Categories & Search

In Rare En Banc Ruling, Second Circuit Holds that Manslaughter is a Categorically Violent Felony, Including Cases of Omission, Potentially Triggering Mandatory Minimums

Our previous summary of the decision is here.

On March 2, 2021, in a rare en banc decision, United States v. Scott, 18-163-cr, the Second Circuit held in a divided 9-5 opinion that New York first-degree manslaughter is categorically a “violent felony” under the Armed Career Criminal Act—subjecting qualifying defendants to the statute’s mandatory minimum sentences— and a “crime of violence” under the Career Offender provision of the Sentencing Guidelines, despite the fact that manslaughter can be carried out by omission. 

The majority opinion was written by Judge Raggi and was joined by Chief Judge Livingston and Judges Cabranes, Chin, Sullivan, Bianco, Park and Nardini in full, and by Judge Menashi in part.  It is the latest in a cascade of opinions to come out of the Supreme Court and the various circuit courts about what kinds of felonies and convictions appropriately trigger certain mandatory and non-mandatory sentencing enhancements, some of which can result in substantial increases to the term of imprisonment. 

The opinions sparked no fewer than two concurring opinions, by recently-appointed Judges Park and Menashi, and two vigorous dissents from Judges Leval and Pooler.  The Second Circuit grants en banc rehearing very infrequently, and the various opinions on this case reveal sharp philosophical differences about mandatory minimums, judicial discretion, and the so-called “categorical approach” to determining whether a particular conviction warrants the application of a sentencing enhancement.


In 1983 Gerald Scott was convicted of first-degree robbery, in violation of N.Y. Penal Law § 160.15, and in 1988 was twice convicted of New York first-degree manslaughter for fatally shooting a victim in one instance and stabbing a victim to death in another, both in violation of N.Y. Penal Law § 125.001.  Then, in 2006, Scott was convicted in the United States District Court for the Southern District of New York of three federal crimes related to his armed robbery of a jewelry store: attempted Hobbs Act robbery, in violation of 18 U.S.C. § 1951; brandishing a firearm during that robbery, in violation of 18 U.S.C. § 924(c)(1)(A)(ii); and being a previously convicted felon in possession of a firearm, in violation of 18 U.S.C. §§ 922(g)(1) and 924(e). 

The district court that sentenced Scott for his federal crimes concluded that, in light of his three prior convictions, Scott warranted sentencing enhancements under both the “Career Offender” section of the Sentencing Guidelines and the ACCA.  The Career Offender provision of the Guidelines provides for an enhancement if the defendant had two prior convictions for “crimes of violence,” while the ACCA imposes a fifteen-year mandatory minimum sentence if the defendant had three prior convictions for a “violent felony.”  The district court sentenced Scott to 22 years in prison, reflecting the ACCA’s mandatory minimum.

In 2016, Scott filed a habeas petition pursuant to 28 U.S.C. § 2255 challenging his conviction and sentence, arguing that his manslaughter convictions did not qualify as predicate violent-crime offenses under the ACCA and the Career Offender Guideline enhancements.  The district court accepted Scott’s argument, ruling that it should not have applied the ACCA’s mandatory minimum because first-degree manslaughter is not a categorical violent felony as it can be committed by omission, and that, under similar reasoning, it was not a crime of violence under the Guidelines.  The district court also concluded that first-degree manslaughter is not a crime of violence under the enumerated offenses clause because it did not fit within any of the generic definitions there listed.  Accordingly, the district court reduced Scott’s sentence to 11 years and 3 months.  The original panel that heard the appeal affirmed, but the Court voted to rehear the case en banc, a fairly unusual step for the Second Circuit.

The En Banc Majority Decision

The majority overruled the original panel’s decision, holding that New York first-degree manslaughter is categorically a violent felony under the ACCA and a crime of violence under the Career Offender Guidelines, reversing the district court’s ruling and remanding with instructions to reinstate the original sentence.

