For many years, purportedly to preserve its renowned sense of collegiality, the Second Circuit heard fewer cases en banc than any other circuit court. On March 2, 2021, however, the Second Circuit issued an en banc opinion that could prove to be a harbinger of a new era featuring more en bancs and potentially less collegiality. In United States v. Scott, the en banc Second Circuit overturned a prior panel decision and held that New York first-degree manslaughter is a categorial crime of violence under the federal Armed Career Criminal Act (“ACCA”), which can trigger significant mandatory minimum sentences. Others have summarized the opinion elsewhere, and the holding itself is not particularly surprising. What is surprising, however, is the unusually harsh language that jumps out at a handful of moments in the majority and dissenting opinions.

By way of background, Gerald Scott pled guilty in 2007 to armed robbery, brandishing a firearm in relation to an armed robbery, and being a felon in possession of a weapon. The felon-in-possession count carried a minimum term of 15 years’ imprisonment under ACCA (due to Scott’s two prior first-degree manslaughter convictions from 1988), and the brandishing count carried a minimum consecutive term of 7 years’ imprisonment, making the aggregate mandatory minimum term of imprisonment a total of 22 years. In 2008, District Judge Laura Taylor Swain sentenced Scott to the minimum term of 22 years’ imprisonment.

Following the Supreme Court’s 2015 decision in Johnson v. United States, however, which held that part of ACCA was unconstitutionally vague, Judge Swain granted Scott’s post-conviction habeas petition and resentenced him to time served (11 years, 3 months). Judge Swain concluded that, in the wake of Johnson, ACCA’s 15-year mandatory minimum did not apply because the first-degree manslaughter convictions that triggered ACCA could be committed not just by action, but by an omission in breach of a duty to act (like when a parent fails to provide medical care for a child), and thus did not necessarily involve the requisite “use of force” under ACCA. Applying the same reasoning, Judge Swain also found that first-degree manslaughter does not trigger the Career Offender Guideline. The Second Circuit affirmed in a panel decision written by Judge Rosemary Pooler and joined by Judge Pierre Leval. Judge Reena Raggi filed a dissenting opinion. The Circuit subsequently granted the government’s motion for rehearing en banc.

Against that backdrop, the question that the en banc Second Circuit decided to take up had nothing to do with the specifics of Scott’s conduct, either in connection with the 2008 robbery and firearms convictions or his prior manslaughter convictions from 1988. Rather, the en banc court limited its analysis to whether New York’s first-degree manslaughter statute—which applies when, “with intent to cause serious physical injury to another person, [one] causes the death of such person or of a third person”—categorically entails, under ACCA and the Career Offender Guideline, “the use, attempted use, or threatened use of physical force against the person of another.” The majority opinion by Judge Raggi (no longer in dissent) provides a simple answer in the affirmative: “New York first-degree manslaughter is a categorically violent crime under both the ACCA and Career Offender Guideline force clauses because, whether a defendant acts by commission or omission, in every instance, it is his intentional use of physical force against the person of another that causes death.” The majority explains this conclusion at some length, but the key concept is that a defendant who causes someone’s death while “intend[ing] to cause serious physical injury” necessarily uses physical force even if the physical force involved is not the defendant’s own.

The conclusion reached by the majority in its en banc decision is debatable, as evidence by the Second Circuit’s decision in which experienced judges came out on opposite sides. The language in both the majority and dissenting opinions, however, is unusually forceful.

At the start of the majority opinion, Judge Raggi states the question in this way: “At issue on this appeal is whether Scott’s manslaughter convictions are for violent crimes.” The majority observes that “[a]n affirmative answer might appear obvious to a man on the street aware of Scott’s conduct.” (In a concurring opinion, Judge Michael Park uses a similar phrase, writing that “[t]he question answers itself to any layperson with common sense.”) Judges Raggi and Park are surely right that a “man on the street” or “layperson with common sense” who is “aware of Scott’s conduct” would describe that conduct as violent. But the question for the Court was decidedly not whether Scott had engaged in violent conduct over thirty years earlier. Instead, the question was whether the offense of first-degree manslaughter in New York is categorically violent, or if it can realistically be committed in ways that do not involve the use of force. If you asked the “man on the street” or the “layperson with common sense” whether first-degree manslaughter is categorically a violent crime in New York the answer could well be less “obvious” than the majority and concurring opinions presume.

In dissent, Judge Leval does not shy away from expressing his own views of the majority decision. Instead, he answers the majority’s multiple references to Scott’s underlying conduct, writing, “[t]he majority opinion devotes considerable ink to telling us what an atrocious beast Scott is.” Judge Pooler, in her own dissent, goes further, explaining that the majority finds the original panel’s “conclusion distasteful because of the nature of the crimes Scott committed,” and that while she does “not minimize the consequences of Scott’s serious crimes, that is simply not a relevant consideration.” This somewhat testy back-and-forth discussion of Scott’s underlying prior offense, despite its irrelevance to the analysis, illustrates a seeming discomfort with the nature of the categorical analysis itself.

