4. Analysis: Precedents and Legal Reasoning
A. Statutory structure makes the intra-Indian provision an exception, not an element
The court begins with the separation between § 1152’s two paragraphs. The first establishes an intelligible general rule: federal enclave criminal law extends to Indian country. The second states exceptions. The majority concludes that the general rule can be accurately described without incorporating the intra-Indian exception into the definition of every offense.
This structure matters because the ordinary rule of statutory interpretation assigns responsibility for establishing a separately stated exception to the party invoking it. The court treats the statute’s organization as evidence of that allocation rather than reasoning backward from the importance of Indian status to the statute’s practical coverage.
McKelvey, Cook, and Cunningham
McKelvey v. United States, 260 U.S. 353, 357 (1922), provides the principal interpretive rule. An indictment based on a general statutory prohibition ordinarily need not negate an exception contained in a separate proviso or distinct clause. The person relying on the exception must raise and establish it.
United States v. Dickson, 40 U.S. (15 Pet.) 141 (1841), establishes the historical pedigree of that approach. Dixon v. United States, 548 U.S. 1 (2006), reaffirms it in modern criminal law. Together, these cases support treating § 1152’s separately stated intra-Indian exception as a defense.
The countervailing principle comes from United States v. Cook, 84 U.S. (17 Wall.) 168 (1872). An apparent exception must sometimes be included in the indictment when it is so integrated into the offense’s definition that the offense cannot be accurately described without it. Prentiss II relied on that principle because the parties’ statuses are central to determining § 1152’s practical reach.
The decisive intervening authority is Cunningham v. Cornell University, 604 U.S. 693 (2025). Although arising outside criminal law, it expressly reaffirms the general exception-as-defense rule and explains that United States v. Cook has been applied narrowly, including where the exception appears within the same sentence defining the offense. The majority concludes that § 1152’s separate paragraphs fit the general rule, not the narrow exception.
This is the opinion’s central interpretive move: a limitation can be important to statutory coverage without becoming an element of the offense. The court also finds no principled basis for treating the intra-Indian exception differently from § 1152’s other separately listed exceptions.
Why § 1152’s distinctive function does not change the analysis
Ruiz argued that § 1152 does not independently define prohibited conduct and therefore falls outside the rule governing exceptions to criminal offenses. The court rejects this distinction. Section 1152 works together with substantive criminal statutes and adds a territorial requirement: the offense must occur in Indian country. Its role in extending federal law does not make ordinary principles of statutory interpretation inapplicable.
Nicoli v. Briggs, 83 F.2d 375 (10th Cir. 1936), supplied language used by Ruiz to characterize the exception rule as applying to statutes stating a clear general offense. The court does not read that language as establishing his proposed categorical limitation. United States v. Carey, 929 F.3d 1092 (9th Cir. 2019), appears as another application recognizing an affirmative defense through analysis of a general prohibition and its exception.
B. The partial overruling of Prentiss II
Understanding the change requires separating the two majority opinions issued in Prentiss II.
-
Judge Henry’s majority opinion treated the Indian or non-Indian statuses of both defendant and victim as essential elements under § 1152.
-
Judge Baldock’s separate majority opinion held that omission of those allegations from an indictment did not eliminate the district court’s subject-matter jurisdiction and was subject to harmless-error review.
The present decision overturns the former rule insofar as it makes the defendant’s non-Indian status an element in a prosecution involving an Indian victim. It does not revive the argument that status allegations determine federal courts’ subject-matter jurisdiction.
The earlier litigation began with United States v. Prentiss, 206 F.3d 960 (10th Cir. 2000), known as Prentiss I, which found the indictment defective for omitting status allegations. After the en banc ruling, Prentiss III concluded that the omission was not harmless. That history demonstrates why the element-versus-defense classification affects both charging practice and trial proof.
Lucas and Famous Smith do not decide the present question
Prentiss II relied heavily on Lucas v. United States, 163 U.S. 612 (1896), and Famous Smith v. United States, 151 U.S. 50 (1894). The present court substantially narrows their relevance.
Lucas v. United States rejected an instruction permitting an inference about a victim’s non-Indian status from race. But it concerned a different statute governing tribal-court jurisdiction, not § 1152 or its predecessors. It therefore does not control the allocation of proof concerning a defendant’s status under § 1152.
