Fair warning: I'm about to be a jerk. But a principled one.
There's a complicated ERISA case. It gets briefed by the parties and argued in May 2009. The panel is split. Two judges write a majority opinion, and one writes a dissent.
It takes a while. Over a year, actually. But finally, things are just about finished. There's some tinkering with the opinions that still needs doing, but it's minor stuff. Things will be ready to file in mere days.
Then one of the judges dies.
What do you do? What can you do?
You obviously can draw a new judge. But all that time and effort has already gone into writing the opinions. Do we really have to reinvent the wheel?
Does it matter whether the deceased judge was one of the two judges in the majority or not? Can you count her vote even when she's no longer with us? Wholly apart from whatever the rules and/or statutes say, does Article III of the Constitution provide an independent limitation on the courts? Appeals have to be heard and decided by such judges: Would counting the vote of a deceased member violate Article III? Do you cease to be an Article III judge upon death? (The text of the Constitution says that judges shall hold office during "good behavior" -- there's nothing about termination upon death, or whether dying counts as bad behavior.)
Those were my thoughts, anyway. I'm not sure that I had a fully worked-out vision of Article III constraints. But I do think that there may be something there. Take an extreme case: Three judge panel, briefs and oral argument happen, two judges say they're voting for plaintiff and one says he's voting for defendant, and after nearly finishing the opinion the two plaintiff votes die in an automobile accident. Would the Constitution allow an appellate court to say that the two votes prevail and hence plaintiff wins? Would that be a decision by an Article III court, when the only current Article III member voted the other way?
I could go on and on about this (and probably already have), but my tentative conclusion is that (1) there's a constitutional minimum here somewhere, but (2) there's also some degree of flexibility, assuming the creation of rules in advance, about what counts as a "final" decision. I think that a court could, consistent with Article III, say that dead judges get a vote, as long as their decision was sufficiently clear in advance pursuant to specific rules that said that certain acts weren't merely "tentative" and were intended to be final even in the face of an intervening death.
Why did I think all this, you might ask? Am I simply feeling a keen sense of mortality today?
No. I had all those thoughts when I read this opinion.
It's not an extreme case like the two-death example I articulated earlier. But it did raise the issue. Since it's very much like the example I started with. Two judges remain (Judges Kleinfeld and Silverman), and the one who dies is Judge Hall, whose long battle with cancer ended February 26, 2011.
Nine days later, Judges Kleinfeld and Silverman do what one might expect in this situation. Particularly since they agreed and Judge Hall was the one who dissented. They publish both their opinion as well as Judge Hall's dissent. Noting Judge Hall's passing, and her instructions to publish her dissent, in a footnote.
Which is a sweet thing, and entirely appropriate. They were, after all, already in the majority. Two judges is a quorum. The case was decided. No need to let Judge Hall's wisdom be wasted.
But was it right?
On this I'm of mixed thoughts. On the one hand, it's a dissent. So it doesn't "mean" anything anyways. It's not a holding. It's just some ideas. Presumably that's just fine.
Plus, as a practical matter, you and I both know that nothing was going to change in the next nine days. The case had been under submission for almost two years. The opinions were written. All that was left was the final proofreading. They were essentially as "final" as they were ever going to be.
On the other hand, where do you draw the line? What about a judge who expresses a view at oral argument and then promptly dies. Do those votes count as well? Does it matter which way the vote goes? Does it matter if this judge never, rarely or often changes his mind? What about a judge who expresses a view in a bench memo prior to oral argument? Does that count? At what point do we invest a decision with sufficient finality so that even when the judge dies, we recognize that vote? We know what the rule is for juries: After polling. What's the appropriate rule in the Court of Appeals?
So those are my theoretical issues. Backed up, again, by Article III concerns. We know that we can (and often do) replace a judge who has died with another judge. I've seen that time and time again in the Ninth Circuit (and elsewhere): someone dies, a new judge is drawn, the new judge listens to oral argument, and a new vote is taken. The tougher question is when you can do otherwise. Was what Judges Kleinfeld and Silverman did here right?
Let me add just one additional piece to the puzzle. Perhaps the most troubling one. We actually have a rule about this in the Ninth Circuit. It's General Order 3.2(g). Which says: "If a member of a three-judge panel becomes unavailable by reason of death, disability, or departure from the court and the case is under submission, the Clerk shall draw a replacement by lot." The case here was clearly under submission. So it seems like, if you follow the rule, you have to draw a replacement.
Not only that, but there's also a Ninth Circuit opinion about this as well -- one that was generated after Judge Ferguson's sudden death in 2008. There, Judge Ferguson joined Judge Reinhardt to hold (briefly) that X was unconstitutional (but defendant entitled to qualified immunity), Judge Smith concurred (arguing that X wasn't unconstitutional), the opinion issued, Judge Ferguson died, a replacement was drawn, the replacement (Judge Tallman) agreed with Judge Smith, and Judge Smith (over the objection of Judge Reinhardt) vacated the prior opinion wrote a new one (joined by Judge Tallman) holding that X wasn't unconstitutional. In that opinion, Judge Smith defended the changed result by arguing that the collaberative process -- including the evaluation of petitions for rehearing -- "strengthens the final quality of those opinions," and cited Ninth Circuit precedent noting that no decision is final until the conclusion of "the period in which either party may petition for rehearing." Those same principles, as well as General Order 3.2(g), arguably equally apply with respect to Judge Hall.
Don't get me wrong. I'm not saying that Judges Kleinfeld and Silverman did anything wrong. They could have said: "We're deciding the case. Here's our opinion. We attach some contrary thoughts from a friend and former colleague. Consider them if you'd like." There's no rule, after all, about attachments or exhibits. There's similarly no pragmatic rule about how an Article III judge decides. The Ninth Circuit could have drawn a replacement, and that replacement could permissibly say: "I may have my own views. But I've decided to articulate Judge Hall's and vote the way she would. Here's her dissent, which I adopt as my own." That's clearly a decision by an Article III could, and done according to the rules.
So I thought it worth considering whether Judge Kleinfeld and Silverman's decision was the same one that I would have adopted in their shoes.