Browse Category

Federal Corner

Federal Corner: Conflict of Interest Arising from Dual Representation


Sheperd I: 27 F. 4th 1075 (5th Cir. 2022)

Sheperd II: 44 F.4th 305, 2022 U.S. App. LEXIS 22357; 2022 WL 3274129 (5th Cir. August 11, 2022)

In a couple of recent opinions, the Fifth Circuit Court of Appeals clarified the rules regarding ineffective assistance of counsel involving conflicts of interest. The decisions address the actions of an attorney who breached his duty of utmost loyalty and harmed a client. But, the decisions also raise the possibility that more ethical attorneys might make similar mistakes much more innocently.

Let’s start at the end with a quick recap of the facts cited by the Court in the second case. Ann Sheperd, the owner of a home‑health agency, was convicted of Medicare fraud. On appeal she complained that her Sixth Amendment right to effective assistance of counsel had been violated. Turns out, Sheperd’s pretrial counsel was also representing one of the Government’s star witnesses. The Court Remanded the case for an evidentiary hearing on whether Sheperd’s lawyer’s conflict of interest—a conflict the Government knew about—adversely affected his representation. The district court’s answer: Yes. (Sheperd II).

Clearly, the position occupied by Sheperd’s counsel is not one that a competent lawyer with a grasp of ethical principles and a casual understanding of conflict of interest would occupy. However, due to the complexity of the rules that address conflict of interest, defense counsel does not have to be oblivious to ethical obligations to occupy similar territory. In addition to being a firm reminder of the duty of utmost loyalty owed to a client, the case is important for how it addressed the second prong of the Strickland analysis (harm), and for the remedy available for this type of harm. If you ever thought about writing a writ, or just hope to avoid being the legitimate subject of one, read these cases.


A.   The Facts

The facts of the case are set out brilliantly in Sheperd I. However, a little editing has been done for brevity, clarity, and to avoid further shaming of the lawyers. The quotes from both cases appear without the footnotes, citations, and quotation marks present in the opinions.

Ann Sheperd owned and operated a home‑health agency. In June 2016, a grand jury indicted her (and several others) with conspiracy to commit healthcare fraud. Sheperd retained counsel. The district court set trial for August 2016. A month later, Sheperd replaced her counsel. The District Court declared the case complex, and relaxed the trial deadlines. Sheperd replaced her counsel again in February 2018. “Attorney A”.

Two months after entering his appearance as Sheperd’s counsel, Attorney A met with FBI agents and a federal prosecutor to discuss Sheperd’s trial. But he was not there acting on Sheperd’s behalf. Attorney A was there to act on another client’s behalf—Okpara. Attorney A had helped Okpara secure a plea deal related to healthcare fraud in a different district court almost a year before. So why did Okpara need Attorney A at the meeting? Because Okpara was Sheperd’s friend and business associate—a relationship the Government wanted to exploit by calling Okpara as a witness against Sheperd.

If representing both Sheperd and Okpara at the same time sounds zany to you, then you wouldn’t be alone. The Government thought it sounded zany, too. In fact, it even pointed out to Attorney A that he had an obvious conflict. Attorney A replied that he hadn’t noticed. Even so, the meeting continued since, according to Attorney A, Sheperd wanted to plead guilty and would do so by the next month, in May. Turns out, the meeting was a success … for Okpara. Afterward, the Government amended Okpara’s plea deal to include 5K1 consideration. That is, in exchange for Okpara having provided the Government with substantial assistance in the investigation or prosecution of another person who has committed an offense, the Government agreed to permit Okpara to receive a reduced sentence.

Nobody told Sheperd about Attorney A’s conflict. Nobody told the district court about it either. But that time was not wasted. The Government used it to bring six more counts of healthcare fraud against Sheperd. The grand jury returned a superseding indictment that charged Sheperd with all six counts. Attorney A, for his part, continued to receive ‑provided discovery. Not until August—with trial looming—did Attorney A start trying to address his conflict. He approached a former state Assistant Attorney General with experience prosecuting healthcare‑fraud cases “Attorney B” about taking over Sheperd’s defense. Attorney A had worked with Attorney B before, and Attorney B got involved, in his words, on August 14.

On August 20, Attorney A finally revealed to Sheperd his conflict and proposed solution. But Sheperd felt burned. She initially refused to retain Attorney B. The district court set a status conference.

The status conference took place on August 27, 2018. Attorney B was not present. Attorney A assured the court that Attorney B would be prepared for trial two weeks later. That same day, Sheperd changed her mind, agreed to let Attorney B represent her, and Attorney A then withdrew as counsel.

A few days later Sheperd’s case was transferred to a new judge. The new judge held a pretrial conference on September 4. At the conference the judge asked the parties if they could move up the trial date. All parties represented they could not, pointing, in part, to Attorney B’s recent appearance in the case. Counsel for defendants also flagged the conflict issue for the district court, but suggested that the trial could proceed so long as Okpara did not testify. After discussing the conflict, the judge stated that he had never seen this type of conflict. The district court then, to the surprise of all parties, moved the trial date up by three days—to Friday, September 7.

On September 6, the day before trial, Attorney B moved for a continuance, arguing, the unfortunate circumstances surrounding the status of Ms. Sheperd’s counsel had denied him significant trial preparation time. He had only worked meaningfully on the case for approximately one week, despite having been involved for three. Alternatively, Attorney B moved to withdraw. That same day, all defendants also filed a joint motion, asking the court to, among other things, continue the trial and exclude Okpara’s testimony. The court did not rule on the motions. Trial began the next day.

A few days into trial, the Government called Okpara as a witness. That brought the conflict‑of‑ interest issue to a head. Defense counsel collectively reasserted the arguments from their pretrial motion: (1) Attorney A could have provided Okpara with privileged information to enhance Okpara’s efforts to cooperate with the Government; (2) if Okpara testified, he could share self‑incriminating statements that Sheperd shared with Attorney A; and (3) the current indictment could have been based on tainted information from Okpara. Attorney A and Sheperd took the stand. The district court ultimately excluded Okpara from testifying.

At the close of trial, Attorney B renewed the defendants’ joint pretrial motion, asking the district court to dismiss the case because of the conflict of interest. The district court denied the motion. The jury found Sheperd guilty on all counts. The district court sentenced her to 30 years in prison. Sheperd timely appealed, alleging several grounds to reverse based on Attorney A’s conflict of interest.

B. The Court Lays Down the Law SHEPERD I

The Court began with a statement of the law, mainly citing Strickland V. Washington, 466 U.S. 668 (1984), when it wrote:

The Constitution guarantees criminal defendants the right to a fair trial. The Supreme Court has explained that one of the safeguards making trials fair is the right to counsel. In all criminal prosecutions, the Sixth Amendment declares, the accused shall enjoy the right to have the Assistance of Counsel for his defense. That’s no hollow guarantee. The Counsel Clause does not countenance empty suits; it requires effective assistance of counsel—the presence of competence and absence of conflicts.

Unfortunately, not all representations meet that standard. At times a representation may suffer from actual ineffectiveness—one that so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result. That standard is satisfied when the defendant can show that (1) her counsel’s representation fell below an objective standard of reasonableness, and (2) that the deficiencies in counsel’s performance were prejudicial to the defense.

United States v. Sheperd, 27 F.4th 1075, 1082‑83 (5th Cir. 2022)(internal quotations omitted).

C. Conflicted Representation Reduces the Burden – SHEPERD I

The Court then continued the legal foundation of ineffective assistance of counsel as related to conflicts of interest.

That two-pronged test has a special application when a defendant argues that her representation was infected with a conflict of interest. One of the most indispensable duties that any counsel owes his client is the duty of loyalty. Counsel breaches that duty when he labors under an actual conflict of interest. And if he breaches the bedrock duty of loyalty, then his representation will fall below the objective standard of reasonableness that the Constitution requires.

But what about prejudice? Defendants still need to show it in actual-conflict cases. They show it, though, in a special way—by showing that their case is subject to a limited presumption of prejudice. Whether it applies boils down to one question: Did the conflict adversely affect counsel’s performance? That’s a lighter burden for defendants. Proving prejudice directly means showing a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. On the other hand, the limited presumption of prejudice in actual-conflict cases turns on whether the conflict adversely affected the representation. One way defendants can answer that mixed question of law and fact is through evidence that counsel’s judgment was actually fettered by concern over the effect of certain trial decisions on other clients.

The Government concedes there is no dispute that Attorney A operated under a conflict of interest. It also concedes that both it and Attorney A were obligated by rules of ethics to bring the conflict to the court’s attention. Still, that’s only the first half of Sheperd’s burden. She still must prove that Attorney A’s conflict adversely affected his performance. And on that question, on this record, we can’t tell one way or the other. (Sheperd I- 27 F.4th at 1083).

United States v. Sheperd, 27 F.4th at 83 (internal quotations omitted).


A. Results on Remand

On remand, the District Court was tasked with con‑ ducting “an evidentiary hearing on whether Shepherd’s lawyer’s conflict of interest—a conflict the Government knew about— adversely affected his representation.” United States v. Sheperd, 2022 U.S. App. LEXIS 22357,

*1, 44 F.4th 305 (5th Cir. 2022). The District Court answered the question in the affirmative, The Fifth Circuit sought out to determine the proper remedy in the case. Id. The Appellate Court did not appear to be impressed by the Government’s argument on the issue:

We ordered the parties to file letter briefs addressing any issues arising from the district court’s findings and conclusions, as well as the appropriate relief if any. The Government now concedes that Sheperd’s Sixth Amendment right to conflict-free counsel was violated and that some remedy may nonetheless be warranted. All that’s left, then, is to craft that remedy. The Government points us to the Supreme Court’s decision in Lafler v. Cooper, where the Court explained that Sixth Amendment remedies should be tailored to the injury suffered from the constitutional violation and should not unnecessarily infringe on competing interests. The remedy must neutralize the taint of the constitutional violation, but not grant a windfall to the defendant or needlessly squander the considerable resources the State properly invested in the criminal prosecution. For example, when a defendant declines a plea offer as a result of ineffective assistance of counsel and then receives a greater sentence as a result of trial, then the remedy could involve resentencing or ordering the Government to reoffer the plea deal, depending on the circumstances.

Applying those principles to this case, the appropriate remedy is to VACATE Sheperd’s convictions and REMAND for a new trial. That remedy may well cause the Government to spend considerable resources. But it is not needless. As the Government admits, it never offered Sheperd a plea deal while she was represented by her conflicted pretrial counsel, or after. So there is no rejected plea deal to measure Sheperd’s harm against. And while Sheperd did reject a plea deal well before her conflicted counsel began representing her, the Government argues we cannot order it to reoffer that bargain without violating separation-of-powers principles. So we’ll give the Government what it asks for, then, by avoiding any constitutional issue.

The constitutional right to counsel is perhaps the central feature of our adversarial system, as it helps make real the Constitution’s other criminal procedure promises. And it is not lost on us that the Government knew Sheperd’s pretrial counsel was conflicted yet delayed informing the district court about it—for months. The Government’s proposed remedy—to keep the convictions intact but remand for new plea negotiations—wouldn’t neutralize the taint of the constitutional violation. After all, what would encourage the Government to offer a reasonable plea when it could hold intact convictions over Sheperd’s head? Nothing.

We VACATE Sheperd’s convictions and REMAND for a new trial.

Id., pp. *1‑4.

Clearly, the Fifth Circuit could not stomach the unethical practice of a lawyer representing a defendant and a snitch in the same case. You might think that this case would have no applicability to you. If you have read this far, you are probably incapable of engaging in the kind of unethical activities demonstrated by Attorney A in Sheperd. However, it is possible for far more noble counsel to find themselves in a similar predicament.

B. Less Obvious But More Frightening Applications

Although the error made by Sheperd’s lawyer might have sounded ethical alarms with almost all lawyers, it is not hard to imagine how good lawyers could innocently straddle the same ethical line. The defendant arrested today may very well be a snitch, or even a witness, down the road. Seeing the name of a current or former client in a new client’s discovery is a common event. When that happens, all work on the case stops until the conflict question is resolved. As with most things legal, the starting point is the applicable rules.