Famous Smith v. United States concerned whether the evidence established that the victim was an Indian. Although it described a jury instruction stating that the government had to establish the victim’s non-Indian status, the Supreme Court expressly declined to endorse the instruction’s legal propositions. The Tenth Circuit thus reads it as an evidentiary-sufficiency decision, not a definitive allocation of pleading or production burdens—and certainly not a decision about the defendant’s Indian status.
The majority also rejects reliance on the reporter’s syllabus as a substitute for the Court’s actual reasoning. Its reference to Sparf v. United States, 156 U.S. 51 (1895), identifies the limited subsequent citation history of Famous Smith v. United States; that observation is supplementary rather than an independent reason to disregard precedent.
Hilderbrand v. United States, 261 F.2d 354 (9th Cir. 1958), and United States v. Torres, 733 F.2d 449 (7th Cir. 1984), supported the earlier court’s emphasis on status allegations and victim status. The en banc majority does not regard those decisions as resolving the narrower question now before it.
Stare decisis and intervening authority
The majority characterizes Prentiss II as a precedent undermined by subsequent Supreme Court guidance, especially Dixon v. United States and Cunningham v. Cornell University. It invokes Pearson v. Callahan, 555 U.S. 223 (2009), for the propositions that stare decisis is not absolute and that reliance interests are comparatively weak for procedural and evidentiary rules that do not structure parties’ primary conduct.
The distinction is consequential: changing who proves Indian status does not change whether child sexual abuse was prohibited. Nevertheless, the court recognizes that defendants may have relied on the former rule in conducting their trials. That litigation reliance remains relevant to the remedy on remand.
C. Why the defendant also bears the ultimate burden of persuasion
Classifying Indian status as an affirmative defense does not, by itself, settle every question about proof. Some defenses require defendants only to produce enough evidence to raise the issue, after which the government must disprove them. The majority expressly considers—and rejects—that arrangement here.
The common-law default and its constitutional boundary
Dixon v. United States supports a default under which a defendant must establish an affirmative defense when Congress has not prescribed a different rule. The opinion also quotes Justice Alito’s concurrence in Ruan v. United States, 597 U.S. 450 (2022), describing the historical allocation of both production and persuasion burdens to defendants. The cited proposition from Ruan v. United States is therefore a concurring statement, not a majority holding.
The constitutional limit comes from Smith v. United States, 568 U.S. 106 (2013), and Patterson v. New York, 432 U.S. 197 (1977): the government cannot escape its obligation to prove an element by assigning the defendant a defense that negates that element.
United States v. Unser, 165 F.3d 755 (10th Cir. 1999), and United States v. Lofton, 776 F.2d 918 (10th Cir. 1985), supply corresponding circuit authority. When defense evidence would negate an offense element, the government retains the ultimate burden on that element.
Here, however, the majority has already concluded that non-Indian status is not an element. Establishing the intra-Indian exception does not deny that the prohibited act occurred or that it occurred in Indian country. It establishes a statutory reason why § 1152 does not extend to the case. On that understanding, the Constitution does not require the government to disprove the defense beyond a reasonable doubt.
United States v. Brown, 128 F.4th 1358 (10th Cir. 2025), describes the initial production burden for a recognized defense as generally not onerous. United States v. Sturm, 672 F.3d 891 (10th Cir. 2012), confirms the government’s responsibility to prove jurisdictional elements, but the majority finds that principle inapplicable to the defendant’s status.
The disagreement with the Fifth and Ninth Circuits
United States v. Hester, 719 F.2d 1041 (9th Cir. 1983), and United States v. Haggerty, 997 F.3d 292 (5th Cir. 2021), both treat Indian status as a defense that the defendant must initially raise. The Tenth Circuit follows them on production but rejects their allocation of ultimate persuasion to the government.
The majority finds the Ninth Circuit’s reasoning underdeveloped. It traces United States v. Hester through United States v. Guess, 629 F.2d 573 (9th Cir. 1980), to Davis v. United States, 160 U.S. 469 (1895). According to the majority, Dixon v. United States later explained that the approach in Davis v. United States reflected the particular statute’s heightened mental-state requirement, not a constitutionally mandated general rule requiring prosecutors to disprove every affirmative defense.
