In United States v. Barrett, the Second Circuit (Winter, Raggi, Droney) rejected a defendant’s argument that his conviction under Title 18, United States Code, Section 924(c), for using firearms in the commission of a violent crime, should be reversed based on the recent Supreme Court decisions in Sessions v. Dimaya, 138 S. Ct. 1204 (2018) and Johnson v. United States, 135 S. Ct. 2551 (2015). Barrett admitted on appeal that the evidence showed that he was a member of a violent robbery conspiracy but still contended that the firearms counts could not be upheld because the underlying Hobbs Act predicates did not categorically fall within the definition of a crime of violence in Section 924(c)(3). While Dimaya and Johnson have changed how courts must construe Section 924(c)(3), those precedents did not lead the Court to reverse Barrett’s conviction, leaving in place his 90-year sentence. This is a major decision for those who practice in the violent crimes area and given the novelty of the issues the decision covers, it seems likely to be the subject of further review, perhaps by the Supreme Court.

The relevant facts are uncomplicated. Barrett participated in a violent robbery crew that committed multiple robberies during the course of the conspiracy. He was convicted of one count of Hobbs Act robbery conspiracy, multiple substantive counts of Hobbs Act robbery and multiple firearms offenses, including a conviction for using a firearm in connection with a murder.

The Court’s Decision

Judge Raggi’s decision addresses two arguments raised by Barrett that relate to his 924(c) convictions.[1] The Court quickly rejected Barrett’s first argument: that three of his firearms convictions should be reversed because the underlying crimes—substantive Hobbs Act robbery counts—were not crimes of violence. Under the Circuit’s recent decision in United States v. Hill, 890 F.3d 51 (2d Cir. 2018), Barrett’s categorical argument failed. In Hill, the Court held that a Hobbs Act robbery is a categorical crime of violence because it has as an element the use, attempted use, or threatened use of physical force against another person. This qualifies as a crime of violence under the “force clause” set out in Section 924(c)(3)(A).

Barrett’s second argument—that the firearm count based on his Hobbs Act conspiracy conviction should be reversed—required much more analysis. In particular, one of the two alternative grounds raised by the Court involved considerable work by the Court in order to affirm and involved abandoning the “categorical approach,” long a hallmark of analysis of whether an offense qualifies as a crime of violence under Section 924(c).

At the outset of its discussion, the Court pointed to a long line of precedent holding that a conspiracy to commit a crime of violence—such as a Hobbs Act robbery—is itself a crime of violence. Courts so ruled because a conspiracy to commit a crime of violence itself poses a substantial risk of violence. However, this precedent depended on the so-called “residual clause” in Section 924(c)(3)(B), which defines a crime of violence as including those offenses that by their nature, “involve[] a substantial risk that physical force against the person or property of another may be used in committing the crime.”

Barrett argued that this line of authority is no longer good law after Dimaya and Johnson. In the context of a deportation proceeding, Dimaya held that the residual clause of the definition of a crime of violence in 18 U.S.C. § 16(b)—which contains the same definition at issue here, Section 924(c)(3)(B)—was unconstitutionally vague because of the “grave uncertainty” about how judges estimate the risk posed by a crime’s “ordinary case,” and what threshold level of risk is needed for a crime to be a violent felony. Dimaya followed Johnson, which invalidated the residual definition of a crime of violence due to the “‘hopeless indeterminacy’ that resulted from tying a judicial assessment of risk to a crime’s hypothetical ‘ordinary case.’” Even the government’s brief acknowledged that Dimaya and Johnson raised “serious constitutional questions” as to the continued validity of the residual clause.

The Court Offers A Reason To Distinguish Dimaya and Johnson

However, the Court persevered and offered two grounds for affirmance notwithstanding Dimaya and Johnson. First, the Court held that even while Dimaya and Johnson prevent a court from looking to the “ordinary case” when making a violent crime determination, it could still make a violent crime determination “by applying the elements of that crime to § 924(c)(3)(A) together with § 924(c)(3)(B) . . . by following the traditional categorical approach.” Here, a conspiracy is an agreement to commit one or more unlawful acts. The unlawful acts at issue—the object of the conspiracy—in a Hobbs Act robbery conspiracy are crimes of violence. As the Court just held, the underlying substantive offense of Hobbs Act robbery is a crime of violence. The Court then looked to the second element of conspiracy—the agreement element’s requirement that two or more people join a common scheme to achieve the object of the conspiracy. Here, the Court held that the agreement element means that in each case of Hobbs Act robbery conspiracy, the risk of force is present. “[W]here, as here, the elements establish an object offense as a categorical crime of violence under § 924(c)(3)(A), the conspiracy itself—by virtue of its agreement element—is a categorical crime of violence under § 924(c)(3)(B).”

The Alternative Basis For Affirmance Does Not Employ The Categorical Approach

Perhaps recognizing that this analysis was subject to critique as incorrect under the categorical approach, the Court offered an alternative basis for affirmance that did not require the use of the categorical approach. In the alternative, the Court held that Section 924(c)(3)(B) can be applied to a defendant’s case-specific conduct if the jury makes the finding that the underlying offense is a crime of violence that creates a risk of physical force being used in its commission. The Court stressed that its analysis was devised in order to save a statute from unconstitutionality:

Submitting § 924(c)(3)(B) determinations to trial juries for conduct-specific determinations thus avoids not only the constitutional vagueness concerns that Dimaya and Johnson located in the categorical ordinary-case standard, but also the Sixth Amendment right-to-trial concern that originally prompted the Supreme Court to mandate a categorical approach to residual definitions of crimes of violence.

