Tuesday, February 10, 2015
In a post last Monday on Prawfsblawg, entitled, On Not Creating Precedent in Plumley v. Austin, Richard M. Re asks, "what’s so wrong with deliberately declining to create precedent?" By his answer, an implied "nothing" because "[d]oing so conserves scarce resources and reduces the risk of mistaken or sloppy precedent," he seems to be asking, "what's the harm?"
There are a couple other bases for finding the practice "wrong," such as whether the practice is legitimate, constitutional, or just. But first, what's the harm in treating some circuit decisions as non-precedential? This is something I discuss in my works on the topic, especially: Overturning the Last Stone: The Final Step in Returning Precedential Status to All Opinions, 10 J. App. Prac. & Process 61 (2009) and Draining the Morass: Ending the Jurisprudentially Unsound Unpublication System, 92 Marq. L. Rev. 685 (2009).
First, deliberately declining to create precedent creates fewer precedents. Fewer precedents means a less definite law. At least since the time of Lord Coke, the law has been viewed as refined by renewed applications. With each new decision, the law is broadened, narrowed, or simply reaffirmed. The common law treats each case as binding but is also concerned about the accrual of such cases and the varying facts to which the rule is applied. This the understanding of precedent of Coke, Blackstone, Kent, Marshall, Story, and Llewellyn. It's how the common law, in principle, works. Never before in common law history has a court been able, at the time of decision, to remove its holding from the body of precedent. And no matter how the court phrases its opinion, it has ultimately been up to the later court to decide whether and how earlier opinions applied.
I am partial to Karl Llewellyn's explanation: "We have discovered that rules alone, mere forms of words, are worthless. We have learned that the concrete instance, the heaping of concrete instances, the present, vital memory of a multitude of concrete instances, is necessary in order to make any general proposition, be it rule of law or any other, mean anything at all." Karl Llewellyn, The Bramble Bush, 66-69 (1930).
The problem is not that there are too many precedents but that there are too few. Judge Posner wrote as much in The Federal Courts: Challenge and Reform, and his experience is echoed in the experiences of the federal judiciary. In a 1998 survey of federal district judges, about a third identified some area of circuit law as inconsistent or difficult to know on account of lack of binding circuit decisions on point. But even more telling than what judges say is what they do. The survey also revealed that nearly two-thirds of lawyers surveyed reviewed unpublishd opinions either generally within their practice area or in researching specific cases. During the citation ban era (1974-2006), courts and litigants frequently cited to unpublished and allegedly non-precedential opinions even in violation of the ban. They were, as Lord Coke might have described it, looking for greater refinement in the law that only comes by seeing it applied. Or as Llewellyn might have said, they saw the published, precedential grains of sand, but they wanted to see the heaps. The citation ban finally ended because it ran counter to a basic understanding of precedent shared by American lawyers and judges alike: each case has value in determining the scope of the law.
More applications of the principles of law to facts, such that those principles are tested and refined, improves our understanding of those principles and gives greater certainty to those seeking to conform their conduct to them. "Mistaken or sloppy precedent" can be corrected by more judicial oversight to their drafting, or should that fail, by the normal processes of the court. While conserving limited resources is important, expediency should not be our highest value. The federal judiciary, a co-equal third branch of our government is allocated a mere two-tenths of one percent of the total federal expenditures. Instead of asking our courts to do with less, we should give them the funds to do more.
Second, issuing some decisions as non-precedential creates the potential for blatantly conflicting published and unpublished opinions. A court may decide in favor of a party today but next year, on the exact same issue, decide exactly the opposite. If the earlier decision is unpublished, the later panel need not acknowledge the earlier decision or give a reason for the change. This was the case in a pair of cases in which the Dallas Area Rapid Transit authority (“DART”) received diametrically opposed decisions from the Fifth Circuit without explanation in a span of just three years. In 1999, a federal district court in the Fifth Circuit held that, “DART is a political subdivision of the state of Texas, and is therefore immune from suit under the Eleventh Amendment," which the Fifth Circuit affirmed without comment in an unpublished opinion. Anderson v. Dallas Area Rapid Transit, No. CA3:97-CV-1834-BC, 1998 U.S. App. LEXIS, 15493 (N.D. Tex. Sept. 29, 1998) aff’d Anderson v. Dallas Area Rapid Transit, 180 F.3d 265, (5th Cir. 1999) (per curiam) (unpublished), cert. denied 529 U.S. 1062 (1999).