Deeper in the opinion, but still in the lengthy introduction, the majority describes the sentencing consequences of its decision, but does so in a somewhat confusing way. The majority explains that “to impose ACCA’s fifteen-year mandatory minimum sentence for Scott being a felon in possession of a firearm, the court had to find that he had three or more prior ‘violent felony’ or ‘serious drug offense’ convictions.” (Emphasis supplied.) This sentence suggests that the finding at issue (that the manslaughter convictions count as violent felonies) was a necessary means to the desired end of imposing a sentence of 15 years or more on Scott (whose conduct, the majority makes clear, well deserved it).

The majority reinforces this impression by adding a footnote stating that “[w]ithout the ACCA enhancement, a person convicted of being a felon in possession of a firearm faces a sentence of between 0 and 10 years.” In combination, the sentence and footnote suggest—certainly to the man on the street—that Scott would have been sent to prison for between 0 and 10 years had the en banc majority not stepped in. While it is technically true that the district court would not have been able to sentence Scott to more than 10 years on the felon-in-possession count without the ACCA enhancement, this suggestion ignores the fact that, due to his separate robbery and brandishing counts of conviction, Scott actually faced a maximum sentence that far exceeded his original 22-year sentence regardless of the ACCA enhancement. In other words, the district court would have been free to impose a lengthy sentence (including the original sentence of 22 years) regardless of how the en banc Court came out.

In another striking passage, the majority references the Circuit’s prior decision in Chrzanoski v. Ashcroft, notes that Chrzanoski was abrogated by the Supreme Court’s decision in United States v. Castleman, and then cites a prior Second Circuit panel’s observation along these lines in a case called Villanueva v. United States. But the majority continues, with a biting observation, stating, “[a]pparently, Villanueva did not make the point [that Castleman overruled Chrzanoski] clearly enough, however, because Scott maintains—incorrectly—that Chrzanoski remains good law with respect to crimes that can be committed by omission.” The observation that “[a]pparently, Villanueva did not make the point clearly enough” strikes a somewhat chiding tone that is amplified by the fact that two of the dissenting judges in the en banc—Judge Leval and Judge Pooler—were on the panel that decided Villanueva. (Judge Leval joined Judge Newman’s opinion for the Court in Villaneuva, and Judge Pooler dissented. Indeed, in her en banc dissent, Judge Pooler explains her disagreement with the majority’s analysis in Villanueva at some length.)

The majority’s treatment of the rule of lenity also strikes a slightly sour note. The majority begins by observing that “Scott’s invocation of the rule is not without some irony”: “There can be no question that he performed physical acts in shooting and stabbing his two manslaughter victims” and “[t]hus, to the extent the lenity rule is grounded in a concern for ‘fair warning,’ Scott can hardly claim that he was not clearly warned.” (Emphasis in original.) The real question, though, may not be whether Scott got “fair warning,” but whether some future defendant has fair warning that ACCA will apply.

Not content to critique Scott (or, really, his lawyers) for invoking the rule of lenity, the majority specifically calls out the dissenting opinions in a seemingly offhand but revealing footnote. In particular, in footnote 35, the majority writes that “[e]ven if some of us might share Judge Leval’s disquiet about mandatory minimum sentences and his preference for broad judicial sentencing discretion, our task here is to identify unresolvable statutory ambiguity, not to tilt the balance in favor of judicial sentencing discretion.” The reference to “some of us” seems to say that not all of the judges in the majority are concerned about mandatory minimum sentences (despite the fact that for some 60 years, “the Judicial Conference has consistently and vigorously opposed mandatory minimums and has supported measures for their repeal or to ameliorate their effects”). Further, the word “disquiet”—meaning worry or uneasiness—may be an understatement. Judge Leval’s dissent argues that “harsh mandatory sentences inevitably become engines of needless injustice,” which might be justified if “the legal procedures in the federal courts delegated the sentencing function to chimpanzees.” (Emphasis supplied.)

Ultimately, we can take at least two lessons from Scott. First, the sharp language may perhaps indicate a widening divide on this famously collegial court. It remains to be seen whether this trend continues in future panel decisions (and en bancs) on sensitive issues.

Second—and perhaps more importantly—the opinion is a reminder of the extent to which substantial mandatory minimum sentences remain a prominent feature of the federal criminal justice system despite decades of sustained criticism, including from the bench. While the First Step Act reduced mandatory minimum sentences for nonviolent drug offenses, there remain a great many cases in which severe mandatory minimum sentences apply, and at the margins, Scott serves to broaden that universe. In a commonly cited statistic, the United States has 5% of the world’s population but 25% of the world’s prison population. Just as the “man on the street” and the “layperson with common sense” would no doubt recognize Scott’s 1988 conduct as violent, they would also no doubt understand how harsh mandatory minimum sentences continue to contribute to the problem of mass incarceration in the United States.