Does it matter whether and how the Supreme Court corrects its mistakes?
The Court, of course, does make mistakes; we all know that the Justices are infallible only because they are final, and not the other way around.1 It could hardly be otherwise. The Court has a shockingly small support staff, quite unlike the vast bureaucracies that surround the legislative and executive branches of government. It nevertheless produces, each year, a completely new and enormous written body of work; operates under tight and near-absolute (albeit self-imposed) deadlines; and issues almost all of its most important and controversial decisions more-or-less simultaneously during a very narrow period in the late spring, just as it is rushing to recess for the summer so that the Justices can begin their vacations and European teaching gigs. Many of the Court’s lower-profile cases, meanwhile, are hard to understand, are decided unanimously (and hence with less rigorous scrutiny), involve technical subjects and abstruse terminology that are unfamiliar to the Justices, and thus sometimes make the eyes glaze over. It would, accordingly, be astonishing if errors and inapt turns of phrase did not find their way into the Court’s opinions.
It also should come as no great surprise that the Justices — or, in any event, the Court’s staff — invest much energy in correcting those errors. It would be intolerable were the decisions of the nation’s highest court to contain for all time such embarrassing errors as, for example, misspellings of the Justices’ names or manifest misstatements of historical fact; we expect the Justices to know who was President of the United States in 1799.2 Professor Richard Lazarus’s revelatory description of the history of these often-unnoticed errors and associated corrections is fascinating, but it is hard to get too worked up over the precise mechanics by which the Court’s Reporter changes “capitol” to “capital.”3 If that were all there is to the problem of the “non-finality” of the Court’s decisions, it would be easy to understand why Lazarus’s article is the first time in two centuries that the problem has attracted sustained attention.
But then Lazarus recounts a remarkable anecdote.4 When he was a young lawyer working in the Solicitor General’s office at the U.S. Justice Department, the Court decided the significant Clean Water Act (CWA) case International Paper Co. v. Ouellette.5 As originally issued in a slip opinion, Justice Powell’s opinion for the Court in Ouelette contained the line: “The CWA also provides its own remedies, including civil and criminal fines for permit violations, and ‘citizen suits’ that allow individuals (including those from affected States) to compel the Environmental Protection Agency (EPA) to enforce a permit.”6 This language, however, troubled the EPA, as the circuits were then divided on whether citizen suits were in fact available to compel EPA action and the matter had not yet been settled; the EPA, presumably, did not want to be subject to such suits.
Accordingly, Lazarus — who had written the brief for the United States and the EPA as amicus in Ouellette7 — wrote a letter to the Court’s Reporter on behalf of his clients explaining that the availability of citizen suits to compel EPA enforcement had not yet been finally decided and urging deletion of the offensive language from the Court’s opinion. The letter made its way to Justice Powell, who obliged the young Lazarus: the language approving citizen suits to compel EPA action was omitted from the final opinion as it appears in the U.S. Reports.
For someone who practices in the Court, this account is eyebrow-raising, and more than a little disturbing, in two related respects.
The first is what it says about the practice followed — or, more precisely, the failure of the Court to follow any ordinary practice — in making what must be recognized as a substantive change of considerable import to its decision. What the Supreme Court says even in its dicta about the meaning of federal statutes like the CWA effectively has the force of law; more often than not, such statements are slavishly followed by the lower courts. So imagine that something similar to Lazarus’s account had occurred prior to the decision in Ouellette. Suppose, for example, that the Ouellette respondents in their reply brief or at oral argument had urged the Court to declare that the CWA authorizes citizen suits to compel EPA enforcement action — that is, urged the Court to state in its opinion the rule initially articulated by Justice Powell. Is it conceivable that the petitioner or the United States as amicus would have sent an ex parte letter to Justice Powell (or to the Reporter, with the request that he pass the letter on to Justice Powell) urging the Court to reject the respondents’ proposed language? Surely not; such a letter, which would fall outside any of the sorts of communication with the Court about a pending case that are contemplated by the Court’s rules, would have been dismissed as wholly inappropriate, and the author likely admonished for sending it. But that is, in practical effect, what actually happened.
The second notable element of the Ouellette affair is its complete lack of transparency. It appears that the Solicitor General’s proposal to change the Court’s opinion was not served on the parties to the case, that the Court did not provide notice of the change to the parties at the time that it occurred, and that the change was not, in fact, ever publicly announced as a change at all. We know it occurred only because, decades after the fact, Lazarus found an account of the Solicitor General’s letter and of Justice Powell’s reaction in that Justice’s official papers, and because of the happenstance that Lazarus himself was involved in the episode while working as an Assistant to the Solicitor General and subsequently became a leading academic with an interest in Supreme Court practice.8 And even so, the Court’s failure to maintain any regular process for implementing and providing notice of such changes means that the crucial explanatory materials, including the letter of the Solicitor General that prompted and elaborated upon the rationale for the change, are not publicly available.
