Introduction

In Purcell v. United States, the Second Circuit reversed the denial of a 28 U.S.C. § 2255 petition and remanded for further proceedings after holding that appellate counsel was constitutionally ineffective for omitting a plainly available and likely dispositive venue challenge to one count of conviction. The panel—Judges Park, Pérez, and Nathan (opinion by Judge Nathan; dissent by Judge Park)—also declined to apply the concurrent sentence doctrine (CSD) to avoid reaching the merits, underscoring that, in the § 2255 context, non-life concurrent sentences present potential custody and collateral consequences that make CSD inapposite absent “reasonable certainty” of no adverse effects.

The petitioner, Lavellous Purcell, had been convicted of multiple sex-trafficking-related offenses. On direct appeal in 2020, the Second Circuit vacated his enticement conviction (Count One) for lack of venue in the Southern District of New York (SDNY) but affirmed the remainder, including a § 2421(a) transportation conviction (Count Two) tied to a single victim (Samantha Vasquez). Appellate counsel had challenged venue as to Count One but not Count Two. In this collateral proceeding, Purcell argued that omission was ineffective assistance; the district court disagreed. The Second Circuit now holds the omission was both deficient and prejudicial under Strickland, and it clarifies when CSD should not be invoked on collateral review.

Two legal developments stand out:

• Appellate counsel’s failure to raise an “obvious and significant” venue challenge to an individual count—particularly when a parallel venue theory succeeded on another count—can be objectively unreasonable and prejudicial under Strickland’s two-prong test.
• The concurrent sentence doctrine should not pretermit review on § 2255 where the petitioner is serving non-life concurrent sentences and cannot be assured with reasonable certainty of no sentencing impact or collateral consequences.

Summary of the Opinion

Purcell’s convictions stemmed from a multi-year prostitution enterprise. He received 216 months (18 years), including concurrent terms: 216 months on Counts One and Five, 120 months on Count Two, and 60 months on Counts Three and Four. On direct appeal, the Second Circuit vacated Count One for lack of venue because the record did not show essential conduct or victim travel in SDNY as to the victims—including Vasquez. Appellate counsel did not press a venue challenge to Count Two; the court treated it as abandoned and affirmed that count.

In this § 2255 action, Purcell argued appellate counsel was ineffective for failing to challenge venue on Count Two. The Second Circuit:

• Declined to apply the concurrent sentence doctrine. Because Purcell is serving non-life concurrent terms, resentencing could change custody time (e.g., under Pepper’s consideration of post-sentencing rehabilitation), and collateral consequences cannot be discounted “with reasonable certainty.”
• Held counsel’s omission was objectively unreasonable. The venue defect on Count Two was “significant and obvious” given the identical evidentiary record regarding Vasquez and the successful venue challenge to Count One. Strategic justifications (brief length, prioritization) did not explain omitting an argument that could have been made succinctly and in parallel.
• Found prejudice. There is a reasonable probability that, had venue been challenged on Count Two, the court would have reversed that count on direct appeal, because the record lacked evidence that either Purcell or Vasquez committed essential conduct in SDNY.

The court reversed the district court’s denial of § 2255 relief and remanded for further proceedings consistent with its opinion. Judge Park dissented, arguing that counsel’s choices were objectively reasonable and that hindsight improperly drove the majority’s Strickland analysis.

