you get into the sorts of issues that others have brought up above. It seems far preferable to me that a judge simply quit working on a case once the recusal bar is met.
I do see lawyers being in a somewhat different position than judges, but we are talking about what judges should do in these circumstances. And even for lawyers the work done should be the bare minimum required to hand off the work to someone else.
May 25, 2011, 12:56 pm