The Fifth Circuit reasoned that Indian status affects the reach of the federal enclave law’s territorial element. The Tenth Circuit rejects that connection: an intra-Indian offense may fall outside § 1152, but that does not negate the fact that it occurred in Indian country. Its disagreement is therefore substantive, not merely terminological.
Why the statute-of-limitations analogy fails
Relying on Musacchio v. United States, 577 U.S. 237 (2016), Ruiz argued that Indian status resembles a limitations defense: neither disputes the criminal nature of the act, and both reflect legislative limits on prosecution.
The majority responds that this description fits many affirmative defenses and does not establish who must prove them. It also relies on Smith v. United States to distinguish the practical difficulty of disproving a defendant-specific event or condition from the more manageable task of proving when an offense occurred.
Information asymmetry as a supporting rationale
The court considers defendants better placed to produce evidence of their own ancestry, tribal relationships, benefits, and community recognition. Requiring the prosecution to negate every possible basis for Indian status could demand proof about an extensive family history and connections to tribes across the country.
Selma, R. & D.R. Co. v. United States, 139 U.S. 560 (1891), supports assigning proof of a particular fact to the party with peculiar knowledge of it. United States v. Hartsock, 347 F.3d 1 (1st Cir. 2003), provides an example of allocating both burdens on an affirmative defense partly because the defendant possessed substantially better information.
This reasoning supplements the majority’s statutory and common-law analysis. It does not create an unrestricted principle permitting courts to shift an element’s burden merely because defendants know more about it.
D. Three meanings of “jurisdiction” must be kept separate
Much of the opinion addresses the defendants’ argument that Indian-status diversity is “jurisdictional.” The majority distinguishes Congress’s authority to enact a law, an offense element connecting conduct to that authority, and a court’s power to adjudicate a federal prosecution.
Congressional authority over Indian-country crime
United States v. Antelope, 430 U.S. 641 (1977), states that Congress has constitutional power to prescribe a criminal code for Indian country. Haaland v. Brackeen, 599 U.S. 255 (2023), and United States v. Lara, 541 U.S. 193 (2004), describe Congress’s authority concerning Indian tribes as plenary.
United States v. Kagama, 118 U.S. 375 (1886), upheld the Major Crimes Act. Its recognition of federal authority over serious intra-Indian crime undermines the argument that diversity of Indian status is constitutionally indispensable to congressional criminal regulation.
The opinion contrasts this setting with United States v. Lopez, 514 U.S. 549 (1995), where the absence of an interstate-commerce connection mattered to Congress’s authority over gun possession near schools. United States v. Chavarria, 140 F.4th 1257 (10th Cir. 2025), United States v. Patton, 451 F.3d 615 (10th Cir. 2006), and Torres v. Lynch, 578 U.S. 452 (2016), explain how jurisdictional elements connect regulated conduct to an enumerated federal power.
For § 1152, the majority identifies the Indian-country location—not the defendant’s status—as the relevant jurisdictional hook. United States v. Jackson, 82 F.4th 943 (10th Cir. 2023), and United States v. Tony, 637 F.3d 1153 (10th Cir. 2011), support treating Indian-country location as a jurisdictional element.
The court expressly leaves aside the victim-status issue addressed in United States v. Reza-Ramos, 816 F.3d 1110 (9th Cir. 2016). Its discussion of that decision should not be mistaken for a holding resolving proof of victim status in other configurations.
The majority also rejects constitutional avoidance and lenity as grounds for the defendants’ construction. Referring to Rewis v. United States, 401 U.S. 808 (1971), it finds no relevant ambiguity about Congress’s authority requiring the proposed narrower interpretation.
Federal courts’ subject-matter jurisdiction
Section 3231 supplies the district courts’ authority to hear federal criminal cases. As United States v. Tony, quoting United States v. White Horse, 316 F.3d 769 (8th Cir. 2003), explains, that statute begins and ends the subject-matter-jurisdiction inquiry in a federal prosecution.
Failure to prove a statutory condition may defeat a prosecution on the merits without depriving the court of adjudicatory authority. United States v. Martin, 147 F.3d 529 (7th Cir. 1998), had already supported that distinction in Prentiss II.
