A few weeks ago, Professor Richard Lazarus posted a fascinating and much-discussed draft article documenting the Supreme Court’s practice of revising its opinions after their initial publication. These often overlooked revisions, Lazarus shows, can extend to important points of law. Partly for that reason, Lazarus proposes various reforms to promote transparency, such as public notice of any post-publication changes.
In reading Lazarus’s paper, I found myself wondering whether there are any limits on the Court’s revision power. I think that there are. In particular, the Court probably has authority to modify the substance of a precedential opinion only when the relevant case is before the Court. Revisions at other times, such as long after the judgment and mandate have issued, seem like advisory opinions.
There are lots of situations where post-publication revisions don’t implicate the Court’s decision-making authority. These include changes to dissents and to at least most concurrences (assuming they don’t have precedential force under the Marks rule). Other unobjectionable changes arise pursuant to motions for reconsideration, which allow the Court to exercise its judicial power while still resolving the case before it. Finally, changes of a non-substantive nature don’t seem problematic. For instance, Lazarus notes a revision that added the “t” to Justice Stevens’s name. That kind of typographical error clearly has no effect on the opinion’s basis or precedential force.
For the opposite extreme, imagine that the Court purported to revise a very old decision. Let’s say, for instance, that a future Court decided to retroactively revise Brown v. Board of Education in two ways: first, to eliminate the famous and famously controversial footnote to psychological studies indicating that segregation had adverse effects on children; and, second, to add the well-known account of Brown set out in volume 1 of Bruce Ackerman’s “We the People” series, which viewed Brown as derivative of the New Deal “constitutional moment.” One of these changes is a subtraction, while the other is an addition. But I would be surprised if anyone viewed either of these revisions as legitimate, even if the changes were accompanied by ample public notice and opportunity to comment. The Court’s decisions aren’t perpetually ongoing works in progress, even though they can be modified through other decisions consistent with judicial practice.
Imposing a temporal limit on the Court’s revision power can be justified in many ways. For one thing, there is an important formal objection to excessive use of the revision power. If a relevant justiciable case is necessary to create a particular precedent under Article III, then one would expect that the same standard of justiciability should also be necessary to revise that precedent. This approach ensures that the Court’s published statements are always linked to concrete judgments. By contrast, late revisions are necessarily post hoc rationalizations — much like the journal articles whose views the Court might be adopting. Even if a later-arising justification were in some sense better than the actual justification in terms of its cogency or clarity, the original justification would still be uniquely valuable as a window into the Court’s actual, contemporaneous decision-making process. In a sense, preserving original judicial opinions, subject to reconsideration in later, separate precedents, is akin to preserving the original Constitution, with amendments reflected at the end of the document. Of course, there are also pragmatic issues at stake, as major or long-delayed revisions could erode the public’s ability to rely on precedent.
In light of the above, I suspect that the basic dividing line between permissible and problematic revisions should be the Court’s authority to act in the relevant case. In the normal course, under Rule 45 the Court is divested of its decision-making authority over a matter when the mandate issues (in state cases), or when a certified copy of the judgment is transmitted to the lower federal court. This generally applicable deadline would run on the order of weeks or, if rehearing is requested, months from the original date of an opinion’s publication. By contrast, the Court currently appears to embrace a multi-year window for revision before an opinion’s publication in the U.S. Reports. In addition to the transparency problems that Lazarus identifies, long-delayed efforts to revise precedent may implicate or even exceed the Court’s authority.