The Federal Circuit held that a voluntary dismissal is not an appealable final decision. But it also held that this lack of a final decision means the case must go on, despite the plaintiff’s voluntary dismissal.
May 9, 2019
Last year, I wrote that the Supreme Court’s decision in Microsoft Corp. v. Baker appeared to hold that the district court proceedings in a case never ended:
The plaintiffs in Baker voluntarily dismissed all of their claims with prejudice, and the district court had nothing else to do. But according to the [Supreme] Court, the voluntary dismissal in Baker was not a “final decision” and thus could not be appealed. A final decision is normally defined as one that marks the end of district court proceedings, leaving nothing else for the district court to do but enforce the judgment. That standard definition suggests that the decision in Baker was final. The Supreme Court said, however, that it wasn’t. So it would seem that the district court proceedings in Baker were interminable.
This was an intentionally silly interpretation of Baker, used to illustrate the problems that stem from building most of the law of federal appellate jurisdiction atop judicial interpretations of the term “final decision.” But now it seems that a court has made Baker’s potential silliness a reality.
Earlier this year, in Princeton Digital Image Corp. v. Office Depot Inc. (Slip opinion, Google Scholar, Westlaw), the Federal Circuit held that a party’s voluntary dismissal of its claim was not an appealable final decision. Simplifying a bit, Princeton Digital involved a breach of contract claim by Adobe Inc. against Princeton Digital Image Corp., which owned a patent that “relates to methods for encoding image data and allegedly covers the encoding of digital images in the JPEG file format.” Adobe licensed the patent from Princeton Digital. And as part of the agreement, Princeton Digital promised not to sue Adobe or its customers for using Adobe products.
Princeton Digital nevertheless sued several Adobe customers, requiring Adobe to intervene in the suits and defend its customers. Adobe also brought a breach of contract claim against Princeton Digital in which Adobe sought to recover the fees it incurred in defending its customers and its fees in bringing the breach-of-contract action itself. But before trial, the district court ruled that Adobe could recover only those fees incurred in defending its customers; it could not recover fees for bringing the breach-of-contract action. Although Adobe had incurred some fees in defending its customers, it apparently believed that those fees alone were not worth proceeding further. So Adobe asked the district court enter judgment against it. The district court did so, and Adobe sought to appeal (among other things) the district court’s decision on which damages were recoverable.
Relying on Baker, the Federal Circuit concluded that the Adobe could not appeal after voluntarily dismissing its claim. The court first noted that Baker applied outside of the class-action context. It also determined that the district court’s decision on fees did not foreclose Adobe from prevailing on its breach-of-contract claim. Granted, the damages might have been less than Adobe wanted. But the decision did not definitively resolve any element of Adobe’s cause of action against it. And “unless the district court has conclusively determined . . . that the plaintiff has failed to satisfy a required element of the cause of action, a voluntarily dismissal lacks finality.” The court accordingly concluded that the district court’s fees decision was not a final one and the court lacked jurisdiction over Adobe’s appeal.
(Sidenote: Baker interpreted 28 U.S.C. § 1291’s grant of appellate jurisdiction over “final decisions” of the district courts. But because Princeton Digital involved a suit relating to patents, the potential source of jurisdiction was 28 U.S.C. § 1295(a)(1). That statute gives the Federal Circuit exclusive jurisdiction to review “final decision[s]” in actions involving patents.)
But then the Federal Circuit did something odd. It said that “[b]ecause the purported final judgment is ineffective, the district court must treat the case as though final judgment had never been entered.” That is, the district court must pretend that Adobe did not voluntarily dismiss its claim. The voluntary dismissal was vacated, even though there was nothing wrong with it (except that it did not allow Adobe to appeal). The court of appeals accordingly instructed the district court to decide the merits of Adobe’s breach-of-contract claim and determine what damages (if any) Adobe is entitled to. “In short, the case must continue until there is a final disposition of the breach claim, at which point there can be an appeal.”
This was a silly outcome.
Granted, this outcome was implied by Baker: if a “final decision” marks the end of district court proceedings, and if a plaintiff’s voluntary dismissal of its claims is not a “final decision,” then a voluntary dismissal does not mark the end of district court proceedings.