In Anderson, and two other unpublished opinions, the Fifth Circuit held that DART was a governmental unit or instrumentality of the State of Texas entitled to qualified immunity. The law on this point seemed so clear that in Williams v. DART, the district court felt this point was "firmly established." The Fifth Circuit disagreed and rejected DART's immunity claim dismissing the unpublished opinions as "neither binding nor persuasive," but failing to give any reason for the different treatment. Williams v. Dallas Area Rapid Transit, 256 F.3d 260, 261 (5th Cir. (Tex.) 2001). This decision drew a strong dissent noting that this kind of unreasoned about-face exposed a flaw in the concept of non-precedential opinions.
A conflict like this between two precedential opinions would be resolved by the second panel distinguishing the present matter from the prior one, or if that proved impossible, by an open declaration of conflict followed by a resolution by the court en banc. Which leads to a third category of harm non-precedential opinions cause.
Third, issuing some decisions as non-precedential increases the likelihood of intra-circuit conflict. Such conflict was especially acute in the citation ban era, because a litigant perceiving a conflict in a circuit's unpublished opinions was prohibited by rule from raising it with the court. For example, in the wake of the U.S. Supreme Court's ruling in Almendarez-Torres v. United States, 523 U.S. 224, 118 S. Ct. 1219 (1998), an ambiguity arose about how to treat a defendant convicted of illegal entry following deportation. Over a two-and-a-half-year period, twenty Ninth Circuit panels ruled on this issue and split three different ways (most remanding for resentencing, some remanding for amendment of the original judgment, and a few foisting the responsibility for determining the proper course of action on the district court). The split continued for over two years, with identically situated defendants receiving different answers from the Circuit. The ongoing intra-circuit conflict was revealed only when a panel in United States v. Rivera-Sanchez, 222 F.3d 1057 (9th Cir. 2000) ordered a litigant to violate the Circuit's non-citation rule and provide a list of these unpublished opinions.
A circuit that does not view its unpublished opinions as binding can simply ignore those decisions for purposes of whether to hear an issue en banc. Yet, the unpublished opinion may be cited for persuasive effect (in all circuits since 2007), which merely increases the chance of creating the separate, conflicting lines of authority as in the Riveria-Sanchez scenario.
Fourth, inter-circuit conflict become more likely, too. In much the same way that intra-circuit conflict can arise undetected or unacknowledged within a circuit, such conflicts can arise between circuits. During the citation ban era, such conflicts were effectively hidden, because citation bans prevented their being raised. But even now, if the unpublished opinion is not treated as establishing the law of the circuit, it can be disregarded within its own circuit and by the other circuits. The Supreme Court takes only a tiny fraction of the cases seeking review each year. Just as with en banc panels, a case that does not establish the law of the circuit is unlikely to be the basis of an apparent conflict even if the conflict it creates is real. In that way, a conflict can exist indefinitely in a manner much like that described in Rivera-Sanchez.
Fifth, declaring some opinions non-precedential allows them to evade Supreme Court Review. As noted above, one line of authority, if present only in unpublished opinions can obscure or deemphasize the nature of the conflict. Two Justices believed that was the case in Waller v. U.S., where Justices White and O'Connor dissented from denial of cert noting that a circuit split existed if one took into account unpublished opinions. 504 U.S. 962, 964-65, 112 S. Ct. 2321 (1992) (White J. and O’Connor J., dissenting) (Mem); see also Hyman v. Rickman, 446 U.S. 989, 990-92 (1980) (Blackmun, Brennan, and Marshall, J., dissenting) (Mem) (dissenting from denial of certiorari on the grounds that the unpublished circuit opinion was in conflict with other circuits on the issue of right to appointed counsel). While the conflict was sufficient to catch individual Justices' attention, it was not sufficient to prompt Supreme Court review, similar to the result in Plumley v. Austin.
Supreme Court review is also less likely due to the signal an unpublished opinion sends. A circuit’s decision not to publish a given decision signals that that decision is routine, even when it is not. For example in United States v. Edge Broad. Co., the Fourth Circuit declared a federal statute limiting lottery advertising unconstitutional in an unpublished opinion. 956 F.2d 263 (per curiam) (4th Cir. 1993). In its reversal of that decision, the Supreme Court expressed surprise and dismay that the Circuit Court could perceive such a ruling as unworthy of publication. 509 U.S. 418, 425 n.3 (1993) (“We deem it remarkable and unusual that although the Court of Appeals affirmed a judgment that an Act of Congress was unconstitutional as applied, the court found it appropriate to announce its judgment in an unpublished per curiam opinion.”)