C. What are the Rules?

  1. Federal courts defer to the State or District Bar Association in which the Court resides. (See for example: Local Criminal Rule 57.8(e) Northern District of Texas, which defers to the Texas Disciplinary Rules of Professional Conduct for the definition of unethical conduct).
  2. Texas Disciplinary Rules of Professional Conduct
    1. Rule 05. Confidentiality of Information
      1. Confidential information includes both privileged information and unprivileged client information. Privileged information refers to the information of a client protected by the lawyer‑client privilege of Rule 503 of the Texas Rules of Evidence or of Rule 503 of the Texas Rules of Criminal Evidence or by the principles of attorney‑client privilege governed by Rule 501 of the Federal Rules of Evidence for United States Courts and Magistrates. Unprivileged client information means all information relating to a client or furnished by the client, other than privileged information, acquired by the lawyer during the course of or by reason of the representation of the client.
      2. Except as permitted by paragraphs (c) and (d), or as required by paragraphs (e) and (f), a lawyer shall not knowingly:
        1. Reveal confidential information of a client or a former client to:
          1. a person that the client has instructed is not to receive the information; or
          2. anyone else, other than the client, the client’s representatives, or the members, associates, or employees of the lawyer’s law firm.
        2. Use confidential information of a client to the disadvantage of the client unless the client consents after consultation.
        3. Use confidential information of a former client to the disadvantage of the former client after the representation is concluded unless the former client consents after consultation or the confidential information has become generally known.
        4. Use privileged information of a client for the advantage of the lawyer or of a third person, unless the client consents after consultation.
    2. Rule 06. Conflict of Interest: General Rule
      1. A lawyer shall not represent opposing parties to the same litigation.
      2. In other situations and except to the extent permitted by paragraph (c), a lawyer shall not represent a person if the representation of that person:
        1. involves a substantially related matter in which that person’s interests are materially and directly adverse to the interests of another client of the lawyer or the lawyer’s firm; or
        2. reasonably appears to be or become adversely limited by the lawyer’s or law firm’s responsibilities to another client or to a third person or by the lawyer’s or law firm’s own interests.
  1. Texas Rules of Evidence
    1. Rule 03 Lawyer‑Client Privilege
      1.  A “client” is a person, public officer, or corporation, association, or other organization or entity ‑ whether public or private ‑ that:
        1. is rendered professional legal services by a lawyer; or
        2. consults a lawyer with a view to obtaining professional legal services from the lawyer.

The definition of “client” under Rule 503 of the Texas Rules of Evidence is broad enough to cover any potential client that discloses any fact relevant to the case. It is clear from these rules that a lawyer could easily get entangled by talking with two defendants regarding the same conspiracy. In that situation, a lawyer cannot use information disclosed by either client to the advantage or disadvantage of either client. The lawyer is unable to do his job.

The likelihood of such a conflict is greatly increased when all defendants are not arrested at the same time. A lawyer could talk to one client months before speaking with another client, with both clients having been charged in the same indictment. According to the Rule 503, it does not matter if the first potential client did not retain the lawyer. If the lawyer consulted with the person with a goal of getting hired, that person is a client.

A similar conflict could arise when a person arrested provides information that implicates others who are subsequently charged. This often occurs in traffic stop or controlled buy situations where the first person arrested is charged individually and then spills the beans on the rest of the gang. A lawyer often does not get the full details of the client’s discussions with the Government until he receives discovery. At that point, the lawyer should have had several conversations with the client where confidential information would have been shared. Under one of these scenarios, a good lawyer could find themselves potentially in the same position occupied by counsel in Sheperd. This is a situation often encountered in smaller jurisdictions, but possible anywhere. Such a conflict affects retained counsel and appointed counsel equally. An appropriate action, declining representation or filing a motion to withdraw, to such conflict might financially burden the ethical lawyer, but the unethical lawyer may suffer greater consequences.


  1. The reduced burden to demonstrate harm under the Strickland analysis is an important distinction for clients who have suffered conflicted representation. It is a “lighter lift” to demonstrate that the conflict affected the representation than it would be to demonstrate that a defendant would have been exonerated or that a lower sentence would have been pronounced, but for the error.
  2. The discussion of the remedy for a conflict of interest in Sheperd II is a wonderfully uncomplicated statement of the law. The remedy must neutralize the taint of the constitutional violation by going back to the point where the case was untainted.
  3. The Government’s lawyers took a scolding. All lawyers have a duty to report an ethical breach, and no party should benefit by abiding the infraction.
  4. If you see a name that rings a bell at any point in the case, find out why before you move forward.
  5. Screen potential clients for ability to pay prior to consultation. In a large conspiracy case, you only get one ticket. Don’t waste it on someone that needs appointed counsel.
  6. Having a client sign a waiver of conflicts is inadequate; it is a band aid on a severed artery.
  7. Decline or withdraw when you should. Being an ethical lawyer may hurt your bottom line, but your bar card is worth more than one case.

Federal Corner: Implications of Bruen for Criminal Defense Practitioners


If you think that constitutional defenses to firearm possession crimes are definitively settled against your client, think again. The Supreme Court changed the game with its recent opinion in New York State Rifle & Pistol Association, Inc. v. Bruen, 142 S. Ct. 2111 (2022), making Fifth Circuit precedent upholding the constitutionality of firearm regulations ripe for re‑examination. And yes, I’m even talking about 18 U.S.C. § 922.

In Bruen, the Court struck down New York’s licensing regime for the public carry of firearms, which required an applicant to demonstrate a “special need for self‑defense,” as violative of the Second Amendment. 142 Ct. at 2122. But the Court’s reasoning implicates far more than New York’s licensing scheme.

First and most importantly, the Court jettisoned “means‑end scrutiny in the Second Amendment context.” Id. at 2127. “[W]hen the Second Amendment’s plain text covers an individual’s conduct, the Constitution presumptively protects that conduct.” Id. at 2126. “To justify its regulation, the government may not simply posit that the regulation promotes an important interest.” Id. “Rather, the government must demonstrate that the regulation is consistent with this Nation’s historical tradition of firearm regulation.” Id. Otherwise, “the Second Amendment’s ‘unqualified command’” controls. Id. (quoting Konigsberg v. State Bar of California, 366 U.S. 36, 49 n.10 (1961)). In so holding, Bruen directly abrogated the Fifth Circuit’s analytical framework for Second Amendment challenges, which had endorsed this precise method of means‑end scrutiny. See Nat’l Rifle Ass’n of Am., Inc. v. Bureau of Alcohol, Tobacco, Firearms, & Explosives, 700 F.3d 185, 195–98 (5th Cir. 2012). Consequently, Bruen also calls into question numerous precedents and other opinions such as:

  • United States McGinnis, 956 F.3d 747, 756 (5th Cir. 2020) (“We need not and do not resolve” whether “the conduct burdened by § 922(g)(8) falls within the Second Amendment right” be‑ cause “§ 922(g)(8) is reasonably adapted” to the government’s interest in “reducing domestic gun abuse”); Mance v. Sessions, 896 F.3d 699, 701–04 (5th Cir. 2018) (even if 18 U.S.C. §§ 922(a)(3) & (b)(3), which prohibit the interstate transfer of handguns, “are not longstanding regulatory measures and are not presumptively lawful regulatory measures,” they are nonetheless constitutional because they “withstand strict scrutiny”);
  • Bezet United States, 714 F. App’x 336, 341 (5th Cir. 2017) (regardless of whether “a desire to obtain weapons that are part of the ordinary military equipment” “sweeps in firearms that are protected by the Second Amendment,” 18 U.S.C. § 922(l) & (r) are constitutional because “they do not substantially burden the core Second Amendment guarantee of acquiring firearms to protect one’s hearth and home,” “trigger only intermediate scrutiny,” and are a “reasonable fit” for achieving “important government objectives, such as cutting off weapons to criminals.”);
  • United States v. Massey, 849 F.3d 262, 265 (5th Cir. 2017) (“The Second Amendment frame‑ work adopted in [NRA] precludes this panel from reexamining . . . the constitutionality of Section 922(g).”); and
  • United States v. May, 538 F. App’x 465, 466 (5th Cir. 2013) (citing to NRA as precedent that precludes constitutional challenge to 18 S.C. 922(g)(3)).

Second, Bruen built on the watershed Second Amendment case that preceded it, District of Columbia v. Heller, 554 U.S. 570 (2008), by holding that “the right of an ordinary, law‑abiding citizen to possess a handgun in the home for self‑defense” extended to carrying “a handgun for self‑defense outside the home.Bruen, 142 S. Ct. at 2122 (emphasis added). Thus, in cases involving the public carry of a firearm, “the government must affirmatively prove that its firearms regulation is part of the historical tradition that delimits the outer bounds of the right to keep and bear arms.” Id. at 2127. “Courts are… entitled to decide a case based on the historical record compiled by the parties,” id. at 2130 n.6, and they are “not obliged to sift the historical materials for evidence to sustain” a statute, id. at 2150. Thus “the burden rests with the government”—not with the defense, and not with the court—“to establish the relevant tradition of regulation.” Id. at 2149 n.25.

Third, “to assess whether modern firearms regulations are consistent with the Second Amendment’s text and historical understanding,” Bruen distinguished between “general societal problem[s] that ha[ve] persisted since the 18th century,” id. at 21341, and “unprecedented societal concerns or dramatic technological changes” prompting “modern regulations that were unimaginable at the founding,” id. at 2132. Evidence that regulations addressing the former will not pass constitutional muster include:

  • A “lack of a distinctly similar historical regulation addressing that problem;”
  • Efforts to address “the societal problem” “through materially different means”; or
  • Attempts “to enact analogous regulations” that “were rejected on constitutional grounds[.]”

Id. at 2131. For the latter, Bruen adopted a “nuanced” test whereby a modern regulation must be “relevantly similar” to a historical analogue to conform with the Second Amendment. Id. at 2132. The majority declined to “provide an exhaustive survey of the features that render regulations relevantly similar,” id., but explicitly noted “how and why the regulations burden a law‑abiding citizen’s right to armed self‑defense” as “central” “metrics,” id. at 2133.

Fourth, establishing a relevant tradition of regulation via proper historical analogues may prove to be a taxing task. “On the one hand, courts should not uphold every modern law that remotely resembles a historical analogue, because doing so risk[s] endorsing outliers that our ancestors would never have accepted.” Id. “On the other hand, analogical reasoning requires only that the government identify a well‑established and representative historical analogue, not a historical twin” or a “dead ringer.” Id. This spectrum offers “little explanation of how stringently” to apply the historical analogue test, id. at 2179 (Breyer, J., dissenting), but Bruen’s analysis makes one thing clear: several pitfalls exist for the government here. “The Court’s historical analysis in this case is over 30 pages long and reviews numerous original sources from over 600 years of English and American history.” Id. at 2179 (Breyer, J., dissenting). Yet:

In each instance, the Court finds a reason to discount the historical evidence’s persuasive force. Some of the laws [the government] has identified are too old. But others are too recent. Still others did not last long enough. Some applied to too few people. Some were enacted for the wrong reasons. Some may have been based on a constitutional rationale that is now impossible to identify. Some arose in historically unique circumstances. And some are not sufficiently analogous to the licensing regime at issue here.

Id. at 2190. In the case of ambiguous historical sources, the majority even endorsed a default interpretation favoring possessors over the government. See id. at 2141 n.11 (“To the extent there are multiple plausible interpretations” of a historical source, “we will favor the one that is more consistent with the Second Amendment’s command.”).