But that twisted reasoning comes from the awkward way in which courts have built most of the law of federal appellate jurisdiction atop interpretations of the term “final decisions.” (I discussed this extensively in the above-quoted article.) Courts have used that term to create a variety of rules governing appellate jurisdiction. And sometimes those rules give meanings to the term that are less than intuitive.
Baker is one example of this. The outcome in Baker was undoubtedly correct—the plaintiffs were trying to circumvent Rule 23(f) and manufacture the appeal of an order that they should not have been able to appeal. But the Court’s reasoning was odd. As Justice Thomas pointed out in his concurrence, district court proceedings in Baker were over. The court had entered an order dismissing all of the plaintiffs’ claims with prejudice. So all issues had been resolved, and there was nothing left for the district court to do. Under any straightforward understanding of the term, Baker reached a “final decision.” But the Court (rightly) thought that the decision at issue should not be appealable. To reach that outcome, the Court gave a new meaning to § 1291: a decision that marks the end of district court proceedings is final, unless that decision involves an attempt to circumvent Rule 23(f). The Court thus defined a “final decision” in a way that prevents parties like the plaintiffs in Baker from manufacturing an appealable decision.
The Princeton Digital court took this reasoning one step further. And in doing so, it stepped into the trap set by Baker by conflating two meanings of the term “final decision”: first as a term of art in the appellate-jurisdiction context, and second in the more everyday sense of describing the end of district court proceedings. Consistent with Baker, the Federal Circuit held that the voluntary dismissal was not appealable because it was not a “final decision.” But because the voluntary dismissal was not a “final decision,” the court concluded that district court proceedings were not over. The case must go on, the court thought, despite Adobe’s voluntarily dismissing its claim. The Federal Circuit never really explained how that voluntary dismissal was vacated. But it was.
Princeton Digital has now petitioned for cert. The case presents a great opportunity for the Supreme Court to re-explain Baker.
If the Court does so, a better way to approach these cases is to fall back on the general rule that winners don’t appeal. The plaintiffs in Baker and Princeton Digital did not lose on the merits. They instead decided that, due to an interlocutory decision (denying class certification in Baker, limiting recoverable fees in Princeton Digital), proceeding with a case was no longer worth their time and money. And these plaintiffs got what they asked for—a voluntary dismissal of their claims, with prejudice. In this sense, they did not lose. And as a general matter, only losers get to appeal.
This isn’t to say that every voluntary dismissal cannot be appealed. At oral argument in Baker, some justices expressed some hesitancy to adopt an absolute rule against appealing voluntary dismissals (the conversation starts on page 10 of the transcript). And an absolute rule isn’t necessary. If a district court makes an interlocutory decision that necessarily resolves a plaintiff’s claim—that is, holds that a plaintiff cannot prove an essential element of its cause of action—then a subsequent voluntary dismissal of that claim should be appealable.
Say, for example, that a district court ruled in limine that all of the plaintiff’s evidence on causation was inadmissible. At that point, the plaintiff cannot prove causation and thus cannot win. Were the defendant to then move for summary judgment, the motion would be granted and the plaintiff could appeal. By voluntarily dismissing the claim, the plaintiff saves the defendant the pro forma step of moving for summary judgment. Voluntary dismissal in these circumstances reaches the same outcome in fewer, simpler steps. And just as if the defendant had moved for summary judgment, the plaintiff in these circumstances should be able to appeal despite the voluntary dismissal.
The Court should accordingly grant cert in Princeton Digital and hold that unless a district court’s interlocutory decision necessarily resolves a claim against a plaintiff, the plaintiff cannot voluntarily dismiss the claim and then seek reversal of an interlocutory decision on appeal. And although these voluntary dismissals are not appealable, they still marks the resolution of the dismissed claim and should not be undone after the plaintiff learns that it cannot appeal.
Links
- Princeton Digital Image Corp. v. Office Depot Inc., 913 F.3d 1342 (Fed. Cir. 2019): Slip opinion, Google Scholar, Westlaw
- Cert petition in Princeton Digital Image Corp. v. Adobe Systems Inc., No. 18-1343: Supreme Court Docket, Petition for cert