The hiding of cases from Supreme Court review also occurs because unpublished cases tend to create a less thorough record, which itself discourages Supreme Court review. For example, in County of Los Angeles v. Kling, the Supreme Court granted cert and issued a summary reversal on a case the Ninth Circuit had decided in a brief, unpublished, non-citeable opinion. 474 U.S. 936, 937-39 (1985). Justice Marshall dissented calling the Ninth Circuit’s practice “plainly wrong” and noting, "the Court of Appeals would have been well advised to discuss the record in greater depth. One reason it failed to do so is that the members of the panel decided that the issues presented by this case did not warrant discussion in a published opinion that could be 'cited to or by the courts of this circuit, save as provided by Rule 21(c).' That decision not to publish the opinion or permit it to be cited-like the decision to promulgate a rule spawning a body of secret law-was plainly wrong."
Justice Marshall continued by chastising the Court for engaging in the same type of shortcut decision making: "The brevity of analysis in the Court of Appeals' unpublished, noncitable opinion, however, does not justify the Court's summary reversal….For, like a court of appeals that issues an opinion that may not be printed or cited, this Court then engages in decision-making without the discipline and accountability that the preparation of opinions requires."
Even when both parties agree that a Circuit decision makes new law, the status of a decision as unpublished can discourage Supreme Court review. In Family Fare, Inc. v. NLRB, both parties agreed that the Sixth Circuit had departed from its previous law in an unpublished opinion. 2006 U.S. Briefs 1536 cert. denied Family Fare, Inc. v. NLRB, 127 S. Ct. 2991 (2007). NLRB liked the change and sought publication or a Supreme Court affirmance to solidify the new interpretation. Family Fare disliked the change and viewed it as exactly the kind of surreptitious change in the law of the circuit that Justice Thomas alludes to in Plumley. Ultimately, The Supreme Court denied cert, probably in significant part because as an unpublished opinion, it was not the formally law of the circuit and did not truly represent a shift in the law. Yet, Family Fare was treated differently than prior litigants, and NLRB likely relied on the decision in future cases to show that the law had changed.
Sixth, creating an opinion on which no one can rely (and which for years no one could even cite) is an invitation to poor reasoning or even strategic, result-based reasoning. Justice Stevens expressed "that occasionally judges will use the unpublished opinion as a device to reach a decision that might be a little hard to justify." Jeffrey Cole & Elaine E. Bucko, A Life Well Lived: An Interview with Justice John Paul Stevens, 32 No. 3 Litigation 8, 67 (2006).
This concern was also expressed by the late-Judge Richard Arnold and quite directly by Judge Wald of the D.C. Circuit: "I have seen judges purposely compromise on an unpublished decision incorporating an agreed-upon result in order to avoid a time-consuming public debate about what law controls. I have even seen wily would be dissenters go along with a result they do not like so long as it is not elevated to a precedent." The Rhetoric of Results and the Results of Rhetoric: Judicial Writings, 62 U. Chi. L. Rev. 1371, 1374 (1995).
A study of asylum cases in one circuit and found considerable strategic decision making surrounding the outcomes of cases and the publication of opinions: "voting and publication are, for some judges, strategically intertwined: for example, judges may be prepared to acquiesce in decisions that run contrary to their own preferences, and to vote with the majority, as long as the decision remains unpublished, but can be driven to dissent if the majority insists upon publication" David S. Law, Strategic Judicial Lawmaking: Ideology, Publication, and Asylum Law in the Ninth Circuit, 73 U. Cinn. L. Rev. 117 (2005).
Finally, the system of unpublished, non-precedential opinions is harmful to both the courts and the litigants before them. It's harmful to the courts, which have been drawn into this very unjudicial exercise of prospectively dividing "worthy" cases from "unworthy" ones. For hundreds of years, a court was expected to abide by, or explain the difference from, a prior case, and a court knew that its decision created a similar obligation on later courts. Now, unmoored from that, they are engaged in a very different process. As the recent article by Adam Liptak suggests, the public concern with unpublished opinions is that a court can issue one-off rulings that it need not every follow again.
It also harms litigants, who look at prior adjudications in the form of unpublished opinions but have no assurance that they will be treated the same or that any explanation will be given for the difference. And often they are not. Individual litigants like those in the cases mentioned above and all the many similar cases they represent, have not been treated equitably or according the system most people believe exists.
But as noted at the outset of this post, these harms are the middle ground problems with non-precedential opinions. One could accept the practice in principle and have serious concerns with the manner in which it is carried out. Or, more deeply, it's fair to question what authority the federal circuits have for ex ante precedent-stripping and whether that practice is constitutional or just. But those will have to wait for other excessively long posts.