Indicia-of-Reliability Standard for Double-Hearsay Co-Conspirator Statements Supporting U.S.S.G. § 2K2.1(b)(1) Firearm-Count Enhancements

Case: United States v. Roache (1st Cir. Mar. 30, 2026) Court: U.S. Court of Appeals for the First Circuit Panel: Aframe, Howard, and Dunlap, JJ.

1. Introduction

In United States v. Roache, the First Circuit addressed when a sentencing court may rely on a co-conspirator’s unsworn, out-of-court statements—relayed through an agent summary (i.e., double hearsay)—to increase a defendant’s advisory guideline range under the firearm-count enhancement in U.S.S.G. § 2K2.1(b)(1).

The defendant, Aizavier Roache, pleaded guilty (without a plea agreement) to conspiring to traffic firearms in violation of 18 U.S.C. § 933(a)(1), (3). At sentencing, the district court applied a six-level enhancement after crediting ATF-interview statements by co-conspirator Travon Brunson that Brunson had purchased twenty-four additional firearms for Roache in 2020–2021—uncharged conduct that, when added to the six charged firearms, pushed the total above the threshold for the six-level increase.

The core issue on appeal was procedural reasonableness: whether the district court clearly erred or otherwise abused its discretion in finding Brunson’s 2021 interview statements sufficiently reliable to support the enhancement by a preponderance of the evidence.

2. Summary of the Opinion

The First Circuit affirmed. It held that the district court did not abuse its discretion in concluding that Brunson’s 2021 statements bore “sufficient indicia of trustworthiness” to satisfy due process and the preponderance standard at sentencing, even though the statements were unsworn, constituted double hearsay, and were not corroborated by the same type of direct documentation later obtained for the 2023 transactions.

The court emphasized three reliability anchors: (1) objective evidence that multiple Brunson-purchased firearms were recovered in Massachusetts; (2) the detailed nature of Brunson’s account, including identifying Roache from a photograph; and (3) strong similarity between the 2021 described trafficking method and the later, documented 2023 conduct (texts, money transfers, purchases, and physical transfers).

3. Analysis

3.1. Precedents Cited

The opinion is principally a sentencing-reliability and standard-of-review decision, built from a chain of First Circuit and Supreme Court authorities:

  • United States v. Rondón-García and United States v. Flores-Machicote: The court reaffirmed that sentencing review recognizes “substantial discretion” in the district court and that uncharged conduct may be considered if proven by a preponderance of the evidence. Rondón-García also supports the propositions that the Federal Rules of Evidence and the Confrontation Clause do not constrain sentencing in the same way they do trials, but that due process still demands reliable information.
  • Buford v. United States: Cited to frame appellate review as deferential both to factfinding (clear error) and to guideline application (due deference), reinforcing that reliability calls are typically for the sentencing judge.
  • United States v. Acevedo-Osorio and United States v. Vargas-Martinez: Used to situate the procedural-reasonableness challenge within the “abuse-of-discretion framework.”
  • United States v. Carvajal: Supplies the broad evidentiary principle at sentencing: a court may rely on “virtually any dependable information” so long as it carries sufficient indicia of reliability.
  • United States v. Franklin and United States v. Castillo-Torres: Central to the holding. Franklin is cited for reviewing reliability determinations for abuse of discretion; Castillo-Torres (also quoted via United States v. Rosa-Borges) anchors the due-process boundary: reversible error occurs if a sentence rests “based solely on unreliable evidence.”
  • United States v. Rosa-Borges and United States v. Ramos-Carreras: These cases express the due process right to be sentenced on “accurate and reliable information,” a framing that the panel used to articulate the governing test.
  • United States v. Colón-Maldonado, United States v. Oquendo-Rivera, and United States v. Fontanez: These precedents provide concrete reliability markers. Colón-Maldonado highlights that a “detailed and nonconclusory” account is a reliability indicator; Oquendo-Rivera recognizes that consistency and objective corroboration bolster trustworthiness; Fontanez underscores the probative value of “objective evidence” corroborating a witness account.
  • United States v. Forbes and Anderson v. Bessemer City: Invoked to distinguish a scenario requiring remand for inconsistent credibility determinations. The panel relied on Forbes to explain that the district court here adequately articulated why it credited core statements while rejecting ancillary ones, and it used the Anderson standard (“internally inconsistent or implausible on its face”) to reject the claim that Brunson’s narrative was facially suspect.
  • United States v. Goncalves: Referenced to acknowledge intra-circuit tension in how standards are articulated; the panel avoided refining doctrine because the dispositive question was factual reliability and deference controlled.
  • Shabshelowitz v. R.I. Dep't of Pub. Safety: Cited on waiver/forfeiture principles when Roache raised a point on appeal that was not pressed below.
  • United States v. Guía-Sendeme, United States v. Melendez-Hiraldo, and United States v. Coplin-Benjamin: These cases were used to justify the sources for the factual narrative (PSR and sentencing record) given a guilty plea without a plea agreement.
  • United States v. Andino-Morales: Cited alongside Buford for “due deference” to the district court’s guideline application and fact-bound judgments.

