Erlinger-Style Jury-Finding Errors Under New Jersey’s Persistent Offender Statute Are Subject to Harmless Error Review (and the Statute Requires Legislative Repair)

1. Introduction

State v. Jamel Carlton sits at the intersection of modern Sixth Amendment sentencing doctrine and New Jersey’s extended-term framework for repeat offenders. The defendant, Jamel Carlton, was convicted by a jury of first-degree aggravated sexual assault and other offenses. At sentencing, the State sought an extended-term sentence under N.J.S.A. 2C:44-3(a) (the “persistent offender” provision). The trial court granted the application based on certified judgments of Carlton’s two prior New York convictions, and imposed an aggregate forty-two-year term subject to NERA.

While Carlton’s appeal was pending, the United States Supreme Court decided Erlinger v. United States, holding that the Fifth and Sixth Amendments require a jury—unanimously and beyond a reasonable doubt—to determine whether prior offenses were “committed on occasions different from one another” for ACCA enhancement purposes. The parties here agreed Carlton’s extended term rested on judge-found facts inconsistent with Erlinger’s constitutional mandate, but disputed the remedy: must such an error be automatically reversible, or may it be affirmed if harmless beyond a reasonable doubt?

The Appellate Division vacated the sentence, reasoning the absence of any jury determination could not be harmless, and it attempted a “saving construction” by effectively substituting “jury” for “court” in the statutory text. The New Jersey Supreme Court granted certification on both sides.

Key issues:

• Whether an Erlinger-type Sixth Amendment error in persistent-offender fact-finding is subject to harmless error review.
• Whether the error in this case was harmless beyond a reasonable doubt on this record.
• What Erlinger implies for the ongoing constitutionality and administration of N.J.S.A. 2C:44-3(a).

Parties: State of New Jersey (appellant/cross-respondent) vs. Jamel Carlton (respondent/cross-appellant).

2. Summary of the Opinion

The Court held that the Sixth Amendment violation—i.e., the sentencing judge (not a jury) finding the predicate facts needed to impose an extended term under N.J.S.A. 2C:44-3(a) after Erlinger—is subject to harmless error review. Applying the federal constitutional harmless-error standard, the Court concluded the error was harmless beyond a reasonable doubt on this record, reinstating the original sentence.

Separately, the Court recognized that N.J.S.A. 2C:44-3(a) as written assigns fact-finding to “the court,” which is “inconsistent with the principles announced in Erlinger.” Without adopting the Appellate Division’s “judicial surgery,” the Court urged the Legislature to amend the statute promptly to align it with controlling federal constitutional doctrine.

3. Analysis

3.1. Precedents Cited (and How They Drove the Outcome)

A. The constitutional jury-factfinding line: from guilt to sentencing

• In re Winship (397 U.S. 358 (1970)) anchors the due process baseline: the State must prove beyond a reasonable doubt every fact necessary to constitute the charged crime. The Carlton Court uses Winship as the constitutional “starting point” for why factfinding allocation (judge vs. jury) matters when punishment turns on additional facts.
• Almendarez-Torres v. United States (523 U.S. 224 (1998)) supplies the “prior conviction” exception: a judge may find the existence of a prior conviction as part of sentencing. Carlton treats Almendarez-Torres as a narrow carve-out that survived later doctrine but only as to the bare “fact of a prior conviction.”
• Apprendi v. New Jersey (530 U.S. 466 (2000)) is the central pivot: except for the fact of a prior conviction, any fact increasing the statutory maximum must be submitted to a jury and proved beyond a reasonable doubt. Carlton adopts Apprendi’s architecture to classify persistent-offender predicates as jury questions when they increase exposure.
• Blakely v. Washington (542 U.S. 296 (2004)) reinforces Apprendi by emphasizing that defendants may insist the prosecutor prove to a jury all facts legally essential to punishment. Carlton uses Blakely to reject the idea that “straightforward” findings can be reassigned to judges merely because they appear easy or document-driven.
• Alleyne v. United States (570 U.S. 99 (2013)) extends Apprendi to facts increasing mandatory minimums. Carlton invokes Alleyne to underline that sentencing exposure includes both ceilings and floors—and that Erlinger fits within that broader principle.
• Erlinger v. United States (602 U.S. 821 (2024)) is the immediate catalyst. It “cements” the limited reach of Almendarez-Torres and holds that even “seemingly straightforward” factual determinations about prior offenses (e.g., whether they occurred on different occasions) must be decided by a unanimous jury beyond a reasonable doubt when they increase sentencing exposure. Carlton treats Erlinger as dispositive on the constitutional defect in New Jersey’s judge-centered persistent-offender finding process.

