The Court was back in action Monday (as am I), hearing argument in two cases, one of which, a dispute over an EEOC regulation's limitations period for federal employees' constructive discharge actions, is possibly the easiest case of the term, and another of which presents two interesting questions of federal criminal appellate procedure, the first being whether sufficiency-of-the-evidence review must be conducted under the rubric of erroneously demanding jury instructions, or under the actual elements of the statute charged, and the second being whether a statute-of-limitations defense not asserted at trial can be raised on direct appeal.
The first of these cases, Green v. Brennan, involves constructive discharge actions, which are odd ducks; they turn on a legal fiction that an employee who suffers an onerously hostile work environment, on the basis of membership in some protected class, and resigns as a result, has been constructively terminated on the basis of protected-class membership. Since the constructive discharge cause of action doesn't accrue, obviously, until resignation - until then there's only a hostile work environment claim - it would make sense that the limitations period to bring such a claim run from resignation. Several circuits, however, have read the text of the applicable EEOC regulation (which is only addressed to federal employees) to start the clock on the date of the last discriminatory resignation-inducing act the employee complains about. That regulation instructs federal employees that they must contact an Equal Employment Opportunity counselor within 45 days "of the date of the matter alleged to be discriminatory" (after which another period runs to sue). Ignoring the fiction in constructive-discharge actions that the resignation is really a discriminatory termination, some circuits read "matter alleged to be discriminatory" to refer, unambiguously, to some real-world employer act, trumping the presumption that limitations periods normally don't begin to run pre-accrual.
The government, none too enamored of this theory, instead defends the decision below on the ground that the plaintiff-employee sought conciliation well over 45 days after he gave notice of his resignation, and argues that the limitations period runs from the date of notice. Amy Howe, at Scotusblog, writes that the plaintiff disagrees with the government's legal position and takes the position that the period runs from the employee's last day of work. As far as I can tell, that's not quite right; both in his briefs and at oral argument, the plaintiff has happily conceded that the period runs from notice of resignation, if in fact there is clear and definitive notice to that effect. He and the government only disagree on the factual question of whether notice of that kind occurred here. That left a Court-appointed amicus to defend the theory of the decision below, and here I agree with Howe that the amicus found, at best, only one taker, Justice Scalia.
I would only add that it is somewhat surprising that, in staving off the amicus's argument for a contrary interpretation of the EEOC regulation than that which the government advances, the government didn't ask for deference to the EEOC's interpretation of the regulation, which I assume accords with the SG's position in this case. Instead, the case was briefed and argued as if it were about a statute as to which no deference doctrine applied. Such reticence on this obvious point may be borne out of the SG's fear that the Court wants to overrule Auer and Seminole Rock (the canonical cases on deference to agency interpretations of their own regulations).
Musacchio v. United States is an altogether more interesting case, one which raises, to my mind at least, fairly deep questions about the purpose of sufficiency-of-the-evidence review. As to the first question in the case, federal law makes it a crime to access a computer without authorization, or exceed authorized access, to certain computers and obtain certain information from them. The difference between accessing without authorization and exceeding authorized access is generally thought to be that the latter requires some excess of actually authorized access, while the former requires the opposite - that there be no authorization at all. Of course, one needn't (and couldn't) do both to commit the crime, but that's just what Musacchio's jury instructions - unobjected to by the government - said. There possibly wasn't sufficient evidence to convict Musacchio of conspiracy under those erroneous instructions, because of (according to Musacchio) a lack of evidence to convict him of encouraging others to exceed authorized access. There was, however, as Musacchio never disputed, sufficient evidence to convict him of encouraging others to access computers without authorization, and on that ground the Fifth Circuit rejected Musacchio's sufficiency challenge.
It's difficult to see, at least in this case of an erroneously instructed extra element, why sufficiency review must make do with Musacchio's jury instructions. As the jury found that Musacchio encouraged his co-conspirators to both access without authorization and exceed authorized access, and as it found the former with sufficient evidence, it found, with sufficient evidence, the necessary elements of the crime. Why should it matter that there was insufficient evidence to find an erroneously instructed extra element? Justices Scalia, Kennedy and Breyer all pressed Musacchio's counsel, Erik Jaffe, on this point (while Justices Alito, Ginsburg, and Sotomayor expressed varying degrees of skepticism of his sufficiency argument on other grounds).
Here, Jaffe's only real answer was that a jury which irrationally found one element may have decided the whole case in an irrational or careless manner, such that one should doubt the reliability of the jury finding on the elements it need did to find. Bracketing, for the moment, whether sufficiency review has anything at all to do with the reliability of an individual jury's findings, it's pretty clear that sufficiency review doesn't inquire so searchingly into the reliability of jury findings as to question the reliability of objectively reasonable findings once one jury finding has been shown to be objectively unreasonable. For example, Jaffe conceded, on questioning by Justice Alito, that an appellate finding of insufficient evidence under one count of an indictment wouldn't call into question a jury's findings of guilt under other counts, even though insufficiency as to one count would raise the same sort of red flags about reliability that insufficiency as to one erroneously charged element raises in Musacchio.
But though the Court clearly sees Musacchio as a poor case in which to treat erroneous instructions as law of the case for purposes of sufficiency review, it's slightly less clear to me that it will hold erroneous instructions should never be law of the case in sufficiency review. The government takes the absolute position that sufficiency review should always be conducted under the rubric of the actual elements of the crime of conviction, somewhat counterintuitively theorizing that sufficiency review really has nothing to do with the reasonableness or reliability of a given jury's findings at all, but rather, solely ensures that the government introduce affirmative evidence on the basis of which a hypothetical jury could find guilt beyond a reasonable doubt. On that theory, errors in instructions should never enter into sufficiency review because sufficiency review isn't about what the real jury did, but just about what evidence of the offense the government adduced.
Ultimately I think the government is right. Suppose that, instead of erroneously requiring that the jury find two independently sufficient elements, the instructions in Musacchio instructed the jury only on exceeding authorized access, omitting the element of which there actually was evidence. If the jury returns a conviction on insufficient evidence of exceeding authorized access, should the conviction be upheld under sufficiency review because the offense may be satisfied by access without authorization, and because sufficient evidence of access without authorization was adduced at trial? One might think not. After all, in that scenario, not only were the jury's findings unreasonable, the jury never actually found an element of Musacchio's offense. That said, while such a conviction shouldn't survive appellate review generally, it should, I think, survive sufficiency review. The real error in such a conviction is an instructional error, not insufficient evidence; it makes no sense for Musacchio to be immune from retrial because there was insufficient evidence to prove an element that shouldn't have been instructed to the jury. Even though the bad instructions may well not have prejudiced Musacchio, inasmuch as there was better evidence to convict under correct instructions than under the bad ones, they do call into question the reliability of the result and entitle Musacchio to a new trial.
Musacchio also is a case about whether a defendant can raise a statute of limitations defense on appeal that he forfeited at trial. Here, Musacchio is also likely to lose. The trouble, of course, with his position is that statutes of limitation have always been understood as affirmative defenses; thus, not only are they not jurisdictional, it's hard to even fit them into plain error review, as it would seem there's no plain error when a defendant is convicted in spite of an affirmative defense he failed to make, however plain that defense may be. Forfeited statute-of-limitations defenses can resurface upon habeas review, as a basis for an ineffective assistance of counsel argument, and a few Justices questioned the government on why a defendant should have to wait until habeas review to raise (in essence) a forfeited statute of limitations defense. But that practically minded concern seems unlikely to carry the day.