Should judicial participation in plea discussions be automatically reversible? “No pressure” if the Justices stick to the Question Presented
on Apr 12, 2013 at 12:47 pm
With apologies to fifty-seven of my fellow “Criminal Law and Procedure Professors” who have filed an amicus brief in support of respondent Anthony Davila in United States v. Davila (set for argument on Monday April 15 – can that really be a coincidence for a felony tax offender?), this looks like a simple case. “Deceptively simple,” Davila’s lawyer Josh Rosenkranz might respond – his brief does a good job of making one pause at the implications of the underlying facts. But the Question on which the Court granted – whether “any degree of judicial participation in plea negotiations in violation of Federal Rule of Criminal Procedure 11(c)(1) automatically requires” reversal (my emphasis) – really does not require examination of these implications. In recent years, the Court has firmly rejected endorsing “automatic reversal” rules, let alone ones based on procedural rules rather than the Constitution, rules that many states do not follow. Expect Monday’s argument to be respectful but one-sided on the Question Presented, with the more defendant-friendly Justices perhaps focused on how best to limit a reversal so as to not endorse the disturbing implications that Davila (and his law professor amici) admirably present.
Facts and background
In a 2009 indictment in the Southern District of Georgia, Davila was charged with filing over 130 false tax returns seeking refunds in the names of Florida state prisoners. According to the government, Davila received over $423,000 from his scheme. But after counsel was appointed, Davila soon filed a motion to discharge him, saying that the lawyer had done nothing other than advise him to “plead guilty.”
At an ex parte hearing on this motion, a magistrate judge made the fateful (and vaguely religious) statements that violated (all now agree) Federal Rule of Criminal Procedure 11(c)(1), which firmly directs that “[t]he court must not participate” in plea discussions. Said the uncircumspect magistrate judge: “[o]ftentimes … the best advice a lawyer can give” is to plead guilty. “[T]here may not be a viable defense.” “It might be a good idea for the Defendant to … plead guilty and go to sentencing.” He continued:
The only thing at your disposal … is the two or three level reduction for acceptance of responsibility. That means you’ve got to go to the cross. You’ve got to tell the probation officer everything you did … regardless of how bad it makes you appear. … Mr. Davila, someone with your criminal history needs a three-level reduction for acceptance. ….In order to get the reduction for acceptance, you’ve got to come to the cross.”
The ex parte nature of this hearing (the prosecutor was not present, although Davila’s lawyer was) arguably solidified the appearance that the judge was acting more like a prosecutor than a neutral arbiter and telling the defendant to plead guilty.
No objection was raised to these statements (Davila now says his lawyer was obviously conflicted). Three months later, Davila and his lawyer (the motion to discharge counsel having been denied) signed a plea agreement, and then appeared before the district judge (not the magistrate who had earlier made the offending remarks) and entered a plea of guilty to one count of false claims conspiracy. At that hearing, Davila agreed that his plea was voluntary and that no one had forced or pressured him to plead guilty.
Davila then asked for, and received, permission to proceed pro se at sentencing. He then immediately moved to withdraw his plea, claiming false statements in the indictment and bad advice from his lawyer. The district judge denied this motion, finding that the plea had been “voluntary, knowing and intelligent.” Significantly (says the Solicitor General), Davila did not allege that his plea decision had been influenced by the magistrate’s earlier remarks.
On appeal, Davila’s lawyer (the same lawyer appointed for trial) attempted to file an “Anders” brief, saying that he did not see any nonfrivolous grounds for appeal. But the Eleventh Circuit directed counsel to address an issue that its own “independent review” had uncovered: whether the “irregularity” in the magistrate judge’s statements required reversal under Rule 11(c)(1). Davila’s lawyer then filed a brief arguing that Davila was indeed entitled to reversal, even though the “irregularity” had not been raised below, because it was “plain error.” The government responded that Davila could not possibly meet the plain error standard. The court of appeals, however, did not engage in any plain or harmless error review. Instead, relying on a 1993 per curiam precedent, the circuit ruled that in a judicial-participation case “the defendant need not show actual prejudice,” and reversal (and reassignment) was required. “While other Circuits recognize harmless error in the context of judicial participation, we do not.” Certiorari was unsurprisingly granted on the Solicitor General’s petition.