B. Harmless error vs. structural error: why this violation was reviewable

• Chapman v. California (386 U.S. 18 (1967)) supplies the governing test: a constitutional error may be deemed harmless only if the court is convinced beyond a reasonable doubt the error did not affect the outcome. Carlton applies Chapman directly to the sentencing-enhancement factfinding error.
• Arizona v. Fulminante (499 U.S. 279 (1991)) provides the broader proposition that most constitutional errors can be harmless. Carlton cites it to resist the Appellate Division’s inclination toward automatic reversal.
• Neder v. United States (527 U.S. 1 (1999)) and Johnson v. United States (520 U.S. 461 (1997)) frame the “very limited class” of structural errors and the “strong presumption” that constitutional errors are harmless-error reviewable. Carlton uses Neder and Johnson to classify this as a reviewable error rather than a structural defect requiring automatic reversal.
• Rose v. Clark (478 U.S. 570 (1986)) is cited (via Neder) for the principle that structural errors are those that “necessarily render a trial fundamentally unfair.” Carlton leverages this to distinguish between (i) deprivation of a jury determination (reviewable in certain contexts), and (ii) complete breakdowns that render proceedings inherently unreliable.
• Washington v. Recuenco (548 U.S. 212 (2006)) is the keystone precedent for the remedy question: the failure to submit a sentencing factor to a jury—an Apprendi-style error—is not structural and is subject to harmless error review. Carlton relies on Recuenco to bring Erlinger errors within the established harmless-error framework.
• State v. Camacho (218 N.J. 533 (2014)) shows New Jersey’s own adoption of the Supreme Court’s structural/harmless distinction. Carlton cites it to confirm compatibility between federal harmless-error doctrine and state appellate practice.

C. Post-Erlinger federal appellate practice (persuasive confirmation)

While the U.S. Supreme Court majority in Erlinger was silent on harmlessness, the Carlton Court looked to the emerging consensus among federal circuits applying harmless error review to Erlinger violations:

• United States v. Campbell, 122 F.4th 624 (6th Cir. 2024)
• United States v. Brown, 136 F.4th 87 (4th Cir. 2025)
• United States v. Valencia, 137 F.4th 331 (5th Cir. 2025)
• United States v. Johnson, 114 F.4th 913 (7th Cir. 2024)
• United States v. Bowling, 135 F.4th 1125 (8th Cir. 2025)
• United States v. Rivers, 134 F.4th 1292 (11th Cir. 2025)

The significance of these citations is methodological: they show that appellate courts are treating Erlinger as a species of Apprendi error and therefore applying Recuenco’s harmlessness framework. Carlton adopts that approach for New Jersey’s analogue statute.

D. New Jersey statutory-interpretation precedents (the lens, not the engine)

The Court invoked multiple state decisions to reaffirm its interpretive method—plain text first, extrinsic aids if ambiguity, legislative intent paramount—while recognizing constitutional constraints:

• State v. Gomes, 253 N.J. 6 (2023)
• State v. Hemenway, 239 N.J. 111 (2019)
• State v. Gandhi, 201 N.J. 161 (2010)
• Richardson v. Bd. of Trs., PFRS, 192 N.J. 189 (2007)
• In re Est. of Jones, 259 N.J. 584 (2025)
• DiProspero v. Penn, 183 N.J. 477 (2005)

These cases did not decide the Sixth Amendment question; rather, they framed why the statute’s text (“the court may…if it finds…”) cannot be read in a vacuum when constitutional law has shifted.

E. Retroactivity / “pipeline” framing

• State v. Wessells, 209 N.J. 395 (2012) and Griffith v. Kentucky, 479 U.S. 314 (1987) are cited to define “pipeline” cases: those pending on direct appeal or otherwise not final when a new rule issues. The concept matters here because Erlinger arrived while Carlton’s appeal was pending, making the Sixth Amendment change applicable to his case.

