Case-or-Controversy Limits and Qualified Immunity

by Michael Dorf

My latest Verdict column discusses the Supreme Court's per curiam decision on Monday in Taylor v. Barkes. The decision reverses a Third Circuit ruling that permitted the widow of a man who killed himself in jail to sue the warden and the head of the state prison administration for failure to adequately train and supervise hired medical staff in evaluating detainees for suicide risk. In light of a conflict in the lower courts and the absence of any SCOTUS precedent recognizing a right to be subject to an adequate suicide prevention protocol, the Court in Barkes holds that the defendants were entitled to qualified immunity (QI).

Without disagreeing with the result per se, I use the Barkes per curiam as an opportunity to criticize various aspects of the Court's QI jurisprudence, including its 2009 decision in Pearson v. Callahan, which overruled the "rigid order of battle" of Saucier v. Katz. In Saucier, the Court had said that when faced with a QI defense, a court ought to first consider whether the plaintiff's claim, if proven, would amount to a constitutional violation. Only if so should the court then consider QI. The rationale for Saucier was that by considering QI w/o first addresing the substantive merits, the courts stunt the development of constitutional law. A decision holding only that a right was not clearly established means that the putative right will be no more (or less) clearly established in the next case. The column questions the reasons given in Pearson for thinking that this phenomenon will be adequately addressed by claims arising in other contexts.

One argument I do not address in the column--but that one sometimes hears in discussions of Saucier--is that the Saucier procedure violates the case-or-controversy norm of Article III. The argument goes like this: If a court is ultimately going to decide a case in favor of the defendant on QI grounds, then the preliminary determination that the plaintiff's constitutional rights were violated is advisory; accordingly, a court should consider QI first; if there is QI, game over; and if there isn't QI because the right at issue was clearly established, then a fortiori there is a constitutional right; thus there is no need to consider the substantive right question at all, and a court shouldn't risk deciding it as an advisory opinion.

I think that the foregoing argument is overstated. Part of the argument for the approach in Saucier is that it makes more sense logically to address the substantive right question first, even if the court ultimately rejects the claim on QI grounds, because addressing the substantive right question will frame the QI inquiry. Moreover, until the court actually finds QI, the substantive right question could have an impact on the outcome, and is therefore not advisory. A ruling doesn't become advisory post hoc. It's only advisory if it is known that it can have no effect on the outcome when the court is considering the question.

Nonetheless, there is a narrower version of the case-or-controversy objection that has some greater force. This narrower version acknowledges that a federal court has the power to consider the substantive right question first but then objects that a ruling in favor of the substantive right will end up being mere dicta if the court next finds that the defendant is entitled to QI. Although there are numerous approaches to distinguishing between holdings and dicta, a standard one defines dicta as a statement that does not state a proposition that is in any way necessary to the holding. If the holding is that the defendant has QI, then a statement that the defendant nonetheless violated the plaintiff's rights surely qualifies as dicta under this standard.

I should be clear that this "it's only dicta" objection is not made by the SCOTUS in Pearson, which leaves district court judges with discretion to address the substantive merits of a claim before the QI issue. So either the SCOTUS thinks that the substantive merits ruling in these circumstances is not dicta or that, if it is dicta, it's justified dicta.

To my mind, each of those positions is plausible. Another definition of dicta is that it's a kind of gratuitous aside. This approach narrows the category of dicta and correspondingly expands the category of holding. A very limited view of the function of the courts would say that the broader approach to holdings threatens case-or-controversy/separation-of-powers values because it gives courts lawmaking powers that go beyond the scope of the particular case. While there is something to that objection, taken too seriously it swallows itself. After all, if courts restrict their statements to those truly necessary to the outcome, they cannot make law at all. They can only say "plaintiff wins" or "defendant wins."

The whole idea of QI strikes me as cutting against the narrow view of courts' lawmaking power. As I discuss in the column, QI rests on the premise that government officials ought not to have to answer in damages for guessing wrong about the lawfulness of their conduct. But that particular bit of legal realism about the limits of the law's determinacy accepts that judicial decisions either do or do not clearly establish legal rules and standards over time. Put differently, QI itself is a response to the fact of judicial lawmaking. It would thus be ironic to limit a court's capacity to clarify the law in the course of a QI case on the ground that judicial lawmaking is illegitimate.