The government’s argument is simple (as reflected by an opening brief of only twenty-eight pages). While Rule 11(c)(1) states a clear direction without exceptions (“the Court must not participate”), Rule 11(h) just as clearly states that “[a] variance from the requirements of this rule is harmless if it does not affect substantial rights.” The Court has previously made clear that unobjected-to Rule 11 violations are also subject to Rule 52’s plain error standard, which requires a showing that the error “affected substantial rights” – i.e., “a prejudicial effect on the outcome.” Thus any “automatic” or per se reversal rule for “any degree” of judicial participation, even if it is a Rule 11(c)(1) violation, is wrong as well as unwarranted. Harmless error review is required, says the government, by the plain language of the Rules, by Court precedent, and by concerns for finality and needless expenditure of resources.
The government points out that the Eleventh Circuit’s contrary ruling can be traced to a 1981 Fifth Circuit case (binding because when the Eleventh Circuit was created, it adopted all pre-1982 Fifth Circuit precedents) that was decided before Rule 11(h) was adopted in 1983 to require harmless error review. The Fifth Circuit’s precedent was itself based on a broad reading of McCarthy v. United States, which the Court later described as overbroad and which the 1983 amendments to Rule 11 were expressly intended (in part) to restrict. Thus (says the government) the holding below is not only inconsistent with the Rules’ plain language, but its origins are plainly suspect.
The government, however, did not bargain for the force that Davila’s Supreme Court advocate, Josh Rosenkranz, would bring to the opposition. Rosenkranz, marshals not only some powerful arguments in an attempt to make the Court at least pause, but also some powerful amici assistance.His basic argument is that this particular case involves an “undisputed and egregious violation” of Rule 11(c)(1)’s “categorical prohibition,” and it is “inappropriate … to speculate” about whether such unseemly judicial pressuring of a criminal defendant may or may not have affected the outcome. He notes that a judicial-participation violation “differs in kind” from other Rule 11 violations, because it “corrupts” the very decision to plead guilty, rather than occurring after that decision is made. And he offers a number of intricate and creative reasons why the “plain language” of the Rules need not be read, or applied, so plainly. (For example, Rule 11(h) directs harmless error review for violations of the Rule’s “requirements,” but it used to say “procedures” and that change was supposed to be “stylistic only.” Because, Davila argues, a flat prohibition against judicial participation is not (unlike other parts of the rule) a “procedure,” Rule 11(h) should not apply.)
Finally, Davila distinguishes contrary precedents that require harmless error review on the ground that this type of Rule 11 violation “inherently impairs” the vital independent process by which defendants may decide to surrender any number of important constitutional rights when they plead guilty. “[N]o published appellate court decision” has ever “let stand a guilty plea entered after a judge urged the defendant to forego a trial and pursue a plea deal.” “A judge who violates the ban alters the entire pretrial dynamic,” and such judicial pressuring is “fundamentally incompatible with a fair criminal process.”
But Rosenkranz’s “fundamentally incompatible with fair process” argument runs headlong into an unblinkable reality: a number of states do not follow the federal rule. Instead, they expressly endorse (and others tolerate) active involvement by trial judges in the plea negotiation and disposition process (a point the government makes in its twenty-one-page reply brief – Davila and his amici at least made the government work!). The federal no-judicial-participation rule has never been viewed as constitutionally required, and there is a strong argument that non-constitutional violations should always be subject to harmless error review (as are, indeed, even some constitutional violations). It would be difficult if not impossible for the Court to rule that Rule 11(c)(1) violations are “structural” (and therefore potentially not subject to harmless error review) without treading on contrary state practices that seem firmly, if perhaps unwisely, embedded.
Davila thus makes what seems to be a strategic concession, arguing that (only) “core” violations of the no-judicial participation rule should be categorically condemned. Or as the amicus brief by fifty-seven law professors puts it (anyone old enough to want to make a ketchup joke here?) “the particular species of Rule 11(c)(1) violations” they characterize as “judicial pressure” violations. But of course, the government does not take the position that no “judicial participation” errors require reversal – the government’s argument, and its original Question Presented, takes issue solely with a categorical “automatic reversal” rule.
Thus Davila appears to work insightfully for an opinion which, while reversing, provides that some “core” violations should not require any examination for particularized prejudice. I would expect Monday’s argument, if it enlivens the Court at all, to focus on whether this fallback position has any traction within the Court.
Ultimately the government reassures the Court that it advocates a “case-by-case approach,” which will sometimes “allow courts to … grant relief.” Meanwhile the Chief Justice has famously advocated that the Court write its opinions narrowly and not decide more than it must. We’ll see on Monday whether the Justices will stick to the narrow Question Presented, or use the grant of certiorari, always rare on any particular topic, as an opportunity to “pressure” each other to delve into the particularized, and quite serious, additional questions that Davila and his amici present.