“Harmless-Error” after Erlinger: The Sixth Circuit’s Two-Track Approach in United States v. Nathaniel Durham

1. Introduction

The consolidated appeal of United States v. Durham & Barnes (Nos. 23-5162/5173, decided 14 Aug 2025) required the Sixth Circuit to apply the Supreme Court’s recent decision in Erlinger v. United States, 602 U.S. 821 (2024). Erlinger held that the Fifth and Sixth Amendments entitle a defendant to have a jury—not a judge—decide whether the three predicate felonies supporting Armed Career Criminal Act (ACCA) enhancement were “committed on occasions different from one another.”

Because the district court sentenced both defendants before Erlinger, the appellate panel had to decide whether the constitutional error (the judge rather than a jury deciding the occasions issue) was harmless. The court reached opposite results: it affirmed Nathaniel Durham’s 15-year ACCA sentence, finding the error harmless beyond a reasonable doubt, but vacated Phillip Barnes’s identical sentence and remanded for resentencing with a jury determination.

2. Summary of the Judgment

• Nathaniel Durham – Ten prior Kentucky second-degree burglaries. The court held that, even under a jury standard, no reasonable juror could find that at least three of the 2010–2011 burglaries were part of the same “occasion.” Harmless error; sentence affirmed.
• Phillip Barnes – Four prior armed robberies of Family Dollar stores in Louisville in 2008–2009. Given temporal proximity, common modus operandi, and shared targets, a reasonable jury could harbor doubt that the August 2008 robberies constituted separate occasions. Error not harmless; sentence vacated and remanded.

The decision clarifies how Sixth-Circuit courts should apply Erlinger-type harmless-error review and synthesises a growing body of intra-circuit precedent (Campbell, Robinson, Cogdill, Kimbrough, Thomas).

3. Detailed Analysis

3.1 Precedents Cited and Their Influence

1. Wooden v. United States, 595 U.S. 360 (2022) – Adopted a “holistic, multi-factor” approach (time, place, character, relationship) to determine “occasions.” Undercut earlier Sixth-Circuit bright-line tests.
2. Erlinger v. United States, 602 U.S. 821 (2024) – Shifted fact-finding to the jury; any judicial determination is constitutional error subject to Chapman harmless-error review on direct appeal.
3. United States v. Campbell, 122 F.4th 624 (6th Cir. 2024) – First post-Erlinger decision to apply harmless-error; stressed that large temporal/physical gaps may alone be decisive.
4. United States v. Robinson, 133 F.4th 712 (6th Cir. 2025) – Found error harmless where predicates were radically different in nature (murder vs. drugs) and decades apart.
5. United States v. Cogdill, 130 F.4th 523 (6th Cir. 2025) – Error not harmless; sparse record and same-county drug crimes three months apart left room for jury doubt.
6. United States v. Kimbrough, 138 F.4th 473 (6th Cir. 2025) – Error not harmless; two burglaries nine days apart, two miles apart, same modus operandi.
7. United States v. Thomas, 142 F.4th 412 (6th Cir. 2025) – Error harmless where offenses occurred years apart and involved different victims.

3.2 Legal Reasoning of the Panel

The court employed a two-step framework:

1. Identify the constitutional error – Judge, not jury, made “different occasions” findings.
2. Apply Chapman harmless-error test – Ask whether, on the whole record, the government proved beyond a reasonable doubt that any rational jury would necessarily find the predicates occurred on distinct occasions.

For Durham, the record contained dates, addresses and victim information for each burglary. Four sample offenses (Oct 19 2010 vs. Jan 27 2011) were three months and 15 miles apart, involving unrelated victims and no evidence of a single continuous spree. The majority analogised to Campbell & Thomas, concluding that time/place factors “decisively” separated the occasions.

For Barnes, the robberies took place within a 25-day window, involved the same retail chain (two at the exact same store) and identical firearm tactics; all four counts were disposed of together. Relying heavily on Kimbrough and Cogdill, the panel found enough overlap for a jury to disagree. Therefore the government could not meet the Chapman burden.

3.3 Impact of the Decision

• Cements a fact-sensitive, defendant-specific approach – Time and distance remain powerful but not exclusive indicators; record completeness is critical.
• Clarifies split within Sixth Circuit precedent – The opinion attempts to reconcile the “pro-government” line (Campbell/Thomas/Robinson) with the “pro-defendant” line (Cogdill/Kimbrough).
• Plea vs. Trial distinction – The government urged adoption of the Fourth Circuit’s Brown standard for guilty-plea cases, but the panel deferred; future litigation will likely test whether pleas require an even higher harmless-error showing.
• Practical Guidance for District Courts – After Durham, sentencing courts must assemble a robust record (dates, addresses, offense narratives) if they hope to survive harmless-error review when a jury finding is absent.
• Potential for Supreme Court review – The tension between judges BUSH and MOORE’s concurring/dissenting opinions signals doctrinal ferment; SCOTUS may revisit how much weight time/place can bear alone.

4. Complex Concepts Simplified

Armed Career Criminal Act (ACCA)

Federal statute adding a 15-year mandatory minimum to §922(g) felon-in-possession offenses when the defendant has three prior “violent felony or serious drug offense” convictions committed on different occasions.

Categorical Approach

Method for deciding whether a prior conviction is a “violent felony” or “serious drug offense” by looking at the statute, not underlying facts.

Different-Occasions Requirement

Statutory phrase in §924(e)(1) requiring that the three predicate crimes be separate criminal episodes. After Erlinger, a jury must decide this beyond a reasonable doubt.

Harmless-Error Doctrine (Chapman)

An appellate court can affirm despite constitutional error if the government proves the error did not affect the outcome beyond a reasonable doubt.

Wooden Factors

1) Time; 2) Place; 3) Character; 4) Relationship (common purpose, scheme, or victim overlap). None is automatically controlling.

5. Conclusion

United States v. Durham crystallises the Sixth Circuit’s post-Erlinger stance: harmless-error review of ACCA occasion findings is intensely factual, record-driven, and may lead to divergent outcomes for co-defendants. The case admonishes prosecutors and district judges to present (or demand) granular evidence on prior crimes, and reminds defense counsel that contesting the “different occasions” element remains fertile ground—especially where predicates are temporally or thematically linked.

Going forward, practitioners should expect: (1) routine jury instructions on the occasions question; (2) aggressive appellate scrutiny of records compiled pre-Erlinger; and (3) continued debate over how much daylight remains between a “reasonable juror” standard and the panel’s own linguistic intuition about what constitutes a single criminal “occasion.”