The majority draws on Santos-Zacaria v. Garland, 598 U.S. 411 (2023), and Fort Bend County. v. Davis, 587 U.S. 541 (2019), to warn against treating every statutory restriction labeled “jurisdictional” as a limitation on adjudicatory power. Harrow v. Dep't of Def., 601 U.S. 480 (2024), and Arbaugh v. Y&H Corp., 546 U.S. 500 (2006), reinforce that caution.
Reconciling older Indian-law decisions
Ex parte Crow Dog, 109 U.S. 556 (1883), held that the predecessor statutory framework did not authorize federal prosecution for the murder of one Indian by another. Its references to limited jurisdiction supported Hebert’s argument, but the majority reads that terminology in historical context.
United States v. Cotton, 535 U.S. 625 (2002), drawing on Steel Co. v. Citizens for a Better Env't, 523 U.S. 83 (1998), explains that older decisions used “jurisdiction” more expansively than modern doctrine permits. The discussion of Ex parte Bain, 121 U.S. 1 (1887), illustrates how historical limits on criminal appellate and habeas review encouraged that broader usage.
Particularly important is Louie v. United States, 254 U.S. 548 (1921). It treated the absence of an Indian-country element as a merits issue concerning federal authority over the offense, rather than a defect in the convicting court’s subject-matter jurisdiction. The majority regards that distinction as strongly supporting its treatment of § 1152’s exceptions.
The defendants also invoked jurisdictional language in Keeble v. United States, 412 U.S. 205 (1973), United States v. Bryant, 579 U.S. 140 (2016), and McGirt v. Oklahoma, 591 U.S. 894 (2020). The majority does not deny those decisions’ substantive allocation of governmental authority. It rejects the inference that their terminology determines the modern procedural classification at issue here.
Finally, Patchak v. Zinke, 583 U.S. 244 (2018), supplies a contrast: legislation expressly directing that actions cannot be filed or maintained in federal court can strip jurisdiction. Section 1152 contains no comparable language. Sebelius v. Auburn Reg'l Med. Ctr., 568 U.S. 145 (2013), further supports requiring a clear congressional statement before treating a statutory restriction as jurisdictional.
E. The federal-state-tribal background remains intact
The decision concerns proof within an established jurisdictional framework, not a wholesale revision of that framework.
-
United States v. McBratney, 104 U.S. 621 (1882), and Draper v. United States, 164 U.S. 240 (1896), establish state authority over non-Indian-on-non-Indian crime in Indian country. Donnelly v. United States, 228 U.S. 243 (1913), explains the historical effect of statehood on that allocation.
-
United States v. Walker, 85 F.4th 973 (10th Cir. 2023), describes § 1152’s practical application to crimes crossing the Indian/non-Indian divide.
-
Oklahoma v. Castro-Huerta, 597 U.S. 629 (2022), establishes concurrent federal and state jurisdiction over crimes by non-Indians against Indians. Its textual description of § 1152 as extending federal law—rather than converting Indian country into an exclusively federal enclave—supports the present court’s reading.
-
United States v. Smith, 178 F.4th 572 (10th Cir. 2026), supplies historical statutory background. United States v. Brice, 167 F.4th 1330 (11th Cir. 2026), quoting United States v. Wadena, 152 F.3d 831 (8th Cir. 1998), explains Congress’s enactment of the Major Crimes Act following Ex parte Crow Dog.
-
United States v. Wheeler, 435 U.S. 313 (1978), provides contextual discussion of traditional tribal justice. It is not a basis for the new burden-of-proof rule.
The court therefore preserves the substantive intra-Indian exception while changing how a defendant establishes entitlement to it.
F. The Indian canon does not determine the outcome
The Indian canon generally favors interpreting ambiguous federal enactments for the benefit of Indians. Ramah Navajo Chapter v. Lujan, 112 F.3d 1455 (10th Cir. 1997), E.E.O.C. v. Cherokee Nation, 871 F.2d 937 (10th Cir. 1989), and Montana v. Blackfeet Tribe, 471 U.S. 759 (1985), supply the governing formulation.
The parties offered competing understandings of tribal benefit. The government emphasized protection of Indian victims and community safety; defendants emphasized tribal self-government and restraint on federal intervention. Oneida County v. Oneida Indian Nation of N.Y. State, 470 U.S. 226 (1985), supplied the canon’s trust-relationship foundation.