To understand the majority’s decision and those of the other judges, it is helpful to start with an explanation of the “categorical approach” that courts use to assess whether a particular crime qualifies to trigger certain federal sentencing enhancements.  The ACCA imposes mandatory minimum sentences for defendants with at least three prior convictions for a “violent felony” or “serious drug offense.” As relevant here, the ACCA defines a “violent felony” as a crime that “has as an element the use, attempted use, or threatened use of physical force against the person of another.”  18 U.S.C. 924(e)(2)(B)(i).  The Career Offender Guidelines provide for a non-mandatory sentencing enhancement if the defendant was previously convicted of at least two felony convictions for a “crime of violence” or “controlled substance offense.” U.S.S.G. § 4B1.1(a).  A “crime of violence” under the Guidelines is defined in the same way as a “violent felony” in the ACCA, so for ease the analysis here refers to both as a “violent felony.”

In New York, a person is guilty of first-degree manslaughter when “[w]ith intent to cause serious physical injury to another person, he causes the death of such person or of a third person.”  N.Y. Penal Law § 125.001.  To most people, it might seem entirely uncontroversial to label first-degree manslaughter as a violent felony—it sounds violent and in most cases involves violence by the defendant against the victim.  But for the purposes of determining whether a crime is a violent felony under the sentencing enhancements, courts apply a mode of legal analysis called the “categorical approach,” which requires the court to identify the minimum criminal conduct necessary for conviction under a particular statute, and then to determine whether that conduct is encompassed by the definition of a violent felony in the sentencing statute.  The defendant’s actual conduct—here, shooting one victim in the head and stabbing another to death—is irrelevant to this analysis.  In other words, if a defendant could theoretically be convicted of violating a statute by committing conduct that would not qualify as a violent felony, then the conviction is not “categorically” a predicate for the sentencing enhancement. 

The issue here was that the manslaughter statute has been interpreted by the courts in New York to include manslaughter committed by omission, meaning the failure to act despite a legal duty to do so.  As Scott, the district court, and the original panel saw it, if “omission” is the minimum conduct necessary to violate the manslaughter statute, then the crime can be committed by taking no action at all.  By contrast, a violent felony requires “the use, attempted use, or threatened use of physical force.”  Manslaughter, according to the original panel’s decision, is therefore not a categorical match for a violent felony.

The en banc majority disagreed.  In a strongly-worded opinion that took pains to highlight the violent and unsympathetic nature of the crimes and the defendant here, the majority took the position that a defendant whose aim is to at least seriously injure and who has caused the death of another can only have achieved that outcome by using violent force.  In the majority’s view, the word “use” in the definition of violent felony refers not to the physical act of the defendant, but to the violent force that was the cause of the victim’s death.  The defendant does not need to perform any particular physical act—or any physical act at all—in order to have “used” that violent force. 

For example, if a parent knowingly lets his child consume poison, with the intent to at least seriously injure that child, he has “used” violent force in the sense that he has “made use of” the violent force of the poison to kill the child.  Omission is not just inaction, the majority emphasized.  It is inaction in the face of a legal duty, for which the law can impose criminal liability.  In the majority’s view, it is irrelevant the means by which the crime was committed, only that it was carried out with the requisite intent and that the defendant in some way made use of or employed force that resulted in death.  

The majority also relied on the Supreme Court’s decision in United States v. Castleman, 572 U.S. 157 (2014).  Castleman involved the interpretation of a misdemeanor domestic violence statute, which required the use of “physical force.”  The Supreme Court in that case stated that “the knowing or intentional causation of bodily injury necessarily involves the use of physical force.”  The Supreme Court’s reasoning in Castleman—although applied in a different context—has been cited by the Second Circuit on more than one occasion to broaden its previous rulings on what constitutes use of physical force.  In the majority’s view, the holding and reasoning in Castleman “compelled” the conclusion that even manslaughter by omission involves the “use” of force because it involves the “knowing or intentional causation of bodily injury,” and is therefore categorically a violent felony under the ACCA and the Career Offender Guidelines.