Finally, a word about 18 U.S.C. § 922. According to Bruen, “the language of the Constitution cannot be interpreted safely except by reference to the common law and to British institutions as they were when the instrument was framed and adopted.” Id. at 2139. For that reason, Bruen anointed “history [b]etween the [Stuart] Restoration [in 1660] and the Glorious Revolution [in 1688]” as “particularly instructive.” Id. at 2140. Certainly “when it contradicts earlier evidence,” “late‑19th‑century evidence cannot provide much insight into the meaning of the Second Amendment.” Id. at 2154 (emphasis added); see also id. at 2154 n.28 (refusing to consider “20th‑century historical evidence” for the same reason). Given this backdrop, Bruen notably did not repeat Heller’s disclaimer that “nothing in our opinion should be taken to cast doubt on longstanding prohibitions on the possession of firearms by felons and the mentally ill, or laws forbidding the carrying of firearms in sensitive places such as schools and government buildings, or laws imposing conditions and qualifications on the commercial sale of arms.” Heller, 554 U.S. at 626–27. Bruen only “assume[d] it settled” that “legislative assemblies, polling places, and courthouses” were “‘sensitive places’ where arms carrying could be prohibited consistent with the Second Amendment,” remaining silent on Heller’s other presumptively lawful restrictions. Id. at 2133 (2022). See also id. at 2156 (“The Second Amendment guaranteed . . . the right to bear commonly used arms in public subject to certain reasonable, well‑defined restrictions” such as limiting “the intent for which one could carry arms, the manner by which one carried arms, or the exceptional circumstances under which one could not carry arms, such as before justices of the peace and other government officials.”). This omission apparently concerned Justice Alito and Justice Kavanaugh (joined by Chief Justice Roberts) sufficiently to prompt their concurrences, which specifically noted that Bruen did not invalidate 18 U.S.C. § 922 restrictions. See id. at 2157–58, 2162. But “prohibitions on the possession of firearms by felons and the mentally ill [and] laws imposing conditions and qualifications on the commercial sale of arms have their origins in the 20th century.” Id. at 2189 (Breyer, J., dissenting) (quoting C. Larson, Four Exceptions in Search of a Theory: District of Columbia v. Heller and Judicial Ipse Dixit, 60 Hastings L. J. 1371, 1374–1379 (2009)). And “[f]ounding‑era legislatures did not strip felons of the right to bear arms simply because of their status as felons.” Id. (quoting Kanter v. Barr, 919 F.3d 437, 451 (7th Cir. 2019) (Barrett, J., dissenting)). Thus, “the disconnect between Heller’s treatment of laws prohibiting, for example, firearms possession by felons or the mentally ill,” and Bruen’s “treatment of New York’s licensing regime” is “hard to square.” Id.

In sum, practitioners should be on high alert to Bruen’s application to a wide range of firearm‑related provisions, including 18 U.S.C. § 922. Already on August 30, 2022, in United States v. Rahimi, No. 21‑11001, the Fifth Circuit heard oral argument concerning Bruen and its impact on the constitutionality of 18 U.S.C. § 922(g) (8). Rahimi likely represents just the beginning of a slew of renewed and reinvigorated constitutional defenses to firearm possession crimes, which finally may prove successful post‑Bruen.

Federal Corner: July/August 2022


In United States v. Vargas, the Fifth Circuit addressed a tricky question—how should a recent Supreme Court decision limiting the deference owed to an agency’s interpretation of its own regulations affect a district court’s deference to the commentary found in the Guidelines Manual?1 After likening Guidelines commentary to “an agency’s interpretation of its own legislative rules,” the Supreme Court announced the following rule in Stinson v. United States: “[C]ommentary in the Guidelines Manual that interprets or explains a Guideline is authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that Guideline.”2 That level of deference may no longer be appropriate. In 2019, the Supreme Court clarified how and when a district court should defer to an agency’s interpretation of its own regulations.3 That opinion—Kisor v. Wilkie—addressed the level of deference described in Stinson and noted that it applies only if “the regulation [at issue] is genuinely ambiguous.”4 Whether Kisor overruled Stinson was an open question in the Fifth Circuit for almost three years, but on May 31, 2022, a three‑judge panel held that Stinson remained good law.5 This resulted in a short opinion authored by Judge Cory Wilson, which affirmed the appellant’s classification as a career offender.6 The panel nevertheless expressed sympathy with Mr. Vargas’s legal claim.7 It also noted a pending circuit split concerning Kisor’s effect on Stinson.8

On appeal, Mr. Vargas attacked the commentary to the Guidelines Manual’s career‑offender provision. He qualified as a career offender, the district court found, “because the instant offense, as well as . . . prior convictions for possession with intent to distribute amphetamine and conspiracy to possess with intent to manufacture and distribute methamphetamine, qualified as controlled substance offenses.”9 The Guidelines Manual, in the substantive text of Section 4B1.2, defines the term “controlled substances offense” to include “an[y] offense under federal or state law, punishable by imprisonment for a term exceeding one year, that prohibits the manufacture, import, export, distribution, or dispensing of a controlled substance . . or the possession of a controlled substance . . . with intent to manufacture, import, export, distribute, or dispense.”10 According to Section 4B1.2’s commentary, this definition “also ‘includes the offenses of aiding and abetting, conspiring, and attempting to commit such offenses.’”11 In light of Kisor, Mr. Vargas pitted the substantive text against the commentary. The “plain text” of Section 4B1.2 was unambiguous, he argued, and did not account for inchoate crimes.12 He then pointed to Kisor’s clarification of the standard announced in Stinson and asked the Fifth Circuit to declare Section 4B1.2’s commentary unworthy of deference.13

To prevail, Mr. Vargas needed to overcome both Stinson and the Fifth Circuit’s 1997 opinion in United States v. Lightbourn. The latter opinion addressed the Sentencing Commission’s statutory authority to de‑ fine drug conspiracies as career‑offender predicates.14 The Fifth Circuit determined that the Commission had appropriately “draw[n] its authority from the general guideline promulgation powers” granted to it by Congress.15 Lightbourn, Judge Wilson noted, “did not cite Stinson or otherwise expressly defer to the Guidelines commentary,” but since Section 4B1.2’s commentary provided “the sole source of authority for including in‑ choate offenses” within the relevant definition, “its hold‑ ing implied deference” to the commentary.16 A published opinion later adopted that implicit holding and thereby “rendered any implication of deference in Lighbourn explicit.”17 Lightbourn and its progeny thus foreclosed Mr. Vargas’s claim. To escape the effect of that authority, he needed to establish “an intervening change in law.”18

Kisor, Mr. Vargas argued, provided just that. There, the Supreme Court addressed the level of deference described in Stinson and explained that it “does not ap‑ ply ‘unless the regulation is genuinely ambiguous.’”19 “[B] efore concluding that a rule is genuinely ambiguous, a court must exhaust all the traditional tools of construction.”20 These include the “careful consideration of ‘the text, structure, history, and purpose of a regulation.’”21 Such analysis “will resolve many seeming ambiguities out of the box,” but if “genuine ambiguity remains,” a court may defer to an agency’s interpretation of its own regulations if that interpretation is “reasonable” in light of the regulation’s text.22

Stinson allowed for greater deference and without regard to ambiguity. There, the Supreme Court “considered whether and when the Sentencing Commission’s commentary to the Guidelines should be given binding interpretive effect” and treated Guidelines commentary as “akin to an agency’s interpretation of its own legislative rules.”23 At the time, that meant Guidelines commentary was “authoritative unless it violates the Constitution or a federal statute, or is inconsistent with, or a plainly erroneous reading of, that Guideline.”24 In Kisor, the Supreme Court rejected this approach as a “caricature” of the appropriate analysis.25

Did Kisor overrule Stinson? Mr. Vargas argued yes, and that development, he claimed, would render Lighbourn’s implicit deference to Section 4B1.2’s commentary improper in light of its unambiguous “plain text.”26 That text—the “controlled substances offense” definition found in Section 4B1.2(b)—did not account for inchoate crimes.27 Given its clarity, Mr. Vargas maintained, the commentary expanding the definition to include conspiracies was no longer “entitled to deference.”28 The panel expressed sympathy with Mr. Vargas’s argument but ultimately determined that Kisor was too vague to overrule Stinson. “If we were writing on a blank slate,” Judge Wilson noted, “we might well agree with Vargas,”29 but to escape Lightbourn, he needed to show that Kisor “unequivocally overrule[d]” the precedent blocking his claim on appeal.30 This, the panel concluded, he could not do, as “Kisor did not discuss the Sentencing Guidelines or Stinson.”31 Although Kisor provided a hint as to how the Supreme Court might rule in the future, “a mere ‘hint’ . . . is not enough to circumvent” the Fifth Circuit’s “rule of orderliness.”32

Other jurists disagree. Sitting en banc, the Third Circuit Court of Appeals considered the same argument advanced by Mr. Vargas and sided with the defendant.33 It found that Kisor had “cut back on what had been understood to be uncritical and broad deference to agency interpretations of regulations” and then applied Kisor’s analysis to Section 4B1.2’s text and commentary.34 The Third Circuit determined that Section 4B1.2(b)’s substantive text unambiguously excluded inchoate crimes and rejected the commentary as unworthy of deference.35 The defendant, who had been classified as a career of‑ fender based in part on a prior “conviction for an attempt to possess with intent to distribute cocaine” was therefore “entitled to be resentenced without being classified as a career offender.”36

The Fifth Circuit may follow the Third Circuit’s lead and rehear the case en banc. Judge Wilson wrote the opinion in Vargas, and Judges Jerry Smith and Gregg Costa signed on to his expression of sympathy concerning Mr. Vargas’s legal claim. A prior unpublished opinion drafted by Judge Edith Jones and joined by Judges Jennifer Walker Elrod and Smith expressed the same sentiment.37 On top of that, Judge Wilson used a footnote in Vargas to acknowledge the existing circuit split.38 Mr. Vargas has since filed a petition for en banc rehearing.39 In the petition, he asked the Fifth Circuit to reconsider the panel’s opinion, and for support, pointed to circuit splits concerning both Kisor’s general effect on Stinson and the specific deference owed to the career‑offender commentary.40 The pending petition may well provide Judge Wilson and his sympathetic colleagues the opportunity to reconsider Kisor’s effect while “writing on a blank slate.”41

Federal Corner: June 2022


While our lives were being upended by a pandemic, the Supreme Court and Congress have upended discrete aspects of federal sentencing that provide opportunity for forging new law that can help our clients. Other circuits have published decisions on three such issues: deference owed to Guidelines commentary, the criminal history bars in the new safety valve criteria, and whether prior marijuana convictions qualify as “controlled substance offenses.” The Fifth Circuit has not yet fully addressed these issues. Spot them, and potentially save your client significant time in prison.

  1. When the guideline commentary increases the advisory Guidelines range, rethink whether that commentary deserves deference.

The Third, Fourth, Sixth, and D.C. Circuits found that §4B1.2 controlled substance offenses do not include inchoate offenses. United States v. Nasir, 17 F.4th 459, 470 (3d Cir. 2021) (en banc); United States v. Campbell, 22 F.4th 438, 440 (4th Cir. 2022); United States v. Havis, 927 F.3d 382, 386 (6th Cir. 2019); United States v. Winstead, 890 F.3d 1082, 1091 (D.C. Cir. 2018).1 They did so based on a recent Supreme Court decision regarding when to give deference to agency decisions: Kisor v. Wilkie, 139 S. Ct. 2400 (2019). This Kisor‑deference argument has the potential to save our clients from higher sentences that result from guideline commentary that is harsher than the guideline text.

What does an agency deference decision have to do with the Guidelines?

The text of the Guidelines comes into effect only after being submitted to Congress for disapproval or modification. See 28 U.S.C. § 994(p). In contrast, the Guidelines commentary is not subject to congressional review.

In 1993, the Supreme Court compared the commentary to “an agency’s interpretation of its own legislative rules.” Stinson v. United States, 508 U.S. 36, 45 (1993).

Unlike Guidelines text, which must be followed, the Supreme Court applied the then‑prevailing agency deference framework of Seminole Rock, also called Auer deference,2 and said that the commentary should be followed unless it is inconsistent with the Guidelines text. Id. at 43. Practically speaking, this meant the Guidelines commentary was almost always applied. So, we defense attorneys gradually were lulled into a routine of applying the commentary without question.

Decades later, in 2019, Kisor “awoke us from our slumber of reflexive deference” to the agency interpretation of regulations. Nasir, 17 F.4th at 472 (Bibas, J., concurring). Kisor “cut[s] back on what had been understood to be uncritical and broad deference to agency interpretations of regulations and explain[s] that Auer, or Seminole Rock, deference should only be applied when a regulation is genuinely ambiguous.” Id. at 470. Rather than reflexive deference, Kisor says “a court must exhaust all the traditional tools of construction” before deciding that a regulation is “genuinely ambiguous[.]” 139 S. Ct. at 2415 (cleaned up). “Doing so will resolve many seeming ambiguities out of the box, without resort to Auer deference.” Id. And even when the regulation is ambiguous, the agency interpretation only warrants deference when it is “reasonable,” “authoritative,” a product of its “substantive expertise,” and a “fair and considered judgment.” Id. at 2416‑18 (cleaned up).

