No Categorical Match: Sixth Circuit En Banc Holds Ohio Robbery Is Not a “Crime of Violence” Under the Career‑Offender Guideline When the Predicate Theft Offense Is Unspecified

Introduction

In United States v. Tyren Cervenak, the Sixth Circuit, sitting en banc, vacated a career‑offender sentence enhancement imposed under U.S.S.G. § 4B1.1 (2021) and held that Ohio robbery, Ohio Rev. Code § 2911.02(A)(2), is not a “crime of violence” under the Guidelines’ enumerated‑offenses clause when the predicate “theft offense” is unspecified in the record. The ruling resolves intra‑circuit tension wrought by two recent panel decisions—United States v. Carter (2023) and United States v. Ivy (2024)—and establishes two consequential rules: (1) Ohio’s robbery statute is “twice divisible” by subsection and by the underlying theft offense incorporated from § 2913.01(K), and (2) when Shepard documents do not reveal which theft offense underlies the robbery, courts must apply the categorical approach to the least culpable covered theft offenses, which extend beyond generic robbery and Guidelines extortion. The court also clarifies that defendants need not produce state cases to satisfy the “realistic probability” inquiry when the statute’s plain text itself is broader than the generic offense.

The decision carries significant implications for federal sentencing within the Sixth Circuit, especially for defendants with prior Ohio robbery convictions whose state records do not specify the predicate theft offense. It also sharpens the circuit’s approach to the “realistic probability” test and declines to certify state‑law questions to the Ohio Supreme Court despite vigorous calls to do so in separate opinions.

Summary of the Opinion

The en banc majority (Judge Mathis) vacated Tyren Cervenak’s 188‑month sentence, holding that his two prior Ohio robbery convictions under § 2911.02(A)(2) do not qualify as “crimes of violence” under U.S.S.G. § 4B1.2(a)(2) (2021). Applying the modified categorical approach to Ohio’s twice‑divisible robbery statute, the court determined that:

Ohio robbery incorporates any one of more than thirty theft offenses listed in § 2913.01(K), many of which do not require obtaining or misappropriating property (e.g., breaking and entering, trespass in a habitation, forgery, personating an officer, cheating in sports).
Because the Shepard documents did not identify the particular theft offense, the court had to presume the least culpable conduct covered by the statute.
That least culpable conduct is broader than both Guidelines extortion (which requires “obtaining something of value from another”) and generic robbery (which requires misappropriation of property under circumstances involving immediate danger to a person).

The court rejected the government’s argument that, when an indictment is silent, Ohio courts treat § 2913.02 “theft” as the default predicate. It also held that defendants need not cite Ohio prosecutions to show overbreadth where the statutory text itself plainly extends beyond the generic offense (clarifying the “realistic probability” inquiry under Gonzales v. Duenas‑Alvarez). Finally, the court found no Fifth or Sixth Amendment infirmity in Ohio’s indictment practices; an indictment that tracks the statute provides adequate notice and can be supplemented by a bill of particulars. Result: the career‑offender enhancement was improper; the sentence was vacated and remanded.

Analysis

1) Precedents Cited and Their Influence

Descamps v. United States and Mathis v. United States: The court faithfully applied the categorical and modified categorical approaches—comparing elements, not facts; treating divisible statutes via Shepard documents; and presuming the least of the acts criminalized when a predicate element is undefined in the record.
Taylor v. United States (1990) and United States v. Camp: The court derived the “generic robbery” definition from common understanding and state‑law consensus because the 2021 Guidelines did not define robbery.
United States v. Yates and United States v. Ivy: The panel’s pre‑existing generic robbery definition (misappropriation plus immediate danger) frames the enumerated‑offense comparison. Ivy added that generic robbery’s force element entails at least recklessness; the en banc majority did not need to revisit that mens rea element because the mismatch turned on the property component.
United States v. Burris (en banc) and United States v. Butts: Burris confirms the categorical methodology; Butts foreclosed the elements‑clause route for Ohio robbery, channeling the dispute to the enumerated‑offenses clause.
Gonzales v. Duenas‑Alvarez, Moncrieffe v. Holder, and Mellouli v. Lynch: The court harmonized these Supreme Court decisions, emphasizing that the “realistic probability” requirement does not bar reliance on statutory text where the text itself unmistakably covers non‑generic conduct. It aligned with many sister circuits that allow text‑only showings of overbreadth.
State law anchors (e.g., State v. Tolliver, Ohio Jury Instructions, and various Ohio appellate decisions): Used to hold that the underlying theft offense is an element that must be proven (supporting “double divisibility”) and to reject a state‑law presumption that § 2913.02 theft is always the predicate.