As part of its reasoning, the Court engaged in a lengthy recitation of the history of the categorical approach, going back to its creation in Taylor v. United States, 495 U.S. 574 (1990). The Court explained that “the mandate for a categorical approach to residual definitions of violent crimes has developed in a singular context: judicial identification of what crimes . . . of prior conviction fit federal definitions of violent crimes so as to expose a defendant to enhanced penalties or other adverse consequences in subsequent federal proceedings.” However, the Supreme Court in its jurisprudence never looked at the residual definition in the context of a pending prosecution—in other words, where the relevant underlying conviction was not a prior conviction, but rather, part of the pending prosecution. A conduct-specific factual inquiry ordinarily raises a Sixth Amendment concern because the finding is made by a judge, not a jury.

But here, no such concern would exist if the jury was asked to make the finding about the nature of the predicate offense. Asking the jury to make this finding would be easy; after all, the predicate offense was one of the offenses of conviction. In other words, both the firearms count and the underlying crime of violence were being tried jointly. So while Dimaya held that similar language demanded a categorical approach, the Circuit here held that the doctrine of constitutional avoidance allowed a different result.

The Court reached this conclusion despite the fact that the words “by its nature” in the residual clause seem to suggest that a categorical offense is required. Again, by relying on the doctrine of constitutional avoidance, the Court held that “because a § 924(c)(3)(B) determination can be made by a trial jury based on a defendant’s real-world conduct without raising either due process or Sixth Amendment concerns, Dimaya and Johnson do not necessarily compel invalidation of Barrett’s conviction on Count Two [the firearms count related to the robbery conspiracy].”

The Court had one more hurdle to clear in affirming: given that the Circuit’s proposal was entirely new, the district court did not submit to the jury this newly framed relevant question of whether the crime committed here, by its nature, posed a risk of violence, as defined under Section 924(c)(3)(B). Therefore, the Court had to conduct a harmless error analysis. Here, the Court held that the error was harmless beyond a reasonable doubt because “violence was the very hallmark of the charged conspiracy.” The Court therefore affirmed.

Analysis

The Court’s opinion in Barrett goes to great lengths both to preserve the conviction and to conclude that the residual clause of Section 924(c)(3)(B) remains constitutional, notwithstanding the Court’s ruling in Dimaya that an identical provision is void for vagueness. The Court’s first basis for affirmance is hard to square with Dimaya, and perhaps for this reason, the Court offers a second basis for affirmance. The alternative basis required the Court to depart from the categorical approach—long the only way in which convictions have been analyzed under the “crime of violence” definition—and to then conduct harmless error analysis to account for the failure to instruct the jury on this subject. There is an apparent conflict with Dimaya, which really did not leave the door open for this sort of analysis—it used the categorical analysis, as have all other courts since 1990, and never suggested that another approach was permitted. Although the Court is right to point out that the categorical approach has been frequently criticized for reaching counter-intuitive results, the Supreme Court remains a believer in the approach.

The Barrett decision also creates a conflict with post-Dimaya decisions in other Circuits. Indeed, the panel acknowledged that both United States v. Eshetu, 898 F.3d 36 (D.C. Cir. 2018) and United States v. Salas, 889 F.3d 681 (10th Cir. 2018) hold that Dimaya renders Section 924(c)(3)(B) unconstitutionally vague. Given all of this, even if the case was correctly decided, a rehearing petition and/or a petition for certiorari seems likely, with the Supreme Court ultimately passing on whether the reanimation of the residual clause undertaken by the Court in Barrett will remain good law. Ultimately, Congress has the authority to fix this problem once and for all by revising or repealing the residual clause. A recent proposal, which was passed by the House of Representatives in early September, seeks to avoid the indeterminacy of the residual clause by enumerating offenses that qualify as a “crime of violence.”[2] The Sentencing Commission, dealing with the residual clause in Guidelines Section 4B1.2 (which is the career offender guideline), decided to simply delete the residual clause in light of Johnson. See U.S.S.G. Amendment 798 (“The Commission determined that the residual clause at §4B1.2 implicates many of the same concerns cited by the Supreme Court in Johnson, and, as a matter of policy, amends §4B1.2(a)(2) to strike the clause.”). As the Commission recognized, removal of the clause “has the advantage of alleviating the considerable application difficulties associated with that clause, as expressed by judges, probation officers, and litigants.” Id.

Yet another option would be for Congress to reconsider whether it is reasonable to require mandatory minimum sentences imposed in a single proceeding to be “stacked” in a way that permits a sentence of 90 years to be imposed in a case such as this. While the crimes here were grave, the sentence would have remained a strong deterrent (65 years’ imprisonment, with a 30-year mandatory minimum), even without an affirmance on the firearms count related to the Hobbs Act conspiracy count. One bill pending in the current Congress, the Sentencing Reform and Correction Act of 2017 (S.1917), would change the law. Section 104(a) of this bill provides that the enhanced 25-year mandatory minimum count for second and successive firearm violations apply only when the violation of Section 924(d) occurs “after a prior conviction under this subsection has become final.” Whether any of these steps will be taken by Congress is unclear as of this writing.