Because Guidelines commentary is treated as an agency’s interpretation of its own legislative rules, several circuits have applied Kisor to the commentary. See Nasir, 17 F.4th at 471‑72. Doing so, they have found, for instance, the §4B1.2(b) text unambiguous, meaning there was no reason to resort to the commentary. Id.

The Fifth Circuit has not squarely addressed how Kisor affects deference to Guidelines commentary. One panel found that it was bound by precedent, United States v. Lightbourn, 115 F.3d 291, 293 (5th Cir. 1997), to follow the §4B1.2 commentary adding conspiracy to the “controlled substance offense” definition. United States v. Goodin, 835 F. App’x 771, 782 (5th Cir. 2021). But the panel indicated that, if given a clean slate, it was inclined to agree with the Third Circuit. Id. at 782 n.1. That amounts to an en banc invitation in the right case.

United States v. Vargas, No. 21‑20140, may be such a case.3 A Fifth Circuit panel heard argument in Vargas in March 2022. Mr. Vargas argued Kisor had abrogated Stinson and that the panel was not bound by Lightbourn. But even if the panel does not agree, the en banc court or the Supreme Court may one day take the issue up and rule that deference is only due to the Guidelines commentary when the Guidelines text is “genuinely ambiguous.” Nasir, 17 F.4th at 471.

A shift to the limited Kisor deference for Guidelines commentary would mean courts would have to work harder to decide whether any given commentary should apply. See, e.g., United States v. Riccardi, 989 F.3d 476 (6th Cir. 2021) (refusing to defer to the §2B1.1 commentary’s definition that “loss” for access device cards means $500 per card). And defense attorneys excel at working hard and using the complexity of issues to our clients’ advantage.

To recap:

  • The old: defer to Guidelines commentary unless inconsistent (Stinson).
  • The break: defer to agency interpretations only if regulation is genuinely ambiguous and interpretation warrants deference (Kisor v. Wilkie).
  • The possible  new: defer to commentary only after exhausting statutory construction tools to determine whether the Guidelines text is ambiguous, and then defer to commentary only if When your client appears to not qualify for safety valve due to too many criminal history points, think again.
  1. We know that having too many criminal history points can disqualify a client from safety valve, and safety valve can be important because it allows a district court to sentence below the mandatory minimum in certain drug cases.4 18 U.S.C. § 3553(f). The question is how many criminal history points is too many.

It used to be more than one criminal history point disqualified a client. 18 U.S.C. § 3553(f)(1) (2018).

Then came the First Step Act of 2018. It changed the criminal history requirements so that a defendant remains eligible if:

the defendant does not have‑‑

  1. more than 4 criminal history points, excluding … 1‑point offense[s]…;
  2. a prior 3‑point offense…;
  3. a prior 2‑point violent offense[.]
    § 3553(f)(1).

In United States v. Lopez, 998 F.3d 431 (9th Cir. 2021), the Ninth Circuit held that the “and” in the new safety‑valve provision means “and.” In other words, a defendant’s criminal history points only make him ineligible for safety valve if he has more than 4 criminal history points excluding 1‑point offenses, a 3‑point offense, and a 2‑point violent offense. If a defendant only has a 3‑point offense and a 2‑point nonviolent offense, he would still be eligible. The Government filed a petition for rehearing en banc in Lopez in August 2021 that is pending.

A panel of the Eleventh Circuit initially ruled the opposite way (finding the “and” is disjunctive) but then vacated that opinion when it granted rehearing en banc. United States v. Garcon, 997 F.3d 1301 (11th Cir. 2021), reh’g en banc granted, opinion vacated, 23 F.4th 1334 (11th Cir. 2022). The Government filed its en banc reply brief in Garcon in May 2022.

The Fifth Circuit heard oral argument on this issue in February 2022 but has not yet issued a decision. United States v. Palomares, No. 21‑40247 (5th Cir.).5 Some district courts in the Western District of Texas have sided with Lopez and imposed sentences below the mandatory minimum sentence. Maybe you can also convince a court that “and” means “and.”

To recap:

  • The old: no safety valve if more than 1 criminal history point.
  • The break: the First Step Act of 2018 amended the safety valve
  • The possible new: only defendants who have more than 4 criminal history points excluding 1‑point offenses, a 3‑point offense, and a 2‑point violent offense are ineligible (Lopez).
  • Or at least: defendants who have more than 4 criminal history points excluding 1‑point offenses, a 3‑point offense, or a 2‑point violent offense are ineligible.
  1. If your client’s pre-2018 marijuana convictions are enhancing the sentence, consider whether those priors included hemp.

In United States v. Bautista, the Ninth Circuit held that a pre‑2018 Arizona marijuana conviction was not a §4B1.2 “controlled substance offense” (CSO). 989 F.3d 698, 703‑ 04 (9th Cir. 2021). As a result, Mr. Bautista’s §2K2.1 sentence of 30 months’ imprisonment was vacated for resentencing under the correct range of 15 to 21 months. Id. at 705. The Ninth Circuit followed these steps to reach the favorable result:

  1. Used substances listed in the federal Controlled Substances Act (CSA) to define “controlled substance” in guideline 4B1.2. United States v. Leal-Vega, 680 F.3d 1160, 1167 (9th Cir. 2012).
  2. Compared the controlled substance definition for the prior conviction to the §4B1.2 controlled substance definition that exists now at sentencing. See 18 U.S.C. § 3553(a)(4)(A) (ii); U.S.S.G. § 1B1.11.

The definition of “marijuana” at the time of the prior Arizona offense included hemp. At the time of sentencing for the new federal offense, the Agricultural Improvement Act of 20186 had removed hemp (cannabis with THC of 0.3% or less) from the “marijuana” definition. Thus, the prior Arizona conviction was broader than the §4B1.2 CSO definition because the former included hemp, and the latter did not.

In a recent unpublished Fifth Circuit decision, Judge Higginson indicated he thought the Bautista analysis was correct and should apply in the Fifth Circuit. United States v. Belducea-Mancinas, No. 20‑50929, 2022 WL 1223800, at *2 (5th Cir. Apr. 26, 2022) (Higginson, J., concurring).7 He thought the district court erred, albeit not plainly, by sentencing Mr. Belducea as a career offender based on pre‑ 2018 marijuana convictions. Id.; see also United States v. Abdulaziz, 998 F.3d 519, 524‑31 (1st Cir. 2021); United States v. Crocco, 15 F.4th 20, 23 n.3 (1st Cir. 2021); United States v. Williams, 850 Fed. App’x 393, 398 (6th Cir. 2021) (unpublished).

Some circuits do not define “controlled substance” in §4B1.2 as a substance in the federal CSA. See Crocco, 15 F.4th at 23 (describing circuit split). But the Fifth Circuit already incorporated the federal CSA in the materially similar drug trafficking offense definition of guideline §2L1.2. United  States v. Gomez-Alvarez, 781 F.3d 787 (5th Cir. 2015) (adopting Leal- Vega). “Because the qualifying prior convictions in § 2L1.2 and § 4B1.2(b) are defined in substantially the same way, cases discussing these definitions are cited interchangeably[.]” United States v. Arayatanon, 980 F.3d 444, 453 n.8 (5th Cir. 2020) (cleaned up). Under the holding and reasoning of Gomez-Alvarez, Judge Higginson finds that the Fifth Circuit defines “controlled substance” in §4B1.2(b) with reference to the CSA. Belducea- Mancinas, 2022 WL 1223800, at *2 (Higginson, J., concurring).

Note: While Bautista and Belducea-Mancinas were in the context of guideline §4B1.2, the argument could apply to other recidivist enhancements or categorizations, such as Armed Career Criminal Act serious drug offenses, and the serious drug felonies that can enhance drug sentences. See, e.g., United States v. Hope, 28 F.4th 487, 504‑05 (4th Cir. 2022) (in the ACCA context).

To recap:

The old: the Controlled Substances Act defined “marijuana” as any part of the cannabis sativa L. plant, regardless of the amount of THC.

The break: the Agriculture Improvement Act of 2018 amended the “marijuana definition” to exclude hemp, which is any part of the cannabis sativa plant containing TCH of 0.3% or less.

The possible new: prior drug convictions for which the least culpable act involved hemp may not be §4B1.2 CSOs or other drug recidivist enhancements.

Federal Corner: The State of Borden in the Fifth Circuit


The state of common Texas offenses almost a year after Borden

Last year, federal criminal defendants won a big victory when the Supreme Court held in Borden v. United States, 141 S.Ct. 1817 (2021), that offenses bearing a mens rea of recklessness lack “the use, attempted use, or threatened use of physical force against the person of another” as an element. The controlling opinion in Borden focuses on the restrictive phrase “against the person of another,” which it holds to imply an “intentional targeting” of the victim, and to exclude accidents. Borden arose in the context  of the Armed Career Criminal Act’s (“ACCA”) definition of “violent felony,” but the language it construed (or very similar language) appears (and is cross‑ referenced) throughout the U.S. Code and Sentencing Guidelines. See 18 U.S.C. 16(a), 18 U.S.C. 924(c) (3)(A), 18 U.S.C. 924(e)(2)(B)(i), 18  U.S.C.  3156(a)(4),  18  U.S.C. 3559(c)(2)(F)(i); USSG 2L1.2, comment. (n. (1)(B))(2015), USSG 4B1.2(a)(1). Most importantly, the language appears in ACCA (which elevates the penalty for firearm possession from a 10‑year maximum to a 15‑year minimum), and in Guideline 4B1.2, which can render the defendant eligible for elevated penalties under the firearm Guideline (USSG 2K2.1) or for the cataclysmic penalties of the career offender Guideline (USSG 4B1.1).

Of course, Borden left circuit courts the task of applying its holding to offenses it did not address. Results in the Fifth Circuit have been mixed, and not everything has been settled. Nonetheless, we do know how some of the most common Texas offenses have fared after Borden. It’s worth remembering that while everything below describes current law, this area of law is notoriously prone to change. Zealous practitioners will no doubt find arguments worth preserving, or that have not yet been addressed, even when the law appears to qualify their clients for an enhanced sentence.

Robbery and Aggravated Robbery

Texas robbery may be committed in either of two basic ways: by causing injury during the course of a theft, or by threatening or placing another in fear during the course of a theft. See Texas Penal Code 29.02(a). Because the Texas offense of robbery‑by‑injury may be committed by the reckless infliction of injury, the Fifth Circuit held in United States v. Ybarra, 2021 WL 3276471 (5th Cir. 2021)(unpublished), that it no longer qualifies as an ACCA “violent felony” under Borden. This is so even if the record establishes that the defendant pleaded guilty to causing such injury intentionally or knowingly. See Ybarra, 2021 WL 3276471, n.1. The three different mental states with which a defendant may commit robbery‑ by‑injury (intent, knowledge and recklessness) in Texas do not define separate offenses, so the crime is not “divisible” for the purposes of criminal history enhancements. See id. (For more on the “divisibility” of prior statutes of conviction for the purpose of criminal history enhancements, see the remarkable case of Mathis v. United States, 136 S.Ct. 2243 (2016), an unsurpassed aid to federal criminal defendants facing recidivism enhancements). The Fifth Circuit has not yet decided whether all forms of aggravated robbery‑by‑injury (e.g. infliction of injury during theft with a deadly weapon, infliction of serious bodily injury during theft, or infliction of injury against a disabled or senior victim during theft) constitute “violent felonies” after Borden. But decisions from the aggravated assault context, see below, strongly suggest that the addition of these aggravating factors will not change the basic analysis.

No such luck for those convicted of robbery‑by‑threat. The Fifth Circuit held in United States v. Garrett, 24 F.4th 485 (5th Cir. 2022), that Texas simple robbery‑by‑ threat qualifies as a “violent felony” under ACCA. This is because Texas robbery‑by‑threat, unlike robbery‑ by‑injury, can only be committed intentionally  or knowingly,  not wrecklessly. And unsurprisingly, the Fifth Circuit held in United States v. Jackson,  F.4th , 2022 WL 951232 (5th Cir. 2022), that aggravated robbery‑by‑threat is likewise a “violent felony” under ACCA. Note, however, that if the government does not produce a record of the defendant’s robbery conviction showing that it stemmed from the statute’s threat prong, ACCA will not apply. See United States v. Balderas, 2022 WL 851768 (5th Cir. 2022)(unpublished).