2) Legal Reasoning

a) Double Divisibility and the Modified Categorical Approach

The Sixth Circuit deemed § 2911.02 “doubly divisible”: (1) by the three robbery subsections (A)(1)–(A)(3), and (2) by which theft offense under § 2913.01(K) supplies the predicate. That second layer matters because Ohio requires the jury (or judge on plea) to find the elements of the particular predicate theft offense. With the indictment silent on the theft predicate, the court could only presume the least culpable conduct encompassed by any predicate theft offense. This maneuver is mandated by the modified categorical approach and the Supreme Court’s “least of the acts criminalized” presumption.

b) No Categorical Match to Guidelines Extortion

Guidelines extortion requires “obtaining something of value from another by the wrongful use of force/fear/threat.” Many theft offenses incorporated into Ohio robbery do not require obtaining anything of value at all (e.g., breaking and entering, trespass in a habitation, forgery). Because § 2911.02(A)(2) adds only a force/threat‑of‑harm overlay to whatever theft offense is at issue, a person can violate Ohio robbery without obtaining property. Thus, Ohio robbery is broader than Guidelines extortion.

c) No Categorical Match to Generic Robbery

The Sixth Circuit’s generic robbery requires the “misappropriation of property under circumstances involving immediate danger to the person.” Again, because Ohio robbery can be predicated on theft offenses that do not entail taking or using another’s property (e.g., personating an officer, certain fraud‑like or trespass‑like predicates), a would‑be Ohio robber need not “misappropriate” property to be convicted. Therefore, Ohio robbery is broader than generic robbery as well.

d) Statutory Text Alone Can Demonstrate Overbreadth—No Case Citations Needed

A significant doctrinal clarification: where the state statute’s text plainly covers conduct beyond the generic offense, a defendant is not required to produce state cases proving that “realistic probability.” The court situates this reading within Supreme Court guidance (Taylor 2022; Mellouli) and aligns the Sixth Circuit with the First, Second, Third, Fourth, Seventh, Eighth, Ninth, Tenth, and Eleventh Circuits. Only the Fifth Circuit’s approach diverges. This clarification reduces the evidentiary burden in categorical‑approach litigation when the overbreadth is textual rather than imaginative.

e) Due Process, Indictments, and the Categorical Approach

The court rejected the claim that constitutional fair‑notice concerns compel a presumption that § 2913.02 theft is always the predicate when an indictment is silent. Under Ohio law, an indictment that tracks statutory language suffices for notice; any lack of specificity can be addressed through a bill of particulars. Critically, the due‑process inquiry (what notice the defendant received) differs from the categorical‑approach inquiry (what the jury necessarily found). Injecting a default theft predicate that the jury never necessarily found would itself risk Sixth Amendment problems; the categorical approach forbids that.

3) Impact and Practical Consequences

Immediate sentencing impact: Prior Ohio robbery convictions under § 2911.02(A)(2), where the record does not specify the predicate theft offense, will not qualify as “crimes of violence” under § 4B1.2(a)(2) (2021). Career‑offender enhancements premised on those priors will be unavailable in the Sixth Circuit.
Record‑specific outcomes remain possible: If Shepard documents specify a predicate theft offense that inherently involves misappropriation (e.g., § 2913.02 theft), some Ohio robbery convictions could still match generic robbery under the enumerated‑offenses clause.
Beyond Ohio: The analytical template (double divisibility; least‑culpable‑conduct presumption across an incorporated theft‑offense menu; reliance on statutory text for overbreadth) will shape categorical‑approach litigation involving other “umbrella” statutes that incorporate multiple predicates.
Refined “realistic probability” doctrine: Within the Sixth Circuit, defendants may rely on statutory text alone to demonstrate overbreadth. This reduces the need to marshal examples of state prosecutions where the plain text already shows the mismatch.
Certification practice: The court declined to certify questions to the Ohio Supreme Court, announcing that existing text and caselaw provide a sufficiently “clear and principled” path. Separate opinions (Judges Nalbandian and Thapar) urge certification in future close cases with constitutional implications; the majority signals that certification is unnecessary where the statutory text does the work.
Interaction with the 2023–2024 Guidelines: The Commission added a robbery definition in late 2023 to address other contexts (including Hobbs Act robbery). This case applied the 2021 manual (controlling at the time of sentencing). Even under the new definition, Ohio robbery’s lack of a property‑taking element when the predicate theft offense is unspecified would likely remain problematic.
Sentencing discretion remains: As Judge Nalbandian underscores, the Guidelines are advisory. District courts may vary under § 3553(a) if they conclude a defendant’s conduct warrants it, even when a prior does not qualify under the categorical approach.
Longer‑term reform: Judge Griffin’s dissent flags proposed Sentencing Commission amendments (noticed Jan. 2, 2025) that would move away from the categorical approach in favor of conduct‑based definitions for “crime of violence” and “controlled substance offense.” If adopted, such changes could eventually reduce the salience of disputes like this one for Guidelines purposes (though ACCA and immigration contexts would remain governed by statute and precedent).