Analysis

Precedents Cited and Their Roles

• Strickland v. Washington, 466 U.S. 668 (1984): The governing two-prong test (deficient performance and prejudice) for ineffective assistance. The panel applies Strickland to appellate counsel (Mayo) and emphasizes assessing performance from counsel’s perspective at the time, with prejudice assessed in light of hindsight.
• Mayo v. Henderson, 13 F.3d 528 (2d Cir. 1994): Establishes that omitting “significant and obvious” appellate issues while pursuing clearly weaker ones can be ineffective. This is the backbone of the performance analysis here.
• Lynch v. Dolce, 789 F.3d 303 (2d Cir. 2015): Reinforces that counsel need not raise every nonfrivolous issue, but omission can be unreasonable if obvious and important. Also confirms prejudice in the appellate context turns on the probability the argument would have succeeded on appeal, not the final sentence.
• Jackson v. Leonardo, 162 F.3d 81 (2d Cir. 1998): Focus on whether the omission reflects plausible strategy; omitting a “sure winner” while advancing weaker issues can be unreasonable.
• Cornell v. Kirkpatrick, 665 F.3d 369 (2d Cir. 2011): Ineffective assistance where trial counsel failed to object to venue; omission stemmed from oversight and lack of legal research, not strategy. The court analogizes Purcell’s situation, emphasizing that venue errors mandate reversal of the affected count(s).
• Brown v. United States, 167 F.3d 109 (2d Cir. 1999), and Farhane v. United States, 121 F.4th 353 (2d Cir. 2024) (en banc): Permit resolving Strickland without remand when the record—here, counsel’s own filing admitting nonstrategic oversight—negates strategic explanations.
• Harrington v. Richter, 562 U.S. 86 (2011) and Yarborough v. Gentry, 540 U.S. 1 (2003): Caution against hindsight bias; presume strategic choice when counsel focuses on some issues to the exclusion of others. The dissent invokes these to argue for deference to counsel’s prioritization; the majority distinguishes them in light of counsel’s admission and the obviousness of the omitted argument.
• Venue authorities: United States v. Ramirez, 420 F.3d 134 (2d Cir. 2005); United States v. Davis, 689 F.3d 179 (2d Cir. 2012); United States v. Lange, 834 F.3d 58 (2d Cir. 2016); United States v. Thompson, 896 F.3d 155 (2d Cir. 2018); United States v. Brennan, 183 F.3d 139 (2d Cir. 1999); United States v. Bozza, 365 F.2d 206 (2d Cir. 1966); United States v. Zeuli, 137 F.2d 845 (2d Cir. 1943). These anchor the rule that the government must prove venue by a preponderance for each count; venue lies where essential conduct elements occur; improper venue is reversible error for the affected count.
• United States v. Mi Sun Cho, 713 F.3d 716 (2d Cir. 2013): Defines “transportation” under § 2421(a) broadly; used by the court to show the sufficiency challenge to Count Two was relatively weak compared to the potent venue argument.
• Standard of review: McCloud v. United States, 987 F.3d 261 (2d Cir. 2021); United States v. Melhuish, 6 F.4th 380 (2d Cir. 2021).
• Concurrent sentence doctrine: Kassir v. United States, 3 F.4th 556 (2d Cir. 2021); Al-’Owhali v. United States, 36 F.4th 461 (2d Cir. 2022); United States v. Vargas, 615 F.2d 952 (2d Cir. 1980). The panel reads these to allow (but not require) application of CSD on collateral review only when custody time and collateral consequences are certainly unaffected. Distinguishes prior applications involving life sentences. Pepper v. United States, 562 U.S. 476 (2011) underscores resentencing variability.
• United States v. Sabhnani, 599 F.3d 215 (2d Cir. 2010): Articulates deference to jury verdicts on sufficiency; helps show why the sufficiency challenge to Count Two was comparatively weak.
• Prior Purcell panel decision: United States v. Purcell, 967 F.3d 159 (2d Cir. 2020): Held venue lacking on Count One as to all victims, including Vasquez, and noted the record did not place Vasquez in SDNY for work done for Purcell. That record baseline makes the Count Two venue omission “obvious.”

Legal Reasoning

1) The Concurrent Sentence Doctrine Does Not Bar Review

The court underscores that CSD is discretionary and rooted in judicial convenience, not a jurisdictional bar. On § 2255 collateral review, it applies only when (a) the challenge will have no effect on custody time and (b) the unreviewed conviction will not yield additional adverse collateral consequences. Applying Vargas, and distinguishing Kassir and Al-’Owhali (both involving life sentences), the panel stresses:

• Resentencing after partial vacatur can change custody time (considering new information and rehabilitation under Pepper).
• Non-life terms leave room for future collateral consequences (recidivist enhancements, credibility/ impeachment, pardon considerations, stigma) that cannot be discounted with “reasonable certainty.”
• The government cited no precedential case applying CSD to avoid a merits review of a conviction in a non-life sentence context.

Result: The court reaches the merits instead of invoking CSD.

2) Strickland’s Performance Prong: Omission of an Obvious, High-Value Venue Argument

The panel applies Mayo’s “significant and obvious” standard to appellate omissions. Several facts converge to make the omission objectively unreasonable:

• The government’s venue proof for Count Two turned on the same victim (Vasquez) and the same evidentiary record in which the prior panel found “nothing in the record” placing her in SDNY for relevant activity.
• Venue is required on a count-by-count basis. A successful venue challenge results in reversal of that count, which is a significant remedy even if concurrent sentences render the practical effect uncertain.
• Appellate counsel had already raised and won a venue argument on Count One. A companion venue argument for Count Two could have been brief, drawing the court’s attention to the same evidentiary gap regarding Vasquez’s presence or essential conduct in SDNY.
• Counsel’s chosen alternative—sufficiency under § 2421(a)—was, in context, weaker (Mi Sun Cho’s broad reading of “transport” undermined it), reinforcing that the omitted venue argument was the stronger play.

The court rejects page-limit and prioritization rationales as post hoc. Counsel’s own petition for rehearing stated the omission was an oversight with no strategic reason. Given that admission and the record’s clarity, the court treats the omission as “sheer neglect,” not plausible strategy. Under Brown and Farhane, no remand was necessary to take testimony from counsel.

3) Strickland’s Prejudice Prong: Reasonable Probability of a Different Outcome on Appeal

The prejudice inquiry focuses on the probability the omitted claim would have succeeded on appeal, not on whether the overall sentence would have changed post-remand. The court reasons:

• On the trial record, venue was as deficient for Count Two as for Count One. The government presented no evidence that either Purcell engaged in essential transport conduct in SDNY or that Vasquez traveled in SDNY for the relevant purposes.
• Accordingly, there is a reasonable probability the court would have reversed the Count Two conviction had venue been raised on direct appeal.