Borden does not appear to help those subjected to USSG 4B1.2 on the basis of Texas robbery convictions. In contrast to ACCA’s definition of “violent felony,” Guideline 4B1.2’s definition of “crime of violence” specifically enumerates “robbery.” That is, a prior conviction can be a “crime of violence” under USSG 4B1.2 if it either has force as an element, or “is… robbery.”1 And the Fifth Circuit has held that Texas robbery, see United States v. Williams, 2022 WL 71826 (5th Cir. 2022)(unpublished), and aggravated robbery, see United States v. Nava, 2021 WL 5095976 (5th Cir. 2021)(unpublished), are both equivalent to “generic robbery” as the Guideline uses the term. There are good reasons to wonder about this conclusion – Texas, in contrast to most states, does not require that the defendant acquire property as a consequence of his or her act of  violence – but the Fifth Circuit isn’t biting on that distinction yet.

Aggravated Assault

As with the Texas robbery statute, defendants can violate the Texas aggravated assault statute by either inflicting injury or threatening it. See Tex. Penal Code 22.01(a), 22.02(a). The Fifth Circuit found in United States v. Combs, 2022 WL 287556 (5th  Cir. 2021)(unpublished), that aggravated‑ assault‑by‑injury lacks the “use of force against the person of another” as an element and accordingly does not constitute a “violent felony” under ACCA. See also United States v. Gomez-Gomez, 23 F.4th 575 (5th Cir. 2022)(same result under 18 U.S.C. §16(a)). Because the three mental states associated with aggravated‑assault‑by‑injury do not represent distinct offenses, it will not matter if the records of conviction show a conviction for intentional or knowing infliction of injury. See United States v. Lara-Garcia, 2021 WL 5272211 (5th Cir. 2021) (unpublished). The Fifth Circuit has held that aggravated‑assault‑ by‑threat represents a violent felony after Borden, see United States v. Lopez, 2022 WL 576407 (5th Cir. 2022)(unpublished), because it requires intentional or knowing conduct. The Fifth Circuit’s post‑ Borden aggravated assault rulings under ACCA, in other words, track its rulings regarding robbery.

Here too, Borden doesn’t seem to help those subjected to an enhanced Guideline range. Guideline 4B1.2 names “aggravated assault” as a “crime of violence.” USSG 4B1.2(a) (2). And the Fifth Circuit has held that the Texas offense falls within the “generic” definition of “aggravated assault.” See United States v. Guillen-Alvarez, 489 F.3d 197, 200‑201 (5th Cir. 2007). It reaffirmed that conclusion after Borden. See United States v. Blackmon, 844 Fed. Appx. 729 (5th Cir. 2021) (unpublished). But note that there is a long‑standing circuit split as to whether Texas aggravated‑assault‑by‑injury qualifies as “generic aggravated assault,” centering on the significance of its reckless mens rea. See United States v. Barcenas-Yanez, 826 F.3d 752 (4th Cir. 2016). Because the Supreme Court has said that it will usually not take up circuit splits on Guideline questions, see Buford v. United States, 532 U.S. 59 (2001), the issue must likely be resolved by the Sentencing Commission, should it ever again enjoy a quorum.

Family Violence

Texas has two felony assault offenses that require a family or dating relationship between the defendant and the victim. One of these requires a prior conviction for family violence, see Tex. Penal Code 22.01(b)(2)(A); the other requires the defendant to impede the victim’s breath or circulation, see Tex. Penal Code 22.01(b)(2)(A). Both offenses may be committed recklessly, so the Fifth Circuit held in United States v. Greer, 20 F.3d 1071 (5th Cir. 2021), that neither is a “crime of violence” under USSG 4B1.2. This holding should exclude both offenses as “violent felonies” under ACCA as well. Setting aside the timing requirements under USSG 4B1.2, all offenses that meet the definition of “violent felonies” ACCA qualify as “crimes of violence” under USSG 4B1.2.

Assault on a Public Servant

Because the Texas offense of assault on a public servant may be committed recklessly, the Fifth Circuit held that it does not satisfy the definition of a “crime of violence” under USSG 4B1.2. See United States v. Bates, 24 F.4th 1017 (5th Cir. 2022). (Older Fifth Circuit authority found that it is not equivalent to the enumerated offense of “aggravated assault.” See United States v. Fierro-Reyna, 466 F.3d 324 (5th Cir. 2006)). As with family violence, this holding would also appear to preclude its use as a “violent felony” under ACCA.


The Fifth Circuit ducked a chance to decide whether murder is a “violent felony” after Borden in United States v. Vickers, 2022 WL 780421 (5th Cir. 2022)(unpublished). There is good reason to think it might not be. Texas murder may be committed by causing death through “an act clearly dangerous to human life” in the course of another felony, a standard that sounds a lot like recklessness. See Texas Penal Code 19.02(b)(3). “Murder” is an enumerated offense under USSG 4B1.2, so disqualifying it as a “crime of violence” under that Guideline may pose more of an uphill battle.


Borden does not appear to affect Texas burglary offenses under either ACCA or USSG 4B1.2. Under pre‑ Borden precedent, Texas burglary constitutes a “violent felony” under the ACCA, see United States v. Herrold, 941 F.3d 173 (5th Cir. 2019) (en banc), but does not represent a “crime of violence” under USSG 4B1.2, see United States v. Reado, 776 F. App’x 261, 262 (5th Cir. 2019)(unpublished).

Drug Trafficking

Likewise, Borden does not seem to affect Texas drug trafficking offenses. Both delivery and possession with intent to deliver qualify as “serious drug offenses” under ACCA, see United States v. Prentice, 956 F.3d 295 (5th Cir. 2020), but do not represent “controlled substance offenses” under USSG 4B1.2, see United States v. Tanksley, 848 F.3d 347 (5th Cir. 2017). There may be some hope for defendants facing ACCA sentences on the basis of drug conviction. Some other circuits have found that state drug laws criminalize a greater range of substances than federal law, rendering some drug convictions from those states overbroad for the purposes of ACCA. See United States v. Ruth, 966 F.3d 642 (7th Cir. 2020); United States v. Hope, 28 F.4th 487 (4th Cir. 2022).

Shut up and give us the chart, Joel:


USSG 4B1.2

Texas Robbery by Injury

No longer qualifies. United States v. Ybarra, 2021 WL 3276471 (5th Cir. 2021)(unpublished)

Qualifies. United States v. Williams, 2022 WL 71826 (5th Cir. 2022)(unpublished).

Texas Aggravated Robbery by Injury

Undecided, but likely does not qualify. See United States v. Ybarra, 2021 WL 3276471 (5th Cir. 2021)(unpublished) and United States v. Gomez-Gomez, 23 F.4th 575 (5th Cir. 2022).

United States v. Nava, 2021 WL 5095976 (5th Cir. 2021)(unpublished).

Texas Robbery and Aggravated Robbery by Threat

Qualifies. United States v. Garrett, 24 F.4th 485 (5th Cir. 2022); United States v. Jackson,  F.4th , 2022 WL 951232 (5th Cir. 2022).

Qualifies. United States v. Williams, 2022 WL 71826 (5th Cir. 2022)(unpublished); United States v. Nava, 2021 WL 5095976 (5th Cir. 2021)(unpublished).

Texas Aggravated Assault by Injury

No longer qualifies. United States v. Combs, 2022 WL 287556 (5th Cir. 2021)(unpublished); United States v. Gomez-Gomez, 23 F.4th 575 (5th Cir. 2022).

Qualifies. United States v. Blackmon, 844 Fed. Appx. 729 (5th Cir. 2021)(unpublished).

Texas Aggravated Assault by Threat

Qualifies. United States v. Lopez, 2022 WL 576407 (5th Cir. 2022)(unpublished).

Qualifies. United States v. Blackmon, 844 Fed. Appx. 729 (5th Cir. 2021)(unpublished).

Texas Felony Family Violence by Recidivism or Impeding Breath or Circulation

Undecided, but implicitly disqualified by United States v. Greer, 20 F.3d 1071 (5th Cir. 2021).

No longer qualifies. United States v. Greer, 20 F.3d 1071 (5th Cir. 2021).

Texas Murder



Texas Assault on a Public Servant

Undecided, but implicitly disqualified by United States v. Bates, 24 F.4th 1017 (5th Cir. 2022).

No longer qualifies. United States v. Bates, 24 F.4th 1017 (5th Cir. 2022).

Texas Burglary

Qualifies. United States v. Herrold, 941 F.3d 173 (5th Cir. 2019)(en banc).

Still doesn’t qualify. United States v. Reado, 776 F. App’x 261, 262 (5th Cir. 2019)(unpublished).

Texas Delivery of a Controlled Substance and Possession with Intent to Deliver

Qualifies. United States v. Prentice, 956 F.3d 295 (5th Cir. 2020).

Still doesn’t qualify. United States v. Tanksley, 848 F.3d 347 (5th Cir. 2017); United States v. Hinkle, 832 F.3d 569 (5th Cir. 2016).

Federal Corner: No Federal Habeas Relief Despite Sleeping Lawyer


The Sixth Amendment right to counsel

“In all criminal prosecutions, the accused shall enjoy the right to . . . have the Assistance of Counsel for his defense.” U.S. Const. amend. VI.

“An accused’s right to be represented by counsel is a fundamental component of our criminal justice system.” United States v. Cronic, 466 U.S. 648, 653 (1984). “[T]he right to the effective assistance of counsel is recognized not for its own sake, but because of the effect it has on the ability of the accused to receive a fair trial.” Id. at 658. The Sixth Amendment requires that the accused have “counsel acting in the role of an advocate.” Id. at 656 (internal quotation omitted).

“The right to counsel exists, and is needed, in order to protect the fundamental right to a fair trial . . . That a person who happens to be a lawyer is present at trial alongside the accused, however, is not enough to satisfy the constitutional command. The Sixth Amendment recognizes the right to the assistance of counsel because it envisions counsel’s playing a role that is critical to the ability of the adversarial system to produce just results.” Strickland v. Washington, 466 U.S. 668, 685 (1985).

In McFarland v. Lumpkin, 26 F.4th 314 (5th Cir. 2022), the Court of Appeals for the Fifth Circuit held that the district court did not err when it found that 28 U.S.C. § 2254(d) precluded federal habeas merits review of petitioner’s claims that his Sixth Amendment right to counsel was violated when trial counsel was asleep and unprepared during his capital murder trial.

Facts underpinning Cronic and Strickland claims of violation of Sixth Amendment right to counsel

Mr. McFarland was convicted as an accomplice to capital murder and sentenced to death by a Texas trial court. According to the Fifth Circuit’s recitation of the facts, “[d]uring trial, the judge confronted a problem. Mr. McFarland’s retained counsel [] was sleeping through significant portions of the trial and otherwise presented as unprepared.”1 Indeed, in his briefing to the Fifth Circuit, Mr. McFarland explained that retained counsel was so obviously unprepared to try a capital case that the trial court appointed second chair counsel (without Mr. McFarland’s consent).2

The trial court, however,  instructed second chair counsel not to make any decisions in the case without the approval of first chair counsel and applicant, and that [first chair counsel] was to remain the “lead lawyer” and  “in charge” (Fifth Circuit opinion). In addition, the McFarland opinion noted it was unknown if the Trial Court actually informed Mr. McFarland expressly about its concern over his lawyer’s preparation and competence. In his briefing, Mr. McFarland further describes how, once trial began, first chair counsel then slept through significant portions of the trial.

Even after second chair counsel was appointed, neither attorney “ever interviewed the State’s key witnesses, the other alleged accomplice [] or [the State’s other key witness].” Defendant’s mitigation presentation in the punishment of the death penalty case totaled 15 minutes.