Judge Park concurred with the majority opinion in full, but wrote separately to condemn the categorical approach, under which courts “are required to ignore the actual facts before them and instead to theorize about whether certain crimes could be committed without using violent force,” which, he maintained, “perverts the will of Congress and leads to inconsistent results, wastes judicial resources, and undermines confidence in the administration of justice.”  The categorical approach has been criticized by numerous judges—many of whom are quoted in Judge Park’s concurrence—who have called on the Supreme Court to revisit its jurisprudence in this area.  Chief Judge Livingston and Judges Cabranes, Sullivan and Nardini joined Judge Park’s concurrence.

Judge Menashi wrote an opinion concurring in the judgment but calling for greater interpretive consistency to the reading of the predicate statute as compared to the relevant enhancement clauses by applying, for example, the “legal meaning” of the language of both, rather than oscillating between a legal interpretation and one that relies on the “ordinary” meaning of the words.  In his view, much of the confusion and interpretive hoops through which the majority has to jump stem from the fact that it is defining “use” by reference to its “ordinary” meaning, rather than its specialized legal meaning under the law.  He argues that the ordinary meaning of “use of physical force” does not include an omission.  But the legal meaning of the phrase does include culpable omissions because the law equates omission with action when a legal duty to act exists.  The majority’s discussion of the ordinary meaning of the term is unhelpful, he argues, especially where it is presumed that Congress legislates against a backdrop of established legal principles and specialized legal meanings.  This interpretative consistency, Judge Menashi maintained, would go a long way to curing the deficiencies of the categorical approach that Judge Park identified in his concurrence.


Judge Leval penned a dissent in which he called for the application of the “rule of lenity” to the question of whether a crime that can be committed by omission is sufficient to trigger the ACCA’s mandatory minimums.  Judges Katzmann, Lohier, and Carney joined this dissent in full; Judge Pooler joined in part.  The rule of lenity requires that criminal statutes give defendants clear notice and warning of the conduct that will be punished, and directs the court to resolve ambiguity in favor of the defendant.  Such a rule is particularly appropriate, he argued, where the statute involves a harsh mandatory sentence imposed for a crime to which it does not clearly apply. 

While Judge Leval took no issue with sentencing Scott to a lengthy prison term, given the nature of his crimes, he voiced deep concern about applying a mandatory 15-year minimum term of imprisonment to every defendant that comes within the ACCA’s scope because of a first-degree manslaughter charge.  At bottom, Judge Leval’s dissent is a denunciation of mandatory minimum sentences, which leave no room for judicial discretion and are “engines of needless injustice.”  Judge Leval noted that the most significant effect of the majority opinion is not that it would set a floor for defendants who deserve long sentences—who, it could be argued, are already likely to receive lengthy sentences given the nature of their crimes and criminal histories—but that it will require a 15-year imprisonment of defendants for whom such a sentence is too harsh and for whom the statute does not give fair notice.

Notably, Judge Leval disagreed with Judge Pooler’s dissent insofar as it concluded that New York first-degree manslaughter is not a “crime of violence” under the Guidelines—which was, after all, a central question at issue on appeal.  This was so because, where the interpretation of the relevant language did not result in a severe, mandatory sentence, but rather left judges free to impose a sentence outside the Guidelines, the application of the rule of lenity was less urgent.  Judge Leval agreed that there were reasonable arguments for classifying manslaughter by inaction as a crime of violence under the Guidelines and would not apply the rule of lenity to them to resolve any ambiguity in favor of the defendant in the absence of a mandatory minimum sentence.

Finally, Judge Pooler wrote a passionate dissent, joined in part by Judges Leval and Carney, in which she disagreed that first-degree manslaughter is categorically a violent felony under either the ACCA or the Career Offender provision of the Guidelines.  In Judge Pooler’s view, the text of the ACCA plainly encompasses only violent, physical force by the defendant—a standard that cannot be satisfied by omission which, by definition, involves the lack of any action.  Castleman, Judge Pooler wrote, was interpreting an entirely different statute and context.  Nothing in that decision, she asserted, was intended to abrogate the definition of physical force set forth in Johnson v. United States, 559 U.S. 133 (2010), which described it as “force strong enough to constitute ‘power.’”  In this way, Judge Pooler reiterated her objection to the ruling in Villanueva v. United States, 893 F.3d 123 (2d Cir. 2018), in which she argued in dissent that the Court should not extend Castleman to the interpretation of the ACCA.