F. The Court’s explicit treatment of prior New Jersey sentencing precedent

• State v. Pierce, 188 N.J. 155 (2006) is directly confronted. Carlton holds that Pierce is abrogated “insofar as it directs a sentencing judge to find, without a jury, the predicate facts required for persistent offender status,” including whether prior offenses were committed on separate occasions. Yet, Pierce “remains intact” on non-Sixth Amendment issues—most importantly, the framework for judicial discretion once eligibility is constitutionally established.

3.2. Legal Reasoning

A. Identifying the constitutional defect after Erlinger

The Court proceeded from a straightforward premise: Carlton’s extended-term eligibility under N.J.S.A. 2C:44-3(a) required factual predicates (age thresholds, two prior convictions on “separate occasions,” “committed at different times,” and a recency requirement tied to the last crime or release from confinement). Those predicate facts increased his sentencing exposure (including the potential of a life term), but they were found by a judge, not a jury.

Under Erlinger, the constitutional line is not merely whether there was a prior conviction, but whether additional factfinding about the nature, timing, separation, or sequencing of past offenses is necessary to increase punishment. Because persistent-offender status demands more than the bare existence of prior convictions—especially whether they were on “separate occasions” / “committed at different times”—the factfinding falls outside the narrow Almendarez-Torres exception and into Apprendi/Blakely/Erlinger territory.

B. Choosing the remedial framework: harmlessness, not automatic reversal

The Appellate Division treated the error as effectively structural because the jury never considered the enhancement predicates at all. The Supreme Court rejected that characterization. Its logic was doctrinal and functional:

• Doctrinal: Recuenco already holds that failure to submit a sentencing factor to a jury is not structural and is subject to harmless error review. The Carlton Court read Erlinger errors as a subtype of Apprendi errors, and thus governed by the same remedial rule.
• Functional (case-specific): The harmless-error inquiry can be performed where the record is complete, the facts are undisputed, and meaningful appellate review is possible. This is consistent with Chapman’s case-setting approach (“in the setting of a particular case”).

C. Applying Chapman: why the error was harmless here

Carlton’s harmlessness holding is highly record-dependent. The Court emphasized:

• The State submitted certified judgments of conviction for the two New York felonies.
• The defense did not object to the documents and conceded the predicate facts.
• The dates showed the convictions were for separate crimes years apart (robbery in 2007; possession of stolen property in 2011), satisfying “committed at different times.”
• The record showed age eligibility (born in 1985; over 21 at the time of the present offense) and the 10-year recency element (release in 2014; current offense in 2018).

On these facts, the Court concluded that “only one outcome would have been possible” had a jury been asked to decide persistent-offender eligibility. Therefore, the Court was convinced beyond a reasonable doubt that the absence of a jury finding did not affect the eligibility outcome in this case.

D. Statutory posture and separation of powers

Having resolved the case on harmlessness, the Court declined to endorse the Appellate Division’s “saving construction” that effectively replaced “the court” with “the jury.” Instead, it took an institutional stance: the statute is now out of alignment with controlling federal constitutional doctrine, and the Legislature should repair it.

The Court supported that call with historical context drawn from the statute’s lineage: the official commentary indicated that assigning extended-term factfinding to the court followed the Model Penal Code view that the issue concerns sentencing, not guilt, and it noted the deliberate departure from the repealed Habitual Offender Act, which provided for a jury trial on prior-conviction allegations absent a waiver.

The upshot is a separation-of-powers message: while courts can decide cases and apply harmless error, they should not be forced into ad hoc procedural invention to keep a statute operational long-term when the statutory text itself points in the constitutionally wrong direction.

3.3. Impact

A. Immediate doctrinal impacts in New Jersey

• Harmlessness is now on the table for Erlinger-based extended-term challenges: defendants can establish a Sixth Amendment violation, but the State may still argue (and sometimes prove) harmlessness beyond a reasonable doubt.
• State v. Pierce is partially abrogated: to the extent it authorized judge-found persistent-offender predicate facts without a jury, it is no longer good law after Erlinger as applied in Carlton. Courts and practitioners must treat persistent-offender eligibility factfinding as constitutionally requiring a jury (or a waiver/admission) even if prior New Jersey practice said otherwise.