State habeas proceedings

In state post-conviction proceedings, Mr. McFarland alleged, amongst other issues, he was actually or constructively denied counsel and deprived of the effective assistance of counsel at trial  in violation of the Sixth Amendment under Strickland and Cronic. The Texas Court of Criminal Appeals denied relief as to all claims, finding that the presence of second chair counsel was sufficient to cure any denial of the right to counsel under Cronic and that trial counsel did develop a strategy sufficient to constitute effective representation under Strickland.3

Federal habeas proceedings in the district court

In federal habeas proceedings, Mr. McFarland re-urged his claims that he was constructively denied counsel and was denied effective assistance of counsel in violation of the Sixth Amendment, in reliance on Cronic and Strickland. Because both claims were adjudicated on the merits by the CCA, merits review by the federal court was prohibited pursuant to 28 U.S.C. § 2254(d), unless the State court’s adjudication “resulted in a decision contrary to, or involved an unreasonable application” of Supreme Court case law or “resulted in a decision that was based on an unreasonable determination of the facts” in light of the state court record.

The district court found that, whether as to Mr. McFarland’s Cronic or Strickland claims, the CCA’s adjudication on the merits neither amounted to an unreasonable application of Cronic and Strickland nor an unreasonable determination of the facts.4 The Federal District Court also found the CCA’s determination that the presence of second chair counsel was sufficient to cure any denial of the right to counsel did not constitute an unreasonable application of Chronic.  The Federal District Court also held the CCA’s finding that trial counsel had  a “trial strategy” did not constitute an unreasonable application of the Strickland test.5

Arguments on appeal to the Fifth Circuit Court of Appeals

Mr. McFarland applied for and was granted a Certificate of Appealability on, amongst other issues, whether the Federal District Court legally erred in determining that the 28 U.S.C. § 2254(d) re-litigation bar precluded merits review of his Cronic and Strickland claims6. As to his Cronic claim, Mr. McFarland argued the Court had failed to engage with his argument that prejudice should not have been presumed based on the circumstances requiring appointment of a second chair alone.7

In support of his circumstances argument, Mr. McFarland pointed to the facts that first chair counsel, as described by the Fifth Circuit, was unprepared and sleeping through significant portions of the trial. The second chair counsel was prohibited by the trial court from taking any actions not approved by first chair counsel and Mr. McFarland who was likely unaware of his own lawyer’s incompetence.8 Mr. McFarland argued the totality of these circumstances rendered any effective representation by second chair counsel impossible, and that he was thus constructively denied counsel under Cronic.9 Indeed, two State court judges had dissented from the majority and would have granted Cronic relief.10

Mr. McFarland also argued that the district court legally erred when it determined that the State court’s adjudication of his Strickland claim did not constitute an unreasonable application of the Supreme Court’s case law. Mr. McFarland urged that the CCA’s finding trial counsel had a strategy was unsupported by the record. Amongst other facts, Mr. McFarland pointed to the fact the record established that first and second chair counsel had very little contact prior to trial and could therefore never have discussed any strategy. Mr. McFarland further argued that the district court should have focused its analysis on whether, applying Strickland, it was reasonable to not investigate the State’s case on guilt as opposed to focusing its inquiry on what other actions trial counsel did undertake.

Mr. McFarland argued that the Fifth Circuit should instead find that the State court’s adjudication constituted an unreasonable application of Cronic and Strickland such that the 28 U.S.C. § 2254(d) relitigation bar was met, thus permitting federal habeas merits review and relief.

No error in the Fifth Circuit

The Fifth Circuit summarily rejected Mr. McFarland’s claim under Cronic: “We are aware of no case where a sleeping co-counsel alone triggers Cronic’s presumption of prejudice.”11 The Fifth Circuit concluded that, based on the absence of such a Supreme Court case, the State’s court’s adjudication that the presence of second chair counsel was sufficient to cure any Sixth Amendment violation did not rise to an unreasonable application of Cronic.12 The Fifth Circuit, however, failed to engage with Mr. McFarland’s circumstances argument, despite clear language from the Supreme Court in Cronic itself in support of such an argument: “Circumstances . . . may be present on some occasions when although counsel is available to assist the accused during trial, the likelihood that any lawyer, even a fully competent one, could provide effective assistance of counsel is so small that a presumption of prejudice is appropriate without inquiry unto the actual conduct of the trial.”13

The Fifth Circuit found Mr. McFarland’s arguments under Strickland failed to survive the court’s “doubly deferential” standard under 28 U.S.C. § 2254(d) and Strickland. As to each of Mr. McFarland’s allegations of deficient performance, the Fifth Circuit found that he could not show that the State court’s determination that the Strickland test was not met was unreasonable.14 For example, the Fifth Circuit found that it was reasonable for the State courts to find that trial counsel’s failure to thoroughly cross-examine the State’s key witnesses (including the State’s sole eyewitness whose description of the perpetrator shifted over time) did not fall below prevailing professional norms because cross-examination is inherently risky and trial counsel did cross those witnesses but on different issues.15


This case is significant because it illustrates how, even in circumstances as egregious as a sleeping lawyer in a capital murder trial, 28 U.S.C. § 2254(d) precludes merits review (let alone relief) in federal habeas. Mr. McFarland’s argument that the circumstances at trial amounted to a constructive denial of the right to counsel finds clear support in the Supreme Court’s language in Cronic. Likewise, the incomplete cross-examination of the State’s key witnesses and truncated mitigation presentation support a Strickland claim. Even where the Sixth Amendment right to counsel is at stake, however, the Fifth Circuit will not disturb a state court’s adjudication of a state turned federal habeas petitioner’s claims.

Federal Corner: Plea agreement or not – You can’t punish me for liking the Yankees and not Red Sox!


United States v. Leroya Cozad, 2022 WL 16492 (10th Cir. Filed January 3, 2022)

The Court in this opinion summarizes a concern defense attorneys have- use a plea agreement or plea to the indictment? This is a very different calculus in Federal Court where the judge does most of the “sentencing” as compared to state court where the prosecutors provide the agreed sentence.

In Cozad, the Court looked at “whether under 18 U.S.C. § 3553(a) it is unreasonable for a district court to impose a harsher sentence based on a defendant’s decision to plead guilty without a plea agreement.” The 10th U.S. Court of Appeals tells us it is unreasonable.

Facts of the Case

Ms. Leroya Cozad was indicted on a single charge of aiding and abetting the making of counterfeit currency. In negotiations with the Government, she offered to enter a guilty plea to four (4) years-probation. The government countered with a “low end” guidelines sentence. She chose to enter a guilty plea to the indictment, without a plea agreement. The probation officer determined that guideline range was 24 to 30 months. Subsequently and prior to sentencing, both parties submitted sentencing memorandums. Ms. Cozad sought probation and the government sought the low end of 24 months, which is what it had sought during plea negotiations.

The District Court rejected those requests and sentenced Ms. Cozad to 27 months, the midpoint of the guidelines. In explaining that decision the district court stated:

I think sometimes there’s sort of an implicit assumption that an individual is entitled to a sentence at the low end of the guideline range, but there is, of course, no such right. And, in practice, one of the factors I’ve used to lean against a low-end guideline range are defendants who submit a plea without a plea agreement, without the agreements that typically happen in a plea agreement.

That’s certainly their right to plead without a plea agreement, but they do not have a right to a low-end guideline range sentence, and so it’s always been my — again this isn’t a hard-and-fast rule by any means that I apply, but it’s always been my practice to say if someone agrees to a plea agreement, the additional conditions that are obtained in that, they’re entitled to additional consideration, which is where I start at a low-end guideline range.

But in my calculation, without a plea agreement, I have always started with looking more at the mid-tier of the guideline range, which is where I think the guidelines initially envisioned that courts would operate, and not giving them the additional credit for actually entering into a plea agreement to do that.

In this case, having thought about it obviously since our hearing last week, having looked through the briefs that have been filed in this case since that time, as well as additional information I’ve received from the probation office and going back through the report as well, it’s my conclusion that, first of all, Ms. Cozad has not shown demonstration that entitles her to a variance from the guideline recommendation of a custodial sentence, and I cannot find that that variance has been earned; and secondly, in light of the matters that I’ve just discussed overall, that a low-end guideline range sentence is not appropriate. [Emphasis in Opinion) Record cite at *2 Rec., vol. III at 42–44 (emphasis added).


The Court began its analysis with discussing that a defendant’s sentence was reviewed for either procedural or substantive reasonableness. The Court focused on the fact that that the district court “described its customary practice of distinguishing defendant’s based on the nature of their plea.” And also that while the “district court made a passing reference to ‘the agreements that typically happen in a plea agreement,’ the court did not specify what those agreements are.”

As we know, and the Court of Appeals discussed, those agreements can be appeal waivers and, fast track type agreements. The Court began by analyzing, the district court’s decision under 18 U.S.C. § 3553(a) stating that there were only four permissible factors that could be considered by statute. And while a court under 18 U.S.C. § 3661 is told that “no limitation shall be placed on the information concerning the background, character and conduct of a person…” That does not mean that all information is relevant to a defendant’s character.

The Court analyzed that the defendant’s need for rehabilitation was impermissible a factor under 18 U.S.C. § 3582(c) See United States v. Story, F.3d 1241, 1247-48 (10th Cir. 2011) or even worse, refusing to testify, in United States v. Cabrera, 811 F.3d 801, 808-09 (6th Cir. 2016) that Court held that a within-guidelines sentence was procedurally unreasonable because it was based on the defendant’s refusal to testify. The Court pointed to several other illegal considerations, such as prison sentences for inability to pay restitution and as Justice Stevens summarized in Rita v. United States, 551 U.S. 338, 365 (2007) for non-relevant factors, when he noted that it would be “unreasonable for a district court to give “harsh sentences to Yankees fans and lenient sentences to Red Sox fans.”

The Court then turned to Ms. Cozad’s sentence and stated that it did not see how the “fact of a defendant’s open plea, standing alone, bears any meaningful relationship to the § 3553(a) factors.” This was because a plea agreement is a contract and either party may reject the other party’s offer or refuse to negotiate entirely. It continued stating that the fact that the parties failed to mutually agree, as here where Ms. Cozad requested probation and the government requested the low end (of the properly calculated guidelines); that failed contractual negotiation and agreement is not something that provides “insight into the defendant’s character.” It also does not constitute one of the proper reasons under the statute to aid a district court in determining a sentence to comply with §3553(a)- a sentence that is sufficient, but not greater than necessary. The court stated that penalizing a defendant on the “absence of a plea agreement alone is arbitrary.”

The government tried to go back to the district courts use of the “other agreements” that a plea agreement may contain. The 10th Cir. has held that a district court could consider a negotiated appellate waiver because it could show acceptance of responsibility beyond the mere guilty plea. Think of a defendant who failed a urinalysis or violated some other item and is in need of demonstrating to a court that he wants to be committed to acceptance of responsibility.

The Court continued though and discussed that the cases that discussed the appellate waiver in a plea agreement as consideration, were, “fundamentally different than the absence of an appeal waiver in an agreement that never happened and whose terms remain purely hypothetical.”

The government’s next reason was that without providing a benefit for “uniformity purposes” and “additional leniency” to individuals who use a plea agreement, defendant’s would not have a “compelling” reason to use a plea agreement.. The Court explained that the Government’s theory was based on a false premise. Because a prosecutor could always file additional charges or seek aggressive interpretation of the guidelines to provide a “compelling reason”. See Fed. R. Crim. P. 11(c)(1)(B),(C). Therefore, with the government’s recommendation, a defendant who refuses to a plea bargain would still risk receiving a higher sentence in many cases. Thus providing a “compelling” reason to accept a plea agreement in some cases.

Lastly, the Court looked at fast track plea agreements. These are plea agreements where the government has provided an agreement with a specific departure under §5K3.1 of the United States Sentencing Guidelines. The Sentencing Commission adopted § 5K3.1 at the direction of Congress, which sought to incentivize early plea deals by accommodating fast track programs within the guidelines. In the author’s experience these usually provide anywhere for 1 to 2 points for offenses like alien smuggling or illegal re-entry. Fast Track Plea Agreements reward defendants with the set reduction by entering a plea usually prior to the indictment and for waiving any pre-trial motion practice. The Court recognized that these programs do not fit into the § 3553(a) factors, but do reflect Congress’ intent. In Cozad, the district court placed its own expression of this intent, where Congress had not spoken.