Complex Concepts Simplified

Categorical approach: A judicial method that compares the statutory elements of a prior conviction to the elements of a generic offense listed in a federal statute or guideline. Courts do not look at what the defendant actually did; they ask what the statute necessarily required for conviction.
Divisible statute: A law that lists alternative elements, creating multiple crimes (e.g., robbery by (A)(1), (A)(2), or (A)(3)). When a statute is divisible, courts may consult Shepard documents (the indictment, plea colloquy, jury instructions) to determine which alternative formed the basis of the conviction.
Elements vs. means: Elements are facts the prosecution must prove beyond a reasonable doubt (and jurors must unanimously find). Means are illustrative ways of committing an element and need not be unanimously found.
Shepard documents: A narrow set of record materials that courts may review to identify the specific statutory elements of a prior conviction (e.g., indictment, jury instructions, plea colloquy). Courts may not look at police reports or underlying facts.
Enumerated‑offenses clause vs. elements clause: The Guidelines’ “crime of violence” definition includes (i) an elements clause (use, attempted use, or threatened use of physical force) and (ii) an enumerated‑offenses clause (listing crimes such as robbery and extortion). This case addresses only the enumerated‑offenses clause.
Generic offense: A “common denominator” definition derived from common law and a survey of state statutes. A state offense qualifies only if its elements are the same as or narrower than the generic definition.
“Realistic probability”: A concept from Duenas‑Alvarez requiring a defendant to show that the state offense covers nongeneric conduct. Many circuits (now including the Sixth) hold that when the statute’s text plainly covers nongeneric conduct, no case examples are needed.

Precedents and Positions in the Separate Opinions

Judge Nalbandian (concurring in part/dissenting in part): Would have certified to the Ohio Supreme Court whether an unspecified “theft offense” equates to § 2913.02(A) theft and whether omitting a specific theft offense in an indictment raises constitutional concerns. Agrees, however, that given the court’s refusal to certify, the majority’s application of the categorical approach is the better reading of Ohio law. Emphasizes the advisory nature of the Guidelines and district courts’ variance authority.
Judge Ritz (concurring in part/dissenting in part): Agrees on double divisibility and that Ohio robbery does not match Guidelines extortion. Parts ways on “generic robbery”: would define the theft component via a broad “misappropriation” concept that sweeps in Ohio’s theft offenses, and he would not graft a recklessness mens rea onto generic robbery’s force element. He also would require defendants to identify state cases to prove overbreadth.
Judge Griffin (dissenting): Decries the categorical approach as producing “absurd” results divorced from common sense and public safety, and catalogs widespread judicial criticism. Notes prospective Sentencing Commission reform to move away from the categorical approach.
Judge Thapar (dissenting): Advances two alternative paths to affirm the enhancement: (i) treat the theft offenses as means (not elements), which would make Ohio robbery a categorical match with generic robbery; or (ii) if they are elements, apply a state‑law “presumption” that unspecified theft offenses are ordinary theft (§ 2913.02), making Ohio robbery a match for Guidelines extortion. He urges certification to the Ohio Supreme Court, contends the majority’s approach imperils Ohio indictment practices, and argues Ohio law requires at least recklessness for the force element after a 2015 statutory change.

Why This Decision Matters

Cervenak delivers a clear, administrable framework for Ohio robbery under the 2021 Guidelines and beyond:

It cements “double divisibility” for Ohio robbery and requires record‑based specificity of the predicate theft offense to qualify as “robbery” or “extortion” under the enumerated‑offenses clause.
It clarifies that statutory text alone can establish overbreadth where the overbreadth is obvious on the face of the statute, easing the “realistic probability” burden for defendants in the Sixth Circuit.
It respects the categorical approach’s “elements only” premise while rejecting efforts to cure record silence with presumptions about what the underlying offense must have been.
It will materially reduce the availability of the career‑offender enhancement for many defendants with prior Ohio robbery convictions unless the Shepard record ties the conviction to a theft‑predicate involving property misappropriation.

Conclusion

United States v. Cervenak is a consequential en banc decision that realigns Sixth Circuit law with the categorical approach’s first principles and with the dominant circuit view on how to apply Duenas‑Alvarez when statutory text alone reveals overbreadth. The court holds that Ohio robbery under § 2911.02(A)(2) is not a “crime of violence” under the 2021 career‑offender guideline’s enumerated‑offenses clause where the predicate theft offense is unspecified because the statute sweeps in predicates that do not involve obtaining or misappropriating property. The opinion also confirms the robbery statute’s double divisibility and disclaims any constitutional necessity to presume a § 2913.02 theft predicate when the indictment is silent.

Going forward, the decision will:

Require precise Shepard documentation to sustain career‑offender enhancements based on Ohio robbery;
Encourage careful charging and plea practices in state court if future federal sentence consequences are foreseeable;
Guide Sixth Circuit courts to accept text‑based showings of overbreadth without demanding state case exemplars; and
Preserve district courts’ latitude to vary under § 3553(a) if a defendant’s actual conduct warrants a higher sentence despite the categorical mismatch.

In the broader legal context, Cervenak is a measured but influential reaffirmation of the categorical approach’s element‑focused discipline and a clear signal that, at least within the Sixth Circuit, statutory text is king when it plainly outruns a generic or enumerated offense.