These oddities and omissions in the Court’s procedures for implementing changes to its opinions, engagingly demonstrated by Lazarus, raise (at least) three questions: (1) How did a Court that spends most of its time stating rules for others come up with such a questionable practice for itself? (2) Do the procedures used by the Court in modifying its decisions matter — which is to say, is the nonfinality problem identified by Lazarus anything more than an interesting curiosity? (3) What, if, anything, should the Court do about correcting its errors?
(1) The first of these questions is the easiest to answer: The haphazard development of the Court’s corrective mechanisms is the product of internal operations that are entirely self-devised, developed in the context of a culture of secrecy. In this, the Court differs in notable respects from the other branches. The Constitution itself dictates at least the outlines of how Congress must operate, and Congress has created rules, like the Administrative Procedure Act, that govern significant operations of the Executive Branch. But there are no similar directives structuring the internal operations of the Court. The Constitution says only that there is to be a Court, offering no guidance at all on how it is to operate (Article III fails even to mention the Chief Justice; the Constitution’s only reference to that person appears in Article I, specifying his or her role, not in running the Court, but in presiding at the impeachment of the President). And although Congress has defined the Court’s jurisdiction, it has made no attempt to specify any aspect of the Court’s internal operations, likely out of concern for treading on the separation of powers.
For its part, the Court’s own internal procedures are largely unwritten, developing in common law fashion and documented, if at all, in internal memoranda that become public only when the papers of a retired or deceased Justice are released. The Court’s published rules principally address the ordinary practice of lawyers appearing before the Court; even an internal matter as fundamental as the Rule of Four, specifying that it takes the votes of four Justices to grant certiorari, does not appear in a statute or formal rule of the Court. And the Court has never reduced to writing internal operating procedures, of the sort issued by most of the federal courts of appeals, that describe how it handles cases internally or modifies decisions after the fact. In this context, where internal developments at the Court take place in a black box up until the moment a decision is issued, it is not surprising that the mechanics of error correction have developed in a slapdash way.
(2) This all may be very interesting, but does it matter? That the Court lacks a transparent and regularized procedure for correcting its typos or changing “insure” to “ensure”9 may reflect badly on the institution, but doesn’t really start one’s blood to boil. And the Court’s lack of transparency means that we literally don’t know how often the more significant, Ouellette-type changes are made. My guess is not very often; the instantly wide circulation and close scrutiny given the Court’s decisions in the SCOTUSblog era mean that someone is likely to notice any such change, at least in the higher-profile cases. We thus can be pretty confident that it would have drawn comment had Chief Justice Roberts, a notable student of the Court’s history, sought to emulate one of his predecessors and, after the fact, added eighteen pages to his opinion in National Federation of Independent Business v. Sebelius.10
But the problem of opinions that change after decision nonetheless warrants attention; as Lazarus shows, it has the potential to cause mischief and embarrassment.11 Consider the circumstance where the Court relies in its opinion on a statement of fact or of historic practice by one of the parties that turns out not to be true. This possibility is not hypothetical. Five years ago, in Nken v. Holder,12 the Court recited and relied upon the Solicitor General’s assurance that the United States has a policy of allowing deportees who successfully challenge their removal from the United States to return to this country.13 That assurance turned out not to be true.14 Upon discovering such an error, could the Solicitor General, or the other party to the litigation, or an interested academic, write a letter to the Court and urge it to modify its opinion?15 Should such a letter be served on the other parties or on amici? Would it matter that the parties might disagree on whether the change, if made, would, or could, undermine the Court’s reasoning, or point to a different outcome altogether?
Or consider the related phenomenon, which has drawn recent academic attention, of amicus filings that make representations of fact falling outside the record of the case.16 The practice isn’t new — it can be traced back more than a century, to the original “Brandeis brief”17 — but such briefs are filed with increasing frequency and their sometimes dubious representations occasionally find their way into the Court’s opinions. If one of these representations subsequently proves to be demonstrably false as a matter of fact, does the Court’s practice allow one of the parties, or amici, or an interested member of the public, to call the error to the Court’s attention? Does it matter if the error is instead discovered by the author of the opinion, or by another Justice? This possibility, too, may not be wholly hypothetical. Judge Posner, for one, has become an advocate of judges using Google to probe the factual background of the cases on which they sit,18 and that practice, if it reaches the Supreme Court (it already may have, for all we know), will make occasional error inevitable.