The Dissent

Judge Park would affirm, emphasizing deference to counsel’s strategic prioritization. His key points:

• Strickland demands objective reasonableness, not reliance on counsel’s subjective mea culpa. Richter and Yarborough urge deference to counsel’s tactical winnowing, especially under word limits.
• Appellate counsel reasonably prioritized arguments targeting the longest sentences (Counts One and Five) and global relief (Fourth Amendment and Confrontation Clause challenges). Count Two’s venue win offered less upside, particularly because venue dismissals are without prejudice (citing Smith v. United States, 599 U.S. 236 (2023)), allowing retrial elsewhere.
• The Count One venue theory focused on Purcell’s conduct location; a Count Two venue theory would have required developing a different factual thread (the victim’s movements), thus not a mere “add-on paragraph.” Combined with the district court’s swift denial of a post-trial venue motion, the issue was not a “dead-bang” winner at the time.

The majority responds, implicitly and explicitly, that counsel’s own admission and the record’s clarity negate plausible strategic explanations; that a concise, parallel Count Two venue argument was readily available; and that the law measures prejudice by appeal success probability, not sentence impact.

Impact

Purcell will have ripple effects across several domains:

• Appellate practice and Strickland claims
  • Heightened attention to “parallel” issues: When a winning theory (e.g., venue) applies to multiple counts on the same record, omitting it for one count may be deemed objectively unreasonable.
  • Prejudice framed at the appellate stage: Litigants and courts should evaluate whether the omitted claim likely would have succeeded on appeal, irrespective of whether resentencing later keeps the aggregate term unchanged.
  • Counsel admissions: Where the record contains explicit statements that an omission was nonstrategic oversight, courts may resolve Strickland without remand for testimony.
• Venue litigation
  • Reaffirmation that venue is count-specific and tied to essential conduct elements. Government must prove, by a preponderance, that either the defendant’s essential conduct or the victim’s travel in response to that conduct occurred in the trial district.
  • Documentary-only proofs linking a victim to numerous jurisdictions, but not to the trial district, are unlikely to carry the day.
• Concurrent sentence doctrine in § 2255
  • Clarifies a practical limit: CSD will rarely bar review where the petitioner serves non-life concurrent sentences. Potential resentencing dynamics (Pepper) and Vargas collateral consequences make it difficult to say with “reasonable certainty” that no adverse effects will follow.
  • Expect fewer CSD dismissals of conviction challenges in non-life concurrent cases; courts will more often reach the merits.
• Prosecutorial and defense strategy
  • Prosecutors should build venue proof count-by-count, anticipating collateral challenges premised on venue gaps for individual victims.
  • Defense counsel should systematically track whether each count’s essential conduct is anchored in the trial district and avoid asymmetrical briefing that leaves a parallel venue argument on the table.

Complex Concepts Simplified

• Venue (criminal): The constitution and Rule 18 require trial in the district where the crime was committed. The government must prove, for each count, by a preponderance (more likely than not) that essential conduct elements occurred in that district.
• Essential conduct element: The actions that constitute the crime itself (e.g., enticement acts; transporting a person for prostitution), as distinct from circumstances or consequences.
• 28 U.S.C. § 2255: A federal mechanism enabling a person in custody to challenge the legality of a sentence or conviction, often on constitutional grounds like ineffective assistance of counsel.
• Ineffective assistance of counsel (Strickland): A two-part test requiring (1) deficient performance—lawyering falling below an objective standard of reasonableness—and (2) prejudice—a reasonable probability that, but for the error, the result would have been different. For appellate claims, prejudice asks whether the omitted argument likely would have succeeded on appeal.
• Concurrent sentence doctrine: A discretionary doctrine permitting courts to decline reviewing a conviction when the sentence runs concurrently with an unchallenged valid sentence and review would have no effect on custody or collateral consequences. In § 2255 cases with non-life sentences, Purcell cautions against its use.
• Pepper resentencing principle: At resentencing after partial vacatur, courts may consider new information such as post-sentencing rehabilitation, which can change the length of custody.

Conclusion

Purcell advances two important propositions in Second Circuit law. First, in the appellate-ineffectiveness context, failing to raise a succinct and clearly available venue challenge to an individual count—especially where a companion venue argument has succeeded—can be constitutionally deficient and prejudicial. The prejudice inquiry is anchored in the likelihood of success on appeal, not the ultimate sentence after remand. Second, the court sharply limits the concurrent sentence doctrine’s gatekeeping role in § 2255 conviction challenges involving non-life concurrent sentences, recognizing that resentencing dynamics and collateral consequences usually foreclose the “reasonable certainty” required to bypass merits review.

The practical lesson is twofold. For the government, venue must be proven count-by-count with concrete evidence tying essential conduct to the trial district. For defense counsel, symmetry in issue-spotting matters: if venue is vulnerable for one count on a given record, it is likely vulnerable for parallel counts involving the same victim and evidence. Purcell thus reinforces both the rigor of venue proof and the professional norms governing appellate advocacy, while preserving meaningful access to collateral review notwithstanding concurrent sentences.