3.2. Legal Reasoning

The court’s reasoning proceeds in a structured sentencing-evidence sequence:

  1. Identify the governing evidentiary constraint at sentencing (reliability, not admissibility). The panel reiterated that sentencing courts may consider broad information—including hearsay and uncharged conduct—because the trial admissibility rules do not control. The limiting principle is due process: the information must be “accurate and reliable,” and findings must be supported by a preponderance.
  2. Frame the appeal as a deferential review of a fact-intensive reliability determination. The panel treated the key issue as factual (“relating to the reliability of the statements”), meaning the district court’s judgment call receives “considerable deference,” reviewed for abuse of discretion.
  3. Apply “sufficient indicia of trustworthiness” to double hearsay. The panel treated the ATF summary of Brunson’s statements as double hearsay but held it could still support an enhancement if reliability indicia exist. Importantly, Roache did not challenge the accuracy of the agent’s summary (the second layer), narrowing the dispute to Brunson’s truthfulness/accuracy.
  4. Identify concrete corroboration and consistency. The court endorsed the district court’s reliance on:
    • Recovered firearms link: fifteen Brunson-purchased firearms recovered in Massachusetts, consistent with trafficking into the state.
    • Detail: Brunson’s specific account (nickname “Boston,” photo identification of Roache, method of selecting guns, payment, meetings, transfers).
    • Modus operandi match: the 2023 documented transactions mirrored the 2021 described pattern (texts, transfers, purchases, handoffs), supporting the truth of the earlier narrative.
  5. Allow partial credibility: rejecting ancillary claims does not negate core reliability. A key practical holding is that a sentencing judge may credit “core elements” of a co-conspirator’s account (the firearm purchases for the defendant) while rejecting other portions (no profit; only buyer was Roache). The panel distinguished United States v. Forbes because the district court here explained the basis for differential crediting rather than making unexplained, inconsistent findings.
  6. Reject “missing records” as dispositive where other indicia suffice. Roache urged a negative inference from the absence of older texts/bank records corroborating 2020–2021 dealings. The district court declined (noting limited phone history), and the panel agreed: the government need only meet the preponderance standard, and corroboration can be circumstantial and pattern-based.
  7. Address subsidiary attacks without letting them collapse the reliability finding. The panel found that (a) Brunson’s later criminal conduct created “cause for concern” but was not fatal given corroboration; (b) Brunson’s undercount of total firearms purchased did not meaningfully conflict with the claim he purchased at least thirty for Roache; and (c) an inaccurate detail about Roache’s residence did not undermine the transactional narrative.

Doctrinally, Roache operationalizes a common sentencing principle: reliability can be established by a web of circumstantial corroboration (objective recoveries + detailed narrative + later similar conduct), even when direct documentary evidence for the earlier period is absent.

3.3. Impact

Three likely effects follow for First Circuit sentencing practice in firearm-trafficking and similar conspiracies:

  • Broader usability of cooperator statements for guideline enhancements—if pattern corroboration exists. The decision confirms that co-conspirator interview statements, even unsworn and summarized by agents, can support substantial enhancements when tied to objective facts (e.g., recoveries) and consistent later conduct.
  • Encouragement of “modus operandi” corroboration. Investigations that can prove a later set of transactions with texts, money transfers, travel, or purchase records may indirectly corroborate earlier alleged conduct by demonstrating the same method of operation.
  • Partial-credit credibility findings are defensible if explained. Sentencing courts may parse a witness’s account—crediting the central factual assertion while rejecting self-serving or implausible details—so long as the court articulates why, reducing the risk of Forbes-type remands.

4. Complex Concepts Simplified

  • “Procedural reasonableness”: Whether the sentencing process was done correctly—especially whether the guideline range was calculated properly and based on reliable facts.
  • “Preponderance of the evidence”: The fact is more likely true than not (just over 50%), a lower burden than “beyond a reasonable doubt.”
  • “Hearsay” and “double hearsay”: Hearsay is an out-of-court statement offered for its truth. Double hearsay is hearsay within hearsay—here, Brunson’s statements repeated through an agent’s summary. At sentencing, hearsay may be considered if reliable.
  • “Indicia of reliability / trustworthiness”: Signs the information is dependable—detail, consistency, objective corroboration (like recovered guns), and confirmation through later proven conduct.
  • U.S.S.G. § 2K2.1(b)(1) firearm-count enhancement: The guideline increases offense level based on the number of firearms involved (here, moving from 6 firearms to 30 firearms raised the enhancement from 2 levels to 6 levels).
  • “Abuse of discretion” review: A highly deferential appellate standard. Even if the appellate court might have weighed evidence differently, it will affirm unless the district court’s call was outside the range of reasonable decisions.

5. Conclusion

United States v. Roache solidifies that, in the First Circuit, a sentencing court may apply a significant firearm-count enhancement based on a co-conspirator’s unsworn, agent-summarized statements about uncharged conduct when the statements are supported by sufficient indicia of reliability—particularly objective firearm recoveries, detailed narrative features (including identification), and demonstrated similarity to later documented transactions. The decision also confirms that district courts may credit the core of a witness’s account while rejecting ancillary assertions, provided the court explains the basis for that parsing.