Ultimately this does not mean that a defendant cannot receive a harsher sentence if she enters a guilty plea without a plea agreement; rather a defendant cannot receive a harsher sentence simply because she did not have a plea agreement. Rule 11(c) prevents a judge from entering into the negotiations of a plea agreement.  See United States v. Carver, 160 F.3d 1266, 1269 (10th Cir. 1998). This does not mean that the government cannot provide the “compelling reasons” for a defendant to choose a plea agreement, only that the court cannot do so. Reflecting on plea decisions in federal court, this happens many times- role agreements, acceptance recommendations-where it may be at risk, and the hardest to obtain, charge bargaining where a charge is offered that does not contain a mandatory minimum. These are some quick examples of “compelling” reasons that can be negotiated in a beneficial plea agreement. However simply entering into a plea agreement in a single count fraud indictment may not provide any incentive or “compelling reason” where the government is recommending the low end of a properly calculated guideline. Here Cozad explains that you should not be punished for solely that act of acceptance and not taking a benefit-less agreement.

Federal Corner: Saving the Confrontation Clause


The Confrontation Clause

One of the greatest trial rights and protections owned by a criminal defendant is the Sixth Amendment right to confront and cross-examine witnesses at trial. The Sixth Amendment of the United States Constitution states that: “In all criminal prosecutions, the accused shall enjoy the right . . . to be confronted with the witnesses against him;”

 – U.S. Const. amend. VI.

The Confrontation Clause and the rule against hearsay found in the rules of evidence protect similar interests.  However, in California v. Green, 339 U.S. 149 (1970), the United States Supreme Court held that the 6th Amendment’s right to confrontation and the hearsay rule in the rules of evidence are not the same.  In doing so, the Court stated the following: “While it may readily be conceded that hearsay rules and the Confrontation Clause are generally designed to protect similar values, it is quite a different thing to suggest that the overlap is complete and that the Confrontation Clause is nothing more or less than a codification of the rules of hearsay and their exceptions as they existed historically at common law.”

The distinction of the confrontation right and the hearsay rule is significant.  Constitutional protections carry more weight than evidentiary rules in trial courts and on appeal. Further, the hearsay rule’s many exceptions do not apply to the confrontation clause.  California v. Green, 339 U.S. 149 (1970); See Barber v. Page, 390 U.S. 719 (1968); Pointer v. Texas, 380 U.S. 400 (1965).  

Crawford v. Washington

In 2004, the United States Supreme Court issued an important opinion in confrontation litigation, Crawford v. Washington, 541 U.S. 36 (2004).  In Crawford, husband, Michael Crawford, and wife, Sylvia Crawford, were charged related to the stabbing of a man. Both Michael and Sylvia gave recorded statements to the police at the police station regarding the incident. Michael admitted to stabbing the man in self-defense, but Sylvia’s statement to the police was inconsistent with Michael’s defense. At Michael’s trial, the State could not compel Sylvia to testify against Michael due to the spousal privilege rule in Washington. Therefore, the State introduced Sylvia’s prior recorded statement under the “statement against interest” exception to the hearsay rule over the Defense’s objection that the recording violated the Confrontation Clause. The Supreme Court in Crawford held that the introduction of Sylvia’s statement at trial without Sylvia appearing to testify in court violated the Confrontation Clause and was inadmissible. The Court held “that any out of court declaration that is testimonial in nature, is inadmissible if the declarant does not testify at trial and the Defendant has not had a prior opportunity to cross examine the witness.” 

Opening the Door Exception to Confrontation

In Hemphill v. New York, 2022 WL 174223 (2022), the Supreme Court had to decide whether the statutory exception to the Confrontation Clause violated the Sixth Amendment. The Confrontation Clause was under attack by New York, which had created an exception to the confrontation clause: “Opening the Door.”

Facts of the Case

A two-year old boy traveling in vehicle was killed by a stray 9-millimeter bullet shot by a person involved in a street fight. Police suspected that either Nicholas Morris or Darrel Hemphill was the shooter. A search of Morris’ apartment yielded 9-millimeter ammunition only and a .357-magnum handgun. Morris was initially charged with the murder of the child, but later was offered and agreed to plea to a charge related to possession of the .357-magnum handgun and dismissal of the murder charge.

Hemphill was then charged with the murder of the child. During Hemphill’s trial, Hemphill used a third-party culpability defense, blaming Morris for the murder.  During opening statement, Hemphill’s counsel told the jury that a search was conducted of Morris’s apartment hours after the shooting, and the police had recovered 9-millimeter ammunition, he ame aliber ammunition that had been used to shoot the boy. 

To controvert the Defense’s opening statement, the prosecution sought to introduce the plea colloquy transcript from Morris’ plea hearing in which Morris had pleaded guilty to possession of the .357-magnum handgun.  The State cited to the Reid Rule, as a judicially and legislatively created exception to the Confrontation Clause in New York that allowed the trial court to admit evidence at trial for the prosecution that would be otherwise inadmissible if the court determines that the defense has “opened the door” to the evidence by creating a misleading impression with the jury.  The Defense objected that the testimony sought by the prosecution (the plea transcript) violated the Confrontation Clause and Crawford v. Washington, because Morris was unavailable to testify and the defense had not had a previous opportunity to cross-examine him.

The trial court found that Hemphill’s attorney “opened the door” during opening statements by telling the jury about the 9-millimeter ammunition that was found in Morris’ apartment on the night of the murder. Therefore, the trial court allowed Morris’ plea colloquy transcript from the possession of the .357-magnum handgun charge into evidence to correct a “false impression” created by the defense.

Question Presented

Whether New York’s “opening the door” rule to the Confrontation Clause is a violation of the Confrontation Clause.   The rule allows the trial court to admit evidence for the prosecution at trial that would be otherwise inadmissible if the court determines that the defense has “opened the door” to the evidence by creating a misleading impression with the jury.


The Court first analyzed some of the history of Confrontation Clause Jurisprudence. In 1980, the Supreme Court held in Ohio v. Roberts, 448 U.S. 56 (1980) that the Confrontation Clause did not bar the admission of statements of an unavailable witness, so long as the statements bear an “adequate ‘indicia of reliability,’” meaning that they fell “with a firmly rooted hearsay exception” or other “particularized guarantees of trustworthiness.”

However, in Crawford v. Washington, 541 U.S. 36 (2004), the Supreme Court abrogated Ohio v. Roberts.  The Court held “that any out of court declaration that is testimonial in nature, is inadmissible if the declarant does not testify at trial and the Defendant has not had a prior opportunity to cross examine the witness.”  

State’s Arguments on Appeal

In its arguments to the Court, the State conceded that Morris’ plea colloquy was testimonial, meaning the Confrontation Clause was implicated.  However, the State argued that New York’s “opening the door” rule was not an exception to the Confrontation Clause.  Instead, the State argued that the “opening the door” rule was a procedural rule, like failing to object to the confrontation clause violation, and thus there was no violation. 

The Court has approved procedural rules and allows the States and the Government to adopt procedural rules related to objections to testimonial evidence.  For example, in Melendez-Dias v. Massachusetts, 557 U.S. 305 (2009), the Court approved “notice and demand” statutes.  These statutes require the State to give notice that it plans on introducing testimonial evidence (such as a lab report) without a sponsoring witness and the defense is given a deadline by which to object to the introduction of the evidence.  Failure to object is considered a waiver of the right to confront the witness. 

The Court’s Analysis – Procedural Rules

First, the Court emphasized that it approves of procedural rules that allow for admission of testimonial evidence.  The Court reiterated its approval of the “notice and demand” statutes discussed in Melendez-Dias.  The Court also approved the rule stated in Illinois v. Allen, 397 U.S. 337 (1970), which allows for removal of a criminal defendant from his trial when despite repeated warnings, he has become so disorderly, disruptive, and disrespectful in court that his trial cannot be cannot be carried on with him in the courtroom. 

The Court’s Analysis – Substantive Rules

However, the Court held that New York’s “opening the door” rule was not a procedural rule, but instead it was substantive.  In other words, the “opening the door” rule was a substantive rule like the one in Ohio v. Roberts, 448 U.S. 56 (1980) that allowed the testimonial statements of an unavailable witness, so long as the statements bear an “adequate ‘indicia of reliability,’” meaning that they fell “with a firmly rooted hearsay exception” or other “particularized guarantees of trustworthiness.”  However, Ohio v. Roberts was rejected in Crawford v. Washington, 541 U.S. 36 (2004).  In rejecting Ohio v. Roberts, Crawford stands for the principal that judges are barred “from substituting their own determinations of reliability for the method the Constitution guarantees.”  In other words, a judge should not substitute her wisdom about reliability for the reliability of cross-examination.

Similarly, the Court held that New York’s “opening the door” rule was substantive, requiring the trial court to weigh evidence. “It was not for the judge to determine whether Hemphill’s theory that Morris was the shooter was unreliable, incredible, or otherwise misleading in light of the State’s proffered, unconfronted plea evidence. Nor, under the Clause, was it the judge’s role to decide that this evidence was reasonably necessary to correct that misleading impression.  Such inquiries are antithetical to the Confrontation Clause.”

The Holding

Because New York’s “opening the door” rule was substantive, requiring the trial court to weigh evidence, the rule violated the Confrontation Clause.  Judges are not allowed to weigh the reliability, credibility, or misleading nature of testimony as a substitute for cross-examination. 

The Authors’ Thoughts

  • The Court properly distinguished between procedural and substantive rules.  Procedural rules like Texas’ Article 38.41 (Certificate of Analysis) which allows the State to give notice that it intends to introduce a laboratory report without a sponsoring witness, are approved because the defense has an opportunity to object to the evidence.  However, rules like New York’s “opening the door” rule rely on judge’s weighing the credibility, reliability, or weight of evidence are substantive in nature, and thus violate the principles set out in Crawford
  • This case is a significant opinion because the Court did not take a step back from Crawford. This is and 8-1 opinion.  Crawford is still the rule of law and confrontation continues to be one of the most important and protected trial rights for a defendant.

Federal Corner: Fifth Circuit Provides a Tool to Contest Firearm Enhancements


Experienced federal practitioners are conditioned to wince upon the mention of a gun in proximity to a drug deal or even just near a stash of drugs intended for sale. Those combinations can produce a two-level enhancement under USSG §2D1.1(b)(1), the drug Guideline, for possession of a firearm in connection with a drug offense; a four-level enhancement under USSG §2K2.1(b)(6), the firearm Guideline for use of a firearm in connection with another felony; a cross reference from the firearm Guideline to the drug Guideline under USSG §2K2.1(c)(1); or, worse, a 5-year mandatory minimum under 18 U.S.C. §924(c). Sometimes, it’s tempting not to fight the Guideline enhancements,1 even when the evidence seems to show nothing more than the mere presence of guns and drugs. The courts have hammered into us that firearms are “tools of the trade”2 when it comes to drug dealing, and the Application Notes to both USSG §2D1.1(b)(1)3 and USSG §2K2.1(b)(6)4 rather strongly encourage their application when guns and drugs are together. But, a recent Fifth Circuit opinion reminds us not to concede the issue too readily.

In United States v. Sincleair, __ F.4th __, No. 20-10495, 2021 WL 5001783 (5th Cir. Oct. 28, 2021), a drug defendant suffered arrest at the home of a downstream customer, that is, a man who bought drugs from the defendant’s own buyer. In fact, these two customers of the defendant were transacting two ounces of methamphetamine when police arrived. The police also found a gun in close proximity to all concerned. Although police ultimately found that the gun was registered to one of the customers (the resident), the district court nonetheless imposed a two-level enhancement for possessing a firearm in connection with the drug offense.

A divided panel of the Fifth Circuit vacated the sentence and remanded. The panel majority did not think the district court was sufficiently clear about the reason for the adjustment. That is, the district court did not clearly say whether the defendant had personally possessed the firearm, or whether, instead, he was vicariously responsible for another’s possession through principles of relevant conduct. The panel said:

It is not clear whether the district court determined that Sincleair personally possessed the firearm or that one of Sincleair’s “unindicted co-conspirators” possessed it during the commission of an offense. The PSR addendum presents both of these options as possibilities, and the district court did not explain which form of possession it attributed to Sincleair. In such a situation, our circuit precedent supports vacating the sentence and remand for the district court to make the appropriate findings.

Id. at *3. Sincleair thus confirms Fifth Circuit precedent. See United States v. Zapata-Lara, 615 F.3d 388 (5th Cir. 2010) (requiring district courts to make explicit findings in support of their Guideline calculations, including those underlying the gun enhancement to USSG §2D1.1).