B. Practical impacts on sentencing procedure

• Increased procedural complexity: unless and until the Legislature amends N.J.S.A. 2C:44-3(a), trial courts face uncertainty about how, when, and by what mechanism a jury should be asked to decide persistent-offender predicates. Carlton signals that relying on the statute’s current “the court … if it finds …” phrasing risks constitutional error.
• Record-building incentives: because harmlessness depends on an unambiguous record, prosecutors may emphasize certified judgments, clear timelines, and admissions; defense counsel may scrutinize “different times”/occasion separation, dates of release, and whether the documents truly resolve the predicate facts beyond a reasonable doubt.
• Appellate sorting: Carlton implies a two-track future: (i) cases with clean documentary proof may be affirmed as harmless; (ii) cases with ambiguous timing/occasion facts may require resentencing or a constitutionally compliant factfinding process.

C. Legislative impact

The Court’s explicit “call upon the Legislature” is a notable institutional move. It predicts that reliance on harmless-error review is an unstable substitute for a statute that accurately assigns factfinding authority. Carlton thus creates pressure for a statutory amendment that:

• clearly allocates predicate factfinding to a jury (or provides an express waiver/admission mechanism),
• specifies procedures and timing (pretrial notice, bifurcation, special verdict forms, etc.), and
• reduces the risk of inconsistent approaches across vicinages and judges.

4. Complex Concepts Simplified

Persistent offender / extended term (N.J.S.A. 2C:44-3(a))

New Jersey allows an enhanced (“extended term”) sentence for certain repeat offenders. Under N.J.S.A. 2C:44-3(a), a defendant can qualify as a “persistent offender” if (among other requirements) they have at least two prior convictions on separate occasions, committed at different times, and the most recent conviction (or release from confinement) is within a specified look-back period. Qualifying exposes the defendant to a higher sentencing range than would otherwise apply.

The Apprendi rule

Apprendi v. New Jersey says: if a fact increases the maximum penalty a defendant can receive, that fact must be found by a jury beyond a reasonable doubt—except for the bare fact of a prior conviction.

The Almendarez-Torres “prior conviction” exception

Almendarez-Torres v. United States permits a judge to find the existence of a prior conviction at sentencing. But Erlinger v. United States emphasizes the exception is narrow: judges cannot decide additional facts about prior crimes (such as whether they occurred on different occasions) when those facts increase sentencing exposure.

Harmless error vs. structural error

• Harmless error (from Chapman v. California): an appellate court may affirm despite a constitutional mistake if it is convinced beyond a reasonable doubt the mistake did not change the outcome.
• Structural error (from Neder v. United States and related cases): a rare kind of error so fundamental that reversal is automatic, without asking whether it changed the result.

Carlton holds the Erlinger-type error here is in the harmless-error category, not structural, following Washington v. Recuenco.

“Saving construction” / “judicial surgery”

This refers to a court interpreting (or effectively rewriting) statutory language to avoid unconstitutionality. The Appellate Division attempted this by substituting “jury” for “court.” The Supreme Court did not decide whether that approach was permissible because it resolved the case via harmless error, but it emphasized the statute remains textually inconsistent with Erlinger and should be amended legislatively.

Pipeline cases

A “pipeline” case is one still on direct appeal or otherwise not final when a new rule is announced (as described in State v. Wessells and Griffith v. Kentucky). Carlton benefitted from Erlinger because his case was still pending when Erlinger was decided.

5. Conclusion

State v. Jamel Carlton establishes two durable propositions for New Jersey sentencing law after Erlinger v. United States: (1) when a judge—not a jury—finds persistent-offender predicate facts that increase sentencing exposure, the error is constitutional, but (2) it is generally subject to harmless error review rather than automatic reversal. On Carlton’s uncontested documentary record, the Court found the error harmless beyond a reasonable doubt and reinstated the extended-term sentence.

At the same time, the Court delivered a clear warning: N.J.S.A. 2C:44-3(a), as written, assigns factfinding to “the court” in a way that conflicts with Erlinger’s jury-right mandate. The opinion therefore functions not only as a case-specific harmlessness decision, but also as a roadmap—and an urgent prompt—for legislative amendment to restore clarity, constitutional compliance, and uniform administration of persistent-offender sentencing in New Jersey.