Last week, there were two appellate opinions, one from the Seventh Circuit and one from the Tenth, that are worth some attention. They’re worth discussing together as well, because while only one is really helpful for defendants, both discuss different conceptions of how to argue comity in class actions.

The first is Smentek v. Dart (7th Cir. 2012). Smentek was a class action filed on behalf of Illinois prisoners who had been denied dental care in violation of the due process clause. It was the third of its kind. How did that happen? In the first two class actions, various trial judges of the Northern District of Illinois had denied certification. Not so the third, who granted it. As Judge Posner noted in the Seventh Circuit’s opinion:

We don’t understand why all three cases were not assigned to the same judge. Besides the usual advantages of consolidation, it would have avoided the problem that has precipitated the appeal in this case, because a single judge would not be of different minds about three identical lawsuits.

Given the Supreme Court’s opinion last year in Smith v. Bayer Corp., the defendant could not ask for an injunction preventing any further class actions from being filed, nor could it argue that a new class action was barred by collateral estoppel. But it could, as the Court had directed, ask the new court to, in the interest of judicial comity, deny certification. Foolproof strategy, right?

As it turns out, not in the Seventh Circuit. The trial court did not take comity into account. Nor, according to Judge Posner, should it have.

The version of comity announced in dictum in Smith v. Bayer Corp. is novel. It does not involve the mutual respect of sovereigns or quasi-sovereigns and it does not appear to be limited to cases in which parallel suits are pending in different courts (or before different judges) of the same sovereign. … The Supreme Court’s opinion cites no authority for the extension of the doctrine of comity to mere disagreement between federal judges, and despite the reference to expecting "federal courts to apply principles of comity to each other’s class certification decisions" (emphasis added), the Court seems really to have been thinking about cases involving federal-state comity, of which Smith v. Bayer Corp. was one.

(Emphasis added.)  The primary problem Judge Posner saw was that, to provide the doctrine with any preclusive effect would simply render it collateral estoppel by another name, an avenue that the Supreme Court had already closed off.

True, the effect of the doctrine of comity, when it is successfully invoked, is preclusive. But unlike res judicata, it is a doctrine that does not require but merely permits preclusion, except (as we’re about to see) when it governs choice of forum. The mandatory comity for which the defendants in our case contend is just another name for collateral estoppel.

Judge Posner, no fan of repetitive class actions (remember, he authored several opinions disapproving of it in Thorogood v. Sears, Roebuck & Co.), conceded that this left open a distinct issue of judge-shopping, one that his colleague Judge Easterbrook had identified long ago in In re Bridgestone/Firestone. In Posner’s words:

Not that there isn’t a serious problem of judge shopping in the disordered realm of class action litigation, a problem well illustrated by this case and its two predecessors taken all together. Without a rule of preclusion, class action lawyers can do what the lawyer here (and the lawyer in Thorogood) did: keep bringing identical class actions with new class representatives until they draw a judge who is willing to certify the class.

So, at this point, it appears that the Supreme Court’s admonition to respect previous denials of certification is of persuasive value only, which means that it is helpful only so long as the court is not inclined to certify a class in the first place, which is to say not helpful at all. At least Judge Posner’s opinion spells out the problem in stark terms, of which one might hope other courts will take notice.  

The second case is Winzler v. Toyota Motor Sales U.S.A., Inc. (10th Cir. 2012). In this case, the plaintiff sued for an injunction requiring essentially requiring a recall of Toyota Corollas that allegedly stalled without warning. After she filed the complaint, Toyota moved to dismiss because the plaintiff had not suffered a stalling incident herself, and therefore had not incurred any injury. ("No-injury" class actions are common in products liability.) The trial court granted the motion.  Toyota also noted that a NHTSA investigation was underway. By the time the plaintiff filed her appeal, Toyota had announced a recall, and it had argued to the Tenth Circuit that the recall mooted the plaintiff’s claims.  (This particular formulation of mootness is referred to as "prudential mootness.")

The Tenth Circuit found merit in Toyota’s position. Like Judge Easterbrook did in Aqua Dots, the Tenth Circuit expressed concern about the fact that suing on top of a recall would add transaction costs that would benefit only the plaintiffs’ lawyers.

Our intervention would, as well, surely add new transaction costs for Toyota and perhaps reduce the incentive manufacturers have to initiate recalls (as Toyota did here), all while offering not even a sliver of additional relief for Ms. Winzler and members of the class she seeks to represent. Perhaps the lawyers would benefit if this would-be class action labored on through certification, summary judgment, and beyond. But it’s hard to see how anyone else could.

But the Tenth Circuit was not just concerned that suing when an administrative recall was available would add transaction costs. It also worried that initiating a redundant action through the courts would undermine the comity between branches of government:

To hold otherwise — to allow a case to proceed simply because there happen to be differences between the remedial process a coordinate branch has selected and those we might choose – would not only require us to ignore the reality that there’s often no one single right way to go about providing equitable relief to an injured party. It would also require us to ignore the reality that there are nearly always (if not always) some differences between Executive, Legislative and Judicial remedial procedures given how differently the three branches operate: by regulation, legislation, and decree. To presume deficiency from difference would no doubt go a long way, as well, toward spelling the end of prudential mootness doctrine and the comity it is supposed to afford our coordinate branches.

At that point, it affirmed the lower court’s dismissal of the case.

So, what can we draw from these two different treatments of the concept of "comity"? Comity as a standalone concept is not a particularly helpful one: as Judge Posner points out, the doctrine of comity really only applies to separate sovereigns. However, comity also has a looser meaning, one that asks courts to respect other governmental bodies, even if they stem from the same sovereign power. And here, comity is at the very least a useful rhetorical device. It reminds the court that it is not the only available means of relief; and that some litigation is simply a waste of time and money. Of course, since comity is not mandatory, what a court chooses to do with that information remains entirely up to it.