Perhaps more significantly, the panel found insufficient evidence to support either a theory of personal possession or of vicarious sentencing liability through relevant conduct. It said:

Moreover, there is not enough in the record to support the firearm enhancement based on Sincleair’s personal possession of the firearm because the PSR did not include sufficient facts establishing a temporal and spatial relationship between the gun, the drug trafficking activity, and Sincleair. The Government (and the probation officer) did not provide any evidence establishing that Sincleair owned the weapon, brought the weapon with him to [downstream buyer]’s house, or had any other connection to it. Neither the PSR nor any other evidence supports a finding of temporal proximity between Sincleair’s drug trafficking activity and the weapon found in [downstream buyer]’s house. The only relevant facts in the PSR are that Sincleair was [direct customer]’s source for methamphetamine, and Sincleair and [direct customer] and their girlfriends were present at [downstream buyer]’s home for a social gathering around the time that [direct customer] sold an ounce of methamphetamine to [downstream buyer]. Thus, the only drug transaction that is documented in the PSR occurred in [downstream buyer]’s home between [direct customer] and [downstream buyer]. Even if it may be inferred that Sincleair sold the methamphetamine to [direct customer], there is no evidence of any temporal proximity between Sincleair’s sale and the presence of the weapon; there is no evidence that the sale occurred on the same day, same week, or even same month as [direct customer]’s sale to [downstream buyer]. There is also no evidence that Sincleair promoted or assisted in the sale in any way. The temporal connection between the firearm and any drug trafficking by Sincleair was thus tenuous at best.

Sincleair, __ F.4th at *4.

Though the opinion doesn’t say as much explicitly, it does seem to bolster a defendant’s argument against personal possession of a firearm – notwithstanding its proximity to both the defendant and the drugs – whether it might just as plausibly have been possessed by another person. It also supplies a good argument against the application of the gun enhancement based on possession by co-defendants. Specifically, it seems to hold that a co-defendant’s possession of a gun in connection with drugs will not justify the adjustment unless the defendant is involved in the particular transaction where the gun is present. Remarkably, this may be so even if he or she previously delivered the very quantity at issue.

Federal Corner: When Does “No” Mean No?


On August 10, 2021, a panel of the United States Court of Appeals for the Fifth Circuit held that a defendant who a jury determined did not know the quantity of drugs involved in a conspiracy could be found guilty of the conspiracy, but could not be sentenced for the quantity of drugs involved in the conspiracy. United States v. Aguirre-Rivera, 8 F.4th 405 (5th Cir., 2021). The case distinguishes between jury questions related to the primary offense versus those relating to punishment.

Background of the Case

Baltazar Aguirre-Rivera was charged with one count of conspiracy to possess with intent to distribute at least one kilogram of heroin in violation of 21 U.S.C. §§841(a)(1), (b)(1)(A)(I), and 846. At the end of Aguirre-Rivera’s trial, the district Court instructed the jury that it could find Aguirre-Rivera guilty only if the government had proven beyond a reasonable doubt: (1) “that two or more persons directly, or indirectly, reached an agreement to possess heroin with intent to distribute the same”; (2) “that the Defendant knew of the unlawful purpose of the agreement”; (3) “that the Defendant joined in the agreement willfully, and that is with the intent to further its unlawful purpose”; (4) “that the overall scope of the conspiracy involved at least one kilogram or more of a mixture or substance containing a detectable amount of heroin”; and(5) “that the Defendant knew, or reasonably should have known, that the scope of the conspiracy involved at least one kilogram or more of a mixture or substance containing a detectable amount of heroin.” See id. at 408.

Jury Questions and Answers

The district court provided the jury with a verdict form containing three questions. The first question asked for a general verdict of “Guilty” or “Not Guilty.” The jury answered this question “Guilty.” The second question asked, “Do you find beyond a reasonable doubt that the overall scope of the conspiracy involved at least one kilogram or more of a mixture or substance containing a detectable amount of heroin?” The jury responded “Yes” to this query. The final question asked, “Do you find beyond a reasonable doubt that the defendant knew or reasonably should have known that the scope of the conspiracy involved at least one kilogram or more of a mixture or substance containing a detectable amount of heroin?” The jury answered, “No.” See id.

Is there a Problem here?

The jury’s answer to the third question seems to fly in the face of the Court’s instructions. The jury found the defendant guilty in spite of the fact that they did not find that the government proved beyond a reasonable doubt that the Defendant knew or should have known the scope of the conspiracy; this was one of the questions the Court instructed the jury must be answered affirmatively in support of a “Guilty” verdict. Based on this discrepancy, Aguirre-Rivera moved for judgment of acquittal on the basis that the jury’s answer to the second special interrogatory contradicted, and therefore undermined, its general verdict of guilty. See id.

“No” Does Not Mean No

Despite the Court’s instructions, the judge did not issue a judgment of acquittal. The district court denied Aguirre-Rivera’s motion because, although the jury’s answer to the second special interrogatory “undermine[d] the fifth element of the jury charge, [it did] not negate an essential element of the jury’s finding of guilt.” See id.

The 5th Circuit addresses the First Issue

Aguirre-Rivera first challenged the district court’s denial of his motion for judgment of acquittal. He contended that the jury’s answer to the second special interrogatory, which found that he neither knew nor should have known that the conspiracy involved one kilogram or more of heroin, directly contradicted the fifth element of the jury charge. According to him, this contradiction undermined one of the elements that was necessary to support his conviction under the statute; and therefore it also undermined the guilty verdict altogether. See id. at 409

The Standard of Review

The panel for the Fifth Circuit started by stating “We review the denial of a motion for judgment of acquittal de novo. United States v. Buluc, 930 f.3d 383, 387 (5th Cir.), cert. denied, 140 S.Ct. 544(2019). ‘Courts consistently vacate convictions when the answers to special interrogatories undermine a finding of guilt the jury made on the general questions.’ United States v. Gonzales, 841 F.3d. 339, 348 (5th Cir. 2016)). If the jury’s answer to the second special interrogatory did undermine an essential element of the charged offense, then the district court should have granted Aguirre-Rivera’s motion for the judgment of acquittal. See id. Our task, then, is to determine whether the jury’s answer to the special interrogatory undermined an essential element of Aguirre-Rivera’s conviction.” See Aguirre-Rivera 8 F.4th at 409.

Element versus Enhancement

The opinion then lays out the elements of the offense charged. “The essential elements of a drug conspiracy are (1) an agreement by two or more persons to violate the narcotics laws; (2) a defendant’s knowledge of the agreement; and (3) his voluntary participation in the agreement.” See id. (quoting United States v. Bargas-Ocampo, 747 f.3d 299, 303 (5th Cir. 2014) (en banc)). The Court then stated, “That is all the government needs to prove to sustain a drug conspiracy conviction under 21 USC §§ 841(a)(1) and 846.” See id. at 410. The Court reasoned that an enhancement is different from an element of the primary offense because it only affects punishment, not guilt. Although it must be submitted to the jury as an element, because it increases the mandatory punishment under Alleyne v. United States, 570 U.S. 99, 103 (2013), a finding of fact that affects only the legally prescribed punishment does not become an element of the conspiracy offense. See id. at 410-11. In so holding, the Court cited United States v. Daniels, 723 F.3d 562 , 573 (5th Cir.) 2013, as supporting its position. On that basis, the Court sustained the district court’s denial of the judgment for acquittal. See id. at 411.

Sentencing Implications

The Court then addressed the impact of the jury’s finding that Aguirre-Rivera did not have knowledge of the scope of the conspiracy. The Court held that even though the conviction was not affected by the jury’s finding, “the sentence most certainly was.” The defendant was sentenced under the guideline range for Conspiracy to Possess with Intent to Distribute 100 Gams or More of Heroin in violation of 21 U.S.C. §§ 841(b)(1)(B). This was improper since the jury’s negative answer to the question regarding the amount of heroin involved in the conspiracy negated any enhancements under § 841(b).” See id.

Legal Basis for Sentencing Implications

The Supreme Court has held “that factual determinations that increase maximum or minimum sentences, other than a prior conviction, must be found by a jury beyond a reasonable doubt.” Apprendi v. New Jersey, 530 U.S. 466, 490 (2000); see also Alleyne v. United States, 133 S.Ct. 2151, 2158, (2013); United States v. Haines, 803 F.3d 713, 738 (5th Cir. 2015).

The Aguirre-Rivera Court held: “Because the quantity of heroin involved” in a drug conspiracy case can affect a defendant’s “minimum sentence[] under §841, it must be found by a jury.” Id. In this case the jury’s answer to the second special interrogatory negated any enhancements under §841(b). Therefore, Aguirre-Rivera could not be subject to any mandatory minimum. He should have been sentenced under  §841(b)(1)(C), which gives the sentencing range for drug conspiracy violations not subject to additional enhancements under §841(b)(1)(A), (B), or (D). Aguirre-Rivera, 8 F.4th at 411.

One More Time, in English

Aguirre-Rivera was charged with participating in a conspiracy involving a kilo or more of heroin under 21 U.S.C §§ 841(a)(1), (b)(1)(A)(I), and 846. The jury found that he did not know the conspiracy involved over a kilo of heroin. So, the district court sentenced Aguirre-Rivera under §841(b)(1)(B) for a conspiracy involving 100 grams or more of heroin, which carries a minimum of 5 years. However, the jury did not make a finding that Aguirre-Rivera knew any amount of heroin involved in the conspiracy, so he could not be charged with even that reduced amount of heroin. Aguirre-Rivera should have been sentenced under §841(b)(1)(C), which is the range of punishment for an offense without an enhancement for the amount of drugs involved in the conspiracy.

The district court sentenced Aguirre-Rivera under a statute which had a range of punishment of 5 years to 40 years. He should have been sentenced within a range of punishment of no more than 20 years. The proper statute for sentencing was §841(b)(1)(C), which has no mandatory minimum sentence.

Was there Harm?

Aguirre-Rivera was sentenced to 60 months in prison. This was the minimum sentence under §841(b)(1)(B). The length of the sentence held significance with the Court. “The court then proceeded to sentence Aguirre-Rivera to 60 months in prison and three years of supervised release- a sentence coinciding almost exactly with the mandatory minimum under §841(b)(1)(B).” Aguirre-Rivera, 8 F.4th at 412. Since the Court held that Aguirre-Rivera should have been sentenced without any mandatory minimum under §841(b)(1)(C), the Court found that there was harm. The Court noted that the district court, despite saying that the sentence would have been the same even if any error was made in the calculation of the guidelines, also stated, “[i]f it turns out that [Aguirre-Rivera’s] lawyers are correct, and if we have lower guidelines, I would be the first to be happy to revisit the case in order to make a correction to any mistake that this may have resulted in.” Id. The Court found that this statement confirmed that Aguirre-Rivera may have been harmed by the error. The Court stated, “Given that the district court expressed willingness to revisit the case and correct any errors inherent in Aguirre-Rivera’s sentence, we cannot say that the government has carried its burden of demonstrating beyond a reasonable doubt that the district court would have imposed the same sentence regardless of any error.” Id. at 412-13.

My Thoughts

In my opinion, the Aguirre-Rivera case is more important to the trial lawyer than it is to the appellate practitioner. The case is a reminder to the trial lawyer, facing the prospects of trying an unwinnable case or pleading to an outlandish guideline range, that there may be an issue to try if the client held a minor role in the conspiracy. This issue is often present in cases involving “mules” or persons  enlisted to perform tasks by the primary conspirators. The results of a jury finding that the defendant lacked knowledge of the scope of the conspiracy at trial could greatly reduce a client’s guidelines just as it did in this case. Winning this issue could go a long way toward earning your client    a deduction for his role in the offense. It would be hard for the judge to find your client was a major player in the conspiracy when a jury found that your client did not know the scope of the conspiracy. See USSG §§2D1.1(a)(5) and 3B1.2. A finding that your client played a minor role in the conspiracy could also eliminate the two-level increase for importation in cases involving methamphetamine under USSG § 2D1.1(b)(5). Remand for sentencing under the correct guidelines was made possible in this case because the district court judge was more interested in the correct application of the law than in protecting against a remand.

1 2 3 12