If a court of appeals committed such an error, the adversely affected party likely would file a petition for rehearing, with a correction typically taking the form of a grant of rehearing and formal issuance of a revised decision. But the Supreme Court’s modern practice is, effectively, never to grant rehearing — a practice so well settled that Supreme Court practitioners reflexively reject any suggestion even to ask for rehearing — and we know from Ouellette that, at least sometimes, such errors are called to the Supreme Court’s attention by less formal means.
This kind of problem was less notable generations ago, when the Court’s opinions were not as widely or quickly disseminated and corrections would have likely overtaken errors shortly after the initial slip opinion arrived in hard copy at the neighborhood law library. But decisions are now scrutinized instantly online across the country at the moment of issuance. This assures that “bench opinions” will be widely read, and that many readers will have strong, instant views on which statements in those opinions contain mistakes that warrant correction. Careful lawyers faced with this situation will be tempted to call the Court’s very helpful Clerk’s office for guidance on what to do — which suggests that the Clerk should be prepared with consistent answers.
(3) So what, if anything, should the Court do about the problem of modifying its opinions in response to perceived error or second thoughts by the authoring Justice? Lazarus offers some sensible proposals, with the degree of procedural protection calibrated according to the nature of the error and the identity of the person calling the mistake to the Court’s attention.19 Perhaps in a generation or two, when the papers of one of the current Justices become available at the Library of Congress, we’ll see whether the Court took any of these suggestions to heart.
Meanwhile, the Justices would do well to focus on the most serious concern that follows from irregularities in communication with the Court and resulting changes in the Court’s decisions: the danger that lack of disclosure will disadvantage uninformed parties and, consequently, distort the Court’s decisionmaking.
One proposal is for the Justices: as Lazarus proposes, the Court should promptly provide to the parties and disclose to the public, in a noticeable and intelligible way, every change that it makes to every opinion after initial publication.20 Perhaps this would be too much information; concededly, there is no burning public demand for a formal announcement every time the Reporter changes “effect” to “affect.” But there can be no doubt that every even arguably substantive change should be publicly announced, and a blanket rule of disclosure would avoid what Lazarus shows could be difficult line-drawing problems.21 It is hard to imagine any serious objection to such a hard-and-fast rule of disclosure, which likely would both lead the Court to think twice about whether proposed changes really are advisable and assure that all parties are aware of changes that affect them — and are able to object to those changes, if only after the fact. Courts of appeals do this now, and the Supreme Court sometimes does, too, with the issuance of errata sheets. Only good can come from making this practice more methodical and comprehensive.
Indeed, the limited nature of a very recent effort by the Court to shine light on correction of an error itself demonstrates the value of a more regularized approach to offering full and complete disclosure. Citing a nonpublic email from the Court’s public information officer to a reporter for the New York Times, Lazarus reveals that the Court alerted selected reporters that Justice Ginsburg had amended a dissent to correct an error and make small stylistic changes.22 That disclosure was a good thing. But the Court did not formally announce or disclose the change by issuing either something labeled an “amended opinion” or an errata sheet, and the disclosure of the correction appears nowhere on the Court’s website, where case documents and materials for the news media (including press releases and media advisories) appear. Consequently, someone who happened not to read press accounts at the time of the disclosure would be unaware that changes had been made and, in any event, would have no way of learning what the changes were short of conducting a word-by-word comparison of the initial decision with the one now available on the Court’s website. There is no reason the Court should make it this difficult to follow its work.23
The other thought is that lawyers proposing any even arguably substantive change to a Court decision should provide counsel for the parties a copy of the corrective communication with the Court.24 This would seem to be good practice in all cases, and surely is when the proposal comes from counsel for a party or amicus — as, for example, it did in Ouellette — who by definition will have a personal stake in the outcome and already will have been in contact with the parties in the course of serving his or her briefs. Doubtless, there is no need for formal service on the parties every time a typo is called to the Reporter’s attention. But the Court’s rules require that every communication with the Court prior to decision be served on the parties; it is difficult to see why a different approach is appropriate for substantive post-decision communications.
Necessarily, much of the Court’s business is veiled in secrecy. The process by which it corrects its errors should not be.
* Charles Rothfeld is a member of Mayer Brown’s Supreme Court and appellate practice and co-director of Yale Law School’s Supreme Court Clinic.