In Judge Pooler’s view, the majority has effectively conflated the mens rea required to commit manslaughter, with the actus reus, the physical act involved in committing a crime, by ruling that it is enough to simply possess the requisite intent without taking any further action.  Nor, she maintains, is it an answer to say that the law imposes criminal culpability for certain acts of omission.  That alone does not transform the omission into an action or use of force.  Judge Pooler was careful to note that she was not minimizing Scott’s actions, but simply contending that the statute he violated was not within the scope of the types of crimes that the ACCA was intended to discourage.

Like Judge Leval, Judge Pooler devoted several paragraphs to denouncing harsh mandatory minimum sentences that she contends strip trial judges of discretion and encourage prosecutors to aggressively pursue the highest possible charge to provide leverage in plea negotiations. 


The Court’s relatively rare en banc opinion revealed sharp differences in approach among the various judges within the circuit.  It also offered an opportunity for the judges to air impassioned views on certain controversial subjects involving sentencing, judicial discretion, and mandatory minimums.  The categorical approach has long been a source of frustration for many judges, who feel that it requires them to take part in a legal fiction that considers a theoretical set of facts, while ignoring the actual and sometimes compelling facts of the case before them.  The type of “angels-on-the-head-of-a-pin” analysis required by the categorical approach seems like an odd way to determine the fate of another human being. 

On the other hand, the Supreme Court has defended the categorical approach as offering greater consistency and predictability in sentencing, honoring the intentions of Congress, and avoiding the often-difficult task of mining records of conviction to engage in an individualized fact-specific inquiry, which could itself present Sixth Amendment concerns.  How exactly would judges determine whether a crime committed by a defendant many years in the past was a crime of violence?  The guilty pleas taken in state court often contain no factual allocution.  Should judges trust police reports that are neither subject to cross-examination nor within the judicial record?  Or should judges accept as true a prosecutor’s rendition of the facts even where the defendant was not asked to agree with those facts?  Should there be judicial fact-finding at sentencing—a mini-trial in federal court about a case from 15 or 20 years before, a case that did not go to trial?  Like Churchill’s aphorism about democracy, the categorical approach is the worst possible approach, except for all of the other possible approaches under current law.

The opinion also highlighted the strong antipathy of several judges to mandatory minimum sentencing statutes.  In the eyes of its many critics, mandatory minimums deprive capable federal judges of exercising their discretion in precisely those cases where it is most appropriate—where the circumstances of the case and the personal history of the defendant are such that a punitive, lengthy sentence is unwarranted or even counterproductive.  Moreover, harsh mandatory minimums permit prosecutors to wield tremendous power, as it is the government that decides how to charge the defendant.  And in a country that devotes tremendous resources to incarcerating more than 2 million people, many have called for Congress to reconsider sentencing regimes that impose harsh, compulsory sentences. 

This is the real fix for the deficiencies of the categorical approach and indeed for the many cases that have been decided under the ACCA in the past:  Congress should reduce or repeal the mandatory minimum sentences that were legislated many years ago and require a level of punishment that is not necessary for deterrence or just punishment.  While courts can vary from a sentence required by the Career Offender Guidelines, the mandatory minimum sentences required by Section 841(b)(1)(A) and (B) (for drug crimes) and by Section 924 (for certain firearms offenses) are overdue for reexamination, as the dissenting judges suggest.  In the alternative, given the congressional inaction caused by the filibuster rule and other circumstances of our modern politics, the Department of Justice could avoid charging and proving the facts needed to impose mandatory minimum sentences.

By Elena Steiger Reich and Harry Sandick