The majority first finds no relevant ambiguity. Under South Carolina v. Catawba Indian Tribe, Inc., 476 U.S. 498 (1986), the canon cannot manufacture ambiguity or override clearly expressed congressional intent.
Alternatively, Negonsott v. Samuels, 507 U.S. 99 (1993), drawing on Bryan v. Itasca County, 426 U.S. 373 (1976), cautions against equating benefits to tribes with benefits to accused Indian defendants while disregarding victims and tribal interests. The majority also emphasizes that reallocating proof does not formally change § 1152’s covered conduct, territory, or exceptions.
The resulting holding is restrained: the canon has, at most, limited relevance here, and the court does not rely on it. The decision does not announce that the canon lacks force in federal criminal Indian-law cases generally.
5. Retrial, Double Jeopardy, and Retroactive Application
A. Why the vacated panel decisions do not bar retrial
Burks v. United States, 437 U.S. 1 (1978), establishes that a determination of evidentiary insufficiency ordinarily bars a second trial. The defendants argued that their successful panel appeals were equivalent to acquittals.
The en banc court rejects that premise because of the appeals’ procedural posture. Under the circuit’s rule, granting rehearing en banc vacated the panel judgments, stayed the mandates, and restored the cases as pending appeals. The en banc decision then expressly vacates the panel opinions. Neither defendant obtained a trial acquittal, and neither retained an operative appellate judgment of insufficiency.
United States v. Scott, 437 U.S. 82 (1978), Evans v. Michigan, 568 U.S. 313 (2013), and Arizona v. Rumsey, 467 U.S. 203 (1984), concern the finality accorded acquittals even when based on legal mistakes. United States v. Jenkins, 420 U.S. 358 (1975), involved proceedings ending without a guilty verdict. The court distinguishes those circumstances from jury convictions still undergoing direct appellate review.
Accordingly, the decision does not authorize reopening a final acquittal. Its conclusion depends on the absence of such an acquittal and the superseding effect of en banc review.
B. Why due process does not categorically prohibit applying the new rule
The Ex Post Facto Clause directly restricts legislatures, while due process imposes related, less extensive restrictions on retroactive judicial interpretations of criminal law. United States v. Budder, 76 F.4th 1007 (10th Cir. 2023), supplies the circuit framework.
Under Rogers v. Tennessee, 532 U.S. 451 (2001), the relevant concern is an interpretation that is unexpected and indefensible by reference to previously expressed law. The majority gives two reasons for rejecting that objection.
-
The ruling changes proof allocation, not the criminal prohibition. The defendants had notice that the charged sexual abuse was unlawful.
-
Treating Indian status as an affirmative defense was already supported by longstanding Supreme Court principles and decisions from other circuits.
United States v. Muskett, 970 F.3d 1233 (10th Cir. 2020), supports the proposition that an existing circuit disagreement can provide notice that a broader construction may ultimately prevail, even where the defendant’s own circuit previously followed a narrower view.
This constitutional conclusion is distinct from the remedial question whether defendants reasonably relied on Prentiss II when deciding what evidence to present. The court allows that concern to be addressed in further proceedings.
C. What remains unresolved on remand
The court returns the cases to their original panels rather than directly ordering new trials. Hebert’s remaining claims concern the absence of an instruction defining non-Indian status and the prosecution’s use of his silence about that status. Ruiz’s remaining claims concern vagueness and a modified Allen instruction.
Those issues were not resolved by the earlier panels and may affect whether further district-court proceedings are appropriate. The en banc court does not decide whether either defendant actually qualifies as an Indian.
Exby-Stolley v. Bd. of Cnty. Comm'rs, 979 F.3d 784 (10th Cir. 2020), supports deciding a recurring legal question to guide proceedings on remand. Kirkendall v. Dep't of Army, 479 F.3d 830 (Fed. Cir. 2007), and United States v. Padilla, 415 F.3d 211 (1st Cir. 2005), explain the relationship between an en banc order and the issues considered by the full court.
Contreras-Bocanegra v. Holder, 678 F.3d 811 (10th Cir. 2012), United States v. Nacchio, 555 F.3d 1234 (10th Cir. 2009), and United States v. Perez, 989 F.2d 1574 (10th Cir. 1993), provide procedural examples of en banc courts vacating or replacing panel rulings after revising circuit precedent.